* [dpdk-dev] [RFC 0/6] New sync modes for ring
@ 2020-02-24 11:35 Konstantin Ananyev
2020-02-24 11:35 ` [dpdk-dev] [RFC 1/6] test/ring: add contention stress test Konstantin Ananyev
` (8 more replies)
0 siblings, 9 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-02-24 11:35 UTC (permalink / raw)
To: dev; +Cc: olivier.matz, Konstantin Ananyev
Upfront note - that RFC is not a complete patch.
It introduces an ABI breakage, plus it doesn't update ring_elem
code properly, etc.
I plan to deal with all these things in later versions.
Right now I seek an initial feedback about proposed ideas.
Would also ask people to repeat performance tests (see below)
on their platforms to confirm the impact.
More and more customers use(/try to use) DPDK based apps within
overcommitted systems (multiple acttive threads over same pysical cores):
VM, container deployments, etc.
One quite common problem they hit: Lock-Holder-Preemption with rte_ring.
LHP is quite a common problem for spin-based sync primitives
(spin-locks, etc.) on overcommitted systems.
The situation gets much worse when some sort of
fair-locking technique is used (ticket-lock, etc.).
As now not only lock-owner but also lock-waiters scheduling
order matters a lot.
This is a well-known problem for kernel within VMs:
http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
The problem with rte_ring is that while head accusion is sort of
un-fair locking, waiting on tail is very similar to ticket lock schema -
tail has to be updated in particular order.
That makes current rte_ring implementation to perform
really pure on some overcommited scenarios.
While it is probably not possible to completely resolve this problem in
userspace only (without some kernel communication/intervention),
removing fairness in tail update can mitigate it significantly.
So this RFC proposes two new optional ring synchronization modes:
1) Head/Tail Sync (HTS) mode
In that mode enqueue/dequeue operation is fully serialized:
only one thread at a time is allowed to perform given op.
As another enhancement provide ability to split enqueue/dequeue
operation into two phases:
- enqueue/dequeue start
- enqueue/dequeue finish
That allows user to inspect objects in the ring without removing
them from it (aka MT safe peek).
2) Relaxed Tail Sync (RTS)
The main difference from original MP/MC algorithm is that
tail value is increased not by every thread that finished enqueue/dequeue,
but only by the last one.
That allows threads to avoid spinning on ring tail value,
leaving actual tail value change to the last thread in the update queue.
Test results on IA (see below) show significant improvements
for average enqueue/dequeue op times on overcommitted systems.
For 'classic' DPDK deployments (one thread per core) original MP/MC
algorithm still shows best numbers, though for 64-bit target
RTS numbers are not that far away.
Numbers were produced by ring_stress_*autotest
(first patch in these series).
X86_64 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
DEQ+ENQ average cycles/obj
MP/MC HTS RTS
1thread@1core(--lcores=6-7) 8.00 8.15 8.99
2thread@2core(--lcores=6-8) 19.14 19.61 20.35
4thread@4core(--lcores=6-10) 29.43 29.79 31.82
8thread@8core(--lcores=6-14) 110.59 192.81 119.50
16thread@16core(--lcores=6-22) 461.03 813.12 495.59
32thread/@32core(--lcores='6-22,55-70') 982.90 1972.38 1160.51
2thread@1core(--lcores='6,(10-11)@7' 20140.50 23.58 25.14
4thread@2core(--lcores='6,(10-11)@7,(20-21)@8' 153680.60 76.88 80.05
8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 280314.32 294.72 318.79
16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 643176.59 1144.02 1175.14
32thread@2core(--lcores='6,(10-25)@7,(30-45)@8' 4264238.80 4627.48 4892.68
8thread@2core(--lcores='6,(10-17)@(7,8))' 321085.98 298.59 307.47
16thread@4core(--lcores='6,(20-35)@(7-10))' 1900705.61 575.35 678.29
32thread@4core(--lcores='6,(20-51)@(7-10))' 5510445.85 2164.36 2714.12
i686 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
DEQ+ENQ average cycles/obj
MP/MC HTS RTS
1thread@1core(--lcores=6-7) 7.85 12.13 11.31
2thread@2core(--lcores=6-8) 17.89 24.52 21.86
8thread@8core(--lcores=6-14) 32.58 354.20 54.58
32thread/@32core(--lcores='6-22,55-70') 813.77 6072.41 2169.91
2thread@1core(--lcores='6,(10-11)@7' 16095.00 36.06 34.74
8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 1140354.54 346.61 361.57
16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 1920417.86 1314.90 1416.65
8thread@2core(--lcores='6,(10-17)@(7,8))' 594358.61 332.70 357.74
32thread@4core(--lcores='6,(20-51)@(7-10))' 5319896.86 2836.44 3028.87
Konstantin Ananyev (6):
test/ring: add contention stress test
ring: rework ring layout to allow new sync schemes
ring: introduce RTS ring mode
test/ring: add contention stress test for RTS ring
ring: introduce HTS ring mode
test/ring: add contention stress test for HTS ring
app/test/Makefile | 3 +
app/test/meson.build | 3 +
app/test/test_pdump.c | 6 +-
app/test/test_ring_hts_stress.c | 28 ++
app/test/test_ring_rts_stress.c | 28 ++
app/test/test_ring_stress.c | 27 ++
app/test/test_ring_stress.h | 477 +++++++++++++++++++
lib/librte_pdump/rte_pdump.c | 2 +-
lib/librte_port/rte_port_ring.c | 12 +-
lib/librte_ring/Makefile | 4 +-
lib/librte_ring/meson.build | 4 +-
lib/librte_ring/rte_ring.c | 84 +++-
lib/librte_ring/rte_ring.h | 619 +++++++++++++++++++++++--
lib/librte_ring/rte_ring_elem.h | 8 +-
lib/librte_ring/rte_ring_hts_generic.h | 228 +++++++++
lib/librte_ring/rte_ring_rts_generic.h | 240 ++++++++++
16 files changed, 1721 insertions(+), 52 deletions(-)
create mode 100644 app/test/test_ring_hts_stress.c
create mode 100644 app/test/test_ring_rts_stress.c
create mode 100644 app/test/test_ring_stress.c
create mode 100644 app/test/test_ring_stress.h
create mode 100644 lib/librte_ring/rte_ring_hts_generic.h
create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [RFC 1/6] test/ring: add contention stress test
2020-02-24 11:35 [dpdk-dev] [RFC 0/6] New sync modes for ring Konstantin Ananyev
@ 2020-02-24 11:35 ` Konstantin Ananyev
2020-02-24 11:35 ` [dpdk-dev] [RFC 2/6] ring: rework ring layout to allow new sync schemes Konstantin Ananyev
` (7 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-02-24 11:35 UTC (permalink / raw)
To: dev; +Cc: olivier.matz, Konstantin Ananyev
Introduce new test-case to measure ring perfomance under contention
(miltiple producers/consumers).
Starts dequeue/enqueue loop on all available slave lcores.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_stress.c | 27 ++
app/test/test_ring_stress.h | 477 ++++++++++++++++++++++++++++++++++++
4 files changed, 506 insertions(+)
create mode 100644 app/test/test_ring_stress.c
create mode 100644 app/test/test_ring_stress.h
diff --git a/app/test/Makefile b/app/test/Makefile
index 1f080d162..4f586d95f 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -78,6 +78,7 @@ SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
ifeq ($(CONFIG_RTE_LIBRTE_TABLE),y)
diff --git a/app/test/meson.build b/app/test/meson.build
index 0a2ce710f..84dde28ad 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -101,6 +101,7 @@ test_sources = files('commands.c',
'test_rib6.c',
'test_ring.c',
'test_ring_perf.c',
+ 'test_ring_stress.c',
'test_rwlock.c',
'test_sched.c',
'test_service_cores.c',
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
new file mode 100644
index 000000000..5689e06c8
--- /dev/null
+++ b/app/test/test_ring_stress.c
@@ -0,0 +1,27 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_mc_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_mp_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num, 0);
+}
+
+REGISTER_TEST_COMMAND(ring_stress_autotest, test_ring_stress);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
new file mode 100644
index 000000000..c6f0bc9f1
--- /dev/null
+++ b/app/test/test_ring_stress.h
@@ -0,0 +1,477 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+
+#include <inttypes.h>
+#include <stddef.h>
+#include <stdalign.h>
+#include <string.h>
+#include <stdio.h>
+#include <unistd.h>
+
+#include <rte_ring.h>
+#include <rte_cycles.h>
+#include <rte_launch.h>
+#include <rte_pause.h>
+#include <rte_random.h>
+#include <rte_malloc.h>
+
+#include "test.h"
+
+/*
+ * Measures performance of ring enqueue/dequeue under high contention
+ */
+
+#define RING_NAME "RING_STRESS"
+#define BULK_NUM 32
+#define RING_SIZE (2 * BULK_NUM * RTE_MAX_LCORE)
+
+enum {
+ WRK_CMD_STOP,
+ WRK_CMD_RUN,
+};
+
+static volatile uint32_t wrk_cmd __rte_cache_aligned;
+
+/* test run-time in seconds */
+static const uint32_t run_time = 60;
+
+struct lcore_stat {
+ uint64_t nb_cycle;
+ struct {
+ uint64_t nb_call;
+ uint64_t nb_obj;
+ uint64_t nb_cycle;
+ uint64_t max_cycle;
+ uint64_t min_cycle;
+ } op;
+};
+
+struct lcore_arg {
+ struct rte_ring *rng;
+ struct lcore_stat stats;
+} __rte_cache_aligned;
+
+struct ring_elem {
+ uint32_t cnt[RTE_CACHE_LINE_SIZE / sizeof(uint32_t)];
+} __rte_cache_aligned;
+
+/*
+ * redefinable functions
+ */
+static uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail);
+
+static uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free);
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num);
+
+
+static void
+lcore_stat_update(struct lcore_stat *ls, uint64_t call, uint64_t obj,
+ uint64_t tm, int32_t prcs)
+{
+ ls->op.nb_call += call;
+ ls->op.nb_obj += obj;
+ ls->op.nb_cycle += tm;
+ if (prcs) {
+ ls->op.max_cycle = RTE_MAX(ls->op.max_cycle, tm);
+ ls->op.min_cycle = RTE_MIN(ls->op.min_cycle, tm);
+ }
+}
+
+static void
+lcore_op_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls)
+{
+
+ ms->op.nb_call += ls->op.nb_call;
+ ms->op.nb_obj += ls->op.nb_obj;
+ ms->op.nb_cycle += ls->op.nb_cycle;
+ ms->op.max_cycle = RTE_MAX(ms->op.max_cycle, ls->op.max_cycle);
+ ms->op.min_cycle = RTE_MIN(ms->op.min_cycle, ls->op.min_cycle);
+}
+
+static void
+lcore_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls)
+{
+ ms->nb_cycle = RTE_MAX(ms->nb_cycle, ls->nb_cycle);
+ lcore_op_stat_aggr(ms, ls);
+}
+
+static void
+lcore_stat_dump(FILE *f, uint32_t lc, const struct lcore_stat *ls)
+{
+ long double st;
+
+ st = (long double)rte_get_timer_hz() / US_PER_S;
+
+ if (lc == UINT32_MAX)
+ fprintf(f, "%s(AGGREGATE)={\n", __func__);
+ else
+ fprintf(f, "%s(lc=%u)={\n", __func__, lc);
+
+ fprintf(f, "\tnb_cycle=%" PRIu64 "(%.2Lf usec),\n",
+ ls->nb_cycle, (long double)ls->nb_cycle / st);
+
+ fprintf(f, "\tDEQ+ENQ={\n");
+
+ fprintf(f, "\t\tnb_call=%" PRIu64 ",\n", ls->op.nb_call);
+ fprintf(f, "\t\tnb_obj=%" PRIu64 ",\n", ls->op.nb_obj);
+ fprintf(f, "\t\tnb_cycle=%" PRIu64 ",\n", ls->op.nb_cycle);
+ fprintf(f, "\t\tobj/call(avg): %.2Lf\n",
+ (long double)ls->op.nb_obj / ls->op.nb_call);
+ fprintf(f, "\t\tcycles/obj(avg): %.2Lf\n",
+ (long double)ls->op.nb_cycle / ls->op.nb_obj);
+ fprintf(f, "\t\tcycles/call(avg): %.2Lf\n",
+ (long double)ls->op.nb_cycle / ls->op.nb_call);
+
+ /* if min/max cycles per call stats was collected */
+ if (ls->op.min_cycle != UINT64_MAX) {
+ fprintf(f, "\t\tmax cycles/call=%" PRIu64 "(%.2Lf usec),\n",
+ ls->op.max_cycle,
+ (long double)ls->op.max_cycle / st);
+ fprintf(f, "\t\tmin cycles/call=%" PRIu64 "(%.2Lf usec),\n",
+ ls->op.min_cycle,
+ (long double)ls->op.min_cycle / st);
+ }
+
+ fprintf(f, "\t},\n");
+ fprintf(f, "};\n");
+}
+
+static void
+fill_ring_elm(struct ring_elem *elm, uint32_t fill)
+{
+ uint32_t i;
+
+ for (i = 0; i != RTE_DIM(elm->cnt); i++)
+ elm->cnt[i] = fill;
+}
+
+static int32_t
+check_updt_elem(struct ring_elem *elm[], uint32_t num,
+ const struct ring_elem *check, const struct ring_elem *fill)
+{
+ uint32_t i;
+
+ static rte_spinlock_t dump_lock;
+
+ for (i = 0; i != num; i++) {
+ if (memcmp(check, elm[i], sizeof(*check)) != 0) {
+ rte_spinlock_lock(&dump_lock);
+ printf("%s(lc=%u, num=%u) failed at %u-th iter, "
+ "offending object: %p\n",
+ __func__, rte_lcore_id(), num, i, elm[i]);
+ rte_memdump(stdout, "expected", check, sizeof(*check));
+ rte_memdump(stdout, "result", elm[i], sizeof(elm[i]));
+ rte_spinlock_unlock(&dump_lock);
+ return -EINVAL;
+ }
+ memcpy(elm[i], fill, sizeof(*elm[i]));
+ }
+
+ return 0;
+}
+
+static int
+check_ring_op(uint32_t exp, uint32_t res, uint32_t lc,
+ const char *fname, const char *opname)
+{
+ if (exp != res) {
+ printf("%s(lc=%u) failure: %s expected: %u, returned %u\n",
+ fname, lc, opname, exp, res);
+ return -ENOSPC;
+ }
+ return 0;
+}
+
+static int
+test_worker_prcs(void *arg)
+{
+ int32_t rc;
+ uint32_t lc, n, num;
+ uint64_t cl, tm0, tm1;
+ struct lcore_arg *la;
+ struct ring_elem def_elm, loc_elm;
+ struct ring_elem *obj[2 * BULK_NUM];
+
+ la = arg;
+ lc = rte_lcore_id();
+
+ fill_ring_elm(&def_elm, UINT32_MAX);
+ fill_ring_elm(&loc_elm, lc);
+
+ while (wrk_cmd != WRK_CMD_RUN) {
+ rte_smp_rmb();
+ rte_pause();
+ }
+
+ cl = rte_rdtsc_precise();
+
+ do {
+ /* num in interval [7/8, 11/8] of BULK_NUM */
+ num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
+
+ /* reset all pointer values */
+ memset(obj, 0, sizeof(obj));
+
+ /* dequeue num elems */
+ tm0 = rte_rdtsc_precise();
+ n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
+ tm0 = rte_rdtsc_precise() - tm0;
+
+ /* check return value and objects */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_dequeue_bulk));
+ if (rc == 0)
+ rc = check_updt_elem(obj, num, &def_elm, &loc_elm);
+ if (rc != 0)
+ break;
+
+ /* enqueue num elems */
+ rte_compiler_barrier();
+ rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
+ if (rc != 0)
+ break;
+
+ tm1 = rte_rdtsc_precise();
+ n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
+ tm1 = rte_rdtsc_precise() - tm1;
+
+ /* check return value */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_enqueue_bulk));
+ if (rc != 0)
+ break;
+
+ lcore_stat_update(&la->stats, 1, num, tm0 + tm1, 1);
+
+ } while (wrk_cmd == WRK_CMD_RUN);
+
+ la->stats.nb_cycle = rte_rdtsc_precise() - cl;
+ return rc;
+}
+
+static int
+test_worker_avg(void *arg)
+{
+ int32_t rc;
+ uint32_t lc, n, num;
+ uint64_t cl;
+ struct lcore_arg *la;
+ struct ring_elem def_elm, loc_elm;
+ struct ring_elem *obj[2 * BULK_NUM];
+
+ la = arg;
+ lc = rte_lcore_id();
+
+ fill_ring_elm(&def_elm, UINT32_MAX);
+ fill_ring_elm(&loc_elm, lc);
+
+ while (wrk_cmd != WRK_CMD_RUN) {
+ rte_smp_rmb();
+ rte_pause();
+ }
+
+ cl = rte_rdtsc_precise();
+
+ do {
+ /* num in interval [7/8, 11/8] of BULK_NUM */
+ num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
+
+ /* reset all pointer values */
+ memset(obj, 0, sizeof(obj));
+
+ /* dequeue num elems */
+ n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
+
+ /* check return value and objects */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_dequeue_bulk));
+ if (rc == 0)
+ rc = check_updt_elem(obj, num, &def_elm, &loc_elm);
+ if (rc != 0)
+ break;
+
+ /* enqueue num elems */
+ rte_compiler_barrier();
+ rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
+ if (rc != 0)
+ break;
+
+ n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
+
+ /* check return value */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_enqueue_bulk));
+ if (rc != 0)
+ break;
+
+ lcore_stat_update(&la->stats, 1, num, 0, 0);
+
+ } while (wrk_cmd == WRK_CMD_RUN);
+
+ /* final stats update */
+ cl = rte_rdtsc_precise() - cl;
+ lcore_stat_update(&la->stats, 0, 0, cl, 0);
+ la->stats.nb_cycle = cl;
+
+ return rc;
+}
+
+static void
+mt1_fini(struct rte_ring *rng, void *data)
+{
+ rte_free(rng);
+ rte_free(data);
+}
+
+static int
+mt1_init(struct rte_ring **rng, void **data, uint32_t num)
+{
+ int32_t rc;
+ size_t sz;
+ uint32_t i, nr;
+ struct rte_ring *r;
+ struct ring_elem *elm;
+ void *p;
+
+ *rng = NULL;
+ *data = NULL;
+
+ sz = num * sizeof(*elm);
+ elm = rte_zmalloc(NULL, sz, alignof(*elm));
+
+ /* alloc ring */
+ nr = 2 * num;
+ sz = rte_ring_get_memsize(nr);
+ r = rte_zmalloc(NULL, sz, alignof(*r));
+ if (r == NULL) {
+ printf("%s: alloca(%zu) for FIFO with %u elems failed",
+ __func__, sz, nr);
+ return -ENOMEM;
+ }
+ rc = _st_ring_init(r, RING_NAME, nr);
+ if (rc != 0) {
+ printf("%s: _st_ring_init(%p, %u) failed, error: %d(%s)\n",
+ __func__, r, nr, rc, strerror(-rc));
+ return rc;
+ }
+
+ for (i = 0; i != num; i++) {
+ fill_ring_elm(elm + i, UINT32_MAX);
+ p = elm + i;
+ if (_st_ring_enqueue_bulk(r, &p, 1, NULL) != 1)
+ break;
+ }
+
+ if (i != num) {
+ printf("%s: _st_ring_enqueue(%p, %u) returned %u\n",
+ __func__, r, num, i);
+ return -ENOSPC;
+ }
+
+ *rng = r;
+ *data = elm;
+ return 0;
+}
+
+static int
+test_mt1(int (*test)(void *))
+{
+ int32_t rc;
+ uint32_t lc, mc;
+ struct rte_ring *r;
+ void *data;
+ struct lcore_arg arg[RTE_MAX_LCORE];
+
+ static const struct lcore_stat init_stat = {
+ .op.min_cycle = UINT64_MAX,
+ };
+
+ rc = mt1_init(&r, &data, RING_SIZE);
+ if (rc != 0) {
+ mt1_fini(r, data);
+ return rc;
+ }
+
+ memset(arg, 0, sizeof(arg));
+
+ /* launch on all slaves */
+ RTE_LCORE_FOREACH_SLAVE(lc) {
+ arg[lc].rng = r;
+ arg[lc].stats = init_stat;
+ rte_eal_remote_launch(test, &arg[lc], lc);
+ }
+
+ /* signal worker to start test */
+ wrk_cmd = WRK_CMD_RUN;
+ rte_smp_wmb();
+
+ usleep(run_time * US_PER_S);
+
+ /* signal worker to start test */
+ wrk_cmd = WRK_CMD_STOP;
+ rte_smp_wmb();
+
+ /* wait for slaves and collect stats. */
+ mc = rte_lcore_id();
+ arg[mc].stats = init_stat;
+
+ rc = 0;
+ RTE_LCORE_FOREACH_SLAVE(lc) {
+ rc |= rte_eal_wait_lcore(lc);
+ lcore_stat_dump(stdout, lc, &arg[lc].stats);
+ lcore_stat_aggr(&arg[mc].stats, &arg[lc].stats);
+ }
+
+ lcore_stat_dump(stdout, UINT32_MAX, &arg[mc].stats);
+ mt1_fini(r, data);
+ return rc;
+}
+
+static int
+test_ring_stress(void)
+{
+ int32_t rc;
+ uint32_t i, k;
+
+ const struct {
+ const char *name;
+ int (*func)(int (*)(void *));
+ int (*wfunc)(void *arg);
+ } tests[] = {
+ {
+ .name = "MT-WRK_ENQ_DEQ-MST_NONE-PRCS",
+ .func = test_mt1,
+ .wfunc = test_worker_prcs,
+ },
+ {
+ .name = "MT-WRK_ENQ_DEQ-MST_NONE-AVG",
+ .func = test_mt1,
+ .wfunc = test_worker_avg,
+ },
+ };
+
+ for (i = 0, k = 0; i != RTE_DIM(tests); i++) {
+
+ printf("TEST %s START\n", tests[i].name);
+
+ rc = tests[i].func(tests[i].wfunc);
+ k += (rc == 0);
+
+ if (rc != 0)
+ printf("TEST-CASE %s FAILED\n", tests[i].name);
+ else
+ printf("TEST-CASE %s OK\n", tests[i].name);
+ }
+
+ printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
+ i, k, i - k);
+ return (k != i);
+}
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [RFC 2/6] ring: rework ring layout to allow new sync schemes
2020-02-24 11:35 [dpdk-dev] [RFC 0/6] New sync modes for ring Konstantin Ananyev
2020-02-24 11:35 ` [dpdk-dev] [RFC 1/6] test/ring: add contention stress test Konstantin Ananyev
@ 2020-02-24 11:35 ` Konstantin Ananyev
2020-02-24 11:35 ` [dpdk-dev] [RFC 3/6] ring: introduce RTS ring mode Konstantin Ananyev
` (6 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-02-24 11:35 UTC (permalink / raw)
To: dev; +Cc: olivier.matz, Konstantin Ananyev
Change from *single* to *sync_type* to allow different
synchronisation schemes to be applied.
Change layout to make sure that *sync_type* and *tail*
will always reside on same offsets.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/test_pdump.c | 6 +--
lib/librte_pdump/rte_pdump.c | 2 +-
lib/librte_port/rte_port_ring.c | 12 +++---
lib/librte_ring/rte_ring.c | 6 ++-
lib/librte_ring/rte_ring.h | 76 ++++++++++++++++++++++++---------
lib/librte_ring/rte_ring_elem.h | 8 ++--
6 files changed, 71 insertions(+), 39 deletions(-)
diff --git a/app/test/test_pdump.c b/app/test/test_pdump.c
index ad183184c..6a1180bcb 100644
--- a/app/test/test_pdump.c
+++ b/app/test/test_pdump.c
@@ -57,8 +57,7 @@ run_pdump_client_tests(void)
if (ret < 0)
return -1;
mp->flags = 0x0000;
- ring_client = rte_ring_create("SR0", RING_SIZE, rte_socket_id(),
- RING_F_SP_ENQ | RING_F_SC_DEQ);
+ ring_client = rte_ring_create("SR0", RING_SIZE, rte_socket_id(), 0);
if (ring_client == NULL) {
printf("rte_ring_create SR0 failed");
return -1;
@@ -71,9 +70,6 @@ run_pdump_client_tests(void)
}
rte_eth_dev_probing_finish(eth_dev);
- ring_client->prod.single = 0;
- ring_client->cons.single = 0;
-
printf("\n***** flags = RTE_PDUMP_FLAG_TX *****\n");
for (itr = 0; itr < NUM_ITR; itr++) {
diff --git a/lib/librte_pdump/rte_pdump.c b/lib/librte_pdump/rte_pdump.c
index 8a01ac510..65364f2c5 100644
--- a/lib/librte_pdump/rte_pdump.c
+++ b/lib/librte_pdump/rte_pdump.c
@@ -380,7 +380,7 @@ pdump_validate_ring_mp(struct rte_ring *ring, struct rte_mempool *mp)
rte_errno = EINVAL;
return -1;
}
- if (ring->prod.single || ring->cons.single) {
+ if (rte_ring_prod_single(ring) || rte_ring_cons_single(ring)) {
PDUMP_LOG(ERR, "ring with either SP or SC settings"
" is not valid for pdump, should have MP and MC settings\n");
rte_errno = EINVAL;
diff --git a/lib/librte_port/rte_port_ring.c b/lib/librte_port/rte_port_ring.c
index 47fcdd06a..2f6c050fa 100644
--- a/lib/librte_port/rte_port_ring.c
+++ b/lib/librte_port/rte_port_ring.c
@@ -44,8 +44,8 @@ rte_port_ring_reader_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->cons.single && is_multi) ||
- (!(conf->ring->cons.single) && !is_multi)) {
+ (rte_ring_cons_single(conf->ring) && is_multi) ||
+ (!rte_ring_cons_single(conf->ring) && !is_multi)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
}
@@ -171,8 +171,8 @@ rte_port_ring_writer_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->prod.single && is_multi) ||
- (!(conf->ring->prod.single) && !is_multi) ||
+ (rte_ring_prod_single(conf->ring) && is_multi) ||
+ (!rte_ring_prod_single(conf->ring) && !is_multi) ||
(conf->tx_burst_sz > RTE_PORT_IN_BURST_SIZE_MAX)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
@@ -440,8 +440,8 @@ rte_port_ring_writer_nodrop_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->prod.single && is_multi) ||
- (!(conf->ring->prod.single) && !is_multi) ||
+ (rte_ring_prod_single(conf->ring) && is_multi) ||
+ (!rte_ring_prod_single(conf->ring) && !is_multi) ||
(conf->tx_burst_sz > RTE_PORT_IN_BURST_SIZE_MAX)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index 77e5de099..fa5733907 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -106,8 +106,10 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
if (ret < 0 || ret >= (int)sizeof(r->name))
return -ENAMETOOLONG;
r->flags = flags;
- r->prod.single = (flags & RING_F_SP_ENQ) ? __IS_SP : __IS_MP;
- r->cons.single = (flags & RING_F_SC_DEQ) ? __IS_SC : __IS_MC;
+ r->prod.sync_type = (flags & RING_F_SP_ENQ) ?
+ RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
+ r->cons.sync_type = (flags & RING_F_SC_DEQ) ?
+ RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
if (flags & RING_F_EXACT_SZ) {
r->size = rte_align32pow2(count + 1);
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index 18fc5d845..aa8c628eb 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -61,11 +61,22 @@ enum rte_ring_queue_behavior {
#define RTE_RING_NAMESIZE (RTE_MEMZONE_NAMESIZE - \
sizeof(RTE_RING_MZ_PREFIX) + 1)
-/* structure to hold a pair of head/tail values and other metadata */
+/** prod/cons sync types */
+enum {
+ RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
+ RTE_RING_SYNC_ST, /**< single thread only */
+};
+
+/**
+ * structure to hold a pair of head/tail values and other metadata.
+ * used by RTE_RING_SYNC_MT, RTE_RING_SYNC_ST sync types.
+ * Depending on sync_type format of that structure might be different,
+ * but offset for *sync_type* and *tail* values should remain the same.
+ */
struct rte_ring_headtail {
- volatile uint32_t head; /**< Prod/consumer head. */
- volatile uint32_t tail; /**< Prod/consumer tail. */
- uint32_t single; /**< True if single prod/cons */
+ uint32_t sync_type; /**< sync type of prod/cons */
+ volatile uint32_t tail __rte_aligned(8); /**< prod/consumer tail. */
+ volatile uint32_t head; /**< prod/consumer head. */
};
/**
@@ -116,11 +127,10 @@ struct rte_ring {
#define RING_F_EXACT_SZ 0x0004
#define RTE_RING_SZ_MASK (0x7fffffffU) /**< Ring size mask */
-/* @internal defines for passing to the enqueue dequeue worker functions */
-#define __IS_SP 1
-#define __IS_MP 0
-#define __IS_SC 1
-#define __IS_MC 0
+#define __IS_SP RTE_RING_SYNC_ST
+#define __IS_MP RTE_RING_SYNC_MT
+#define __IS_SC RTE_RING_SYNC_ST
+#define __IS_MC RTE_RING_SYNC_MT
/**
* Calculate the memory size needed for a ring
@@ -420,7 +430,7 @@ rte_ring_mp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_MP, free_space);
+ RTE_RING_SYNC_MT, free_space);
}
/**
@@ -443,7 +453,7 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_SP, free_space);
+ RTE_RING_SYNC_ST, free_space);
}
/**
@@ -470,7 +480,7 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->prod.single, free_space);
+ r->prod.sync_type, free_space);
}
/**
@@ -554,7 +564,7 @@ rte_ring_mc_dequeue_bulk(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_MC, available);
+ RTE_RING_SYNC_MT, available);
}
/**
@@ -578,7 +588,7 @@ rte_ring_sc_dequeue_bulk(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_SC, available);
+ RTE_RING_SYNC_ST, available);
}
/**
@@ -605,7 +615,7 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
/**
@@ -777,6 +787,30 @@ rte_ring_get_capacity(const struct rte_ring *r)
return r->capacity;
}
+static inline uint32_t
+rte_ring_get_prod_sync_type(const struct rte_ring *r)
+{
+ return r->prod.sync_type;
+}
+
+static inline int
+rte_ring_prod_single(const struct rte_ring *r)
+{
+ return (rte_ring_get_prod_sync_type(r) == RTE_RING_SYNC_ST);
+}
+
+static inline uint32_t
+rte_ring_get_cons_sync_type(const struct rte_ring *r)
+{
+ return r->cons.sync_type;
+}
+
+static inline int
+rte_ring_cons_single(const struct rte_ring *r)
+{
+ return (rte_ring_get_cons_sync_type(r) == RTE_RING_SYNC_ST);
+}
+
/**
* Dump the status of all rings on the console
*
@@ -820,7 +854,7 @@ rte_ring_mp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_MP, free_space);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_MT, free_space);
}
/**
@@ -843,7 +877,7 @@ rte_ring_sp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_SP, free_space);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST, free_space);
}
/**
@@ -870,7 +904,7 @@ rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE,
- r->prod.single, free_space);
+ r->prod.sync_type, free_space);
}
/**
@@ -898,7 +932,7 @@ rte_ring_mc_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_MC, available);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_MT, available);
}
/**
@@ -923,7 +957,7 @@ rte_ring_sc_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_SC, available);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST, available);
}
/**
@@ -951,7 +985,7 @@ rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
{
return __rte_ring_do_dequeue(r, obj_table, n,
RTE_RING_QUEUE_VARIABLE,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 3976757ed..ff7a28ea5 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -570,7 +570,7 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->prod.single, free_space);
+ RTE_RING_QUEUE_FIXED, r->prod.sync_type, free_space);
}
/**
@@ -734,7 +734,7 @@ rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
unsigned int esize, unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->cons.single, available);
+ RTE_RING_QUEUE_FIXED, r->cons.sync_type, available);
}
/**
@@ -902,7 +902,7 @@ rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_VARIABLE, r->prod.single, free_space);
+ RTE_RING_QUEUE_VARIABLE, r->prod.sync_type, free_space);
}
/**
@@ -995,7 +995,7 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
{
return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
RTE_RING_QUEUE_VARIABLE,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
#ifdef __cplusplus
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [RFC 3/6] ring: introduce RTS ring mode
2020-02-24 11:35 [dpdk-dev] [RFC 0/6] New sync modes for ring Konstantin Ananyev
2020-02-24 11:35 ` [dpdk-dev] [RFC 1/6] test/ring: add contention stress test Konstantin Ananyev
2020-02-24 11:35 ` [dpdk-dev] [RFC 2/6] ring: rework ring layout to allow new sync schemes Konstantin Ananyev
@ 2020-02-24 11:35 ` Konstantin Ananyev
2020-02-24 11:35 ` [dpdk-dev] [RFC 4/6] test/ring: add contention stress test for RTS ring Konstantin Ananyev
` (5 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-02-24 11:35 UTC (permalink / raw)
To: dev; +Cc: olivier.matz, Konstantin Ananyev
Introduce relaxed tail sync (RTS) mode for MT ring synchronization.
Aim to reduce stall times in case when ring is used on
overcommited cpus (multiple active threads on the same cpu).
The main difference from original MP/MC algorithm is that
tail value is increased not by every thread that finished enqueue/dequeue,
but only by the last one.
That allows threads to avoid spinning on ring tail value,
leaving actual tail value change to the last thread in the update queue.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 3 +-
lib/librte_ring/meson.build | 3 +-
lib/librte_ring/rte_ring.c | 75 ++++++-
lib/librte_ring/rte_ring.h | 300 +++++++++++++++++++++++--
lib/librte_ring/rte_ring_rts_generic.h | 240 ++++++++++++++++++++
5 files changed, 598 insertions(+), 23 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 917c560ad..4f90344f4 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -18,6 +18,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_RING) := rte_ring.c
SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_elem.h \
rte_ring_generic.h \
- rte_ring_c11_mem.h
+ rte_ring_c11_mem.h \
+ rte_ring_rts_generic.h
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index f2f3ccc88..dc8d7dbea 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -5,7 +5,8 @@ sources = files('rte_ring.c')
headers = files('rte_ring.h',
'rte_ring_elem.h',
'rte_ring_c11_mem.h',
- 'rte_ring_generic.h')
+ 'rte_ring_generic.h',
+ 'rte_ring_rts_generic.h')
# rte_ring_create_elem and rte_ring_get_memsize_elem are experimental
allow_experimental_apis = true
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index fa5733907..1ce0af3e5 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -45,6 +45,9 @@ EAL_REGISTER_TAILQ(rte_ring_tailq)
/* true if x is a power of 2 */
#define POWEROF2(x) ((((x)-1) & (x)) == 0)
+/* by default set head/tail distance as 1/8 of ring capacity */
+#define HTD_MAX_DEF 8
+
/* return the size of memory occupied by a ring */
ssize_t
rte_ring_get_memsize_elem(unsigned int esize, unsigned int count)
@@ -82,8 +85,56 @@ rte_ring_get_memsize(unsigned int count)
void
rte_ring_reset(struct rte_ring *r)
{
- r->prod.head = r->cons.head = 0;
- r->prod.tail = r->cons.tail = 0;
+ memset((void *)(uintptr_t)&r->prod.tail, 0,
+ offsetof(struct rte_ring, pad1) -
+ offsetof(struct rte_ring, prod.tail));
+ memset((void *)(uintptr_t)&r->cons.tail, 0,
+ offsetof(struct rte_ring, pad2) -
+ offsetof(struct rte_ring, cons.tail));
+}
+
+/*
+ * helper function, calculates sync_type values for prod and cons
+ * based on input flags. Returns zero at success or negative
+ * errno value otherwise.
+ */
+static int
+get_sync_type(uint32_t flags, uint32_t *prod_st, uint32_t *cons_st)
+{
+ static const uint32_t prod_st_flags =
+ (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
+ static const uint32_t cons_st_flags =
+ (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
+
+ switch (flags & prod_st_flags) {
+ case 0:
+ *prod_st = RTE_RING_SYNC_MT;
+ break;
+ case RING_F_SP_ENQ:
+ *prod_st = RTE_RING_SYNC_ST;
+ break;
+ case RING_F_MP_RTS_ENQ:
+ *prod_st = RTE_RING_SYNC_MT_RTS;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ switch (flags & cons_st_flags) {
+ case 0:
+ *cons_st = RTE_RING_SYNC_MT;
+ break;
+ case RING_F_SC_DEQ:
+ *cons_st = RTE_RING_SYNC_ST;
+ break;
+ case RING_F_MC_RTS_DEQ:
+ *cons_st = RTE_RING_SYNC_MT_RTS;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
}
int
@@ -100,16 +151,20 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
RTE_CACHE_LINE_MASK) != 0);
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
+ offsetof(struct rte_ring_rts_headtail, sync_type));
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
+ offsetof(struct rte_ring_rts_headtail, tail.val.pos));
+
/* init the ring structure */
memset(r, 0, sizeof(*r));
ret = strlcpy(r->name, name, sizeof(r->name));
if (ret < 0 || ret >= (int)sizeof(r->name))
return -ENAMETOOLONG;
r->flags = flags;
- r->prod.sync_type = (flags & RING_F_SP_ENQ) ?
- RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
- r->cons.sync_type = (flags & RING_F_SC_DEQ) ?
- RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
+ ret = get_sync_type(flags, &r->prod.sync_type, &r->cons.sync_type);
+ if (ret != 0)
+ return ret;
if (flags & RING_F_EXACT_SZ) {
r->size = rte_align32pow2(count + 1);
@@ -126,8 +181,12 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
r->mask = count - 1;
r->capacity = r->mask;
}
- r->prod.head = r->cons.head = 0;
- r->prod.tail = r->cons.tail = 0;
+
+ /* set default values for head-tail distance */
+ if (flags & RING_F_MP_RTS_ENQ)
+ rte_ring_set_prod_htd_max(r, r->capacity / HTD_MAX_DEF);
+ if (flags & RING_F_MC_RTS_DEQ)
+ rte_ring_set_cons_htd_max(r, r->capacity / HTD_MAX_DEF);
return 0;
}
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index aa8c628eb..a130aeb9d 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -65,13 +65,15 @@ enum rte_ring_queue_behavior {
enum {
RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
RTE_RING_SYNC_ST, /**< single thread only */
+ RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
};
/**
* structure to hold a pair of head/tail values and other metadata.
* used by RTE_RING_SYNC_MT, RTE_RING_SYNC_ST sync types.
- * Depending on sync_type format of that structure might be different,
- * but offset for *sync_type* and *tail* values should remain the same.
+ * Depending on sync_type format of that structure might differ
+ * depending on the sync mechanism selelcted, but offsets for
+ * *sync_type* and *tail* values should always remain the same.
*/
struct rte_ring_headtail {
uint32_t sync_type; /**< sync type of prod/cons */
@@ -79,6 +81,21 @@ struct rte_ring_headtail {
volatile uint32_t head; /**< prod/consumer head. */
};
+union rte_ring_ht_poscnt {
+ uint64_t raw;
+ struct {
+ uint32_t pos; /**< head/tail position */
+ uint32_t cnt; /**< head/tail reference counter */
+ } val;
+};
+
+struct rte_ring_rts_headtail {
+ uint32_t sync_type; /**< sync type of prod/cons */
+ uint32_t htd_max; /**< max allowed distance between head/tail */
+ volatile union rte_ring_ht_poscnt tail;
+ volatile union rte_ring_ht_poscnt head;
+};
+
/**
* An RTE ring structure.
*
@@ -106,11 +123,21 @@ struct rte_ring {
char pad0 __rte_cache_aligned; /**< empty cache line */
/** Ring producer status. */
- struct rte_ring_headtail prod __rte_cache_aligned;
+ RTE_STD_C11
+ union {
+ struct rte_ring_headtail prod;
+ struct rte_ring_rts_headtail rts_prod;
+ } __rte_cache_aligned;
+
char pad1 __rte_cache_aligned; /**< empty cache line */
/** Ring consumer status. */
- struct rte_ring_headtail cons __rte_cache_aligned;
+ RTE_STD_C11
+ union {
+ struct rte_ring_headtail cons;
+ struct rte_ring_rts_headtail rts_cons;
+ } __rte_cache_aligned;
+
char pad2 __rte_cache_aligned; /**< empty cache line */
};
@@ -127,6 +154,9 @@ struct rte_ring {
#define RING_F_EXACT_SZ 0x0004
#define RTE_RING_SZ_MASK (0x7fffffffU) /**< Ring size mask */
+#define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS". */
+#define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC RTS". */
+
#define __IS_SP RTE_RING_SYNC_ST
#define __IS_MP RTE_RING_SYNC_MT
#define __IS_SC RTE_RING_SYNC_ST
@@ -407,6 +437,82 @@ __rte_ring_do_dequeue(struct rte_ring *r, void **obj_table,
return n;
}
+#include <rte_ring_rts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_enqueue(struct rte_ring *r, void * const *obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_rts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_rts_update_tail(&r->rts_prod);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_dequeue(struct rte_ring *r, void **obj_table,
+ unsigned int n, enum rte_ring_queue_behavior behavior,
+ unsigned int *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_rts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_rts_update_tail(&r->rts_cons);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
/**
* Enqueue several objects on the ring (multi-producers safe).
*
@@ -456,6 +562,29 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
RTE_RING_SYNC_ST, free_space);
}
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+static __rte_always_inline unsigned int
+rte_ring_rts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ free_space);
+}
+
/**
* Enqueue several objects on a ring.
*
@@ -479,8 +608,18 @@ static __rte_always_inline unsigned int
rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_bulk(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_bulk(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_rts_enqueue_bulk(r, obj_table, n, free_space);
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -591,6 +730,29 @@ rte_ring_sc_dequeue_bulk(struct rte_ring *r, void **obj_table,
RTE_RING_SYNC_ST, available);
}
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+static __rte_always_inline unsigned int
+rte_ring_rts_dequeue_bulk(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ available);
+}
+
/**
* Dequeue several objects from a ring.
*
@@ -614,8 +776,18 @@ static __rte_always_inline unsigned int
rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
unsigned int *available)
{
- return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_bulk(r, obj_table, n, available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_bulk(r, obj_table, n, available);
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_rts_dequeue_bulk(r, obj_table, n, available);
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -811,6 +983,42 @@ rte_ring_cons_single(const struct rte_ring *r)
return (rte_ring_get_cons_sync_type(r) == RTE_RING_SYNC_ST);
}
+static inline uint32_t
+rte_ring_get_prod_htd_max(const struct rte_ring *r)
+{
+ if (r->prod.sync_type == RTE_RING_SYNC_MT_RTS)
+ return r->rts_prod.htd_max;
+ return UINT32_MAX;
+}
+
+static inline int
+rte_ring_set_prod_htd_max(struct rte_ring *r, uint32_t v)
+{
+ if (r->prod.sync_type != RTE_RING_SYNC_MT_RTS)
+ return -ENOTSUP;
+
+ r->rts_prod.htd_max = v;
+ return 0;
+}
+
+static inline uint32_t
+rte_ring_get_cons_htd_max(const struct rte_ring *r)
+{
+ if (r->cons.sync_type == RTE_RING_SYNC_MT_RTS)
+ return r->rts_prod.htd_max;
+ return UINT32_MAX;
+}
+
+static inline int
+rte_ring_set_cons_htd_max(struct rte_ring *r, uint32_t v)
+{
+ if (r->cons.sync_type != RTE_RING_SYNC_MT_RTS)
+ return -ENOTSUP;
+
+ r->rts_cons.htd_max = v;
+ return 0;
+}
+
/**
* Dump the status of all rings on the console
*
@@ -880,6 +1088,29 @@ rte_ring_sp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST, free_space);
}
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+static __rte_always_inline unsigned
+rte_ring_rts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
/**
* Enqueue several objects on a ring.
*
@@ -903,8 +1134,18 @@ static __rte_always_inline unsigned
rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE,
- r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_burst(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_burst(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_rts_enqueue_burst(r, obj_table, n, free_space);
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -960,6 +1201,30 @@ rte_ring_sc_dequeue_burst(struct rte_ring *r, void **obj_table,
RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST, available);
}
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+static __rte_always_inline unsigned
+rte_ring_rts_dequeue_burst(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
/**
* Dequeue multiple objects from a ring up to a maximum number.
*
@@ -983,9 +1248,18 @@ static __rte_always_inline unsigned
rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
- return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_burst(r, obj_table, n, available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_burst(r, obj_table, n, available);
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_rts_dequeue_burst(r, obj_table, n, available);
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_rts_generic.h b/lib/librte_ring/rte_ring_rts_generic.h
new file mode 100644
index 000000000..add8630b2
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts_generic.h
@@ -0,0 +1,240 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_GENERIC_H_
+#define _RTE_RING_RTS_GENERIC_H_
+
+/**
+ * @file rte_ring_rts_generic.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for Relaxed Tail Sync (RTS) ring mode.
+ * The main idea remains the same as for our original MP/MC synchronization
+ * mechanism.
+ * The main difference is that tail value is increased not
+ * by every thread that finished enqueue/dequeue,
+ * but only by the last one doing enqueue/dequeue.
+ * That allows threads to skip spinning on tail value,
+ * leaving actual tail value change to last thread in the update queue.
+ * RTS requires 2 64-bit CAS for each enqueue(/dequeue) operation:
+ * one for head update, second for tail update.
+ * As a gain it allows thread to avoid spinning/waiting on tail value.
+ * In comparision original MP/MC algorithm requires one 32-bit CAS
+ * for head update and waiting/spinning on tail value.
+ *
+ * Brief outline:
+ * - introduce refcnt for both head and tail.
+ * - increment head.refcnt for each head.value update
+ * - write head:value and head:refcnt atomically (64-bit CAS)
+ * - move tail.value ahead only when tail.refcnt + 1 == head.refcnt
+ * - increment tail.refcnt when each enqueue/dequeue op finishes
+ * (no matter is tail:value going to change or not)
+ * - write tail.value and tail.recnt atomically (64-bit CAS)
+ *
+ * To avoid head/tail starvation:
+ * - limit max allowed distance between head and tail value (HTD_MAX).
+ * I.E. thread is allowed to proceed with changing head.value,
+ * only when: head.value - tail.value <= HTD_MAX
+ * HTD_MAX is an optional parameter.
+ * With HTD_MAX == 0 we'll have ring with fully synchronized
+ * head/tail (like HTS).
+ * With HTD_MAX == UINT32_MAX - no limitation.
+ * By default HTD_MAX == ring.capacity / 8.
+ */
+
+/**
+ * @internal This function updates tail values.
+ */
+static __rte_always_inline void
+__rte_ring_rts_update_tail(struct rte_ring_rts_headtail *ht)
+{
+ union rte_ring_ht_poscnt h, ot, nt;
+
+ /*
+ * If there are other enqueues/dequeues in progress that
+ * might preceded us, then don't update tail with new value.
+ */
+
+ do {
+ ot.raw = ht->tail.raw;
+ rte_smp_rmb();
+
+ /* on 32-bit systems we have to do atomic read here */
+ h.raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->head.raw);
+
+ nt.raw = ot.raw;
+ if (++nt.val.cnt == h.val.cnt)
+ nt.val.pos = h.val.pos;
+
+ } while (rte_atomic64_cmpset(&ht->tail.raw, ot.raw, nt.raw) == 0);
+}
+
+/**
+ * @internal This function waits till head/tail distance wouldn't
+ * exceed pre-defined max value.
+ */
+static __rte_always_inline void
+__rte_ring_rts_head_wait(const struct rte_ring_rts_headtail *ht,
+ union rte_ring_ht_poscnt *h)
+{
+ uint32_t max;
+
+ max = ht->htd_max;
+ h->raw = ht->head.raw;
+ rte_smp_rmb();
+
+ while (h->val.pos - ht->tail.val.pos > max) {
+ rte_pause();
+ h->raw = ht->head.raw;
+ rte_smp_rmb();
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue.
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_rts_move_prod_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* read prod head (may spin on prod tail) */
+ __rte_ring_rts_head_wait(&r->rts_prod, &oh);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - oh.val.pos;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ return 0;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (rte_atomic64_cmpset(&r->rts_prod.head.raw,
+ oh.raw, nh.raw) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_rts_move_cons_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* read cons head (may spin on cons tail) */
+ __rte_ring_rts_head_wait(&r->rts_cons, &oh);
+
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - oh.val.pos;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ return 0;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (rte_atomic64_cmpset(&r->rts_cons.head.raw,
+ oh.raw, nh.raw) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+#endif /* _RTE_RING_RTS_GENERIC_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [RFC 4/6] test/ring: add contention stress test for RTS ring
2020-02-24 11:35 [dpdk-dev] [RFC 0/6] New sync modes for ring Konstantin Ananyev
` (2 preceding siblings ...)
2020-02-24 11:35 ` [dpdk-dev] [RFC 3/6] ring: introduce RTS ring mode Konstantin Ananyev
@ 2020-02-24 11:35 ` Konstantin Ananyev
2020-02-24 11:35 ` [dpdk-dev] [RFC 5/6] ring: introduce HTS ring mode Konstantin Ananyev
` (4 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-02-24 11:35 UTC (permalink / raw)
To: dev; +Cc: olivier.matz, Konstantin Ananyev
Introduce new test case to test RTS ring mode under contention.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_rts_stress.c | 28 ++++++++++++++++++++++++++++
3 files changed, 30 insertions(+)
create mode 100644 app/test/test_ring_rts_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index 4f586d95f..d22b9f702 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -78,6 +78,7 @@ SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
diff --git a/app/test/meson.build b/app/test/meson.build
index 84dde28ad..fa4fb4b51 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -101,6 +101,7 @@ test_sources = files('commands.c',
'test_rib6.c',
'test_ring.c',
'test_ring_perf.c',
+ 'test_ring_rts_stress.c',
'test_ring_stress.c',
'test_rwlock.c',
'test_sched.c',
diff --git a/app/test/test_ring_rts_stress.c b/app/test/test_ring_rts_stress.c
new file mode 100644
index 000000000..525d222b2
--- /dev/null
+++ b/app/test/test_ring_rts_stress.c
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_rts_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_rts_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_RTS_ENQ | RING_F_MC_RTS_DEQ);
+}
+
+REGISTER_TEST_COMMAND(ring_stress_rts_autotest, test_ring_stress);
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [RFC 5/6] ring: introduce HTS ring mode
2020-02-24 11:35 [dpdk-dev] [RFC 0/6] New sync modes for ring Konstantin Ananyev
` (3 preceding siblings ...)
2020-02-24 11:35 ` [dpdk-dev] [RFC 4/6] test/ring: add contention stress test for RTS ring Konstantin Ananyev
@ 2020-02-24 11:35 ` Konstantin Ananyev
2020-03-25 20:44 ` Honnappa Nagarahalli
2020-02-24 11:35 ` [dpdk-dev] [RFC 6/6] test/ring: add contention stress test for HTS ring Konstantin Ananyev
` (3 subsequent siblings)
8 siblings, 1 reply; 146+ messages in thread
From: Konstantin Ananyev @ 2020-02-24 11:35 UTC (permalink / raw)
To: dev; +Cc: olivier.matz, Konstantin Ananyev
Introduce head/tail sync mode for MT ring synchronization.
In that mode enqueue/dequeue operation is fully serialized:
only one thread at a time is allowed to perform given op.
Suppose to reduce stall times in case when ring is used on
overcommitted cpus (multiple active threads on the same cpu).
As another enhancement provide ability to split enqueue/dequeue
operation into two phases:
- enqueue/dequeue start
- enqueue/dequeue finish
That allows user to inspect objects in the ring without removing
them from it (aka MT safe peek).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 1 +
lib/librte_ring/meson.build | 1 +
lib/librte_ring/rte_ring.c | 15 +-
lib/librte_ring/rte_ring.h | 259 ++++++++++++++++++++++++-
lib/librte_ring/rte_ring_hts_generic.h | 228 ++++++++++++++++++++++
5 files changed, 500 insertions(+), 4 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_hts_generic.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 4f90344f4..0c7f8f918 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -19,6 +19,7 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_elem.h \
rte_ring_generic.h \
rte_ring_c11_mem.h \
+ rte_ring_hts_generic.h \
rte_ring_rts_generic.h
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index dc8d7dbea..5aa673199 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -6,6 +6,7 @@ headers = files('rte_ring.h',
'rte_ring_elem.h',
'rte_ring_c11_mem.h',
'rte_ring_generic.h',
+ 'rte_ring_hts_generic.h',
'rte_ring_rts_generic.h')
# rte_ring_create_elem and rte_ring_get_memsize_elem are experimental
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index 1ce0af3e5..d3b948667 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -102,9 +102,9 @@ static int
get_sync_type(uint32_t flags, uint32_t *prod_st, uint32_t *cons_st)
{
static const uint32_t prod_st_flags =
- (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
+ (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ | RING_F_MP_HTS_ENQ);
static const uint32_t cons_st_flags =
- (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
+ (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ | RING_F_MC_HTS_DEQ);
switch (flags & prod_st_flags) {
case 0:
@@ -116,6 +116,9 @@ get_sync_type(uint32_t flags, uint32_t *prod_st, uint32_t *cons_st)
case RING_F_MP_RTS_ENQ:
*prod_st = RTE_RING_SYNC_MT_RTS;
break;
+ case RING_F_MP_HTS_ENQ:
+ *prod_st = RTE_RING_SYNC_MT_HTS;
+ break;
default:
return -EINVAL;
}
@@ -130,6 +133,9 @@ get_sync_type(uint32_t flags, uint32_t *prod_st, uint32_t *cons_st)
case RING_F_MC_RTS_DEQ:
*cons_st = RTE_RING_SYNC_MT_RTS;
break;
+ case RING_F_MC_HTS_DEQ:
+ *cons_st = RTE_RING_SYNC_MT_HTS;
+ break;
default:
return -EINVAL;
}
@@ -151,6 +157,11 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
RTE_CACHE_LINE_MASK) != 0);
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
+ offsetof(struct rte_ring_hts_headtail, sync_type));
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
+ offsetof(struct rte_ring_hts_headtail, ht.pos.tail));
+
RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
offsetof(struct rte_ring_rts_headtail, sync_type));
RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index a130aeb9d..52edcea11 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -66,11 +66,11 @@ enum {
RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
RTE_RING_SYNC_ST, /**< single thread only */
RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
+ RTE_RING_SYNC_MT_HTS, /**< multi-thread head/tail sync */
};
/**
- * structure to hold a pair of head/tail values and other metadata.
- * used by RTE_RING_SYNC_MT, RTE_RING_SYNC_ST sync types.
+ * Structure to hold a pair of head/tail values and other metadata.
* Depending on sync_type format of that structure might differ
* depending on the sync mechanism selelcted, but offsets for
* *sync_type* and *tail* values should always remain the same.
@@ -96,6 +96,19 @@ struct rte_ring_rts_headtail {
volatile union rte_ring_ht_poscnt head;
};
+union rte_ring_ht_pos {
+ uint64_t raw;
+ struct {
+ uint32_t tail; /**< tail position */
+ uint32_t head; /**< head position */
+ } pos;
+};
+
+struct rte_ring_hts_headtail {
+ uint32_t sync_type; /**< sync type of prod/cons */
+ volatile union rte_ring_ht_pos ht __rte_aligned(8);
+};
+
/**
* An RTE ring structure.
*
@@ -126,6 +139,7 @@ struct rte_ring {
RTE_STD_C11
union {
struct rte_ring_headtail prod;
+ struct rte_ring_hts_headtail hts_prod;
struct rte_ring_rts_headtail rts_prod;
} __rte_cache_aligned;
@@ -135,6 +149,7 @@ struct rte_ring {
RTE_STD_C11
union {
struct rte_ring_headtail cons;
+ struct rte_ring_hts_headtail hts_cons;
struct rte_ring_rts_headtail rts_cons;
} __rte_cache_aligned;
@@ -157,6 +172,9 @@ struct rte_ring {
#define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS". */
#define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC RTS". */
+#define RING_F_MP_HTS_ENQ 0x0020 /**< The default enqueue is "MP HTS". */
+#define RING_F_MC_HTS_DEQ 0x0040 /**< The default dequeue is "MC HTS". */
+
#define __IS_SP RTE_RING_SYNC_ST
#define __IS_MP RTE_RING_SYNC_MT
#define __IS_SC RTE_RING_SYNC_ST
@@ -513,6 +531,82 @@ __rte_ring_do_rts_dequeue(struct rte_ring *r, void **obj_table,
return n;
}
+#include <rte_ring_hts_generic.h>
+
+/**
+ * @internal Start to enqueue several objects on the HTS ring.
+ * Note that user has to call appropriate enqueue_finish()
+ * to complete given enqueue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_enqueue_start(struct rte_ring *r, void * const *obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_hts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0)
+ ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Start to dequeue several objects from the HTS ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_dequeue_start(struct rte_ring *r, void **obj_table,
+ unsigned int n, enum rte_ring_queue_behavior behavior,
+ unsigned int *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_hts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0)
+ DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
/**
* Enqueue several objects on the ring (multi-producers safe).
*
@@ -585,6 +679,47 @@ rte_ring_rts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
free_space);
}
+/**
+ * Start to enqueue several objects on the HTS ring (multi-producers safe).
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+static __rte_always_inline unsigned int
+rte_ring_hts_enqueue_bulk_start(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue_start(r, obj_table, n,
+ RTE_RING_QUEUE_FIXED, free_space);
+}
+
+static __rte_always_inline void
+rte_ring_hts_enqueue_finish(struct rte_ring *r, unsigned int n)
+{
+ __rte_ring_hts_update_tail(&r->hts_prod, n, 1);
+}
+
+static __rte_always_inline unsigned int
+rte_ring_hts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ n = rte_ring_hts_enqueue_bulk_start(r, obj_table, n, free_space);
+ if (n != 0)
+ rte_ring_hts_enqueue_finish(r, n);
+ return n;
+}
+
/**
* Enqueue several objects on a ring.
*
@@ -615,6 +750,8 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
return rte_ring_sp_enqueue_bulk(r, obj_table, n, free_space);
case RTE_RING_SYNC_MT_RTS:
return rte_ring_rts_enqueue_bulk(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_hts_enqueue_bulk(r, obj_table, n, free_space);
}
/* valid ring should never reach this point */
@@ -753,6 +890,47 @@ rte_ring_rts_dequeue_bulk(struct rte_ring *r, void **obj_table,
available);
}
+/**
+ * Start to dequeue several objects from an HTS ring (multi-consumers safe).
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+static __rte_always_inline unsigned int
+rte_ring_hts_dequeue_bulk_start(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue_start(r, obj_table, n,
+ RTE_RING_QUEUE_FIXED, available);
+}
+
+static __rte_always_inline void
+rte_ring_hts_dequeue_finish(struct rte_ring *r, unsigned int n)
+{
+ __rte_ring_hts_update_tail(&r->hts_cons, n, 0);
+}
+
+static __rte_always_inline unsigned int
+rte_ring_hts_dequeue_bulk(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ n = rte_ring_hts_dequeue_bulk_start(r, obj_table, n, available);
+ if (n != 0)
+ rte_ring_hts_dequeue_finish(r, n);
+ return n;
+}
+
/**
* Dequeue several objects from a ring.
*
@@ -783,6 +961,8 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
return rte_ring_sc_dequeue_bulk(r, obj_table, n, available);
case RTE_RING_SYNC_MT_RTS:
return rte_ring_rts_dequeue_bulk(r, obj_table, n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_hts_dequeue_bulk(r, obj_table, n, available);
}
/* valid ring should never reach this point */
@@ -1111,6 +1291,41 @@ rte_ring_rts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
RTE_RING_QUEUE_VARIABLE, free_space);
}
+/**
+ * Start to enqueue several objects on the HTS ring (multi-producers safe).
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+static __rte_always_inline unsigned int
+rte_ring_hts_enqueue_burst_start(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue_start(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+static __rte_always_inline unsigned int
+rte_ring_hts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ n = rte_ring_hts_enqueue_burst_start(r, obj_table, n, free_space);
+ if (n != 0)
+ rte_ring_hts_enqueue_finish(r, n);
+ return n;
+}
+
/**
* Enqueue several objects on a ring.
*
@@ -1141,6 +1356,8 @@ rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
return rte_ring_sp_enqueue_burst(r, obj_table, n, free_space);
case RTE_RING_SYNC_MT_RTS:
return rte_ring_rts_enqueue_burst(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_hts_enqueue_burst(r, obj_table, n, free_space);
}
/* valid ring should never reach this point */
@@ -1225,6 +1442,42 @@ rte_ring_rts_dequeue_burst(struct rte_ring *r, void **obj_table,
return __rte_ring_do_rts_dequeue(r, obj_table, n,
RTE_RING_QUEUE_VARIABLE, available);
}
+
+/**
+ * Start to dequeue several objects from an HTS ring (multi-consumers safe).
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+static __rte_always_inline unsigned int
+rte_ring_hts_dequeue_burst_start(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue_start(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+static __rte_always_inline unsigned int
+rte_ring_hts_dequeue_burst(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ n = rte_ring_hts_dequeue_burst_start(r, obj_table, n, available);
+ if (n != 0)
+ rte_ring_hts_dequeue_finish(r, n);
+ return n;
+}
+
/**
* Dequeue multiple objects from a ring up to a maximum number.
*
@@ -1255,6 +1508,8 @@ rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
return rte_ring_sc_dequeue_burst(r, obj_table, n, available);
case RTE_RING_SYNC_MT_RTS:
return rte_ring_rts_dequeue_burst(r, obj_table, n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_hts_dequeue_burst(r, obj_table, n, available);
}
/* valid ring should never reach this point */
diff --git a/lib/librte_ring/rte_ring_hts_generic.h b/lib/librte_ring/rte_ring_hts_generic.h
new file mode 100644
index 000000000..7e447e30b
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts_generic.h
@@ -0,0 +1,228 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2020 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_GENERIC_H_
+#define _RTE_RING_HTS_GENERIC_H_
+
+/**
+ * @file rte_ring_hts_generic.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for head/tail sync (HTS) ring mode.
+ * In that mode enqueue/dequeue operation is fully serialized:
+ * only one thread at a time is allowed to perform given op.
+ * This is achieved by thread is allowed to proceed with changing head.value
+ * only when head.value == tail.value.
+ * Both head and tail values are updated atomically (as one 64-bit value).
+ * As another enhancement that provides ability to split enqueue/dequeue
+ * operation into two phases:
+ * - enqueue/dequeue start
+ * - enqueue/dequeue finish
+ * That allows user to inspect objects in the ring without removing
+ * them from it (aka MT safe peek).
+ * As an example:
+ * // read 1 elem from the ring:
+ * n = rte_ring_hts_dequeue_bulk_start(ring, &obj, 1, NULL);
+ * if (n != 0) {
+ * //examined object
+ * if (object_examine(obj) == KEEP)
+ * //decided to keep it in the ring.
+ * rte_ring_hts_dequeue_finish(ring, 0);
+ * else
+ * //decided to remove it in the ring.
+ * rte_ring_hts_dequeue_finish(ring, n);
+ * }
+ * Note that between _start_ and _finish_ the ring is sort of locked -
+ * none other thread can proceed with enqueue(/dequeue) operation till
+ * _finish_ will complete.
+ */
+
+static __rte_always_inline void
+__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
+ uint32_t enqueue)
+{
+ uint32_t n;
+ union rte_ring_ht_pos p;
+
+ if (enqueue)
+ rte_smp_wmb();
+ else
+ rte_smp_rmb();
+
+ p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
+
+ n = p.pos.head - p.pos.tail;
+ RTE_ASSERT(n >= num);
+ RTE_SET_USED(n);
+
+ p.pos.head = p.pos.tail + num;
+ p.pos.tail = p.pos.head;
+
+ rte_atomic64_set((rte_atomic64_t *)(uintptr_t)&ht->ht.raw, p.raw);
+}
+
+/**
+ * @internal waits till tail will become equal to head.
+ * Means no writer/reader is active for that ring.
+ * Suppose to work as serialization point.
+ */
+static __rte_always_inline void
+__rte_ring_hts_head_wait(const struct rte_ring_hts_headtail *ht,
+ union rte_ring_ht_pos *p)
+{
+ p->raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->ht.raw);
+
+ while (p->pos.head != p->pos.tail) {
+ rte_pause();
+ p->raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->ht.raw);
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_prod_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_prod, &op);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - op.pos.head;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ return 0;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (rte_atomic64_cmpset(&r->hts_prod.ht.raw,
+ op.raw, np.raw) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_cons_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_cons, &op);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - op.pos.head;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ return 0;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (rte_atomic64_cmpset(&r->hts_cons.ht.raw,
+ op.raw, np.raw) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+#endif /* _RTE_RING_HTS_GENERIC_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [RFC 6/6] test/ring: add contention stress test for HTS ring
2020-02-24 11:35 [dpdk-dev] [RFC 0/6] New sync modes for ring Konstantin Ananyev
` (4 preceding siblings ...)
2020-02-24 11:35 ` [dpdk-dev] [RFC 5/6] ring: introduce HTS ring mode Konstantin Ananyev
@ 2020-02-24 11:35 ` Konstantin Ananyev
2020-02-24 16:59 ` [dpdk-dev] [RFC 0/6] New sync modes for ring Stephen Hemminger
` (2 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-02-24 11:35 UTC (permalink / raw)
To: dev; +Cc: olivier.matz, Konstantin Ananyev
Introduce new test case to test HTS ring mode under contention.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_hts_stress.c | 28 ++++++++++++++++++++++++++++
3 files changed, 30 insertions(+)
create mode 100644 app/test/test_ring_hts_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index d22b9f702..ff151cd55 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -77,6 +77,7 @@ SRCS-y += test_external_mem.c
SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
+SRCS-y += test_ring_hts_stress.c
SRCS-y += test_ring_perf.c
SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
diff --git a/app/test/meson.build b/app/test/meson.build
index fa4fb4b51..7e58fa999 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -100,6 +100,7 @@ test_sources = files('commands.c',
'test_rib.c',
'test_rib6.c',
'test_ring.c',
+ 'test_ring_hts_stress.c',
'test_ring_perf.c',
'test_ring_rts_stress.c',
'test_ring_stress.c',
diff --git a/app/test/test_ring_hts_stress.c b/app/test/test_ring_hts_stress.c
new file mode 100644
index 000000000..b7f2d21fc
--- /dev/null
+++ b/app/test/test_ring_hts_stress.c
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_hts_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_hts_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_HTS_ENQ | RING_F_MC_HTS_DEQ);
+}
+
+REGISTER_TEST_COMMAND(ring_stress_hts_autotest, test_ring_stress);
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-24 11:35 [dpdk-dev] [RFC 0/6] New sync modes for ring Konstantin Ananyev
` (5 preceding siblings ...)
2020-02-24 11:35 ` [dpdk-dev] [RFC 6/6] test/ring: add contention stress test for HTS ring Konstantin Ananyev
@ 2020-02-24 16:59 ` Stephen Hemminger
2020-02-24 17:59 ` Jerin Jacob
2020-03-25 20:43 ` Honnappa Nagarahalli
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
8 siblings, 1 reply; 146+ messages in thread
From: Stephen Hemminger @ 2020-02-24 16:59 UTC (permalink / raw)
To: Konstantin Ananyev; +Cc: dev, olivier.matz
On Mon, 24 Feb 2020 11:35:09 +0000
Konstantin Ananyev <konstantin.ananyev@intel.com> wrote:
> Upfront note - that RFC is not a complete patch.
> It introduces an ABI breakage, plus it doesn't update ring_elem
> code properly, etc.
> I plan to deal with all these things in later versions.
> Right now I seek an initial feedback about proposed ideas.
> Would also ask people to repeat performance tests (see below)
> on their platforms to confirm the impact.
>
> More and more customers use(/try to use) DPDK based apps within
> overcommitted systems (multiple acttive threads over same pysical cores):
> VM, container deployments, etc.
> One quite common problem they hit: Lock-Holder-Preemption with rte_ring.
> LHP is quite a common problem for spin-based sync primitives
> (spin-locks, etc.) on overcommitted systems.
> The situation gets much worse when some sort of
> fair-locking technique is used (ticket-lock, etc.).
> As now not only lock-owner but also lock-waiters scheduling
> order matters a lot.
> This is a well-known problem for kernel within VMs:
> http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
> The problem with rte_ring is that while head accusion is sort of
> un-fair locking, waiting on tail is very similar to ticket lock schema -
> tail has to be updated in particular order.
> That makes current rte_ring implementation to perform
> really pure on some overcommited scenarios.
Rather than reform rte_ring to fit this scenario, it would make
more sense to me to introduce another primitive. The current lockless
ring performs very well for the isolated thread model that DPDK
was built around. This looks like a case of customers violating
the usage model of the DPDK and then being surprised at the fallout.
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-24 16:59 ` [dpdk-dev] [RFC 0/6] New sync modes for ring Stephen Hemminger
@ 2020-02-24 17:59 ` Jerin Jacob
2020-02-24 19:35 ` Stephen Hemminger
2020-02-25 0:58 ` Honnappa Nagarahalli
0 siblings, 2 replies; 146+ messages in thread
From: Jerin Jacob @ 2020-02-24 17:59 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: Konstantin Ananyev, dpdk-dev, Olivier Matz
On Mon, Feb 24, 2020 at 10:29 PM Stephen Hemminger
<stephen@networkplumber.org> wrote:
>
> On Mon, 24 Feb 2020 11:35:09 +0000
> Konstantin Ananyev <konstantin.ananyev@intel.com> wrote:
>
> > Upfront note - that RFC is not a complete patch.
> > It introduces an ABI breakage, plus it doesn't update ring_elem
> > code properly, etc.
> > I plan to deal with all these things in later versions.
> > Right now I seek an initial feedback about proposed ideas.
> > Would also ask people to repeat performance tests (see below)
> > on their platforms to confirm the impact.
> >
> > More and more customers use(/try to use) DPDK based apps within
> > overcommitted systems (multiple acttive threads over same pysical cores):
> > VM, container deployments, etc.
> > One quite common problem they hit: Lock-Holder-Preemption with rte_ring.
> > LHP is quite a common problem for spin-based sync primitives
> > (spin-locks, etc.) on overcommitted systems.
> > The situation gets much worse when some sort of
> > fair-locking technique is used (ticket-lock, etc.).
> > As now not only lock-owner but also lock-waiters scheduling
> > order matters a lot.
> > This is a well-known problem for kernel within VMs:
> > http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> > https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
> > The problem with rte_ring is that while head accusion is sort of
> > un-fair locking, waiting on tail is very similar to ticket lock schema -
> > tail has to be updated in particular order.
> > That makes current rte_ring implementation to perform
> > really pure on some overcommited scenarios.
>
> Rather than reform rte_ring to fit this scenario, it would make
> more sense to me to introduce another primitive. The current lockless
> ring performs very well for the isolated thread model that DPDK
> was built around. This looks like a case of customers violating
> the usage model of the DPDK and then being surprised at the fallout.
I agree with Stephen here.
I think, adding more runtime check in the enqueue() and dequeue() will
have a bad effect on the low-end cores too.
But I agree with the problem statement that in the virtualization use
case, It may be possible to have N virtual cores runs on a physical
core.
IMO, The best solution would be keeping the ring API same and have a
different flavor in "compile-time". Something like
liburcu did for accommodating different flavors.
i.e urcu-qsbr.h and urcu-bp.h will identical definition of API. The
application can simply include ONE header file in a C file based on
the flavor.
If need both at runtime. Need to have function pointer or so in the
application and define the function in different c file by including
the approaite flavor in C file.
#include <urcu-qsbr.h> /* QSBR RCU flavor */
#include <urcu-bp.h> /* Bulletproof RCU flavor */
>
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-24 17:59 ` Jerin Jacob
@ 2020-02-24 19:35 ` Stephen Hemminger
2020-02-24 20:52 ` Honnappa Nagarahalli
2020-02-25 13:41 ` Ananyev, Konstantin
2020-02-25 0:58 ` Honnappa Nagarahalli
1 sibling, 2 replies; 146+ messages in thread
From: Stephen Hemminger @ 2020-02-24 19:35 UTC (permalink / raw)
To: Jerin Jacob; +Cc: Konstantin Ananyev, dpdk-dev, Olivier Matz
On Mon, 24 Feb 2020 23:29:57 +0530
Jerin Jacob <jerinjacobk@gmail.com> wrote:
> On Mon, Feb 24, 2020 at 10:29 PM Stephen Hemminger
> <stephen@networkplumber.org> wrote:
> >
> > On Mon, 24 Feb 2020 11:35:09 +0000
> > Konstantin Ananyev <konstantin.ananyev@intel.com> wrote:
> >
> > > Upfront note - that RFC is not a complete patch.
> > > It introduces an ABI breakage, plus it doesn't update ring_elem
> > > code properly, etc.
> > > I plan to deal with all these things in later versions.
> > > Right now I seek an initial feedback about proposed ideas.
> > > Would also ask people to repeat performance tests (see below)
> > > on their platforms to confirm the impact.
> > >
> > > More and more customers use(/try to use) DPDK based apps within
> > > overcommitted systems (multiple acttive threads over same pysical cores):
> > > VM, container deployments, etc.
> > > One quite common problem they hit: Lock-Holder-Preemption with rte_ring.
> > > LHP is quite a common problem for spin-based sync primitives
> > > (spin-locks, etc.) on overcommitted systems.
> > > The situation gets much worse when some sort of
> > > fair-locking technique is used (ticket-lock, etc.).
> > > As now not only lock-owner but also lock-waiters scheduling
> > > order matters a lot.
> > > This is a well-known problem for kernel within VMs:
> > > http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> > > https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
> > > The problem with rte_ring is that while head accusion is sort of
> > > un-fair locking, waiting on tail is very similar to ticket lock schema -
> > > tail has to be updated in particular order.
> > > That makes current rte_ring implementation to perform
> > > really pure on some overcommited scenarios.
> >
> > Rather than reform rte_ring to fit this scenario, it would make
> > more sense to me to introduce another primitive. The current lockless
> > ring performs very well for the isolated thread model that DPDK
> > was built around. This looks like a case of customers violating
> > the usage model of the DPDK and then being surprised at the fallout.
>
> I agree with Stephen here.
>
> I think, adding more runtime check in the enqueue() and dequeue() will
> have a bad effect on the low-end cores too.
> But I agree with the problem statement that in the virtualization use
> case, It may be possible to have N virtual cores runs on a physical
> core.
>
> IMO, The best solution would be keeping the ring API same and have a
> different flavor in "compile-time". Something like
> liburcu did for accommodating different flavors.
>
> i.e urcu-qsbr.h and urcu-bp.h will identical definition of API. The
> application can simply include ONE header file in a C file based on
> the flavor.
> If need both at runtime. Need to have function pointer or so in the
> application and define the function in different c file by including
> the approaite flavor in C file.
This would also be a good time to consider the tradeoffs of the
heavy use of inlining that is done in rte_ring vs the impact that
has on API/ABI stability.
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-24 19:35 ` Stephen Hemminger
@ 2020-02-24 20:52 ` Honnappa Nagarahalli
2020-02-25 11:45 ` Ananyev, Konstantin
2020-02-25 13:41 ` Ananyev, Konstantin
1 sibling, 1 reply; 146+ messages in thread
From: Honnappa Nagarahalli @ 2020-02-24 20:52 UTC (permalink / raw)
To: Stephen Hemminger, Jerin Jacob
Cc: Konstantin Ananyev, dpdk-dev, Olivier Matz, Honnappa Nagarahalli, nd, nd
> -----Original Message-----
> From: dev <dev-bounces@dpdk.org> On Behalf Of Stephen Hemminger
> Sent: Monday, February 24, 2020 1:35 PM
> To: Jerin Jacob <jerinjacobk@gmail.com>
> Cc: Konstantin Ananyev <konstantin.ananyev@intel.com>; dpdk-dev
> <dev@dpdk.org>; Olivier Matz <olivier.matz@6wind.com>
> Subject: Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
>
> On Mon, 24 Feb 2020 23:29:57 +0530
> Jerin Jacob <jerinjacobk@gmail.com> wrote:
>
> > On Mon, Feb 24, 2020 at 10:29 PM Stephen Hemminger
> > <stephen@networkplumber.org> wrote:
> > >
> > > On Mon, 24 Feb 2020 11:35:09 +0000
> > > Konstantin Ananyev <konstantin.ananyev@intel.com> wrote:
> > >
> > > > Upfront note - that RFC is not a complete patch.
> > > > It introduces an ABI breakage, plus it doesn't update ring_elem
> > > > code properly, etc.
> > > > I plan to deal with all these things in later versions.
> > > > Right now I seek an initial feedback about proposed ideas.
> > > > Would also ask people to repeat performance tests (see below) on
> > > > their platforms to confirm the impact.
> > > >
> > > > More and more customers use(/try to use) DPDK based apps within
> > > > overcommitted systems (multiple acttive threads over same pysical
> cores):
> > > > VM, container deployments, etc.
> > > > One quite common problem they hit: Lock-Holder-Preemption with
> rte_ring.
> > > > LHP is quite a common problem for spin-based sync primitives
> > > > (spin-locks, etc.) on overcommitted systems.
> > > > The situation gets much worse when some sort of fair-locking
> > > > technique is used (ticket-lock, etc.).
> > > > As now not only lock-owner but also lock-waiters scheduling order
> > > > matters a lot.
> > > > This is a well-known problem for kernel within VMs:
> > > > http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> > > > https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
> > > > The problem with rte_ring is that while head accusion is sort of
> > > > un-fair locking, waiting on tail is very similar to ticket lock
> > > > schema - tail has to be updated in particular order.
> > > > That makes current rte_ring implementation to perform really pure
> > > > on some overcommited scenarios.
> > >
> > > Rather than reform rte_ring to fit this scenario, it would make more
> > > sense to me to introduce another primitive. The current lockless
> > > ring performs very well for the isolated thread model that DPDK was
> > > built around. This looks like a case of customers violating the
> > > usage model of the DPDK and then being surprised at the fallout.
> >
> > I agree with Stephen here.
> >
> > I think, adding more runtime check in the enqueue() and dequeue() will
> > have a bad effect on the low-end cores too.
> > But I agree with the problem statement that in the virtualization use
> > case, It may be possible to have N virtual cores runs on a physical
> > core.
> >
> > IMO, The best solution would be keeping the ring API same and have a
> > different flavor in "compile-time". Something like liburcu did for
> > accommodating different flavors.
> >
> > i.e urcu-qsbr.h and urcu-bp.h will identical definition of API. The
> > application can simply include ONE header file in a C file based on
> > the flavor.
> > If need both at runtime. Need to have function pointer or so in the
> > application and define the function in different c file by including
> > the approaite flavor in C file.
>
> This would also be a good time to consider the tradeoffs of the heavy use of
> inlining that is done in rte_ring vs the impact that has on API/ABI stability.
>
I was working on few requirements in rte_ring library for RCU defer APIs. RFC is at https://patchwork.dpdk.org/cover/66020/.
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-24 17:59 ` Jerin Jacob
2020-02-24 19:35 ` Stephen Hemminger
@ 2020-02-25 0:58 ` Honnappa Nagarahalli
2020-02-25 15:14 ` Ananyev, Konstantin
1 sibling, 1 reply; 146+ messages in thread
From: Honnappa Nagarahalli @ 2020-02-25 0:58 UTC (permalink / raw)
To: Jerin Jacob, Stephen Hemminger
Cc: Konstantin Ananyev, dpdk-dev, Olivier Matz, Honnappa Nagarahalli, nd, nd
<snip>
>
> On Mon, Feb 24, 2020 at 10:29 PM Stephen Hemminger
> <stephen@networkplumber.org> wrote:
> >
> > On Mon, 24 Feb 2020 11:35:09 +0000
> > Konstantin Ananyev <konstantin.ananyev@intel.com> wrote:
> >
> > > Upfront note - that RFC is not a complete patch.
> > > It introduces an ABI breakage, plus it doesn't update ring_elem code
> > > properly, etc.
> > > I plan to deal with all these things in later versions.
> > > Right now I seek an initial feedback about proposed ideas.
> > > Would also ask people to repeat performance tests (see below) on
> > > their platforms to confirm the impact.
> > >
> > > More and more customers use(/try to use) DPDK based apps within
> > > overcommitted systems (multiple acttive threads over same pysical cores):
> > > VM, container deployments, etc.
> > > One quite common problem they hit: Lock-Holder-Preemption with
> rte_ring.
> > > LHP is quite a common problem for spin-based sync primitives
> > > (spin-locks, etc.) on overcommitted systems.
> > > The situation gets much worse when some sort of fair-locking
> > > technique is used (ticket-lock, etc.).
> > > As now not only lock-owner but also lock-waiters scheduling order
> > > matters a lot.
> > > This is a well-known problem for kernel within VMs:
> > > http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> > > https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
These slides seem to indicate that the problems are mitigated through the Hypervisor configuration. Do we still need to address the issues?
> > > The problem with rte_ring is that while head accusion is sort of
> > > un-fair locking, waiting on tail is very similar to ticket lock
> > > schema - tail has to be updated in particular order.
> > > That makes current rte_ring implementation to perform really pure on
> > > some overcommited scenarios.
> >
> > Rather than reform rte_ring to fit this scenario, it would make more
> > sense to me to introduce another primitive. The current lockless ring
> > performs very well for the isolated thread model that DPDK was built
> > around. This looks like a case of customers violating the usage model
> > of the DPDK and then being surprised at the fallout.
>
> I agree with Stephen here.
>
> I think, adding more runtime check in the enqueue() and dequeue() will have a
> bad effect on the low-end cores too.
> But I agree with the problem statement that in the virtualization use case, It
> may be possible to have N virtual cores runs on a physical core.
It is hard to imagine that there are data plane applications deployed in such environments. Wouldn't this affect the performance terribly?
>
> IMO, The best solution would be keeping the ring API same and have a
> different flavor in "compile-time". Something like liburcu did for
> accommodating different flavors.
>
> i.e urcu-qsbr.h and urcu-bp.h will identical definition of API. The application
> can simply include ONE header file in a C file based on the flavor.
> If need both at runtime. Need to have function pointer or so in the application
> and define the function in different c file by including the approaite flavor in C
> file.
>
> #include <urcu-qsbr.h> /* QSBR RCU flavor */ #include <urcu-bp.h> /*
> Bulletproof RCU flavor */
>
>
>
>
>
>
>
>
>
>
>
>
>
> >
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-24 20:52 ` Honnappa Nagarahalli
@ 2020-02-25 11:45 ` Ananyev, Konstantin
0 siblings, 0 replies; 146+ messages in thread
From: Ananyev, Konstantin @ 2020-02-25 11:45 UTC (permalink / raw)
To: Honnappa Nagarahalli, Stephen Hemminger, Jerin Jacob
Cc: dpdk-dev, Olivier Matz, nd, nd
> > -----Original Message-----
> > From: dev <dev-bounces@dpdk.org> On Behalf Of Stephen Hemminger
> > Sent: Monday, February 24, 2020 1:35 PM
> > To: Jerin Jacob <jerinjacobk@gmail.com>
> > Cc: Konstantin Ananyev <konstantin.ananyev@intel.com>; dpdk-dev
> > <dev@dpdk.org>; Olivier Matz <olivier.matz@6wind.com>
> > Subject: Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
> >
> > On Mon, 24 Feb 2020 23:29:57 +0530
> > Jerin Jacob <jerinjacobk@gmail.com> wrote:
> >
> > > On Mon, Feb 24, 2020 at 10:29 PM Stephen Hemminger
> > > <stephen@networkplumber.org> wrote:
> > > >
> > > > On Mon, 24 Feb 2020 11:35:09 +0000
> > > > Konstantin Ananyev <konstantin.ananyev@intel.com> wrote:
> > > >
> > > > > Upfront note - that RFC is not a complete patch.
> > > > > It introduces an ABI breakage, plus it doesn't update ring_elem
> > > > > code properly, etc.
> > > > > I plan to deal with all these things in later versions.
> > > > > Right now I seek an initial feedback about proposed ideas.
> > > > > Would also ask people to repeat performance tests (see below) on
> > > > > their platforms to confirm the impact.
> > > > >
> > > > > More and more customers use(/try to use) DPDK based apps within
> > > > > overcommitted systems (multiple acttive threads over same pysical
> > cores):
> > > > > VM, container deployments, etc.
> > > > > One quite common problem they hit: Lock-Holder-Preemption with
> > rte_ring.
> > > > > LHP is quite a common problem for spin-based sync primitives
> > > > > (spin-locks, etc.) on overcommitted systems.
> > > > > The situation gets much worse when some sort of fair-locking
> > > > > technique is used (ticket-lock, etc.).
> > > > > As now not only lock-owner but also lock-waiters scheduling order
> > > > > matters a lot.
> > > > > This is a well-known problem for kernel within VMs:
> > > > > http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> > > > > https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
> > > > > The problem with rte_ring is that while head accusion is sort of
> > > > > un-fair locking, waiting on tail is very similar to ticket lock
> > > > > schema - tail has to be updated in particular order.
> > > > > That makes current rte_ring implementation to perform really pure
> > > > > on some overcommited scenarios.
> > > >
> > > > Rather than reform rte_ring to fit this scenario, it would make more
> > > > sense to me to introduce another primitive. The current lockless
> > > > ring performs very well for the isolated thread model that DPDK was
> > > > built around. This looks like a case of customers violating the
> > > > usage model of the DPDK and then being surprised at the fallout.
> > >
> > > I agree with Stephen here.
> > >
> > > I think, adding more runtime check in the enqueue() and dequeue() will
> > > have a bad effect on the low-end cores too.
> > > But I agree with the problem statement that in the virtualization use
> > > case, It may be possible to have N virtual cores runs on a physical
> > > core.
> > >
> > > IMO, The best solution would be keeping the ring API same and have a
> > > different flavor in "compile-time". Something like liburcu did for
> > > accommodating different flavors.
> > >
> > > i.e urcu-qsbr.h and urcu-bp.h will identical definition of API. The
> > > application can simply include ONE header file in a C file based on
> > > the flavor.
> > > If need both at runtime. Need to have function pointer or so in the
> > > application and define the function in different c file by including
> > > the approaite flavor in C file.
> >
> > This would also be a good time to consider the tradeoffs of the heavy use of
> > inlining that is done in rte_ring vs the impact that has on API/ABI stability.
> >
> I was working on few requirements in rte_ring library for RCU defer APIs. RFC is at https://patchwork.dpdk.org/cover/66020/.
Yep, noticed your patch, seems we sort of collided here.
As I understand you patch aims to provide functionality similar to my HTS one.
Will try to look at yours one in next fee days, hopefully we can end-up with
some common denominator.
Konstantin
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-24 19:35 ` Stephen Hemminger
2020-02-24 20:52 ` Honnappa Nagarahalli
@ 2020-02-25 13:41 ` Ananyev, Konstantin
2020-02-26 16:53 ` Morten Brørup
2020-02-27 10:31 ` Jerin Jacob
1 sibling, 2 replies; 146+ messages in thread
From: Ananyev, Konstantin @ 2020-02-25 13:41 UTC (permalink / raw)
To: Stephen Hemminger, Jerin Jacob; +Cc: dpdk-dev, Olivier Matz, drc
> > > > Upfront note - that RFC is not a complete patch.
> > > > It introduces an ABI breakage, plus it doesn't update ring_elem
> > > > code properly, etc.
> > > > I plan to deal with all these things in later versions.
> > > > Right now I seek an initial feedback about proposed ideas.
> > > > Would also ask people to repeat performance tests (see below)
> > > > on their platforms to confirm the impact.
> > > >
> > > > More and more customers use(/try to use) DPDK based apps within
> > > > overcommitted systems (multiple acttive threads over same pysical cores):
> > > > VM, container deployments, etc.
> > > > One quite common problem they hit: Lock-Holder-Preemption with rte_ring.
> > > > LHP is quite a common problem for spin-based sync primitives
> > > > (spin-locks, etc.) on overcommitted systems.
> > > > The situation gets much worse when some sort of
> > > > fair-locking technique is used (ticket-lock, etc.).
> > > > As now not only lock-owner but also lock-waiters scheduling
> > > > order matters a lot.
> > > > This is a well-known problem for kernel within VMs:
> > > > http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> > > > https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
> > > > The problem with rte_ring is that while head accusion is sort of
> > > > un-fair locking, waiting on tail is very similar to ticket lock schema -
> > > > tail has to be updated in particular order.
> > > > That makes current rte_ring implementation to perform
> > > > really pure on some overcommited scenarios.
> > >
> > > Rather than reform rte_ring to fit this scenario, it would make
> > > more sense to me to introduce another primitive.
I don't see much advantages it will bring us.
As a disadvantages, for developers and maintainers - code duplication,
for end users - extra code churn and removed ability to mix and match
different sync modes in one ring.
> The current lockless
> > > ring performs very well for the isolated thread model that DPDK
> > > was built around. This looks like a case of customers violating
> > > the usage model of the DPDK and then being surprised at the fallout.
For customers using isolated thread model - nothing should change
(both in terms of API and performance).
Existing sync modes MP/MC,SP/SC kept untouched, set up in the same
way (via flags and _init_), and MP/MC remains as default one.
From other side I don't see why we should ignore customers that want to use
their DPDK apps in different deployment scenarios.
> >
> > I agree with Stephen here.
> >
> > I think, adding more runtime check in the enqueue() and dequeue() will
> > have a bad effect on the low-end cores too.
We do have a run-time check in our current enqueue()/dequeue implementation.
In fact we support both modes: we have generic rte_ring_enqueue(/dequeue)_bulk(/burst)
where sync behaviour is determined at runtime by value of prod(/cons).single.
Or user can call rte_ring_(mp/sp)_enqueue_* functions directly.
This RFC follows exactly the same paradigm:
rte_ring_enqueue(/dequeue)_bulk(/burst) kept generic and it's
behaviour is determined at runtime, by value of prod(/cons).sync_type.
Or user can call enqueue/dequeue with particular sync mode directly:
rte_ring_(mp/sp/rts/hts)_enqueue_(bulk/burst)*.
The only thing that changed:
Format of prod/cons now could differ depending on mode selected at _init_.
So you can't create a ring for let say SP mode and then in the middle of data-path
change your mind and start using MP_RTS mode.
For existing modes (SP/MP, SC/MC) format remains the same and user can still
use them interchangeably, though of course that is an error prone practice.
> > But I agree with the problem statement that in the virtualization use
> > case, It may be possible to have N virtual cores runs on a physical
> > core.
> >
> > IMO, The best solution would be keeping the ring API same and have a
> > different flavor in "compile-time". Something like
> > liburcu did for accommodating different flavors.
> >
> > i.e urcu-qsbr.h and urcu-bp.h will identical definition of API. The
> > application can simply include ONE header file in a C file based on
> > the flavor.
I don't think it is a flexible enough approach.
In one app user might need to have several rings with different sync modes.
Or even user might need a ring with different sync modes for enqueue/dequeue.
> > If need both at runtime. Need to have function pointer or so in the
> > application and define the function in different c file by including
> > the approaite flavor in C file.
Big issue with function pointers here would be DPDK MP model.
AFAIK, rte_ring is quite popular mechanism for IPC between DPDK apps.
To support such model, we'll need to split rte_ring data into 'shared'
and 'private' and initialize private one for every process that is going to use it.
That sounds like a massive change, and I am not sure the required effort will worth it.
BTW, if user just calls API functions without trying to access structure internals directly,
I don't think it would be a big difference for him what is inside:
indirect function call or inlined switch(...) {}.
> This would also be a good time to consider the tradeoffs of the
> heavy use of inlining that is done in rte_ring vs the impact that
> has on API/ABI stability.
Yes, hiding rte_ring implementation inside .c would help a lot
in terms of ABI maintenance and would make our future life easier.
The question is what is the price for it in terms of performance,
and are we ready to pay it. Not to mention that it would cause
changes in many other libs/apps...
So I think it should be a subject for a separate discussion.
But, agree it would be good at least to measure the performance
impact of such change.
If I'll have some spare cycles, will give it a try.
Meanwhile, can I ask Jerin and other guys to repeat tests from this RFC
on their HW? Before continuing discussion would probably be good to know
does the suggested patch work as expected across different platforms.
Thanks
Konstantin
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-25 0:58 ` Honnappa Nagarahalli
@ 2020-02-25 15:14 ` Ananyev, Konstantin
0 siblings, 0 replies; 146+ messages in thread
From: Ananyev, Konstantin @ 2020-02-25 15:14 UTC (permalink / raw)
To: Honnappa Nagarahalli, Jerin Jacob, Stephen Hemminger
Cc: dpdk-dev, Olivier Matz, nd, nd, wang.yong19
> > > > Upfront note - that RFC is not a complete patch.
> > > > It introduces an ABI breakage, plus it doesn't update ring_elem code
> > > > properly, etc.
> > > > I plan to deal with all these things in later versions.
> > > > Right now I seek an initial feedback about proposed ideas.
> > > > Would also ask people to repeat performance tests (see below) on
> > > > their platforms to confirm the impact.
> > > >
> > > > More and more customers use(/try to use) DPDK based apps within
> > > > overcommitted systems (multiple acttive threads over same pysical cores):
> > > > VM, container deployments, etc.
> > > > One quite common problem they hit: Lock-Holder-Preemption with
> > rte_ring.
> > > > LHP is quite a common problem for spin-based sync primitives
> > > > (spin-locks, etc.) on overcommitted systems.
> > > > The situation gets much worse when some sort of fair-locking
> > > > technique is used (ticket-lock, etc.).
> > > > As now not only lock-owner but also lock-waiters scheduling order
> > > > matters a lot.
> > > > This is a well-known problem for kernel within VMs:
> > > > http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> > > > https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
> These slides seem to indicate that the problems are mitigated through the Hypervisor configuration. Do we still need to address the issues?
I am not really an expert here, but AFAIK current mitigations deal mostly with guest kernel:
linux implements PV version of spinlocks (unfair and/or based on hypercall availability),
hypervisor might make decision itself based on is guest in user/kernel mode,
plus on some special cpu instructions.
We do spin in user-space mode.
Might be hypervisors became smarter these days, but so far,
I heard about few different customers that hit such problem.
As an example, NA DPDK summit presentation:
https://dpdkna2019.sched.com/event/WYBG/dpdk-containers-challenges-solutions-wang-yong-zte
page 16 (problem #4) describes same issue.
>
> > > > The problem with rte_ring is that while head accusion is sort of
> > > > un-fair locking, waiting on tail is very similar to ticket lock
> > > > schema - tail has to be updated in particular order.
> > > > That makes current rte_ring implementation to perform really pure on
> > > > some overcommited scenarios.
> > >
> > > Rather than reform rte_ring to fit this scenario, it would make more
> > > sense to me to introduce another primitive. The current lockless ring
> > > performs very well for the isolated thread model that DPDK was built
> > > around. This looks like a case of customers violating the usage model
> > > of the DPDK and then being surprised at the fallout.
> >
> > I agree with Stephen here.
> >
> > I think, adding more runtime check in the enqueue() and dequeue() will have a
> > bad effect on the low-end cores too.
> > But I agree with the problem statement that in the virtualization use case, It
> > may be possible to have N virtual cores runs on a physical core.
> It is hard to imagine that there are data plane applications deployed in such environments. Wouldn't this affect the performance terribly?
It wouldn't reach same performance as isolated threads,
but for some tasks it might be enough.
AFAIK, one quite common scenario - few isolated threads(/processes) doing
actual IO and then spread packets over dozens(/hundreds) non-isolated
consumers.
>
> >
> > IMO, The best solution would be keeping the ring API same and have a
> > different flavor in "compile-time". Something like liburcu did for
> > accommodating different flavors.
> >
> > i.e urcu-qsbr.h and urcu-bp.h will identical definition of API. The application
> > can simply include ONE header file in a C file based on the flavor.
> > If need both at runtime. Need to have function pointer or so in the application
> > and define the function in different c file by including the approaite flavor in C
> > file.
> >
> > #include <urcu-qsbr.h> /* QSBR RCU flavor */ #include <urcu-bp.h> /*
> > Bulletproof RCU flavor */
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-25 13:41 ` Ananyev, Konstantin
@ 2020-02-26 16:53 ` Morten Brørup
2020-02-27 10:31 ` Jerin Jacob
1 sibling, 0 replies; 146+ messages in thread
From: Morten Brørup @ 2020-02-26 16:53 UTC (permalink / raw)
To: Ananyev, Konstantin, Stephen Hemminger, Jerin Jacob
Cc: dpdk-dev, Olivier Matz, drc
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Ananyev,
> Konstantin
>
<snip>
> > > > > More and more customers use(/try to use) DPDK based apps within
> > > > > overcommitted systems (multiple acttive threads over same
> pysical cores):
> > > > > VM, container deployments, etc.
<snip>
> > > > > That makes current rte_ring implementation to perform
> > > > > really pure on some overcommited scenarios.
> > > >
> > > > Rather than reform rte_ring to fit this scenario, it would make
> > > > more sense to me to introduce another primitive.
>
> I don't see much advantages it will bring us.
> As a disadvantages, for developers and maintainers - code duplication,
> for end users - extra code churn and removed ability to mix and match
> different sync modes in one ring.
>
I strongly agree with Konstantin on this.
Please consider this discussion at a higher abstraction level:
As DPDK applications grow in number and popularity, people will deploy them in overcommitted systems. In this scenario, I consider it extremely unlikely that the user is able to dedicate N physical cores to N specific lcores of the M total lcores of the DPDK application. In the typical hypervisor scenario, the user will be able to assign a number of virtual CPUs to the virtual machine running the DPDK application, and these vCPUs will either be dedicated to the virtual machine or shared with other virtual machines.
DPDK is currently designed for dedicated CPUs only. If a user runs it on shared vCPUs, the user is violating an important DPDK precondition, and DPDK's behavior is undefined!
Adding the ability to run DPDK applications on shared vCPUs would be a great improvement.
I prefer that support for this is as painless as possible for the DPDK application developer.
Perhaps run-time detection during the EAL initialization could be a solution. The EAL would then configure relevant libraries to use the appropriate synchronization primitives.
Med venlig hilsen / kind regards
- Morten Brørup
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-25 13:41 ` Ananyev, Konstantin
2020-02-26 16:53 ` Morten Brørup
@ 2020-02-27 10:31 ` Jerin Jacob
2020-02-28 0:17 ` David Christensen
1 sibling, 1 reply; 146+ messages in thread
From: Jerin Jacob @ 2020-02-27 10:31 UTC (permalink / raw)
To: Ananyev, Konstantin; +Cc: Stephen Hemminger, dpdk-dev, Olivier Matz, drc
On Tue, Feb 25, 2020 at 7:11 PM Ananyev, Konstantin
<konstantin.ananyev@intel.com> wrote:
> We do have a run-time check in our current enqueue()/dequeue implementation.
> In fact we support both modes: we have generic rte_ring_enqueue(/dequeue)_bulk(/burst)
> where sync behaviour is determined at runtime by value of prod(/cons).single.
> Or user can call rte_ring_(mp/sp)_enqueue_* functions directly.
> This RFC follows exactly the same paradigm:
> rte_ring_enqueue(/dequeue)_bulk(/burst) kept generic and it's
> behaviour is determined at runtime, by value of prod(/cons).sync_type.
> Or user can call enqueue/dequeue with particular sync mode directly:
> rte_ring_(mp/sp/rts/hts)_enqueue_(bulk/burst)*.
> The only thing that changed:
> Format of prod/cons now could differ depending on mode selected at _init_.
> So you can't create a ring for let say SP mode and then in the middle of data-path
> change your mind and start using MP_RTS mode.
> For existing modes (SP/MP, SC/MC) format remains the same and user can still
> use them interchangeably, though of course that is an error prone practice.
Makes sense.
>
> > > But I agree with the problem statement that in the virtualization use
> > > case, It may be possible to have N virtual cores runs on a physical
> > > core.
> > >
> > > IMO, The best solution would be keeping the ring API same and have a
> > > different flavor in "compile-time". Something like
> > > liburcu did for accommodating different flavors.
> > >
> > > i.e urcu-qsbr.h and urcu-bp.h will identical definition of API. The
> > > application can simply include ONE header file in a C file based on
> > > the flavor.
>
> I don't think it is a flexible enough approach.
> In one app user might need to have several rings with different sync modes.
> Or even user might need a ring with different sync modes for enqueue/dequeue.
Ack.
> Yes, hiding rte_ring implementation inside .c would help a lot
> in terms of ABI maintenance and would make our future life easier.
> The question is what is the price for it in terms of performance,
> and are we ready to pay it. Not to mention that it would cause
> changes in many other libs/apps...
> So I think it should be a subject for a separate discussion.
> But, agree it would be good at least to measure the performance
> impact of such change.
> If I'll have some spare cycles, will give it a try.
> Meanwhile, can I ask Jerin and other guys to repeat tests from this RFC
> on their HW? Before continuing discussion would probably be good to know
> does the suggested patch work as expected across different platforms.
I tested on an arm64 HW. The former section is without the
patch(20.02) and later one with this patch.
I agree with Konstantin that getting more platform tests will be good
early so that we can focus on the approach
to avoid back and forth latter.
RTE>>ring_perf_autotest // without path
### Testing single element enq/deq ###
legacy APIs: SP/SC: single: 289.78
legacy APIs: MP/MC: single: 516.20
### Testing burst enq/deq ###
legacy APIs: SP/SC: burst (size: 8): 312.88
legacy APIs: SP/SC: burst (size: 32): 426.72
legacy APIs: MP/MC: burst (size: 8): 510.95
legacy APIs: MP/MC: burst (size: 32): 702.01
### Testing bulk enq/deq ###
legacy APIs: SP/SC: bulk (size: 8): 306.74
legacy APIs: SP/SC: bulk (size: 32): 411.56
legacy APIs: MP/MC: bulk (size: 8): 501.32
legacy APIs: MP/MC: bulk (size: 32): 693.07
### Testing empty bulk deq ###
legacy APIs: SP/SC: bulk (size: 8): 7.00
legacy APIs: MP/MC: bulk (size: 8): 7.00
### Testing using two physical cores ###
legacy APIs: SP/SC: bulk (size: 8): 74.36
legacy APIs: MP/MC: bulk (size: 8): 110.18
legacy APIs: SP/SC: bulk (size: 32): 23.04
legacy APIs: MP/MC: bulk (size: 32): 32.29
### Testing using all slave nodes ##
Bulk enq/dequeue count on size 8
Core [8] count = 293741
Core [9] count = 293741
Total count (size: 8): 587482
Bulk enq/dequeue count on size 32
Core [8] count = 244909
Core [9] count = 244909
Total count (size: 32): 1077300
### Testing single element enq/deq ###
elem APIs: element size 16B: SP/SC: single: 255.37
elem APIs: element size 16B: MP/MC: single: 456.68
### Testing burst enq/deq ###
elem APIs: element size 16B: SP/SC: burst (size: 8): 291.99
elem APIs: element size 16B: SP/SC: burst (size: 32): 456.25
elem APIs: element size 16B: MP/MC: burst (size: 8): 497.77
elem APIs: element size 16B: MP/MC: burst (size: 32): 680.87
### Testing bulk enq/deq ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 284.40
elem APIs: element size 16B: SP/SC: bulk (size: 32): 453.17
elem APIs: element size 16B: MP/MC: bulk (size: 8): 485.77
elem APIs: element size 16B: MP/MC: bulk (size: 32): 675.08
### Testing empty bulk deq ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 8.00
elem APIs: element size 16B: MP/MC: bulk (size: 8): 7.00
### Testing using two physical cores ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 74.45
elem APIs: element size 16B: MP/MC: bulk (size: 8): 105.91
elem APIs: element size 16B: SP/SC: bulk (size: 32): 22.92
elem APIs: element size 16B: MP/MC: bulk (size: 32): 31.55
### Testing using all slave nodes ###
Bulk enq/dequeue count on size 8
Core [8] count = 308724
Core [9] count = 308723
Total count (size: 8): 617447
Bulk enq/dequeue count on size 32
Core [8] count = 214269
Core [9] count = 214269
Total count (size: 32): 1045985
RTE>>ring_perf_autotest // with patch
### Testing single element enq/deq ###
legacy APIs: SP/SC: single: 289.78
legacy APIs: MP/MC: single: 475.76
### Testing burst enq/deq ###
legacy APIs: SP/SC: burst (size: 8): 323.91
legacy APIs: SP/SC: burst (size: 32): 424.60
legacy APIs: MP/MC: burst (size: 8): 523.00
legacy APIs: MP/MC: burst (size: 32): 717.09
### Testing bulk enq/deq ###
legacy APIs: SP/SC: bulk (size: 8): 317.74
legacy APIs: SP/SC: bulk (size: 32): 413.57
legacy APIs: MP/MC: bulk (size: 8): 512.89
legacy APIs: MP/MC: bulk (size: 32): 712.45
### Testing empty bulk deq ###
legacy APIs: SP/SC: bulk (size: 8): 7.00
legacy APIs: MP/MC: bulk (size: 8): 7.00
### Testing using two physical cores ###
legacy APIs: SP/SC: bulk (size: 8): 74.82
legacy APIs: MP/MC: bulk (size: 8): 96.45
legacy APIs: SP/SC: bulk (size: 32): 22.97
legacy APIs: MP/MC: bulk (size: 32): 32.52
### Testing using all slave nodes ###
Bulk enq/dequeue count on size 8
Core [8] count = 283928
Core [9] count = 283927
Total count (size: 8): 567855
Bulk enq/dequeue count on size 32
Core [8] count = 223916
Core [9] count = 223915
Total count (size: 32): 1015686
### Testing single element enq/deq ###
elem APIs: element size 16B: SP/SC: single: 267.65
elem APIs: element size 16B: MP/MC: single: 439.06
### Testing burst enq/deq ###
elem APIs: element size 16B: SP/SC: burst (size: 8): 302.44
elem APIs: element size 16B: SP/SC: burst (size: 32): 466.31
elem APIs: element size 16B: MP/MC: burst (size: 8): 502.51
elem APIs: element size 16B: MP/MC: burst (size: 32): 695.81
### Testing bulk enq/deq ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 295.15
elem APIs: element size 16B: SP/SC: bulk (size: 32): 462.77
elem APIs: element size 16B: MP/MC: bulk (size: 8): 496.89
elem APIs: element size 16B: MP/MC: bulk (size: 32): 690.46
### Testing empty bulk deq ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 7.50
elem APIs: element size 16B: MP/MC: bulk (size: 8): 7.44
### Testing using two physical cores ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 65.85
elem APIs: element size 16B: MP/MC: bulk (size: 8): 103.80
elem APIs: element size 16B: SP/SC: bulk (size: 32): 23.27
elem APIs: element size 16B: MP/MC: bulk (size: 32): 31.17
### Testing using all slave nodes ###
Bulk enq/dequeue count on size 8
Core [8] count = 304223
Core [9] count = 304221
Total count (size: 8): 608444
Bulk enq/dequeue count on size 32
Core [8] count = 214856
Core [9] count = 214855
Total count (size: 32): 1038155
Test OK
RTE>>quit
> Thanks
> Konstantin
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-27 10:31 ` Jerin Jacob
@ 2020-02-28 0:17 ` David Christensen
2020-03-20 16:45 ` Ananyev, Konstantin
0 siblings, 1 reply; 146+ messages in thread
From: David Christensen @ 2020-02-28 0:17 UTC (permalink / raw)
To: Jerin Jacob, Ananyev, Konstantin
Cc: Stephen Hemminger, dpdk-dev, Olivier Matz
> On Tue, Feb 25, 2020 at 7:11 PM Ananyev, Konstantin
> <konstantin.ananyev@intel.com> wrote:
>
>> We do have a run-time check in our current enqueue()/dequeue implementation.
>> In fact we support both modes: we have generic rte_ring_enqueue(/dequeue)_bulk(/burst)
>> where sync behaviour is determined at runtime by value of prod(/cons).single.
>> Or user can call rte_ring_(mp/sp)_enqueue_* functions directly.
>> This RFC follows exactly the same paradigm:
>> rte_ring_enqueue(/dequeue)_bulk(/burst) kept generic and it's
>> behaviour is determined at runtime, by value of prod(/cons).sync_type.
>> Or user can call enqueue/dequeue with particular sync mode directly:
>> rte_ring_(mp/sp/rts/hts)_enqueue_(bulk/burst)*.
>> The only thing that changed:
>> Format of prod/cons now could differ depending on mode selected at _init_.
>> So you can't create a ring for let say SP mode and then in the middle of data-path
>> change your mind and start using MP_RTS mode.
>> For existing modes (SP/MP, SC/MC) format remains the same and user can still
>> use them interchangeably, though of course that is an error prone practice.
>
> Makes sense.
>
>
>>
>>>> But I agree with the problem statement that in the virtualization use
>>>> case, It may be possible to have N virtual cores runs on a physical
>>>> core.
>>>>
>>>> IMO, The best solution would be keeping the ring API same and have a
>>>> different flavor in "compile-time". Something like
>>>> liburcu did for accommodating different flavors.
>>>>
>>>> i.e urcu-qsbr.h and urcu-bp.h will identical definition of API. The
>>>> application can simply include ONE header file in a C file based on
>>>> the flavor.
>>
>> I don't think it is a flexible enough approach.
>> In one app user might need to have several rings with different sync modes.
>> Or even user might need a ring with different sync modes for enqueue/dequeue.
>
> Ack.
>
>
>> Yes, hiding rte_ring implementation inside .c would help a lot
>> in terms of ABI maintenance and would make our future life easier.
>> The question is what is the price for it in terms of performance,
>> and are we ready to pay it. Not to mention that it would cause
>> changes in many other libs/apps...
>> So I think it should be a subject for a separate discussion.
>> But, agree it would be good at least to measure the performance
>> impact of such change.
>> If I'll have some spare cycles, will give it a try.
>> Meanwhile, can I ask Jerin and other guys to repeat tests from this RFC
>> on their HW? Before continuing discussion would probably be good to know
>> does the suggested patch work as expected across different platforms.
>
>
> I tested on an arm64 HW. The former section is without the
> patch(20.02) and later one with this patch.
> I agree with Konstantin that getting more platform tests will be good
> early so that we can focus on the approach
> to avoid back and forth latter.
>
>
> RTE>>ring_perf_autotest // without path
>
> ### Testing single element enq/deq ###
> legacy APIs: SP/SC: single: 289.78
> legacy APIs: MP/MC: single: 516.20
>
> ### Testing burst enq/deq ###
> legacy APIs: SP/SC: burst (size: 8): 312.88
> legacy APIs: SP/SC: burst (size: 32): 426.72
> legacy APIs: MP/MC: burst (size: 8): 510.95
> legacy APIs: MP/MC: burst (size: 32): 702.01
>
> ### Testing bulk enq/deq ###
> legacy APIs: SP/SC: bulk (size: 8): 306.74
> legacy APIs: SP/SC: bulk (size: 32): 411.56
> legacy APIs: MP/MC: bulk (size: 8): 501.32
> legacy APIs: MP/MC: bulk (size: 32): 693.07
>
> ### Testing empty bulk deq ###
> legacy APIs: SP/SC: bulk (size: 8): 7.00
> legacy APIs: MP/MC: bulk (size: 8): 7.00
>
> ### Testing using two physical cores ###
> legacy APIs: SP/SC: bulk (size: 8): 74.36
> legacy APIs: MP/MC: bulk (size: 8): 110.18
> legacy APIs: SP/SC: bulk (size: 32): 23.04
> legacy APIs: MP/MC: bulk (size: 32): 32.29
>
> ### Testing using all slave nodes ##
> Bulk enq/dequeue count on size 8
> Core [8] count = 293741
> Core [9] count = 293741
> Total count (size: 8): 587482
>
> Bulk enq/dequeue count on size 32
> Core [8] count = 244909
> Core [9] count = 244909
> Total count (size: 32): 1077300
>
> ### Testing single element enq/deq ###
> elem APIs: element size 16B: SP/SC: single: 255.37
> elem APIs: element size 16B: MP/MC: single: 456.68
>
> ### Testing burst enq/deq ###
> elem APIs: element size 16B: SP/SC: burst (size: 8): 291.99
> elem APIs: element size 16B: SP/SC: burst (size: 32): 456.25
> elem APIs: element size 16B: MP/MC: burst (size: 8): 497.77
> elem APIs: element size 16B: MP/MC: burst (size: 32): 680.87
>
> ### Testing bulk enq/deq ###
> elem APIs: element size 16B: SP/SC: bulk (size: 8): 284.40
> elem APIs: element size 16B: SP/SC: bulk (size: 32): 453.17
> elem APIs: element size 16B: MP/MC: bulk (size: 8): 485.77
> elem APIs: element size 16B: MP/MC: bulk (size: 32): 675.08
>
> ### Testing empty bulk deq ###
> elem APIs: element size 16B: SP/SC: bulk (size: 8): 8.00
> elem APIs: element size 16B: MP/MC: bulk (size: 8): 7.00
>
> ### Testing using two physical cores ###
> elem APIs: element size 16B: SP/SC: bulk (size: 8): 74.45
> elem APIs: element size 16B: MP/MC: bulk (size: 8): 105.91
> elem APIs: element size 16B: SP/SC: bulk (size: 32): 22.92
> elem APIs: element size 16B: MP/MC: bulk (size: 32): 31.55
>
> ### Testing using all slave nodes ###
>
> Bulk enq/dequeue count on size 8
> Core [8] count = 308724
> Core [9] count = 308723
> Total count (size: 8): 617447
>
> Bulk enq/dequeue count on size 32
> Core [8] count = 214269
> Core [9] count = 214269
> Total count (size: 32): 1045985
>
> RTE>>ring_perf_autotest // with patch
>
> ### Testing single element enq/deq ###
> legacy APIs: SP/SC: single: 289.78
> legacy APIs: MP/MC: single: 475.76
>
> ### Testing burst enq/deq ###
> legacy APIs: SP/SC: burst (size: 8): 323.91
> legacy APIs: SP/SC: burst (size: 32): 424.60
> legacy APIs: MP/MC: burst (size: 8): 523.00
> legacy APIs: MP/MC: burst (size: 32): 717.09
>
> ### Testing bulk enq/deq ###
> legacy APIs: SP/SC: bulk (size: 8): 317.74
> legacy APIs: SP/SC: bulk (size: 32): 413.57
> legacy APIs: MP/MC: bulk (size: 8): 512.89
> legacy APIs: MP/MC: bulk (size: 32): 712.45
>
> ### Testing empty bulk deq ###
> legacy APIs: SP/SC: bulk (size: 8): 7.00
> legacy APIs: MP/MC: bulk (size: 8): 7.00
>
> ### Testing using two physical cores ###
> legacy APIs: SP/SC: bulk (size: 8): 74.82
> legacy APIs: MP/MC: bulk (size: 8): 96.45
> legacy APIs: SP/SC: bulk (size: 32): 22.97
> legacy APIs: MP/MC: bulk (size: 32): 32.52
>
> ### Testing using all slave nodes ###
>
> Bulk enq/dequeue count on size 8
> Core [8] count = 283928
> Core [9] count = 283927
> Total count (size: 8): 567855
>
> Bulk enq/dequeue count on size 32
> Core [8] count = 223916
> Core [9] count = 223915
> Total count (size: 32): 1015686
>
> ### Testing single element enq/deq ###
> elem APIs: element size 16B: SP/SC: single: 267.65
> elem APIs: element size 16B: MP/MC: single: 439.06
>
> ### Testing burst enq/deq ###
> elem APIs: element size 16B: SP/SC: burst (size: 8): 302.44
> elem APIs: element size 16B: SP/SC: burst (size: 32): 466.31
> elem APIs: element size 16B: MP/MC: burst (size: 8): 502.51
> elem APIs: element size 16B: MP/MC: burst (size: 32): 695.81
>
> ### Testing bulk enq/deq ###
> elem APIs: element size 16B: SP/SC: bulk (size: 8): 295.15
> elem APIs: element size 16B: SP/SC: bulk (size: 32): 462.77
> elem APIs: element size 16B: MP/MC: bulk (size: 8): 496.89
> elem APIs: element size 16B: MP/MC: bulk (size: 32): 690.46
>
> ### Testing empty bulk deq ###
> elem APIs: element size 16B: SP/SC: bulk (size: 8): 7.50
> elem APIs: element size 16B: MP/MC: bulk (size: 8): 7.44
>
> ### Testing using two physical cores ###
> elem APIs: element size 16B: SP/SC: bulk (size: 8): 65.85
> elem APIs: element size 16B: MP/MC: bulk (size: 8): 103.80
> elem APIs: element size 16B: SP/SC: bulk (size: 32): 23.27
> elem APIs: element size 16B: MP/MC: bulk (size: 32): 31.17
>
> ### Testing using all slave nodes ###
>
> Bulk enq/dequeue count on size 8
> Core [8] count = 304223
> Core [9] count = 304221
> Total count (size: 8): 608444
>
> Bulk enq/dequeue count on size 32
> Core [8] count = 214856
> Core [9] count = 214855
> Total count (size: 32): 1038155
> Test OK
> RTE>>quit
>
>
Encountered a couple of different build errors with these patches on my
Power 9 system:
In file included from ../lib/librte_ring/rte_ring.h:534,
from ../drivers/mempool/ring/rte_mempool_ring.c:9:
../lib/librte_ring/rte_ring_hts_generic.h: In function
‘__rte_ring_hts_update_tail’:
../lib/librte_ring/rte_ring_hts_generic.h:61:2: warning: implicit
declaration of function ‘RTE_ASSERT’; did you mean ‘RTE_STR’?
[-Wimplicit-function-declaration]
RTE_ASSERT(n >= num);
^~~~~~~~~~
RTE_STR
Fixed by adding "#include <rte_debug.h>" to rte_ring.h.
Also encountered:
In file included from ../app/test/test_ring_hts_stress.c:5:
../app/test/test_ring_stress.h: In function ‘check_updt_elem’:
../app/test/test_ring_stress.h:162:9: error: unknown type name
‘rte_spinlock_t’
static rte_spinlock_t dump_lock;
^~~~~~~~~~~~~~
../app/test/test_ring_stress.h:166:4: warning: implicit declaration of
function ‘rte_spinlock_lock’; did you mean ‘rte_calloc_socket’?
[-Wimplicit-function-declaration]
rte_spinlock_lock(&dump_lock);
^~~~~~~~~~~~~~~~~
rte_calloc_socket
../app/test/test_ring_stress.h:166:4: warning: nested extern declaration
of ‘rte_spinlock_lock’ [-Wnested-externs]
../app/test/test_ring_stress.h:172:4: warning: implicit declaration of
function ‘rte_spinlock_unlock’; did you mean ‘pthread_rwlock_unlock’?
[-Wimplicit-function-declaration]
rte_spinlock_unlock(&dump_lock);
^~~~~~~~~~~~~~~~~~~
pthread_rwlock_unlock
../app/test/test_ring_stress.h:172:4: warning: nested extern declaration
of ‘rte_spinlock_unlock’ [-Wnested-externs]
Fixed by adding "#include <rte_spinlock.h>" to test_ring_stress.h.
Autoperf test results
---------------------
RTE>>ring_perf_autotest // DPDK 20.02, without patch
### Testing single element enq/deq ###
legacy APIs: SP/SC: single: 42.14
legacy APIs: MP/MC: single: 56.26
### Testing burst enq/deq ###
legacy APIs: SP/SC: burst (size: 8): 43.59
legacy APIs: SP/SC: burst (size: 32): 49.87
legacy APIs: MP/MC: burst (size: 8): 58.43
legacy APIs: MP/MC: burst (size: 32): 65.68
### Testing bulk enq/deq ###
legacy APIs: SP/SC: bulk (size: 8): 43.59
legacy APIs: SP/SC: bulk (size: 32): 49.85
legacy APIs: MP/MC: bulk (size: 8): 58.43
legacy APIs: MP/MC: bulk (size: 32): 65.60
### Testing empty bulk deq ###
legacy APIs: SP/SC: bulk (size: 8): 7.16
legacy APIs: MP/MC: bulk (size: 8): 7.16
### Testing using two hyperthreads ###
legacy APIs: SP/SC: bulk (size: 8): 12.46
legacy APIs: MP/MC: bulk (size: 8): 16.20
legacy APIs: SP/SC: bulk (size: 32): 3.21
legacy APIs: MP/MC: bulk (size: 32): 3.73
### Testing using two physical cores ###
legacy APIs: SP/SC: bulk (size: 8): 33.34
legacy APIs: MP/MC: bulk (size: 8): 37.99
legacy APIs: SP/SC: bulk (size: 32): 10.19
legacy APIs: MP/MC: bulk (size: 32): 11.90
### Testing using two NUMA nodes ###
legacy APIs: SP/SC: bulk (size: 8): 49.50
legacy APIs: MP/MC: bulk (size: 8): 63.65
legacy APIs: SP/SC: bulk (size: 32): 12.49
legacy APIs: MP/MC: bulk (size: 32): 23.53
### Testing using all slave nodes ###
Bulk enq/dequeue count on size 8
Core [4] count = 5604
Core [5] count = 5563
Core [6] count = 5576
Core [7] count = 5630
Core [8] count = 5643
Core [9] count = 5727
Core [10] count = 5698
Core [11] count = 5711
Core [64] count = 5259
Core [65] count = 5322
Core [66] count = 5321
Core [67] count = 5310
Core [68] count = 4350
Core [69] count = 4455
Core [70] count = 4546
Core [71] count = 4475
Total count (size: 8): 84190
Bulk enq/dequeue count on size 32
Core [4] count = 5543
Core [5] count = 5555
Core [6] count = 5596
Core [7] count = 5584
Core [8] count = 5613
Core [9] count = 5686
Core [10] count = 5689
Core [11] count = 5677
Core [64] count = 5228
Core [65] count = 5389
Core [66] count = 5406
Core [67] count = 5359
Core [68] count = 4554
Core [69] count = 4673
Core [70] count = 4675
Core [71] count = 4644
Total count (size: 32): 169061
### Testing single element enq/deq ###
elem APIs: element size 16B: SP/SC: single: 42.84
elem APIs: element size 16B: MP/MC: single: 56.77
### Testing burst enq/deq ###
elem APIs: element size 16B: SP/SC: burst (size: 8): 44.98
elem APIs: element size 16B: SP/SC: burst (size: 32): 59.16
elem APIs: element size 16B: MP/MC: burst (size: 8): 60.58
elem APIs: element size 16B: MP/MC: burst (size: 32): 74.75
### Testing bulk enq/deq ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 45.01
elem APIs: element size 16B: SP/SC: bulk (size: 32): 59.08
elem APIs: element size 16B: MP/MC: bulk (size: 8): 60.58
elem APIs: element size 16B: MP/MC: bulk (size: 32): 74.76
### Testing empty bulk deq ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 7.16
elem APIs: element size 16B: MP/MC: bulk (size: 8): 7.16
### Testing using two hyperthreads ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 12.18
elem APIs: element size 16B: MP/MC: bulk (size: 8): 15.44
elem APIs: element size 16B: SP/SC: bulk (size: 32): 3.22
elem APIs: element size 16B: MP/MC: bulk (size: 32): 3.97
### Testing using two physical cores ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 42.07
elem APIs: element size 16B: MP/MC: bulk (size: 8): 44.50
elem APIs: element size 16B: SP/SC: bulk (size: 32): 10.73
elem APIs: element size 16B: MP/MC: bulk (size: 32): 11.73
### Testing using two NUMA nodes ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 49.55
elem APIs: element size 16B: MP/MC: bulk (size: 8): 93.10
elem APIs: element size 16B: SP/SC: bulk (size: 32): 12.33
elem APIs: element size 16B: MP/MC: bulk (size: 32): 27.10
### Testing using all slave nodes ###
Bulk enq/dequeue count on size 8
Core [4] count = 5489
Core [5] count = 5559
Core [6] count = 5566
Core [7] count = 5577
Core [8] count = 5645
Core [9] count = 5699
Core [10] count = 5695
Core [11] count = 5733
Core [64] count = 5202
Core [65] count = 5284
Core [66] count = 5319
Core [67] count = 5349
Core [68] count = 4331
Core [69] count = 4465
Core [70] count = 4484
Core [71] count = 4439
Total count (size: 8): 83836
Bulk enq/dequeue count on size 32
Core [4] count = 5567
Core [5] count = 5492
Core [6] count = 5517
Core [7] count = 5515
Core [8] count = 5593
Core [9] count = 5650
Core [10] count = 5694
Core [11] count = 5665
Core [64] count = 5236
Core [65] count = 5319
Core [66] count = 5333
Core [67] count = 5304
Core [68] count = 4608
Core [69] count = 4669
Core [70] count = 4690
Core [71] count = 4654
Total count (size: 32): 168342
Test OK
---------------------
RTE>>ring_perf_autotest // DPDK 20.02, without patch
### Testing single element enq/deq ###
legacy APIs: SP/SC: single: 42.18
legacy APIs: MP/MC: single: 56.26
### Testing burst enq/deq ###
legacy APIs: SP/SC: burst (size: 8): 43.60
legacy APIs: SP/SC: burst (size: 32): 49.86
legacy APIs: MP/MC: burst (size: 8): 58.43
legacy APIs: MP/MC: burst (size: 32): 65.67
### Testing bulk enq/deq ###
legacy APIs: SP/SC: bulk (size: 8): 43.59
legacy APIs: SP/SC: bulk (size: 32): 49.86
legacy APIs: MP/MC: bulk (size: 8): 58.43
legacy APIs: MP/MC: bulk (size: 32): 65.63
### Testing empty bulk deq ###
legacy APIs: SP/SC: bulk (size: 8): 7.16
legacy APIs: MP/MC: bulk (size: 8): 7.16
### Testing using two hyperthreads ###
legacy APIs: SP/SC: bulk (size: 8): 12.07
legacy APIs: MP/MC: bulk (size: 8): 16.24
legacy APIs: SP/SC: bulk (size: 32): 3.20
legacy APIs: MP/MC: bulk (size: 32): 3.72
### Testing using two physical cores ###
legacy APIs: SP/SC: bulk (size: 8): 33.41
legacy APIs: MP/MC: bulk (size: 8): 38.01
legacy APIs: SP/SC: bulk (size: 32): 10.23
legacy APIs: MP/MC: bulk (size: 32): 11.90
### Testing using two NUMA nodes ###
legacy APIs: SP/SC: bulk (size: 8): 49.27
legacy APIs: MP/MC: bulk (size: 8): 64.80
legacy APIs: SP/SC: bulk (size: 32): 12.45
legacy APIs: MP/MC: bulk (size: 32): 23.11
### Testing using all slave nodes ###
Bulk enq/dequeue count on size 8
Core [4] count = 5637
Core [5] count = 5599
Core [6] count = 5623
Core [7] count = 5627
Core [8] count = 5723
Core [9] count = 5758
Core [10] count = 5714
Core [11] count = 5724
Core [64] count = 5310
Core [65] count = 5438
Core [66] count = 5448
Core [67] count = 5374
Core [68] count = 4441
Core [69] count = 4550
Core [70] count = 4550
Core [71] count = 4558
Total count (size: 8): 85074
Bulk enq/dequeue count on size 32
Core [4] count = 5608
Core [5] count = 5623
Core [6] count = 5590
Core [7] count = 5658
Core [8] count = 5680
Core [9] count = 5738
Core [10] count = 5692
Core [11] count = 5712
Core [64] count = 5273
Core [65] count = 5363
Core [66] count = 5341
Core [67] count = 5349
Core [68] count = 4591
Core [69] count = 4673
Core [70] count = 4698
Core [71] count = 4687
Total count (size: 32): 170350
### Testing single element enq/deq ###
elem APIs: element size 16B: SP/SC: single: 42.82
elem APIs: element size 16B: MP/MC: single: 56.79
### Testing burst enq/deq ###
elem APIs: element size 16B: SP/SC: burst (size: 8): 44.99
elem APIs: element size 16B: SP/SC: burst (size: 32): 59.00
elem APIs: element size 16B: MP/MC: burst (size: 8): 60.59
elem APIs: element size 16B: MP/MC: burst (size: 32): 74.78
### Testing bulk enq/deq ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 44.97
elem APIs: element size 16B: SP/SC: bulk (size: 32): 58.91
elem APIs: element size 16B: MP/MC: bulk (size: 8): 60.60
elem APIs: element size 16B: MP/MC: bulk (size: 32): 74.61
### Testing empty bulk deq ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 7.16
elem APIs: element size 16B: MP/MC: bulk (size: 8): 7.16
### Testing using two hyperthreads ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 12.18
elem APIs: element size 16B: MP/MC: bulk (size: 8): 15.41
elem APIs: element size 16B: SP/SC: bulk (size: 32): 3.19
elem APIs: element size 16B: MP/MC: bulk (size: 32): 4.06
### Testing using two physical cores ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 42.08
elem APIs: element size 16B: MP/MC: bulk (size: 8): 44.52
elem APIs: element size 16B: SP/SC: bulk (size: 32): 10.73
elem APIs: element size 16B: MP/MC: bulk (size: 32): 12.39
### Testing using two NUMA nodes ###
elem APIs: element size 16B: SP/SC: bulk (size: 8): 49.65
elem APIs: element size 16B: MP/MC: bulk (size: 8): 93.27
elem APIs: element size 16B: SP/SC: bulk (size: 32): 12.38
elem APIs: element size 16B: MP/MC: bulk (size: 32): 27.19
### Testing using all slave nodes ###
Bulk enq/dequeue count on size 8
Core [4] count = 5629
Core [5] count = 5585
Core [6] count = 5676
Core [7] count = 5604
Core [8] count = 5639
Core [9] count = 5731
Core [10] count = 5694
Core [11] count = 5707
Core [64] count = 5254
Core [65] count = 5331
Core [66] count = 5340
Core [67] count = 5355
Core [68] count = 4339
Core [69] count = 4481
Core [70] count = 4504
Core [71] count = 4507
Total count (size: 8): 84376
Bulk enq/dequeue count on size 32
Core [4] count = 5518
Core [5] count = 5493
Core [6] count = 5559
Core [7] count = 5484
Core [8] count = 5623
Core [9] count = 5669
Core [10] count = 5661
Core [11] count = 5658
Core [64] count = 5207
Core [65] count = 5305
Core [66] count = 5273
Core [67] count = 5303
Core [68] count = 4542
Core [69] count = 4682
Core [70] count = 4672
Core [71] count = 4609
Total count (size: 32): 168634
Test OK
Dave
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-28 0:17 ` David Christensen
@ 2020-03-20 16:45 ` Ananyev, Konstantin
0 siblings, 0 replies; 146+ messages in thread
From: Ananyev, Konstantin @ 2020-03-20 16:45 UTC (permalink / raw)
To: David Christensen, Jerin Jacob; +Cc: Stephen Hemminger, dpdk-dev, Olivier Matz
> >
> >
> > I tested on an arm64 HW. The former section is without the
> > patch(20.02) and later one with this patch.
> > I agree with Konstantin that getting more platform tests will be good
> > early so that we can focus on the approach
> > to avoid back and forth latter.
> >
> >
> > RTE>>ring_perf_autotest // without path
> >
> > ### Testing single element enq/deq ###
> > legacy APIs: SP/SC: single: 289.78
> > legacy APIs: MP/MC: single: 516.20
> >
> > ### Testing burst enq/deq ###
> > legacy APIs: SP/SC: burst (size: 8): 312.88
> > legacy APIs: SP/SC: burst (size: 32): 426.72
> > legacy APIs: MP/MC: burst (size: 8): 510.95
> > legacy APIs: MP/MC: burst (size: 32): 702.01
> >
> > ### Testing bulk enq/deq ###
> > legacy APIs: SP/SC: bulk (size: 8): 306.74
> > legacy APIs: SP/SC: bulk (size: 32): 411.56
> > legacy APIs: MP/MC: bulk (size: 8): 501.32
> > legacy APIs: MP/MC: bulk (size: 32): 693.07
> >
> > ### Testing empty bulk deq ###
> > legacy APIs: SP/SC: bulk (size: 8): 7.00
> > legacy APIs: MP/MC: bulk (size: 8): 7.00
> >
> > ### Testing using two physical cores ###
> > legacy APIs: SP/SC: bulk (size: 8): 74.36
> > legacy APIs: MP/MC: bulk (size: 8): 110.18
> > legacy APIs: SP/SC: bulk (size: 32): 23.04
> > legacy APIs: MP/MC: bulk (size: 32): 32.29
> >
> > ### Testing using all slave nodes ##
> > Bulk enq/dequeue count on size 8
> > Core [8] count = 293741
> > Core [9] count = 293741
> > Total count (size: 8): 587482
> >
> > Bulk enq/dequeue count on size 32
> > Core [8] count = 244909
> > Core [9] count = 244909
> > Total count (size: 32): 1077300
> >
> > ### Testing single element enq/deq ###
> > elem APIs: element size 16B: SP/SC: single: 255.37
> > elem APIs: element size 16B: MP/MC: single: 456.68
> >
> > ### Testing burst enq/deq ###
> > elem APIs: element size 16B: SP/SC: burst (size: 8): 291.99
> > elem APIs: element size 16B: SP/SC: burst (size: 32): 456.25
> > elem APIs: element size 16B: MP/MC: burst (size: 8): 497.77
> > elem APIs: element size 16B: MP/MC: burst (size: 32): 680.87
> >
> > ### Testing bulk enq/deq ###
> > elem APIs: element size 16B: SP/SC: bulk (size: 8): 284.40
> > elem APIs: element size 16B: SP/SC: bulk (size: 32): 453.17
> > elem APIs: element size 16B: MP/MC: bulk (size: 8): 485.77
> > elem APIs: element size 16B: MP/MC: bulk (size: 32): 675.08
> >
> > ### Testing empty bulk deq ###
> > elem APIs: element size 16B: SP/SC: bulk (size: 8): 8.00
> > elem APIs: element size 16B: MP/MC: bulk (size: 8): 7.00
> >
> > ### Testing using two physical cores ###
> > elem APIs: element size 16B: SP/SC: bulk (size: 8): 74.45
> > elem APIs: element size 16B: MP/MC: bulk (size: 8): 105.91
> > elem APIs: element size 16B: SP/SC: bulk (size: 32): 22.92
> > elem APIs: element size 16B: MP/MC: bulk (size: 32): 31.55
> >
> > ### Testing using all slave nodes ###
> >
> > Bulk enq/dequeue count on size 8
> > Core [8] count = 308724
> > Core [9] count = 308723
> > Total count (size: 8): 617447
> >
> > Bulk enq/dequeue count on size 32
> > Core [8] count = 214269
> > Core [9] count = 214269
> > Total count (size: 32): 1045985
> >
> > RTE>>ring_perf_autotest // with patch
> >
> > ### Testing single element enq/deq ###
> > legacy APIs: SP/SC: single: 289.78
> > legacy APIs: MP/MC: single: 475.76
> >
> > ### Testing burst enq/deq ###
> > legacy APIs: SP/SC: burst (size: 8): 323.91
> > legacy APIs: SP/SC: burst (size: 32): 424.60
> > legacy APIs: MP/MC: burst (size: 8): 523.00
> > legacy APIs: MP/MC: burst (size: 32): 717.09
> >
> > ### Testing bulk enq/deq ###
> > legacy APIs: SP/SC: bulk (size: 8): 317.74
> > legacy APIs: SP/SC: bulk (size: 32): 413.57
> > legacy APIs: MP/MC: bulk (size: 8): 512.89
> > legacy APIs: MP/MC: bulk (size: 32): 712.45
> >
> > ### Testing empty bulk deq ###
> > legacy APIs: SP/SC: bulk (size: 8): 7.00
> > legacy APIs: MP/MC: bulk (size: 8): 7.00
> >
> > ### Testing using two physical cores ###
> > legacy APIs: SP/SC: bulk (size: 8): 74.82
> > legacy APIs: MP/MC: bulk (size: 8): 96.45
> > legacy APIs: SP/SC: bulk (size: 32): 22.97
> > legacy APIs: MP/MC: bulk (size: 32): 32.52
> >
> > ### Testing using all slave nodes ###
> >
> > Bulk enq/dequeue count on size 8
> > Core [8] count = 283928
> > Core [9] count = 283927
> > Total count (size: 8): 567855
> >
> > Bulk enq/dequeue count on size 32
> > Core [8] count = 223916
> > Core [9] count = 223915
> > Total count (size: 32): 1015686
> >
> > ### Testing single element enq/deq ###
> > elem APIs: element size 16B: SP/SC: single: 267.65
> > elem APIs: element size 16B: MP/MC: single: 439.06
> >
> > ### Testing burst enq/deq ###
> > elem APIs: element size 16B: SP/SC: burst (size: 8): 302.44
> > elem APIs: element size 16B: SP/SC: burst (size: 32): 466.31
> > elem APIs: element size 16B: MP/MC: burst (size: 8): 502.51
> > elem APIs: element size 16B: MP/MC: burst (size: 32): 695.81
> >
> > ### Testing bulk enq/deq ###
> > elem APIs: element size 16B: SP/SC: bulk (size: 8): 295.15
> > elem APIs: element size 16B: SP/SC: bulk (size: 32): 462.77
> > elem APIs: element size 16B: MP/MC: bulk (size: 8): 496.89
> > elem APIs: element size 16B: MP/MC: bulk (size: 32): 690.46
> >
> > ### Testing empty bulk deq ###
> > elem APIs: element size 16B: SP/SC: bulk (size: 8): 7.50
> > elem APIs: element size 16B: MP/MC: bulk (size: 8): 7.44
> >
> > ### Testing using two physical cores ###
> > elem APIs: element size 16B: SP/SC: bulk (size: 8): 65.85
> > elem APIs: element size 16B: MP/MC: bulk (size: 8): 103.80
> > elem APIs: element size 16B: SP/SC: bulk (size: 32): 23.27
> > elem APIs: element size 16B: MP/MC: bulk (size: 32): 31.17
> >
> > ### Testing using all slave nodes ###
> >
> > Bulk enq/dequeue count on size 8
> > Core [8] count = 304223
> > Core [9] count = 304221
> > Total count (size: 8): 608444
> >
> > Bulk enq/dequeue count on size 32
> > Core [8] count = 214856
> > Core [9] count = 214855
> > Total count (size: 32): 1038155
> > Test OK
> > RTE>>quit
> >
> >
>
> Encountered a couple of different build errors with these patches on my
> Power 9 system:
>
> In file included from ../lib/librte_ring/rte_ring.h:534,
> from ../drivers/mempool/ring/rte_mempool_ring.c:9:
> ../lib/librte_ring/rte_ring_hts_generic.h: In function
> ‘__rte_ring_hts_update_tail’:
> ../lib/librte_ring/rte_ring_hts_generic.h:61:2: warning: implicit
> declaration of function ‘RTE_ASSERT’; did you mean ‘RTE_STR’?
> [-Wimplicit-function-declaration]
> RTE_ASSERT(n >= num);
> ^~~~~~~~~~
> RTE_STR
>
> Fixed by adding "#include <rte_debug.h>" to rte_ring.h.
>
> Also encountered:
>
> In file included from ../app/test/test_ring_hts_stress.c:5:
> ../app/test/test_ring_stress.h: In function ‘check_updt_elem’:
> ../app/test/test_ring_stress.h:162:9: error: unknown type name
> ‘rte_spinlock_t’
> static rte_spinlock_t dump_lock;
> ^~~~~~~~~~~~~~
> ../app/test/test_ring_stress.h:166:4: warning: implicit declaration of
> function ‘rte_spinlock_lock’; did you mean ‘rte_calloc_socket’?
> [-Wimplicit-function-declaration]
> rte_spinlock_lock(&dump_lock);
> ^~~~~~~~~~~~~~~~~
> rte_calloc_socket
> ../app/test/test_ring_stress.h:166:4: warning: nested extern declaration
> of ‘rte_spinlock_lock’ [-Wnested-externs]
> ../app/test/test_ring_stress.h:172:4: warning: implicit declaration of
> function ‘rte_spinlock_unlock’; did you mean ‘pthread_rwlock_unlock’?
> [-Wimplicit-function-declaration]
> rte_spinlock_unlock(&dump_lock);
> ^~~~~~~~~~~~~~~~~~~
> pthread_rwlock_unlock
> ../app/test/test_ring_stress.h:172:4: warning: nested extern declaration
> of ‘rte_spinlock_unlock’ [-Wnested-externs]
>
> Fixed by adding "#include <rte_spinlock.h>" to test_ring_stress.h.
Thanks a lot for trying it guys.
About compilation issues - will to address it in V1.
Konstantin
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-02-24 11:35 [dpdk-dev] [RFC 0/6] New sync modes for ring Konstantin Ananyev
` (6 preceding siblings ...)
2020-02-24 16:59 ` [dpdk-dev] [RFC 0/6] New sync modes for ring Stephen Hemminger
@ 2020-03-25 20:43 ` Honnappa Nagarahalli
2020-03-26 1:50 ` Ananyev, Konstantin
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
8 siblings, 1 reply; 146+ messages in thread
From: Honnappa Nagarahalli @ 2020-03-25 20:43 UTC (permalink / raw)
To: Konstantin Ananyev, dev; +Cc: olivier.matz, nd, Honnappa Nagarahalli, nd
<snip>
> Subject: [dpdk-dev] [RFC 0/6] New sync modes for ring
>
> Upfront note - that RFC is not a complete patch.
> It introduces an ABI breakage, plus it doesn't update ring_elem code properly,
As per the current rules, these changes (in the current form) will be accepted only for 20.11 release. How do we address this for immediate requirements like RCU defer APIs?
I suggest that we move forward with my RFC (taking into consideration your feedback) to make progress on RCU APIs.
> etc.
> I plan to deal with all these things in later versions.
> Right now I seek an initial feedback about proposed ideas.
> Would also ask people to repeat performance tests (see below) on their
> platforms to confirm the impact.
>
> More and more customers use(/try to use) DPDK based apps within
> overcommitted systems (multiple acttive threads over same pysical cores):
> VM, container deployments, etc.
> One quite common problem they hit: Lock-Holder-Preemption with rte_ring.
> LHP is quite a common problem for spin-based sync primitives (spin-locks, etc.)
> on overcommitted systems.
> The situation gets much worse when some sort of fair-locking technique is
> used (ticket-lock, etc.).
> As now not only lock-owner but also lock-waiters scheduling order matters a
> lot.
> This is a well-known problem for kernel within VMs:
> http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
> The problem with rte_ring is that while head accusion is sort of un-fair locking,
> waiting on tail is very similar to ticket lock schema - tail has to be updated in
> particular order.
> That makes current rte_ring implementation to perform really pure on some
> overcommited scenarios.
> While it is probably not possible to completely resolve this problem in
> userspace only (without some kernel communication/intervention), removing
> fairness in tail update can mitigate it significantly.
> So this RFC proposes two new optional ring synchronization modes:
> 1) Head/Tail Sync (HTS) mode
> In that mode enqueue/dequeue operation is fully serialized:
> only one thread at a time is allowed to perform given op.
> As another enhancement provide ability to split enqueue/dequeue
> operation into two phases:
> - enqueue/dequeue start
> - enqueue/dequeue finish
> That allows user to inspect objects in the ring without removing
> them from it (aka MT safe peek).
IMO, this will not address the problem described above. For ex: when a producer updates the head and gets scheduled out, other producers have to spin. The problem is probably worse as with non-HTS case moving of the head and copying of the ring elements can happen in parallel between the producers (similarly for consumers).
IMO, HTS should not be a configurable flag. In RCU requirement, a MP enqueue and HTS dequeue are required.
> 2) Relaxed Tail Sync (RTS)
> The main difference from original MP/MC algorithm is that tail value is
> increased not by every thread that finished enqueue/dequeue, but only by the
> last one.
> That allows threads to avoid spinning on ring tail value, leaving actual tail value
> change to the last thread in the update queue.
This can be a configurable flag on the ring.
I am not sure how this solves the problem you have stated above completely. Updating the count from all intermediate threads is still required to update the value of the head. But yes, it reduces the severity of the problem by not enforcing the order in which the tail is updated.
I also think it introduces the problem on the other side of the ring because the tail is not updated soon enough (the other side has to wait longer for the elements to become available). It also introduces another configuration parameter (HTD_MAX_DEF) which they have to deal with.
Users have to still implement the current hypervisor related solutions.
IMO, we should run the benchmark for this on an over committed setup to understand the benefits.
>
> Test results on IA (see below) show significant improvements for average
> enqueue/dequeue op times on overcommitted systems.
> For 'classic' DPDK deployments (one thread per core) original MP/MC
> algorithm still shows best numbers, though for 64-bit target RTS numbers are
> not that far away.
> Numbers were produced by ring_stress_*autotest (first patch in these series).
>
> X86_64 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
> DEQ+ENQ average cycles/obj
>
> MP/MC HTS RTS
> 1thread@1core(--lcores=6-7) 8.00 8.15 8.99
> 2thread@2core(--lcores=6-8) 19.14 19.61 20.35
> 4thread@4core(--lcores=6-10) 29.43 29.79 31.82
> 8thread@8core(--lcores=6-14) 110.59 192.81 119.50
> 16thread@16core(--lcores=6-22) 461.03 813.12 495.59
> 32thread/@32core(--lcores='6-22,55-70') 982.90 1972.38 1160.51
>
> 2thread@1core(--lcores='6,(10-11)@7' 20140.50 23.58 25.14
> 4thread@2core(--lcores='6,(10-11)@7,(20-21)@8' 153680.60 76.88 80.05
> 8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 280314.32 294.72 318.79
> 16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 643176.59 1144.02
> 1175.14 32thread@2core(--lcores='6,(10-25)@7,(30-45)@8' 4264238.80
> 4627.48 4892.68
>
> 8thread@2core(--lcores='6,(10-17)@(7,8))' 321085.98 298.59 307.47
> 16thread@4core(--lcores='6,(20-35)@(7-10))' 1900705.61 575.35 678.29
> 32thread@4core(--lcores='6,(20-51)@(7-10))' 5510445.85 2164.36 2714.12
>
> i686 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
> DEQ+ENQ average cycles/obj
>
> MP/MC HTS RTS
> 1thread@1core(--lcores=6-7) 7.85 12.13 11.31
> 2thread@2core(--lcores=6-8) 17.89 24.52 21.86
> 8thread@8core(--lcores=6-14) 32.58 354.20 54.58
> 32thread/@32core(--lcores='6-22,55-70') 813.77 6072.41 2169.91
>
> 2thread@1core(--lcores='6,(10-11)@7' 16095.00 36.06 34.74
> 8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 1140354.54 346.61 361.57
> 16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 1920417.86 1314.90
> 1416.65
>
> 8thread@2core(--lcores='6,(10-17)@(7,8))' 594358.61 332.70 357.74
> 32thread@4core(--lcores='6,(20-51)@(7-10))' 5319896.86 2836.44 3028.87
>
> Konstantin Ananyev (6):
> test/ring: add contention stress test
> ring: rework ring layout to allow new sync schemes
> ring: introduce RTS ring mode
> test/ring: add contention stress test for RTS ring
> ring: introduce HTS ring mode
> test/ring: add contention stress test for HTS ring
>
> app/test/Makefile | 3 +
> app/test/meson.build | 3 +
> app/test/test_pdump.c | 6 +-
> app/test/test_ring_hts_stress.c | 28 ++
> app/test/test_ring_rts_stress.c | 28 ++
> app/test/test_ring_stress.c | 27 ++
> app/test/test_ring_stress.h | 477 +++++++++++++++++++
> lib/librte_pdump/rte_pdump.c | 2 +-
> lib/librte_port/rte_port_ring.c | 12 +-
> lib/librte_ring/Makefile | 4 +-
> lib/librte_ring/meson.build | 4 +-
> lib/librte_ring/rte_ring.c | 84 +++-
> lib/librte_ring/rte_ring.h | 619 +++++++++++++++++++++++--
> lib/librte_ring/rte_ring_elem.h | 8 +-
> lib/librte_ring/rte_ring_hts_generic.h | 228 +++++++++
> lib/librte_ring/rte_ring_rts_generic.h | 240 ++++++++++
> 16 files changed, 1721 insertions(+), 52 deletions(-) create mode 100644
> app/test/test_ring_hts_stress.c create mode 100644
> app/test/test_ring_rts_stress.c create mode 100644
> app/test/test_ring_stress.c create mode 100644 app/test/test_ring_stress.h
> create mode 100644 lib/librte_ring/rte_ring_hts_generic.h
> create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
>
> --
> 2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 5/6] ring: introduce HTS ring mode
2020-02-24 11:35 ` [dpdk-dev] [RFC 5/6] ring: introduce HTS ring mode Konstantin Ananyev
@ 2020-03-25 20:44 ` Honnappa Nagarahalli
2020-03-26 12:26 ` Ananyev, Konstantin
0 siblings, 1 reply; 146+ messages in thread
From: Honnappa Nagarahalli @ 2020-03-25 20:44 UTC (permalink / raw)
To: Konstantin Ananyev, dev; +Cc: olivier.matz, nd, Honnappa Nagarahalli, nd
<snip>
>
> Introduce head/tail sync mode for MT ring synchronization.
> In that mode enqueue/dequeue operation is fully serialized:
> only one thread at a time is allowed to perform given op.
> Suppose to reduce stall times in case when ring is used on overcommitted
> cpus (multiple active threads on the same cpu).
> As another enhancement provide ability to split enqueue/dequeue operation
> into two phases:
> - enqueue/dequeue start
> - enqueue/dequeue finish
> That allows user to inspect objects in the ring without removing them from it
> (aka MT safe peek).
>
> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> ---
> lib/librte_ring/Makefile | 1 +
> lib/librte_ring/meson.build | 1 +
> lib/librte_ring/rte_ring.c | 15 +-
> lib/librte_ring/rte_ring.h | 259 ++++++++++++++++++++++++-
> lib/librte_ring/rte_ring_hts_generic.h | 228 ++++++++++++++++++++++
> 5 files changed, 500 insertions(+), 4 deletions(-) create mode 100644
> lib/librte_ring/rte_ring_hts_generic.h
>
> diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile index
> 4f90344f4..0c7f8f918 100644
> --- a/lib/librte_ring/Makefile
> +++ b/lib/librte_ring/Makefile
> @@ -19,6 +19,7 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include :=
> rte_ring.h \
> rte_ring_elem.h \
> rte_ring_generic.h \
> rte_ring_c11_mem.h \
> + rte_ring_hts_generic.h \
> rte_ring_rts_generic.h
>
> include $(RTE_SDK)/mk/rte.lib.mk
> diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build index
> dc8d7dbea..5aa673199 100644
> --- a/lib/librte_ring/meson.build
> +++ b/lib/librte_ring/meson.build
> @@ -6,6 +6,7 @@ headers = files('rte_ring.h',
> 'rte_ring_elem.h',
> 'rte_ring_c11_mem.h',
> 'rte_ring_generic.h',
> + 'rte_ring_hts_generic.h',
> 'rte_ring_rts_generic.h')
>
> # rte_ring_create_elem and rte_ring_get_memsize_elem are experimental
> diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c index
> 1ce0af3e5..d3b948667 100644
> --- a/lib/librte_ring/rte_ring.c
> +++ b/lib/librte_ring/rte_ring.c
> @@ -102,9 +102,9 @@ static int
> get_sync_type(uint32_t flags, uint32_t *prod_st, uint32_t *cons_st) {
> static const uint32_t prod_st_flags =
> - (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
> + (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ |
> RING_F_MP_HTS_ENQ);
> static const uint32_t cons_st_flags =
> - (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
> + (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ |
> RING_F_MC_HTS_DEQ);
>
> switch (flags & prod_st_flags) {
> case 0:
> @@ -116,6 +116,9 @@ get_sync_type(uint32_t flags, uint32_t *prod_st,
> uint32_t *cons_st)
> case RING_F_MP_RTS_ENQ:
> *prod_st = RTE_RING_SYNC_MT_RTS;
> break;
> + case RING_F_MP_HTS_ENQ:
> + *prod_st = RTE_RING_SYNC_MT_HTS;
> + break;
> default:
> return -EINVAL;
> }
> @@ -130,6 +133,9 @@ get_sync_type(uint32_t flags, uint32_t *prod_st,
> uint32_t *cons_st)
> case RING_F_MC_RTS_DEQ:
> *cons_st = RTE_RING_SYNC_MT_RTS;
> break;
> + case RING_F_MC_HTS_DEQ:
> + *cons_st = RTE_RING_SYNC_MT_HTS;
> + break;
> default:
> return -EINVAL;
> }
> @@ -151,6 +157,11 @@ rte_ring_init(struct rte_ring *r, const char *name,
> unsigned count,
> RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
> RTE_CACHE_LINE_MASK) != 0);
>
> + RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
> + offsetof(struct rte_ring_hts_headtail, sync_type));
> + RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
> + offsetof(struct rte_ring_hts_headtail, ht.pos.tail));
> +
> RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
> offsetof(struct rte_ring_rts_headtail, sync_type));
> RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) != diff --git
> a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h index
> a130aeb9d..52edcea11 100644
> --- a/lib/librte_ring/rte_ring.h
> +++ b/lib/librte_ring/rte_ring.h
> @@ -66,11 +66,11 @@ enum {
> RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
> RTE_RING_SYNC_ST, /**< single thread only */
> RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
> + RTE_RING_SYNC_MT_HTS, /**< multi-thread head/tail sync */
> };
>
> /**
> - * structure to hold a pair of head/tail values and other metadata.
> - * used by RTE_RING_SYNC_MT, RTE_RING_SYNC_ST sync types.
> + * Structure to hold a pair of head/tail values and other metadata.
> * Depending on sync_type format of that structure might differ
> * depending on the sync mechanism selelcted, but offsets for
> * *sync_type* and *tail* values should always remain the same.
> @@ -96,6 +96,19 @@ struct rte_ring_rts_headtail {
> volatile union rte_ring_ht_poscnt head; };
>
> +union rte_ring_ht_pos {
> + uint64_t raw;
> + struct {
> + uint32_t tail; /**< tail position */
> + uint32_t head; /**< head position */
> + } pos;
> +};
> +
> +struct rte_ring_hts_headtail {
> + uint32_t sync_type; /**< sync type of prod/cons */
> + volatile union rte_ring_ht_pos ht __rte_aligned(8); };
> +
> /**
> * An RTE ring structure.
> *
> @@ -126,6 +139,7 @@ struct rte_ring {
> RTE_STD_C11
> union {
> struct rte_ring_headtail prod;
> + struct rte_ring_hts_headtail hts_prod;
> struct rte_ring_rts_headtail rts_prod;
> } __rte_cache_aligned;
>
> @@ -135,6 +149,7 @@ struct rte_ring {
> RTE_STD_C11
> union {
> struct rte_ring_headtail cons;
> + struct rte_ring_hts_headtail hts_cons;
> struct rte_ring_rts_headtail rts_cons;
> } __rte_cache_aligned;
>
> @@ -157,6 +172,9 @@ struct rte_ring {
> #define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS".
> */ #define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC
> RTS". */
>
> +#define RING_F_MP_HTS_ENQ 0x0020 /**< The default enqueue is "MP
> HTS".
> +*/ #define RING_F_MC_HTS_DEQ 0x0040 /**< The default dequeue is "MC
> +HTS". */
> +
> #define __IS_SP RTE_RING_SYNC_ST
> #define __IS_MP RTE_RING_SYNC_MT
> #define __IS_SC RTE_RING_SYNC_ST
> @@ -513,6 +531,82 @@ __rte_ring_do_rts_dequeue(struct rte_ring *r, void
> **obj_table,
> return n;
> }
>
> +#include <rte_ring_hts_generic.h>
> +
> +/**
> + * @internal Start to enqueue several objects on the HTS ring.
> + * Note that user has to call appropriate enqueue_finish()
> + * to complete given enqueue operation.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to add in the ring from the obj_table.
> + * @param behavior
> + * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
> + * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from
> ring
> + * @param free_space
> + * returns the amount of space after the enqueue operation has finished
> + * @return
> + * Actual number of objects enqueued.
> + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> + */
> +static __rte_always_inline unsigned int
> +__rte_ring_do_hts_enqueue_start(struct rte_ring *r, void * const *obj_table,
> + uint32_t n, enum rte_ring_queue_behavior behavior,
> + uint32_t *free_space)
> +{
> + uint32_t free, head;
> +
> + n = __rte_ring_hts_move_prod_head(r, n, behavior, &head, &free);
> +
> + if (n != 0)
> + ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
> +
> + if (free_space != NULL)
> + *free_space = free - n;
> + return n;
> +}
rte_ring.h is becoming too big. May be we should move these functions to another HTS specific file. But leave the top level API in rte_ring.h. Similarly for RTS.
> +
> +/**
> + * @internal Start to dequeue several objects from the HTS ring.
> + * Note that user has to call appropriate dequeue_finish()
> + * to complete given dequeue operation.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to pull from the ring.
> + * @param behavior
> + * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
> + * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from
> ring
> + * @param available
> + * returns the number of remaining ring entries after the dequeue has
> finished
> + * @return
> + * - Actual number of objects dequeued.
> + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> + */
> +static __rte_always_inline unsigned int
> +__rte_ring_do_hts_dequeue_start(struct rte_ring *r, void **obj_table,
> + unsigned int n, enum rte_ring_queue_behavior behavior,
> + unsigned int *available)
> +{
> + uint32_t entries, head;
> +
> + n = __rte_ring_hts_move_cons_head(r, n, behavior, &head, &entries);
> +
> + if (n != 0)
> + DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
> +
> + if (available != NULL)
> + *available = entries - n;
> + return n;
> +}
> +
> /**
> * Enqueue several objects on the ring (multi-producers safe).
> *
> @@ -585,6 +679,47 @@ rte_ring_rts_enqueue_bulk(struct rte_ring *r, void *
> const *obj_table,
> free_space);
> }
>
> +/**
> + * Start to enqueue several objects on the HTS ring (multi-producers safe).
> + * Note that user has to call appropriate dequeue_finish()
> + * to complete given dequeue operation.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to add in the ring from the obj_table.
> + * @param free_space
> + * if non-NULL, returns the amount of space in the ring after the
> + * enqueue operation has finished.
> + * @return
> + * The number of objects enqueued, either 0 or n
> + */
> +static __rte_always_inline unsigned int
> +rte_ring_hts_enqueue_bulk_start(struct rte_ring *r, void * const *obj_table,
> + unsigned int n, unsigned int *free_space) {
> + return __rte_ring_do_hts_enqueue_start(r, obj_table, n,
> + RTE_RING_QUEUE_FIXED, free_space);
> +}
I do not clearly understand the requirements on the enqueue_start and enqueue_finish in the form they are here.
IMO, only requirement for these APIs is to provide the ability to avoid intermediate memcpys.
> +
> +static __rte_always_inline void
> +rte_ring_hts_enqueue_finish(struct rte_ring *r, unsigned int n) {
> + __rte_ring_hts_update_tail(&r->hts_prod, n, 1); }
> +
> +static __rte_always_inline unsigned int
> +rte_ring_hts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
> + unsigned int n, unsigned int *free_space) {
> + n = rte_ring_hts_enqueue_bulk_start(r, obj_table, n, free_space);
> + if (n != 0)
> + rte_ring_hts_enqueue_finish(r, n);
> + return n;
> +}
> +
> /**
> * Enqueue several objects on a ring.
> *
> @@ -615,6 +750,8 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void * const
> *obj_table,
> return rte_ring_sp_enqueue_bulk(r, obj_table, n, free_space);
> case RTE_RING_SYNC_MT_RTS:
> return rte_ring_rts_enqueue_bulk(r, obj_table, n, free_space);
> + case RTE_RING_SYNC_MT_HTS:
> + return rte_ring_hts_enqueue_bulk(r, obj_table, n,
> free_space);
> }
>
> /* valid ring should never reach this point */ @@ -753,6 +890,47 @@
> rte_ring_rts_dequeue_bulk(struct rte_ring *r, void **obj_table,
> available);
> }
>
> +/**
> + * Start to dequeue several objects from an HTS ring (multi-consumers safe).
> + * Note that user has to call appropriate dequeue_finish()
> + * to complete given dequeue operation.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects) that will be filled.
> + * @param n
> + * The number of objects to dequeue from the ring to the obj_table.
> + * @param available
> + * If non-NULL, returns the number of remaining ring entries after the
> + * dequeue has finished.
> + * @return
> + * The number of objects dequeued, either 0 or n
> + */
> +static __rte_always_inline unsigned int
> +rte_ring_hts_dequeue_bulk_start(struct rte_ring *r, void **obj_table,
> + unsigned int n, unsigned int *available) {
> + return __rte_ring_do_hts_dequeue_start(r, obj_table, n,
> + RTE_RING_QUEUE_FIXED, available);
> +}
IMO, we should look to provide the ability to avoid intermediate copies when the data from the ring needs to be distributed to different locations.
My proposal in its form is complicated. But, I am thinking that, if the return values are abstracted in a structure, it might look much simple.
> +
> +static __rte_always_inline void
> +rte_ring_hts_dequeue_finish(struct rte_ring *r, unsigned int n) {
> + __rte_ring_hts_update_tail(&r->hts_cons, n, 0); }
> +
> +static __rte_always_inline unsigned int
> +rte_ring_hts_dequeue_bulk(struct rte_ring *r, void **obj_table,
> + unsigned int n, unsigned int *available) {
> + n = rte_ring_hts_dequeue_bulk_start(r, obj_table, n, available);
> + if (n != 0)
> + rte_ring_hts_dequeue_finish(r, n);
> + return n;
> +}
> +
> /**
> * Dequeue several objects from a ring.
> *
> @@ -783,6 +961,8 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void
> **obj_table, unsigned int n,
> return rte_ring_sc_dequeue_bulk(r, obj_table, n, available);
> case RTE_RING_SYNC_MT_RTS:
> return rte_ring_rts_dequeue_bulk(r, obj_table, n, available);
> + case RTE_RING_SYNC_MT_HTS:
> + return rte_ring_hts_dequeue_bulk(r, obj_table, n, available);
> }
>
> /* valid ring should never reach this point */ @@ -1111,6 +1291,41
> @@ rte_ring_rts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
> RTE_RING_QUEUE_VARIABLE, free_space); }
>
> +/**
> + * Start to enqueue several objects on the HTS ring (multi-producers safe).
> + * Note that user has to call appropriate dequeue_finish()
> + * to complete given dequeue operation.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to add in the ring from the obj_table.
> + * @param free_space
> + * if non-NULL, returns the amount of space in the ring after the
> + * enqueue operation has finished.
> + * @return
> + * The number of objects enqueued, either 0 or n
> + */
> +static __rte_always_inline unsigned int
> +rte_ring_hts_enqueue_burst_start(struct rte_ring *r, void * const *obj_table,
> + unsigned int n, unsigned int *free_space) {
> + return __rte_ring_do_hts_enqueue_start(r, obj_table, n,
> + RTE_RING_QUEUE_VARIABLE, free_space); }
> +
rte_ring_hts_enqueue_burst_finish is not implemented. It requires the 'n' returned from ' rte_ring_hts_enqueue_burst_start' to be passed. We can't completely avoid passing correct information between xxx_start and xxx_finish APIs.
> +static __rte_always_inline unsigned int
> +rte_ring_hts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
> + unsigned int n, unsigned int *free_space) {
> + n = rte_ring_hts_enqueue_burst_start(r, obj_table, n, free_space);
> + if (n != 0)
> + rte_ring_hts_enqueue_finish(r, n);
> + return n;
> +}
> +
> /**
> * Enqueue several objects on a ring.
> *
> @@ -1141,6 +1356,8 @@ rte_ring_enqueue_burst(struct rte_ring *r, void *
> const *obj_table,
> return rte_ring_sp_enqueue_burst(r, obj_table, n,
> free_space);
> case RTE_RING_SYNC_MT_RTS:
> return rte_ring_rts_enqueue_burst(r, obj_table, n,
> free_space);
> + case RTE_RING_SYNC_MT_HTS:
> + return rte_ring_hts_enqueue_burst(r, obj_table, n,
> free_space);
> }
>
> /* valid ring should never reach this point */ @@ -1225,6 +1442,42
> @@ rte_ring_rts_dequeue_burst(struct rte_ring *r, void **obj_table,
> return __rte_ring_do_rts_dequeue(r, obj_table, n,
> RTE_RING_QUEUE_VARIABLE, available); }
> +
> +/**
> + * Start to dequeue several objects from an HTS ring (multi-consumers safe).
> + * Note that user has to call appropriate dequeue_finish()
> + * to complete given dequeue operation.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects) that will be filled.
> + * @param n
> + * The number of objects to dequeue from the ring to the obj_table.
> + * @param available
> + * If non-NULL, returns the number of remaining ring entries after the
> + * dequeue has finished.
> + * @return
> + * The number of objects dequeued, either 0 or n
> + */
> +static __rte_always_inline unsigned int
> +rte_ring_hts_dequeue_burst_start(struct rte_ring *r, void **obj_table,
> + unsigned int n, unsigned int *available) {
> + return __rte_ring_do_hts_dequeue_start(r, obj_table, n,
> + RTE_RING_QUEUE_VARIABLE, available);
> +}
> +
> +static __rte_always_inline unsigned int
> +rte_ring_hts_dequeue_burst(struct rte_ring *r, void **obj_table,
> + unsigned int n, unsigned int *available) {
> + n = rte_ring_hts_dequeue_burst_start(r, obj_table, n, available);
> + if (n != 0)
> + rte_ring_hts_dequeue_finish(r, n);
> + return n;
> +}
> +
> /**
> * Dequeue multiple objects from a ring up to a maximum number.
> *
> @@ -1255,6 +1508,8 @@ rte_ring_dequeue_burst(struct rte_ring *r, void
> **obj_table,
> return rte_ring_sc_dequeue_burst(r, obj_table, n, available);
> case RTE_RING_SYNC_MT_RTS:
> return rte_ring_rts_dequeue_burst(r, obj_table, n, available);
> + case RTE_RING_SYNC_MT_HTS:
> + return rte_ring_hts_dequeue_burst(r, obj_table, n, available);
> }
>
> /* valid ring should never reach this point */ diff --git
> a/lib/librte_ring/rte_ring_hts_generic.h
> b/lib/librte_ring/rte_ring_hts_generic.h
> new file mode 100644
> index 000000000..7e447e30b
> --- /dev/null
> +++ b/lib/librte_ring/rte_ring_hts_generic.h
> @@ -0,0 +1,228 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + *
> + * Copyright (c) 2010-2020 Intel Corporation
> + * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
> + * All rights reserved.
> + * Derived from FreeBSD's bufring.h
> + * Used as BSD-3 Licensed with permission from Kip Macy.
> + */
> +
> +#ifndef _RTE_RING_HTS_GENERIC_H_
> +#define _RTE_RING_HTS_GENERIC_H_
> +
> +/**
> + * @file rte_ring_hts_generic.h
> + * It is not recommended to include this file directly,
> + * include <rte_ring.h> instead.
> + * Contains internal helper functions for head/tail sync (HTS) ring mode.
> + * In that mode enqueue/dequeue operation is fully serialized:
> + * only one thread at a time is allowed to perform given op.
> + * This is achieved by thread is allowed to proceed with changing
> +head.value
> + * only when head.value == tail.value.
> + * Both head and tail values are updated atomically (as one 64-bit value).
> + * As another enhancement that provides ability to split
> +enqueue/dequeue
> + * operation into two phases:
> + * - enqueue/dequeue start
> + * - enqueue/dequeue finish
> + * That allows user to inspect objects in the ring without removing
> + * them from it (aka MT safe peek).
> + * As an example:
> + * // read 1 elem from the ring:
> + * n = rte_ring_hts_dequeue_bulk_start(ring, &obj, 1, NULL);
> + * if (n != 0) {
> + * //examined object
> + * if (object_examine(obj) == KEEP)
> + * //decided to keep it in the ring.
> + * rte_ring_hts_dequeue_finish(ring, 0);
> + * else
> + * //decided to remove it in the ring.
> + * rte_ring_hts_dequeue_finish(ring, n);
> + * }
> + * Note that between _start_ and _finish_ the ring is sort of locked -
> + * none other thread can proceed with enqueue(/dequeue) operation till
> + * _finish_ will complete.
This means it does not solve the problem for over committed systems. Do you agree?
> + */
> +
> +static __rte_always_inline void
> +__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
> + uint32_t enqueue)
> +{
> + uint32_t n;
> + union rte_ring_ht_pos p;
> +
> + if (enqueue)
> + rte_smp_wmb();
> + else
> + rte_smp_rmb();
> +
> + p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
> +
> + n = p.pos.head - p.pos.tail;
> + RTE_ASSERT(n >= num);
> + RTE_SET_USED(n);
> +
> + p.pos.head = p.pos.tail + num;
> + p.pos.tail = p.pos.head;
> +
> + rte_atomic64_set((rte_atomic64_t *)(uintptr_t)&ht->ht.raw, p.raw); }
> +
> +/**
> + * @internal waits till tail will become equal to head.
> + * Means no writer/reader is active for that ring.
> + * Suppose to work as serialization point.
> + */
> +static __rte_always_inline void
> +__rte_ring_hts_head_wait(const struct rte_ring_hts_headtail *ht,
> + union rte_ring_ht_pos *p)
> +{
> + p->raw = rte_atomic64_read((rte_atomic64_t *)
> + (uintptr_t)&ht->ht.raw);
> +
> + while (p->pos.head != p->pos.tail) {
> + rte_pause();
> + p->raw = rte_atomic64_read((rte_atomic64_t *)
> + (uintptr_t)&ht->ht.raw);
> + }
> +}
> +
> +/**
> + * @internal This function updates the producer head for enqueue
> + *
> + * @param r
> + * A pointer to the ring structure
> + * @param is_sp
> + * Indicates whether multi-producer path is needed or not
> + * @param n
> + * The number of elements we will want to enqueue, i.e. how far should the
> + * head be moved
> + * @param behavior
> + * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
> + * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from
> ring
> + * @param old_head
> + * Returns head value as it was before the move, i.e. where enqueue starts
> + * @param new_head
> + * Returns the current/new head value i.e. where enqueue finishes
> + * @param free_entries
> + * Returns the amount of free space in the ring BEFORE head was moved
> + * @return
> + * Actual number of objects enqueued.
> + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> + */
> +static __rte_always_inline unsigned int
> +__rte_ring_hts_move_prod_head(struct rte_ring *r, unsigned int num,
> + enum rte_ring_queue_behavior behavior, uint32_t *old_head,
> + uint32_t *free_entries)
> +{
> + uint32_t n;
> + union rte_ring_ht_pos np, op;
> +
> + const uint32_t capacity = r->capacity;
> +
> + do {
> + /* Reset n to the initial burst count */
> + n = num;
> +
> + /* wait for tail to be equal to head */
> + __rte_ring_hts_head_wait(&r->hts_prod, &op);
> +
> + /* add rmb barrier to avoid load/load reorder in weak
> + * memory model. It is noop on x86
> + */
> + rte_smp_rmb();
> +
> + /*
> + * The subtraction is done between two unsigned 32bits value
> + * (the result is always modulo 32 bits even if we have
> + * *old_head > cons_tail). So 'free_entries' is always between
> 0
> + * and capacity (which is < size).
> + */
> + *free_entries = capacity + r->cons.tail - op.pos.head;
> +
> + /* check that we have enough room in ring */
> + if (unlikely(n > *free_entries))
> + n = (behavior == RTE_RING_QUEUE_FIXED) ?
> + 0 : *free_entries;
> +
> + if (n == 0)
> + return 0;
> +
> + np.pos.tail = op.pos.tail;
> + np.pos.head = op.pos.head + n;
> +
> + } while (rte_atomic64_cmpset(&r->hts_prod.ht.raw,
> + op.raw, np.raw) == 0);
> +
> + *old_head = op.pos.head;
> + return n;
> +}
> +
> +/**
> + * @internal This function updates the consumer head for dequeue
> + *
> + * @param r
> + * A pointer to the ring structure
> + * @param is_sc
> + * Indicates whether multi-consumer path is needed or not
> + * @param n
> + * The number of elements we will want to enqueue, i.e. how far should the
> + * head be moved
> + * @param behavior
> + * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
> + * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from
> ring
> + * @param old_head
> + * Returns head value as it was before the move, i.e. where dequeue starts
> + * @param new_head
> + * Returns the current/new head value i.e. where dequeue finishes
> + * @param entries
> + * Returns the number of entries in the ring BEFORE head was moved
> + * @return
> + * - Actual number of objects dequeued.
> + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> + */
> +static __rte_always_inline unsigned int
> +__rte_ring_hts_move_cons_head(struct rte_ring *r, unsigned int num,
> + enum rte_ring_queue_behavior behavior, uint32_t *old_head,
> + uint32_t *entries)
> +{
> + uint32_t n;
> + union rte_ring_ht_pos np, op;
> +
> + /* move cons.head atomically */
> + do {
> + /* Restore n as it may change every loop */
> + n = num;
> +
> + /* wait for tail to be equal to head */
> + __rte_ring_hts_head_wait(&r->hts_cons, &op);
> +
> + /* add rmb barrier to avoid load/load reorder in weak
> + * memory model. It is noop on x86
> + */
> + rte_smp_rmb();
> +
> + /* The subtraction is done between two unsigned 32bits value
> + * (the result is always modulo 32 bits even if we have
> + * cons_head > prod_tail). So 'entries' is always between 0
> + * and size(ring)-1.
> + */
> + *entries = r->prod.tail - op.pos.head;
> +
> + /* Set the actual entries for dequeue */
> + if (n > *entries)
> + n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 :
> *entries;
> +
> + if (unlikely(n == 0))
> + return 0;
> +
> + np.pos.tail = op.pos.tail;
> + np.pos.head = op.pos.head + n;
> +
> + } while (rte_atomic64_cmpset(&r->hts_cons.ht.raw,
> + op.raw, np.raw) == 0);
> +
> + *old_head = op.pos.head;
> + return n;
> +}
> +
> +#endif /* _RTE_RING_HTS_GENERIC_H_ */
> --
> 2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-03-25 20:43 ` Honnappa Nagarahalli
@ 2020-03-26 1:50 ` Ananyev, Konstantin
2020-03-30 21:29 ` Honnappa Nagarahalli
0 siblings, 1 reply; 146+ messages in thread
From: Ananyev, Konstantin @ 2020-03-26 1:50 UTC (permalink / raw)
To: Honnappa Nagarahalli, dev; +Cc: olivier.matz, nd, nd
>
> <snip>
>
> > Subject: [dpdk-dev] [RFC 0/6] New sync modes for ring
> >
> > Upfront note - that RFC is not a complete patch.
> > It introduces an ABI breakage, plus it doesn't update ring_elem code properly,
> As per the current rules, these changes (in the current form) will be accepted only for 20.11 release. How do we address this for immediate
> requirements like RCU defer APIs?
I think I found a way to introduce these new modes without API/ABI breakage.
Working on v1 right now. Plan to submit it by end of that week/start of next one.
> I suggest that we move forward with my RFC (taking into consideration your feedback) to make progress on RCU APIs.
>
> > etc.
> > I plan to deal with all these things in later versions.
> > Right now I seek an initial feedback about proposed ideas.
> > Would also ask people to repeat performance tests (see below) on their
> > platforms to confirm the impact.
> >
> > More and more customers use(/try to use) DPDK based apps within
> > overcommitted systems (multiple acttive threads over same pysical cores):
> > VM, container deployments, etc.
> > One quite common problem they hit: Lock-Holder-Preemption with rte_ring.
> > LHP is quite a common problem for spin-based sync primitives (spin-locks, etc.)
> > on overcommitted systems.
> > The situation gets much worse when some sort of fair-locking technique is
> > used (ticket-lock, etc.).
> > As now not only lock-owner but also lock-waiters scheduling order matters a
> > lot.
> > This is a well-known problem for kernel within VMs:
> > http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> > https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
> > The problem with rte_ring is that while head accusion is sort of un-fair locking,
> > waiting on tail is very similar to ticket lock schema - tail has to be updated in
> > particular order.
> > That makes current rte_ring implementation to perform really pure on some
> > overcommited scenarios.
> > While it is probably not possible to completely resolve this problem in
> > userspace only (without some kernel communication/intervention), removing
> > fairness in tail update can mitigate it significantly.
> > So this RFC proposes two new optional ring synchronization modes:
> > 1) Head/Tail Sync (HTS) mode
> > In that mode enqueue/dequeue operation is fully serialized:
> > only one thread at a time is allowed to perform given op.
> > As another enhancement provide ability to split enqueue/dequeue
> > operation into two phases:
> > - enqueue/dequeue start
> > - enqueue/dequeue finish
> > That allows user to inspect objects in the ring without removing
> > them from it (aka MT safe peek).
> IMO, this will not address the problem described above.
It does, please see the results produced by ring_stress_*autotest below.
Let say for test-case: 8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' it shows:
avg number of cycles per object for enqueue /dequeue:
MP/MC: 280314.32
HTS: 294.72
RTS: 318.79
Customer who tried it reported similar level of improvement.
Actually if you have time - would be very interesting to see what numbers will be on ARM boxes.
To reproduce, just:
$cat ring_tests_u4
ring_stress_autotest
ring_stress_hts_autotest
ring_stress_rts_autotest
/app/test/dpdk-test --lcores='6,(10-13)@7,(20-23)@8' -n 4 < ring_tests_u4 2>&1 | tee res1
Then look at the ' AGGREGATE' stats.
Right now it is a bit too verbose, so probably the easiest thing to extract same numbers quickly:
grep 'cycles/obj' res1 | grep 'cycles/obj' | cat -n | awk '{if ($(1)%9==0) print $(NF);}'
280314.32
1057833.55
294.72
480.10
318.79
461.52
First 2 numbers will be for MP/MC, next 2 for HTS, last 2 for RTS.
> For ex: when a producer updates the head and gets scheduled out, other producers
> have to spin.
Sure, as I wrote in original cover letter:
" While it is probably not possible to completely resolve this problem in
userspace only (without some kernel communication/intervention),
removing fairness in tail update can mitigate it significantly."
Results from the ring_stress_*_autotest confirm that.
> The problem is probably worse as with non-HTS case moving of the head and copying of the ring elements can happen in
> parallel between the producers (similarly for consumers).
Yes as we serialize the ring, we remove possibility of simultaneous copy.
That's why for 'normal' cases (one thread per core) original MP/MC is usually faster.
Though on overcommitted scenarios current MP/MC performance degrades dramatically.
The main problem with current MP/MC implementation is in that tail update
have to be done in strict order (sort of fair locking scheme).
Which means that we have much-much worse LHP manifestation,
then when we use unfair schemes.
With serialized ring (HTS) we remove that ordering completely
(same idea as switch from fair to unfair locking for PV spin-locks).
> IMO, HTS should not be a configurable flag.
Why?
> In RCU requirement, a MP enqueue and HTS dequeue are required.
This is supported, user can specify different modes for consumer and producer:
(0 | RING_F_MC_HTS_DEQ).
Then it is up to the user either to call generic rte_ring_enqueue/rte_ring_dequeue,
or specify mode manually by function name:
rte_ring_mp_enqueue_bulk/ rte_ring_hts_dequeue_bulk.
>
> > 2) Relaxed Tail Sync (RTS)
> > The main difference from original MP/MC algorithm is that tail value is
> > increased not by every thread that finished enqueue/dequeue, but only by the
> > last one.
> > That allows threads to avoid spinning on ring tail value, leaving actual tail value
> > change to the last thread in the update queue.
> This can be a configurable flag on the ring.
> I am not sure how this solves the problem you have stated above completely. Updating the count from all intermediate threads is still
> required to update the value of the head. But yes, it reduces the severity of the problem by not enforcing the order in which the tail is
> updated.
As I said above, main source of slowdown here -
that we have to update tail in particular order.
So the main objective (same as for HTS) is to remove
that ordering.
> I also think it introduces the problem on the other side of the ring because the tail is not updated soon enough (the other side has to wait
> longer for the elements to become available).
Yes, producer/consumer starvation.
That's why we need max allowed Head-Tail-Distance (htd_max) -
to limit how far head can go away from tail.
> It also introduces another configuration parameter (HTD_MAX_DEF) which they have to deal
> with.
If user doesn't provide any value, it will be set by default to ring.capacity / 8.
From my measurements works quite well.
Though there possibility for the user to set another value, if needed.
> Users have to still implement the current hypervisor related solutions.
Didn't get what you trying to say with that phrase.
> IMO, we should run the benchmark for this on an over committed setup to understand the benefits.
That's why I created ring_stress_*autotest test-cases and collected numbers provided below.
I suppose they clearly show the problem on overcommitted scenarios,
and how RTS/HTS improve that situation.
Would appreciate if you repeat these tests on your machines.
>
> >
> > Test results on IA (see below) show significant improvements for average
> > enqueue/dequeue op times on overcommitted systems.
> > For 'classic' DPDK deployments (one thread per core) original MP/MC
> > algorithm still shows best numbers, though for 64-bit target RTS numbers are
> > not that far away.
> > Numbers were produced by ring_stress_*autotest (first patch in these series).
> >
> > X86_64 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
> > DEQ+ENQ average cycles/obj
> >
> > MP/MC HTS RTS
> > 1thread@1core(--lcores=6-7) 8.00 8.15 8.99
> > 2thread@2core(--lcores=6-8) 19.14 19.61 20.35
> > 4thread@4core(--lcores=6-10) 29.43 29.79 31.82
> > 8thread@8core(--lcores=6-14) 110.59 192.81 119.50
> > 16thread@16core(--lcores=6-22) 461.03 813.12 495.59
> > 32thread/@32core(--lcores='6-22,55-70') 982.90 1972.38 1160.51
> >
> > 2thread@1core(--lcores='6,(10-11)@7' 20140.50 23.58 25.14
> > 4thread@2core(--lcores='6,(10-11)@7,(20-21)@8' 153680.60 76.88 80.05
> > 8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 280314.32 294.72 318.79
> > 16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 643176.59 1144.02
> > 1175.14 32thread@2core(--lcores='6,(10-25)@7,(30-45)@8' 4264238.80
> > 4627.48 4892.68
> >
> > 8thread@2core(--lcores='6,(10-17)@(7,8))' 321085.98 298.59 307.47
> > 16thread@4core(--lcores='6,(20-35)@(7-10))' 1900705.61 575.35 678.29
> > 32thread@4core(--lcores='6,(20-51)@(7-10))' 5510445.85 2164.36 2714.12
> >
> > i686 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
> > DEQ+ENQ average cycles/obj
> >
> > MP/MC HTS RTS
> > 1thread@1core(--lcores=6-7) 7.85 12.13 11.31
> > 2thread@2core(--lcores=6-8) 17.89 24.52 21.86
> > 8thread@8core(--lcores=6-14) 32.58 354.20 54.58
> > 32thread/@32core(--lcores='6-22,55-70') 813.77 6072.41 2169.91
> >
> > 2thread@1core(--lcores='6,(10-11)@7' 16095.00 36.06 34.74
> > 8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 1140354.54 346.61 361.57
> > 16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 1920417.86 1314.90
> > 1416.65
> >
> > 8thread@2core(--lcores='6,(10-17)@(7,8))' 594358.61 332.70 357.74
> > 32thread@4core(--lcores='6,(20-51)@(7-10))' 5319896.86 2836.44 3028.87
> >
> > Konstantin Ananyev (6):
> > test/ring: add contention stress test
> > ring: rework ring layout to allow new sync schemes
> > ring: introduce RTS ring mode
> > test/ring: add contention stress test for RTS ring
> > ring: introduce HTS ring mode
> > test/ring: add contention stress test for HTS ring
> >
> > app/test/Makefile | 3 +
> > app/test/meson.build | 3 +
> > app/test/test_pdump.c | 6 +-
> > app/test/test_ring_hts_stress.c | 28 ++
> > app/test/test_ring_rts_stress.c | 28 ++
> > app/test/test_ring_stress.c | 27 ++
> > app/test/test_ring_stress.h | 477 +++++++++++++++++++
> > lib/librte_pdump/rte_pdump.c | 2 +-
> > lib/librte_port/rte_port_ring.c | 12 +-
> > lib/librte_ring/Makefile | 4 +-
> > lib/librte_ring/meson.build | 4 +-
> > lib/librte_ring/rte_ring.c | 84 +++-
> > lib/librte_ring/rte_ring.h | 619 +++++++++++++++++++++++--
> > lib/librte_ring/rte_ring_elem.h | 8 +-
> > lib/librte_ring/rte_ring_hts_generic.h | 228 +++++++++
> > lib/librte_ring/rte_ring_rts_generic.h | 240 ++++++++++
> > 16 files changed, 1721 insertions(+), 52 deletions(-) create mode 100644
> > app/test/test_ring_hts_stress.c create mode 100644
> > app/test/test_ring_rts_stress.c create mode 100644
> > app/test/test_ring_stress.c create mode 100644 app/test/test_ring_stress.h
> > create mode 100644 lib/librte_ring/rte_ring_hts_generic.h
> > create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
> >
> > --
> > 2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 5/6] ring: introduce HTS ring mode
2020-03-25 20:44 ` Honnappa Nagarahalli
@ 2020-03-26 12:26 ` Ananyev, Konstantin
0 siblings, 0 replies; 146+ messages in thread
From: Ananyev, Konstantin @ 2020-03-26 12:26 UTC (permalink / raw)
To: Honnappa Nagarahalli, dev; +Cc: olivier.matz, nd, nd
> > Introduce head/tail sync mode for MT ring synchronization.
> > In that mode enqueue/dequeue operation is fully serialized:
> > only one thread at a time is allowed to perform given op.
> > Suppose to reduce stall times in case when ring is used on overcommitted
> > cpus (multiple active threads on the same cpu).
> > As another enhancement provide ability to split enqueue/dequeue operation
> > into two phases:
> > - enqueue/dequeue start
> > - enqueue/dequeue finish
> > That allows user to inspect objects in the ring without removing them from it
> > (aka MT safe peek).
> >
> > Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> > ---
> > lib/librte_ring/Makefile | 1 +
> > lib/librte_ring/meson.build | 1 +
> > lib/librte_ring/rte_ring.c | 15 +-
> > lib/librte_ring/rte_ring.h | 259 ++++++++++++++++++++++++-
> > lib/librte_ring/rte_ring_hts_generic.h | 228 ++++++++++++++++++++++
> > 5 files changed, 500 insertions(+), 4 deletions(-) create mode 100644
> > lib/librte_ring/rte_ring_hts_generic.h
> >
> > diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile index
> > 4f90344f4..0c7f8f918 100644
> > --- a/lib/librte_ring/Makefile
> > +++ b/lib/librte_ring/Makefile
> > @@ -19,6 +19,7 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include :=
> > rte_ring.h \
> > rte_ring_elem.h \
> > rte_ring_generic.h \
> > rte_ring_c11_mem.h \
> > + rte_ring_hts_generic.h \
> > rte_ring_rts_generic.h
> >
> > include $(RTE_SDK)/mk/rte.lib.mk
> > diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build index
> > dc8d7dbea..5aa673199 100644
> > --- a/lib/librte_ring/meson.build
> > +++ b/lib/librte_ring/meson.build
> > @@ -6,6 +6,7 @@ headers = files('rte_ring.h',
> > 'rte_ring_elem.h',
> > 'rte_ring_c11_mem.h',
> > 'rte_ring_generic.h',
> > + 'rte_ring_hts_generic.h',
> > 'rte_ring_rts_generic.h')
> >
> > # rte_ring_create_elem and rte_ring_get_memsize_elem are experimental
> > diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c index
> > 1ce0af3e5..d3b948667 100644
> > --- a/lib/librte_ring/rte_ring.c
> > +++ b/lib/librte_ring/rte_ring.c
> > @@ -102,9 +102,9 @@ static int
> > get_sync_type(uint32_t flags, uint32_t *prod_st, uint32_t *cons_st) {
> > static const uint32_t prod_st_flags =
> > - (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
> > + (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ |
> > RING_F_MP_HTS_ENQ);
> > static const uint32_t cons_st_flags =
> > - (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
> > + (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ |
> > RING_F_MC_HTS_DEQ);
> >
> > switch (flags & prod_st_flags) {
> > case 0:
> > @@ -116,6 +116,9 @@ get_sync_type(uint32_t flags, uint32_t *prod_st,
> > uint32_t *cons_st)
> > case RING_F_MP_RTS_ENQ:
> > *prod_st = RTE_RING_SYNC_MT_RTS;
> > break;
> > + case RING_F_MP_HTS_ENQ:
> > + *prod_st = RTE_RING_SYNC_MT_HTS;
> > + break;
> > default:
> > return -EINVAL;
> > }
> > @@ -130,6 +133,9 @@ get_sync_type(uint32_t flags, uint32_t *prod_st,
> > uint32_t *cons_st)
> > case RING_F_MC_RTS_DEQ:
> > *cons_st = RTE_RING_SYNC_MT_RTS;
> > break;
> > + case RING_F_MC_HTS_DEQ:
> > + *cons_st = RTE_RING_SYNC_MT_HTS;
> > + break;
> > default:
> > return -EINVAL;
> > }
> > @@ -151,6 +157,11 @@ rte_ring_init(struct rte_ring *r, const char *name,
> > unsigned count,
> > RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
> > RTE_CACHE_LINE_MASK) != 0);
> >
> > + RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
> > + offsetof(struct rte_ring_hts_headtail, sync_type));
> > + RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
> > + offsetof(struct rte_ring_hts_headtail, ht.pos.tail));
> > +
> > RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
> > offsetof(struct rte_ring_rts_headtail, sync_type));
> > RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) != diff --git
> > a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h index
> > a130aeb9d..52edcea11 100644
> > --- a/lib/librte_ring/rte_ring.h
> > +++ b/lib/librte_ring/rte_ring.h
> > @@ -66,11 +66,11 @@ enum {
> > RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
> > RTE_RING_SYNC_ST, /**< single thread only */
> > RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
> > + RTE_RING_SYNC_MT_HTS, /**< multi-thread head/tail sync */
> > };
> >
> > /**
> > - * structure to hold a pair of head/tail values and other metadata.
> > - * used by RTE_RING_SYNC_MT, RTE_RING_SYNC_ST sync types.
> > + * Structure to hold a pair of head/tail values and other metadata.
> > * Depending on sync_type format of that structure might differ
> > * depending on the sync mechanism selelcted, but offsets for
> > * *sync_type* and *tail* values should always remain the same.
> > @@ -96,6 +96,19 @@ struct rte_ring_rts_headtail {
> > volatile union rte_ring_ht_poscnt head; };
> >
> > +union rte_ring_ht_pos {
> > + uint64_t raw;
> > + struct {
> > + uint32_t tail; /**< tail position */
> > + uint32_t head; /**< head position */
> > + } pos;
> > +};
> > +
> > +struct rte_ring_hts_headtail {
> > + uint32_t sync_type; /**< sync type of prod/cons */
> > + volatile union rte_ring_ht_pos ht __rte_aligned(8); };
> > +
> > /**
> > * An RTE ring structure.
> > *
> > @@ -126,6 +139,7 @@ struct rte_ring {
> > RTE_STD_C11
> > union {
> > struct rte_ring_headtail prod;
> > + struct rte_ring_hts_headtail hts_prod;
> > struct rte_ring_rts_headtail rts_prod;
> > } __rte_cache_aligned;
> >
> > @@ -135,6 +149,7 @@ struct rte_ring {
> > RTE_STD_C11
> > union {
> > struct rte_ring_headtail cons;
> > + struct rte_ring_hts_headtail hts_cons;
> > struct rte_ring_rts_headtail rts_cons;
> > } __rte_cache_aligned;
> >
> > @@ -157,6 +172,9 @@ struct rte_ring {
> > #define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS".
> > */ #define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC
> > RTS". */
> >
> > +#define RING_F_MP_HTS_ENQ 0x0020 /**< The default enqueue is "MP
> > HTS".
> > +*/ #define RING_F_MC_HTS_DEQ 0x0040 /**< The default dequeue is "MC
> > +HTS". */
> > +
> > #define __IS_SP RTE_RING_SYNC_ST
> > #define __IS_MP RTE_RING_SYNC_MT
> > #define __IS_SC RTE_RING_SYNC_ST
> > @@ -513,6 +531,82 @@ __rte_ring_do_rts_dequeue(struct rte_ring *r, void
> > **obj_table,
> > return n;
> > }
> >
> > +#include <rte_ring_hts_generic.h>
> > +
> > +/**
> > + * @internal Start to enqueue several objects on the HTS ring.
> > + * Note that user has to call appropriate enqueue_finish()
> > + * to complete given enqueue operation.
> > + *
> > + * @param r
> > + * A pointer to the ring structure.
> > + * @param obj_table
> > + * A pointer to a table of void * pointers (objects).
> > + * @param n
> > + * The number of objects to add in the ring from the obj_table.
> > + * @param behavior
> > + * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
> > + * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from
> > ring
> > + * @param free_space
> > + * returns the amount of space after the enqueue operation has finished
> > + * @return
> > + * Actual number of objects enqueued.
> > + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> > + */
> > +static __rte_always_inline unsigned int
> > +__rte_ring_do_hts_enqueue_start(struct rte_ring *r, void * const *obj_table,
> > + uint32_t n, enum rte_ring_queue_behavior behavior,
> > + uint32_t *free_space)
> > +{
> > + uint32_t free, head;
> > +
> > + n = __rte_ring_hts_move_prod_head(r, n, behavior, &head, &free);
> > +
> > + if (n != 0)
> > + ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
> > +
> > + if (free_space != NULL)
> > + *free_space = free - n;
> > + return n;
> > +}
> rte_ring.h is becoming too big. May be we should move these functions to another HTS specific file. But leave the top level API in rte_ring.h.
> Similarly for RTS.
Good point, will try in v1.
>
> > +
> > +/**
> > + * @internal Start to dequeue several objects from the HTS ring.
> > + * Note that user has to call appropriate dequeue_finish()
> > + * to complete given dequeue operation.
> > + *
> > + * @param r
> > + * A pointer to the ring structure.
> > + * @param obj_table
> > + * A pointer to a table of void * pointers (objects).
> > + * @param n
> > + * The number of objects to pull from the ring.
> > + * @param behavior
> > + * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
> > + * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from
> > ring
> > + * @param available
> > + * returns the number of remaining ring entries after the dequeue has
> > finished
> > + * @return
> > + * - Actual number of objects dequeued.
> > + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> > + */
> > +static __rte_always_inline unsigned int
> > +__rte_ring_do_hts_dequeue_start(struct rte_ring *r, void **obj_table,
> > + unsigned int n, enum rte_ring_queue_behavior behavior,
> > + unsigned int *available)
> > +{
> > + uint32_t entries, head;
> > +
> > + n = __rte_ring_hts_move_cons_head(r, n, behavior, &head, &entries);
> > +
> > + if (n != 0)
> > + DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
> > +
> > + if (available != NULL)
> > + *available = entries - n;
> > + return n;
> > +}
> > +
> > /**
> > * Enqueue several objects on the ring (multi-producers safe).
> > *
> > @@ -585,6 +679,47 @@ rte_ring_rts_enqueue_bulk(struct rte_ring *r, void *
> > const *obj_table,
> > free_space);
> > }
> >
> > +/**
> > + * Start to enqueue several objects on the HTS ring (multi-producers safe).
> > + * Note that user has to call appropriate dequeue_finish()
> > + * to complete given dequeue operation.
> > + *
> > + * @param r
> > + * A pointer to the ring structure.
> > + * @param obj_table
> > + * A pointer to a table of void * pointers (objects).
> > + * @param n
> > + * The number of objects to add in the ring from the obj_table.
> > + * @param free_space
> > + * if non-NULL, returns the amount of space in the ring after the
> > + * enqueue operation has finished.
> > + * @return
> > + * The number of objects enqueued, either 0 or n
> > + */
> > +static __rte_always_inline unsigned int
> > +rte_ring_hts_enqueue_bulk_start(struct rte_ring *r, void * const *obj_table,
> > + unsigned int n, unsigned int *free_space) {
> > + return __rte_ring_do_hts_enqueue_start(r, obj_table, n,
> > + RTE_RING_QUEUE_FIXED, free_space);
> > +}
> I do not clearly understand the requirements on the enqueue_start and enqueue_finish in the form they are here.
> IMO, only requirement for these APIs is to provide the ability to avoid intermediate memcpys.
I think the main objective here is to provide 'MT safe peek' functionality.
The requirement is to split let say dequeue operation into two parts:
1. start - copy N elems into provided by user data buffer and guarantee that
these elems will remain in the ring till finish().
2. finish - remove M(<=N) elems from the ring.
For enqueue it a mirror:
1. start - reserve space for N elems in the ring.
2. finish - copy M (<=N) to the ring.
>
> > +
> > +static __rte_always_inline void
> > +rte_ring_hts_enqueue_finish(struct rte_ring *r, unsigned int n) {
> > + __rte_ring_hts_update_tail(&r->hts_prod, n, 1); }
> > +
> > +static __rte_always_inline unsigned int
> > +rte_ring_hts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
> > + unsigned int n, unsigned int *free_space) {
> > + n = rte_ring_hts_enqueue_bulk_start(r, obj_table, n, free_space);
> > + if (n != 0)
> > + rte_ring_hts_enqueue_finish(r, n);
> > + return n;
> > +}
> > +
> > /**
> > * Enqueue several objects on a ring.
> > *
> > @@ -615,6 +750,8 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void * const
> > *obj_table,
> > return rte_ring_sp_enqueue_bulk(r, obj_table, n, free_space);
> > case RTE_RING_SYNC_MT_RTS:
> > return rte_ring_rts_enqueue_bulk(r, obj_table, n, free_space);
> > + case RTE_RING_SYNC_MT_HTS:
> > + return rte_ring_hts_enqueue_bulk(r, obj_table, n,
> > free_space);
> > }
> >
> > /* valid ring should never reach this point */ @@ -753,6 +890,47 @@
> > rte_ring_rts_dequeue_bulk(struct rte_ring *r, void **obj_table,
> > available);
> > }
> >
> > +/**
> > + * Start to dequeue several objects from an HTS ring (multi-consumers safe).
> > + * Note that user has to call appropriate dequeue_finish()
> > + * to complete given dequeue operation.
> > + *
> > + * @param r
> > + * A pointer to the ring structure.
> > + * @param obj_table
> > + * A pointer to a table of void * pointers (objects) that will be filled.
> > + * @param n
> > + * The number of objects to dequeue from the ring to the obj_table.
> > + * @param available
> > + * If non-NULL, returns the number of remaining ring entries after the
> > + * dequeue has finished.
> > + * @return
> > + * The number of objects dequeued, either 0 or n
> > + */
> > +static __rte_always_inline unsigned int
> > +rte_ring_hts_dequeue_bulk_start(struct rte_ring *r, void **obj_table,
> > + unsigned int n, unsigned int *available) {
> > + return __rte_ring_do_hts_dequeue_start(r, obj_table, n,
> > + RTE_RING_QUEUE_FIXED, available);
> > +}
> IMO, we should look to provide the ability to avoid intermediate copies when the data from the ring needs to be distributed to different
> locations.
As I said in other thread - I am not sure it would provide any gain in terms of performance.
Unless we have a case with bulk transfers and big size elems.
If you still strongly feel SG is needed here, I think it should be an add-on API not the main and only one.
> My proposal in its form is complicated. But, I am thinking that, if the return values are abstracted in a structure, it might look much simple.
>
> > +
> > +static __rte_always_inline void
> > +rte_ring_hts_dequeue_finish(struct rte_ring *r, unsigned int n) {
> > + __rte_ring_hts_update_tail(&r->hts_cons, n, 0); }
> > +
> > +static __rte_always_inline unsigned int
> > +rte_ring_hts_dequeue_bulk(struct rte_ring *r, void **obj_table,
> > + unsigned int n, unsigned int *available) {
> > + n = rte_ring_hts_dequeue_bulk_start(r, obj_table, n, available);
> > + if (n != 0)
> > + rte_ring_hts_dequeue_finish(r, n);
> > + return n;
> > +}
> > +
> > /**
> > * Dequeue several objects from a ring.
> > *
> > @@ -783,6 +961,8 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void
> > **obj_table, unsigned int n,
> > return rte_ring_sc_dequeue_bulk(r, obj_table, n, available);
> > case RTE_RING_SYNC_MT_RTS:
> > return rte_ring_rts_dequeue_bulk(r, obj_table, n, available);
> > + case RTE_RING_SYNC_MT_HTS:
> > + return rte_ring_hts_dequeue_bulk(r, obj_table, n, available);
> > }
> >
> > /* valid ring should never reach this point */ @@ -1111,6 +1291,41
> > @@ rte_ring_rts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
> > RTE_RING_QUEUE_VARIABLE, free_space); }
> >
> > +/**
> > + * Start to enqueue several objects on the HTS ring (multi-producers safe).
> > + * Note that user has to call appropriate dequeue_finish()
> > + * to complete given dequeue operation.
> > + *
> > + * @param r
> > + * A pointer to the ring structure.
> > + * @param obj_table
> > + * A pointer to a table of void * pointers (objects).
> > + * @param n
> > + * The number of objects to add in the ring from the obj_table.
> > + * @param free_space
> > + * if non-NULL, returns the amount of space in the ring after the
> > + * enqueue operation has finished.
> > + * @return
> > + * The number of objects enqueued, either 0 or n
> > + */
> > +static __rte_always_inline unsigned int
> > +rte_ring_hts_enqueue_burst_start(struct rte_ring *r, void * const *obj_table,
> > + unsigned int n, unsigned int *free_space) {
> > + return __rte_ring_do_hts_enqueue_start(r, obj_table, n,
> > + RTE_RING_QUEUE_VARIABLE, free_space); }
> > +
> rte_ring_hts_enqueue_burst_finish is not implemented.
No need to, finish() is identical for both _bulk and _burst.
That's why we have 2 starts:
rte_ring_hts_enqueue_bulk_start()
rte_ring_hts_enqueue_burst_start()
and one finish:
rte_ring_hts_enqueue_finish().
Same story for dequeue.
> It requires the 'n' returned from ' rte_ring_hts_enqueue_burst_start' to be passed.
Yes, it requires some m <= n to be passed.
That's the whole point of peek - we want to be able to inspect N elems
possibly without retrieving them from the ring.
I.E.: inspect N, retrieve M <=N.
> We can't completely avoid passing correct information between xxx_start and xxx_finish APIs.
Yes we can't.
But in that model we can check that provided by finish() value is valid,
plus we don't provide user direct access to the contents of the ring,
and don't require him to specify head/tail values directly.
> > +static __rte_always_inline unsigned int
> > +rte_ring_hts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
> > + unsigned int n, unsigned int *free_space) {
> > + n = rte_ring_hts_enqueue_burst_start(r, obj_table, n, free_space);
> > + if (n != 0)
> > + rte_ring_hts_enqueue_finish(r, n);
> > + return n;
> > +}
> > +
> > /**
> > * Enqueue several objects on a ring.
> > *
> > @@ -1141,6 +1356,8 @@ rte_ring_enqueue_burst(struct rte_ring *r, void *
> > const *obj_table,
> > return rte_ring_sp_enqueue_burst(r, obj_table, n,
> > free_space);
> > case RTE_RING_SYNC_MT_RTS:
> > return rte_ring_rts_enqueue_burst(r, obj_table, n,
> > free_space);
> > + case RTE_RING_SYNC_MT_HTS:
> > + return rte_ring_hts_enqueue_burst(r, obj_table, n,
> > free_space);
> > }
> >
> > /* valid ring should never reach this point */ @@ -1225,6 +1442,42
> > @@ rte_ring_rts_dequeue_burst(struct rte_ring *r, void **obj_table,
> > return __rte_ring_do_rts_dequeue(r, obj_table, n,
> > RTE_RING_QUEUE_VARIABLE, available); }
> > +
> > +/**
> > + * Start to dequeue several objects from an HTS ring (multi-consumers safe).
> > + * Note that user has to call appropriate dequeue_finish()
> > + * to complete given dequeue operation.
> > + *
> > + * @param r
> > + * A pointer to the ring structure.
> > + * @param obj_table
> > + * A pointer to a table of void * pointers (objects) that will be filled.
> > + * @param n
> > + * The number of objects to dequeue from the ring to the obj_table.
> > + * @param available
> > + * If non-NULL, returns the number of remaining ring entries after the
> > + * dequeue has finished.
> > + * @return
> > + * The number of objects dequeued, either 0 or n
> > + */
> > +static __rte_always_inline unsigned int
> > +rte_ring_hts_dequeue_burst_start(struct rte_ring *r, void **obj_table,
> > + unsigned int n, unsigned int *available) {
> > + return __rte_ring_do_hts_dequeue_start(r, obj_table, n,
> > + RTE_RING_QUEUE_VARIABLE, available);
> > +}
> > +
> > +static __rte_always_inline unsigned int
> > +rte_ring_hts_dequeue_burst(struct rte_ring *r, void **obj_table,
> > + unsigned int n, unsigned int *available) {
> > + n = rte_ring_hts_dequeue_burst_start(r, obj_table, n, available);
> > + if (n != 0)
> > + rte_ring_hts_dequeue_finish(r, n);
> > + return n;
> > +}
> > +
> > /**
> > * Dequeue multiple objects from a ring up to a maximum number.
> > *
> > @@ -1255,6 +1508,8 @@ rte_ring_dequeue_burst(struct rte_ring *r, void
> > **obj_table,
> > return rte_ring_sc_dequeue_burst(r, obj_table, n, available);
> > case RTE_RING_SYNC_MT_RTS:
> > return rte_ring_rts_dequeue_burst(r, obj_table, n, available);
> > + case RTE_RING_SYNC_MT_HTS:
> > + return rte_ring_hts_dequeue_burst(r, obj_table, n, available);
> > }
> >
> > /* valid ring should never reach this point */ diff --git
> > a/lib/librte_ring/rte_ring_hts_generic.h
> > b/lib/librte_ring/rte_ring_hts_generic.h
> > new file mode 100644
> > index 000000000..7e447e30b
> > --- /dev/null
> > +++ b/lib/librte_ring/rte_ring_hts_generic.h
> > @@ -0,0 +1,228 @@
> > +/* SPDX-License-Identifier: BSD-3-Clause
> > + *
> > + * Copyright (c) 2010-2020 Intel Corporation
> > + * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
> > + * All rights reserved.
> > + * Derived from FreeBSD's bufring.h
> > + * Used as BSD-3 Licensed with permission from Kip Macy.
> > + */
> > +
> > +#ifndef _RTE_RING_HTS_GENERIC_H_
> > +#define _RTE_RING_HTS_GENERIC_H_
> > +
> > +/**
> > + * @file rte_ring_hts_generic.h
> > + * It is not recommended to include this file directly,
> > + * include <rte_ring.h> instead.
> > + * Contains internal helper functions for head/tail sync (HTS) ring mode.
> > + * In that mode enqueue/dequeue operation is fully serialized:
> > + * only one thread at a time is allowed to perform given op.
> > + * This is achieved by thread is allowed to proceed with changing
> > +head.value
> > + * only when head.value == tail.value.
> > + * Both head and tail values are updated atomically (as one 64-bit value).
> > + * As another enhancement that provides ability to split
> > +enqueue/dequeue
> > + * operation into two phases:
> > + * - enqueue/dequeue start
> > + * - enqueue/dequeue finish
> > + * That allows user to inspect objects in the ring without removing
> > + * them from it (aka MT safe peek).
> > + * As an example:
> > + * // read 1 elem from the ring:
> > + * n = rte_ring_hts_dequeue_bulk_start(ring, &obj, 1, NULL);
> > + * if (n != 0) {
> > + * //examined object
> > + * if (object_examine(obj) == KEEP)
> > + * //decided to keep it in the ring.
> > + * rte_ring_hts_dequeue_finish(ring, 0);
> > + * else
> > + * //decided to remove it in the ring.
> > + * rte_ring_hts_dequeue_finish(ring, n);
> > + * }
> > + * Note that between _start_ and _finish_ the ring is sort of locked -
> > + * none other thread can proceed with enqueue(/dequeue) operation till
> > + * _finish_ will complete.
> This means it does not solve the problem for over committed systems. Do you agree?
I never stated that serialized ring fixes the problem completely.
I said that current approach mitigates is quite well.
And yes, I still believe that statement is correct.
See other thread for more detailed discussion.
>
> > + */
> > +
> > +static __rte_always_inline void
> > +__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
> > + uint32_t enqueue)
> > +{
> > + uint32_t n;
> > + union rte_ring_ht_pos p;
> > +
> > + if (enqueue)
> > + rte_smp_wmb();
> > + else
> > + rte_smp_rmb();
> > +
> > + p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
> > +
> > + n = p.pos.head - p.pos.tail;
> > + RTE_ASSERT(n >= num);
> > + RTE_SET_USED(n);
> > +
> > + p.pos.head = p.pos.tail + num;
> > + p.pos.tail = p.pos.head;
> > +
> > + rte_atomic64_set((rte_atomic64_t *)(uintptr_t)&ht->ht.raw, p.raw); }
> > +
> > +/**
> > + * @internal waits till tail will become equal to head.
> > + * Means no writer/reader is active for that ring.
> > + * Suppose to work as serialization point.
> > + */
> > +static __rte_always_inline void
> > +__rte_ring_hts_head_wait(const struct rte_ring_hts_headtail *ht,
> > + union rte_ring_ht_pos *p)
> > +{
> > + p->raw = rte_atomic64_read((rte_atomic64_t *)
> > + (uintptr_t)&ht->ht.raw);
> > +
> > + while (p->pos.head != p->pos.tail) {
> > + rte_pause();
> > + p->raw = rte_atomic64_read((rte_atomic64_t *)
> > + (uintptr_t)&ht->ht.raw);
> > + }
> > +}
> > +
> > +/**
> > + * @internal This function updates the producer head for enqueue
> > + *
> > + * @param r
> > + * A pointer to the ring structure
> > + * @param is_sp
> > + * Indicates whether multi-producer path is needed or not
> > + * @param n
> > + * The number of elements we will want to enqueue, i.e. how far should the
> > + * head be moved
> > + * @param behavior
> > + * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
> > + * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from
> > ring
> > + * @param old_head
> > + * Returns head value as it was before the move, i.e. where enqueue starts
> > + * @param new_head
> > + * Returns the current/new head value i.e. where enqueue finishes
> > + * @param free_entries
> > + * Returns the amount of free space in the ring BEFORE head was moved
> > + * @return
> > + * Actual number of objects enqueued.
> > + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> > + */
> > +static __rte_always_inline unsigned int
> > +__rte_ring_hts_move_prod_head(struct rte_ring *r, unsigned int num,
> > + enum rte_ring_queue_behavior behavior, uint32_t *old_head,
> > + uint32_t *free_entries)
> > +{
> > + uint32_t n;
> > + union rte_ring_ht_pos np, op;
> > +
> > + const uint32_t capacity = r->capacity;
> > +
> > + do {
> > + /* Reset n to the initial burst count */
> > + n = num;
> > +
> > + /* wait for tail to be equal to head */
> > + __rte_ring_hts_head_wait(&r->hts_prod, &op);
> > +
> > + /* add rmb barrier to avoid load/load reorder in weak
> > + * memory model. It is noop on x86
> > + */
> > + rte_smp_rmb();
> > +
> > + /*
> > + * The subtraction is done between two unsigned 32bits value
> > + * (the result is always modulo 32 bits even if we have
> > + * *old_head > cons_tail). So 'free_entries' is always between
> > 0
> > + * and capacity (which is < size).
> > + */
> > + *free_entries = capacity + r->cons.tail - op.pos.head;
> > +
> > + /* check that we have enough room in ring */
> > + if (unlikely(n > *free_entries))
> > + n = (behavior == RTE_RING_QUEUE_FIXED) ?
> > + 0 : *free_entries;
> > +
> > + if (n == 0)
> > + return 0;
> > +
> > + np.pos.tail = op.pos.tail;
> > + np.pos.head = op.pos.head + n;
> > +
> > + } while (rte_atomic64_cmpset(&r->hts_prod.ht.raw,
> > + op.raw, np.raw) == 0);
> > +
> > + *old_head = op.pos.head;
> > + return n;
> > +}
> > +
> > +/**
> > + * @internal This function updates the consumer head for dequeue
> > + *
> > + * @param r
> > + * A pointer to the ring structure
> > + * @param is_sc
> > + * Indicates whether multi-consumer path is needed or not
> > + * @param n
> > + * The number of elements we will want to enqueue, i.e. how far should the
> > + * head be moved
> > + * @param behavior
> > + * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
> > + * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from
> > ring
> > + * @param old_head
> > + * Returns head value as it was before the move, i.e. where dequeue starts
> > + * @param new_head
> > + * Returns the current/new head value i.e. where dequeue finishes
> > + * @param entries
> > + * Returns the number of entries in the ring BEFORE head was moved
> > + * @return
> > + * - Actual number of objects dequeued.
> > + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> > + */
> > +static __rte_always_inline unsigned int
> > +__rte_ring_hts_move_cons_head(struct rte_ring *r, unsigned int num,
> > + enum rte_ring_queue_behavior behavior, uint32_t *old_head,
> > + uint32_t *entries)
> > +{
> > + uint32_t n;
> > + union rte_ring_ht_pos np, op;
> > +
> > + /* move cons.head atomically */
> > + do {
> > + /* Restore n as it may change every loop */
> > + n = num;
> > +
> > + /* wait for tail to be equal to head */
> > + __rte_ring_hts_head_wait(&r->hts_cons, &op);
> > +
> > + /* add rmb barrier to avoid load/load reorder in weak
> > + * memory model. It is noop on x86
> > + */
> > + rte_smp_rmb();
> > +
> > + /* The subtraction is done between two unsigned 32bits value
> > + * (the result is always modulo 32 bits even if we have
> > + * cons_head > prod_tail). So 'entries' is always between 0
> > + * and size(ring)-1.
> > + */
> > + *entries = r->prod.tail - op.pos.head;
> > +
> > + /* Set the actual entries for dequeue */
> > + if (n > *entries)
> > + n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 :
> > *entries;
> > +
> > + if (unlikely(n == 0))
> > + return 0;
> > +
> > + np.pos.tail = op.pos.tail;
> > + np.pos.head = op.pos.head + n;
> > +
> > + } while (rte_atomic64_cmpset(&r->hts_cons.ht.raw,
> > + op.raw, np.raw) == 0);
> > +
> > + *old_head = op.pos.head;
> > + return n;
> > +}
> > +
> > +#endif /* _RTE_RING_HTS_GENERIC_H_ */
> > --
> > 2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-03-26 1:50 ` Ananyev, Konstantin
@ 2020-03-30 21:29 ` Honnappa Nagarahalli
2020-03-30 23:37 ` Honnappa Nagarahalli
0 siblings, 1 reply; 146+ messages in thread
From: Honnappa Nagarahalli @ 2020-03-30 21:29 UTC (permalink / raw)
To: Ananyev, Konstantin, dev; +Cc: olivier.matz, nd, Honnappa Nagarahalli, nd
<snip>
> >
> > > Subject: [dpdk-dev] [RFC 0/6] New sync modes for ring
> > >
> > > Upfront note - that RFC is not a complete patch.
> > > It introduces an ABI breakage, plus it doesn't update ring_elem code
> > > properly,
> > As per the current rules, these changes (in the current form) will be
> > accepted only for 20.11 release. How do we address this for immediate
> requirements like RCU defer APIs?
>
> I think I found a way to introduce these new modes without API/ABI breakage.
> Working on v1 right now. Plan to submit it by end of that week/start of next
> one.
ok
>
> > I suggest that we move forward with my RFC (taking into consideration your
> feedback) to make progress on RCU APIs.
> >
> > > etc.
> > > I plan to deal with all these things in later versions.
> > > Right now I seek an initial feedback about proposed ideas.
> > > Would also ask people to repeat performance tests (see below) on
> > > their platforms to confirm the impact.
> > >
> > > More and more customers use(/try to use) DPDK based apps within
> > > overcommitted systems (multiple acttive threads over same pysical cores):
> > > VM, container deployments, etc.
> > > One quite common problem they hit: Lock-Holder-Preemption with
> rte_ring.
> > > LHP is quite a common problem for spin-based sync primitives
> > > (spin-locks, etc.) on overcommitted systems.
> > > The situation gets much worse when some sort of fair-locking
> > > technique is used (ticket-lock, etc.).
> > > As now not only lock-owner but also lock-waiters scheduling order
> > > matters a lot.
> > > This is a well-known problem for kernel within VMs:
> > > http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
> > > https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
> > > The problem with rte_ring is that while head accusion is sort of
> > > un-fair locking, waiting on tail is very similar to ticket lock
> > > schema - tail has to be updated in particular order.
> > > That makes current rte_ring implementation to perform really pure on
> > > some overcommited scenarios.
> > > While it is probably not possible to completely resolve this problem
> > > in userspace only (without some kernel communication/intervention),
> > > removing fairness in tail update can mitigate it significantly.
> > > So this RFC proposes two new optional ring synchronization modes:
> > > 1) Head/Tail Sync (HTS) mode
> > > In that mode enqueue/dequeue operation is fully serialized:
> > > only one thread at a time is allowed to perform given op.
> > > As another enhancement provide ability to split enqueue/dequeue
> > > operation into two phases:
> > > - enqueue/dequeue start
> > > - enqueue/dequeue finish
> > > That allows user to inspect objects in the ring without removing
> > > them from it (aka MT safe peek).
> > IMO, this will not address the problem described above.
>
> It does, please see the results produced by ring_stress_*autotest below.
> Let say for test-case: 8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' it
Had not looked at these tests. Please see the numbers below.
> shows:
> avg number of cycles per object for enqueue /dequeue:
> MP/MC: 280314.32
> HTS: 294.72
> RTS: 318.79
>
> Customer who tried it reported similar level of improvement.
Is this tested with the VM/Container setup described in the slides you referred to?
> Actually if you have time - would be very interesting to see what numbers will
> be on ARM boxes.
> To reproduce, just:
> $cat ring_tests_u4
> ring_stress_autotest
> ring_stress_hts_autotest
> ring_stress_rts_autotest
>
> /app/test/dpdk-test --lcores='6,(10-13)@7,(20-23)@8' -n 4 < ring_tests_u4
> 2>&1 | tee res1
>
> Then look at the ' AGGREGATE' stats.
> Right now it is a bit too verbose, so probably the easiest thing to extract same
> numbers quickly:
> grep 'cycles/obj' res1 | grep 'cycles/obj' | cat -n | awk '{if ($(1)%9==0) print
> $(NF);}'
> 280314.32
> 1057833.55
> 294.72
> 480.10
> 318.79
> 461.52
>
> First 2 numbers will be for MP/MC, next 2 for HTS, last 2 for RTS.
12305.05
12027.09
3.59
7.37
4.41
7.98
>
> > For ex: when a producer updates the head and gets scheduled out, other
> > producers have to spin.
>
> Sure, as I wrote in original cover letter:
> " While it is probably not possible to completely resolve this problem in
> userspace only (without some kernel communication/intervention), removing
> fairness in tail update can mitigate it significantly."
> Results from the ring_stress_*_autotest confirm that.
>
> > The problem is probably worse as with non-HTS case moving of the head
> > and copying of the ring elements can happen in parallel between the
> producers (similarly for consumers).
>
> Yes as we serialize the ring, we remove possibility of simultaneous copy.
> That's why for 'normal' cases (one thread per core) original MP/MC is usually
> faster.
> Though on overcommitted scenarios current MP/MC performance degrades
> dramatically.
> The main problem with current MP/MC implementation is in that tail update
> have to be done in strict order (sort of fair locking scheme).
> Which means that we have much-much worse LHP manifestation, then when
> we use unfair schemes.
> With serialized ring (HTS) we remove that ordering completely (same idea as
> switch from fair to unfair locking for PV spin-locks).
>
> > IMO, HTS should not be a configurable flag.
>
> Why?
>
> > In RCU requirement, a MP enqueue and HTS dequeue are required.
>
> This is supported, user can specify different modes for consumer and
> producer:
> (0 | RING_F_MC_HTS_DEQ).
> Then it is up to the user either to call generic
> rte_ring_enqueue/rte_ring_dequeue,
> or specify mode manually by function name:
> rte_ring_mp_enqueue_bulk/ rte_ring_hts_dequeue_bulk.
Ok, that should be good.
>
> >
> > > 2) Relaxed Tail Sync (RTS)
> > > The main difference from original MP/MC algorithm is that tail value
> > > is increased not by every thread that finished enqueue/dequeue, but
> > > only by the last one.
> > > That allows threads to avoid spinning on ring tail value, leaving
> > > actual tail value change to the last thread in the update queue.
> > This can be a configurable flag on the ring.
> > I am not sure how this solves the problem you have stated above
> > completely. Updating the count from all intermediate threads is still
> > required to update the value of the head. But yes, it reduces the severity of
> the problem by not enforcing the order in which the tail is updated.
>
> As I said above, main source of slowdown here - that we have to update tail in
> particular order.
> So the main objective (same as for HTS) is to remove that ordering.
>
> > I also think it introduces the problem on the other side of the ring
> > because the tail is not updated soon enough (the other side has to wait
> longer for the elements to become available).
>
> Yes, producer/consumer starvation.
> That's why we need max allowed Head-Tail-Distance (htd_max) - to limit how
> far head can go away from tail.
>
> > It also introduces another configuration parameter (HTD_MAX_DEF) which
> > they have to deal with.
>
> If user doesn't provide any value, it will be set by default to ring.capacity / 8.
> From my measurements works quite well.
> Though there possibility for the user to set another value, if needed.
>
> > Users have to still implement the current hypervisor related solutions.
>
> Didn't get what you trying to say with that phrase.
The references you provided talked about resolving LHP by doing co-scheduling of vCPUs (which I think could be applied to DPDK applications). I am saying that we still need such mechanisms along with these solutions.
>
> > IMO, we should run the benchmark for this on an over committed setup to
> understand the benefits.
>
> That's why I created ring_stress_*autotest test-cases and collected numbers
> provided below.
> I suppose they clearly show the problem on overcommitted scenarios, and
> how RTS/HTS improve that situation.
> Would appreciate if you repeat these tests on your machines.
>
> >
> > >
> > > Test results on IA (see below) show significant improvements for
> > > average enqueue/dequeue op times on overcommitted systems.
> > > For 'classic' DPDK deployments (one thread per core) original MP/MC
> > > algorithm still shows best numbers, though for 64-bit target RTS
> > > numbers are not that far away.
> > > Numbers were produced by ring_stress_*autotest (first patch in these
> series).
> > >
> > > X86_64 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
> > > DEQ+ENQ average cycles/obj
> > >
> > > MP/MC HTS RTS
> > > 1thread@1core(--lcores=6-7) 8.00 8.15 8.99
> > > 2thread@2core(--lcores=6-8) 19.14 19.61 20.35
> > > 4thread@4core(--lcores=6-10) 29.43 29.79 31.82
> > > 8thread@8core(--lcores=6-14) 110.59 192.81 119.50
> > > 16thread@16core(--lcores=6-22) 461.03 813.12 495.59
> > > 32thread/@32core(--lcores='6-22,55-70') 982.90 1972.38 1160.51
> > >
> > > 2thread@1core(--lcores='6,(10-11)@7' 20140.50 23.58 25.14
> > > 4thread@2core(--lcores='6,(10-11)@7,(20-21)@8' 153680.60 76.88
> 80.05
> > > 8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 280314.32 294.72
> > > 318.79 16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 643176.59
> > > 1144.02
> > > 1175.14 32thread@2core(--lcores='6,(10-25)@7,(30-45)@8' 4264238.80
> > > 4627.48 4892.68
> > >
> > > 8thread@2core(--lcores='6,(10-17)@(7,8))' 321085.98 298.59 307.47
> > > 16thread@4core(--lcores='6,(20-35)@(7-10))' 1900705.61 575.35
> 678.29
> > > 32thread@4core(--lcores='6,(20-51)@(7-10))' 5510445.85 2164.36
> 2714.12
> > >
> > > i686 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
> > > DEQ+ENQ average cycles/obj
> > >
> > > MP/MC HTS RTS
> > > 1thread@1core(--lcores=6-7) 7.85 12.13 11.31
> > > 2thread@2core(--lcores=6-8) 17.89 24.52 21.86
> > > 8thread@8core(--lcores=6-14) 32.58 354.20 54.58
> > > 32thread/@32core(--lcores='6-22,55-70') 813.77 6072.41 2169.91
> > >
> > > 2thread@1core(--lcores='6,(10-11)@7' 16095.00 36.06 34.74
> > > 8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 1140354.54 346.61
> > > 361.57 16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 1920417.86
> > > 1314.90
> > > 1416.65
> > >
> > > 8thread@2core(--lcores='6,(10-17)@(7,8))' 594358.61 332.70 357.74
> > > 32thread@4core(--lcores='6,(20-51)@(7-10))' 5319896.86 2836.44
> 3028.87
> > >
> > > Konstantin Ananyev (6):
> > > test/ring: add contention stress test
> > > ring: rework ring layout to allow new sync schemes
> > > ring: introduce RTS ring mode
> > > test/ring: add contention stress test for RTS ring
> > > ring: introduce HTS ring mode
> > > test/ring: add contention stress test for HTS ring
> > >
> > > app/test/Makefile | 3 +
> > > app/test/meson.build | 3 +
> > > app/test/test_pdump.c | 6 +-
> > > app/test/test_ring_hts_stress.c | 28 ++
> > > app/test/test_ring_rts_stress.c | 28 ++
> > > app/test/test_ring_stress.c | 27 ++
> > > app/test/test_ring_stress.h | 477 +++++++++++++++++++
> > > lib/librte_pdump/rte_pdump.c | 2 +-
> > > lib/librte_port/rte_port_ring.c | 12 +-
> > > lib/librte_ring/Makefile | 4 +-
> > > lib/librte_ring/meson.build | 4 +-
> > > lib/librte_ring/rte_ring.c | 84 +++-
> > > lib/librte_ring/rte_ring.h | 619 +++++++++++++++++++++++--
> > > lib/librte_ring/rte_ring_elem.h | 8 +-
> > > lib/librte_ring/rte_ring_hts_generic.h | 228 +++++++++
> > > lib/librte_ring/rte_ring_rts_generic.h | 240 ++++++++++
> > > 16 files changed, 1721 insertions(+), 52 deletions(-) create mode
> > > 100644 app/test/test_ring_hts_stress.c create mode 100644
> > > app/test/test_ring_rts_stress.c create mode 100644
> > > app/test/test_ring_stress.c create mode 100644
> > > app/test/test_ring_stress.h create mode 100644
> > > lib/librte_ring/rte_ring_hts_generic.h
> > > create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
> > >
> > > --
> > > 2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-03-30 21:29 ` Honnappa Nagarahalli
@ 2020-03-30 23:37 ` Honnappa Nagarahalli
2020-03-31 17:21 ` Ananyev, Konstantin
0 siblings, 1 reply; 146+ messages in thread
From: Honnappa Nagarahalli @ 2020-03-30 23:37 UTC (permalink / raw)
To: Ananyev, Konstantin, dev; +Cc: olivier.matz, nd, Honnappa Nagarahalli, nd
<snip>
>
> > >
> > > > Subject: [dpdk-dev] [RFC 0/6] New sync modes for ring
> > > >
> > > > Upfront note - that RFC is not a complete patch.
> > > > It introduces an ABI breakage, plus it doesn't update ring_elem
> > > > code properly,
> > > As per the current rules, these changes (in the current form) will
> > > be accepted only for 20.11 release. How do we address this for
> > > immediate
> > requirements like RCU defer APIs?
> >
> > I think I found a way to introduce these new modes without API/ABI
> breakage.
> > Working on v1 right now. Plan to submit it by end of that week/start
> > of next one.
> ok
RCU defer APIs require the rte_ring_xxx_elem versions. I guess you are adding those as well.
>
<snip>
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v1 0/8] New sync modes for ring
2020-02-24 11:35 [dpdk-dev] [RFC 0/6] New sync modes for ring Konstantin Ananyev
` (7 preceding siblings ...)
2020-03-25 20:43 ` Honnappa Nagarahalli
@ 2020-03-31 16:43 ` Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 1/8] test/ring: add contention stress test Konstantin Ananyev
` (8 more replies)
8 siblings, 9 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-03-31 16:43 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, Konstantin Ananyev
RFC - V1 changes:
1. remove ABI brekage (at least I hope I did)
2. Add support for ring_elem
3. Rework peek related API a bit
4. Rework test to make it less verbose and unite all test-cases
in one command
5. Add new test-case for MT peek API
TODO list:
1. Add C11 atomics support
2. Update docs
These days many customers use(/try to use) DPDK based apps within
overcommitted systems (multiple acttive threads over same pysical cores):
VM, container deployments, etc.
One quite common problem they hit:
Lock-Holder-Preemption/Lock-Waiter-Preemption with rte_ring.
LHP/LWP are quite a common problem for spin-based sync primitives
(spin-locks, etc.) on overcommitted systems.
The situation gets much worse when some sort of
fair-locking technique is used (ticket-lock, etc.).
As now not only lock-owner but also lock-waiters scheduling
order matters a lot.
This is a well-known problem for kernel within VMs:
http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
The problem with rte_ring is that while head accusion is sort of
un-fair locking, waiting on tail is very similar to ticket lock schema -
tail has to be updated in particular order.
That makes current rte_ring implementation to perform
really pure on some overcommited scenarios.
While it is probably not possible to completely resolve LHP problem in
userspace only (without some kernel communication/intervention),
removing fairness in tail update can mitigate it significantly.
So this RFC proposes two new optional ring synchronization modes:
1) Head/Tail Sync (HTS) mode
In that mode enqueue/dequeue operation is fully serialized:
only one thread at a time is allowed to perform given op.
As another enhancement provide ability to split enqueue/dequeue
operation into two phases:
- enqueue/dequeue start
- enqueue/dequeue finish
That allows user to inspect objects in the ring without removing
them from it (aka MT safe peek).
2) Relaxed Tail Sync (RTS)
The main difference from original MP/MC algorithm is that
tail value is increased not by every thread that finished enqueue/dequeue,
but only by the last one.
That allows threads to avoid spinning on ring tail value,
leaving actual tail value change to the last thread in the update queue.
Note that these new sync modes are optional.
For current rte_ring users nothing should change
(both in terms of API/ABI and performance).
Existing sync modes MP/MC,SP/SC kept untouched, set up in the same
way (via flags and _init_), and MP/MC remains as default one.
The only thing that changed:
Format of prod/cons now could differ depending on mode selected at _init_.
So user has to stick with one sync model through whole ring lifetime.
In other words, user can't create a ring for let say SP mode and then
in the middle of data-path change his mind and start using MP_RTS mode.
For existing modes (SP/MP, SC/MC) format remains the same and
user can still use them interchangeably, though of course it is an
error prone practice.
Test results on IA (see below) show significant improvements
for average enqueue/dequeue op times on overcommitted systems.
For 'classic' DPDK deployments (one thread per core) original MP/MC
algorithm still shows best numbers, though for 64-bit target
RTS numbers are not that far away.
Numbers were produced by new UT test-case: ring_stress_autotest, i.e.:
echo ring_stress_autotest | ./dpdk-test -n 4 --lcores='...'
X86_64 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
DEQ+ENQ average cycles/obj
MP/MC HTS RTS
1thread@1core(--lcores=6-7) 8.00 8.15 8.99
2thread@2core(--lcores=6-8) 19.14 19.61 20.35
4thread@4core(--lcores=6-10) 29.43 29.79 31.82
8thread@8core(--lcores=6-14) 110.59 192.81 119.50
16thread@16core(--lcores=6-22) 461.03 813.12 495.59
32thread/@32core(--lcores='6-22,55-70') 982.90 1972.38 1160.51
2thread@1core(--lcores='6,(10-11)@7' 20140.50 23.58 25.14
4thread@2core(--lcores='6,(10-11)@7,(20-21)@8' 153680.60 76.88 80.05
8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 280314.32 294.72 318.79
16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 643176.59 1144.02 1175.14
32thread@2core(--lcores='6,(10-25)@7,(30-45)@8' 4264238.80 4627.48 4892.68
8thread@2core(--lcores='6,(10-17)@(7,8))' 321085.98 298.59 307.47
16thread@4core(--lcores='6,(20-35)@(7-10))' 1900705.61 575.35 678.29
32thread@4core(--lcores='6,(20-51)@(7-10))' 5510445.85 2164.36 2714.12
i686 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
DEQ+ENQ average cycles/obj
MP/MC HTS RTS
1thread@1core(--lcores=6-7) 7.85 12.13 11.31
2thread@2core(--lcores=6-8) 17.89 24.52 21.86
8thread@8core(--lcores=6-14) 32.58 354.20 54.58
32thread/@32core(--lcores='6-22,55-70') 813.77 6072.41 2169.91
2thread@1core(--lcores='6,(10-11)@7' 16095.00 36.06 34.74
8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 1140354.54 346.61 361.57
16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 1920417.86 1314.90 1416.65
8thread@2core(--lcores='6,(10-17)@(7,8))' 594358.61 332.70 357.74
32thread@4core(--lcores='6,(20-51)@(7-10))' 5319896.86 2836.44 3028.87
Konstantin Ananyev (8):
test/ring: add contention stress test
ring: prepare ring to allow new sync schemes
ring: introduce RTS ring mode
test/ring: add contention stress test for RTS ring
ring: introduce HTS ring mode
test/ring: add contention stress test for HTS ring
ring: introduce peek style API
test/ring: add stress test for MT peek API
app/test/Makefile | 5 +
app/test/meson.build | 5 +
app/test/test_pdump.c | 6 +-
app/test/test_ring_hts_stress.c | 32 ++
app/test/test_ring_mpmc_stress.c | 31 ++
app/test/test_ring_peek_stress.c | 43 +++
app/test/test_ring_rts_stress.c | 32 ++
app/test/test_ring_stress.c | 57 ++++
app/test/test_ring_stress.h | 37 +++
app/test/test_ring_stress_impl.h | 436 +++++++++++++++++++++++++
lib/librte_pdump/rte_pdump.c | 2 +-
lib/librte_port/rte_port_ring.c | 12 +-
lib/librte_ring/Makefile | 9 +-
lib/librte_ring/meson.build | 9 +-
lib/librte_ring/rte_ring.c | 114 ++++++-
lib/librte_ring/rte_ring.h | 243 ++++++++++++--
lib/librte_ring/rte_ring_c11_mem.h | 44 +++
lib/librte_ring/rte_ring_elem.h | 105 +++++-
lib/librte_ring/rte_ring_generic.h | 48 +++
lib/librte_ring/rte_ring_hts.h | 210 ++++++++++++
lib/librte_ring/rte_ring_hts_elem.h | 205 ++++++++++++
lib/librte_ring/rte_ring_hts_generic.h | 235 +++++++++++++
lib/librte_ring/rte_ring_peek.h | 379 +++++++++++++++++++++
lib/librte_ring/rte_ring_rts.h | 316 ++++++++++++++++++
lib/librte_ring/rte_ring_rts_elem.h | 205 ++++++++++++
lib/librte_ring/rte_ring_rts_generic.h | 210 ++++++++++++
26 files changed, 2974 insertions(+), 56 deletions(-)
create mode 100644 app/test/test_ring_hts_stress.c
create mode 100644 app/test/test_ring_mpmc_stress.c
create mode 100644 app/test/test_ring_peek_stress.c
create mode 100644 app/test/test_ring_rts_stress.c
create mode 100644 app/test/test_ring_stress.c
create mode 100644 app/test/test_ring_stress.h
create mode 100644 app/test/test_ring_stress_impl.h
create mode 100644 lib/librte_ring/rte_ring_hts.h
create mode 100644 lib/librte_ring/rte_ring_hts_elem.h
create mode 100644 lib/librte_ring/rte_ring_hts_generic.h
create mode 100644 lib/librte_ring/rte_ring_peek.h
create mode 100644 lib/librte_ring/rte_ring_rts.h
create mode 100644 lib/librte_ring/rte_ring_rts_elem.h
create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v1 1/8] test/ring: add contention stress test
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
@ 2020-03-31 16:43 ` Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 2/8] ring: prepare ring to allow new sync schemes Konstantin Ananyev
` (7 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-03-31 16:43 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, Konstantin Ananyev
Introduce new test-case to measure ring perfomance under contention
(miltiple producers/consumers).
Starts dequeue/enqueue loop on all available slave lcores.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 2 +
app/test/meson.build | 2 +
app/test/test_ring_mpmc_stress.c | 31 +++
app/test/test_ring_stress.c | 48 ++++
app/test/test_ring_stress.h | 34 +++
app/test/test_ring_stress_impl.h | 436 +++++++++++++++++++++++++++++++
6 files changed, 553 insertions(+)
create mode 100644 app/test/test_ring_mpmc_stress.c
create mode 100644 app/test/test_ring_stress.c
create mode 100644 app/test/test_ring_stress.h
create mode 100644 app/test/test_ring_stress_impl.h
diff --git a/app/test/Makefile b/app/test/Makefile
index 1f080d162..4eefaa887 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -77,7 +77,9 @@ SRCS-y += test_external_mem.c
SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
+SRCS-y += test_ring_mpmc_stress.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
ifeq ($(CONFIG_RTE_LIBRTE_TABLE),y)
diff --git a/app/test/meson.build b/app/test/meson.build
index 351d29cb6..827b04886 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -100,7 +100,9 @@ test_sources = files('commands.c',
'test_rib.c',
'test_rib6.c',
'test_ring.c',
+ 'test_ring_mpmc_stress.c',
'test_ring_perf.c',
+ 'test_ring_stress.c',
'test_rwlock.c',
'test_sched.c',
'test_service_cores.c',
diff --git a/app/test/test_ring_mpmc_stress.c b/app/test/test_ring_mpmc_stress.c
new file mode 100644
index 000000000..1524b0248
--- /dev/null
+++ b/app/test/test_ring_mpmc_stress.c
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_mc_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_mp_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num, 0);
+}
+
+const struct test test_ring_mpmc_stress = {
+ .name = "MP/MC",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
new file mode 100644
index 000000000..60706f799
--- /dev/null
+++ b/app/test/test_ring_stress.c
@@ -0,0 +1,48 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress.h"
+
+static int
+run_test(const struct test *test)
+{
+ int32_t rc;
+ uint32_t i, k;
+
+ for (i = 0, k = 0; i != test->nb_case; i++) {
+
+ printf("TEST-CASE %s %s START\n",
+ test->name, test->cases[i].name);
+
+ rc = test->cases[i].func(test->cases[i].wfunc);
+ k += (rc == 0);
+
+ if (rc != 0)
+ printf("TEST-CASE %s %s FAILED\n",
+ test->name, test->cases[i].name);
+ else
+ printf("TEST-CASE %s %s OK\n",
+ test->name, test->cases[i].name);
+ }
+
+ return k;
+}
+
+static int
+test_ring_stress(void)
+{
+ uint32_t n, k;
+
+ n = 0;
+ k = 0;
+
+ n += test_ring_mpmc_stress.nb_case;
+ k += run_test(&test_ring_mpmc_stress);
+
+ printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
+ n, k, n - k);
+ return (k != n);
+}
+
+REGISTER_TEST_COMMAND(ring_stress_autotest, test_ring_stress);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
new file mode 100644
index 000000000..5ab121fe4
--- /dev/null
+++ b/app/test/test_ring_stress.h
@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+
+#include <inttypes.h>
+#include <stddef.h>
+#include <stdalign.h>
+#include <string.h>
+#include <stdio.h>
+#include <unistd.h>
+
+#include <rte_ring.h>
+#include <rte_cycles.h>
+#include <rte_launch.h>
+#include <rte_pause.h>
+#include <rte_random.h>
+#include <rte_malloc.h>
+
+#include "test.h"
+
+struct test_case {
+ const char *name;
+ int (*func)(int (*)(void *));
+ int (*wfunc)(void *arg);
+};
+
+struct test {
+ const char *name;
+ uint32_t nb_case;
+ const struct test_case *cases;
+};
+
+extern const struct test test_ring_mpmc_stress;
diff --git a/app/test/test_ring_stress_impl.h b/app/test/test_ring_stress_impl.h
new file mode 100644
index 000000000..0d6f0d2ae
--- /dev/null
+++ b/app/test/test_ring_stress_impl.h
@@ -0,0 +1,436 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress.h"
+
+/*
+ * Measures performance of ring enqueue/dequeue under high contention
+ */
+
+#define RING_NAME "RING_STRESS"
+#define BULK_NUM 32
+#define RING_SIZE (2 * BULK_NUM * RTE_MAX_LCORE)
+
+enum {
+ WRK_CMD_STOP,
+ WRK_CMD_RUN,
+};
+
+static volatile uint32_t wrk_cmd __rte_cache_aligned;
+
+/* test run-time in seconds */
+static const uint32_t run_time = 60;
+static const uint32_t verbose;
+
+struct lcore_stat {
+ uint64_t nb_cycle;
+ struct {
+ uint64_t nb_call;
+ uint64_t nb_obj;
+ uint64_t nb_cycle;
+ uint64_t max_cycle;
+ uint64_t min_cycle;
+ } op;
+};
+
+struct lcore_arg {
+ struct rte_ring *rng;
+ struct lcore_stat stats;
+} __rte_cache_aligned;
+
+struct ring_elem {
+ uint32_t cnt[RTE_CACHE_LINE_SIZE / sizeof(uint32_t)];
+} __rte_cache_aligned;
+
+/*
+ * redefinable functions
+ */
+static uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail);
+
+static uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free);
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num);
+
+
+static void
+lcore_stat_update(struct lcore_stat *ls, uint64_t call, uint64_t obj,
+ uint64_t tm, int32_t prcs)
+{
+ ls->op.nb_call += call;
+ ls->op.nb_obj += obj;
+ ls->op.nb_cycle += tm;
+ if (prcs) {
+ ls->op.max_cycle = RTE_MAX(ls->op.max_cycle, tm);
+ ls->op.min_cycle = RTE_MIN(ls->op.min_cycle, tm);
+ }
+}
+
+static void
+lcore_op_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls)
+{
+
+ ms->op.nb_call += ls->op.nb_call;
+ ms->op.nb_obj += ls->op.nb_obj;
+ ms->op.nb_cycle += ls->op.nb_cycle;
+ ms->op.max_cycle = RTE_MAX(ms->op.max_cycle, ls->op.max_cycle);
+ ms->op.min_cycle = RTE_MIN(ms->op.min_cycle, ls->op.min_cycle);
+}
+
+static void
+lcore_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls)
+{
+ ms->nb_cycle = RTE_MAX(ms->nb_cycle, ls->nb_cycle);
+ lcore_op_stat_aggr(ms, ls);
+}
+
+static void
+lcore_stat_dump(FILE *f, uint32_t lc, const struct lcore_stat *ls)
+{
+ long double st;
+
+ st = (long double)rte_get_timer_hz() / US_PER_S;
+
+ if (lc == UINT32_MAX)
+ fprintf(f, "%s(AGGREGATE)={\n", __func__);
+ else
+ fprintf(f, "%s(lc=%u)={\n", __func__, lc);
+
+ fprintf(f, "\tnb_cycle=%" PRIu64 "(%.2Lf usec),\n",
+ ls->nb_cycle, (long double)ls->nb_cycle / st);
+
+ fprintf(f, "\tDEQ+ENQ={\n");
+
+ fprintf(f, "\t\tnb_call=%" PRIu64 ",\n", ls->op.nb_call);
+ fprintf(f, "\t\tnb_obj=%" PRIu64 ",\n", ls->op.nb_obj);
+ fprintf(f, "\t\tnb_cycle=%" PRIu64 ",\n", ls->op.nb_cycle);
+ fprintf(f, "\t\tobj/call(avg): %.2Lf\n",
+ (long double)ls->op.nb_obj / ls->op.nb_call);
+ fprintf(f, "\t\tcycles/obj(avg): %.2Lf\n",
+ (long double)ls->op.nb_cycle / ls->op.nb_obj);
+ fprintf(f, "\t\tcycles/call(avg): %.2Lf\n",
+ (long double)ls->op.nb_cycle / ls->op.nb_call);
+
+ /* if min/max cycles per call stats was collected */
+ if (ls->op.min_cycle != UINT64_MAX) {
+ fprintf(f, "\t\tmax cycles/call=%" PRIu64 "(%.2Lf usec),\n",
+ ls->op.max_cycle,
+ (long double)ls->op.max_cycle / st);
+ fprintf(f, "\t\tmin cycles/call=%" PRIu64 "(%.2Lf usec),\n",
+ ls->op.min_cycle,
+ (long double)ls->op.min_cycle / st);
+ }
+
+ fprintf(f, "\t},\n");
+ fprintf(f, "};\n");
+}
+
+static void
+fill_ring_elm(struct ring_elem *elm, uint32_t fill)
+{
+ uint32_t i;
+
+ for (i = 0; i != RTE_DIM(elm->cnt); i++)
+ elm->cnt[i] = fill;
+}
+
+static int32_t
+check_updt_elem(struct ring_elem *elm[], uint32_t num,
+ const struct ring_elem *check, const struct ring_elem *fill)
+{
+ uint32_t i;
+
+ static rte_spinlock_t dump_lock;
+
+ for (i = 0; i != num; i++) {
+ if (memcmp(check, elm[i], sizeof(*check)) != 0) {
+ rte_spinlock_lock(&dump_lock);
+ printf("%s(lc=%u, num=%u) failed at %u-th iter, "
+ "offending object: %p\n",
+ __func__, rte_lcore_id(), num, i, elm[i]);
+ rte_memdump(stdout, "expected", check, sizeof(*check));
+ rte_memdump(stdout, "result", elm[i], sizeof(elm[i]));
+ rte_spinlock_unlock(&dump_lock);
+ return -EINVAL;
+ }
+ memcpy(elm[i], fill, sizeof(*elm[i]));
+ }
+
+ return 0;
+}
+
+static int
+check_ring_op(uint32_t exp, uint32_t res, uint32_t lc,
+ const char *fname, const char *opname)
+{
+ if (exp != res) {
+ printf("%s(lc=%u) failure: %s expected: %u, returned %u\n",
+ fname, lc, opname, exp, res);
+ return -ENOSPC;
+ }
+ return 0;
+}
+
+static int
+test_worker_prcs(void *arg)
+{
+ int32_t rc;
+ uint32_t lc, n, num;
+ uint64_t cl, tm0, tm1;
+ struct lcore_arg *la;
+ struct ring_elem def_elm, loc_elm;
+ struct ring_elem *obj[2 * BULK_NUM];
+
+ la = arg;
+ lc = rte_lcore_id();
+
+ fill_ring_elm(&def_elm, UINT32_MAX);
+ fill_ring_elm(&loc_elm, lc);
+
+ while (wrk_cmd != WRK_CMD_RUN) {
+ rte_smp_rmb();
+ rte_pause();
+ }
+
+ cl = rte_rdtsc_precise();
+
+ do {
+ /* num in interval [7/8, 11/8] of BULK_NUM */
+ num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
+
+ /* reset all pointer values */
+ memset(obj, 0, sizeof(obj));
+
+ /* dequeue num elems */
+ tm0 = rte_rdtsc_precise();
+ n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
+ tm0 = rte_rdtsc_precise() - tm0;
+
+ /* check return value and objects */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_dequeue_bulk));
+ if (rc == 0)
+ rc = check_updt_elem(obj, num, &def_elm, &loc_elm);
+ if (rc != 0)
+ break;
+
+ /* enqueue num elems */
+ rte_compiler_barrier();
+ rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
+ if (rc != 0)
+ break;
+
+ tm1 = rte_rdtsc_precise();
+ n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
+ tm1 = rte_rdtsc_precise() - tm1;
+
+ /* check return value */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_enqueue_bulk));
+ if (rc != 0)
+ break;
+
+ lcore_stat_update(&la->stats, 1, num, tm0 + tm1, 1);
+
+ } while (wrk_cmd == WRK_CMD_RUN);
+
+ la->stats.nb_cycle = rte_rdtsc_precise() - cl;
+ return rc;
+}
+
+static int
+test_worker_avg(void *arg)
+{
+ int32_t rc;
+ uint32_t lc, n, num;
+ uint64_t cl;
+ struct lcore_arg *la;
+ struct ring_elem def_elm, loc_elm;
+ struct ring_elem *obj[2 * BULK_NUM];
+
+ la = arg;
+ lc = rte_lcore_id();
+
+ fill_ring_elm(&def_elm, UINT32_MAX);
+ fill_ring_elm(&loc_elm, lc);
+
+ while (wrk_cmd != WRK_CMD_RUN) {
+ rte_smp_rmb();
+ rte_pause();
+ }
+
+ cl = rte_rdtsc_precise();
+
+ do {
+ /* num in interval [7/8, 11/8] of BULK_NUM */
+ num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
+
+ /* reset all pointer values */
+ memset(obj, 0, sizeof(obj));
+
+ /* dequeue num elems */
+ n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
+
+ /* check return value and objects */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_dequeue_bulk));
+ if (rc == 0)
+ rc = check_updt_elem(obj, num, &def_elm, &loc_elm);
+ if (rc != 0)
+ break;
+
+ /* enqueue num elems */
+ rte_compiler_barrier();
+ rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
+ if (rc != 0)
+ break;
+
+ n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
+
+ /* check return value */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_enqueue_bulk));
+ if (rc != 0)
+ break;
+
+ lcore_stat_update(&la->stats, 1, num, 0, 0);
+
+ } while (wrk_cmd == WRK_CMD_RUN);
+
+ /* final stats update */
+ cl = rte_rdtsc_precise() - cl;
+ lcore_stat_update(&la->stats, 0, 0, cl, 0);
+ la->stats.nb_cycle = cl;
+
+ return rc;
+}
+
+static void
+mt1_fini(struct rte_ring *rng, void *data)
+{
+ rte_free(rng);
+ rte_free(data);
+}
+
+static int
+mt1_init(struct rte_ring **rng, void **data, uint32_t num)
+{
+ int32_t rc;
+ size_t sz;
+ uint32_t i, nr;
+ struct rte_ring *r;
+ struct ring_elem *elm;
+ void *p;
+
+ *rng = NULL;
+ *data = NULL;
+
+ sz = num * sizeof(*elm);
+ elm = rte_zmalloc(NULL, sz, alignof(*elm));
+
+ /* alloc ring */
+ nr = 2 * num;
+ sz = rte_ring_get_memsize(nr);
+ r = rte_zmalloc(NULL, sz, alignof(*r));
+ if (r == NULL) {
+ printf("%s: alloca(%zu) for FIFO with %u elems failed",
+ __func__, sz, nr);
+ return -ENOMEM;
+ }
+ rc = _st_ring_init(r, RING_NAME, nr);
+ if (rc != 0) {
+ printf("%s: _st_ring_init(%p, %u) failed, error: %d(%s)\n",
+ __func__, r, nr, rc, strerror(-rc));
+ return rc;
+ }
+
+ for (i = 0; i != num; i++) {
+ fill_ring_elm(elm + i, UINT32_MAX);
+ p = elm + i;
+ if (_st_ring_enqueue_bulk(r, &p, 1, NULL) != 1)
+ break;
+ }
+
+ if (i != num) {
+ printf("%s: _st_ring_enqueue(%p, %u) returned %u\n",
+ __func__, r, num, i);
+ return -ENOSPC;
+ }
+
+ *rng = r;
+ *data = elm;
+ return 0;
+}
+
+static int
+test_mt1(int (*test)(void *))
+{
+ int32_t rc;
+ uint32_t lc, mc;
+ struct rte_ring *r;
+ void *data;
+ struct lcore_arg arg[RTE_MAX_LCORE];
+
+ static const struct lcore_stat init_stat = {
+ .op.min_cycle = UINT64_MAX,
+ };
+
+ rc = mt1_init(&r, &data, RING_SIZE);
+ if (rc != 0) {
+ mt1_fini(r, data);
+ return rc;
+ }
+
+ memset(arg, 0, sizeof(arg));
+
+ /* launch on all slaves */
+ RTE_LCORE_FOREACH_SLAVE(lc) {
+ arg[lc].rng = r;
+ arg[lc].stats = init_stat;
+ rte_eal_remote_launch(test, &arg[lc], lc);
+ }
+
+ /* signal worker to start test */
+ wrk_cmd = WRK_CMD_RUN;
+ rte_smp_wmb();
+
+ usleep(run_time * US_PER_S);
+
+ /* signal worker to start test */
+ wrk_cmd = WRK_CMD_STOP;
+ rte_smp_wmb();
+
+ /* wait for slaves and collect stats. */
+ mc = rte_lcore_id();
+ arg[mc].stats = init_stat;
+
+ rc = 0;
+ RTE_LCORE_FOREACH_SLAVE(lc) {
+ rc |= rte_eal_wait_lcore(lc);
+ lcore_stat_aggr(&arg[mc].stats, &arg[lc].stats);
+ if (verbose != 0)
+ lcore_stat_dump(stdout, lc, &arg[lc].stats);
+ }
+
+ lcore_stat_dump(stdout, UINT32_MAX, &arg[mc].stats);
+ mt1_fini(r, data);
+ return rc;
+}
+
+static const struct test_case tests[] = {
+ {
+ .name = "MT-WRK_ENQ_DEQ-MST_NONE-PRCS",
+ .func = test_mt1,
+ .wfunc = test_worker_prcs,
+ },
+ {
+ .name = "MT-WRK_ENQ_DEQ-MST_NONE-AVG",
+ .func = test_mt1,
+ .wfunc = test_worker_avg,
+ },
+};
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v1 2/8] ring: prepare ring to allow new sync schemes
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 1/8] test/ring: add contention stress test Konstantin Ananyev
@ 2020-03-31 16:43 ` Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 3/8] ring: introduce RTS ring mode Konstantin Ananyev
` (6 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-03-31 16:43 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, Konstantin Ananyev
Change from *single* to *sync_type* to allow different
synchronisation schemes to be applied.
Mark *single* as deprecated in comments.
Add new functions to allow user to query ring sync types.
Replace direct access to *single* with appopriate function call.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/test_pdump.c | 6 +-
lib/librte_pdump/rte_pdump.c | 2 +-
lib/librte_port/rte_port_ring.c | 12 ++--
lib/librte_ring/rte_ring.c | 6 +-
lib/librte_ring/rte_ring.h | 113 ++++++++++++++++++++++++++------
lib/librte_ring/rte_ring_elem.h | 8 +--
6 files changed, 108 insertions(+), 39 deletions(-)
diff --git a/app/test/test_pdump.c b/app/test/test_pdump.c
index ad183184c..6a1180bcb 100644
--- a/app/test/test_pdump.c
+++ b/app/test/test_pdump.c
@@ -57,8 +57,7 @@ run_pdump_client_tests(void)
if (ret < 0)
return -1;
mp->flags = 0x0000;
- ring_client = rte_ring_create("SR0", RING_SIZE, rte_socket_id(),
- RING_F_SP_ENQ | RING_F_SC_DEQ);
+ ring_client = rte_ring_create("SR0", RING_SIZE, rte_socket_id(), 0);
if (ring_client == NULL) {
printf("rte_ring_create SR0 failed");
return -1;
@@ -71,9 +70,6 @@ run_pdump_client_tests(void)
}
rte_eth_dev_probing_finish(eth_dev);
- ring_client->prod.single = 0;
- ring_client->cons.single = 0;
-
printf("\n***** flags = RTE_PDUMP_FLAG_TX *****\n");
for (itr = 0; itr < NUM_ITR; itr++) {
diff --git a/lib/librte_pdump/rte_pdump.c b/lib/librte_pdump/rte_pdump.c
index 8a01ac510..65364f2c5 100644
--- a/lib/librte_pdump/rte_pdump.c
+++ b/lib/librte_pdump/rte_pdump.c
@@ -380,7 +380,7 @@ pdump_validate_ring_mp(struct rte_ring *ring, struct rte_mempool *mp)
rte_errno = EINVAL;
return -1;
}
- if (ring->prod.single || ring->cons.single) {
+ if (rte_ring_prod_single(ring) || rte_ring_cons_single(ring)) {
PDUMP_LOG(ERR, "ring with either SP or SC settings"
" is not valid for pdump, should have MP and MC settings\n");
rte_errno = EINVAL;
diff --git a/lib/librte_port/rte_port_ring.c b/lib/librte_port/rte_port_ring.c
index 47fcdd06a..2f6c050fa 100644
--- a/lib/librte_port/rte_port_ring.c
+++ b/lib/librte_port/rte_port_ring.c
@@ -44,8 +44,8 @@ rte_port_ring_reader_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->cons.single && is_multi) ||
- (!(conf->ring->cons.single) && !is_multi)) {
+ (rte_ring_cons_single(conf->ring) && is_multi) ||
+ (!rte_ring_cons_single(conf->ring) && !is_multi)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
}
@@ -171,8 +171,8 @@ rte_port_ring_writer_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->prod.single && is_multi) ||
- (!(conf->ring->prod.single) && !is_multi) ||
+ (rte_ring_prod_single(conf->ring) && is_multi) ||
+ (!rte_ring_prod_single(conf->ring) && !is_multi) ||
(conf->tx_burst_sz > RTE_PORT_IN_BURST_SIZE_MAX)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
@@ -440,8 +440,8 @@ rte_port_ring_writer_nodrop_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->prod.single && is_multi) ||
- (!(conf->ring->prod.single) && !is_multi) ||
+ (rte_ring_prod_single(conf->ring) && is_multi) ||
+ (!rte_ring_prod_single(conf->ring) && !is_multi) ||
(conf->tx_burst_sz > RTE_PORT_IN_BURST_SIZE_MAX)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index 77e5de099..fa5733907 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -106,8 +106,10 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
if (ret < 0 || ret >= (int)sizeof(r->name))
return -ENAMETOOLONG;
r->flags = flags;
- r->prod.single = (flags & RING_F_SP_ENQ) ? __IS_SP : __IS_MP;
- r->cons.single = (flags & RING_F_SC_DEQ) ? __IS_SC : __IS_MC;
+ r->prod.sync_type = (flags & RING_F_SP_ENQ) ?
+ RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
+ r->cons.sync_type = (flags & RING_F_SC_DEQ) ?
+ RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
if (flags & RING_F_EXACT_SZ) {
r->size = rte_align32pow2(count + 1);
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index 18fc5d845..d4775a063 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -61,11 +61,27 @@ enum rte_ring_queue_behavior {
#define RTE_RING_NAMESIZE (RTE_MEMZONE_NAMESIZE - \
sizeof(RTE_RING_MZ_PREFIX) + 1)
-/* structure to hold a pair of head/tail values and other metadata */
+/** prod/cons sync types */
+enum rte_ring_sync_type {
+ RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
+ RTE_RING_SYNC_ST, /**< single thread only */
+};
+
+/**
+ * structure to hold a pair of head/tail values and other metadata.
+ * Depending on sync_type format of that structure might be different,
+ * but offset for *sync_type* and *tail* values should remain the same.
+ */
struct rte_ring_headtail {
- volatile uint32_t head; /**< Prod/consumer head. */
- volatile uint32_t tail; /**< Prod/consumer tail. */
- uint32_t single; /**< True if single prod/cons */
+ volatile uint32_t head; /**< prod/consumer head. */
+ volatile uint32_t tail; /**< prod/consumer tail. */
+ RTE_STD_C11
+ union {
+ /** sync type of prod/cons */
+ enum rte_ring_sync_type sync_type;
+ /** deprecated - True if single prod/cons */
+ uint32_t single;
+ };
};
/**
@@ -116,11 +132,10 @@ struct rte_ring {
#define RING_F_EXACT_SZ 0x0004
#define RTE_RING_SZ_MASK (0x7fffffffU) /**< Ring size mask */
-/* @internal defines for passing to the enqueue dequeue worker functions */
-#define __IS_SP 1
-#define __IS_MP 0
-#define __IS_SC 1
-#define __IS_MC 0
+#define __IS_SP RTE_RING_SYNC_ST
+#define __IS_MP RTE_RING_SYNC_MT
+#define __IS_SC RTE_RING_SYNC_ST
+#define __IS_MC RTE_RING_SYNC_MT
/**
* Calculate the memory size needed for a ring
@@ -420,7 +435,7 @@ rte_ring_mp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_MP, free_space);
+ RTE_RING_SYNC_MT, free_space);
}
/**
@@ -443,7 +458,7 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_SP, free_space);
+ RTE_RING_SYNC_ST, free_space);
}
/**
@@ -470,7 +485,7 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->prod.single, free_space);
+ r->prod.sync_type, free_space);
}
/**
@@ -554,7 +569,7 @@ rte_ring_mc_dequeue_bulk(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_MC, available);
+ RTE_RING_SYNC_MT, available);
}
/**
@@ -578,7 +593,7 @@ rte_ring_sc_dequeue_bulk(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_SC, available);
+ RTE_RING_SYNC_ST, available);
}
/**
@@ -605,7 +620,7 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
/**
@@ -777,6 +792,62 @@ rte_ring_get_capacity(const struct rte_ring *r)
return r->capacity;
}
+/**
+ * Return sync type used by producer in the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Producer sync type value.
+ */
+static inline enum rte_ring_sync_type
+rte_ring_get_prod_sync_type(const struct rte_ring *r)
+{
+ return r->prod.sync_type;
+}
+
+/**
+ * Check is the ring for single producer.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * true if ring is SP, zero otherwise.
+ */
+static inline int
+rte_ring_prod_single(const struct rte_ring *r)
+{
+ return (rte_ring_get_prod_sync_type(r) == RTE_RING_SYNC_ST);
+}
+
+/**
+ * Return sync type used by consumer in the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Consumer sync type value.
+ */
+static inline enum rte_ring_sync_type
+rte_ring_get_cons_sync_type(const struct rte_ring *r)
+{
+ return r->cons.sync_type;
+}
+
+/**
+ * Check is the ring for single consumer.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * true if ring is SC, zero otherwise.
+ */
+static inline int
+rte_ring_cons_single(const struct rte_ring *r)
+{
+ return (rte_ring_get_cons_sync_type(r) == RTE_RING_SYNC_ST);
+}
+
/**
* Dump the status of all rings on the console
*
@@ -820,7 +891,7 @@ rte_ring_mp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_MP, free_space);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_MT, free_space);
}
/**
@@ -843,7 +914,7 @@ rte_ring_sp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_SP, free_space);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST, free_space);
}
/**
@@ -870,7 +941,7 @@ rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE,
- r->prod.single, free_space);
+ r->prod.sync_type, free_space);
}
/**
@@ -898,7 +969,7 @@ rte_ring_mc_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_MC, available);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_MT, available);
}
/**
@@ -923,7 +994,7 @@ rte_ring_sc_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_SC, available);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST, available);
}
/**
@@ -951,7 +1022,7 @@ rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
{
return __rte_ring_do_dequeue(r, obj_table, n,
RTE_RING_QUEUE_VARIABLE,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 663addc73..28f9836e6 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -570,7 +570,7 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->prod.single, free_space);
+ RTE_RING_QUEUE_FIXED, r->prod.sync_type, free_space);
}
/**
@@ -734,7 +734,7 @@ rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
unsigned int esize, unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->cons.single, available);
+ RTE_RING_QUEUE_FIXED, r->cons.sync_type, available);
}
/**
@@ -902,7 +902,7 @@ rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_VARIABLE, r->prod.single, free_space);
+ RTE_RING_QUEUE_VARIABLE, r->prod.sync_type, free_space);
}
/**
@@ -995,7 +995,7 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
{
return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
RTE_RING_QUEUE_VARIABLE,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
#ifdef __cplusplus
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v1 3/8] ring: introduce RTS ring mode
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 1/8] test/ring: add contention stress test Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 2/8] ring: prepare ring to allow new sync schemes Konstantin Ananyev
@ 2020-03-31 16:43 ` Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 4/8] test/ring: add contention stress test for RTS ring Konstantin Ananyev
` (5 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-03-31 16:43 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, Konstantin Ananyev
Introduce relaxed tail sync (RTS) mode for MT ring synchronization.
Aim to reduce stall times in case when ring is used on
overcommited cpus (multiple active threads on the same cpu).
The main difference from original MP/MC algorithm is that
tail value is increased not by every thread that finished enqueue/dequeue,
but only by the last one.
That allows threads to avoid spinning on ring tail value,
leaving actual tail value change to the last thread in the update queue.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 5 +-
lib/librte_ring/meson.build | 5 +-
lib/librte_ring/rte_ring.c | 100 +++++++-
lib/librte_ring/rte_ring.h | 109 ++++++++-
lib/librte_ring/rte_ring_elem.h | 86 ++++++-
lib/librte_ring/rte_ring_rts.h | 316 +++++++++++++++++++++++++
lib/librte_ring/rte_ring_rts_elem.h | 205 ++++++++++++++++
lib/librte_ring/rte_ring_rts_generic.h | 210 ++++++++++++++++
8 files changed, 1007 insertions(+), 29 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_rts.h
create mode 100644 lib/librte_ring/rte_ring_rts_elem.h
create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 917c560ad..8f5c284cc 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -18,6 +18,9 @@ SRCS-$(CONFIG_RTE_LIBRTE_RING) := rte_ring.c
SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_elem.h \
rte_ring_generic.h \
- rte_ring_c11_mem.h
+ rte_ring_c11_mem.h \
+ rte_ring_rts.h \
+ rte_ring_rts_elem.h \
+ rte_ring_rts_generic.h
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index f2f3ccc88..612936afb 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -5,7 +5,10 @@ sources = files('rte_ring.c')
headers = files('rte_ring.h',
'rte_ring_elem.h',
'rte_ring_c11_mem.h',
- 'rte_ring_generic.h')
+ 'rte_ring_generic.h',
+ 'rte_ring_rts.h',
+ 'rte_ring_rts_elem.h',
+ 'rte_ring_rts_generic.h')
# rte_ring_create_elem and rte_ring_get_memsize_elem are experimental
allow_experimental_apis = true
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index fa5733907..222eec0fb 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -45,6 +45,9 @@ EAL_REGISTER_TAILQ(rte_ring_tailq)
/* true if x is a power of 2 */
#define POWEROF2(x) ((((x)-1) & (x)) == 0)
+/* by default set head/tail distance as 1/8 of ring capacity */
+#define HTD_MAX_DEF 8
+
/* return the size of memory occupied by a ring */
ssize_t
rte_ring_get_memsize_elem(unsigned int esize, unsigned int count)
@@ -79,11 +82,84 @@ rte_ring_get_memsize(unsigned int count)
return rte_ring_get_memsize_elem(sizeof(void *), count);
}
+/*
+ * internal helper function to reset prod/cons head-tail values.
+ */
+static void
+reset_headtail(void *p)
+{
+ struct rte_ring_headtail *ht;
+ struct rte_ring_rts_headtail *ht_rts;
+
+ ht = p;
+ ht_rts = p;
+
+ switch (ht->sync_type) {
+ case RTE_RING_SYNC_MT:
+ case RTE_RING_SYNC_ST:
+ ht->head = 0;
+ ht->tail = 0;
+ break;
+ case RTE_RING_SYNC_MT_RTS:
+ ht_rts->head.raw = 0;
+ ht_rts->tail.raw = 0;
+ break;
+ default:
+ /* unknown sync mode */
+ RTE_ASSERT(0);
+ }
+}
+
void
rte_ring_reset(struct rte_ring *r)
{
- r->prod.head = r->cons.head = 0;
- r->prod.tail = r->cons.tail = 0;
+ reset_headtail(&r->prod);
+ reset_headtail(&r->cons);
+}
+
+/*
+ * helper function, calculates sync_type values for prod and cons
+ * based on input flags. Returns zero at success or negative
+ * errno value otherwise.
+ */
+static int
+get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
+ enum rte_ring_sync_type *cons_st)
+{
+ static const uint32_t prod_st_flags =
+ (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
+ static const uint32_t cons_st_flags =
+ (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
+
+ switch (flags & prod_st_flags) {
+ case 0:
+ *prod_st = RTE_RING_SYNC_MT;
+ break;
+ case RING_F_SP_ENQ:
+ *prod_st = RTE_RING_SYNC_ST;
+ break;
+ case RING_F_MP_RTS_ENQ:
+ *prod_st = RTE_RING_SYNC_MT_RTS;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ switch (flags & cons_st_flags) {
+ case 0:
+ *cons_st = RTE_RING_SYNC_MT;
+ break;
+ case RING_F_SC_DEQ:
+ *cons_st = RTE_RING_SYNC_ST;
+ break;
+ case RING_F_MC_RTS_DEQ:
+ *cons_st = RTE_RING_SYNC_MT_RTS;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
}
int
@@ -100,16 +176,20 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
RTE_CACHE_LINE_MASK) != 0);
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
+ offsetof(struct rte_ring_rts_headtail, sync_type));
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
+ offsetof(struct rte_ring_rts_headtail, tail.val.pos));
+
/* init the ring structure */
memset(r, 0, sizeof(*r));
ret = strlcpy(r->name, name, sizeof(r->name));
if (ret < 0 || ret >= (int)sizeof(r->name))
return -ENAMETOOLONG;
r->flags = flags;
- r->prod.sync_type = (flags & RING_F_SP_ENQ) ?
- RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
- r->cons.sync_type = (flags & RING_F_SC_DEQ) ?
- RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
+ ret = get_sync_type(flags, &r->prod.sync_type, &r->cons.sync_type);
+ if (ret != 0)
+ return ret;
if (flags & RING_F_EXACT_SZ) {
r->size = rte_align32pow2(count + 1);
@@ -126,8 +206,12 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
r->mask = count - 1;
r->capacity = r->mask;
}
- r->prod.head = r->cons.head = 0;
- r->prod.tail = r->cons.tail = 0;
+
+ /* set default values for head-tail distance */
+ if (flags & RING_F_MP_RTS_ENQ)
+ rte_ring_set_prod_htd_max(r, r->capacity / HTD_MAX_DEF);
+ if (flags & RING_F_MC_RTS_DEQ)
+ rte_ring_set_cons_htd_max(r, r->capacity / HTD_MAX_DEF);
return 0;
}
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index d4775a063..2b42a0211 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -65,10 +65,13 @@ enum rte_ring_queue_behavior {
enum rte_ring_sync_type {
RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
RTE_RING_SYNC_ST, /**< single thread only */
+#ifdef ALLOW_EXPERIMENTAL_API
+ RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
+#endif
};
/**
- * structure to hold a pair of head/tail values and other metadata.
+ * structures to hold a pair of head/tail values and other metadata.
* Depending on sync_type format of that structure might be different,
* but offset for *sync_type* and *tail* values should remain the same.
*/
@@ -84,6 +87,21 @@ struct rte_ring_headtail {
};
};
+union rte_ring_ht_poscnt {
+ uint64_t raw;
+ struct {
+ uint32_t cnt; /**< head/tail reference counter */
+ uint32_t pos; /**< head/tail position */
+ } val;
+};
+
+struct rte_ring_rts_headtail {
+ volatile union rte_ring_ht_poscnt tail;
+ enum rte_ring_sync_type sync_type; /**< sync type of prod/cons */
+ uint32_t htd_max; /**< max allowed distance between head/tail */
+ volatile union rte_ring_ht_poscnt head;
+};
+
/**
* An RTE ring structure.
*
@@ -111,11 +129,21 @@ struct rte_ring {
char pad0 __rte_cache_aligned; /**< empty cache line */
/** Ring producer status. */
- struct rte_ring_headtail prod __rte_cache_aligned;
+ RTE_STD_C11
+ union {
+ struct rte_ring_headtail prod;
+ struct rte_ring_rts_headtail rts_prod;
+ } __rte_cache_aligned;
+
char pad1 __rte_cache_aligned; /**< empty cache line */
/** Ring consumer status. */
- struct rte_ring_headtail cons __rte_cache_aligned;
+ RTE_STD_C11
+ union {
+ struct rte_ring_headtail cons;
+ struct rte_ring_rts_headtail rts_cons;
+ } __rte_cache_aligned;
+
char pad2 __rte_cache_aligned; /**< empty cache line */
};
@@ -132,6 +160,9 @@ struct rte_ring {
#define RING_F_EXACT_SZ 0x0004
#define RTE_RING_SZ_MASK (0x7fffffffU) /**< Ring size mask */
+#define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS". */
+#define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC RTS". */
+
#define __IS_SP RTE_RING_SYNC_ST
#define __IS_MP RTE_RING_SYNC_MT
#define __IS_SC RTE_RING_SYNC_ST
@@ -461,6 +492,10 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
RTE_RING_SYNC_ST, free_space);
}
+#ifdef ALLOW_EXPERIMENTAL_API
+#include <rte_ring_rts.h>
+#endif
+
/**
* Enqueue several objects on a ring.
*
@@ -484,8 +519,21 @@ static __rte_always_inline unsigned int
rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_bulk(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_bulk(r, obj_table, n, free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_bulk(r, obj_table, n,
+ free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -619,8 +667,20 @@ static __rte_always_inline unsigned int
rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
unsigned int *available)
{
- return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_bulk(r, obj_table, n, available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_bulk(r, obj_table, n, available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_bulk(r, obj_table, n, available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -940,8 +1000,21 @@ static __rte_always_inline unsigned
rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE,
- r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_burst(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_burst(r, obj_table, n, free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_burst(r, obj_table, n,
+ free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -1020,9 +1093,21 @@ static __rte_always_inline unsigned
rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
- return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_burst(r, obj_table, n, available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_burst(r, obj_table, n, available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_burst(r, obj_table, n,
+ available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 28f9836e6..5de0850dc 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -542,6 +542,8 @@ rte_ring_sp_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
RTE_RING_QUEUE_FIXED, __IS_SP, free_space);
}
+#include <rte_ring_rts_elem.h>
+
/**
* Enqueue several objects on a ring.
*
@@ -571,6 +573,26 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
RTE_RING_QUEUE_FIXED, r->prod.sync_type, free_space);
+
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (free_space != NULL)
+ *free_space = 0;
+ return 0;
}
/**
@@ -733,8 +755,25 @@ static __rte_always_inline unsigned int
rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
unsigned int esize, unsigned int n, unsigned int *available)
{
- return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_bulk_elem(r, obj_table, esize, n,
+ available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_bulk_elem(r, obj_table, esize, n,
+ available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_bulk_elem(r, obj_table, esize,
+ n, available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (available != NULL)
+ *available = 0;
+ return 0;
}
/**
@@ -901,8 +940,25 @@ static __rte_always_inline unsigned
rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_VARIABLE, r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_burst_elem(r, obj_table, esize, n,
+ free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_burst_elem(r, obj_table, esize, n,
+ free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_burst_elem(r, obj_table, esize,
+ n, free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (free_space != NULL)
+ *free_space = 0;
+ return 0;
}
/**
@@ -993,9 +1049,25 @@ static __rte_always_inline unsigned int
rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
unsigned int esize, unsigned int n, unsigned int *available)
{
- return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_VARIABLE,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_burst_elem(r, obj_table, esize, n,
+ available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_burst_elem(r, obj_table, esize, n,
+ available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_burst_elem(r, obj_table, esize,
+ n, available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (available != NULL)
+ *available = 0;
+ return 0;
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_rts.h b/lib/librte_ring/rte_ring_rts.h
new file mode 100644
index 000000000..18404fe48
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts.h
@@ -0,0 +1,316 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_H_
+#define _RTE_RING_RTS_H_
+
+/**
+ * @file rte_ring_rts.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring.h> instead.
+ *
+ * Contains functions for Relaxed Tail Sync (RTS) ring mode.
+ * The main idea remains the same as for our original MP/MC synchronization
+ * mechanism.
+ * The main difference is that tail value is increased not
+ * by every thread that finished enqueue/dequeue,
+ * but only by the last one doing enqueue/dequeue.
+ * That allows threads to skip spinning on tail value,
+ * leaving actual tail value change to last thread in the update queue.
+ * RTS requires 2 64-bit CAS for each enqueue(/dequeue) operation:
+ * one for head update, second for tail update.
+ * As a gain it allows thread to avoid spinning/waiting on tail value.
+ * In comparision original MP/MC algorithm requires one 32-bit CAS
+ * for head update and waiting/spinning on tail value.
+ *
+ * Brief outline:
+ * - introduce refcnt for both head and tail.
+ * - increment head.refcnt for each head.value update
+ * - write head:value and head:refcnt atomically (64-bit CAS)
+ * - move tail.value ahead only when tail.refcnt + 1 == head.refcnt
+ * - increment tail.refcnt when each enqueue/dequeue op finishes
+ * (no matter is tail:value going to change or not)
+ * - write tail.value and tail.recnt atomically (64-bit CAS)
+ *
+ * To avoid producer/consumer starvation:
+ * - limit max allowed distance between head and tail value (HTD_MAX).
+ * I.E. thread is allowed to proceed with changing head.value,
+ * only when: head.value - tail.value <= HTD_MAX
+ * HTD_MAX is an optional parameter.
+ * With HTD_MAX == 0 we'll have fully serialized ring -
+ * i.e. only one thread at a time will be able to enqueue/dequeue
+ * to/from the ring.
+ * With HTD_MAX >= ring.capacity - no limitation.
+ * By default HTD_MAX == ring.capacity / 8.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_rts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_enqueue(struct rte_ring *r, void * const *obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_rts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_rts_update_tail(&r->rts_prod);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_dequeue(struct rte_ring *r, void **obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_rts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_rts_update_tail(&r->rts_cons);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_rts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_rts_dequeue_bulk(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ available);
+}
+
+/**
+ * Return producer max Head-Tail-Distance (HTD).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Producer HTD value, if producer is set in appropriate sync mode,
+ * or UINT32_MAX otherwise.
+ */
+__rte_experimental
+static inline uint32_t
+rte_ring_get_prod_htd_max(const struct rte_ring *r)
+{
+ if (r->prod.sync_type == RTE_RING_SYNC_MT_RTS)
+ return r->rts_prod.htd_max;
+ return UINT32_MAX;
+}
+
+/**
+ * Set producer max Head-Tail-Distance (HTD).
+ * Note that producer has to use appropriate sync mode (RTS).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param v
+ * new HTD value to setup.
+ * @return
+ * Zero on success, or negative error code otherwise.
+ */
+__rte_experimental
+static inline int
+rte_ring_set_prod_htd_max(struct rte_ring *r, uint32_t v)
+{
+ if (r->prod.sync_type != RTE_RING_SYNC_MT_RTS)
+ return -ENOTSUP;
+
+ r->rts_prod.htd_max = v;
+ return 0;
+}
+
+/**
+ * Return consumer max Head-Tail-Distance (HTD).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Consumer HTD value, if consumer is set in appropriate sync mode,
+ * or UINT32_MAX otherwise.
+ */
+__rte_experimental
+static inline uint32_t
+rte_ring_get_cons_htd_max(const struct rte_ring *r)
+{
+ if (r->cons.sync_type == RTE_RING_SYNC_MT_RTS)
+ return r->rts_cons.htd_max;
+ return UINT32_MAX;
+}
+
+/**
+ * Set consumer max Head-Tail-Distance (HTD).
+ * Note that consumer has to use appropriate sync mode (RTS).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param v
+ * new HTD value to setup.
+ * @return
+ * Zero on success, or negative error code otherwise.
+ */
+__rte_experimental
+static inline int
+rte_ring_set_cons_htd_max(struct rte_ring *r, uint32_t v)
+{
+ if (r->cons.sync_type != RTE_RING_SYNC_MT_RTS)
+ return -ENOTSUP;
+
+ r->rts_cons.htd_max = v;
+ return 0;
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_rts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_rts_dequeue_burst(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_RTS_H_ */
diff --git a/lib/librte_ring/rte_ring_rts_elem.h b/lib/librte_ring/rte_ring_rts_elem.h
new file mode 100644
index 000000000..71a331b23
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts_elem.h
@@ -0,0 +1,205 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_ELEM_H_
+#define _RTE_RING_RTS_ELEM_H_
+
+/**
+ * @file rte_ring_rts_elem.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ *
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring_elem.h> instead.
+ * Contains *ring_elem* functions for Relaxed Tail Sync (RTS) ring mode.
+ * for more details please refer to <rte_ring_rts.h>.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_rts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_enqueue_elem(struct rte_ring *r, void * const *obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_rts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ __rte_ring_enqueue_elems(r, head, obj_table, esize, n);
+ __rte_ring_rts_update_tail(&r->rts_prod);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_dequeue_elem(struct rte_ring *r, void **obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_rts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ __rte_ring_dequeue_elems(r, head, obj_table, esize, n);
+ __rte_ring_rts_update_tail(&r->rts_cons);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_rts_enqueue_bulk_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_rts_dequeue_bulk_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, available);
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_rts_enqueue_burst_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_rts_dequeue_burst_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_RTS_ELEM_H_ */
diff --git a/lib/librte_ring/rte_ring_rts_generic.h b/lib/librte_ring/rte_ring_rts_generic.h
new file mode 100644
index 000000000..31a37924c
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts_generic.h
@@ -0,0 +1,210 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_GENERIC_H_
+#define _RTE_RING_RTS_GENERIC_H_
+
+/**
+ * @file rte_ring_rts_generic.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for Relaxed Tail Sync (RTS) ring mode.
+ * For more information please refer to <rte_ring_rts.h>.
+ */
+
+/**
+ * @internal This function updates tail values.
+ */
+static __rte_always_inline void
+__rte_ring_rts_update_tail(struct rte_ring_rts_headtail *ht)
+{
+ union rte_ring_ht_poscnt h, ot, nt;
+
+ /*
+ * If there are other enqueues/dequeues in progress that
+ * might preceded us, then don't update tail with new value.
+ */
+
+ do {
+ ot.raw = ht->tail.raw;
+ rte_smp_rmb();
+
+ /* on 32-bit systems we have to do atomic read here */
+ h.raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->head.raw);
+
+ nt.raw = ot.raw;
+ if (++nt.val.cnt == h.val.cnt)
+ nt.val.pos = h.val.pos;
+
+ } while (rte_atomic64_cmpset(&ht->tail.raw, ot.raw, nt.raw) == 0);
+}
+
+/**
+ * @internal This function waits till head/tail distance wouldn't
+ * exceed pre-defined max value.
+ */
+static __rte_always_inline void
+__rte_ring_rts_head_wait(const struct rte_ring_rts_headtail *ht,
+ union rte_ring_ht_poscnt *h)
+{
+ uint32_t max;
+
+ max = ht->htd_max;
+ h->raw = ht->head.raw;
+ rte_smp_rmb();
+
+ while (h->val.pos - ht->tail.val.pos > max) {
+ rte_pause();
+ h->raw = ht->head.raw;
+ rte_smp_rmb();
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue.
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_rts_move_prod_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* read prod head (may spin on prod tail) */
+ __rte_ring_rts_head_wait(&r->rts_prod, &oh);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - oh.val.pos;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ return 0;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (rte_atomic64_cmpset(&r->rts_prod.head.raw,
+ oh.raw, nh.raw) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_rts_move_cons_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* read cons head (may spin on cons tail) */
+ __rte_ring_rts_head_wait(&r->rts_cons, &oh);
+
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - oh.val.pos;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ return 0;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (rte_atomic64_cmpset(&r->rts_cons.head.raw,
+ oh.raw, nh.raw) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+#endif /* _RTE_RING_RTS_GENERIC_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v1 4/8] test/ring: add contention stress test for RTS ring
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
` (2 preceding siblings ...)
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 3/8] ring: introduce RTS ring mode Konstantin Ananyev
@ 2020-03-31 16:43 ` Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 5/8] ring: introduce HTS ring mode Konstantin Ananyev
` (4 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-03-31 16:43 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, Konstantin Ananyev
Introduce new test case to test RTS ring mode under contention.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_rts_stress.c | 32 ++++++++++++++++++++++++++++++++
app/test/test_ring_stress.c | 3 +++
app/test/test_ring_stress.h | 1 +
5 files changed, 38 insertions(+)
create mode 100644 app/test/test_ring_rts_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index 4eefaa887..3e15f3791 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -79,6 +79,7 @@ SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
SRCS-y += test_ring_mpmc_stress.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
diff --git a/app/test/meson.build b/app/test/meson.build
index 827b04886..bb67a49f0 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -102,6 +102,7 @@ test_sources = files('commands.c',
'test_ring.c',
'test_ring_mpmc_stress.c',
'test_ring_perf.c',
+ 'test_ring_rts_stress.c',
'test_ring_stress.c',
'test_rwlock.c',
'test_sched.c',
diff --git a/app/test/test_ring_rts_stress.c b/app/test/test_ring_rts_stress.c
new file mode 100644
index 000000000..f5255f24c
--- /dev/null
+++ b/app/test/test_ring_rts_stress.c
@@ -0,0 +1,32 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_mc_rts_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_mp_rts_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_RTS_ENQ | RING_F_MC_RTS_DEQ);
+}
+
+const struct test test_ring_rts_stress = {
+ .name = "MT_RTS",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
index 60706f799..eab395e30 100644
--- a/app/test/test_ring_stress.c
+++ b/app/test/test_ring_stress.c
@@ -40,6 +40,9 @@ test_ring_stress(void)
n += test_ring_mpmc_stress.nb_case;
k += run_test(&test_ring_mpmc_stress);
+ n += test_ring_rts_stress.nb_case;
+ k += run_test(&test_ring_rts_stress);
+
printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
n, k, n - k);
return (k != n);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
index 5ab121fe4..206f97cb6 100644
--- a/app/test/test_ring_stress.h
+++ b/app/test/test_ring_stress.h
@@ -32,3 +32,4 @@ struct test {
};
extern const struct test test_ring_mpmc_stress;
+extern const struct test test_ring_rts_stress;
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v1 5/8] ring: introduce HTS ring mode
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
` (3 preceding siblings ...)
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 4/8] test/ring: add contention stress test for RTS ring Konstantin Ananyev
@ 2020-03-31 16:43 ` Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 6/8] test/ring: add contention stress test for HTS ring Konstantin Ananyev
` (3 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-03-31 16:43 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, Konstantin Ananyev
Introduce head/tail sync mode for MT ring synchronization.
In that mode enqueue/dequeue operation is fully serialized:
only one thread at a time is allowed to perform given op.
Suppose to reduce stall times in case when ring is used on
overcommitted cpus (multiple active threads on the same cpu).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 3 +
lib/librte_ring/meson.build | 3 +
lib/librte_ring/rte_ring.c | 20 ++-
lib/librte_ring/rte_ring.h | 31 ++++
lib/librte_ring/rte_ring_elem.h | 13 ++
lib/librte_ring/rte_ring_hts.h | 210 +++++++++++++++++++++++++
lib/librte_ring/rte_ring_hts_elem.h | 205 ++++++++++++++++++++++++
lib/librte_ring/rte_ring_hts_generic.h | 198 +++++++++++++++++++++++
8 files changed, 681 insertions(+), 2 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_hts.h
create mode 100644 lib/librte_ring/rte_ring_hts_elem.h
create mode 100644 lib/librte_ring/rte_ring_hts_generic.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 8f5c284cc..6fe500f0d 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -19,6 +19,9 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_elem.h \
rte_ring_generic.h \
rte_ring_c11_mem.h \
+ rte_ring_hts.h \
+ rte_ring_hts_elem.h \
+ rte_ring_hts_generic.h \
rte_ring_rts.h \
rte_ring_rts_elem.h \
rte_ring_rts_generic.h
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index 612936afb..8e86e037a 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -6,6 +6,9 @@ headers = files('rte_ring.h',
'rte_ring_elem.h',
'rte_ring_c11_mem.h',
'rte_ring_generic.h',
+ 'rte_ring_hts.h',
+ 'rte_ring_hts_elem.h',
+ 'rte_ring_hts_generic.h',
'rte_ring_rts.h',
'rte_ring_rts_elem.h',
'rte_ring_rts_generic.h')
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index 222eec0fb..ebe5ccf0d 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -89,9 +89,11 @@ static void
reset_headtail(void *p)
{
struct rte_ring_headtail *ht;
+ struct rte_ring_hts_headtail *ht_hts;
struct rte_ring_rts_headtail *ht_rts;
ht = p;
+ ht_hts = p;
ht_rts = p;
switch (ht->sync_type) {
@@ -104,6 +106,9 @@ reset_headtail(void *p)
ht_rts->head.raw = 0;
ht_rts->tail.raw = 0;
break;
+ case RTE_RING_SYNC_MT_HTS:
+ ht_hts->ht.raw = 0;
+ break;
default:
/* unknown sync mode */
RTE_ASSERT(0);
@@ -127,9 +132,9 @@ get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
enum rte_ring_sync_type *cons_st)
{
static const uint32_t prod_st_flags =
- (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
+ (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ | RING_F_MP_HTS_ENQ);
static const uint32_t cons_st_flags =
- (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
+ (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ | RING_F_MC_HTS_DEQ);
switch (flags & prod_st_flags) {
case 0:
@@ -141,6 +146,9 @@ get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
case RING_F_MP_RTS_ENQ:
*prod_st = RTE_RING_SYNC_MT_RTS;
break;
+ case RING_F_MP_HTS_ENQ:
+ *prod_st = RTE_RING_SYNC_MT_HTS;
+ break;
default:
return -EINVAL;
}
@@ -155,6 +163,9 @@ get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
case RING_F_MC_RTS_DEQ:
*cons_st = RTE_RING_SYNC_MT_RTS;
break;
+ case RING_F_MC_HTS_DEQ:
+ *cons_st = RTE_RING_SYNC_MT_HTS;
+ break;
default:
return -EINVAL;
}
@@ -176,6 +187,11 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
RTE_CACHE_LINE_MASK) != 0);
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
+ offsetof(struct rte_ring_hts_headtail, sync_type));
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
+ offsetof(struct rte_ring_hts_headtail, ht.pos.tail));
+
RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
offsetof(struct rte_ring_rts_headtail, sync_type));
RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index 2b42a0211..f295bd7ce 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -67,6 +67,7 @@ enum rte_ring_sync_type {
RTE_RING_SYNC_ST, /**< single thread only */
#ifdef ALLOW_EXPERIMENTAL_API
RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
+ RTE_RING_SYNC_MT_HTS, /**< multi-thread head/tail sync */
#endif
};
@@ -102,6 +103,19 @@ struct rte_ring_rts_headtail {
volatile union rte_ring_ht_poscnt head;
};
+union rte_ring_ht_pos {
+ uint64_t raw;
+ struct {
+ uint32_t head; /**< head position */
+ uint32_t tail; /**< tail position */
+ } pos;
+};
+
+struct rte_ring_hts_headtail {
+ volatile union rte_ring_ht_pos ht;
+ enum rte_ring_sync_type sync_type; /**< sync type of prod/cons */
+};
+
/**
* An RTE ring structure.
*
@@ -132,6 +146,7 @@ struct rte_ring {
RTE_STD_C11
union {
struct rte_ring_headtail prod;
+ struct rte_ring_hts_headtail hts_prod;
struct rte_ring_rts_headtail rts_prod;
} __rte_cache_aligned;
@@ -141,6 +156,7 @@ struct rte_ring {
RTE_STD_C11
union {
struct rte_ring_headtail cons;
+ struct rte_ring_hts_headtail hts_cons;
struct rte_ring_rts_headtail rts_cons;
} __rte_cache_aligned;
@@ -163,6 +179,9 @@ struct rte_ring {
#define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS". */
#define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC RTS". */
+#define RING_F_MP_HTS_ENQ 0x0020 /**< The default enqueue is "MP HTS". */
+#define RING_F_MC_HTS_DEQ 0x0040 /**< The default dequeue is "MC HTS". */
+
#define __IS_SP RTE_RING_SYNC_ST
#define __IS_MP RTE_RING_SYNC_MT
#define __IS_SC RTE_RING_SYNC_ST
@@ -493,6 +512,7 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
}
#ifdef ALLOW_EXPERIMENTAL_API
+#include <rte_ring_hts.h>
#include <rte_ring_rts.h>
#endif
@@ -528,6 +548,9 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_bulk(r, obj_table, n,
free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_bulk(r, obj_table, n,
+ free_space);
#endif
}
@@ -675,6 +698,8 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
#ifdef ALLOW_EXPERIMENTAL_API
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_bulk(r, obj_table, n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_bulk(r, obj_table, n, available);
#endif
}
@@ -1009,6 +1034,9 @@ rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_burst(r, obj_table, n,
free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_burst(r, obj_table, n,
+ free_space);
#endif
}
@@ -1102,6 +1130,9 @@ rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_burst(r, obj_table, n,
available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_burst(r, obj_table, n,
+ available);
#endif
}
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 5de0850dc..010a564c1 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -542,6 +542,7 @@ rte_ring_sp_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
RTE_RING_QUEUE_FIXED, __IS_SP, free_space);
}
+#include <rte_ring_hts_elem.h>
#include <rte_ring_rts_elem.h>
/**
@@ -585,6 +586,9 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_bulk_elem(r, obj_table, esize, n,
free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
#endif
}
@@ -766,6 +770,9 @@ rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_bulk_elem(r, obj_table, esize,
n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_bulk_elem(r, obj_table, esize,
+ n, available);
#endif
}
@@ -951,6 +958,9 @@ rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_burst_elem(r, obj_table, esize,
n, free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_burst_elem(r, obj_table, esize,
+ n, free_space);
#endif
}
@@ -1060,6 +1070,9 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_burst_elem(r, obj_table, esize,
n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_burst_elem(r, obj_table, esize,
+ n, available);
#endif
}
diff --git a/lib/librte_ring/rte_ring_hts.h b/lib/librte_ring/rte_ring_hts.h
new file mode 100644
index 000000000..062d7be6c
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts.h
@@ -0,0 +1,210 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_H_
+#define _RTE_RING_HTS_H_
+
+/**
+ * @file rte_ring_hts.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring.h> instead.
+ *
+ * Contains functions for serialized, aka Head-Tail Sync (HTS) ring mode.
+ * In that mode enqueue/dequeue operation is fully serialized:
+ * at any given moement only one enqueue/dequeue operation can proceed.
+ * This is achieved by thread is allowed to proceed with changing head.value
+ * only when head.value == tail.value.
+ * Both head and tail values are updated atomically (as one 64-bit value).
+ * To achieve that 64-bit CAS is used by head update routine.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_hts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_enqueue(struct rte_ring *r, void * const *obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_hts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_hts_update_tail(&r->hts_prod, n, 1);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_dequeue(struct rte_ring *r, void **obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_hts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_hts_update_tail(&r->hts_cons, n, 0);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_hts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_hts_dequeue_bulk(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ available);
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_hts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_hts_dequeue_burst(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_HTS_H_ */
diff --git a/lib/librte_ring/rte_ring_hts_elem.h b/lib/librte_ring/rte_ring_hts_elem.h
new file mode 100644
index 000000000..34f0d121d
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts_elem.h
@@ -0,0 +1,205 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_ELEM_H_
+#define _RTE_RING_HTS_ELEM_H_
+
+/**
+ * @file rte_ring_hts_elem.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring_elem.h> instead.
+ *
+ * Contains *ring_elem* functions for Head-Tail Sync (HTS) ring mode.
+ * for more details please refer to <rte_ring_hts.h>.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_hts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_enqueue_elem(struct rte_ring *r, void * const *obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_hts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ __rte_ring_enqueue_elems(r, head, obj_table, esize, n);
+ __rte_ring_hts_update_tail(&r->hts_prod, n, 1);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_dequeue_elem(struct rte_ring *r, void **obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_hts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ __rte_ring_dequeue_elems(r, head, obj_table, esize, n);
+ __rte_ring_hts_update_tail(&r->hts_cons, n, 0);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_hts_enqueue_bulk_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_hts_dequeue_bulk_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, available);
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_hts_enqueue_burst_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_hts_dequeue_burst_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_HTS_ELEM_H_ */
diff --git a/lib/librte_ring/rte_ring_hts_generic.h b/lib/librte_ring/rte_ring_hts_generic.h
new file mode 100644
index 000000000..0b3931ffa
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts_generic.h
@@ -0,0 +1,198 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2020 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_GENERIC_H_
+#define _RTE_RING_HTS_GENERIC_H_
+
+/**
+ * @file rte_ring_hts_generic.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for head/tail sync (HTS) ring mode.
+ * For more information please refer to <rte_ring_hts.h>.
+ */
+
+static __rte_always_inline void
+__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
+ uint32_t enqueue)
+{
+ union rte_ring_ht_pos p;
+
+ if (enqueue)
+ rte_smp_wmb();
+ else
+ rte_smp_rmb();
+
+ p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
+
+ p.pos.head = p.pos.tail + num;
+ p.pos.tail = p.pos.head;
+
+ rte_atomic64_set((rte_atomic64_t *)(uintptr_t)&ht->ht.raw, p.raw);
+}
+
+/**
+ * @internal waits till tail will become equal to head.
+ * Means no writer/reader is active for that ring.
+ * Suppose to work as serialization point.
+ */
+static __rte_always_inline void
+__rte_ring_hts_head_wait(const struct rte_ring_hts_headtail *ht,
+ union rte_ring_ht_pos *p)
+{
+ p->raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->ht.raw);
+
+ while (p->pos.head != p->pos.tail) {
+ rte_pause();
+ p->raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->ht.raw);
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_prod_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_prod, &op);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - op.pos.head;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ return 0;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (rte_atomic64_cmpset(&r->hts_prod.ht.raw,
+ op.raw, np.raw) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_cons_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_cons, &op);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - op.pos.head;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ return 0;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (rte_atomic64_cmpset(&r->hts_cons.ht.raw,
+ op.raw, np.raw) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+#endif /* _RTE_RING_HTS_GENERIC_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v1 6/8] test/ring: add contention stress test for HTS ring
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
` (4 preceding siblings ...)
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 5/8] ring: introduce HTS ring mode Konstantin Ananyev
@ 2020-03-31 16:43 ` Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 7/8] ring: introduce peek style API Konstantin Ananyev
` (2 subsequent siblings)
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-03-31 16:43 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, Konstantin Ananyev
Introduce new test case to test HTS ring mode under contention.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_hts_stress.c | 32 ++++++++++++++++++++++++++++++++
app/test/test_ring_stress.c | 3 +++
app/test/test_ring_stress.h | 1 +
5 files changed, 38 insertions(+)
create mode 100644 app/test/test_ring_hts_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index 3e15f3791..72f64e30e 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -78,6 +78,7 @@ SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
SRCS-y += test_ring_mpmc_stress.c
+SRCS-y += test_ring_hts_stress.c
SRCS-y += test_ring_perf.c
SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
diff --git a/app/test/meson.build b/app/test/meson.build
index bb67a49f0..d6504a08a 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -101,6 +101,7 @@ test_sources = files('commands.c',
'test_rib6.c',
'test_ring.c',
'test_ring_mpmc_stress.c',
+ 'test_ring_hts_stress.c',
'test_ring_perf.c',
'test_ring_rts_stress.c',
'test_ring_stress.c',
diff --git a/app/test/test_ring_hts_stress.c b/app/test/test_ring_hts_stress.c
new file mode 100644
index 000000000..edc9175cb
--- /dev/null
+++ b/app/test/test_ring_hts_stress.c
@@ -0,0 +1,32 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_mc_hts_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_mp_hts_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_HTS_ENQ | RING_F_MC_HTS_DEQ);
+}
+
+const struct test test_ring_hts_stress = {
+ .name = "MT_HTS",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
index eab395e30..29a1368d7 100644
--- a/app/test/test_ring_stress.c
+++ b/app/test/test_ring_stress.c
@@ -43,6 +43,9 @@ test_ring_stress(void)
n += test_ring_rts_stress.nb_case;
k += run_test(&test_ring_rts_stress);
+ n += test_ring_hts_stress.nb_case;
+ k += run_test(&test_ring_hts_stress);
+
printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
n, k, n - k);
return (k != n);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
index 206f97cb6..de5c3750e 100644
--- a/app/test/test_ring_stress.h
+++ b/app/test/test_ring_stress.h
@@ -33,3 +33,4 @@ struct test {
extern const struct test test_ring_mpmc_stress;
extern const struct test test_ring_rts_stress;
+extern const struct test test_ring_hts_stress;
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v1 7/8] ring: introduce peek style API
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
` (5 preceding siblings ...)
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 6/8] test/ring: add contention stress test for HTS ring Konstantin Ananyev
@ 2020-03-31 16:43 ` Konstantin Ananyev
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 8/8] test/ring: add stress test for MT peek API Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-03-31 16:43 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, Konstantin Ananyev
For rings with producer/consumer in RTE_RING_SYNC_ST, RTE_RING_SYNC_MT_HTS
mode, provide an ability to split enqueue/dequeue operation
into two phases:
- enqueue/dequeue start
- enqueue/dequeue finish
That allows user to inspect objects in the ring without removing
them from it (aka MT safe peek).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 1 +
lib/librte_ring/meson.build | 1 +
lib/librte_ring/rte_ring_c11_mem.h | 44 +++
lib/librte_ring/rte_ring_elem.h | 4 +
lib/librte_ring/rte_ring_generic.h | 48 ++++
lib/librte_ring/rte_ring_hts_generic.h | 47 ++-
lib/librte_ring/rte_ring_peek.h | 379 +++++++++++++++++++++++++
7 files changed, 519 insertions(+), 5 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_peek.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 6fe500f0d..5f8662737 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -22,6 +22,7 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_hts.h \
rte_ring_hts_elem.h \
rte_ring_hts_generic.h \
+ rte_ring_peek.h \
rte_ring_rts.h \
rte_ring_rts_elem.h \
rte_ring_rts_generic.h
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index 8e86e037a..f5f84dc6e 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -9,6 +9,7 @@ headers = files('rte_ring.h',
'rte_ring_hts.h',
'rte_ring_hts_elem.h',
'rte_ring_hts_generic.h',
+ 'rte_ring_peek.h',
'rte_ring_rts.h',
'rte_ring_rts_elem.h',
'rte_ring_rts_generic.h')
diff --git a/lib/librte_ring/rte_ring_c11_mem.h b/lib/librte_ring/rte_ring_c11_mem.h
index 0fb73a337..bb3096721 100644
--- a/lib/librte_ring/rte_ring_c11_mem.h
+++ b/lib/librte_ring/rte_ring_c11_mem.h
@@ -10,6 +10,50 @@
#ifndef _RTE_RING_C11_MEM_H_
#define _RTE_RING_C11_MEM_H_
+/**
+ * @internal get current tail value.
+ * This function should be used only for single thread producer/consumer.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_st_get_tail(struct rte_ring_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t h, n, t;
+
+ h = ht->head;
+ t = ht->tail;
+ n = h - t;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = h;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail.
+ * This function should be used only for single thread producer/consumer.
+ * Should be used only in conjunction with __rte_ring_st_get_tail.
+ */
+static __rte_always_inline void
+__rte_ring_st_set_head_tail(struct rte_ring_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
+{
+ uint32_t pos;
+
+ RTE_SET_USED(enqueue);
+
+ pos = tail + num;
+ ht->head = pos;
+ __atomic_store_n(&ht->tail, pos, __ATOMIC_RELEASE);
+}
+
static __rte_always_inline void
update_tail(struct rte_ring_headtail *ht, uint32_t old_val, uint32_t new_val,
uint32_t single, uint32_t enqueue)
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 010a564c1..5bf7c1c1b 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -1083,6 +1083,10 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
return 0;
}
+#ifdef ALLOW_EXPERIMENTAL_API
+#include <rte_ring_peek.h>
+#endif
+
#ifdef __cplusplus
}
#endif
diff --git a/lib/librte_ring/rte_ring_generic.h b/lib/librte_ring/rte_ring_generic.h
index 953cdbbd5..9f5fdf13b 100644
--- a/lib/librte_ring/rte_ring_generic.h
+++ b/lib/librte_ring/rte_ring_generic.h
@@ -10,6 +10,54 @@
#ifndef _RTE_RING_GENERIC_H_
#define _RTE_RING_GENERIC_H_
+/**
+ * @internal get current tail value.
+ * This function should be used only for single thread producer/consumer.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_st_get_tail(struct rte_ring_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t h, n, t;
+
+ h = ht->head;
+ t = ht->tail;
+ n = h - t;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = h;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail.
+ * This function should be used only for single thread producer/consumer.
+ * Should be used only in conjunction with __rte_ring_st_get_tail.
+ */
+static __rte_always_inline void
+__rte_ring_st_set_head_tail(struct rte_ring_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
+{
+ uint32_t pos;
+
+ pos = tail + num;
+
+ if (enqueue)
+ rte_smp_wmb();
+ else
+ rte_smp_rmb();
+
+ ht->head = pos;
+ ht->tail = pos;
+}
+
static __rte_always_inline void
update_tail(struct rte_ring_headtail *ht, uint32_t old_val, uint32_t new_val,
uint32_t single, uint32_t enqueue)
diff --git a/lib/librte_ring/rte_ring_hts_generic.h b/lib/librte_ring/rte_ring_hts_generic.h
index 0b3931ffa..7eac761f9 100644
--- a/lib/librte_ring/rte_ring_hts_generic.h
+++ b/lib/librte_ring/rte_ring_hts_generic.h
@@ -18,9 +18,38 @@
* For more information please refer to <rte_ring_hts.h>.
*/
+/**
+ * @internal get current tail value.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_hts_get_tail(struct rte_ring_hts_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t n;
+ union rte_ring_ht_pos p;
+
+ p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
+ n = p.pos.head - p.pos.tail;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = p.pos.tail;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail as one atomic 64 bit operation.
+ * Should be used only in conjunction with __rte_ring_hts_get_tail.
+ */
static __rte_always_inline void
-__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
- uint32_t enqueue)
+__rte_ring_hts_set_head_tail(struct rte_ring_hts_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
{
union rte_ring_ht_pos p;
@@ -29,14 +58,22 @@ __rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
else
rte_smp_rmb();
- p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
-
- p.pos.head = p.pos.tail + num;
+ p.pos.head = tail + num;
p.pos.tail = p.pos.head;
rte_atomic64_set((rte_atomic64_t *)(uintptr_t)&ht->ht.raw, p.raw);
}
+static __rte_always_inline void
+__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
+ uint32_t enqueue)
+{
+ uint32_t tail;
+
+ num = __rte_ring_hts_get_tail(ht, &tail, num);
+ __rte_ring_hts_set_head_tail(ht, tail, num, enqueue);
+}
+
/**
* @internal waits till tail will become equal to head.
* Means no writer/reader is active for that ring.
diff --git a/lib/librte_ring/rte_ring_peek.h b/lib/librte_ring/rte_ring_peek.h
new file mode 100644
index 000000000..baefd2f7b
--- /dev/null
+++ b/lib/librte_ring/rte_ring_peek.h
@@ -0,0 +1,379 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_PEEK_H_
+#define _RTE_RING_PEEK_H_
+
+/**
+ * @file
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring_elem.h> instead.
+ *
+ * Ring Peek AP
+ * Introduction of rte_ring with serialized producer/consumer (HTS sync mode)
+ * makes possible to split public enqueue/dequeue API into two phases:
+ * - enqueue/dequeue start
+ * - enqueue/dequeue finish
+ * That allows user to inspect objects in the ring without removing them
+ * from it (aka MT safe peek).
+ * Note that right now this new API is avaialble only for two sync modes:
+ * 1) Single Producer/Single Consumer (RTE_RING_SYNC_ST)
+ * 2) Serialized Producer/Serialized Consumer (RTE_RING_SYNC_MT_HTS).
+ * It is a user responsibility to create/init ring with appropriate sync
+ * modes selected.
+ * As an example:
+ * // read 1 elem from the ring:
+ * n = rte_ring_hts_dequeue_bulk_start(ring, &obj, 1, NULL);
+ * if (n != 0) {
+ * //examine object
+ * if (object_examine(obj) == KEEP)
+ * //decided to keep it in the ring.
+ * rte_ring_hts_dequeue_finish(ring, 0);
+ * else
+ * //decided to remove it from the ring.
+ * rte_ring_hts_dequeue_finish(ring, n);
+ * }
+ * Note that between _start_ and _finish_ the ring is sort of locked -
+ * none other thread can proceed with enqueue(/dequeue) operation till
+ * _finish_ will complete.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/**
+ * @internal This function moves prod head value.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_enqueue_start(struct rte_ring *r, uint32_t n,
+ enum rte_ring_queue_behavior behavior, uint32_t *free_space)
+{
+ uint32_t free, head, next;
+
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_move_prod_head(r, RTE_RING_SYNC_ST, n,
+ behavior, &head, &next, &free);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ n = __rte_ring_hts_move_prod_head(r, n, behavior,
+ &head, &free);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ n = 0;
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * Start to enqueue several objects on the ring.
+ * Note that no actual objects are put in the queue by this function,
+ * it just reserves for user such ability.
+ * User has to call appropriate enqueue_finish() to copy objects into the
+ * queue and complete given enqueue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects that can be enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_enqueue_bulk_start(struct rte_ring *r, unsigned int n,
+ unsigned int *free_space)
+{
+ return __rte_ring_do_enqueue_start(r, n, RTE_RING_QUEUE_FIXED,
+ free_space);
+}
+
+/**
+ * Start to enqueue several objects on the ring.
+ * Note that no actual objects are put in the queue by this function,
+ * it just reserves for user such ability.
+ * User has to call appropriate enqueue_finish() to copy objects into the
+ * queue and complete given enqueue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * Actual number of objects that can be enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_enqueue_burst_start(struct rte_ring *r, unsigned int n,
+ unsigned int *free_space)
+{
+ return __rte_ring_do_enqueue_start(r, n, RTE_RING_QUEUE_VARIABLE,
+ free_space);
+}
+
+/**
+ * Complete to enqueue several objects on the ring.
+ * Note that number of objects to enqueue should not exceed previous
+ * enqueue_start return value.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of objects.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to add to the ring from the obj_table.
+ */
+__rte_experimental
+static __rte_always_inline void
+rte_ring_enqueue_elem_finish(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n)
+{
+ uint32_t tail;
+
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_st_get_tail(&r->prod, &tail, n);
+ if (n != 0)
+ __rte_ring_enqueue_elems(r, tail, obj_table, esize, n);
+ __rte_ring_st_set_head_tail(&r->prod, tail, n, 1);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ n = __rte_ring_hts_get_tail(&r->hts_prod, &tail, n);
+ if (n != 0)
+ __rte_ring_enqueue_elems(r, tail, obj_table, esize, n);
+ __rte_ring_hts_set_head_tail(&r->hts_prod, tail, n, 1);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ }
+}
+
+/**
+ * Complete to enqueue several objects on the ring.
+ * Note that number of objects to enqueue should not exceed previous
+ * enqueue_start return value.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of objects.
+ * @param n
+ * The number of objects to add to the ring from the obj_table.
+ */
+__rte_experimental
+static __rte_always_inline void
+rte_ring_enqueue_finish(struct rte_ring *r, void * const *obj_table,
+ unsigned int n)
+{
+ rte_ring_enqueue_elem_finish(r, obj_table, sizeof(uintptr_t), n);
+}
+
+/**
+ * @internal This function moves cons head value and copies up to *n*
+ * objects from the ring to the user provided obj_table.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_dequeue_start(struct rte_ring *r, void **obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t avail, head, next;
+
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_move_cons_head(r, RTE_RING_SYNC_ST, n,
+ behavior, &head, &next, &avail);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ n = __rte_ring_hts_move_cons_head(r, n, behavior,
+ &head, &avail);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ n = 0;
+ }
+
+ if (n != 0)
+ __rte_ring_dequeue_elems(r, head, obj_table, esize, n);
+
+ if (available != NULL)
+ *available = avail - n;
+ return n;
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_bulk_elem_start(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_dequeue_start(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, available);
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * Actual number of objects dequeued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_bulk_start(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return rte_ring_dequeue_bulk_elem_start(r, obj_table, sizeof(uintptr_t),
+ n, available);
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The actual number of objects dequeued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_burst_elem_start(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_dequeue_start(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The actual number of objects dequeued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_burst_start(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return rte_ring_dequeue_burst_elem_start(r, obj_table,
+ sizeof(uintptr_t), n, available);
+}
+
+/**
+ * Complete to dequeue several objects from the ring.
+ * Note that number of objects to dequeue should not exceed previous
+ * dequeue_start return value.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param n
+ * The number of objects to add to the ring from the obj_table.
+ */
+__rte_experimental
+static __rte_always_inline void
+rte_ring_dequeue_finish(struct rte_ring *r, unsigned int n)
+{
+ uint32_t tail;
+
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_st_get_tail(&r->cons, &tail, n);
+ __rte_ring_st_set_head_tail(&r->cons, tail, n, 0);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ __rte_ring_hts_update_tail(&r->hts_cons, n, 0);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ }
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_PEEK_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v1 8/8] test/ring: add stress test for MT peek API
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
` (6 preceding siblings ...)
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 7/8] ring: introduce peek style API Konstantin Ananyev
@ 2020-03-31 16:43 ` Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
8 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-03-31 16:43 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, Konstantin Ananyev
Introduce new test case to test MT peek API.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_peek_stress.c | 43 ++++++++++++++++++++++++++++++++
app/test/test_ring_stress.c | 3 +++
app/test/test_ring_stress.h | 1 +
5 files changed, 49 insertions(+)
create mode 100644 app/test/test_ring_peek_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index 72f64e30e..f4c987a98 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -80,6 +80,7 @@ SRCS-y += test_ring.c
SRCS-y += test_ring_mpmc_stress.c
SRCS-y += test_ring_hts_stress.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_peek_stress.c
SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
diff --git a/app/test/meson.build b/app/test/meson.build
index d6504a08a..cfcf5b81d 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -102,6 +102,7 @@ test_sources = files('commands.c',
'test_ring.c',
'test_ring_mpmc_stress.c',
'test_ring_hts_stress.c',
+ 'test_ring_peek_stress.c',
'test_ring_perf.c',
'test_ring_rts_stress.c',
'test_ring_stress.c',
diff --git a/app/test/test_ring_peek_stress.c b/app/test/test_ring_peek_stress.c
new file mode 100644
index 000000000..cfc82d728
--- /dev/null
+++ b/app/test/test_ring_peek_stress.c
@@ -0,0 +1,43 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+#include <rte_ring_elem.h>
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ uint32_t m;
+
+ m = rte_ring_dequeue_bulk_start(r, obj, n, avail);
+ n = (m == n) ? n : 0;
+ rte_ring_dequeue_finish(r, n);
+ return n;
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ uint32_t m;
+
+ m = rte_ring_enqueue_bulk_start(r, n, free);
+ n = (m == n) ? n : 0;
+ rte_ring_enqueue_finish(r, obj, n);
+ return n;
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_HTS_ENQ | RING_F_MC_HTS_DEQ);
+}
+
+const struct test test_ring_peek_stress = {
+ .name = "MT_PEEK",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
index 29a1368d7..853fcc190 100644
--- a/app/test/test_ring_stress.c
+++ b/app/test/test_ring_stress.c
@@ -46,6 +46,9 @@ test_ring_stress(void)
n += test_ring_hts_stress.nb_case;
k += run_test(&test_ring_hts_stress);
+ n += test_ring_peek_stress.nb_case;
+ k += run_test(&test_ring_peek_stress);
+
printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
n, k, n - k);
return (k != n);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
index de5c3750e..8fac7f4d2 100644
--- a/app/test/test_ring_stress.h
+++ b/app/test/test_ring_stress.h
@@ -34,3 +34,4 @@ struct test {
extern const struct test test_ring_mpmc_stress;
extern const struct test test_ring_rts_stress;
extern const struct test test_ring_hts_stress;
+extern const struct test test_ring_peek_stress;
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [RFC 0/6] New sync modes for ring
2020-03-30 23:37 ` Honnappa Nagarahalli
@ 2020-03-31 17:21 ` Ananyev, Konstantin
0 siblings, 0 replies; 146+ messages in thread
From: Ananyev, Konstantin @ 2020-03-31 17:21 UTC (permalink / raw)
To: Honnappa Nagarahalli, dev; +Cc: olivier.matz, nd, nd
> <snip>
> >
> > > >
> > > > > Subject: [dpdk-dev] [RFC 0/6] New sync modes for ring
> > > > >
> > > > > Upfront note - that RFC is not a complete patch.
> > > > > It introduces an ABI breakage, plus it doesn't update ring_elem
> > > > > code properly,
> > > > As per the current rules, these changes (in the current form) will
> > > > be accepted only for 20.11 release. How do we address this for
> > > > immediate
> > > requirements like RCU defer APIs?
> > >
> > > I think I found a way to introduce these new modes without API/ABI
> > breakage.
> > > Working on v1 right now. Plan to submit it by end of that week/start
> > > of next one.
> > ok
> RCU defer APIs require the rte_ring_xxx_elem versions. I guess you are adding those as well.
Yes, I added it into V1, please have a look.
Also I made 'legacy' peek API (enqueue/dequeue_start/finish) to call
'elem' peek API (enqueue/dequeue_elem_start/finish).
About naming: thought about changing start/finish to reserve/commit,
but decided to left as it is for now - in case you would like to go ahead
with SG API and use reserve/commit there.
Konstantin
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v2 0/9] New sync modes for ring
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 0/8] " Konstantin Ananyev
` (7 preceding siblings ...)
2020-03-31 16:43 ` [dpdk-dev] [PATCH v1 8/8] test/ring: add stress test for MT peek API Konstantin Ananyev
@ 2020-04-02 22:09 ` Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 1/9] test/ring: add contention stress test Konstantin Ananyev
` (9 more replies)
8 siblings, 10 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-02 22:09 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
V1 - v2 changes:
1. Fix compilation issues
2. Add C11 atomics support
3. Updates devtools/libabigail.abignore (workaround)
RFC - V1 changes:
1. remove ABI brekage (at least I hope I did)
2. Add support for ring_elem
3. Rework peek related API a bit
4. Rework test to make it less verbose and unite all test-cases
in one command
5. Add new test-case for MT peek API
TODO list:
1. Update docs
These days more and more customers use(/try to use) DPDK based apps within
overcommitted systems (multiple acttive threads over same pysical cores):
VM, container deployments, etc.
One quite common problem they hit:
Lock-Holder-Preemption/Lock-Waiter-Preemption with rte_ring.
LHP is quite a common problem for spin-based sync primitives
(spin-locks, etc.) on overcommitted systems.
The situation gets much worse when some sort of
fair-locking technique is used (ticket-lock, etc.).
As now not only lock-owner but also lock-waiters scheduling
order matters a lot (LWP).
These two problems are well-known for kernel within VMs:
http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
The problem with rte_ring is that while head accusion is sort of
un-fair locking, waiting on tail is very similar to ticket lock schema -
tail has to be updated in particular order.
So while head update exibits only LHP scneario,
tail wait and update can cause an LWP.
That makes current rte_ring implementation to perform
really pure on some overcommited scenarios.
While it is probably not possible to completely resolve LHP problem in
userspace only (without some kernel communication/intervention),
removing fairness in tail update can mitigate current LWP significantly.
This RFC proposes two new optional ring synchronization modes:
1) Head/Tail Sync (HTS) mode
In that mode enqueue/dequeue operation is fully serialized:
only one thread at a time is allowed to perform given op.
As another enhancement provide ability to split enqueue/dequeue
operation into two phases:
- enqueue/dequeue start
- enqueue/dequeue finish
That allows user to inspect objects in the ring without removing
them from it (aka MT safe peek).
2) Relaxed Tail Sync (RTS)
The main difference from original MP/MC algorithm is that
tail value is increased not by every thread that finished enqueue/dequeue,
but only by the last one.
That allows threads to avoid spinning on ring tail value,
leaving actual tail value change to the last thread in the update queue.
Note that these new sync modes are optional.
For current rte_ring users nothing should change
(both in terms of API/ABI and performance).
Existing sync modes MP/MC,SP/SC kept untouched, set up in the same
way (via flags and _init_), and MP/MC remains as default one.
The only thing that changed:
Format of prod/cons now could differ depending on mode selected at _init_.
So user has to stick with one sync model through whole ring lifetime.
In other words, user can't create a ring for let say SP mode and then
in the middle of data-path change his mind and start using MP_RTS mode.
For existing modes (SP/MP, SC/MC) format remains the same and
user can still use them interchangeably, though of course it is an
error prone practice.
Test results on IA (see below) show significant improvements
for average enqueue/dequeue op times on overcommitted systems.
For 'classic' DPDK deployments (one thread per core) original MP/MC
algorithm still shows best numbers, though for 64-bit target
RTS numbers are not that far away.
Numbers were produced by new UT test-case: ring_stress_autotest, i.e.:
echo ring_stress_autotest | ./dpdk-test -n 4 --lcores='...'
X86_64 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
DEQ+ENQ average cycles/obj
MP/MC HTS RTS
1thread@1core(--lcores=6-7) 8.00 8.15 8.99
2thread@2core(--lcores=6-8) 19.14 19.61 20.35
4thread@4core(--lcores=6-10) 29.43 29.79 31.82
8thread@8core(--lcores=6-14) 110.59 192.81 119.50
16thread@16core(--lcores=6-22) 461.03 813.12 495.59
32thread/@32core(--lcores='6-22,55-70') 982.90 1972.38 1160.51
2thread@1core(--lcores='6,(10-11)@7' 20140.50 23.58 25.14
4thread@2core(--lcores='6,(10-11)@7,(20-21)@8' 153680.60 76.88 80.05
8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 280314.32 294.72 318.79
16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 643176.59 1144.02 1175.14
32thread@2core(--lcores='6,(10-25)@7,(30-45)@8' 4264238.80 4627.48 4892.68
8thread@2core(--lcores='6,(10-17)@(7,8))' 321085.98 298.59 307.47
16thread@4core(--lcores='6,(20-35)@(7-10))' 1900705.61 575.35 678.29
32thread@4core(--lcores='6,(20-51)@(7-10))' 5510445.85 2164.36 2714.12
i686 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
DEQ+ENQ average cycles/obj
MP/MC HTS RTS
1thread@1core(--lcores=6-7) 7.85 12.13 11.31
2thread@2core(--lcores=6-8) 17.89 24.52 21.86
8thread@8core(--lcores=6-14) 32.58 354.20 54.58
32thread/@32core(--lcores='6-22,55-70') 813.77 6072.41 2169.91
2thread@1core(--lcores='6,(10-11)@7' 16095.00 36.06 34.74
8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 1140354.54 346.61 361.57
16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 1920417.86 1314.90 1416.65
8thread@2core(--lcores='6,(10-17)@(7,8))' 594358.61 332.70 357.74
32thread@4core(--lcores='6,(20-51)@(7-10))' 5319896.86 2836.44 3028.87
Konstantin Ananyev (9):
test/ring: add contention stress test
ring: prepare ring to allow new sync schemes
ring: introduce RTS ring mode
test/ring: add contention stress test for RTS ring
ring: introduce HTS ring mode
test/ring: add contention stress test for HTS ring
ring: introduce peek style API
test/ring: add stress test for MT peek API
ring: add C11 memory model for new sync modes
app/test/Makefile | 5 +
app/test/meson.build | 5 +
app/test/test_pdump.c | 6 +-
app/test/test_ring_hts_stress.c | 32 ++
app/test/test_ring_mpmc_stress.c | 31 ++
app/test/test_ring_peek_stress.c | 43 +++
app/test/test_ring_rts_stress.c | 32 ++
app/test/test_ring_stress.c | 57 ++++
app/test/test_ring_stress.h | 38 +++
app/test/test_ring_stress_impl.h | 444 +++++++++++++++++++++++++
devtools/libabigail.abignore | 4 +
lib/librte_pdump/rte_pdump.c | 2 +-
lib/librte_port/rte_port_ring.c | 12 +-
lib/librte_ring/Makefile | 11 +-
lib/librte_ring/meson.build | 11 +-
lib/librte_ring/rte_ring.c | 114 ++++++-
lib/librte_ring/rte_ring.h | 244 ++++++++++++--
lib/librte_ring/rte_ring_c11_mem.h | 44 +++
lib/librte_ring/rte_ring_elem.h | 105 +++++-
lib/librte_ring/rte_ring_generic.h | 48 +++
lib/librte_ring/rte_ring_hts.h | 214 ++++++++++++
lib/librte_ring/rte_ring_hts_c11_mem.h | 222 +++++++++++++
lib/librte_ring/rte_ring_hts_elem.h | 209 ++++++++++++
lib/librte_ring/rte_ring_hts_generic.h | 235 +++++++++++++
lib/librte_ring/rte_ring_peek.h | 379 +++++++++++++++++++++
lib/librte_ring/rte_ring_rts.h | 320 ++++++++++++++++++
lib/librte_ring/rte_ring_rts_c11_mem.h | 198 +++++++++++
lib/librte_ring/rte_ring_rts_elem.h | 209 ++++++++++++
lib/librte_ring/rte_ring_rts_generic.h | 210 ++++++++++++
29 files changed, 3428 insertions(+), 56 deletions(-)
create mode 100644 app/test/test_ring_hts_stress.c
create mode 100644 app/test/test_ring_mpmc_stress.c
create mode 100644 app/test/test_ring_peek_stress.c
create mode 100644 app/test/test_ring_rts_stress.c
create mode 100644 app/test/test_ring_stress.c
create mode 100644 app/test/test_ring_stress.h
create mode 100644 app/test/test_ring_stress_impl.h
create mode 100644 lib/librte_ring/rte_ring_hts.h
create mode 100644 lib/librte_ring/rte_ring_hts_c11_mem.h
create mode 100644 lib/librte_ring/rte_ring_hts_elem.h
create mode 100644 lib/librte_ring/rte_ring_hts_generic.h
create mode 100644 lib/librte_ring/rte_ring_peek.h
create mode 100644 lib/librte_ring/rte_ring_rts.h
create mode 100644 lib/librte_ring/rte_ring_rts_c11_mem.h
create mode 100644 lib/librte_ring/rte_ring_rts_elem.h
create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v2 1/9] test/ring: add contention stress test
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
@ 2020-04-02 22:09 ` Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 2/9] ring: prepare ring to allow new sync schemes Konstantin Ananyev
` (8 subsequent siblings)
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-02 22:09 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce new test-case to measure ring perfomance under contention
(miltiple producers/consumers).
Starts dequeue/enqueue loop on all available slave lcores.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 2 +
app/test/meson.build | 2 +
app/test/test_ring_mpmc_stress.c | 31 +++
app/test/test_ring_stress.c | 48 ++++
app/test/test_ring_stress.h | 35 +++
app/test/test_ring_stress_impl.h | 444 +++++++++++++++++++++++++++++++
6 files changed, 562 insertions(+)
create mode 100644 app/test/test_ring_mpmc_stress.c
create mode 100644 app/test/test_ring_stress.c
create mode 100644 app/test/test_ring_stress.h
create mode 100644 app/test/test_ring_stress_impl.h
diff --git a/app/test/Makefile b/app/test/Makefile
index 1f080d162..4eefaa887 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -77,7 +77,9 @@ SRCS-y += test_external_mem.c
SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
+SRCS-y += test_ring_mpmc_stress.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
ifeq ($(CONFIG_RTE_LIBRTE_TABLE),y)
diff --git a/app/test/meson.build b/app/test/meson.build
index 351d29cb6..827b04886 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -100,7 +100,9 @@ test_sources = files('commands.c',
'test_rib.c',
'test_rib6.c',
'test_ring.c',
+ 'test_ring_mpmc_stress.c',
'test_ring_perf.c',
+ 'test_ring_stress.c',
'test_rwlock.c',
'test_sched.c',
'test_service_cores.c',
diff --git a/app/test/test_ring_mpmc_stress.c b/app/test/test_ring_mpmc_stress.c
new file mode 100644
index 000000000..1524b0248
--- /dev/null
+++ b/app/test/test_ring_mpmc_stress.c
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_mc_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_mp_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num, 0);
+}
+
+const struct test test_ring_mpmc_stress = {
+ .name = "MP/MC",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
new file mode 100644
index 000000000..60706f799
--- /dev/null
+++ b/app/test/test_ring_stress.c
@@ -0,0 +1,48 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress.h"
+
+static int
+run_test(const struct test *test)
+{
+ int32_t rc;
+ uint32_t i, k;
+
+ for (i = 0, k = 0; i != test->nb_case; i++) {
+
+ printf("TEST-CASE %s %s START\n",
+ test->name, test->cases[i].name);
+
+ rc = test->cases[i].func(test->cases[i].wfunc);
+ k += (rc == 0);
+
+ if (rc != 0)
+ printf("TEST-CASE %s %s FAILED\n",
+ test->name, test->cases[i].name);
+ else
+ printf("TEST-CASE %s %s OK\n",
+ test->name, test->cases[i].name);
+ }
+
+ return k;
+}
+
+static int
+test_ring_stress(void)
+{
+ uint32_t n, k;
+
+ n = 0;
+ k = 0;
+
+ n += test_ring_mpmc_stress.nb_case;
+ k += run_test(&test_ring_mpmc_stress);
+
+ printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
+ n, k, n - k);
+ return (k != n);
+}
+
+REGISTER_TEST_COMMAND(ring_stress_autotest, test_ring_stress);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
new file mode 100644
index 000000000..60eac6216
--- /dev/null
+++ b/app/test/test_ring_stress.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+
+#include <inttypes.h>
+#include <stddef.h>
+#include <stdalign.h>
+#include <string.h>
+#include <stdio.h>
+#include <unistd.h>
+
+#include <rte_ring.h>
+#include <rte_cycles.h>
+#include <rte_launch.h>
+#include <rte_pause.h>
+#include <rte_random.h>
+#include <rte_malloc.h>
+#include <rte_spinlock.h>
+
+#include "test.h"
+
+struct test_case {
+ const char *name;
+ int (*func)(int (*)(void *));
+ int (*wfunc)(void *arg);
+};
+
+struct test {
+ const char *name;
+ uint32_t nb_case;
+ const struct test_case *cases;
+};
+
+extern const struct test test_ring_mpmc_stress;
diff --git a/app/test/test_ring_stress_impl.h b/app/test/test_ring_stress_impl.h
new file mode 100644
index 000000000..11476d28c
--- /dev/null
+++ b/app/test/test_ring_stress_impl.h
@@ -0,0 +1,444 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress.h"
+
+/*
+ * Measures performance of ring enqueue/dequeue under high contention
+ */
+
+#define RING_NAME "RING_STRESS"
+#define BULK_NUM 32
+#define RING_SIZE (2 * BULK_NUM * RTE_MAX_LCORE)
+
+enum {
+ WRK_CMD_STOP,
+ WRK_CMD_RUN,
+};
+
+static volatile uint32_t wrk_cmd __rte_cache_aligned;
+
+/* test run-time in seconds */
+static const uint32_t run_time = 60;
+static const uint32_t verbose;
+
+struct lcore_stat {
+ uint64_t nb_cycle;
+ struct {
+ uint64_t nb_call;
+ uint64_t nb_obj;
+ uint64_t nb_cycle;
+ uint64_t max_cycle;
+ uint64_t min_cycle;
+ } op;
+};
+
+struct lcore_arg {
+ struct rte_ring *rng;
+ struct lcore_stat stats;
+} __rte_cache_aligned;
+
+struct ring_elem {
+ uint32_t cnt[RTE_CACHE_LINE_SIZE / sizeof(uint32_t)];
+} __rte_cache_aligned;
+
+/*
+ * redefinable functions
+ */
+static uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail);
+
+static uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free);
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num);
+
+
+static void
+lcore_stat_update(struct lcore_stat *ls, uint64_t call, uint64_t obj,
+ uint64_t tm, int32_t prcs)
+{
+ ls->op.nb_call += call;
+ ls->op.nb_obj += obj;
+ ls->op.nb_cycle += tm;
+ if (prcs) {
+ ls->op.max_cycle = RTE_MAX(ls->op.max_cycle, tm);
+ ls->op.min_cycle = RTE_MIN(ls->op.min_cycle, tm);
+ }
+}
+
+static void
+lcore_op_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls)
+{
+
+ ms->op.nb_call += ls->op.nb_call;
+ ms->op.nb_obj += ls->op.nb_obj;
+ ms->op.nb_cycle += ls->op.nb_cycle;
+ ms->op.max_cycle = RTE_MAX(ms->op.max_cycle, ls->op.max_cycle);
+ ms->op.min_cycle = RTE_MIN(ms->op.min_cycle, ls->op.min_cycle);
+}
+
+static void
+lcore_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls)
+{
+ ms->nb_cycle = RTE_MAX(ms->nb_cycle, ls->nb_cycle);
+ lcore_op_stat_aggr(ms, ls);
+}
+
+static void
+lcore_stat_dump(FILE *f, uint32_t lc, const struct lcore_stat *ls)
+{
+ long double st;
+
+ st = (long double)rte_get_timer_hz() / US_PER_S;
+
+ if (lc == UINT32_MAX)
+ fprintf(f, "%s(AGGREGATE)={\n", __func__);
+ else
+ fprintf(f, "%s(lc=%u)={\n", __func__, lc);
+
+ fprintf(f, "\tnb_cycle=%" PRIu64 "(%.2Lf usec),\n",
+ ls->nb_cycle, (long double)ls->nb_cycle / st);
+
+ fprintf(f, "\tDEQ+ENQ={\n");
+
+ fprintf(f, "\t\tnb_call=%" PRIu64 ",\n", ls->op.nb_call);
+ fprintf(f, "\t\tnb_obj=%" PRIu64 ",\n", ls->op.nb_obj);
+ fprintf(f, "\t\tnb_cycle=%" PRIu64 ",\n", ls->op.nb_cycle);
+ fprintf(f, "\t\tobj/call(avg): %.2Lf\n",
+ (long double)ls->op.nb_obj / ls->op.nb_call);
+ fprintf(f, "\t\tcycles/obj(avg): %.2Lf\n",
+ (long double)ls->op.nb_cycle / ls->op.nb_obj);
+ fprintf(f, "\t\tcycles/call(avg): %.2Lf\n",
+ (long double)ls->op.nb_cycle / ls->op.nb_call);
+
+ /* if min/max cycles per call stats was collected */
+ if (ls->op.min_cycle != UINT64_MAX) {
+ fprintf(f, "\t\tmax cycles/call=%" PRIu64 "(%.2Lf usec),\n",
+ ls->op.max_cycle,
+ (long double)ls->op.max_cycle / st);
+ fprintf(f, "\t\tmin cycles/call=%" PRIu64 "(%.2Lf usec),\n",
+ ls->op.min_cycle,
+ (long double)ls->op.min_cycle / st);
+ }
+
+ fprintf(f, "\t},\n");
+ fprintf(f, "};\n");
+}
+
+static void
+fill_ring_elm(struct ring_elem *elm, uint32_t fill)
+{
+ uint32_t i;
+
+ for (i = 0; i != RTE_DIM(elm->cnt); i++)
+ elm->cnt[i] = fill;
+}
+
+static int32_t
+check_updt_elem(struct ring_elem *elm[], uint32_t num,
+ const struct ring_elem *check, const struct ring_elem *fill)
+{
+ uint32_t i;
+
+ static rte_spinlock_t dump_lock;
+
+ for (i = 0; i != num; i++) {
+ if (memcmp(check, elm[i], sizeof(*check)) != 0) {
+ rte_spinlock_lock(&dump_lock);
+ printf("%s(lc=%u, num=%u) failed at %u-th iter, "
+ "offending object: %p\n",
+ __func__, rte_lcore_id(), num, i, elm[i]);
+ rte_memdump(stdout, "expected", check, sizeof(*check));
+ rte_memdump(stdout, "result", elm[i], sizeof(elm[i]));
+ rte_spinlock_unlock(&dump_lock);
+ return -EINVAL;
+ }
+ memcpy(elm[i], fill, sizeof(*elm[i]));
+ }
+
+ return 0;
+}
+
+static int
+check_ring_op(uint32_t exp, uint32_t res, uint32_t lc,
+ const char *fname, const char *opname)
+{
+ if (exp != res) {
+ printf("%s(lc=%u) failure: %s expected: %u, returned %u\n",
+ fname, lc, opname, exp, res);
+ return -ENOSPC;
+ }
+ return 0;
+}
+
+static int
+test_worker_prcs(void *arg)
+{
+ int32_t rc;
+ uint32_t lc, n, num;
+ uint64_t cl, tm0, tm1;
+ struct lcore_arg *la;
+ struct ring_elem def_elm, loc_elm;
+ struct ring_elem *obj[2 * BULK_NUM];
+
+ la = arg;
+ lc = rte_lcore_id();
+
+ fill_ring_elm(&def_elm, UINT32_MAX);
+ fill_ring_elm(&loc_elm, lc);
+
+ while (wrk_cmd != WRK_CMD_RUN) {
+ rte_smp_rmb();
+ rte_pause();
+ }
+
+ cl = rte_rdtsc_precise();
+
+ do {
+ /* num in interval [7/8, 11/8] of BULK_NUM */
+ num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
+
+ /* reset all pointer values */
+ memset(obj, 0, sizeof(obj));
+
+ /* dequeue num elems */
+ tm0 = rte_rdtsc_precise();
+ n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
+ tm0 = rte_rdtsc_precise() - tm0;
+
+ /* check return value and objects */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_dequeue_bulk));
+ if (rc == 0)
+ rc = check_updt_elem(obj, num, &def_elm, &loc_elm);
+ if (rc != 0)
+ break;
+
+ /* enqueue num elems */
+ rte_compiler_barrier();
+ rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
+ if (rc != 0)
+ break;
+
+ tm1 = rte_rdtsc_precise();
+ n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
+ tm1 = rte_rdtsc_precise() - tm1;
+
+ /* check return value */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_enqueue_bulk));
+ if (rc != 0)
+ break;
+
+ lcore_stat_update(&la->stats, 1, num, tm0 + tm1, 1);
+
+ } while (wrk_cmd == WRK_CMD_RUN);
+
+ la->stats.nb_cycle = rte_rdtsc_precise() - cl;
+ return rc;
+}
+
+static int
+test_worker_avg(void *arg)
+{
+ int32_t rc;
+ uint32_t lc, n, num;
+ uint64_t cl;
+ struct lcore_arg *la;
+ struct ring_elem def_elm, loc_elm;
+ struct ring_elem *obj[2 * BULK_NUM];
+
+ la = arg;
+ lc = rte_lcore_id();
+
+ fill_ring_elm(&def_elm, UINT32_MAX);
+ fill_ring_elm(&loc_elm, lc);
+
+ while (wrk_cmd != WRK_CMD_RUN) {
+ rte_smp_rmb();
+ rte_pause();
+ }
+
+ cl = rte_rdtsc_precise();
+
+ do {
+ /* num in interval [7/8, 11/8] of BULK_NUM */
+ num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
+
+ /* reset all pointer values */
+ memset(obj, 0, sizeof(obj));
+
+ /* dequeue num elems */
+ n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
+
+ /* check return value and objects */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_dequeue_bulk));
+ if (rc == 0)
+ rc = check_updt_elem(obj, num, &def_elm, &loc_elm);
+ if (rc != 0)
+ break;
+
+ /* enqueue num elems */
+ rte_compiler_barrier();
+ rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
+ if (rc != 0)
+ break;
+
+ n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
+
+ /* check return value */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_enqueue_bulk));
+ if (rc != 0)
+ break;
+
+ lcore_stat_update(&la->stats, 1, num, 0, 0);
+
+ } while (wrk_cmd == WRK_CMD_RUN);
+
+ /* final stats update */
+ cl = rte_rdtsc_precise() - cl;
+ lcore_stat_update(&la->stats, 0, 0, cl, 0);
+ la->stats.nb_cycle = cl;
+
+ return rc;
+}
+
+static void
+mt1_fini(struct rte_ring *rng, void *data)
+{
+ rte_free(rng);
+ rte_free(data);
+}
+
+static int
+mt1_init(struct rte_ring **rng, void **data, uint32_t num)
+{
+ int32_t rc;
+ size_t sz;
+ uint32_t i, nr;
+ struct rte_ring *r;
+ struct ring_elem *elm;
+ void *p;
+
+ *rng = NULL;
+ *data = NULL;
+
+ sz = num * sizeof(*elm);
+ elm = rte_zmalloc(NULL, sz, __alignof__(*elm));
+ if (elm == NULL) {
+ printf("%s: alloc(%zu) for %u elems data failed",
+ __func__, sz, num);
+ return -ENOMEM;
+ }
+
+ *data = elm;
+
+ /* alloc ring */
+ nr = 2 * num;
+ sz = rte_ring_get_memsize(nr);
+ r = rte_zmalloc(NULL, sz, __alignof__(*r));
+ if (r == NULL) {
+ printf("%s: alloc(%zu) for FIFO with %u elems failed",
+ __func__, sz, nr);
+ return -ENOMEM;
+ }
+
+ *rng = r;
+
+ rc = _st_ring_init(r, RING_NAME, nr);
+ if (rc != 0) {
+ printf("%s: _st_ring_init(%p, %u) failed, error: %d(%s)\n",
+ __func__, r, nr, rc, strerror(-rc));
+ return rc;
+ }
+
+ for (i = 0; i != num; i++) {
+ fill_ring_elm(elm + i, UINT32_MAX);
+ p = elm + i;
+ if (_st_ring_enqueue_bulk(r, &p, 1, NULL) != 1)
+ break;
+ }
+
+ if (i != num) {
+ printf("%s: _st_ring_enqueue(%p, %u) returned %u\n",
+ __func__, r, num, i);
+ return -ENOSPC;
+ }
+
+ return 0;
+}
+
+static int
+test_mt1(int (*test)(void *))
+{
+ int32_t rc;
+ uint32_t lc, mc;
+ struct rte_ring *r;
+ void *data;
+ struct lcore_arg arg[RTE_MAX_LCORE];
+
+ static const struct lcore_stat init_stat = {
+ .op.min_cycle = UINT64_MAX,
+ };
+
+ rc = mt1_init(&r, &data, RING_SIZE);
+ if (rc != 0) {
+ mt1_fini(r, data);
+ return rc;
+ }
+
+ memset(arg, 0, sizeof(arg));
+
+ /* launch on all slaves */
+ RTE_LCORE_FOREACH_SLAVE(lc) {
+ arg[lc].rng = r;
+ arg[lc].stats = init_stat;
+ rte_eal_remote_launch(test, &arg[lc], lc);
+ }
+
+ /* signal worker to start test */
+ wrk_cmd = WRK_CMD_RUN;
+ rte_smp_wmb();
+
+ usleep(run_time * US_PER_S);
+
+ /* signal worker to start test */
+ wrk_cmd = WRK_CMD_STOP;
+ rte_smp_wmb();
+
+ /* wait for slaves and collect stats. */
+ mc = rte_lcore_id();
+ arg[mc].stats = init_stat;
+
+ rc = 0;
+ RTE_LCORE_FOREACH_SLAVE(lc) {
+ rc |= rte_eal_wait_lcore(lc);
+ lcore_stat_aggr(&arg[mc].stats, &arg[lc].stats);
+ if (verbose != 0)
+ lcore_stat_dump(stdout, lc, &arg[lc].stats);
+ }
+
+ lcore_stat_dump(stdout, UINT32_MAX, &arg[mc].stats);
+ mt1_fini(r, data);
+ return rc;
+}
+
+static const struct test_case tests[] = {
+ {
+ .name = "MT-WRK_ENQ_DEQ-MST_NONE-PRCS",
+ .func = test_mt1,
+ .wfunc = test_worker_prcs,
+ },
+ {
+ .name = "MT-WRK_ENQ_DEQ-MST_NONE-AVG",
+ .func = test_mt1,
+ .wfunc = test_worker_avg,
+ },
+};
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v2 2/9] ring: prepare ring to allow new sync schemes
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 1/9] test/ring: add contention stress test Konstantin Ananyev
@ 2020-04-02 22:09 ` Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 3/9] ring: introduce RTS ring mode Konstantin Ananyev
` (7 subsequent siblings)
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-02 22:09 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Change from *single* to *sync_type* to allow different
synchronisation schemes to be applied.
Mark *single* as deprecated in comments.
Add new functions to allow user to query ring sync types.
Replace direct access to *single* with appopriate function call.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/test_pdump.c | 6 +-
lib/librte_pdump/rte_pdump.c | 2 +-
lib/librte_port/rte_port_ring.c | 12 ++--
lib/librte_ring/rte_ring.c | 6 +-
lib/librte_ring/rte_ring.h | 113 ++++++++++++++++++++++++++------
lib/librte_ring/rte_ring_elem.h | 8 +--
6 files changed, 108 insertions(+), 39 deletions(-)
diff --git a/app/test/test_pdump.c b/app/test/test_pdump.c
index ad183184c..6a1180bcb 100644
--- a/app/test/test_pdump.c
+++ b/app/test/test_pdump.c
@@ -57,8 +57,7 @@ run_pdump_client_tests(void)
if (ret < 0)
return -1;
mp->flags = 0x0000;
- ring_client = rte_ring_create("SR0", RING_SIZE, rte_socket_id(),
- RING_F_SP_ENQ | RING_F_SC_DEQ);
+ ring_client = rte_ring_create("SR0", RING_SIZE, rte_socket_id(), 0);
if (ring_client == NULL) {
printf("rte_ring_create SR0 failed");
return -1;
@@ -71,9 +70,6 @@ run_pdump_client_tests(void)
}
rte_eth_dev_probing_finish(eth_dev);
- ring_client->prod.single = 0;
- ring_client->cons.single = 0;
-
printf("\n***** flags = RTE_PDUMP_FLAG_TX *****\n");
for (itr = 0; itr < NUM_ITR; itr++) {
diff --git a/lib/librte_pdump/rte_pdump.c b/lib/librte_pdump/rte_pdump.c
index 8a01ac510..65364f2c5 100644
--- a/lib/librte_pdump/rte_pdump.c
+++ b/lib/librte_pdump/rte_pdump.c
@@ -380,7 +380,7 @@ pdump_validate_ring_mp(struct rte_ring *ring, struct rte_mempool *mp)
rte_errno = EINVAL;
return -1;
}
- if (ring->prod.single || ring->cons.single) {
+ if (rte_ring_prod_single(ring) || rte_ring_cons_single(ring)) {
PDUMP_LOG(ERR, "ring with either SP or SC settings"
" is not valid for pdump, should have MP and MC settings\n");
rte_errno = EINVAL;
diff --git a/lib/librte_port/rte_port_ring.c b/lib/librte_port/rte_port_ring.c
index 47fcdd06a..2f6c050fa 100644
--- a/lib/librte_port/rte_port_ring.c
+++ b/lib/librte_port/rte_port_ring.c
@@ -44,8 +44,8 @@ rte_port_ring_reader_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->cons.single && is_multi) ||
- (!(conf->ring->cons.single) && !is_multi)) {
+ (rte_ring_cons_single(conf->ring) && is_multi) ||
+ (!rte_ring_cons_single(conf->ring) && !is_multi)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
}
@@ -171,8 +171,8 @@ rte_port_ring_writer_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->prod.single && is_multi) ||
- (!(conf->ring->prod.single) && !is_multi) ||
+ (rte_ring_prod_single(conf->ring) && is_multi) ||
+ (!rte_ring_prod_single(conf->ring) && !is_multi) ||
(conf->tx_burst_sz > RTE_PORT_IN_BURST_SIZE_MAX)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
@@ -440,8 +440,8 @@ rte_port_ring_writer_nodrop_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->prod.single && is_multi) ||
- (!(conf->ring->prod.single) && !is_multi) ||
+ (rte_ring_prod_single(conf->ring) && is_multi) ||
+ (!rte_ring_prod_single(conf->ring) && !is_multi) ||
(conf->tx_burst_sz > RTE_PORT_IN_BURST_SIZE_MAX)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index 77e5de099..fa5733907 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -106,8 +106,10 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
if (ret < 0 || ret >= (int)sizeof(r->name))
return -ENAMETOOLONG;
r->flags = flags;
- r->prod.single = (flags & RING_F_SP_ENQ) ? __IS_SP : __IS_MP;
- r->cons.single = (flags & RING_F_SC_DEQ) ? __IS_SC : __IS_MC;
+ r->prod.sync_type = (flags & RING_F_SP_ENQ) ?
+ RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
+ r->cons.sync_type = (flags & RING_F_SC_DEQ) ?
+ RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
if (flags & RING_F_EXACT_SZ) {
r->size = rte_align32pow2(count + 1);
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index 18fc5d845..d4775a063 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -61,11 +61,27 @@ enum rte_ring_queue_behavior {
#define RTE_RING_NAMESIZE (RTE_MEMZONE_NAMESIZE - \
sizeof(RTE_RING_MZ_PREFIX) + 1)
-/* structure to hold a pair of head/tail values and other metadata */
+/** prod/cons sync types */
+enum rte_ring_sync_type {
+ RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
+ RTE_RING_SYNC_ST, /**< single thread only */
+};
+
+/**
+ * structure to hold a pair of head/tail values and other metadata.
+ * Depending on sync_type format of that structure might be different,
+ * but offset for *sync_type* and *tail* values should remain the same.
+ */
struct rte_ring_headtail {
- volatile uint32_t head; /**< Prod/consumer head. */
- volatile uint32_t tail; /**< Prod/consumer tail. */
- uint32_t single; /**< True if single prod/cons */
+ volatile uint32_t head; /**< prod/consumer head. */
+ volatile uint32_t tail; /**< prod/consumer tail. */
+ RTE_STD_C11
+ union {
+ /** sync type of prod/cons */
+ enum rte_ring_sync_type sync_type;
+ /** deprecated - True if single prod/cons */
+ uint32_t single;
+ };
};
/**
@@ -116,11 +132,10 @@ struct rte_ring {
#define RING_F_EXACT_SZ 0x0004
#define RTE_RING_SZ_MASK (0x7fffffffU) /**< Ring size mask */
-/* @internal defines for passing to the enqueue dequeue worker functions */
-#define __IS_SP 1
-#define __IS_MP 0
-#define __IS_SC 1
-#define __IS_MC 0
+#define __IS_SP RTE_RING_SYNC_ST
+#define __IS_MP RTE_RING_SYNC_MT
+#define __IS_SC RTE_RING_SYNC_ST
+#define __IS_MC RTE_RING_SYNC_MT
/**
* Calculate the memory size needed for a ring
@@ -420,7 +435,7 @@ rte_ring_mp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_MP, free_space);
+ RTE_RING_SYNC_MT, free_space);
}
/**
@@ -443,7 +458,7 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_SP, free_space);
+ RTE_RING_SYNC_ST, free_space);
}
/**
@@ -470,7 +485,7 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->prod.single, free_space);
+ r->prod.sync_type, free_space);
}
/**
@@ -554,7 +569,7 @@ rte_ring_mc_dequeue_bulk(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_MC, available);
+ RTE_RING_SYNC_MT, available);
}
/**
@@ -578,7 +593,7 @@ rte_ring_sc_dequeue_bulk(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_SC, available);
+ RTE_RING_SYNC_ST, available);
}
/**
@@ -605,7 +620,7 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
/**
@@ -777,6 +792,62 @@ rte_ring_get_capacity(const struct rte_ring *r)
return r->capacity;
}
+/**
+ * Return sync type used by producer in the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Producer sync type value.
+ */
+static inline enum rte_ring_sync_type
+rte_ring_get_prod_sync_type(const struct rte_ring *r)
+{
+ return r->prod.sync_type;
+}
+
+/**
+ * Check is the ring for single producer.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * true if ring is SP, zero otherwise.
+ */
+static inline int
+rte_ring_prod_single(const struct rte_ring *r)
+{
+ return (rte_ring_get_prod_sync_type(r) == RTE_RING_SYNC_ST);
+}
+
+/**
+ * Return sync type used by consumer in the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Consumer sync type value.
+ */
+static inline enum rte_ring_sync_type
+rte_ring_get_cons_sync_type(const struct rte_ring *r)
+{
+ return r->cons.sync_type;
+}
+
+/**
+ * Check is the ring for single consumer.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * true if ring is SC, zero otherwise.
+ */
+static inline int
+rte_ring_cons_single(const struct rte_ring *r)
+{
+ return (rte_ring_get_cons_sync_type(r) == RTE_RING_SYNC_ST);
+}
+
/**
* Dump the status of all rings on the console
*
@@ -820,7 +891,7 @@ rte_ring_mp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_MP, free_space);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_MT, free_space);
}
/**
@@ -843,7 +914,7 @@ rte_ring_sp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_SP, free_space);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST, free_space);
}
/**
@@ -870,7 +941,7 @@ rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE,
- r->prod.single, free_space);
+ r->prod.sync_type, free_space);
}
/**
@@ -898,7 +969,7 @@ rte_ring_mc_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_MC, available);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_MT, available);
}
/**
@@ -923,7 +994,7 @@ rte_ring_sc_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_SC, available);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST, available);
}
/**
@@ -951,7 +1022,7 @@ rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
{
return __rte_ring_do_dequeue(r, obj_table, n,
RTE_RING_QUEUE_VARIABLE,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 663addc73..28f9836e6 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -570,7 +570,7 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->prod.single, free_space);
+ RTE_RING_QUEUE_FIXED, r->prod.sync_type, free_space);
}
/**
@@ -734,7 +734,7 @@ rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
unsigned int esize, unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->cons.single, available);
+ RTE_RING_QUEUE_FIXED, r->cons.sync_type, available);
}
/**
@@ -902,7 +902,7 @@ rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_VARIABLE, r->prod.single, free_space);
+ RTE_RING_QUEUE_VARIABLE, r->prod.sync_type, free_space);
}
/**
@@ -995,7 +995,7 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
{
return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
RTE_RING_QUEUE_VARIABLE,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
#ifdef __cplusplus
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v2 3/9] ring: introduce RTS ring mode
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 1/9] test/ring: add contention stress test Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 2/9] ring: prepare ring to allow new sync schemes Konstantin Ananyev
@ 2020-04-02 22:09 ` Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 4/9] test/ring: add contention stress test for RTS ring Konstantin Ananyev
` (6 subsequent siblings)
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-02 22:09 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce relaxed tail sync (RTS) mode for MT ring synchronization.
Aim to reduce stall times in case when ring is used on
overcommited cpus (multiple active threads on the same cpu).
The main difference from original MP/MC algorithm is that
tail value is increased not by every thread that finished enqueue/dequeue,
but only by the last one.
That allows threads to avoid spinning on ring tail value,
leaving actual tail value change to the last thread in the update queue.
check-abi.sh reports what I believe is a false-positive about
ring cons/prod changes. As a workaround, devtools/libabigail.abignore is
updated to suppress *struct ring* related errors.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
devtools/libabigail.abignore | 4 +
lib/librte_ring/Makefile | 5 +-
lib/librte_ring/meson.build | 5 +-
lib/librte_ring/rte_ring.c | 100 +++++++-
lib/librte_ring/rte_ring.h | 110 ++++++++-
lib/librte_ring/rte_ring_elem.h | 86 ++++++-
lib/librte_ring/rte_ring_rts.h | 316 +++++++++++++++++++++++++
lib/librte_ring/rte_ring_rts_elem.h | 205 ++++++++++++++++
lib/librte_ring/rte_ring_rts_generic.h | 210 ++++++++++++++++
9 files changed, 1012 insertions(+), 29 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_rts.h
create mode 100644 lib/librte_ring/rte_ring_rts_elem.h
create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
diff --git a/devtools/libabigail.abignore b/devtools/libabigail.abignore
index a59df8f13..ece014111 100644
--- a/devtools/libabigail.abignore
+++ b/devtools/libabigail.abignore
@@ -11,3 +11,7 @@
type_kind = enum
name = rte_crypto_asym_xform_type
changed_enumerators = RTE_CRYPTO_ASYM_XFORM_TYPE_LIST_END
+; Ignore updates of ring prod/cons
+[suppress_type]
+ type_kind = struct
+ name = rte_ring
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 917c560ad..8f5c284cc 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -18,6 +18,9 @@ SRCS-$(CONFIG_RTE_LIBRTE_RING) := rte_ring.c
SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_elem.h \
rte_ring_generic.h \
- rte_ring_c11_mem.h
+ rte_ring_c11_mem.h \
+ rte_ring_rts.h \
+ rte_ring_rts_elem.h \
+ rte_ring_rts_generic.h
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index f2f3ccc88..612936afb 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -5,7 +5,10 @@ sources = files('rte_ring.c')
headers = files('rte_ring.h',
'rte_ring_elem.h',
'rte_ring_c11_mem.h',
- 'rte_ring_generic.h')
+ 'rte_ring_generic.h',
+ 'rte_ring_rts.h',
+ 'rte_ring_rts_elem.h',
+ 'rte_ring_rts_generic.h')
# rte_ring_create_elem and rte_ring_get_memsize_elem are experimental
allow_experimental_apis = true
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index fa5733907..222eec0fb 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -45,6 +45,9 @@ EAL_REGISTER_TAILQ(rte_ring_tailq)
/* true if x is a power of 2 */
#define POWEROF2(x) ((((x)-1) & (x)) == 0)
+/* by default set head/tail distance as 1/8 of ring capacity */
+#define HTD_MAX_DEF 8
+
/* return the size of memory occupied by a ring */
ssize_t
rte_ring_get_memsize_elem(unsigned int esize, unsigned int count)
@@ -79,11 +82,84 @@ rte_ring_get_memsize(unsigned int count)
return rte_ring_get_memsize_elem(sizeof(void *), count);
}
+/*
+ * internal helper function to reset prod/cons head-tail values.
+ */
+static void
+reset_headtail(void *p)
+{
+ struct rte_ring_headtail *ht;
+ struct rte_ring_rts_headtail *ht_rts;
+
+ ht = p;
+ ht_rts = p;
+
+ switch (ht->sync_type) {
+ case RTE_RING_SYNC_MT:
+ case RTE_RING_SYNC_ST:
+ ht->head = 0;
+ ht->tail = 0;
+ break;
+ case RTE_RING_SYNC_MT_RTS:
+ ht_rts->head.raw = 0;
+ ht_rts->tail.raw = 0;
+ break;
+ default:
+ /* unknown sync mode */
+ RTE_ASSERT(0);
+ }
+}
+
void
rte_ring_reset(struct rte_ring *r)
{
- r->prod.head = r->cons.head = 0;
- r->prod.tail = r->cons.tail = 0;
+ reset_headtail(&r->prod);
+ reset_headtail(&r->cons);
+}
+
+/*
+ * helper function, calculates sync_type values for prod and cons
+ * based on input flags. Returns zero at success or negative
+ * errno value otherwise.
+ */
+static int
+get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
+ enum rte_ring_sync_type *cons_st)
+{
+ static const uint32_t prod_st_flags =
+ (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
+ static const uint32_t cons_st_flags =
+ (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
+
+ switch (flags & prod_st_flags) {
+ case 0:
+ *prod_st = RTE_RING_SYNC_MT;
+ break;
+ case RING_F_SP_ENQ:
+ *prod_st = RTE_RING_SYNC_ST;
+ break;
+ case RING_F_MP_RTS_ENQ:
+ *prod_st = RTE_RING_SYNC_MT_RTS;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ switch (flags & cons_st_flags) {
+ case 0:
+ *cons_st = RTE_RING_SYNC_MT;
+ break;
+ case RING_F_SC_DEQ:
+ *cons_st = RTE_RING_SYNC_ST;
+ break;
+ case RING_F_MC_RTS_DEQ:
+ *cons_st = RTE_RING_SYNC_MT_RTS;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
}
int
@@ -100,16 +176,20 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
RTE_CACHE_LINE_MASK) != 0);
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
+ offsetof(struct rte_ring_rts_headtail, sync_type));
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
+ offsetof(struct rte_ring_rts_headtail, tail.val.pos));
+
/* init the ring structure */
memset(r, 0, sizeof(*r));
ret = strlcpy(r->name, name, sizeof(r->name));
if (ret < 0 || ret >= (int)sizeof(r->name))
return -ENAMETOOLONG;
r->flags = flags;
- r->prod.sync_type = (flags & RING_F_SP_ENQ) ?
- RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
- r->cons.sync_type = (flags & RING_F_SC_DEQ) ?
- RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
+ ret = get_sync_type(flags, &r->prod.sync_type, &r->cons.sync_type);
+ if (ret != 0)
+ return ret;
if (flags & RING_F_EXACT_SZ) {
r->size = rte_align32pow2(count + 1);
@@ -126,8 +206,12 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
r->mask = count - 1;
r->capacity = r->mask;
}
- r->prod.head = r->cons.head = 0;
- r->prod.tail = r->cons.tail = 0;
+
+ /* set default values for head-tail distance */
+ if (flags & RING_F_MP_RTS_ENQ)
+ rte_ring_set_prod_htd_max(r, r->capacity / HTD_MAX_DEF);
+ if (flags & RING_F_MC_RTS_DEQ)
+ rte_ring_set_cons_htd_max(r, r->capacity / HTD_MAX_DEF);
return 0;
}
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index d4775a063..f6f084d79 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -48,6 +48,7 @@ extern "C" {
#include <rte_branch_prediction.h>
#include <rte_memzone.h>
#include <rte_pause.h>
+#include <rte_debug.h>
#define RTE_TAILQ_RING_NAME "RTE_RING"
@@ -65,10 +66,13 @@ enum rte_ring_queue_behavior {
enum rte_ring_sync_type {
RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
RTE_RING_SYNC_ST, /**< single thread only */
+#ifdef ALLOW_EXPERIMENTAL_API
+ RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
+#endif
};
/**
- * structure to hold a pair of head/tail values and other metadata.
+ * structures to hold a pair of head/tail values and other metadata.
* Depending on sync_type format of that structure might be different,
* but offset for *sync_type* and *tail* values should remain the same.
*/
@@ -84,6 +88,21 @@ struct rte_ring_headtail {
};
};
+union rte_ring_ht_poscnt {
+ uint64_t raw;
+ struct {
+ uint32_t cnt; /**< head/tail reference counter */
+ uint32_t pos; /**< head/tail position */
+ } val;
+};
+
+struct rte_ring_rts_headtail {
+ volatile union rte_ring_ht_poscnt tail;
+ enum rte_ring_sync_type sync_type; /**< sync type of prod/cons */
+ uint32_t htd_max; /**< max allowed distance between head/tail */
+ volatile union rte_ring_ht_poscnt head;
+};
+
/**
* An RTE ring structure.
*
@@ -111,11 +130,21 @@ struct rte_ring {
char pad0 __rte_cache_aligned; /**< empty cache line */
/** Ring producer status. */
- struct rte_ring_headtail prod __rte_cache_aligned;
+ RTE_STD_C11
+ union {
+ struct rte_ring_headtail prod;
+ struct rte_ring_rts_headtail rts_prod;
+ } __rte_cache_aligned;
+
char pad1 __rte_cache_aligned; /**< empty cache line */
/** Ring consumer status. */
- struct rte_ring_headtail cons __rte_cache_aligned;
+ RTE_STD_C11
+ union {
+ struct rte_ring_headtail cons;
+ struct rte_ring_rts_headtail rts_cons;
+ } __rte_cache_aligned;
+
char pad2 __rte_cache_aligned; /**< empty cache line */
};
@@ -132,6 +161,9 @@ struct rte_ring {
#define RING_F_EXACT_SZ 0x0004
#define RTE_RING_SZ_MASK (0x7fffffffU) /**< Ring size mask */
+#define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS". */
+#define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC RTS". */
+
#define __IS_SP RTE_RING_SYNC_ST
#define __IS_MP RTE_RING_SYNC_MT
#define __IS_SC RTE_RING_SYNC_ST
@@ -461,6 +493,10 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
RTE_RING_SYNC_ST, free_space);
}
+#ifdef ALLOW_EXPERIMENTAL_API
+#include <rte_ring_rts.h>
+#endif
+
/**
* Enqueue several objects on a ring.
*
@@ -484,8 +520,21 @@ static __rte_always_inline unsigned int
rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_bulk(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_bulk(r, obj_table, n, free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_bulk(r, obj_table, n,
+ free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -619,8 +668,20 @@ static __rte_always_inline unsigned int
rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
unsigned int *available)
{
- return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_bulk(r, obj_table, n, available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_bulk(r, obj_table, n, available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_bulk(r, obj_table, n, available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -940,8 +1001,21 @@ static __rte_always_inline unsigned
rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE,
- r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_burst(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_burst(r, obj_table, n, free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_burst(r, obj_table, n,
+ free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -1020,9 +1094,21 @@ static __rte_always_inline unsigned
rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
- return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_burst(r, obj_table, n, available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_burst(r, obj_table, n, available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_burst(r, obj_table, n,
+ available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 28f9836e6..5de0850dc 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -542,6 +542,8 @@ rte_ring_sp_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
RTE_RING_QUEUE_FIXED, __IS_SP, free_space);
}
+#include <rte_ring_rts_elem.h>
+
/**
* Enqueue several objects on a ring.
*
@@ -571,6 +573,26 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
RTE_RING_QUEUE_FIXED, r->prod.sync_type, free_space);
+
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (free_space != NULL)
+ *free_space = 0;
+ return 0;
}
/**
@@ -733,8 +755,25 @@ static __rte_always_inline unsigned int
rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
unsigned int esize, unsigned int n, unsigned int *available)
{
- return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_bulk_elem(r, obj_table, esize, n,
+ available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_bulk_elem(r, obj_table, esize, n,
+ available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_bulk_elem(r, obj_table, esize,
+ n, available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (available != NULL)
+ *available = 0;
+ return 0;
}
/**
@@ -901,8 +940,25 @@ static __rte_always_inline unsigned
rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_VARIABLE, r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_burst_elem(r, obj_table, esize, n,
+ free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_burst_elem(r, obj_table, esize, n,
+ free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_burst_elem(r, obj_table, esize,
+ n, free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (free_space != NULL)
+ *free_space = 0;
+ return 0;
}
/**
@@ -993,9 +1049,25 @@ static __rte_always_inline unsigned int
rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
unsigned int esize, unsigned int n, unsigned int *available)
{
- return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_VARIABLE,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_burst_elem(r, obj_table, esize, n,
+ available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_burst_elem(r, obj_table, esize, n,
+ available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_burst_elem(r, obj_table, esize,
+ n, available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (available != NULL)
+ *available = 0;
+ return 0;
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_rts.h b/lib/librte_ring/rte_ring_rts.h
new file mode 100644
index 000000000..18404fe48
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts.h
@@ -0,0 +1,316 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_H_
+#define _RTE_RING_RTS_H_
+
+/**
+ * @file rte_ring_rts.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring.h> instead.
+ *
+ * Contains functions for Relaxed Tail Sync (RTS) ring mode.
+ * The main idea remains the same as for our original MP/MC synchronization
+ * mechanism.
+ * The main difference is that tail value is increased not
+ * by every thread that finished enqueue/dequeue,
+ * but only by the last one doing enqueue/dequeue.
+ * That allows threads to skip spinning on tail value,
+ * leaving actual tail value change to last thread in the update queue.
+ * RTS requires 2 64-bit CAS for each enqueue(/dequeue) operation:
+ * one for head update, second for tail update.
+ * As a gain it allows thread to avoid spinning/waiting on tail value.
+ * In comparision original MP/MC algorithm requires one 32-bit CAS
+ * for head update and waiting/spinning on tail value.
+ *
+ * Brief outline:
+ * - introduce refcnt for both head and tail.
+ * - increment head.refcnt for each head.value update
+ * - write head:value and head:refcnt atomically (64-bit CAS)
+ * - move tail.value ahead only when tail.refcnt + 1 == head.refcnt
+ * - increment tail.refcnt when each enqueue/dequeue op finishes
+ * (no matter is tail:value going to change or not)
+ * - write tail.value and tail.recnt atomically (64-bit CAS)
+ *
+ * To avoid producer/consumer starvation:
+ * - limit max allowed distance between head and tail value (HTD_MAX).
+ * I.E. thread is allowed to proceed with changing head.value,
+ * only when: head.value - tail.value <= HTD_MAX
+ * HTD_MAX is an optional parameter.
+ * With HTD_MAX == 0 we'll have fully serialized ring -
+ * i.e. only one thread at a time will be able to enqueue/dequeue
+ * to/from the ring.
+ * With HTD_MAX >= ring.capacity - no limitation.
+ * By default HTD_MAX == ring.capacity / 8.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_rts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_enqueue(struct rte_ring *r, void * const *obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_rts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_rts_update_tail(&r->rts_prod);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_dequeue(struct rte_ring *r, void **obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_rts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_rts_update_tail(&r->rts_cons);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_rts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_rts_dequeue_bulk(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ available);
+}
+
+/**
+ * Return producer max Head-Tail-Distance (HTD).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Producer HTD value, if producer is set in appropriate sync mode,
+ * or UINT32_MAX otherwise.
+ */
+__rte_experimental
+static inline uint32_t
+rte_ring_get_prod_htd_max(const struct rte_ring *r)
+{
+ if (r->prod.sync_type == RTE_RING_SYNC_MT_RTS)
+ return r->rts_prod.htd_max;
+ return UINT32_MAX;
+}
+
+/**
+ * Set producer max Head-Tail-Distance (HTD).
+ * Note that producer has to use appropriate sync mode (RTS).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param v
+ * new HTD value to setup.
+ * @return
+ * Zero on success, or negative error code otherwise.
+ */
+__rte_experimental
+static inline int
+rte_ring_set_prod_htd_max(struct rte_ring *r, uint32_t v)
+{
+ if (r->prod.sync_type != RTE_RING_SYNC_MT_RTS)
+ return -ENOTSUP;
+
+ r->rts_prod.htd_max = v;
+ return 0;
+}
+
+/**
+ * Return consumer max Head-Tail-Distance (HTD).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Consumer HTD value, if consumer is set in appropriate sync mode,
+ * or UINT32_MAX otherwise.
+ */
+__rte_experimental
+static inline uint32_t
+rte_ring_get_cons_htd_max(const struct rte_ring *r)
+{
+ if (r->cons.sync_type == RTE_RING_SYNC_MT_RTS)
+ return r->rts_cons.htd_max;
+ return UINT32_MAX;
+}
+
+/**
+ * Set consumer max Head-Tail-Distance (HTD).
+ * Note that consumer has to use appropriate sync mode (RTS).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param v
+ * new HTD value to setup.
+ * @return
+ * Zero on success, or negative error code otherwise.
+ */
+__rte_experimental
+static inline int
+rte_ring_set_cons_htd_max(struct rte_ring *r, uint32_t v)
+{
+ if (r->cons.sync_type != RTE_RING_SYNC_MT_RTS)
+ return -ENOTSUP;
+
+ r->rts_cons.htd_max = v;
+ return 0;
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_rts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_rts_dequeue_burst(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_RTS_H_ */
diff --git a/lib/librte_ring/rte_ring_rts_elem.h b/lib/librte_ring/rte_ring_rts_elem.h
new file mode 100644
index 000000000..71a331b23
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts_elem.h
@@ -0,0 +1,205 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_ELEM_H_
+#define _RTE_RING_RTS_ELEM_H_
+
+/**
+ * @file rte_ring_rts_elem.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ *
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring_elem.h> instead.
+ * Contains *ring_elem* functions for Relaxed Tail Sync (RTS) ring mode.
+ * for more details please refer to <rte_ring_rts.h>.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_rts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_enqueue_elem(struct rte_ring *r, void * const *obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_rts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ __rte_ring_enqueue_elems(r, head, obj_table, esize, n);
+ __rte_ring_rts_update_tail(&r->rts_prod);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_dequeue_elem(struct rte_ring *r, void **obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_rts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ __rte_ring_dequeue_elems(r, head, obj_table, esize, n);
+ __rte_ring_rts_update_tail(&r->rts_cons);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_rts_enqueue_bulk_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_rts_dequeue_bulk_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, available);
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_rts_enqueue_burst_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_rts_dequeue_burst_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_RTS_ELEM_H_ */
diff --git a/lib/librte_ring/rte_ring_rts_generic.h b/lib/librte_ring/rte_ring_rts_generic.h
new file mode 100644
index 000000000..31a37924c
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts_generic.h
@@ -0,0 +1,210 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_GENERIC_H_
+#define _RTE_RING_RTS_GENERIC_H_
+
+/**
+ * @file rte_ring_rts_generic.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for Relaxed Tail Sync (RTS) ring mode.
+ * For more information please refer to <rte_ring_rts.h>.
+ */
+
+/**
+ * @internal This function updates tail values.
+ */
+static __rte_always_inline void
+__rte_ring_rts_update_tail(struct rte_ring_rts_headtail *ht)
+{
+ union rte_ring_ht_poscnt h, ot, nt;
+
+ /*
+ * If there are other enqueues/dequeues in progress that
+ * might preceded us, then don't update tail with new value.
+ */
+
+ do {
+ ot.raw = ht->tail.raw;
+ rte_smp_rmb();
+
+ /* on 32-bit systems we have to do atomic read here */
+ h.raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->head.raw);
+
+ nt.raw = ot.raw;
+ if (++nt.val.cnt == h.val.cnt)
+ nt.val.pos = h.val.pos;
+
+ } while (rte_atomic64_cmpset(&ht->tail.raw, ot.raw, nt.raw) == 0);
+}
+
+/**
+ * @internal This function waits till head/tail distance wouldn't
+ * exceed pre-defined max value.
+ */
+static __rte_always_inline void
+__rte_ring_rts_head_wait(const struct rte_ring_rts_headtail *ht,
+ union rte_ring_ht_poscnt *h)
+{
+ uint32_t max;
+
+ max = ht->htd_max;
+ h->raw = ht->head.raw;
+ rte_smp_rmb();
+
+ while (h->val.pos - ht->tail.val.pos > max) {
+ rte_pause();
+ h->raw = ht->head.raw;
+ rte_smp_rmb();
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue.
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_rts_move_prod_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* read prod head (may spin on prod tail) */
+ __rte_ring_rts_head_wait(&r->rts_prod, &oh);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - oh.val.pos;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ return 0;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (rte_atomic64_cmpset(&r->rts_prod.head.raw,
+ oh.raw, nh.raw) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_rts_move_cons_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* read cons head (may spin on cons tail) */
+ __rte_ring_rts_head_wait(&r->rts_cons, &oh);
+
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - oh.val.pos;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ return 0;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (rte_atomic64_cmpset(&r->rts_cons.head.raw,
+ oh.raw, nh.raw) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+#endif /* _RTE_RING_RTS_GENERIC_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v2 4/9] test/ring: add contention stress test for RTS ring
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
` (2 preceding siblings ...)
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 3/9] ring: introduce RTS ring mode Konstantin Ananyev
@ 2020-04-02 22:09 ` Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 5/9] ring: introduce HTS ring mode Konstantin Ananyev
` (5 subsequent siblings)
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-02 22:09 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce new test case to test RTS ring mode under contention.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_rts_stress.c | 32 ++++++++++++++++++++++++++++++++
app/test/test_ring_stress.c | 3 +++
app/test/test_ring_stress.h | 1 +
5 files changed, 38 insertions(+)
create mode 100644 app/test/test_ring_rts_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index 4eefaa887..3e15f3791 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -79,6 +79,7 @@ SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
SRCS-y += test_ring_mpmc_stress.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
diff --git a/app/test/meson.build b/app/test/meson.build
index 827b04886..bb67a49f0 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -102,6 +102,7 @@ test_sources = files('commands.c',
'test_ring.c',
'test_ring_mpmc_stress.c',
'test_ring_perf.c',
+ 'test_ring_rts_stress.c',
'test_ring_stress.c',
'test_rwlock.c',
'test_sched.c',
diff --git a/app/test/test_ring_rts_stress.c b/app/test/test_ring_rts_stress.c
new file mode 100644
index 000000000..f5255f24c
--- /dev/null
+++ b/app/test/test_ring_rts_stress.c
@@ -0,0 +1,32 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_mc_rts_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_mp_rts_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_RTS_ENQ | RING_F_MC_RTS_DEQ);
+}
+
+const struct test test_ring_rts_stress = {
+ .name = "MT_RTS",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
index 60706f799..eab395e30 100644
--- a/app/test/test_ring_stress.c
+++ b/app/test/test_ring_stress.c
@@ -40,6 +40,9 @@ test_ring_stress(void)
n += test_ring_mpmc_stress.nb_case;
k += run_test(&test_ring_mpmc_stress);
+ n += test_ring_rts_stress.nb_case;
+ k += run_test(&test_ring_rts_stress);
+
printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
n, k, n - k);
return (k != n);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
index 60eac6216..32aae2072 100644
--- a/app/test/test_ring_stress.h
+++ b/app/test/test_ring_stress.h
@@ -33,3 +33,4 @@ struct test {
};
extern const struct test test_ring_mpmc_stress;
+extern const struct test test_ring_rts_stress;
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v2 5/9] ring: introduce HTS ring mode
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
` (3 preceding siblings ...)
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 4/9] test/ring: add contention stress test for RTS ring Konstantin Ananyev
@ 2020-04-02 22:09 ` Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 6/9] test/ring: add contention stress test for HTS ring Konstantin Ananyev
` (4 subsequent siblings)
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-02 22:09 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce head/tail sync mode for MT ring synchronization.
In that mode enqueue/dequeue operation is fully serialized:
only one thread at a time is allowed to perform given op.
Suppose to reduce stall times in case when ring is used on
overcommitted cpus (multiple active threads on the same cpu).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 3 +
lib/librte_ring/meson.build | 3 +
lib/librte_ring/rte_ring.c | 20 ++-
lib/librte_ring/rte_ring.h | 31 ++++
lib/librte_ring/rte_ring_elem.h | 13 ++
lib/librte_ring/rte_ring_hts.h | 210 +++++++++++++++++++++++++
lib/librte_ring/rte_ring_hts_elem.h | 205 ++++++++++++++++++++++++
lib/librte_ring/rte_ring_hts_generic.h | 198 +++++++++++++++++++++++
8 files changed, 681 insertions(+), 2 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_hts.h
create mode 100644 lib/librte_ring/rte_ring_hts_elem.h
create mode 100644 lib/librte_ring/rte_ring_hts_generic.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 8f5c284cc..6fe500f0d 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -19,6 +19,9 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_elem.h \
rte_ring_generic.h \
rte_ring_c11_mem.h \
+ rte_ring_hts.h \
+ rte_ring_hts_elem.h \
+ rte_ring_hts_generic.h \
rte_ring_rts.h \
rte_ring_rts_elem.h \
rte_ring_rts_generic.h
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index 612936afb..8e86e037a 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -6,6 +6,9 @@ headers = files('rte_ring.h',
'rte_ring_elem.h',
'rte_ring_c11_mem.h',
'rte_ring_generic.h',
+ 'rte_ring_hts.h',
+ 'rte_ring_hts_elem.h',
+ 'rte_ring_hts_generic.h',
'rte_ring_rts.h',
'rte_ring_rts_elem.h',
'rte_ring_rts_generic.h')
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index 222eec0fb..ebe5ccf0d 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -89,9 +89,11 @@ static void
reset_headtail(void *p)
{
struct rte_ring_headtail *ht;
+ struct rte_ring_hts_headtail *ht_hts;
struct rte_ring_rts_headtail *ht_rts;
ht = p;
+ ht_hts = p;
ht_rts = p;
switch (ht->sync_type) {
@@ -104,6 +106,9 @@ reset_headtail(void *p)
ht_rts->head.raw = 0;
ht_rts->tail.raw = 0;
break;
+ case RTE_RING_SYNC_MT_HTS:
+ ht_hts->ht.raw = 0;
+ break;
default:
/* unknown sync mode */
RTE_ASSERT(0);
@@ -127,9 +132,9 @@ get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
enum rte_ring_sync_type *cons_st)
{
static const uint32_t prod_st_flags =
- (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
+ (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ | RING_F_MP_HTS_ENQ);
static const uint32_t cons_st_flags =
- (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
+ (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ | RING_F_MC_HTS_DEQ);
switch (flags & prod_st_flags) {
case 0:
@@ -141,6 +146,9 @@ get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
case RING_F_MP_RTS_ENQ:
*prod_st = RTE_RING_SYNC_MT_RTS;
break;
+ case RING_F_MP_HTS_ENQ:
+ *prod_st = RTE_RING_SYNC_MT_HTS;
+ break;
default:
return -EINVAL;
}
@@ -155,6 +163,9 @@ get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
case RING_F_MC_RTS_DEQ:
*cons_st = RTE_RING_SYNC_MT_RTS;
break;
+ case RING_F_MC_HTS_DEQ:
+ *cons_st = RTE_RING_SYNC_MT_HTS;
+ break;
default:
return -EINVAL;
}
@@ -176,6 +187,11 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
RTE_CACHE_LINE_MASK) != 0);
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
+ offsetof(struct rte_ring_hts_headtail, sync_type));
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
+ offsetof(struct rte_ring_hts_headtail, ht.pos.tail));
+
RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
offsetof(struct rte_ring_rts_headtail, sync_type));
RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index f6f084d79..6e4213afa 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -68,6 +68,7 @@ enum rte_ring_sync_type {
RTE_RING_SYNC_ST, /**< single thread only */
#ifdef ALLOW_EXPERIMENTAL_API
RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
+ RTE_RING_SYNC_MT_HTS, /**< multi-thread head/tail sync */
#endif
};
@@ -103,6 +104,19 @@ struct rte_ring_rts_headtail {
volatile union rte_ring_ht_poscnt head;
};
+union rte_ring_ht_pos {
+ uint64_t raw;
+ struct {
+ uint32_t head; /**< head position */
+ uint32_t tail; /**< tail position */
+ } pos;
+};
+
+struct rte_ring_hts_headtail {
+ volatile union rte_ring_ht_pos ht;
+ enum rte_ring_sync_type sync_type; /**< sync type of prod/cons */
+};
+
/**
* An RTE ring structure.
*
@@ -133,6 +147,7 @@ struct rte_ring {
RTE_STD_C11
union {
struct rte_ring_headtail prod;
+ struct rte_ring_hts_headtail hts_prod;
struct rte_ring_rts_headtail rts_prod;
} __rte_cache_aligned;
@@ -142,6 +157,7 @@ struct rte_ring {
RTE_STD_C11
union {
struct rte_ring_headtail cons;
+ struct rte_ring_hts_headtail hts_cons;
struct rte_ring_rts_headtail rts_cons;
} __rte_cache_aligned;
@@ -164,6 +180,9 @@ struct rte_ring {
#define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS". */
#define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC RTS". */
+#define RING_F_MP_HTS_ENQ 0x0020 /**< The default enqueue is "MP HTS". */
+#define RING_F_MC_HTS_DEQ 0x0040 /**< The default dequeue is "MC HTS". */
+
#define __IS_SP RTE_RING_SYNC_ST
#define __IS_MP RTE_RING_SYNC_MT
#define __IS_SC RTE_RING_SYNC_ST
@@ -494,6 +513,7 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
}
#ifdef ALLOW_EXPERIMENTAL_API
+#include <rte_ring_hts.h>
#include <rte_ring_rts.h>
#endif
@@ -529,6 +549,9 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_bulk(r, obj_table, n,
free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_bulk(r, obj_table, n,
+ free_space);
#endif
}
@@ -676,6 +699,8 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
#ifdef ALLOW_EXPERIMENTAL_API
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_bulk(r, obj_table, n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_bulk(r, obj_table, n, available);
#endif
}
@@ -1010,6 +1035,9 @@ rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_burst(r, obj_table, n,
free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_burst(r, obj_table, n,
+ free_space);
#endif
}
@@ -1103,6 +1131,9 @@ rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_burst(r, obj_table, n,
available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_burst(r, obj_table, n,
+ available);
#endif
}
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 5de0850dc..010a564c1 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -542,6 +542,7 @@ rte_ring_sp_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
RTE_RING_QUEUE_FIXED, __IS_SP, free_space);
}
+#include <rte_ring_hts_elem.h>
#include <rte_ring_rts_elem.h>
/**
@@ -585,6 +586,9 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_bulk_elem(r, obj_table, esize, n,
free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
#endif
}
@@ -766,6 +770,9 @@ rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_bulk_elem(r, obj_table, esize,
n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_bulk_elem(r, obj_table, esize,
+ n, available);
#endif
}
@@ -951,6 +958,9 @@ rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_burst_elem(r, obj_table, esize,
n, free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_burst_elem(r, obj_table, esize,
+ n, free_space);
#endif
}
@@ -1060,6 +1070,9 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_burst_elem(r, obj_table, esize,
n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_burst_elem(r, obj_table, esize,
+ n, available);
#endif
}
diff --git a/lib/librte_ring/rte_ring_hts.h b/lib/librte_ring/rte_ring_hts.h
new file mode 100644
index 000000000..062d7be6c
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts.h
@@ -0,0 +1,210 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_H_
+#define _RTE_RING_HTS_H_
+
+/**
+ * @file rte_ring_hts.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring.h> instead.
+ *
+ * Contains functions for serialized, aka Head-Tail Sync (HTS) ring mode.
+ * In that mode enqueue/dequeue operation is fully serialized:
+ * at any given moement only one enqueue/dequeue operation can proceed.
+ * This is achieved by thread is allowed to proceed with changing head.value
+ * only when head.value == tail.value.
+ * Both head and tail values are updated atomically (as one 64-bit value).
+ * To achieve that 64-bit CAS is used by head update routine.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_hts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_enqueue(struct rte_ring *r, void * const *obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_hts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_hts_update_tail(&r->hts_prod, n, 1);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_dequeue(struct rte_ring *r, void **obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_hts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_hts_update_tail(&r->hts_cons, n, 0);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_hts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_hts_dequeue_bulk(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ available);
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_hts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_hts_dequeue_burst(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_HTS_H_ */
diff --git a/lib/librte_ring/rte_ring_hts_elem.h b/lib/librte_ring/rte_ring_hts_elem.h
new file mode 100644
index 000000000..34f0d121d
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts_elem.h
@@ -0,0 +1,205 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_ELEM_H_
+#define _RTE_RING_HTS_ELEM_H_
+
+/**
+ * @file rte_ring_hts_elem.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring_elem.h> instead.
+ *
+ * Contains *ring_elem* functions for Head-Tail Sync (HTS) ring mode.
+ * for more details please refer to <rte_ring_hts.h>.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_hts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_enqueue_elem(struct rte_ring *r, void * const *obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_hts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ __rte_ring_enqueue_elems(r, head, obj_table, esize, n);
+ __rte_ring_hts_update_tail(&r->hts_prod, n, 1);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_dequeue_elem(struct rte_ring *r, void **obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_hts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ __rte_ring_dequeue_elems(r, head, obj_table, esize, n);
+ __rte_ring_hts_update_tail(&r->hts_cons, n, 0);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_hts_enqueue_bulk_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_hts_dequeue_bulk_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, available);
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_hts_enqueue_burst_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_hts_dequeue_burst_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_HTS_ELEM_H_ */
diff --git a/lib/librte_ring/rte_ring_hts_generic.h b/lib/librte_ring/rte_ring_hts_generic.h
new file mode 100644
index 000000000..0b3931ffa
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts_generic.h
@@ -0,0 +1,198 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2020 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_GENERIC_H_
+#define _RTE_RING_HTS_GENERIC_H_
+
+/**
+ * @file rte_ring_hts_generic.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for head/tail sync (HTS) ring mode.
+ * For more information please refer to <rte_ring_hts.h>.
+ */
+
+static __rte_always_inline void
+__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
+ uint32_t enqueue)
+{
+ union rte_ring_ht_pos p;
+
+ if (enqueue)
+ rte_smp_wmb();
+ else
+ rte_smp_rmb();
+
+ p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
+
+ p.pos.head = p.pos.tail + num;
+ p.pos.tail = p.pos.head;
+
+ rte_atomic64_set((rte_atomic64_t *)(uintptr_t)&ht->ht.raw, p.raw);
+}
+
+/**
+ * @internal waits till tail will become equal to head.
+ * Means no writer/reader is active for that ring.
+ * Suppose to work as serialization point.
+ */
+static __rte_always_inline void
+__rte_ring_hts_head_wait(const struct rte_ring_hts_headtail *ht,
+ union rte_ring_ht_pos *p)
+{
+ p->raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->ht.raw);
+
+ while (p->pos.head != p->pos.tail) {
+ rte_pause();
+ p->raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->ht.raw);
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_prod_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_prod, &op);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - op.pos.head;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ return 0;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (rte_atomic64_cmpset(&r->hts_prod.ht.raw,
+ op.raw, np.raw) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_cons_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_cons, &op);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - op.pos.head;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ return 0;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (rte_atomic64_cmpset(&r->hts_cons.ht.raw,
+ op.raw, np.raw) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+#endif /* _RTE_RING_HTS_GENERIC_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v2 6/9] test/ring: add contention stress test for HTS ring
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
` (4 preceding siblings ...)
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 5/9] ring: introduce HTS ring mode Konstantin Ananyev
@ 2020-04-02 22:09 ` Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 7/9] ring: introduce peek style API Konstantin Ananyev
` (3 subsequent siblings)
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-02 22:09 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce new test case to test HTS ring mode under contention.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_hts_stress.c | 32 ++++++++++++++++++++++++++++++++
app/test/test_ring_stress.c | 3 +++
app/test/test_ring_stress.h | 1 +
5 files changed, 38 insertions(+)
create mode 100644 app/test/test_ring_hts_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index 3e15f3791..72f64e30e 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -78,6 +78,7 @@ SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
SRCS-y += test_ring_mpmc_stress.c
+SRCS-y += test_ring_hts_stress.c
SRCS-y += test_ring_perf.c
SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
diff --git a/app/test/meson.build b/app/test/meson.build
index bb67a49f0..d6504a08a 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -101,6 +101,7 @@ test_sources = files('commands.c',
'test_rib6.c',
'test_ring.c',
'test_ring_mpmc_stress.c',
+ 'test_ring_hts_stress.c',
'test_ring_perf.c',
'test_ring_rts_stress.c',
'test_ring_stress.c',
diff --git a/app/test/test_ring_hts_stress.c b/app/test/test_ring_hts_stress.c
new file mode 100644
index 000000000..edc9175cb
--- /dev/null
+++ b/app/test/test_ring_hts_stress.c
@@ -0,0 +1,32 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_mc_hts_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_mp_hts_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_HTS_ENQ | RING_F_MC_HTS_DEQ);
+}
+
+const struct test test_ring_hts_stress = {
+ .name = "MT_HTS",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
index eab395e30..29a1368d7 100644
--- a/app/test/test_ring_stress.c
+++ b/app/test/test_ring_stress.c
@@ -43,6 +43,9 @@ test_ring_stress(void)
n += test_ring_rts_stress.nb_case;
k += run_test(&test_ring_rts_stress);
+ n += test_ring_hts_stress.nb_case;
+ k += run_test(&test_ring_hts_stress);
+
printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
n, k, n - k);
return (k != n);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
index 32aae2072..9a87c7f7b 100644
--- a/app/test/test_ring_stress.h
+++ b/app/test/test_ring_stress.h
@@ -34,3 +34,4 @@ struct test {
extern const struct test test_ring_mpmc_stress;
extern const struct test test_ring_rts_stress;
+extern const struct test test_ring_hts_stress;
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v2 7/9] ring: introduce peek style API
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
` (5 preceding siblings ...)
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 6/9] test/ring: add contention stress test for HTS ring Konstantin Ananyev
@ 2020-04-02 22:09 ` Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 8/9] test/ring: add stress test for MT peek API Konstantin Ananyev
` (2 subsequent siblings)
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-02 22:09 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
For rings with producer/consumer in RTE_RING_SYNC_ST, RTE_RING_SYNC_MT_HTS
mode, provide an ability to split enqueue/dequeue operation
into two phases:
- enqueue/dequeue start
- enqueue/dequeue finish
That allows user to inspect objects in the ring without removing
them from it (aka MT safe peek).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 1 +
lib/librte_ring/meson.build | 1 +
lib/librte_ring/rte_ring_c11_mem.h | 44 +++
lib/librte_ring/rte_ring_elem.h | 4 +
lib/librte_ring/rte_ring_generic.h | 48 ++++
lib/librte_ring/rte_ring_hts_generic.h | 47 ++-
lib/librte_ring/rte_ring_peek.h | 379 +++++++++++++++++++++++++
7 files changed, 519 insertions(+), 5 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_peek.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 6fe500f0d..5f8662737 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -22,6 +22,7 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_hts.h \
rte_ring_hts_elem.h \
rte_ring_hts_generic.h \
+ rte_ring_peek.h \
rte_ring_rts.h \
rte_ring_rts_elem.h \
rte_ring_rts_generic.h
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index 8e86e037a..f5f84dc6e 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -9,6 +9,7 @@ headers = files('rte_ring.h',
'rte_ring_hts.h',
'rte_ring_hts_elem.h',
'rte_ring_hts_generic.h',
+ 'rte_ring_peek.h',
'rte_ring_rts.h',
'rte_ring_rts_elem.h',
'rte_ring_rts_generic.h')
diff --git a/lib/librte_ring/rte_ring_c11_mem.h b/lib/librte_ring/rte_ring_c11_mem.h
index 0fb73a337..bb3096721 100644
--- a/lib/librte_ring/rte_ring_c11_mem.h
+++ b/lib/librte_ring/rte_ring_c11_mem.h
@@ -10,6 +10,50 @@
#ifndef _RTE_RING_C11_MEM_H_
#define _RTE_RING_C11_MEM_H_
+/**
+ * @internal get current tail value.
+ * This function should be used only for single thread producer/consumer.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_st_get_tail(struct rte_ring_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t h, n, t;
+
+ h = ht->head;
+ t = ht->tail;
+ n = h - t;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = h;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail.
+ * This function should be used only for single thread producer/consumer.
+ * Should be used only in conjunction with __rte_ring_st_get_tail.
+ */
+static __rte_always_inline void
+__rte_ring_st_set_head_tail(struct rte_ring_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
+{
+ uint32_t pos;
+
+ RTE_SET_USED(enqueue);
+
+ pos = tail + num;
+ ht->head = pos;
+ __atomic_store_n(&ht->tail, pos, __ATOMIC_RELEASE);
+}
+
static __rte_always_inline void
update_tail(struct rte_ring_headtail *ht, uint32_t old_val, uint32_t new_val,
uint32_t single, uint32_t enqueue)
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 010a564c1..5bf7c1c1b 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -1083,6 +1083,10 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
return 0;
}
+#ifdef ALLOW_EXPERIMENTAL_API
+#include <rte_ring_peek.h>
+#endif
+
#ifdef __cplusplus
}
#endif
diff --git a/lib/librte_ring/rte_ring_generic.h b/lib/librte_ring/rte_ring_generic.h
index 953cdbbd5..9f5fdf13b 100644
--- a/lib/librte_ring/rte_ring_generic.h
+++ b/lib/librte_ring/rte_ring_generic.h
@@ -10,6 +10,54 @@
#ifndef _RTE_RING_GENERIC_H_
#define _RTE_RING_GENERIC_H_
+/**
+ * @internal get current tail value.
+ * This function should be used only for single thread producer/consumer.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_st_get_tail(struct rte_ring_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t h, n, t;
+
+ h = ht->head;
+ t = ht->tail;
+ n = h - t;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = h;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail.
+ * This function should be used only for single thread producer/consumer.
+ * Should be used only in conjunction with __rte_ring_st_get_tail.
+ */
+static __rte_always_inline void
+__rte_ring_st_set_head_tail(struct rte_ring_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
+{
+ uint32_t pos;
+
+ pos = tail + num;
+
+ if (enqueue)
+ rte_smp_wmb();
+ else
+ rte_smp_rmb();
+
+ ht->head = pos;
+ ht->tail = pos;
+}
+
static __rte_always_inline void
update_tail(struct rte_ring_headtail *ht, uint32_t old_val, uint32_t new_val,
uint32_t single, uint32_t enqueue)
diff --git a/lib/librte_ring/rte_ring_hts_generic.h b/lib/librte_ring/rte_ring_hts_generic.h
index 0b3931ffa..7eac761f9 100644
--- a/lib/librte_ring/rte_ring_hts_generic.h
+++ b/lib/librte_ring/rte_ring_hts_generic.h
@@ -18,9 +18,38 @@
* For more information please refer to <rte_ring_hts.h>.
*/
+/**
+ * @internal get current tail value.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_hts_get_tail(struct rte_ring_hts_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t n;
+ union rte_ring_ht_pos p;
+
+ p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
+ n = p.pos.head - p.pos.tail;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = p.pos.tail;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail as one atomic 64 bit operation.
+ * Should be used only in conjunction with __rte_ring_hts_get_tail.
+ */
static __rte_always_inline void
-__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
- uint32_t enqueue)
+__rte_ring_hts_set_head_tail(struct rte_ring_hts_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
{
union rte_ring_ht_pos p;
@@ -29,14 +58,22 @@ __rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
else
rte_smp_rmb();
- p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
-
- p.pos.head = p.pos.tail + num;
+ p.pos.head = tail + num;
p.pos.tail = p.pos.head;
rte_atomic64_set((rte_atomic64_t *)(uintptr_t)&ht->ht.raw, p.raw);
}
+static __rte_always_inline void
+__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
+ uint32_t enqueue)
+{
+ uint32_t tail;
+
+ num = __rte_ring_hts_get_tail(ht, &tail, num);
+ __rte_ring_hts_set_head_tail(ht, tail, num, enqueue);
+}
+
/**
* @internal waits till tail will become equal to head.
* Means no writer/reader is active for that ring.
diff --git a/lib/librte_ring/rte_ring_peek.h b/lib/librte_ring/rte_ring_peek.h
new file mode 100644
index 000000000..baefd2f7b
--- /dev/null
+++ b/lib/librte_ring/rte_ring_peek.h
@@ -0,0 +1,379 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_PEEK_H_
+#define _RTE_RING_PEEK_H_
+
+/**
+ * @file
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring_elem.h> instead.
+ *
+ * Ring Peek AP
+ * Introduction of rte_ring with serialized producer/consumer (HTS sync mode)
+ * makes possible to split public enqueue/dequeue API into two phases:
+ * - enqueue/dequeue start
+ * - enqueue/dequeue finish
+ * That allows user to inspect objects in the ring without removing them
+ * from it (aka MT safe peek).
+ * Note that right now this new API is avaialble only for two sync modes:
+ * 1) Single Producer/Single Consumer (RTE_RING_SYNC_ST)
+ * 2) Serialized Producer/Serialized Consumer (RTE_RING_SYNC_MT_HTS).
+ * It is a user responsibility to create/init ring with appropriate sync
+ * modes selected.
+ * As an example:
+ * // read 1 elem from the ring:
+ * n = rte_ring_hts_dequeue_bulk_start(ring, &obj, 1, NULL);
+ * if (n != 0) {
+ * //examine object
+ * if (object_examine(obj) == KEEP)
+ * //decided to keep it in the ring.
+ * rte_ring_hts_dequeue_finish(ring, 0);
+ * else
+ * //decided to remove it from the ring.
+ * rte_ring_hts_dequeue_finish(ring, n);
+ * }
+ * Note that between _start_ and _finish_ the ring is sort of locked -
+ * none other thread can proceed with enqueue(/dequeue) operation till
+ * _finish_ will complete.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/**
+ * @internal This function moves prod head value.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_enqueue_start(struct rte_ring *r, uint32_t n,
+ enum rte_ring_queue_behavior behavior, uint32_t *free_space)
+{
+ uint32_t free, head, next;
+
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_move_prod_head(r, RTE_RING_SYNC_ST, n,
+ behavior, &head, &next, &free);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ n = __rte_ring_hts_move_prod_head(r, n, behavior,
+ &head, &free);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ n = 0;
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * Start to enqueue several objects on the ring.
+ * Note that no actual objects are put in the queue by this function,
+ * it just reserves for user such ability.
+ * User has to call appropriate enqueue_finish() to copy objects into the
+ * queue and complete given enqueue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects that can be enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_enqueue_bulk_start(struct rte_ring *r, unsigned int n,
+ unsigned int *free_space)
+{
+ return __rte_ring_do_enqueue_start(r, n, RTE_RING_QUEUE_FIXED,
+ free_space);
+}
+
+/**
+ * Start to enqueue several objects on the ring.
+ * Note that no actual objects are put in the queue by this function,
+ * it just reserves for user such ability.
+ * User has to call appropriate enqueue_finish() to copy objects into the
+ * queue and complete given enqueue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * Actual number of objects that can be enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_enqueue_burst_start(struct rte_ring *r, unsigned int n,
+ unsigned int *free_space)
+{
+ return __rte_ring_do_enqueue_start(r, n, RTE_RING_QUEUE_VARIABLE,
+ free_space);
+}
+
+/**
+ * Complete to enqueue several objects on the ring.
+ * Note that number of objects to enqueue should not exceed previous
+ * enqueue_start return value.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of objects.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to add to the ring from the obj_table.
+ */
+__rte_experimental
+static __rte_always_inline void
+rte_ring_enqueue_elem_finish(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n)
+{
+ uint32_t tail;
+
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_st_get_tail(&r->prod, &tail, n);
+ if (n != 0)
+ __rte_ring_enqueue_elems(r, tail, obj_table, esize, n);
+ __rte_ring_st_set_head_tail(&r->prod, tail, n, 1);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ n = __rte_ring_hts_get_tail(&r->hts_prod, &tail, n);
+ if (n != 0)
+ __rte_ring_enqueue_elems(r, tail, obj_table, esize, n);
+ __rte_ring_hts_set_head_tail(&r->hts_prod, tail, n, 1);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ }
+}
+
+/**
+ * Complete to enqueue several objects on the ring.
+ * Note that number of objects to enqueue should not exceed previous
+ * enqueue_start return value.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of objects.
+ * @param n
+ * The number of objects to add to the ring from the obj_table.
+ */
+__rte_experimental
+static __rte_always_inline void
+rte_ring_enqueue_finish(struct rte_ring *r, void * const *obj_table,
+ unsigned int n)
+{
+ rte_ring_enqueue_elem_finish(r, obj_table, sizeof(uintptr_t), n);
+}
+
+/**
+ * @internal This function moves cons head value and copies up to *n*
+ * objects from the ring to the user provided obj_table.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_dequeue_start(struct rte_ring *r, void **obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t avail, head, next;
+
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_move_cons_head(r, RTE_RING_SYNC_ST, n,
+ behavior, &head, &next, &avail);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ n = __rte_ring_hts_move_cons_head(r, n, behavior,
+ &head, &avail);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ n = 0;
+ }
+
+ if (n != 0)
+ __rte_ring_dequeue_elems(r, head, obj_table, esize, n);
+
+ if (available != NULL)
+ *available = avail - n;
+ return n;
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_bulk_elem_start(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_dequeue_start(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, available);
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * Actual number of objects dequeued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_bulk_start(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return rte_ring_dequeue_bulk_elem_start(r, obj_table, sizeof(uintptr_t),
+ n, available);
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The actual number of objects dequeued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_burst_elem_start(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_dequeue_start(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The actual number of objects dequeued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_burst_start(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return rte_ring_dequeue_burst_elem_start(r, obj_table,
+ sizeof(uintptr_t), n, available);
+}
+
+/**
+ * Complete to dequeue several objects from the ring.
+ * Note that number of objects to dequeue should not exceed previous
+ * dequeue_start return value.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param n
+ * The number of objects to add to the ring from the obj_table.
+ */
+__rte_experimental
+static __rte_always_inline void
+rte_ring_dequeue_finish(struct rte_ring *r, unsigned int n)
+{
+ uint32_t tail;
+
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_st_get_tail(&r->cons, &tail, n);
+ __rte_ring_st_set_head_tail(&r->cons, tail, n, 0);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ __rte_ring_hts_update_tail(&r->hts_cons, n, 0);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ }
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_PEEK_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v2 8/9] test/ring: add stress test for MT peek API
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
` (6 preceding siblings ...)
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 7/9] ring: introduce peek style API Konstantin Ananyev
@ 2020-04-02 22:09 ` Konstantin Ananyev
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 9/9] ring: add C11 memory model for new sync modes Konstantin Ananyev
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-02 22:09 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce new test case to test MT peek API.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_peek_stress.c | 43 ++++++++++++++++++++++++++++++++
app/test/test_ring_stress.c | 3 +++
app/test/test_ring_stress.h | 1 +
5 files changed, 49 insertions(+)
create mode 100644 app/test/test_ring_peek_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index 72f64e30e..f4c987a98 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -80,6 +80,7 @@ SRCS-y += test_ring.c
SRCS-y += test_ring_mpmc_stress.c
SRCS-y += test_ring_hts_stress.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_peek_stress.c
SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
diff --git a/app/test/meson.build b/app/test/meson.build
index d6504a08a..cfcf5b81d 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -102,6 +102,7 @@ test_sources = files('commands.c',
'test_ring.c',
'test_ring_mpmc_stress.c',
'test_ring_hts_stress.c',
+ 'test_ring_peek_stress.c',
'test_ring_perf.c',
'test_ring_rts_stress.c',
'test_ring_stress.c',
diff --git a/app/test/test_ring_peek_stress.c b/app/test/test_ring_peek_stress.c
new file mode 100644
index 000000000..cfc82d728
--- /dev/null
+++ b/app/test/test_ring_peek_stress.c
@@ -0,0 +1,43 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+#include <rte_ring_elem.h>
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ uint32_t m;
+
+ m = rte_ring_dequeue_bulk_start(r, obj, n, avail);
+ n = (m == n) ? n : 0;
+ rte_ring_dequeue_finish(r, n);
+ return n;
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ uint32_t m;
+
+ m = rte_ring_enqueue_bulk_start(r, n, free);
+ n = (m == n) ? n : 0;
+ rte_ring_enqueue_finish(r, obj, n);
+ return n;
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_HTS_ENQ | RING_F_MC_HTS_DEQ);
+}
+
+const struct test test_ring_peek_stress = {
+ .name = "MT_PEEK",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
index 29a1368d7..853fcc190 100644
--- a/app/test/test_ring_stress.c
+++ b/app/test/test_ring_stress.c
@@ -46,6 +46,9 @@ test_ring_stress(void)
n += test_ring_hts_stress.nb_case;
k += run_test(&test_ring_hts_stress);
+ n += test_ring_peek_stress.nb_case;
+ k += run_test(&test_ring_peek_stress);
+
printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
n, k, n - k);
return (k != n);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
index 9a87c7f7b..60953ce47 100644
--- a/app/test/test_ring_stress.h
+++ b/app/test/test_ring_stress.h
@@ -35,3 +35,4 @@ struct test {
extern const struct test test_ring_mpmc_stress;
extern const struct test test_ring_rts_stress;
extern const struct test test_ring_hts_stress;
+extern const struct test test_ring_peek_stress;
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v2 9/9] ring: add C11 memory model for new sync modes
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
` (7 preceding siblings ...)
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 8/9] test/ring: add stress test for MT peek API Konstantin Ananyev
@ 2020-04-02 22:09 ` Konstantin Ananyev
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-02 22:09 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Add C11 atomics based implementation for RTS and HTS
head/tail update primitivies.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 4 +-
lib/librte_ring/meson.build | 2 +
lib/librte_ring/rte_ring_hts.h | 4 +
lib/librte_ring/rte_ring_hts_c11_mem.h | 222 +++++++++++++++++++++++++
lib/librte_ring/rte_ring_hts_elem.h | 4 +
lib/librte_ring/rte_ring_rts.h | 4 +
lib/librte_ring/rte_ring_rts_c11_mem.h | 198 ++++++++++++++++++++++
lib/librte_ring/rte_ring_rts_elem.h | 4 +
8 files changed, 441 insertions(+), 1 deletion(-)
create mode 100644 lib/librte_ring/rte_ring_hts_c11_mem.h
create mode 100644 lib/librte_ring/rte_ring_rts_c11_mem.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 5f8662737..927d105bf 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -22,9 +22,11 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_hts.h \
rte_ring_hts_elem.h \
rte_ring_hts_generic.h \
+ rte_ring_hts_c11_mem.h \
rte_ring_peek.h \
rte_ring_rts.h \
rte_ring_rts_elem.h \
- rte_ring_rts_generic.h
+ rte_ring_rts_generic.h \
+ rte_ring_rts_c11_mem.h
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index f5f84dc6e..f2e37a8e4 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -7,10 +7,12 @@ headers = files('rte_ring.h',
'rte_ring_c11_mem.h',
'rte_ring_generic.h',
'rte_ring_hts.h',
+ 'rte_ring_hts_c11_mem.h',
'rte_ring_hts_elem.h',
'rte_ring_hts_generic.h',
'rte_ring_peek.h',
'rte_ring_rts.h',
+ 'rte_ring_rts_c11_mem.h',
'rte_ring_rts_elem.h',
'rte_ring_rts_generic.h')
diff --git a/lib/librte_ring/rte_ring_hts.h b/lib/librte_ring/rte_ring_hts.h
index 062d7be6c..ddaa47ff1 100644
--- a/lib/librte_ring/rte_ring_hts.h
+++ b/lib/librte_ring/rte_ring_hts.h
@@ -29,7 +29,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_hts_c11_mem.h>
+#else
#include <rte_ring_hts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the HTS ring.
diff --git a/lib/librte_ring/rte_ring_hts_c11_mem.h b/lib/librte_ring/rte_ring_hts_c11_mem.h
new file mode 100644
index 000000000..ce0f15f8f
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts_c11_mem.h
@@ -0,0 +1,222 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2020 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_C11_MEM_H_
+#define _RTE_RING_HTS_C11_MEM_H_
+
+/**
+ * @file rte_ring_hts_c11_mem.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for head/tail sync (HTS) ring mode.
+ * For more information please refer to <rte_ring_hts.h>.
+ */
+
+/**
+ * @internal get current tail value.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_hts_get_tail(struct rte_ring_hts_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t n;
+ union rte_ring_ht_pos p;
+
+ p.raw = __atomic_load_n(&ht->ht.raw, __ATOMIC_RELAXED);
+ n = p.pos.head - p.pos.tail;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = p.pos.tail;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail as one atomic 64 bit operation.
+ * Should be used only in conjunction with __rte_ring_hts_get_tail.
+ */
+static __rte_always_inline void
+__rte_ring_hts_set_head_tail(struct rte_ring_hts_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
+{
+ union rte_ring_ht_pos p;
+
+ RTE_SET_USED(enqueue);
+
+ p.pos.head = tail + num;
+ p.pos.tail = p.pos.head;
+
+ __atomic_store_n(&ht->ht.raw, p.raw, __ATOMIC_RELEASE);
+}
+
+static __rte_always_inline void
+__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
+ uint32_t enqueue)
+{
+ uint32_t tail;
+
+ num = __rte_ring_hts_get_tail(ht, &tail, num);
+ __rte_ring_hts_set_head_tail(ht, tail, num, enqueue);
+}
+
+/**
+ * @internal waits till tail will become equal to head.
+ * Means no writer/reader is active for that ring.
+ * Suppose to work as serialization point.
+ */
+static __rte_always_inline void
+__rte_ring_hts_head_wait(const struct rte_ring_hts_headtail *ht,
+ union rte_ring_ht_pos *p)
+{
+ p->raw = __atomic_load_n(&ht->ht.raw, __ATOMIC_ACQUIRE);
+
+ while (p->pos.head != p->pos.tail) {
+ rte_pause();
+ p->raw = __atomic_load_n(&ht->ht.raw, __ATOMIC_ACQUIRE);
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_prod_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* wait for tail to be equal to head, , acquire point */
+ __rte_ring_hts_head_wait(&r->hts_prod, &op);
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - op.pos.head;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ return 0;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (__atomic_compare_exchange_n(&r->hts_prod.ht.raw,
+ &op.raw, np.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_cons_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_cons, &op);
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - op.pos.head;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ return 0;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (__atomic_compare_exchange_n(&r->hts_cons.ht.raw,
+ &op.raw, np.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+#endif /* _RTE_RING_HTS_C11_MEM_H_ */
diff --git a/lib/librte_ring/rte_ring_hts_elem.h b/lib/librte_ring/rte_ring_hts_elem.h
index 34f0d121d..1e9a49c7a 100644
--- a/lib/librte_ring/rte_ring_hts_elem.h
+++ b/lib/librte_ring/rte_ring_hts_elem.h
@@ -24,7 +24,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_hts_c11_mem.h>
+#else
#include <rte_ring_hts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the HTS ring.
diff --git a/lib/librte_ring/rte_ring_rts.h b/lib/librte_ring/rte_ring_rts.h
index 18404fe48..28b2d25f5 100644
--- a/lib/librte_ring/rte_ring_rts.h
+++ b/lib/librte_ring/rte_ring_rts.h
@@ -55,7 +55,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_rts_c11_mem.h>
+#else
#include <rte_ring_rts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the RTS ring.
diff --git a/lib/librte_ring/rte_ring_rts_c11_mem.h b/lib/librte_ring/rte_ring_rts_c11_mem.h
new file mode 100644
index 000000000..19d3ea288
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts_c11_mem.h
@@ -0,0 +1,198 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_C11_MEM_H_
+#define _RTE_RING_RTS_C11_MEM_H_
+
+/**
+ * @file rte_ring_rts_c11_mem.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for Relaxed Tail Sync (RTS) ring mode.
+ * For more information please refer to <rte_ring_rts.h>.
+ */
+
+/**
+ * @internal This function updates tail values.
+ */
+static __rte_always_inline void
+__rte_ring_rts_update_tail(struct rte_ring_rts_headtail *ht)
+{
+ union rte_ring_ht_poscnt h, ot, nt;
+
+ /*
+ * If there are other enqueues/dequeues in progress that
+ * might preceded us, then don't update tail with new value.
+ */
+
+ ot.raw = __atomic_load_n(&ht->tail.raw, __ATOMIC_ACQUIRE);
+
+ do {
+ /* on 32-bit systems we have to do atomic read here */
+ h.raw = __atomic_load_n(&ht->head.raw, __ATOMIC_RELAXED);
+
+ nt.raw = ot.raw;
+ if (++nt.val.cnt == h.val.cnt)
+ nt.val.pos = h.val.pos;
+
+ } while (__atomic_compare_exchange_n(&ht->tail.raw, &ot.raw, nt.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_ACQUIRE) == 0);
+}
+
+/**
+ * @internal This function waits till head/tail distance wouldn't
+ * exceed pre-defined max value.
+ */
+static __rte_always_inline void
+__rte_ring_rts_head_wait(const struct rte_ring_rts_headtail *ht,
+ union rte_ring_ht_poscnt *h)
+{
+ uint32_t max;
+
+ max = ht->htd_max;
+ h->raw = __atomic_load_n(&ht->head.raw, __ATOMIC_ACQUIRE);
+
+ while (h->val.pos - ht->tail.val.pos > max) {
+ rte_pause();
+ h->raw = __atomic_load_n(&ht->head.raw, __ATOMIC_ACQUIRE);
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue.
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_rts_move_prod_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* read prod head (may spin on prod tail, acquire point) */
+ __rte_ring_rts_head_wait(&r->rts_prod, &oh);
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - oh.val.pos;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ return 0;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (__atomic_compare_exchange_n(&r->rts_prod.head.raw,
+ &oh.raw, nh.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_rts_move_cons_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* read cons head (may spin on cons tail, acquire point) */
+ __rte_ring_rts_head_wait(&r->rts_cons, &oh);
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - oh.val.pos;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ return 0;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (__atomic_compare_exchange_n(&r->rts_cons.head.raw,
+ &oh.raw, nh.raw,
+ 1, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+#endif /* _RTE_RING_RTS_C11_MEM_H_ */
diff --git a/lib/librte_ring/rte_ring_rts_elem.h b/lib/librte_ring/rte_ring_rts_elem.h
index 71a331b23..23d8aeec7 100644
--- a/lib/librte_ring/rte_ring_rts_elem.h
+++ b/lib/librte_ring/rte_ring_rts_elem.h
@@ -24,7 +24,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_rts_c11_mem.h>
+#else
#include <rte_ring_rts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the RTS ring.
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v3 0/9] New sync modes for ring
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 0/9] New sync modes for ring Konstantin Ananyev
` (8 preceding siblings ...)
2020-04-02 22:09 ` [dpdk-dev] [PATCH v2 9/9] ring: add C11 memory model for new sync modes Konstantin Ananyev
@ 2020-04-03 17:42 ` Konstantin Ananyev
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 1/9] test/ring: add contention stress test Konstantin Ananyev
` (9 more replies)
9 siblings, 10 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-03 17:42 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
V2 - V3
1. Few more compilation fixes (for gcc 4.8.X)
2. Extra update devtools/libabigail.abignore (workaround)
V1 - v2 changes:
1. Fix compilation issues
2. Add C11 atomics support
3. Updates devtools/libabigail.abignore (workaround)
RFC - V1 changes:
1. remove ABI brekage (at least I hope I did)
2. Add support for ring_elem
3. Rework peek related API a bit
4. Rework test to make it less verbose and unite all test-cases
in one command
5. Add new test-case for MT peek API
TODO list:
1. Update docs
These days more and more customers use(/try to use) DPDK based apps within
overcommitted systems (multiple acttive threads over same pysical cores):
VM, container deployments, etc.
One quite common problem they hit:
Lock-Holder-Preemption/Lock-Waiter-Preemption with rte_ring.
LHP is quite a common problem for spin-based sync primitives
(spin-locks, etc.) on overcommitted systems.
The situation gets much worse when some sort of
fair-locking technique is used (ticket-lock, etc.).
As now not only lock-owner but also lock-waiters scheduling
order matters a lot (LWP).
These two problems are well-known for kernel within VMs:
http://www-archive.xenproject.org/files/xensummitboston08/LHP.pdf
https://www.cs.hs-rm.de/~kaiser/events/wamos2017/Slides/selcuk.pdf
The problem with rte_ring is that while head accusion is sort of
un-fair locking, waiting on tail is very similar to ticket lock schema -
tail has to be updated in particular order.
That makes current rte_ring implementation to perform
really pure on some overcommited scenarios.
It is probably not possible to completely resolve LHP problem in
userspace only (without some kernel communication/intervention).
But removing fairness at tail update helps to avoid LWP and
can mitigate the situation significantly.
This patch proposes two new optional ring synchronization modes:
1) Head/Tail Sync (HTS) mode
In that mode enqueue/dequeue operation is fully serialized:
only one thread at a time is allowed to perform given op.
As another enhancement provide ability to split enqueue/dequeue
operation into two phases:
- enqueue/dequeue start
- enqueue/dequeue finish
That allows user to inspect objects in the ring without removing
them from it (aka MT safe peek).
2) Relaxed Tail Sync (RTS)
The main difference from original MP/MC algorithm is that
tail value is increased not by every thread that finished enqueue/dequeue,
but only by the last one.
That allows threads to avoid spinning on ring tail value,
leaving actual tail value change to the last thread in the update queue.
Note that these new sync modes are optional.
For current rte_ring users nothing should change
(both in terms of API/ABI and performance).
Existing sync modes MP/MC,SP/SC kept untouched, set up in the same
way (via flags and _init_), and MP/MC remains as default one.
The only thing that changed:
Format of prod/cons now could differ depending on mode selected at _init_.
So user has to stick with one sync model through whole ring lifetime.
In other words, user can't create a ring for let say SP mode and then
in the middle of data-path change his mind and start using MP_RTS mode.
For existing modes (SP/MP, SC/MC) format remains the same and
user can still use them interchangeably, though of course it is an
error prone practice.
Test results on IA (see below) show significant improvements
for average enqueue/dequeue op times on overcommitted systems.
For 'classic' DPDK deployments (one thread per core) original MP/MC
algorithm still shows best numbers, though for 64-bit target
RTS numbers are not that far away.
Numbers were produced by new UT test-case: ring_stress_autotest, i.e.:
echo ring_stress_autotest | ./dpdk-test -n 4 --lcores='...'
X86_64 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
DEQ+ENQ average cycles/obj
MP/MC HTS RTS
1thread@1core(--lcores=6-7) 8.00 8.15 8.99
2thread@2core(--lcores=6-8) 19.14 19.61 20.35
4thread@4core(--lcores=6-10) 29.43 29.79 31.82
8thread@8core(--lcores=6-14) 110.59 192.81 119.50
16thread@16core(--lcores=6-22) 461.03 813.12 495.59
32thread/@32core(--lcores='6-22,55-70') 982.90 1972.38 1160.51
2thread@1core(--lcores='6,(10-11)@7' 20140.50 23.58 25.14
4thread@2core(--lcores='6,(10-11)@7,(20-21)@8' 153680.60 76.88 80.05
8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 280314.32 294.72 318.79
16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 643176.59 1144.02 1175.14
32thread@2core(--lcores='6,(10-25)@7,(30-45)@8' 4264238.80 4627.48 4892.68
8thread@2core(--lcores='6,(10-17)@(7,8))' 321085.98 298.59 307.47
16thread@4core(--lcores='6,(20-35)@(7-10))' 1900705.61 575.35 678.29
32thread@4core(--lcores='6,(20-51)@(7-10))' 5510445.85 2164.36 2714.12
i686 @ Intel(R) Xeon(R) Platinum 8160 CPU @ 2.10GHz
DEQ+ENQ average cycles/obj
MP/MC HTS RTS
1thread@1core(--lcores=6-7) 7.85 12.13 11.31
2thread@2core(--lcores=6-8) 17.89 24.52 21.86
8thread@8core(--lcores=6-14) 32.58 354.20 54.58
32thread/@32core(--lcores='6-22,55-70') 813.77 6072.41 2169.91
2thread@1core(--lcores='6,(10-11)@7' 16095.00 36.06 34.74
8thread@2core(--lcores='6,(10-13)@7,(20-23)@8' 1140354.54 346.61 361.57
16thread@2core(--lcores='6,(10-17)@7,(20-27)@8' 1920417.86 1314.90 1416.65
8thread@2core(--lcores='6,(10-17)@(7,8))' 594358.61 332.70 357.74
32thread@4core(--lcores='6,(20-51)@(7-10))' 5319896.86 2836.44 3028.87
Konstantin Ananyev (9):
test/ring: add contention stress test
ring: prepare ring to allow new sync schemes
ring: introduce RTS ring mode
test/ring: add contention stress test for RTS ring
ring: introduce HTS ring mode
test/ring: add contention stress test for HTS ring
ring: introduce peek style API
test/ring: add stress test for MT peek API
ring: add C11 memory model for new sync modes
app/test/Makefile | 5 +
app/test/meson.build | 5 +
app/test/test_pdump.c | 6 +-
app/test/test_ring_hts_stress.c | 32 ++
app/test/test_ring_mpmc_stress.c | 31 ++
app/test/test_ring_peek_stress.c | 43 +++
app/test/test_ring_rts_stress.c | 32 ++
app/test/test_ring_stress.c | 57 ++++
app/test/test_ring_stress.h | 38 +++
app/test/test_ring_stress_impl.h | 444 +++++++++++++++++++++++++
devtools/libabigail.abignore | 7 +
lib/librte_pdump/rte_pdump.c | 2 +-
lib/librte_port/rte_port_ring.c | 12 +-
lib/librte_ring/Makefile | 11 +-
lib/librte_ring/meson.build | 11 +-
lib/librte_ring/rte_ring.c | 114 ++++++-
lib/librte_ring/rte_ring.h | 244 ++++++++++++--
lib/librte_ring/rte_ring_c11_mem.h | 44 +++
lib/librte_ring/rte_ring_elem.h | 105 +++++-
lib/librte_ring/rte_ring_generic.h | 48 +++
lib/librte_ring/rte_ring_hts.h | 214 ++++++++++++
lib/librte_ring/rte_ring_hts_c11_mem.h | 222 +++++++++++++
lib/librte_ring/rte_ring_hts_elem.h | 209 ++++++++++++
lib/librte_ring/rte_ring_hts_generic.h | 235 +++++++++++++
lib/librte_ring/rte_ring_peek.h | 379 +++++++++++++++++++++
lib/librte_ring/rte_ring_rts.h | 320 ++++++++++++++++++
lib/librte_ring/rte_ring_rts_c11_mem.h | 198 +++++++++++
lib/librte_ring/rte_ring_rts_elem.h | 209 ++++++++++++
lib/librte_ring/rte_ring_rts_generic.h | 210 ++++++++++++
29 files changed, 3431 insertions(+), 56 deletions(-)
create mode 100644 app/test/test_ring_hts_stress.c
create mode 100644 app/test/test_ring_mpmc_stress.c
create mode 100644 app/test/test_ring_peek_stress.c
create mode 100644 app/test/test_ring_rts_stress.c
create mode 100644 app/test/test_ring_stress.c
create mode 100644 app/test/test_ring_stress.h
create mode 100644 app/test/test_ring_stress_impl.h
create mode 100644 lib/librte_ring/rte_ring_hts.h
create mode 100644 lib/librte_ring/rte_ring_hts_c11_mem.h
create mode 100644 lib/librte_ring/rte_ring_hts_elem.h
create mode 100644 lib/librte_ring/rte_ring_hts_generic.h
create mode 100644 lib/librte_ring/rte_ring_peek.h
create mode 100644 lib/librte_ring/rte_ring_rts.h
create mode 100644 lib/librte_ring/rte_ring_rts_c11_mem.h
create mode 100644 lib/librte_ring/rte_ring_rts_elem.h
create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v3 1/9] test/ring: add contention stress test
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
@ 2020-04-03 17:42 ` Konstantin Ananyev
2020-04-08 4:59 ` Honnappa Nagarahalli
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 2/9] ring: prepare ring to allow new sync schemes Konstantin Ananyev
` (8 subsequent siblings)
9 siblings, 1 reply; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-03 17:42 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce new test-case to measure ring perfomance under contention
(miltiple producers/consumers).
Starts dequeue/enqueue loop on all available slave lcores.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 2 +
app/test/meson.build | 2 +
app/test/test_ring_mpmc_stress.c | 31 +++
app/test/test_ring_stress.c | 48 ++++
app/test/test_ring_stress.h | 35 +++
app/test/test_ring_stress_impl.h | 444 +++++++++++++++++++++++++++++++
6 files changed, 562 insertions(+)
create mode 100644 app/test/test_ring_mpmc_stress.c
create mode 100644 app/test/test_ring_stress.c
create mode 100644 app/test/test_ring_stress.h
create mode 100644 app/test/test_ring_stress_impl.h
diff --git a/app/test/Makefile b/app/test/Makefile
index 1f080d162..4eefaa887 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -77,7 +77,9 @@ SRCS-y += test_external_mem.c
SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
+SRCS-y += test_ring_mpmc_stress.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
ifeq ($(CONFIG_RTE_LIBRTE_TABLE),y)
diff --git a/app/test/meson.build b/app/test/meson.build
index 351d29cb6..827b04886 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -100,7 +100,9 @@ test_sources = files('commands.c',
'test_rib.c',
'test_rib6.c',
'test_ring.c',
+ 'test_ring_mpmc_stress.c',
'test_ring_perf.c',
+ 'test_ring_stress.c',
'test_rwlock.c',
'test_sched.c',
'test_service_cores.c',
diff --git a/app/test/test_ring_mpmc_stress.c b/app/test/test_ring_mpmc_stress.c
new file mode 100644
index 000000000..1524b0248
--- /dev/null
+++ b/app/test/test_ring_mpmc_stress.c
@@ -0,0 +1,31 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_mc_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_mp_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num, 0);
+}
+
+const struct test test_ring_mpmc_stress = {
+ .name = "MP/MC",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
new file mode 100644
index 000000000..60706f799
--- /dev/null
+++ b/app/test/test_ring_stress.c
@@ -0,0 +1,48 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress.h"
+
+static int
+run_test(const struct test *test)
+{
+ int32_t rc;
+ uint32_t i, k;
+
+ for (i = 0, k = 0; i != test->nb_case; i++) {
+
+ printf("TEST-CASE %s %s START\n",
+ test->name, test->cases[i].name);
+
+ rc = test->cases[i].func(test->cases[i].wfunc);
+ k += (rc == 0);
+
+ if (rc != 0)
+ printf("TEST-CASE %s %s FAILED\n",
+ test->name, test->cases[i].name);
+ else
+ printf("TEST-CASE %s %s OK\n",
+ test->name, test->cases[i].name);
+ }
+
+ return k;
+}
+
+static int
+test_ring_stress(void)
+{
+ uint32_t n, k;
+
+ n = 0;
+ k = 0;
+
+ n += test_ring_mpmc_stress.nb_case;
+ k += run_test(&test_ring_mpmc_stress);
+
+ printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
+ n, k, n - k);
+ return (k != n);
+}
+
+REGISTER_TEST_COMMAND(ring_stress_autotest, test_ring_stress);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
new file mode 100644
index 000000000..60eac6216
--- /dev/null
+++ b/app/test/test_ring_stress.h
@@ -0,0 +1,35 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+
+#include <inttypes.h>
+#include <stddef.h>
+#include <stdalign.h>
+#include <string.h>
+#include <stdio.h>
+#include <unistd.h>
+
+#include <rte_ring.h>
+#include <rte_cycles.h>
+#include <rte_launch.h>
+#include <rte_pause.h>
+#include <rte_random.h>
+#include <rte_malloc.h>
+#include <rte_spinlock.h>
+
+#include "test.h"
+
+struct test_case {
+ const char *name;
+ int (*func)(int (*)(void *));
+ int (*wfunc)(void *arg);
+};
+
+struct test {
+ const char *name;
+ uint32_t nb_case;
+ const struct test_case *cases;
+};
+
+extern const struct test test_ring_mpmc_stress;
diff --git a/app/test/test_ring_stress_impl.h b/app/test/test_ring_stress_impl.h
new file mode 100644
index 000000000..11476d28c
--- /dev/null
+++ b/app/test/test_ring_stress_impl.h
@@ -0,0 +1,444 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress.h"
+
+/*
+ * Measures performance of ring enqueue/dequeue under high contention
+ */
+
+#define RING_NAME "RING_STRESS"
+#define BULK_NUM 32
+#define RING_SIZE (2 * BULK_NUM * RTE_MAX_LCORE)
+
+enum {
+ WRK_CMD_STOP,
+ WRK_CMD_RUN,
+};
+
+static volatile uint32_t wrk_cmd __rte_cache_aligned;
+
+/* test run-time in seconds */
+static const uint32_t run_time = 60;
+static const uint32_t verbose;
+
+struct lcore_stat {
+ uint64_t nb_cycle;
+ struct {
+ uint64_t nb_call;
+ uint64_t nb_obj;
+ uint64_t nb_cycle;
+ uint64_t max_cycle;
+ uint64_t min_cycle;
+ } op;
+};
+
+struct lcore_arg {
+ struct rte_ring *rng;
+ struct lcore_stat stats;
+} __rte_cache_aligned;
+
+struct ring_elem {
+ uint32_t cnt[RTE_CACHE_LINE_SIZE / sizeof(uint32_t)];
+} __rte_cache_aligned;
+
+/*
+ * redefinable functions
+ */
+static uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail);
+
+static uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free);
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num);
+
+
+static void
+lcore_stat_update(struct lcore_stat *ls, uint64_t call, uint64_t obj,
+ uint64_t tm, int32_t prcs)
+{
+ ls->op.nb_call += call;
+ ls->op.nb_obj += obj;
+ ls->op.nb_cycle += tm;
+ if (prcs) {
+ ls->op.max_cycle = RTE_MAX(ls->op.max_cycle, tm);
+ ls->op.min_cycle = RTE_MIN(ls->op.min_cycle, tm);
+ }
+}
+
+static void
+lcore_op_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls)
+{
+
+ ms->op.nb_call += ls->op.nb_call;
+ ms->op.nb_obj += ls->op.nb_obj;
+ ms->op.nb_cycle += ls->op.nb_cycle;
+ ms->op.max_cycle = RTE_MAX(ms->op.max_cycle, ls->op.max_cycle);
+ ms->op.min_cycle = RTE_MIN(ms->op.min_cycle, ls->op.min_cycle);
+}
+
+static void
+lcore_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls)
+{
+ ms->nb_cycle = RTE_MAX(ms->nb_cycle, ls->nb_cycle);
+ lcore_op_stat_aggr(ms, ls);
+}
+
+static void
+lcore_stat_dump(FILE *f, uint32_t lc, const struct lcore_stat *ls)
+{
+ long double st;
+
+ st = (long double)rte_get_timer_hz() / US_PER_S;
+
+ if (lc == UINT32_MAX)
+ fprintf(f, "%s(AGGREGATE)={\n", __func__);
+ else
+ fprintf(f, "%s(lc=%u)={\n", __func__, lc);
+
+ fprintf(f, "\tnb_cycle=%" PRIu64 "(%.2Lf usec),\n",
+ ls->nb_cycle, (long double)ls->nb_cycle / st);
+
+ fprintf(f, "\tDEQ+ENQ={\n");
+
+ fprintf(f, "\t\tnb_call=%" PRIu64 ",\n", ls->op.nb_call);
+ fprintf(f, "\t\tnb_obj=%" PRIu64 ",\n", ls->op.nb_obj);
+ fprintf(f, "\t\tnb_cycle=%" PRIu64 ",\n", ls->op.nb_cycle);
+ fprintf(f, "\t\tobj/call(avg): %.2Lf\n",
+ (long double)ls->op.nb_obj / ls->op.nb_call);
+ fprintf(f, "\t\tcycles/obj(avg): %.2Lf\n",
+ (long double)ls->op.nb_cycle / ls->op.nb_obj);
+ fprintf(f, "\t\tcycles/call(avg): %.2Lf\n",
+ (long double)ls->op.nb_cycle / ls->op.nb_call);
+
+ /* if min/max cycles per call stats was collected */
+ if (ls->op.min_cycle != UINT64_MAX) {
+ fprintf(f, "\t\tmax cycles/call=%" PRIu64 "(%.2Lf usec),\n",
+ ls->op.max_cycle,
+ (long double)ls->op.max_cycle / st);
+ fprintf(f, "\t\tmin cycles/call=%" PRIu64 "(%.2Lf usec),\n",
+ ls->op.min_cycle,
+ (long double)ls->op.min_cycle / st);
+ }
+
+ fprintf(f, "\t},\n");
+ fprintf(f, "};\n");
+}
+
+static void
+fill_ring_elm(struct ring_elem *elm, uint32_t fill)
+{
+ uint32_t i;
+
+ for (i = 0; i != RTE_DIM(elm->cnt); i++)
+ elm->cnt[i] = fill;
+}
+
+static int32_t
+check_updt_elem(struct ring_elem *elm[], uint32_t num,
+ const struct ring_elem *check, const struct ring_elem *fill)
+{
+ uint32_t i;
+
+ static rte_spinlock_t dump_lock;
+
+ for (i = 0; i != num; i++) {
+ if (memcmp(check, elm[i], sizeof(*check)) != 0) {
+ rte_spinlock_lock(&dump_lock);
+ printf("%s(lc=%u, num=%u) failed at %u-th iter, "
+ "offending object: %p\n",
+ __func__, rte_lcore_id(), num, i, elm[i]);
+ rte_memdump(stdout, "expected", check, sizeof(*check));
+ rte_memdump(stdout, "result", elm[i], sizeof(elm[i]));
+ rte_spinlock_unlock(&dump_lock);
+ return -EINVAL;
+ }
+ memcpy(elm[i], fill, sizeof(*elm[i]));
+ }
+
+ return 0;
+}
+
+static int
+check_ring_op(uint32_t exp, uint32_t res, uint32_t lc,
+ const char *fname, const char *opname)
+{
+ if (exp != res) {
+ printf("%s(lc=%u) failure: %s expected: %u, returned %u\n",
+ fname, lc, opname, exp, res);
+ return -ENOSPC;
+ }
+ return 0;
+}
+
+static int
+test_worker_prcs(void *arg)
+{
+ int32_t rc;
+ uint32_t lc, n, num;
+ uint64_t cl, tm0, tm1;
+ struct lcore_arg *la;
+ struct ring_elem def_elm, loc_elm;
+ struct ring_elem *obj[2 * BULK_NUM];
+
+ la = arg;
+ lc = rte_lcore_id();
+
+ fill_ring_elm(&def_elm, UINT32_MAX);
+ fill_ring_elm(&loc_elm, lc);
+
+ while (wrk_cmd != WRK_CMD_RUN) {
+ rte_smp_rmb();
+ rte_pause();
+ }
+
+ cl = rte_rdtsc_precise();
+
+ do {
+ /* num in interval [7/8, 11/8] of BULK_NUM */
+ num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
+
+ /* reset all pointer values */
+ memset(obj, 0, sizeof(obj));
+
+ /* dequeue num elems */
+ tm0 = rte_rdtsc_precise();
+ n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
+ tm0 = rte_rdtsc_precise() - tm0;
+
+ /* check return value and objects */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_dequeue_bulk));
+ if (rc == 0)
+ rc = check_updt_elem(obj, num, &def_elm, &loc_elm);
+ if (rc != 0)
+ break;
+
+ /* enqueue num elems */
+ rte_compiler_barrier();
+ rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
+ if (rc != 0)
+ break;
+
+ tm1 = rte_rdtsc_precise();
+ n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
+ tm1 = rte_rdtsc_precise() - tm1;
+
+ /* check return value */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_enqueue_bulk));
+ if (rc != 0)
+ break;
+
+ lcore_stat_update(&la->stats, 1, num, tm0 + tm1, 1);
+
+ } while (wrk_cmd == WRK_CMD_RUN);
+
+ la->stats.nb_cycle = rte_rdtsc_precise() - cl;
+ return rc;
+}
+
+static int
+test_worker_avg(void *arg)
+{
+ int32_t rc;
+ uint32_t lc, n, num;
+ uint64_t cl;
+ struct lcore_arg *la;
+ struct ring_elem def_elm, loc_elm;
+ struct ring_elem *obj[2 * BULK_NUM];
+
+ la = arg;
+ lc = rte_lcore_id();
+
+ fill_ring_elm(&def_elm, UINT32_MAX);
+ fill_ring_elm(&loc_elm, lc);
+
+ while (wrk_cmd != WRK_CMD_RUN) {
+ rte_smp_rmb();
+ rte_pause();
+ }
+
+ cl = rte_rdtsc_precise();
+
+ do {
+ /* num in interval [7/8, 11/8] of BULK_NUM */
+ num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
+
+ /* reset all pointer values */
+ memset(obj, 0, sizeof(obj));
+
+ /* dequeue num elems */
+ n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
+
+ /* check return value and objects */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_dequeue_bulk));
+ if (rc == 0)
+ rc = check_updt_elem(obj, num, &def_elm, &loc_elm);
+ if (rc != 0)
+ break;
+
+ /* enqueue num elems */
+ rte_compiler_barrier();
+ rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
+ if (rc != 0)
+ break;
+
+ n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
+
+ /* check return value */
+ rc = check_ring_op(num, n, lc, __func__,
+ RTE_STR(_st_ring_enqueue_bulk));
+ if (rc != 0)
+ break;
+
+ lcore_stat_update(&la->stats, 1, num, 0, 0);
+
+ } while (wrk_cmd == WRK_CMD_RUN);
+
+ /* final stats update */
+ cl = rte_rdtsc_precise() - cl;
+ lcore_stat_update(&la->stats, 0, 0, cl, 0);
+ la->stats.nb_cycle = cl;
+
+ return rc;
+}
+
+static void
+mt1_fini(struct rte_ring *rng, void *data)
+{
+ rte_free(rng);
+ rte_free(data);
+}
+
+static int
+mt1_init(struct rte_ring **rng, void **data, uint32_t num)
+{
+ int32_t rc;
+ size_t sz;
+ uint32_t i, nr;
+ struct rte_ring *r;
+ struct ring_elem *elm;
+ void *p;
+
+ *rng = NULL;
+ *data = NULL;
+
+ sz = num * sizeof(*elm);
+ elm = rte_zmalloc(NULL, sz, __alignof__(*elm));
+ if (elm == NULL) {
+ printf("%s: alloc(%zu) for %u elems data failed",
+ __func__, sz, num);
+ return -ENOMEM;
+ }
+
+ *data = elm;
+
+ /* alloc ring */
+ nr = 2 * num;
+ sz = rte_ring_get_memsize(nr);
+ r = rte_zmalloc(NULL, sz, __alignof__(*r));
+ if (r == NULL) {
+ printf("%s: alloc(%zu) for FIFO with %u elems failed",
+ __func__, sz, nr);
+ return -ENOMEM;
+ }
+
+ *rng = r;
+
+ rc = _st_ring_init(r, RING_NAME, nr);
+ if (rc != 0) {
+ printf("%s: _st_ring_init(%p, %u) failed, error: %d(%s)\n",
+ __func__, r, nr, rc, strerror(-rc));
+ return rc;
+ }
+
+ for (i = 0; i != num; i++) {
+ fill_ring_elm(elm + i, UINT32_MAX);
+ p = elm + i;
+ if (_st_ring_enqueue_bulk(r, &p, 1, NULL) != 1)
+ break;
+ }
+
+ if (i != num) {
+ printf("%s: _st_ring_enqueue(%p, %u) returned %u\n",
+ __func__, r, num, i);
+ return -ENOSPC;
+ }
+
+ return 0;
+}
+
+static int
+test_mt1(int (*test)(void *))
+{
+ int32_t rc;
+ uint32_t lc, mc;
+ struct rte_ring *r;
+ void *data;
+ struct lcore_arg arg[RTE_MAX_LCORE];
+
+ static const struct lcore_stat init_stat = {
+ .op.min_cycle = UINT64_MAX,
+ };
+
+ rc = mt1_init(&r, &data, RING_SIZE);
+ if (rc != 0) {
+ mt1_fini(r, data);
+ return rc;
+ }
+
+ memset(arg, 0, sizeof(arg));
+
+ /* launch on all slaves */
+ RTE_LCORE_FOREACH_SLAVE(lc) {
+ arg[lc].rng = r;
+ arg[lc].stats = init_stat;
+ rte_eal_remote_launch(test, &arg[lc], lc);
+ }
+
+ /* signal worker to start test */
+ wrk_cmd = WRK_CMD_RUN;
+ rte_smp_wmb();
+
+ usleep(run_time * US_PER_S);
+
+ /* signal worker to start test */
+ wrk_cmd = WRK_CMD_STOP;
+ rte_smp_wmb();
+
+ /* wait for slaves and collect stats. */
+ mc = rte_lcore_id();
+ arg[mc].stats = init_stat;
+
+ rc = 0;
+ RTE_LCORE_FOREACH_SLAVE(lc) {
+ rc |= rte_eal_wait_lcore(lc);
+ lcore_stat_aggr(&arg[mc].stats, &arg[lc].stats);
+ if (verbose != 0)
+ lcore_stat_dump(stdout, lc, &arg[lc].stats);
+ }
+
+ lcore_stat_dump(stdout, UINT32_MAX, &arg[mc].stats);
+ mt1_fini(r, data);
+ return rc;
+}
+
+static const struct test_case tests[] = {
+ {
+ .name = "MT-WRK_ENQ_DEQ-MST_NONE-PRCS",
+ .func = test_mt1,
+ .wfunc = test_worker_prcs,
+ },
+ {
+ .name = "MT-WRK_ENQ_DEQ-MST_NONE-AVG",
+ .func = test_mt1,
+ .wfunc = test_worker_avg,
+ },
+};
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v3 2/9] ring: prepare ring to allow new sync schemes
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 1/9] test/ring: add contention stress test Konstantin Ananyev
@ 2020-04-03 17:42 ` Konstantin Ananyev
2020-04-08 4:59 ` Honnappa Nagarahalli
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 3/9] ring: introduce RTS ring mode Konstantin Ananyev
` (7 subsequent siblings)
9 siblings, 1 reply; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-03 17:42 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Change from *single* to *sync_type* to allow different
synchronisation schemes to be applied.
Mark *single* as deprecated in comments.
Add new functions to allow user to query ring sync types.
Replace direct access to *single* with appopriate function call.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/test_pdump.c | 6 +-
lib/librte_pdump/rte_pdump.c | 2 +-
lib/librte_port/rte_port_ring.c | 12 ++--
lib/librte_ring/rte_ring.c | 6 +-
lib/librte_ring/rte_ring.h | 113 ++++++++++++++++++++++++++------
lib/librte_ring/rte_ring_elem.h | 8 +--
6 files changed, 108 insertions(+), 39 deletions(-)
diff --git a/app/test/test_pdump.c b/app/test/test_pdump.c
index ad183184c..6a1180bcb 100644
--- a/app/test/test_pdump.c
+++ b/app/test/test_pdump.c
@@ -57,8 +57,7 @@ run_pdump_client_tests(void)
if (ret < 0)
return -1;
mp->flags = 0x0000;
- ring_client = rte_ring_create("SR0", RING_SIZE, rte_socket_id(),
- RING_F_SP_ENQ | RING_F_SC_DEQ);
+ ring_client = rte_ring_create("SR0", RING_SIZE, rte_socket_id(), 0);
if (ring_client == NULL) {
printf("rte_ring_create SR0 failed");
return -1;
@@ -71,9 +70,6 @@ run_pdump_client_tests(void)
}
rte_eth_dev_probing_finish(eth_dev);
- ring_client->prod.single = 0;
- ring_client->cons.single = 0;
-
printf("\n***** flags = RTE_PDUMP_FLAG_TX *****\n");
for (itr = 0; itr < NUM_ITR; itr++) {
diff --git a/lib/librte_pdump/rte_pdump.c b/lib/librte_pdump/rte_pdump.c
index 8a01ac510..65364f2c5 100644
--- a/lib/librte_pdump/rte_pdump.c
+++ b/lib/librte_pdump/rte_pdump.c
@@ -380,7 +380,7 @@ pdump_validate_ring_mp(struct rte_ring *ring, struct rte_mempool *mp)
rte_errno = EINVAL;
return -1;
}
- if (ring->prod.single || ring->cons.single) {
+ if (rte_ring_prod_single(ring) || rte_ring_cons_single(ring)) {
PDUMP_LOG(ERR, "ring with either SP or SC settings"
" is not valid for pdump, should have MP and MC settings\n");
rte_errno = EINVAL;
diff --git a/lib/librte_port/rte_port_ring.c b/lib/librte_port/rte_port_ring.c
index 47fcdd06a..2f6c050fa 100644
--- a/lib/librte_port/rte_port_ring.c
+++ b/lib/librte_port/rte_port_ring.c
@@ -44,8 +44,8 @@ rte_port_ring_reader_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->cons.single && is_multi) ||
- (!(conf->ring->cons.single) && !is_multi)) {
+ (rte_ring_cons_single(conf->ring) && is_multi) ||
+ (!rte_ring_cons_single(conf->ring) && !is_multi)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
}
@@ -171,8 +171,8 @@ rte_port_ring_writer_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->prod.single && is_multi) ||
- (!(conf->ring->prod.single) && !is_multi) ||
+ (rte_ring_prod_single(conf->ring) && is_multi) ||
+ (!rte_ring_prod_single(conf->ring) && !is_multi) ||
(conf->tx_burst_sz > RTE_PORT_IN_BURST_SIZE_MAX)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
@@ -440,8 +440,8 @@ rte_port_ring_writer_nodrop_create_internal(void *params, int socket_id,
/* Check input parameters */
if ((conf == NULL) ||
(conf->ring == NULL) ||
- (conf->ring->prod.single && is_multi) ||
- (!(conf->ring->prod.single) && !is_multi) ||
+ (rte_ring_prod_single(conf->ring) && is_multi) ||
+ (!rte_ring_prod_single(conf->ring) && !is_multi) ||
(conf->tx_burst_sz > RTE_PORT_IN_BURST_SIZE_MAX)) {
RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
return NULL;
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index 77e5de099..fa5733907 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -106,8 +106,10 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
if (ret < 0 || ret >= (int)sizeof(r->name))
return -ENAMETOOLONG;
r->flags = flags;
- r->prod.single = (flags & RING_F_SP_ENQ) ? __IS_SP : __IS_MP;
- r->cons.single = (flags & RING_F_SC_DEQ) ? __IS_SC : __IS_MC;
+ r->prod.sync_type = (flags & RING_F_SP_ENQ) ?
+ RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
+ r->cons.sync_type = (flags & RING_F_SC_DEQ) ?
+ RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
if (flags & RING_F_EXACT_SZ) {
r->size = rte_align32pow2(count + 1);
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index 18fc5d845..d4775a063 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -61,11 +61,27 @@ enum rte_ring_queue_behavior {
#define RTE_RING_NAMESIZE (RTE_MEMZONE_NAMESIZE - \
sizeof(RTE_RING_MZ_PREFIX) + 1)
-/* structure to hold a pair of head/tail values and other metadata */
+/** prod/cons sync types */
+enum rte_ring_sync_type {
+ RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
+ RTE_RING_SYNC_ST, /**< single thread only */
+};
+
+/**
+ * structure to hold a pair of head/tail values and other metadata.
+ * Depending on sync_type format of that structure might be different,
+ * but offset for *sync_type* and *tail* values should remain the same.
+ */
struct rte_ring_headtail {
- volatile uint32_t head; /**< Prod/consumer head. */
- volatile uint32_t tail; /**< Prod/consumer tail. */
- uint32_t single; /**< True if single prod/cons */
+ volatile uint32_t head; /**< prod/consumer head. */
+ volatile uint32_t tail; /**< prod/consumer tail. */
+ RTE_STD_C11
+ union {
+ /** sync type of prod/cons */
+ enum rte_ring_sync_type sync_type;
+ /** deprecated - True if single prod/cons */
+ uint32_t single;
+ };
};
/**
@@ -116,11 +132,10 @@ struct rte_ring {
#define RING_F_EXACT_SZ 0x0004
#define RTE_RING_SZ_MASK (0x7fffffffU) /**< Ring size mask */
-/* @internal defines for passing to the enqueue dequeue worker functions */
-#define __IS_SP 1
-#define __IS_MP 0
-#define __IS_SC 1
-#define __IS_MC 0
+#define __IS_SP RTE_RING_SYNC_ST
+#define __IS_MP RTE_RING_SYNC_MT
+#define __IS_SC RTE_RING_SYNC_ST
+#define __IS_MC RTE_RING_SYNC_MT
/**
* Calculate the memory size needed for a ring
@@ -420,7 +435,7 @@ rte_ring_mp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_MP, free_space);
+ RTE_RING_SYNC_MT, free_space);
}
/**
@@ -443,7 +458,7 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_SP, free_space);
+ RTE_RING_SYNC_ST, free_space);
}
/**
@@ -470,7 +485,7 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->prod.single, free_space);
+ r->prod.sync_type, free_space);
}
/**
@@ -554,7 +569,7 @@ rte_ring_mc_dequeue_bulk(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_MC, available);
+ RTE_RING_SYNC_MT, available);
}
/**
@@ -578,7 +593,7 @@ rte_ring_sc_dequeue_bulk(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- __IS_SC, available);
+ RTE_RING_SYNC_ST, available);
}
/**
@@ -605,7 +620,7 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
/**
@@ -777,6 +792,62 @@ rte_ring_get_capacity(const struct rte_ring *r)
return r->capacity;
}
+/**
+ * Return sync type used by producer in the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Producer sync type value.
+ */
+static inline enum rte_ring_sync_type
+rte_ring_get_prod_sync_type(const struct rte_ring *r)
+{
+ return r->prod.sync_type;
+}
+
+/**
+ * Check is the ring for single producer.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * true if ring is SP, zero otherwise.
+ */
+static inline int
+rte_ring_prod_single(const struct rte_ring *r)
+{
+ return (rte_ring_get_prod_sync_type(r) == RTE_RING_SYNC_ST);
+}
+
+/**
+ * Return sync type used by consumer in the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Consumer sync type value.
+ */
+static inline enum rte_ring_sync_type
+rte_ring_get_cons_sync_type(const struct rte_ring *r)
+{
+ return r->cons.sync_type;
+}
+
+/**
+ * Check is the ring for single consumer.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * true if ring is SC, zero otherwise.
+ */
+static inline int
+rte_ring_cons_single(const struct rte_ring *r)
+{
+ return (rte_ring_get_cons_sync_type(r) == RTE_RING_SYNC_ST);
+}
+
/**
* Dump the status of all rings on the console
*
@@ -820,7 +891,7 @@ rte_ring_mp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_MP, free_space);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_MT, free_space);
}
/**
@@ -843,7 +914,7 @@ rte_ring_sp_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_SP, free_space);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST, free_space);
}
/**
@@ -870,7 +941,7 @@ rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE,
- r->prod.single, free_space);
+ r->prod.sync_type, free_space);
}
/**
@@ -898,7 +969,7 @@ rte_ring_mc_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_MC, available);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_MT, available);
}
/**
@@ -923,7 +994,7 @@ rte_ring_sc_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE, __IS_SC, available);
+ RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST, available);
}
/**
@@ -951,7 +1022,7 @@ rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
{
return __rte_ring_do_dequeue(r, obj_table, n,
RTE_RING_QUEUE_VARIABLE,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 663addc73..28f9836e6 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -570,7 +570,7 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->prod.single, free_space);
+ RTE_RING_QUEUE_FIXED, r->prod.sync_type, free_space);
}
/**
@@ -734,7 +734,7 @@ rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
unsigned int esize, unsigned int n, unsigned int *available)
{
return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->cons.single, available);
+ RTE_RING_QUEUE_FIXED, r->cons.sync_type, available);
}
/**
@@ -902,7 +902,7 @@ rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_VARIABLE, r->prod.single, free_space);
+ RTE_RING_QUEUE_VARIABLE, r->prod.sync_type, free_space);
}
/**
@@ -995,7 +995,7 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
{
return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
RTE_RING_QUEUE_VARIABLE,
- r->cons.single, available);
+ r->cons.sync_type, available);
}
#ifdef __cplusplus
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v3 3/9] ring: introduce RTS ring mode
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 1/9] test/ring: add contention stress test Konstantin Ananyev
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 2/9] ring: prepare ring to allow new sync schemes Konstantin Ananyev
@ 2020-04-03 17:42 ` Konstantin Ananyev
2020-04-04 17:27 ` Wang, Haiyue
2020-04-08 5:00 ` Honnappa Nagarahalli
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 4/9] test/ring: add contention stress test for RTS ring Konstantin Ananyev
` (6 subsequent siblings)
9 siblings, 2 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-03 17:42 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce relaxed tail sync (RTS) mode for MT ring synchronization.
Aim to reduce stall times in case when ring is used on
overcommited cpus (multiple active threads on the same cpu).
The main difference from original MP/MC algorithm is that
tail value is increased not by every thread that finished enqueue/dequeue,
but only by the last one.
That allows threads to avoid spinning on ring tail value,
leaving actual tail value change to the last thread in the update queue.
check-abi.sh reports what I believe is a false-positive about
ring cons/prod changes. As a workaround, devtools/libabigail.abignore is
updated to suppress *struct ring* related errors.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
devtools/libabigail.abignore | 7 +
lib/librte_ring/Makefile | 5 +-
lib/librte_ring/meson.build | 5 +-
lib/librte_ring/rte_ring.c | 100 +++++++-
lib/librte_ring/rte_ring.h | 110 ++++++++-
lib/librte_ring/rte_ring_elem.h | 86 ++++++-
lib/librte_ring/rte_ring_rts.h | 316 +++++++++++++++++++++++++
lib/librte_ring/rte_ring_rts_elem.h | 205 ++++++++++++++++
lib/librte_ring/rte_ring_rts_generic.h | 210 ++++++++++++++++
9 files changed, 1015 insertions(+), 29 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_rts.h
create mode 100644 lib/librte_ring/rte_ring_rts_elem.h
create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
diff --git a/devtools/libabigail.abignore b/devtools/libabigail.abignore
index a59df8f13..cd86d89ca 100644
--- a/devtools/libabigail.abignore
+++ b/devtools/libabigail.abignore
@@ -11,3 +11,10 @@
type_kind = enum
name = rte_crypto_asym_xform_type
changed_enumerators = RTE_CRYPTO_ASYM_XFORM_TYPE_LIST_END
+; Ignore updates of ring prod/cons
+[suppress_type]
+ type_kind = struct
+ name = rte_ring
+[suppress_type]
+ type_kind = struct
+ name = rte_event_ring
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 917c560ad..8f5c284cc 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -18,6 +18,9 @@ SRCS-$(CONFIG_RTE_LIBRTE_RING) := rte_ring.c
SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_elem.h \
rte_ring_generic.h \
- rte_ring_c11_mem.h
+ rte_ring_c11_mem.h \
+ rte_ring_rts.h \
+ rte_ring_rts_elem.h \
+ rte_ring_rts_generic.h
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index f2f3ccc88..612936afb 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -5,7 +5,10 @@ sources = files('rte_ring.c')
headers = files('rte_ring.h',
'rte_ring_elem.h',
'rte_ring_c11_mem.h',
- 'rte_ring_generic.h')
+ 'rte_ring_generic.h',
+ 'rte_ring_rts.h',
+ 'rte_ring_rts_elem.h',
+ 'rte_ring_rts_generic.h')
# rte_ring_create_elem and rte_ring_get_memsize_elem are experimental
allow_experimental_apis = true
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index fa5733907..222eec0fb 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -45,6 +45,9 @@ EAL_REGISTER_TAILQ(rte_ring_tailq)
/* true if x is a power of 2 */
#define POWEROF2(x) ((((x)-1) & (x)) == 0)
+/* by default set head/tail distance as 1/8 of ring capacity */
+#define HTD_MAX_DEF 8
+
/* return the size of memory occupied by a ring */
ssize_t
rte_ring_get_memsize_elem(unsigned int esize, unsigned int count)
@@ -79,11 +82,84 @@ rte_ring_get_memsize(unsigned int count)
return rte_ring_get_memsize_elem(sizeof(void *), count);
}
+/*
+ * internal helper function to reset prod/cons head-tail values.
+ */
+static void
+reset_headtail(void *p)
+{
+ struct rte_ring_headtail *ht;
+ struct rte_ring_rts_headtail *ht_rts;
+
+ ht = p;
+ ht_rts = p;
+
+ switch (ht->sync_type) {
+ case RTE_RING_SYNC_MT:
+ case RTE_RING_SYNC_ST:
+ ht->head = 0;
+ ht->tail = 0;
+ break;
+ case RTE_RING_SYNC_MT_RTS:
+ ht_rts->head.raw = 0;
+ ht_rts->tail.raw = 0;
+ break;
+ default:
+ /* unknown sync mode */
+ RTE_ASSERT(0);
+ }
+}
+
void
rte_ring_reset(struct rte_ring *r)
{
- r->prod.head = r->cons.head = 0;
- r->prod.tail = r->cons.tail = 0;
+ reset_headtail(&r->prod);
+ reset_headtail(&r->cons);
+}
+
+/*
+ * helper function, calculates sync_type values for prod and cons
+ * based on input flags. Returns zero at success or negative
+ * errno value otherwise.
+ */
+static int
+get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
+ enum rte_ring_sync_type *cons_st)
+{
+ static const uint32_t prod_st_flags =
+ (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
+ static const uint32_t cons_st_flags =
+ (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
+
+ switch (flags & prod_st_flags) {
+ case 0:
+ *prod_st = RTE_RING_SYNC_MT;
+ break;
+ case RING_F_SP_ENQ:
+ *prod_st = RTE_RING_SYNC_ST;
+ break;
+ case RING_F_MP_RTS_ENQ:
+ *prod_st = RTE_RING_SYNC_MT_RTS;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ switch (flags & cons_st_flags) {
+ case 0:
+ *cons_st = RTE_RING_SYNC_MT;
+ break;
+ case RING_F_SC_DEQ:
+ *cons_st = RTE_RING_SYNC_ST;
+ break;
+ case RING_F_MC_RTS_DEQ:
+ *cons_st = RTE_RING_SYNC_MT_RTS;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
}
int
@@ -100,16 +176,20 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
RTE_CACHE_LINE_MASK) != 0);
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
+ offsetof(struct rte_ring_rts_headtail, sync_type));
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
+ offsetof(struct rte_ring_rts_headtail, tail.val.pos));
+
/* init the ring structure */
memset(r, 0, sizeof(*r));
ret = strlcpy(r->name, name, sizeof(r->name));
if (ret < 0 || ret >= (int)sizeof(r->name))
return -ENAMETOOLONG;
r->flags = flags;
- r->prod.sync_type = (flags & RING_F_SP_ENQ) ?
- RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
- r->cons.sync_type = (flags & RING_F_SC_DEQ) ?
- RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
+ ret = get_sync_type(flags, &r->prod.sync_type, &r->cons.sync_type);
+ if (ret != 0)
+ return ret;
if (flags & RING_F_EXACT_SZ) {
r->size = rte_align32pow2(count + 1);
@@ -126,8 +206,12 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
r->mask = count - 1;
r->capacity = r->mask;
}
- r->prod.head = r->cons.head = 0;
- r->prod.tail = r->cons.tail = 0;
+
+ /* set default values for head-tail distance */
+ if (flags & RING_F_MP_RTS_ENQ)
+ rte_ring_set_prod_htd_max(r, r->capacity / HTD_MAX_DEF);
+ if (flags & RING_F_MC_RTS_DEQ)
+ rte_ring_set_cons_htd_max(r, r->capacity / HTD_MAX_DEF);
return 0;
}
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index d4775a063..f6f084d79 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -48,6 +48,7 @@ extern "C" {
#include <rte_branch_prediction.h>
#include <rte_memzone.h>
#include <rte_pause.h>
+#include <rte_debug.h>
#define RTE_TAILQ_RING_NAME "RTE_RING"
@@ -65,10 +66,13 @@ enum rte_ring_queue_behavior {
enum rte_ring_sync_type {
RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
RTE_RING_SYNC_ST, /**< single thread only */
+#ifdef ALLOW_EXPERIMENTAL_API
+ RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
+#endif
};
/**
- * structure to hold a pair of head/tail values and other metadata.
+ * structures to hold a pair of head/tail values and other metadata.
* Depending on sync_type format of that structure might be different,
* but offset for *sync_type* and *tail* values should remain the same.
*/
@@ -84,6 +88,21 @@ struct rte_ring_headtail {
};
};
+union rte_ring_ht_poscnt {
+ uint64_t raw;
+ struct {
+ uint32_t cnt; /**< head/tail reference counter */
+ uint32_t pos; /**< head/tail position */
+ } val;
+};
+
+struct rte_ring_rts_headtail {
+ volatile union rte_ring_ht_poscnt tail;
+ enum rte_ring_sync_type sync_type; /**< sync type of prod/cons */
+ uint32_t htd_max; /**< max allowed distance between head/tail */
+ volatile union rte_ring_ht_poscnt head;
+};
+
/**
* An RTE ring structure.
*
@@ -111,11 +130,21 @@ struct rte_ring {
char pad0 __rte_cache_aligned; /**< empty cache line */
/** Ring producer status. */
- struct rte_ring_headtail prod __rte_cache_aligned;
+ RTE_STD_C11
+ union {
+ struct rte_ring_headtail prod;
+ struct rte_ring_rts_headtail rts_prod;
+ } __rte_cache_aligned;
+
char pad1 __rte_cache_aligned; /**< empty cache line */
/** Ring consumer status. */
- struct rte_ring_headtail cons __rte_cache_aligned;
+ RTE_STD_C11
+ union {
+ struct rte_ring_headtail cons;
+ struct rte_ring_rts_headtail rts_cons;
+ } __rte_cache_aligned;
+
char pad2 __rte_cache_aligned; /**< empty cache line */
};
@@ -132,6 +161,9 @@ struct rte_ring {
#define RING_F_EXACT_SZ 0x0004
#define RTE_RING_SZ_MASK (0x7fffffffU) /**< Ring size mask */
+#define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS". */
+#define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC RTS". */
+
#define __IS_SP RTE_RING_SYNC_ST
#define __IS_MP RTE_RING_SYNC_MT
#define __IS_SC RTE_RING_SYNC_ST
@@ -461,6 +493,10 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
RTE_RING_SYNC_ST, free_space);
}
+#ifdef ALLOW_EXPERIMENTAL_API
+#include <rte_ring_rts.h>
+#endif
+
/**
* Enqueue several objects on a ring.
*
@@ -484,8 +520,21 @@ static __rte_always_inline unsigned int
rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_bulk(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_bulk(r, obj_table, n, free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_bulk(r, obj_table, n,
+ free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -619,8 +668,20 @@ static __rte_always_inline unsigned int
rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
unsigned int *available)
{
- return __rte_ring_do_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_bulk(r, obj_table, n, available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_bulk(r, obj_table, n, available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_bulk(r, obj_table, n, available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -940,8 +1001,21 @@ static __rte_always_inline unsigned
rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue(r, obj_table, n, RTE_RING_QUEUE_VARIABLE,
- r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_burst(r, obj_table, n, free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_burst(r, obj_table, n, free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_burst(r, obj_table, n,
+ free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
/**
@@ -1020,9 +1094,21 @@ static __rte_always_inline unsigned
rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
unsigned int n, unsigned int *available)
{
- return __rte_ring_do_dequeue(r, obj_table, n,
- RTE_RING_QUEUE_VARIABLE,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_burst(r, obj_table, n, available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_burst(r, obj_table, n, available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_burst(r, obj_table, n,
+ available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ return 0;
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 28f9836e6..5de0850dc 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -542,6 +542,8 @@ rte_ring_sp_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
RTE_RING_QUEUE_FIXED, __IS_SP, free_space);
}
+#include <rte_ring_rts_elem.h>
+
/**
* Enqueue several objects on a ring.
*
@@ -571,6 +573,26 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
{
return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
RTE_RING_QUEUE_FIXED, r->prod.sync_type, free_space);
+
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (free_space != NULL)
+ *free_space = 0;
+ return 0;
}
/**
@@ -733,8 +755,25 @@ static __rte_always_inline unsigned int
rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
unsigned int esize, unsigned int n, unsigned int *available)
{
- return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_FIXED, r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_bulk_elem(r, obj_table, esize, n,
+ available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_bulk_elem(r, obj_table, esize, n,
+ available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_bulk_elem(r, obj_table, esize,
+ n, available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (available != NULL)
+ *available = 0;
+ return 0;
}
/**
@@ -901,8 +940,25 @@ static __rte_always_inline unsigned
rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
unsigned int esize, unsigned int n, unsigned int *free_space)
{
- return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_VARIABLE, r->prod.sync_type, free_space);
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mp_enqueue_burst_elem(r, obj_table, esize, n,
+ free_space);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sp_enqueue_burst_elem(r, obj_table, esize, n,
+ free_space);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mp_rts_enqueue_burst_elem(r, obj_table, esize,
+ n, free_space);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (free_space != NULL)
+ *free_space = 0;
+ return 0;
}
/**
@@ -993,9 +1049,25 @@ static __rte_always_inline unsigned int
rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
unsigned int esize, unsigned int n, unsigned int *available)
{
- return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
- RTE_RING_QUEUE_VARIABLE,
- r->cons.sync_type, available);
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_MT:
+ return rte_ring_mc_dequeue_burst_elem(r, obj_table, esize, n,
+ available);
+ case RTE_RING_SYNC_ST:
+ return rte_ring_sc_dequeue_burst_elem(r, obj_table, esize, n,
+ available);
+#ifdef ALLOW_EXPERIMENTAL_API
+ case RTE_RING_SYNC_MT_RTS:
+ return rte_ring_mc_rts_dequeue_burst_elem(r, obj_table, esize,
+ n, available);
+#endif
+ }
+
+ /* valid ring should never reach this point */
+ RTE_ASSERT(0);
+ if (available != NULL)
+ *available = 0;
+ return 0;
}
#ifdef __cplusplus
diff --git a/lib/librte_ring/rte_ring_rts.h b/lib/librte_ring/rte_ring_rts.h
new file mode 100644
index 000000000..18404fe48
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts.h
@@ -0,0 +1,316 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_H_
+#define _RTE_RING_RTS_H_
+
+/**
+ * @file rte_ring_rts.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring.h> instead.
+ *
+ * Contains functions for Relaxed Tail Sync (RTS) ring mode.
+ * The main idea remains the same as for our original MP/MC synchronization
+ * mechanism.
+ * The main difference is that tail value is increased not
+ * by every thread that finished enqueue/dequeue,
+ * but only by the last one doing enqueue/dequeue.
+ * That allows threads to skip spinning on tail value,
+ * leaving actual tail value change to last thread in the update queue.
+ * RTS requires 2 64-bit CAS for each enqueue(/dequeue) operation:
+ * one for head update, second for tail update.
+ * As a gain it allows thread to avoid spinning/waiting on tail value.
+ * In comparision original MP/MC algorithm requires one 32-bit CAS
+ * for head update and waiting/spinning on tail value.
+ *
+ * Brief outline:
+ * - introduce refcnt for both head and tail.
+ * - increment head.refcnt for each head.value update
+ * - write head:value and head:refcnt atomically (64-bit CAS)
+ * - move tail.value ahead only when tail.refcnt + 1 == head.refcnt
+ * - increment tail.refcnt when each enqueue/dequeue op finishes
+ * (no matter is tail:value going to change or not)
+ * - write tail.value and tail.recnt atomically (64-bit CAS)
+ *
+ * To avoid producer/consumer starvation:
+ * - limit max allowed distance between head and tail value (HTD_MAX).
+ * I.E. thread is allowed to proceed with changing head.value,
+ * only when: head.value - tail.value <= HTD_MAX
+ * HTD_MAX is an optional parameter.
+ * With HTD_MAX == 0 we'll have fully serialized ring -
+ * i.e. only one thread at a time will be able to enqueue/dequeue
+ * to/from the ring.
+ * With HTD_MAX >= ring.capacity - no limitation.
+ * By default HTD_MAX == ring.capacity / 8.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_rts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_enqueue(struct rte_ring *r, void * const *obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_rts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_rts_update_tail(&r->rts_prod);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_dequeue(struct rte_ring *r, void **obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_rts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_rts_update_tail(&r->rts_cons);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_rts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_rts_dequeue_bulk(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ available);
+}
+
+/**
+ * Return producer max Head-Tail-Distance (HTD).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Producer HTD value, if producer is set in appropriate sync mode,
+ * or UINT32_MAX otherwise.
+ */
+__rte_experimental
+static inline uint32_t
+rte_ring_get_prod_htd_max(const struct rte_ring *r)
+{
+ if (r->prod.sync_type == RTE_RING_SYNC_MT_RTS)
+ return r->rts_prod.htd_max;
+ return UINT32_MAX;
+}
+
+/**
+ * Set producer max Head-Tail-Distance (HTD).
+ * Note that producer has to use appropriate sync mode (RTS).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param v
+ * new HTD value to setup.
+ * @return
+ * Zero on success, or negative error code otherwise.
+ */
+__rte_experimental
+static inline int
+rte_ring_set_prod_htd_max(struct rte_ring *r, uint32_t v)
+{
+ if (r->prod.sync_type != RTE_RING_SYNC_MT_RTS)
+ return -ENOTSUP;
+
+ r->rts_prod.htd_max = v;
+ return 0;
+}
+
+/**
+ * Return consumer max Head-Tail-Distance (HTD).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @return
+ * Consumer HTD value, if consumer is set in appropriate sync mode,
+ * or UINT32_MAX otherwise.
+ */
+__rte_experimental
+static inline uint32_t
+rte_ring_get_cons_htd_max(const struct rte_ring *r)
+{
+ if (r->cons.sync_type == RTE_RING_SYNC_MT_RTS)
+ return r->rts_cons.htd_max;
+ return UINT32_MAX;
+}
+
+/**
+ * Set consumer max Head-Tail-Distance (HTD).
+ * Note that consumer has to use appropriate sync mode (RTS).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param v
+ * new HTD value to setup.
+ * @return
+ * Zero on success, or negative error code otherwise.
+ */
+__rte_experimental
+static inline int
+rte_ring_set_cons_htd_max(struct rte_ring *r, uint32_t v)
+{
+ if (r->cons.sync_type != RTE_RING_SYNC_MT_RTS)
+ return -ENOTSUP;
+
+ r->rts_cons.htd_max = v;
+ return 0;
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_rts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_rts_dequeue_burst(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_RTS_H_ */
diff --git a/lib/librte_ring/rte_ring_rts_elem.h b/lib/librte_ring/rte_ring_rts_elem.h
new file mode 100644
index 000000000..71a331b23
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts_elem.h
@@ -0,0 +1,205 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_ELEM_H_
+#define _RTE_RING_RTS_ELEM_H_
+
+/**
+ * @file rte_ring_rts_elem.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ *
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring_elem.h> instead.
+ * Contains *ring_elem* functions for Relaxed Tail Sync (RTS) ring mode.
+ * for more details please refer to <rte_ring_rts.h>.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_rts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_enqueue_elem(struct rte_ring *r, void * const *obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_rts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ __rte_ring_enqueue_elems(r, head, obj_table, esize, n);
+ __rte_ring_rts_update_tail(&r->rts_prod);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the RTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_rts_dequeue_elem(struct rte_ring *r, void **obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_rts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ __rte_ring_dequeue_elems(r, head, obj_table, esize, n);
+ __rte_ring_rts_update_tail(&r->rts_cons);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_rts_enqueue_bulk_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_rts_dequeue_bulk_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, available);
+}
+
+/**
+ * Enqueue several objects on the RTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_rts_enqueue_burst_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_rts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an RTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_rts_dequeue_burst_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_rts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_RTS_ELEM_H_ */
diff --git a/lib/librte_ring/rte_ring_rts_generic.h b/lib/librte_ring/rte_ring_rts_generic.h
new file mode 100644
index 000000000..f88460d47
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts_generic.h
@@ -0,0 +1,210 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_GENERIC_H_
+#define _RTE_RING_RTS_GENERIC_H_
+
+/**
+ * @file rte_ring_rts_generic.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for Relaxed Tail Sync (RTS) ring mode.
+ * For more information please refer to <rte_ring_rts.h>.
+ */
+
+/**
+ * @internal This function updates tail values.
+ */
+static __rte_always_inline void
+__rte_ring_rts_update_tail(struct rte_ring_rts_headtail *ht)
+{
+ union rte_ring_ht_poscnt h, ot, nt;
+
+ /*
+ * If there are other enqueues/dequeues in progress that
+ * might preceded us, then don't update tail with new value.
+ */
+
+ do {
+ ot.raw = ht->tail.raw;
+ rte_smp_rmb();
+
+ /* on 32-bit systems we have to do atomic read here */
+ h.raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->head.raw);
+
+ nt.raw = ot.raw;
+ if (++nt.val.cnt == h.val.cnt)
+ nt.val.pos = h.val.pos;
+
+ } while (rte_atomic64_cmpset(&ht->tail.raw, ot.raw, nt.raw) == 0);
+}
+
+/**
+ * @internal This function waits till head/tail distance wouldn't
+ * exceed pre-defined max value.
+ */
+static __rte_always_inline void
+__rte_ring_rts_head_wait(const struct rte_ring_rts_headtail *ht,
+ union rte_ring_ht_poscnt *h)
+{
+ uint32_t max;
+
+ max = ht->htd_max;
+ h->raw = ht->head.raw;
+ rte_smp_rmb();
+
+ while (h->val.pos - ht->tail.val.pos > max) {
+ rte_pause();
+ h->raw = ht->head.raw;
+ rte_smp_rmb();
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue.
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_rts_move_prod_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* read prod head (may spin on prod tail) */
+ __rte_ring_rts_head_wait(&r->rts_prod, &oh);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - oh.val.pos;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ break;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (rte_atomic64_cmpset(&r->rts_prod.head.raw,
+ oh.raw, nh.raw) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_rts_move_cons_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* read cons head (may spin on cons tail) */
+ __rte_ring_rts_head_wait(&r->rts_cons, &oh);
+
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - oh.val.pos;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ break;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (rte_atomic64_cmpset(&r->rts_cons.head.raw,
+ oh.raw, nh.raw) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+#endif /* _RTE_RING_RTS_GENERIC_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v3 4/9] test/ring: add contention stress test for RTS ring
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
` (2 preceding siblings ...)
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 3/9] ring: introduce RTS ring mode Konstantin Ananyev
@ 2020-04-03 17:42 ` Konstantin Ananyev
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 5/9] ring: introduce HTS ring mode Konstantin Ananyev
` (5 subsequent siblings)
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-03 17:42 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce new test case to test RTS ring mode under contention.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_rts_stress.c | 32 ++++++++++++++++++++++++++++++++
app/test/test_ring_stress.c | 3 +++
app/test/test_ring_stress.h | 1 +
5 files changed, 38 insertions(+)
create mode 100644 app/test/test_ring_rts_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index 4eefaa887..3e15f3791 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -79,6 +79,7 @@ SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
SRCS-y += test_ring_mpmc_stress.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
diff --git a/app/test/meson.build b/app/test/meson.build
index 827b04886..bb67a49f0 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -102,6 +102,7 @@ test_sources = files('commands.c',
'test_ring.c',
'test_ring_mpmc_stress.c',
'test_ring_perf.c',
+ 'test_ring_rts_stress.c',
'test_ring_stress.c',
'test_rwlock.c',
'test_sched.c',
diff --git a/app/test/test_ring_rts_stress.c b/app/test/test_ring_rts_stress.c
new file mode 100644
index 000000000..f5255f24c
--- /dev/null
+++ b/app/test/test_ring_rts_stress.c
@@ -0,0 +1,32 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_mc_rts_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_mp_rts_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_RTS_ENQ | RING_F_MC_RTS_DEQ);
+}
+
+const struct test test_ring_rts_stress = {
+ .name = "MT_RTS",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
index 60706f799..eab395e30 100644
--- a/app/test/test_ring_stress.c
+++ b/app/test/test_ring_stress.c
@@ -40,6 +40,9 @@ test_ring_stress(void)
n += test_ring_mpmc_stress.nb_case;
k += run_test(&test_ring_mpmc_stress);
+ n += test_ring_rts_stress.nb_case;
+ k += run_test(&test_ring_rts_stress);
+
printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
n, k, n - k);
return (k != n);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
index 60eac6216..32aae2072 100644
--- a/app/test/test_ring_stress.h
+++ b/app/test/test_ring_stress.h
@@ -33,3 +33,4 @@ struct test {
};
extern const struct test test_ring_mpmc_stress;
+extern const struct test test_ring_rts_stress;
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v3 5/9] ring: introduce HTS ring mode
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
` (3 preceding siblings ...)
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 4/9] test/ring: add contention stress test for RTS ring Konstantin Ananyev
@ 2020-04-03 17:42 ` Konstantin Ananyev
2020-04-13 23:27 ` Honnappa Nagarahalli
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 6/9] test/ring: add contention stress test for HTS ring Konstantin Ananyev
` (4 subsequent siblings)
9 siblings, 1 reply; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-03 17:42 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce head/tail sync mode for MT ring synchronization.
In that mode enqueue/dequeue operation is fully serialized:
only one thread at a time is allowed to perform given op.
Suppose to reduce stall times in case when ring is used on
overcommitted cpus (multiple active threads on the same cpu).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 3 +
lib/librte_ring/meson.build | 3 +
lib/librte_ring/rte_ring.c | 20 ++-
lib/librte_ring/rte_ring.h | 31 ++++
lib/librte_ring/rte_ring_elem.h | 13 ++
lib/librte_ring/rte_ring_hts.h | 210 +++++++++++++++++++++++++
lib/librte_ring/rte_ring_hts_elem.h | 205 ++++++++++++++++++++++++
lib/librte_ring/rte_ring_hts_generic.h | 198 +++++++++++++++++++++++
8 files changed, 681 insertions(+), 2 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_hts.h
create mode 100644 lib/librte_ring/rte_ring_hts_elem.h
create mode 100644 lib/librte_ring/rte_ring_hts_generic.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 8f5c284cc..6fe500f0d 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -19,6 +19,9 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_elem.h \
rte_ring_generic.h \
rte_ring_c11_mem.h \
+ rte_ring_hts.h \
+ rte_ring_hts_elem.h \
+ rte_ring_hts_generic.h \
rte_ring_rts.h \
rte_ring_rts_elem.h \
rte_ring_rts_generic.h
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index 612936afb..8e86e037a 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -6,6 +6,9 @@ headers = files('rte_ring.h',
'rte_ring_elem.h',
'rte_ring_c11_mem.h',
'rte_ring_generic.h',
+ 'rte_ring_hts.h',
+ 'rte_ring_hts_elem.h',
+ 'rte_ring_hts_generic.h',
'rte_ring_rts.h',
'rte_ring_rts_elem.h',
'rte_ring_rts_generic.h')
diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c
index 222eec0fb..ebe5ccf0d 100644
--- a/lib/librte_ring/rte_ring.c
+++ b/lib/librte_ring/rte_ring.c
@@ -89,9 +89,11 @@ static void
reset_headtail(void *p)
{
struct rte_ring_headtail *ht;
+ struct rte_ring_hts_headtail *ht_hts;
struct rte_ring_rts_headtail *ht_rts;
ht = p;
+ ht_hts = p;
ht_rts = p;
switch (ht->sync_type) {
@@ -104,6 +106,9 @@ reset_headtail(void *p)
ht_rts->head.raw = 0;
ht_rts->tail.raw = 0;
break;
+ case RTE_RING_SYNC_MT_HTS:
+ ht_hts->ht.raw = 0;
+ break;
default:
/* unknown sync mode */
RTE_ASSERT(0);
@@ -127,9 +132,9 @@ get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
enum rte_ring_sync_type *cons_st)
{
static const uint32_t prod_st_flags =
- (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
+ (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ | RING_F_MP_HTS_ENQ);
static const uint32_t cons_st_flags =
- (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
+ (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ | RING_F_MC_HTS_DEQ);
switch (flags & prod_st_flags) {
case 0:
@@ -141,6 +146,9 @@ get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
case RING_F_MP_RTS_ENQ:
*prod_st = RTE_RING_SYNC_MT_RTS;
break;
+ case RING_F_MP_HTS_ENQ:
+ *prod_st = RTE_RING_SYNC_MT_HTS;
+ break;
default:
return -EINVAL;
}
@@ -155,6 +163,9 @@ get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
case RING_F_MC_RTS_DEQ:
*cons_st = RTE_RING_SYNC_MT_RTS;
break;
+ case RING_F_MC_HTS_DEQ:
+ *cons_st = RTE_RING_SYNC_MT_HTS;
+ break;
default:
return -EINVAL;
}
@@ -176,6 +187,11 @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
RTE_CACHE_LINE_MASK) != 0);
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
+ offsetof(struct rte_ring_hts_headtail, sync_type));
+ RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
+ offsetof(struct rte_ring_hts_headtail, ht.pos.tail));
+
RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
offsetof(struct rte_ring_rts_headtail, sync_type));
RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h
index f6f084d79..6e4213afa 100644
--- a/lib/librte_ring/rte_ring.h
+++ b/lib/librte_ring/rte_ring.h
@@ -68,6 +68,7 @@ enum rte_ring_sync_type {
RTE_RING_SYNC_ST, /**< single thread only */
#ifdef ALLOW_EXPERIMENTAL_API
RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
+ RTE_RING_SYNC_MT_HTS, /**< multi-thread head/tail sync */
#endif
};
@@ -103,6 +104,19 @@ struct rte_ring_rts_headtail {
volatile union rte_ring_ht_poscnt head;
};
+union rte_ring_ht_pos {
+ uint64_t raw;
+ struct {
+ uint32_t head; /**< head position */
+ uint32_t tail; /**< tail position */
+ } pos;
+};
+
+struct rte_ring_hts_headtail {
+ volatile union rte_ring_ht_pos ht;
+ enum rte_ring_sync_type sync_type; /**< sync type of prod/cons */
+};
+
/**
* An RTE ring structure.
*
@@ -133,6 +147,7 @@ struct rte_ring {
RTE_STD_C11
union {
struct rte_ring_headtail prod;
+ struct rte_ring_hts_headtail hts_prod;
struct rte_ring_rts_headtail rts_prod;
} __rte_cache_aligned;
@@ -142,6 +157,7 @@ struct rte_ring {
RTE_STD_C11
union {
struct rte_ring_headtail cons;
+ struct rte_ring_hts_headtail hts_cons;
struct rte_ring_rts_headtail rts_cons;
} __rte_cache_aligned;
@@ -164,6 +180,9 @@ struct rte_ring {
#define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS". */
#define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC RTS". */
+#define RING_F_MP_HTS_ENQ 0x0020 /**< The default enqueue is "MP HTS". */
+#define RING_F_MC_HTS_DEQ 0x0040 /**< The default dequeue is "MC HTS". */
+
#define __IS_SP RTE_RING_SYNC_ST
#define __IS_MP RTE_RING_SYNC_MT
#define __IS_SC RTE_RING_SYNC_ST
@@ -494,6 +513,7 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
}
#ifdef ALLOW_EXPERIMENTAL_API
+#include <rte_ring_hts.h>
#include <rte_ring_rts.h>
#endif
@@ -529,6 +549,9 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_bulk(r, obj_table, n,
free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_bulk(r, obj_table, n,
+ free_space);
#endif
}
@@ -676,6 +699,8 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
#ifdef ALLOW_EXPERIMENTAL_API
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_bulk(r, obj_table, n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_bulk(r, obj_table, n, available);
#endif
}
@@ -1010,6 +1035,9 @@ rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_burst(r, obj_table, n,
free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_burst(r, obj_table, n,
+ free_space);
#endif
}
@@ -1103,6 +1131,9 @@ rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_burst(r, obj_table, n,
available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_burst(r, obj_table, n,
+ available);
#endif
}
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 5de0850dc..010a564c1 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -542,6 +542,7 @@ rte_ring_sp_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
RTE_RING_QUEUE_FIXED, __IS_SP, free_space);
}
+#include <rte_ring_hts_elem.h>
#include <rte_ring_rts_elem.h>
/**
@@ -585,6 +586,9 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_bulk_elem(r, obj_table, esize, n,
free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_bulk_elem(r, obj_table, esize, n,
+ free_space);
#endif
}
@@ -766,6 +770,9 @@ rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_bulk_elem(r, obj_table, esize,
n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_bulk_elem(r, obj_table, esize,
+ n, available);
#endif
}
@@ -951,6 +958,9 @@ rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mp_rts_enqueue_burst_elem(r, obj_table, esize,
n, free_space);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mp_hts_enqueue_burst_elem(r, obj_table, esize,
+ n, free_space);
#endif
}
@@ -1060,6 +1070,9 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
case RTE_RING_SYNC_MT_RTS:
return rte_ring_mc_rts_dequeue_burst_elem(r, obj_table, esize,
n, available);
+ case RTE_RING_SYNC_MT_HTS:
+ return rte_ring_mc_hts_dequeue_burst_elem(r, obj_table, esize,
+ n, available);
#endif
}
diff --git a/lib/librte_ring/rte_ring_hts.h b/lib/librte_ring/rte_ring_hts.h
new file mode 100644
index 000000000..062d7be6c
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts.h
@@ -0,0 +1,210 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_H_
+#define _RTE_RING_HTS_H_
+
+/**
+ * @file rte_ring_hts.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring.h> instead.
+ *
+ * Contains functions for serialized, aka Head-Tail Sync (HTS) ring mode.
+ * In that mode enqueue/dequeue operation is fully serialized:
+ * at any given moement only one enqueue/dequeue operation can proceed.
+ * This is achieved by thread is allowed to proceed with changing head.value
+ * only when head.value == tail.value.
+ * Both head and tail values are updated atomically (as one 64-bit value).
+ * To achieve that 64-bit CAS is used by head update routine.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_hts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_enqueue(struct rte_ring *r, void * const *obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_hts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_hts_update_tail(&r->hts_prod, n, 1);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_dequeue(struct rte_ring *r, void **obj_table,
+ uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_hts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
+ __rte_ring_hts_update_tail(&r->hts_cons, n, 0);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_hts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_hts_dequeue_bulk(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue(r, obj_table, n, RTE_RING_QUEUE_FIXED,
+ available);
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_hts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
+ unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_hts_dequeue_burst(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue(r, obj_table, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_HTS_H_ */
diff --git a/lib/librte_ring/rte_ring_hts_elem.h b/lib/librte_ring/rte_ring_hts_elem.h
new file mode 100644
index 000000000..34f0d121d
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts_elem.h
@@ -0,0 +1,205 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_ELEM_H_
+#define _RTE_RING_HTS_ELEM_H_
+
+/**
+ * @file rte_ring_hts_elem.h
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring_elem.h> instead.
+ *
+ * Contains *ring_elem* functions for Head-Tail Sync (HTS) ring mode.
+ * for more details please refer to <rte_ring_hts.h>.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include <rte_ring_hts_generic.h>
+
+/**
+ * @internal Enqueue several objects on the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param free_space
+ * returns the amount of space after the enqueue operation has finished
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_enqueue_elem(struct rte_ring *r, void * const *obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *free_space)
+{
+ uint32_t free, head;
+
+ n = __rte_ring_hts_move_prod_head(r, n, behavior, &head, &free);
+
+ if (n != 0) {
+ __rte_ring_enqueue_elems(r, head, obj_table, esize, n);
+ __rte_ring_hts_update_tail(&r->hts_prod, n, 1);
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * @internal Dequeue several objects from the HTS ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to pull from the ring.
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param available
+ * returns the number of remaining ring entries after the dequeue has finished
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_hts_dequeue_elem(struct rte_ring *r, void **obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t entries, head;
+
+ n = __rte_ring_hts_move_cons_head(r, n, behavior, &head, &entries);
+
+ if (n != 0) {
+ __rte_ring_dequeue_elems(r, head, obj_table, esize, n);
+ __rte_ring_hts_update_tail(&r->hts_cons, n, 0);
+ }
+
+ if (available != NULL)
+ *available = entries - n;
+ return n;
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mp_hts_enqueue_bulk_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_mc_hts_dequeue_bulk_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, available);
+}
+
+/**
+ * Enqueue several objects on the HTS ring (multi-producers safe).
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects).
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * - n: Actual number of objects enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mp_hts_enqueue_burst_elem(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n, unsigned int *free_space)
+{
+ return __rte_ring_do_hts_enqueue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, free_space);
+}
+
+/**
+ * Dequeue several objects from an HTS ring (multi-consumers safe).
+ * When the requested objects are more than the available objects,
+ * only dequeue the actual number of objects.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * - n: Actual number of objects dequeued, 0 if ring is empty
+ */
+__rte_experimental
+static __rte_always_inline unsigned
+rte_ring_mc_hts_dequeue_burst_elem(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_hts_dequeue_elem(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_HTS_ELEM_H_ */
diff --git a/lib/librte_ring/rte_ring_hts_generic.h b/lib/librte_ring/rte_ring_hts_generic.h
new file mode 100644
index 000000000..da08f1d94
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts_generic.h
@@ -0,0 +1,198 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2020 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_GENERIC_H_
+#define _RTE_RING_HTS_GENERIC_H_
+
+/**
+ * @file rte_ring_hts_generic.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for head/tail sync (HTS) ring mode.
+ * For more information please refer to <rte_ring_hts.h>.
+ */
+
+static __rte_always_inline void
+__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
+ uint32_t enqueue)
+{
+ union rte_ring_ht_pos p;
+
+ if (enqueue)
+ rte_smp_wmb();
+ else
+ rte_smp_rmb();
+
+ p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
+
+ p.pos.head = p.pos.tail + num;
+ p.pos.tail = p.pos.head;
+
+ rte_atomic64_set((rte_atomic64_t *)(uintptr_t)&ht->ht.raw, p.raw);
+}
+
+/**
+ * @internal waits till tail will become equal to head.
+ * Means no writer/reader is active for that ring.
+ * Suppose to work as serialization point.
+ */
+static __rte_always_inline void
+__rte_ring_hts_head_wait(const struct rte_ring_hts_headtail *ht,
+ union rte_ring_ht_pos *p)
+{
+ p->raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->ht.raw);
+
+ while (p->pos.head != p->pos.tail) {
+ rte_pause();
+ p->raw = rte_atomic64_read((rte_atomic64_t *)
+ (uintptr_t)&ht->ht.raw);
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_prod_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_prod, &op);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - op.pos.head;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ break;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (rte_atomic64_cmpset(&r->hts_prod.ht.raw,
+ op.raw, np.raw) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_cons_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_cons, &op);
+
+ /* add rmb barrier to avoid load/load reorder in weak
+ * memory model. It is noop on x86
+ */
+ rte_smp_rmb();
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - op.pos.head;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ break;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (rte_atomic64_cmpset(&r->hts_cons.ht.raw,
+ op.raw, np.raw) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+#endif /* _RTE_RING_HTS_GENERIC_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v3 6/9] test/ring: add contention stress test for HTS ring
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
` (4 preceding siblings ...)
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 5/9] ring: introduce HTS ring mode Konstantin Ananyev
@ 2020-04-03 17:42 ` Konstantin Ananyev
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 7/9] ring: introduce peek style API Konstantin Ananyev
` (3 subsequent siblings)
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-03 17:42 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce new test case to test HTS ring mode under contention.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_hts_stress.c | 32 ++++++++++++++++++++++++++++++++
app/test/test_ring_stress.c | 3 +++
app/test/test_ring_stress.h | 1 +
5 files changed, 38 insertions(+)
create mode 100644 app/test/test_ring_hts_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index 3e15f3791..72f64e30e 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -78,6 +78,7 @@ SRCS-y += test_rand_perf.c
SRCS-y += test_ring.c
SRCS-y += test_ring_mpmc_stress.c
+SRCS-y += test_ring_hts_stress.c
SRCS-y += test_ring_perf.c
SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
diff --git a/app/test/meson.build b/app/test/meson.build
index bb67a49f0..d6504a08a 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -101,6 +101,7 @@ test_sources = files('commands.c',
'test_rib6.c',
'test_ring.c',
'test_ring_mpmc_stress.c',
+ 'test_ring_hts_stress.c',
'test_ring_perf.c',
'test_ring_rts_stress.c',
'test_ring_stress.c',
diff --git a/app/test/test_ring_hts_stress.c b/app/test/test_ring_hts_stress.c
new file mode 100644
index 000000000..edc9175cb
--- /dev/null
+++ b/app/test/test_ring_hts_stress.c
@@ -0,0 +1,32 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ return rte_ring_mc_hts_dequeue_bulk(r, obj, n, avail);
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ return rte_ring_mp_hts_enqueue_bulk(r, obj, n, free);
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_HTS_ENQ | RING_F_MC_HTS_DEQ);
+}
+
+const struct test test_ring_hts_stress = {
+ .name = "MT_HTS",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
index eab395e30..29a1368d7 100644
--- a/app/test/test_ring_stress.c
+++ b/app/test/test_ring_stress.c
@@ -43,6 +43,9 @@ test_ring_stress(void)
n += test_ring_rts_stress.nb_case;
k += run_test(&test_ring_rts_stress);
+ n += test_ring_hts_stress.nb_case;
+ k += run_test(&test_ring_hts_stress);
+
printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
n, k, n - k);
return (k != n);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
index 32aae2072..9a87c7f7b 100644
--- a/app/test/test_ring_stress.h
+++ b/app/test/test_ring_stress.h
@@ -34,3 +34,4 @@ struct test {
extern const struct test test_ring_mpmc_stress;
extern const struct test test_ring_rts_stress;
+extern const struct test test_ring_hts_stress;
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v3 7/9] ring: introduce peek style API
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
` (5 preceding siblings ...)
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 6/9] test/ring: add contention stress test for HTS ring Konstantin Ananyev
@ 2020-04-03 17:42 ` Konstantin Ananyev
2020-04-14 3:45 ` Honnappa Nagarahalli
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 8/9] test/ring: add stress test for MT peek API Konstantin Ananyev
` (2 subsequent siblings)
9 siblings, 1 reply; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-03 17:42 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
For rings with producer/consumer in RTE_RING_SYNC_ST, RTE_RING_SYNC_MT_HTS
mode, provide an ability to split enqueue/dequeue operation
into two phases:
- enqueue/dequeue start
- enqueue/dequeue finish
That allows user to inspect objects in the ring without removing
them from it (aka MT safe peek).
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 1 +
lib/librte_ring/meson.build | 1 +
lib/librte_ring/rte_ring_c11_mem.h | 44 +++
lib/librte_ring/rte_ring_elem.h | 4 +
lib/librte_ring/rte_ring_generic.h | 48 ++++
lib/librte_ring/rte_ring_hts_generic.h | 47 ++-
lib/librte_ring/rte_ring_peek.h | 379 +++++++++++++++++++++++++
7 files changed, 519 insertions(+), 5 deletions(-)
create mode 100644 lib/librte_ring/rte_ring_peek.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 6fe500f0d..5f8662737 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -22,6 +22,7 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_hts.h \
rte_ring_hts_elem.h \
rte_ring_hts_generic.h \
+ rte_ring_peek.h \
rte_ring_rts.h \
rte_ring_rts_elem.h \
rte_ring_rts_generic.h
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index 8e86e037a..f5f84dc6e 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -9,6 +9,7 @@ headers = files('rte_ring.h',
'rte_ring_hts.h',
'rte_ring_hts_elem.h',
'rte_ring_hts_generic.h',
+ 'rte_ring_peek.h',
'rte_ring_rts.h',
'rte_ring_rts_elem.h',
'rte_ring_rts_generic.h')
diff --git a/lib/librte_ring/rte_ring_c11_mem.h b/lib/librte_ring/rte_ring_c11_mem.h
index 0fb73a337..bb3096721 100644
--- a/lib/librte_ring/rte_ring_c11_mem.h
+++ b/lib/librte_ring/rte_ring_c11_mem.h
@@ -10,6 +10,50 @@
#ifndef _RTE_RING_C11_MEM_H_
#define _RTE_RING_C11_MEM_H_
+/**
+ * @internal get current tail value.
+ * This function should be used only for single thread producer/consumer.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_st_get_tail(struct rte_ring_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t h, n, t;
+
+ h = ht->head;
+ t = ht->tail;
+ n = h - t;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = h;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail.
+ * This function should be used only for single thread producer/consumer.
+ * Should be used only in conjunction with __rte_ring_st_get_tail.
+ */
+static __rte_always_inline void
+__rte_ring_st_set_head_tail(struct rte_ring_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
+{
+ uint32_t pos;
+
+ RTE_SET_USED(enqueue);
+
+ pos = tail + num;
+ ht->head = pos;
+ __atomic_store_n(&ht->tail, pos, __ATOMIC_RELEASE);
+}
+
static __rte_always_inline void
update_tail(struct rte_ring_headtail *ht, uint32_t old_val, uint32_t new_val,
uint32_t single, uint32_t enqueue)
diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
index 010a564c1..5bf7c1c1b 100644
--- a/lib/librte_ring/rte_ring_elem.h
+++ b/lib/librte_ring/rte_ring_elem.h
@@ -1083,6 +1083,10 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
return 0;
}
+#ifdef ALLOW_EXPERIMENTAL_API
+#include <rte_ring_peek.h>
+#endif
+
#ifdef __cplusplus
}
#endif
diff --git a/lib/librte_ring/rte_ring_generic.h b/lib/librte_ring/rte_ring_generic.h
index 953cdbbd5..9f5fdf13b 100644
--- a/lib/librte_ring/rte_ring_generic.h
+++ b/lib/librte_ring/rte_ring_generic.h
@@ -10,6 +10,54 @@
#ifndef _RTE_RING_GENERIC_H_
#define _RTE_RING_GENERIC_H_
+/**
+ * @internal get current tail value.
+ * This function should be used only for single thread producer/consumer.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_st_get_tail(struct rte_ring_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t h, n, t;
+
+ h = ht->head;
+ t = ht->tail;
+ n = h - t;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = h;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail.
+ * This function should be used only for single thread producer/consumer.
+ * Should be used only in conjunction with __rte_ring_st_get_tail.
+ */
+static __rte_always_inline void
+__rte_ring_st_set_head_tail(struct rte_ring_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
+{
+ uint32_t pos;
+
+ pos = tail + num;
+
+ if (enqueue)
+ rte_smp_wmb();
+ else
+ rte_smp_rmb();
+
+ ht->head = pos;
+ ht->tail = pos;
+}
+
static __rte_always_inline void
update_tail(struct rte_ring_headtail *ht, uint32_t old_val, uint32_t new_val,
uint32_t single, uint32_t enqueue)
diff --git a/lib/librte_ring/rte_ring_hts_generic.h b/lib/librte_ring/rte_ring_hts_generic.h
index da08f1d94..8e699c006 100644
--- a/lib/librte_ring/rte_ring_hts_generic.h
+++ b/lib/librte_ring/rte_ring_hts_generic.h
@@ -18,9 +18,38 @@
* For more information please refer to <rte_ring_hts.h>.
*/
+/**
+ * @internal get current tail value.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_hts_get_tail(struct rte_ring_hts_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t n;
+ union rte_ring_ht_pos p;
+
+ p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
+ n = p.pos.head - p.pos.tail;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = p.pos.tail;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail as one atomic 64 bit operation.
+ * Should be used only in conjunction with __rte_ring_hts_get_tail.
+ */
static __rte_always_inline void
-__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
- uint32_t enqueue)
+__rte_ring_hts_set_head_tail(struct rte_ring_hts_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
{
union rte_ring_ht_pos p;
@@ -29,14 +58,22 @@ __rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
else
rte_smp_rmb();
- p.raw = rte_atomic64_read((rte_atomic64_t *)(uintptr_t)&ht->ht.raw);
-
- p.pos.head = p.pos.tail + num;
+ p.pos.head = tail + num;
p.pos.tail = p.pos.head;
rte_atomic64_set((rte_atomic64_t *)(uintptr_t)&ht->ht.raw, p.raw);
}
+static __rte_always_inline void
+__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
+ uint32_t enqueue)
+{
+ uint32_t tail;
+
+ num = __rte_ring_hts_get_tail(ht, &tail, num);
+ __rte_ring_hts_set_head_tail(ht, tail, num, enqueue);
+}
+
/**
* @internal waits till tail will become equal to head.
* Means no writer/reader is active for that ring.
diff --git a/lib/librte_ring/rte_ring_peek.h b/lib/librte_ring/rte_ring_peek.h
new file mode 100644
index 000000000..baefd2f7b
--- /dev/null
+++ b/lib/librte_ring/rte_ring_peek.h
@@ -0,0 +1,379 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_PEEK_H_
+#define _RTE_RING_PEEK_H_
+
+/**
+ * @file
+ * @b EXPERIMENTAL: this API may change without prior notice
+ * It is not recommended to include this file directly.
+ * Please include <rte_ring_elem.h> instead.
+ *
+ * Ring Peek AP
+ * Introduction of rte_ring with serialized producer/consumer (HTS sync mode)
+ * makes possible to split public enqueue/dequeue API into two phases:
+ * - enqueue/dequeue start
+ * - enqueue/dequeue finish
+ * That allows user to inspect objects in the ring without removing them
+ * from it (aka MT safe peek).
+ * Note that right now this new API is avaialble only for two sync modes:
+ * 1) Single Producer/Single Consumer (RTE_RING_SYNC_ST)
+ * 2) Serialized Producer/Serialized Consumer (RTE_RING_SYNC_MT_HTS).
+ * It is a user responsibility to create/init ring with appropriate sync
+ * modes selected.
+ * As an example:
+ * // read 1 elem from the ring:
+ * n = rte_ring_hts_dequeue_bulk_start(ring, &obj, 1, NULL);
+ * if (n != 0) {
+ * //examine object
+ * if (object_examine(obj) == KEEP)
+ * //decided to keep it in the ring.
+ * rte_ring_hts_dequeue_finish(ring, 0);
+ * else
+ * //decided to remove it from the ring.
+ * rte_ring_hts_dequeue_finish(ring, n);
+ * }
+ * Note that between _start_ and _finish_ the ring is sort of locked -
+ * none other thread can proceed with enqueue(/dequeue) operation till
+ * _finish_ will complete.
+ */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/**
+ * @internal This function moves prod head value.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_enqueue_start(struct rte_ring *r, uint32_t n,
+ enum rte_ring_queue_behavior behavior, uint32_t *free_space)
+{
+ uint32_t free, head, next;
+
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_move_prod_head(r, RTE_RING_SYNC_ST, n,
+ behavior, &head, &next, &free);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ n = __rte_ring_hts_move_prod_head(r, n, behavior,
+ &head, &free);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ n = 0;
+ }
+
+ if (free_space != NULL)
+ *free_space = free - n;
+ return n;
+}
+
+/**
+ * Start to enqueue several objects on the ring.
+ * Note that no actual objects are put in the queue by this function,
+ * it just reserves for user such ability.
+ * User has to call appropriate enqueue_finish() to copy objects into the
+ * queue and complete given enqueue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * The number of objects that can be enqueued, either 0 or n
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_enqueue_bulk_start(struct rte_ring *r, unsigned int n,
+ unsigned int *free_space)
+{
+ return __rte_ring_do_enqueue_start(r, n, RTE_RING_QUEUE_FIXED,
+ free_space);
+}
+
+/**
+ * Start to enqueue several objects on the ring.
+ * Note that no actual objects are put in the queue by this function,
+ * it just reserves for user such ability.
+ * User has to call appropriate enqueue_finish() to copy objects into the
+ * queue and complete given enqueue operation.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param n
+ * The number of objects to add in the ring from the obj_table.
+ * @param free_space
+ * if non-NULL, returns the amount of space in the ring after the
+ * enqueue operation has finished.
+ * @return
+ * Actual number of objects that can be enqueued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_enqueue_burst_start(struct rte_ring *r, unsigned int n,
+ unsigned int *free_space)
+{
+ return __rte_ring_do_enqueue_start(r, n, RTE_RING_QUEUE_VARIABLE,
+ free_space);
+}
+
+/**
+ * Complete to enqueue several objects on the ring.
+ * Note that number of objects to enqueue should not exceed previous
+ * enqueue_start return value.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of objects.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to add to the ring from the obj_table.
+ */
+__rte_experimental
+static __rte_always_inline void
+rte_ring_enqueue_elem_finish(struct rte_ring *r, void * const *obj_table,
+ unsigned int esize, unsigned int n)
+{
+ uint32_t tail;
+
+ switch (r->prod.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_st_get_tail(&r->prod, &tail, n);
+ if (n != 0)
+ __rte_ring_enqueue_elems(r, tail, obj_table, esize, n);
+ __rte_ring_st_set_head_tail(&r->prod, tail, n, 1);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ n = __rte_ring_hts_get_tail(&r->hts_prod, &tail, n);
+ if (n != 0)
+ __rte_ring_enqueue_elems(r, tail, obj_table, esize, n);
+ __rte_ring_hts_set_head_tail(&r->hts_prod, tail, n, 1);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ }
+}
+
+/**
+ * Complete to enqueue several objects on the ring.
+ * Note that number of objects to enqueue should not exceed previous
+ * enqueue_start return value.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of objects.
+ * @param n
+ * The number of objects to add to the ring from the obj_table.
+ */
+__rte_experimental
+static __rte_always_inline void
+rte_ring_enqueue_finish(struct rte_ring *r, void * const *obj_table,
+ unsigned int n)
+{
+ rte_ring_enqueue_elem_finish(r, obj_table, sizeof(uintptr_t), n);
+}
+
+/**
+ * @internal This function moves cons head value and copies up to *n*
+ * objects from the ring to the user provided obj_table.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_do_dequeue_start(struct rte_ring *r, void **obj_table,
+ uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
+ uint32_t *available)
+{
+ uint32_t avail, head, next;
+
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_move_cons_head(r, RTE_RING_SYNC_ST, n,
+ behavior, &head, &next, &avail);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ n = __rte_ring_hts_move_cons_head(r, n, behavior,
+ &head, &avail);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ n = 0;
+ }
+
+ if (n != 0)
+ __rte_ring_dequeue_elems(r, head, obj_table, esize, n);
+
+ if (available != NULL)
+ *available = avail - n;
+ return n;
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The number of objects dequeued, either 0 or n.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_bulk_elem_start(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_dequeue_start(r, obj_table, esize, n,
+ RTE_RING_QUEUE_FIXED, available);
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * Actual number of objects dequeued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_bulk_start(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return rte_ring_dequeue_bulk_elem_start(r, obj_table, sizeof(uintptr_t),
+ n, available);
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param esize
+ * The size of ring element, in bytes. It must be a multiple of 4.
+ * This must be the same value used while creating the ring. Otherwise
+ * the results are undefined.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The actual number of objects dequeued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_burst_elem_start(struct rte_ring *r, void **obj_table,
+ unsigned int esize, unsigned int n, unsigned int *available)
+{
+ return __rte_ring_do_dequeue_start(r, obj_table, esize, n,
+ RTE_RING_QUEUE_VARIABLE, available);
+}
+
+/**
+ * Start to dequeue several objects from the ring.
+ * Note that user has to call appropriate dequeue_finish()
+ * to complete given dequeue operation and actually remove objects the ring.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param obj_table
+ * A pointer to a table of void * pointers (objects) that will be filled.
+ * @param n
+ * The number of objects to dequeue from the ring to the obj_table.
+ * @param available
+ * If non-NULL, returns the number of remaining ring entries after the
+ * dequeue has finished.
+ * @return
+ * The actual number of objects dequeued.
+ */
+__rte_experimental
+static __rte_always_inline unsigned int
+rte_ring_dequeue_burst_start(struct rte_ring *r, void **obj_table,
+ unsigned int n, unsigned int *available)
+{
+ return rte_ring_dequeue_burst_elem_start(r, obj_table,
+ sizeof(uintptr_t), n, available);
+}
+
+/**
+ * Complete to dequeue several objects from the ring.
+ * Note that number of objects to dequeue should not exceed previous
+ * dequeue_start return value.
+ *
+ * @param r
+ * A pointer to the ring structure.
+ * @param n
+ * The number of objects to add to the ring from the obj_table.
+ */
+__rte_experimental
+static __rte_always_inline void
+rte_ring_dequeue_finish(struct rte_ring *r, unsigned int n)
+{
+ uint32_t tail;
+
+ switch (r->cons.sync_type) {
+ case RTE_RING_SYNC_ST:
+ n = __rte_ring_st_get_tail(&r->cons, &tail, n);
+ __rte_ring_st_set_head_tail(&r->cons, tail, n, 0);
+ break;
+ case RTE_RING_SYNC_MT_HTS:
+ __rte_ring_hts_update_tail(&r->hts_cons, n, 0);
+ break;
+ default:
+ /* unsupported mode, shouldn't be here */
+ RTE_ASSERT(0);
+ }
+}
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_RING_PEEK_H_ */
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v3 8/9] test/ring: add stress test for MT peek API
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
` (6 preceding siblings ...)
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 7/9] ring: introduce peek style API Konstantin Ananyev
@ 2020-04-03 17:42 ` Konstantin Ananyev
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 9/9] ring: add C11 memory model for new sync modes Konstantin Ananyev
2020-04-17 13:36 ` [dpdk-dev] [PATCH v4 0/9] New sync modes for ring Konstantin Ananyev
9 siblings, 0 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-03 17:42 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Introduce new test case to test MT peek API.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
app/test/Makefile | 1 +
app/test/meson.build | 1 +
app/test/test_ring_peek_stress.c | 43 ++++++++++++++++++++++++++++++++
app/test/test_ring_stress.c | 3 +++
app/test/test_ring_stress.h | 1 +
5 files changed, 49 insertions(+)
create mode 100644 app/test/test_ring_peek_stress.c
diff --git a/app/test/Makefile b/app/test/Makefile
index 72f64e30e..f4c987a98 100644
--- a/app/test/Makefile
+++ b/app/test/Makefile
@@ -80,6 +80,7 @@ SRCS-y += test_ring.c
SRCS-y += test_ring_mpmc_stress.c
SRCS-y += test_ring_hts_stress.c
SRCS-y += test_ring_perf.c
+SRCS-y += test_ring_peek_stress.c
SRCS-y += test_ring_rts_stress.c
SRCS-y += test_ring_stress.c
SRCS-y += test_pmd_perf.c
diff --git a/app/test/meson.build b/app/test/meson.build
index d6504a08a..cfcf5b81d 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -102,6 +102,7 @@ test_sources = files('commands.c',
'test_ring.c',
'test_ring_mpmc_stress.c',
'test_ring_hts_stress.c',
+ 'test_ring_peek_stress.c',
'test_ring_perf.c',
'test_ring_rts_stress.c',
'test_ring_stress.c',
diff --git a/app/test/test_ring_peek_stress.c b/app/test/test_ring_peek_stress.c
new file mode 100644
index 000000000..cfc82d728
--- /dev/null
+++ b/app/test/test_ring_peek_stress.c
@@ -0,0 +1,43 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2020 Intel Corporation
+ */
+
+#include "test_ring_stress_impl.h"
+#include <rte_ring_elem.h>
+
+static inline uint32_t
+_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
+ uint32_t *avail)
+{
+ uint32_t m;
+
+ m = rte_ring_dequeue_bulk_start(r, obj, n, avail);
+ n = (m == n) ? n : 0;
+ rte_ring_dequeue_finish(r, n);
+ return n;
+}
+
+static inline uint32_t
+_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
+ uint32_t *free)
+{
+ uint32_t m;
+
+ m = rte_ring_enqueue_bulk_start(r, n, free);
+ n = (m == n) ? n : 0;
+ rte_ring_enqueue_finish(r, obj, n);
+ return n;
+}
+
+static int
+_st_ring_init(struct rte_ring *r, const char *name, uint32_t num)
+{
+ return rte_ring_init(r, name, num,
+ RING_F_MP_HTS_ENQ | RING_F_MC_HTS_DEQ);
+}
+
+const struct test test_ring_peek_stress = {
+ .name = "MT_PEEK",
+ .nb_case = RTE_DIM(tests),
+ .cases = tests,
+};
diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c
index 29a1368d7..853fcc190 100644
--- a/app/test/test_ring_stress.c
+++ b/app/test/test_ring_stress.c
@@ -46,6 +46,9 @@ test_ring_stress(void)
n += test_ring_hts_stress.nb_case;
k += run_test(&test_ring_hts_stress);
+ n += test_ring_peek_stress.nb_case;
+ k += run_test(&test_ring_peek_stress);
+
printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
n, k, n - k);
return (k != n);
diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h
index 9a87c7f7b..60953ce47 100644
--- a/app/test/test_ring_stress.h
+++ b/app/test/test_ring_stress.h
@@ -35,3 +35,4 @@ struct test {
extern const struct test test_ring_mpmc_stress;
extern const struct test test_ring_rts_stress;
extern const struct test test_ring_hts_stress;
+extern const struct test test_ring_peek_stress;
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] [PATCH v3 9/9] ring: add C11 memory model for new sync modes
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 0/9] New sync modes for ring Konstantin Ananyev
` (7 preceding siblings ...)
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 8/9] test/ring: add stress test for MT peek API Konstantin Ananyev
@ 2020-04-03 17:42 ` Konstantin Ananyev
2020-04-04 14:16 ` [dpdk-dev] 回复:[PATCH " 周介龙
2020-04-14 4:28 ` [dpdk-dev] [PATCH " Honnappa Nagarahalli
2020-04-17 13:36 ` [dpdk-dev] [PATCH v4 0/9] New sync modes for ring Konstantin Ananyev
9 siblings, 2 replies; 146+ messages in thread
From: Konstantin Ananyev @ 2020-04-03 17:42 UTC (permalink / raw)
To: dev; +Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Konstantin Ananyev
Add C11 atomics based implementation for RTS and HTS
head/tail update primitivies.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 4 +-
lib/librte_ring/meson.build | 2 +
lib/librte_ring/rte_ring_hts.h | 4 +
lib/librte_ring/rte_ring_hts_c11_mem.h | 222 +++++++++++++++++++++++++
lib/librte_ring/rte_ring_hts_elem.h | 4 +
lib/librte_ring/rte_ring_rts.h | 4 +
lib/librte_ring/rte_ring_rts_c11_mem.h | 198 ++++++++++++++++++++++
lib/librte_ring/rte_ring_rts_elem.h | 4 +
8 files changed, 441 insertions(+), 1 deletion(-)
create mode 100644 lib/librte_ring/rte_ring_hts_c11_mem.h
create mode 100644 lib/librte_ring/rte_ring_rts_c11_mem.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 5f8662737..927d105bf 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -22,9 +22,11 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_hts.h \
rte_ring_hts_elem.h \
rte_ring_hts_generic.h \
+ rte_ring_hts_c11_mem.h \
rte_ring_peek.h \
rte_ring_rts.h \
rte_ring_rts_elem.h \
- rte_ring_rts_generic.h
+ rte_ring_rts_generic.h \
+ rte_ring_rts_c11_mem.h
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index f5f84dc6e..f2e37a8e4 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -7,10 +7,12 @@ headers = files('rte_ring.h',
'rte_ring_c11_mem.h',
'rte_ring_generic.h',
'rte_ring_hts.h',
+ 'rte_ring_hts_c11_mem.h',
'rte_ring_hts_elem.h',
'rte_ring_hts_generic.h',
'rte_ring_peek.h',
'rte_ring_rts.h',
+ 'rte_ring_rts_c11_mem.h',
'rte_ring_rts_elem.h',
'rte_ring_rts_generic.h')
diff --git a/lib/librte_ring/rte_ring_hts.h b/lib/librte_ring/rte_ring_hts.h
index 062d7be6c..ddaa47ff1 100644
--- a/lib/librte_ring/rte_ring_hts.h
+++ b/lib/librte_ring/rte_ring_hts.h
@@ -29,7 +29,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_hts_c11_mem.h>
+#else
#include <rte_ring_hts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the HTS ring.
diff --git a/lib/librte_ring/rte_ring_hts_c11_mem.h b/lib/librte_ring/rte_ring_hts_c11_mem.h
new file mode 100644
index 000000000..0218d0e7d
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts_c11_mem.h
@@ -0,0 +1,222 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2020 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_C11_MEM_H_
+#define _RTE_RING_HTS_C11_MEM_H_
+
+/**
+ * @file rte_ring_hts_c11_mem.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for head/tail sync (HTS) ring mode.
+ * For more information please refer to <rte_ring_hts.h>.
+ */
+
+/**
+ * @internal get current tail value.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_hts_get_tail(struct rte_ring_hts_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t n;
+ union rte_ring_ht_pos p;
+
+ p.raw = __atomic_load_n(&ht->ht.raw, __ATOMIC_RELAXED);
+ n = p.pos.head - p.pos.tail;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = p.pos.tail;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail as one atomic 64 bit operation.
+ * Should be used only in conjunction with __rte_ring_hts_get_tail.
+ */
+static __rte_always_inline void
+__rte_ring_hts_set_head_tail(struct rte_ring_hts_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
+{
+ union rte_ring_ht_pos p;
+
+ RTE_SET_USED(enqueue);
+
+ p.pos.head = tail + num;
+ p.pos.tail = p.pos.head;
+
+ __atomic_store_n(&ht->ht.raw, p.raw, __ATOMIC_RELEASE);
+}
+
+static __rte_always_inline void
+__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
+ uint32_t enqueue)
+{
+ uint32_t tail;
+
+ num = __rte_ring_hts_get_tail(ht, &tail, num);
+ __rte_ring_hts_set_head_tail(ht, tail, num, enqueue);
+}
+
+/**
+ * @internal waits till tail will become equal to head.
+ * Means no writer/reader is active for that ring.
+ * Suppose to work as serialization point.
+ */
+static __rte_always_inline void
+__rte_ring_hts_head_wait(const struct rte_ring_hts_headtail *ht,
+ union rte_ring_ht_pos *p)
+{
+ p->raw = __atomic_load_n(&ht->ht.raw, __ATOMIC_ACQUIRE);
+
+ while (p->pos.head != p->pos.tail) {
+ rte_pause();
+ p->raw = __atomic_load_n(&ht->ht.raw, __ATOMIC_ACQUIRE);
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_prod_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* wait for tail to be equal to head, , acquire point */
+ __rte_ring_hts_head_wait(&r->hts_prod, &op);
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - op.pos.head;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ break;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (__atomic_compare_exchange_n(&r->hts_prod.ht.raw,
+ &op.raw, np.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_cons_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_cons, &op);
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - op.pos.head;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ break;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (__atomic_compare_exchange_n(&r->hts_cons.ht.raw,
+ &op.raw, np.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+#endif /* _RTE_RING_HTS_C11_MEM_H_ */
diff --git a/lib/librte_ring/rte_ring_hts_elem.h b/lib/librte_ring/rte_ring_hts_elem.h
index 34f0d121d..1e9a49c7a 100644
--- a/lib/librte_ring/rte_ring_hts_elem.h
+++ b/lib/librte_ring/rte_ring_hts_elem.h
@@ -24,7 +24,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_hts_c11_mem.h>
+#else
#include <rte_ring_hts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the HTS ring.
diff --git a/lib/librte_ring/rte_ring_rts.h b/lib/librte_ring/rte_ring_rts.h
index 18404fe48..28b2d25f5 100644
--- a/lib/librte_ring/rte_ring_rts.h
+++ b/lib/librte_ring/rte_ring_rts.h
@@ -55,7 +55,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_rts_c11_mem.h>
+#else
#include <rte_ring_rts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the RTS ring.
diff --git a/lib/librte_ring/rte_ring_rts_c11_mem.h b/lib/librte_ring/rte_ring_rts_c11_mem.h
new file mode 100644
index 000000000..b72901497
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts_c11_mem.h
@@ -0,0 +1,198 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_C11_MEM_H_
+#define _RTE_RING_RTS_C11_MEM_H_
+
+/**
+ * @file rte_ring_rts_c11_mem.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for Relaxed Tail Sync (RTS) ring mode.
+ * For more information please refer to <rte_ring_rts.h>.
+ */
+
+/**
+ * @internal This function updates tail values.
+ */
+static __rte_always_inline void
+__rte_ring_rts_update_tail(struct rte_ring_rts_headtail *ht)
+{
+ union rte_ring_ht_poscnt h, ot, nt;
+
+ /*
+ * If there are other enqueues/dequeues in progress that
+ * might preceded us, then don't update tail with new value.
+ */
+
+ ot.raw = __atomic_load_n(&ht->tail.raw, __ATOMIC_ACQUIRE);
+
+ do {
+ /* on 32-bit systems we have to do atomic read here */
+ h.raw = __atomic_load_n(&ht->head.raw, __ATOMIC_RELAXED);
+
+ nt.raw = ot.raw;
+ if (++nt.val.cnt == h.val.cnt)
+ nt.val.pos = h.val.pos;
+
+ } while (__atomic_compare_exchange_n(&ht->tail.raw, &ot.raw, nt.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_ACQUIRE) == 0);
+}
+
+/**
+ * @internal This function waits till head/tail distance wouldn't
+ * exceed pre-defined max value.
+ */
+static __rte_always_inline void
+__rte_ring_rts_head_wait(const struct rte_ring_rts_headtail *ht,
+ union rte_ring_ht_poscnt *h)
+{
+ uint32_t max;
+
+ max = ht->htd_max;
+ h->raw = __atomic_load_n(&ht->head.raw, __ATOMIC_ACQUIRE);
+
+ while (h->val.pos - ht->tail.val.pos > max) {
+ rte_pause();
+ h->raw = __atomic_load_n(&ht->head.raw, __ATOMIC_ACQUIRE);
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue.
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_rts_move_prod_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* read prod head (may spin on prod tail, acquire point) */
+ __rte_ring_rts_head_wait(&r->rts_prod, &oh);
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - oh.val.pos;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ break;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (__atomic_compare_exchange_n(&r->rts_prod.head.raw,
+ &oh.raw, nh.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_rts_move_cons_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* read cons head (may spin on cons tail, acquire point) */
+ __rte_ring_rts_head_wait(&r->rts_cons, &oh);
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - oh.val.pos;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ break;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (__atomic_compare_exchange_n(&r->rts_cons.head.raw,
+ &oh.raw, nh.raw,
+ 1, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+#endif /* _RTE_RING_RTS_C11_MEM_H_ */
diff --git a/lib/librte_ring/rte_ring_rts_elem.h b/lib/librte_ring/rte_ring_rts_elem.h
index 71a331b23..23d8aeec7 100644
--- a/lib/librte_ring/rte_ring_rts_elem.h
+++ b/lib/librte_ring/rte_ring_rts_elem.h
@@ -24,7 +24,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_rts_c11_mem.h>
+#else
#include <rte_ring_rts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the RTS ring.
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* [dpdk-dev] 回复:[PATCH v3 9/9] ring: add C11 memory model for new sync modes
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 9/9] ring: add C11 memory model for new sync modes Konstantin Ananyev
@ 2020-04-04 14:16 ` 周介龙
2020-04-14 4:28 ` [dpdk-dev] [PATCH " Honnappa Nagarahalli
1 sibling, 0 replies; 146+ messages in thread
From: 周介龙 @ 2020-04-04 14:16 UTC (permalink / raw)
To: dev, Konstantin Ananyev
Cc: honnappa.nagarahalli, david.marchand, Konstantin Ananyev
Hi,
This patchset really helps my case a lot.
We have a case which would have several hundreds of threads
get/put concurrently on single mempool(ring-based) without cache.
Even on a simplified test case in which 32 threads putting into mempool
concurrently for 1024 times respectively, it works extremely slow and
would be finished after tens of seconds. With new sync modes introduced
by this patchset, the case could finish in 2816979 cycles(RTS) and
2279615(HTS) cycles.
It would be great if we could have simple APIs to enable these sync modes
for structures based on rte_ring like mempool.
------------------------------------------------------------------
发件人:Konstantin Ananyev <konstantin.ananyev@intel.com>
发送时间:2020年4月4日(星期六) 01:43
收件人:dev <dev@dpdk.org>
抄 送:honnappa.nagarahalli <honnappa.nagarahalli@arm.com>; david.marchand <david.marchand@redhat.com>; 周介龙(卓腾) <jielong.zjl@antfin.com>; Konstantin Ananyev <konstantin.ananyev@intel.com>
主 题:[PATCH v3 9/9] ring: add C11 memory model for new sync modes
Add C11 atomics based implementation for RTS and HTS
head/tail update primitivies.
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
lib/librte_ring/Makefile | 4 +-
lib/librte_ring/meson.build | 2 +
lib/librte_ring/rte_ring_hts.h | 4 +
lib/librte_ring/rte_ring_hts_c11_mem.h | 222 +++++++++++++++++++++++++
lib/librte_ring/rte_ring_hts_elem.h | 4 +
lib/librte_ring/rte_ring_rts.h | 4 +
lib/librte_ring/rte_ring_rts_c11_mem.h | 198 ++++++++++++++++++++++
lib/librte_ring/rte_ring_rts_elem.h | 4 +
8 files changed, 441 insertions(+), 1 deletion(-)
create mode 100644 lib/librte_ring/rte_ring_hts_c11_mem.h
create mode 100644 lib/librte_ring/rte_ring_rts_c11_mem.h
diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile
index 5f8662737..927d105bf 100644
--- a/lib/librte_ring/Makefile
+++ b/lib/librte_ring/Makefile
@@ -22,9 +22,11 @@ SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
rte_ring_hts.h \
rte_ring_hts_elem.h \
rte_ring_hts_generic.h \
+ rte_ring_hts_c11_mem.h \
rte_ring_peek.h \
rte_ring_rts.h \
rte_ring_rts_elem.h \
- rte_ring_rts_generic.h
+ rte_ring_rts_generic.h \
+ rte_ring_rts_c11_mem.h
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
index f5f84dc6e..f2e37a8e4 100644
--- a/lib/librte_ring/meson.build
+++ b/lib/librte_ring/meson.build
@@ -7,10 +7,12 @@ headers = files('rte_ring.h',
'rte_ring_c11_mem.h',
'rte_ring_generic.h',
'rte_ring_hts.h',
+ 'rte_ring_hts_c11_mem.h',
'rte_ring_hts_elem.h',
'rte_ring_hts_generic.h',
'rte_ring_peek.h',
'rte_ring_rts.h',
+ 'rte_ring_rts_c11_mem.h',
'rte_ring_rts_elem.h',
'rte_ring_rts_generic.h')
diff --git a/lib/librte_ring/rte_ring_hts.h b/lib/librte_ring/rte_ring_hts.h
index 062d7be6c..ddaa47ff1 100644
--- a/lib/librte_ring/rte_ring_hts.h
+++ b/lib/librte_ring/rte_ring_hts.h
@@ -29,7 +29,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_hts_c11_mem.h>
+#else
#include <rte_ring_hts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the HTS ring.
diff --git a/lib/librte_ring/rte_ring_hts_c11_mem.h b/lib/librte_ring/rte_ring_hts_c11_mem.h
new file mode 100644
index 000000000..0218d0e7d
--- /dev/null
+++ b/lib/librte_ring/rte_ring_hts_c11_mem.h
@@ -0,0 +1,222 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2020 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_HTS_C11_MEM_H_
+#define _RTE_RING_HTS_C11_MEM_H_
+
+/**
+ * @file rte_ring_hts_c11_mem.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for head/tail sync (HTS) ring mode.
+ * For more information please refer to <rte_ring_hts.h>.
+ */
+
+/**
+ * @internal get current tail value.
+ * Check that user didn't request to move tail above the head.
+ * In that situation:
+ * - return zero, that will cause abort any pending changes and
+ * return head to its previous position.
+ * - throw an assert in debug mode.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_hts_get_tail(struct rte_ring_hts_headtail *ht, uint32_t *tail,
+ uint32_t num)
+{
+ uint32_t n;
+ union rte_ring_ht_pos p;
+
+ p.raw = __atomic_load_n(&ht->ht.raw, __ATOMIC_RELAXED);
+ n = p.pos.head - p.pos.tail;
+
+ RTE_ASSERT(n >= num);
+ num = (n >= num) ? num : 0;
+
+ *tail = p.pos.tail;
+ return num;
+}
+
+/**
+ * @internal set new values for head and tail as one atomic 64 bit operation.
+ * Should be used only in conjunction with __rte_ring_hts_get_tail.
+ */
+static __rte_always_inline void
+__rte_ring_hts_set_head_tail(struct rte_ring_hts_headtail *ht, uint32_t tail,
+ uint32_t num, uint32_t enqueue)
+{
+ union rte_ring_ht_pos p;
+
+ RTE_SET_USED(enqueue);
+
+ p.pos.head = tail + num;
+ p.pos.tail = p.pos.head;
+
+ __atomic_store_n(&ht->ht.raw, p.raw, __ATOMIC_RELEASE);
+}
+
+static __rte_always_inline void
+__rte_ring_hts_update_tail(struct rte_ring_hts_headtail *ht, uint32_t num,
+ uint32_t enqueue)
+{
+ uint32_t tail;
+
+ num = __rte_ring_hts_get_tail(ht, &tail, num);
+ __rte_ring_hts_set_head_tail(ht, tail, num, enqueue);
+}
+
+/**
+ * @internal waits till tail will become equal to head.
+ * Means no writer/reader is active for that ring.
+ * Suppose to work as serialization point.
+ */
+static __rte_always_inline void
+__rte_ring_hts_head_wait(const struct rte_ring_hts_headtail *ht,
+ union rte_ring_ht_pos *p)
+{
+ p->raw = __atomic_load_n(&ht->ht.raw, __ATOMIC_ACQUIRE);
+
+ while (p->pos.head != p->pos.tail) {
+ rte_pause();
+ p->raw = __atomic_load_n(&ht->ht.raw, __ATOMIC_ACQUIRE);
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_prod_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* wait for tail to be equal to head, , acquire point */
+ __rte_ring_hts_head_wait(&r->hts_prod, &op);
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - op.pos.head;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ break;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (__atomic_compare_exchange_n(&r->hts_prod.ht.raw,
+ &op.raw, np.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_hts_move_cons_head(struct rte_ring *r, unsigned int num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_pos np, op;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* wait for tail to be equal to head */
+ __rte_ring_hts_head_wait(&r->hts_cons, &op);
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - op.pos.head;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ break;
+
+ np.pos.tail = op.pos.tail;
+ np.pos.head = op.pos.head + n;
+
+ } while (__atomic_compare_exchange_n(&r->hts_cons.ht.raw,
+ &op.raw, np.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = op.pos.head;
+ return n;
+}
+
+#endif /* _RTE_RING_HTS_C11_MEM_H_ */
diff --git a/lib/librte_ring/rte_ring_hts_elem.h b/lib/librte_ring/rte_ring_hts_elem.h
index 34f0d121d..1e9a49c7a 100644
--- a/lib/librte_ring/rte_ring_hts_elem.h
+++ b/lib/librte_ring/rte_ring_hts_elem.h
@@ -24,7 +24,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_hts_c11_mem.h>
+#else
#include <rte_ring_hts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the HTS ring.
diff --git a/lib/librte_ring/rte_ring_rts.h b/lib/librte_ring/rte_ring_rts.h
index 18404fe48..28b2d25f5 100644
--- a/lib/librte_ring/rte_ring_rts.h
+++ b/lib/librte_ring/rte_ring_rts.h
@@ -55,7 +55,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_rts_c11_mem.h>
+#else
#include <rte_ring_rts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the RTS ring.
diff --git a/lib/librte_ring/rte_ring_rts_c11_mem.h b/lib/librte_ring/rte_ring_rts_c11_mem.h
new file mode 100644
index 000000000..b72901497
--- /dev/null
+++ b/lib/librte_ring/rte_ring_rts_c11_mem.h
@@ -0,0 +1,198 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2010-2017 Intel Corporation
+ * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
+ * All rights reserved.
+ * Derived from FreeBSD's bufring.h
+ * Used as BSD-3 Licensed with permission from Kip Macy.
+ */
+
+#ifndef _RTE_RING_RTS_C11_MEM_H_
+#define _RTE_RING_RTS_C11_MEM_H_
+
+/**
+ * @file rte_ring_rts_c11_mem.h
+ * It is not recommended to include this file directly,
+ * include <rte_ring.h> instead.
+ * Contains internal helper functions for Relaxed Tail Sync (RTS) ring mode.
+ * For more information please refer to <rte_ring_rts.h>.
+ */
+
+/**
+ * @internal This function updates tail values.
+ */
+static __rte_always_inline void
+__rte_ring_rts_update_tail(struct rte_ring_rts_headtail *ht)
+{
+ union rte_ring_ht_poscnt h, ot, nt;
+
+ /*
+ * If there are other enqueues/dequeues in progress that
+ * might preceded us, then don't update tail with new value.
+ */
+
+ ot.raw = __atomic_load_n(&ht->tail.raw, __ATOMIC_ACQUIRE);
+
+ do {
+ /* on 32-bit systems we have to do atomic read here */
+ h.raw = __atomic_load_n(&ht->head.raw, __ATOMIC_RELAXED);
+
+ nt.raw = ot.raw;
+ if (++nt.val.cnt == h.val.cnt)
+ nt.val.pos = h.val.pos;
+
+ } while (__atomic_compare_exchange_n(&ht->tail.raw, &ot.raw, nt.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_ACQUIRE) == 0);
+}
+
+/**
+ * @internal This function waits till head/tail distance wouldn't
+ * exceed pre-defined max value.
+ */
+static __rte_always_inline void
+__rte_ring_rts_head_wait(const struct rte_ring_rts_headtail *ht,
+ union rte_ring_ht_poscnt *h)
+{
+ uint32_t max;
+
+ max = ht->htd_max;
+ h->raw = __atomic_load_n(&ht->head.raw, __ATOMIC_ACQUIRE);
+
+ while (h->val.pos - ht->tail.val.pos > max) {
+ rte_pause();
+ h->raw = __atomic_load_n(&ht->head.raw, __ATOMIC_ACQUIRE);
+ }
+}
+
+/**
+ * @internal This function updates the producer head for enqueue.
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sp
+ * Indicates whether multi-producer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where enqueue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where enqueue finishes
+ * @param free_entries
+ * Returns the amount of free space in the ring BEFORE head was moved
+ * @return
+ * Actual number of objects enqueued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline uint32_t
+__rte_ring_rts_move_prod_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *free_entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ const uint32_t capacity = r->capacity;
+
+ do {
+ /* Reset n to the initial burst count */
+ n = num;
+
+ /* read prod head (may spin on prod tail, acquire point) */
+ __rte_ring_rts_head_wait(&r->rts_prod, &oh);
+
+ /*
+ * The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * *old_head > cons_tail). So 'free_entries' is always between 0
+ * and capacity (which is < size).
+ */
+ *free_entries = capacity + r->cons.tail - oh.val.pos;
+
+ /* check that we have enough room in ring */
+ if (unlikely(n > *free_entries))
+ n = (behavior == RTE_RING_QUEUE_FIXED) ?
+ 0 : *free_entries;
+
+ if (n == 0)
+ break;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (__atomic_compare_exchange_n(&r->rts_prod.head.raw,
+ &oh.raw, nh.raw,
+ 0, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+/**
+ * @internal This function updates the consumer head for dequeue
+ *
+ * @param r
+ * A pointer to the ring structure
+ * @param is_sc
+ * Indicates whether multi-consumer path is needed or not
+ * @param n
+ * The number of elements we will want to enqueue, i.e. how far should the
+ * head be moved
+ * @param behavior
+ * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
+ * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from ring
+ * @param old_head
+ * Returns head value as it was before the move, i.e. where dequeue starts
+ * @param new_head
+ * Returns the current/new head value i.e. where dequeue finishes
+ * @param entries
+ * Returns the number of entries in the ring BEFORE head was moved
+ * @return
+ * - Actual number of objects dequeued.
+ * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
+ */
+static __rte_always_inline unsigned int
+__rte_ring_rts_move_cons_head(struct rte_ring *r, uint32_t num,
+ enum rte_ring_queue_behavior behavior, uint32_t *old_head,
+ uint32_t *entries)
+{
+ uint32_t n;
+ union rte_ring_ht_poscnt nh, oh;
+
+ /* move cons.head atomically */
+ do {
+ /* Restore n as it may change every loop */
+ n = num;
+
+ /* read cons head (may spin on cons tail, acquire point) */
+ __rte_ring_rts_head_wait(&r->rts_cons, &oh);
+
+ /* The subtraction is done between two unsigned 32bits value
+ * (the result is always modulo 32 bits even if we have
+ * cons_head > prod_tail). So 'entries' is always between 0
+ * and size(ring)-1.
+ */
+ *entries = r->prod.tail - oh.val.pos;
+
+ /* Set the actual entries for dequeue */
+ if (n > *entries)
+ n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 : *entries;
+
+ if (unlikely(n == 0))
+ break;
+
+ nh.val.pos = oh.val.pos + n;
+ nh.val.cnt = oh.val.cnt + 1;
+
+ } while (__atomic_compare_exchange_n(&r->rts_cons.head.raw,
+ &oh.raw, nh.raw,
+ 1, __ATOMIC_RELEASE, __ATOMIC_RELAXED) == 0);
+
+ *old_head = oh.val.pos;
+ return n;
+}
+
+#endif /* _RTE_RING_RTS_C11_MEM_H_ */
diff --git a/lib/librte_ring/rte_ring_rts_elem.h b/lib/librte_ring/rte_ring_rts_elem.h
index 71a331b23..23d8aeec7 100644
--- a/lib/librte_ring/rte_ring_rts_elem.h
+++ b/lib/librte_ring/rte_ring_rts_elem.h
@@ -24,7 +24,11 @@
extern "C" {
#endif
+#ifdef RTE_USE_C11_MEM_MODEL
+#include <rte_ring_rts_c11_mem.h>
+#else
#include <rte_ring_rts_generic.h>
+#endif
/**
* @internal Enqueue several objects on the RTS ring.
--
2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [PATCH v3 3/9] ring: introduce RTS ring mode
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 3/9] ring: introduce RTS ring mode Konstantin Ananyev
@ 2020-04-04 17:27 ` Wang, Haiyue
2020-04-08 5:00 ` Honnappa Nagarahalli
1 sibling, 0 replies; 146+ messages in thread
From: Wang, Haiyue @ 2020-04-04 17:27 UTC (permalink / raw)
To: Ananyev, Konstantin, dev
Cc: honnappa.nagarahalli, david.marchand, jielong.zjl, Ananyev, Konstantin
> -----Original Message-----
> From: dev <dev-bounces@dpdk.org> On Behalf Of Konstantin Ananyev
> Sent: Saturday, April 4, 2020 01:42
> To: dev@dpdk.org
> Cc: honnappa.nagarahalli@arm.com; david.marchand@redhat.com; jielong.zjl@antfin.com; Ananyev,
> Konstantin <konstantin.ananyev@intel.com>
> Subject: [dpdk-dev] [PATCH v3 3/9] ring: introduce RTS ring mode
>
> Introduce relaxed tail sync (RTS) mode for MT ring synchronization.
> Aim to reduce stall times in case when ring is used on
> overcommited cpus (multiple active threads on the same cpu).
> The main difference from original MP/MC algorithm is that
> tail value is increased not by every thread that finished enqueue/dequeue,
> but only by the last one.
> That allows threads to avoid spinning on ring tail value,
> leaving actual tail value change to the last thread in the update queue.
>
> check-abi.sh reports what I believe is a false-positive about
> ring cons/prod changes. As a workaround, devtools/libabigail.abignore is
> updated to suppress *struct ring* related errors.
>
> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> ---
> devtools/libabigail.abignore | 7 +
> lib/librte_ring/Makefile | 5 +-
> lib/librte_ring/meson.build | 5 +-
> lib/librte_ring/rte_ring.c | 100 +++++++-
> lib/librte_ring/rte_ring.h | 110 ++++++++-
> lib/librte_ring/rte_ring_elem.h | 86 ++++++-
> lib/librte_ring/rte_ring_rts.h | 316 +++++++++++++++++++++++++
> lib/librte_ring/rte_ring_rts_elem.h | 205 ++++++++++++++++
> lib/librte_ring/rte_ring_rts_generic.h | 210 ++++++++++++++++
> 9 files changed, 1015 insertions(+), 29 deletions(-)
> create mode 100644 lib/librte_ring/rte_ring_rts.h
> create mode 100644 lib/librte_ring/rte_ring_rts_elem.h
> create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
>
> #ifdef __cplusplus
> diff --git a/lib/librte_ring/rte_ring_rts.h b/lib/librte_ring/rte_ring_rts.h
> new file mode 100644
> index 000000000..18404fe48
> --- /dev/null
> +++ b/lib/librte_ring/rte_ring_rts.h
> @@ -0,0 +1,316 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + *
> + * Copyright (c) 2010-2017 Intel Corporation
> + * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
> + * All rights reserved.
> + * Derived from FreeBSD's bufring.h
Find that it is buf_ring.h in real ;-)
> + * Used as BSD-3 Licensed with permission from Kip Macy.
> + */
> --
> 2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [PATCH v3 1/9] test/ring: add contention stress test
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 1/9] test/ring: add contention stress test Konstantin Ananyev
@ 2020-04-08 4:59 ` Honnappa Nagarahalli
2020-04-09 12:36 ` Ananyev, Konstantin
0 siblings, 1 reply; 146+ messages in thread
From: Honnappa Nagarahalli @ 2020-04-08 4:59 UTC (permalink / raw)
To: Konstantin Ananyev, dev
Cc: david.marchand, jielong.zjl, nd, Honnappa Nagarahalli, nd
<snip>
> Subject: [PATCH v3 1/9] test/ring: add contention stress test
Minor, would 'add stress test for overcommitted use case' sound better?
>
> Introduce new test-case to measure ring perfomance under contention
Minor, 'over committed' seems to the word commonly used from the references you provided. Does it make sense to use that?
> (miltiple producers/consumers).
^^^^^^^ multiple
> Starts dequeue/enqueue loop on all available slave lcores.
>
> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> ---
> app/test/Makefile | 2 +
> app/test/meson.build | 2 +
> app/test/test_ring_mpmc_stress.c | 31 +++
> app/test/test_ring_stress.c | 48 ++++
> app/test/test_ring_stress.h | 35 +++
> app/test/test_ring_stress_impl.h | 444 +++++++++++++++++++++++++++++++
Would be good to change the file names to indicate that these tests are for over-committed usecase/configuration.
These are performance tests, better to have 'perf' or 'performance' in their names.
> 6 files changed, 562 insertions(+)
> create mode 100644 app/test/test_ring_mpmc_stress.c create mode 100644
> app/test/test_ring_stress.c create mode 100644 app/test/test_ring_stress.h
> create mode 100644 app/test/test_ring_stress_impl.h
>
> diff --git a/app/test/Makefile b/app/test/Makefile index
> 1f080d162..4eefaa887 100644
> --- a/app/test/Makefile
> +++ b/app/test/Makefile
> @@ -77,7 +77,9 @@ SRCS-y += test_external_mem.c SRCS-y +=
> test_rand_perf.c
>
> SRCS-y += test_ring.c
> +SRCS-y += test_ring_mpmc_stress.c
> SRCS-y += test_ring_perf.c
> +SRCS-y += test_ring_stress.c
> SRCS-y += test_pmd_perf.c
>
> ifeq ($(CONFIG_RTE_LIBRTE_TABLE),y)
> diff --git a/app/test/meson.build b/app/test/meson.build index
> 351d29cb6..827b04886 100644
> --- a/app/test/meson.build
> +++ b/app/test/meson.build
> @@ -100,7 +100,9 @@ test_sources = files('commands.c',
> 'test_rib.c',
> 'test_rib6.c',
> 'test_ring.c',
> + 'test_ring_mpmc_stress.c',
> 'test_ring_perf.c',
> + 'test_ring_stress.c',
> 'test_rwlock.c',
> 'test_sched.c',
> 'test_service_cores.c',
> diff --git a/app/test/test_ring_mpmc_stress.c
> b/app/test/test_ring_mpmc_stress.c
> new file mode 100644
> index 000000000..1524b0248
> --- /dev/null
> +++ b/app/test/test_ring_mpmc_stress.c
> @@ -0,0 +1,31 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2020 Intel Corporation
> + */
> +
> +#include "test_ring_stress_impl.h"
> +
> +static inline uint32_t
> +_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
> + uint32_t *avail)
> +{
> + return rte_ring_mc_dequeue_bulk(r, obj, n, avail); }
> +
> +static inline uint32_t
> +_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
> + uint32_t *free)
> +{
> + return rte_ring_mp_enqueue_bulk(r, obj, n, free); }
> +
> +static int
> +_st_ring_init(struct rte_ring *r, const char *name, uint32_t num) {
> + return rte_ring_init(r, name, num, 0); }
> +
> +const struct test test_ring_mpmc_stress = {
> + .name = "MP/MC",
> + .nb_case = RTE_DIM(tests),
> + .cases = tests,
> +};
> diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c new file
> mode 100644 index 000000000..60706f799
> --- /dev/null
> +++ b/app/test/test_ring_stress.c
> @@ -0,0 +1,48 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2020 Intel Corporation
> + */
> +
> +#include "test_ring_stress.h"
> +
> +static int
> +run_test(const struct test *test)
> +{
> + int32_t rc;
> + uint32_t i, k;
> +
> + for (i = 0, k = 0; i != test->nb_case; i++) {
> +
> + printf("TEST-CASE %s %s START\n",
> + test->name, test->cases[i].name);
> +
> + rc = test->cases[i].func(test->cases[i].wfunc);
> + k += (rc == 0);
> +
> + if (rc != 0)
> + printf("TEST-CASE %s %s FAILED\n",
> + test->name, test->cases[i].name);
> + else
> + printf("TEST-CASE %s %s OK\n",
> + test->name, test->cases[i].name);
> + }
> +
> + return k;
> +}
> +
> +static int
> +test_ring_stress(void)
> +{
> + uint32_t n, k;
> +
> + n = 0;
> + k = 0;
> +
> + n += test_ring_mpmc_stress.nb_case;
> + k += run_test(&test_ring_mpmc_stress);
> +
> + printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
> + n, k, n - k);
> + return (k != n);
> +}
> +
> +REGISTER_TEST_COMMAND(ring_stress_autotest, test_ring_stress);
> diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h new file
> mode 100644 index 000000000..60eac6216
> --- /dev/null
> +++ b/app/test/test_ring_stress.h
> @@ -0,0 +1,35 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2020 Intel Corporation
> + */
> +
> +
> +#include <inttypes.h>
> +#include <stddef.h>
> +#include <stdalign.h>
> +#include <string.h>
> +#include <stdio.h>
> +#include <unistd.h>
> +
> +#include <rte_ring.h>
> +#include <rte_cycles.h>
> +#include <rte_launch.h>
> +#include <rte_pause.h>
> +#include <rte_random.h>
> +#include <rte_malloc.h>
> +#include <rte_spinlock.h>
> +
> +#include "test.h"
> +
> +struct test_case {
> + const char *name;
> + int (*func)(int (*)(void *));
> + int (*wfunc)(void *arg);
> +};
> +
> +struct test {
> + const char *name;
> + uint32_t nb_case;
> + const struct test_case *cases;
> +};
> +
> +extern const struct test test_ring_mpmc_stress;
> diff --git a/app/test/test_ring_stress_impl.h
> b/app/test/test_ring_stress_impl.h
> new file mode 100644
> index 000000000..11476d28c
> --- /dev/null
> +++ b/app/test/test_ring_stress_impl.h
> @@ -0,0 +1,444 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + * Copyright(c) 2020 Intel Corporation
> + */
> +
> +#include "test_ring_stress.h"
> +
> +/*
> + * Measures performance of ring enqueue/dequeue under high contention
> +*/
> +
> +#define RING_NAME "RING_STRESS"
> +#define BULK_NUM 32
> +#define RING_SIZE (2 * BULK_NUM * RTE_MAX_LCORE)
> +
> +enum {
> + WRK_CMD_STOP,
> + WRK_CMD_RUN,
> +};
> +
> +static volatile uint32_t wrk_cmd __rte_cache_aligned;
> +
> +/* test run-time in seconds */
> +static const uint32_t run_time = 60;
> +static const uint32_t verbose;
> +
> +struct lcore_stat {
> + uint64_t nb_cycle;
> + struct {
> + uint64_t nb_call;
> + uint64_t nb_obj;
> + uint64_t nb_cycle;
> + uint64_t max_cycle;
> + uint64_t min_cycle;
> + } op;
> +};
> +
> +struct lcore_arg {
> + struct rte_ring *rng;
> + struct lcore_stat stats;
> +} __rte_cache_aligned;
> +
> +struct ring_elem {
> + uint32_t cnt[RTE_CACHE_LINE_SIZE / sizeof(uint32_t)]; }
> +__rte_cache_aligned;
> +
> +/*
> + * redefinable functions
> + */
> +static uint32_t
> +_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
> + uint32_t *avail);
> +
> +static uint32_t
> +_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
> + uint32_t *free);
> +
> +static int
> +_st_ring_init(struct rte_ring *r, const char *name, uint32_t num);
> +
> +
> +static void
> +lcore_stat_update(struct lcore_stat *ls, uint64_t call, uint64_t obj,
> + uint64_t tm, int32_t prcs)
> +{
> + ls->op.nb_call += call;
> + ls->op.nb_obj += obj;
> + ls->op.nb_cycle += tm;
> + if (prcs) {
> + ls->op.max_cycle = RTE_MAX(ls->op.max_cycle, tm);
> + ls->op.min_cycle = RTE_MIN(ls->op.min_cycle, tm);
> + }
> +}
> +
> +static void
> +lcore_op_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls)
> +{
> +
> + ms->op.nb_call += ls->op.nb_call;
> + ms->op.nb_obj += ls->op.nb_obj;
> + ms->op.nb_cycle += ls->op.nb_cycle;
> + ms->op.max_cycle = RTE_MAX(ms->op.max_cycle, ls->op.max_cycle);
> + ms->op.min_cycle = RTE_MIN(ms->op.min_cycle, ls->op.min_cycle); }
> +
> +static void
> +lcore_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls) {
> + ms->nb_cycle = RTE_MAX(ms->nb_cycle, ls->nb_cycle);
> + lcore_op_stat_aggr(ms, ls);
> +}
> +
> +static void
> +lcore_stat_dump(FILE *f, uint32_t lc, const struct lcore_stat *ls) {
> + long double st;
> +
> + st = (long double)rte_get_timer_hz() / US_PER_S;
> +
> + if (lc == UINT32_MAX)
> + fprintf(f, "%s(AGGREGATE)={\n", __func__);
> + else
> + fprintf(f, "%s(lc=%u)={\n", __func__, lc);
> +
> + fprintf(f, "\tnb_cycle=%" PRIu64 "(%.2Lf usec),\n",
> + ls->nb_cycle, (long double)ls->nb_cycle / st);
> +
> + fprintf(f, "\tDEQ+ENQ={\n");
> +
> + fprintf(f, "\t\tnb_call=%" PRIu64 ",\n", ls->op.nb_call);
> + fprintf(f, "\t\tnb_obj=%" PRIu64 ",\n", ls->op.nb_obj);
> + fprintf(f, "\t\tnb_cycle=%" PRIu64 ",\n", ls->op.nb_cycle);
> + fprintf(f, "\t\tobj/call(avg): %.2Lf\n",
> + (long double)ls->op.nb_obj / ls->op.nb_call);
> + fprintf(f, "\t\tcycles/obj(avg): %.2Lf\n",
> + (long double)ls->op.nb_cycle / ls->op.nb_obj);
> + fprintf(f, "\t\tcycles/call(avg): %.2Lf\n",
> + (long double)ls->op.nb_cycle / ls->op.nb_call);
> +
> + /* if min/max cycles per call stats was collected */
> + if (ls->op.min_cycle != UINT64_MAX) {
> + fprintf(f, "\t\tmax cycles/call=%" PRIu64 "(%.2Lf usec),\n",
> + ls->op.max_cycle,
> + (long double)ls->op.max_cycle / st);
> + fprintf(f, "\t\tmin cycles/call=%" PRIu64 "(%.2Lf usec),\n",
> + ls->op.min_cycle,
> + (long double)ls->op.min_cycle / st);
> + }
> +
> + fprintf(f, "\t},\n");
> + fprintf(f, "};\n");
> +}
> +
> +static void
> +fill_ring_elm(struct ring_elem *elm, uint32_t fill) {
> + uint32_t i;
> +
> + for (i = 0; i != RTE_DIM(elm->cnt); i++)
> + elm->cnt[i] = fill;
> +}
> +
> +static int32_t
> +check_updt_elem(struct ring_elem *elm[], uint32_t num,
> + const struct ring_elem *check, const struct ring_elem *fill) {
> + uint32_t i;
> +
> + static rte_spinlock_t dump_lock;
> +
> + for (i = 0; i != num; i++) {
> + if (memcmp(check, elm[i], sizeof(*check)) != 0) {
> + rte_spinlock_lock(&dump_lock);
> + printf("%s(lc=%u, num=%u) failed at %u-th iter, "
> + "offending object: %p\n",
> + __func__, rte_lcore_id(), num, i, elm[i]);
> + rte_memdump(stdout, "expected", check,
> sizeof(*check));
> + rte_memdump(stdout, "result", elm[i], sizeof(elm[i]));
> + rte_spinlock_unlock(&dump_lock);
> + return -EINVAL;
> + }
> + memcpy(elm[i], fill, sizeof(*elm[i]));
> + }
> +
> + return 0;
> +}
> +
> +static int
> +check_ring_op(uint32_t exp, uint32_t res, uint32_t lc,
minor, lcore instead of lc would be better
> + const char *fname, const char *opname) {
> + if (exp != res) {
> + printf("%s(lc=%u) failure: %s expected: %u, returned %u\n",
Suggest using lcore in the printf
> + fname, lc, opname, exp, res);
> + return -ENOSPC;
> + }
> + return 0;
> +}
> +
> +static int
> +test_worker_prcs(void *arg)
> +{
> + int32_t rc;
> + uint32_t lc, n, num;
minor, lcore instead of lc would be better
> + uint64_t cl, tm0, tm1;
> + struct lcore_arg *la;
> + struct ring_elem def_elm, loc_elm;
> + struct ring_elem *obj[2 * BULK_NUM];
> +
> + la = arg;
> + lc = rte_lcore_id();
> +
> + fill_ring_elm(&def_elm, UINT32_MAX);
> + fill_ring_elm(&loc_elm, lc);
> +
> + while (wrk_cmd != WRK_CMD_RUN) {
> + rte_smp_rmb();
> + rte_pause();
> + }
> +
> + cl = rte_rdtsc_precise();
> +
> + do {
> + /* num in interval [7/8, 11/8] of BULK_NUM */
> + num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
> +
> + /* reset all pointer values */
> + memset(obj, 0, sizeof(obj));
> +
> + /* dequeue num elems */
> + tm0 = rte_rdtsc_precise();
> + n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
> + tm0 = rte_rdtsc_precise() - tm0;
> +
> + /* check return value and objects */
> + rc = check_ring_op(num, n, lc, __func__,
> + RTE_STR(_st_ring_dequeue_bulk));
> + if (rc == 0)
> + rc = check_updt_elem(obj, num, &def_elm,
> &loc_elm);
> + if (rc != 0)
> + break;
Since this seems like a performance test, should we skip validating the objects?
Did these tests run on Travis CI? I believe Travis CI has trouble running stress/performance tests if they take too much time.
The RTS and HTS tests should be added to functional tests.
> +
> + /* enqueue num elems */
> + rte_compiler_barrier();
> + rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
> + if (rc != 0)
> + break;
> +
> + tm1 = rte_rdtsc_precise();
> + n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
> + tm1 = rte_rdtsc_precise() - tm1;
> +
> + /* check return value */
> + rc = check_ring_op(num, n, lc, __func__,
> + RTE_STR(_st_ring_enqueue_bulk));
> + if (rc != 0)
> + break;
> +
> + lcore_stat_update(&la->stats, 1, num, tm0 + tm1, 1);
> +
> + } while (wrk_cmd == WRK_CMD_RUN);
> +
> + la->stats.nb_cycle = rte_rdtsc_precise() - cl;
> + return rc;
> +}
> +
> +static int
> +test_worker_avg(void *arg)
> +{
> + int32_t rc;
> + uint32_t lc, n, num;
> + uint64_t cl;
> + struct lcore_arg *la;
> + struct ring_elem def_elm, loc_elm;
> + struct ring_elem *obj[2 * BULK_NUM];
> +
> + la = arg;
> + lc = rte_lcore_id();
> +
> + fill_ring_elm(&def_elm, UINT32_MAX);
> + fill_ring_elm(&loc_elm, lc);
> +
> + while (wrk_cmd != WRK_CMD_RUN) {
> + rte_smp_rmb();
> + rte_pause();
> + }
> +
> + cl = rte_rdtsc_precise();
> +
> + do {
> + /* num in interval [7/8, 11/8] of BULK_NUM */
> + num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
> +
> + /* reset all pointer values */
> + memset(obj, 0, sizeof(obj));
> +
> + /* dequeue num elems */
> + n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
> +
> + /* check return value and objects */
> + rc = check_ring_op(num, n, lc, __func__,
> + RTE_STR(_st_ring_dequeue_bulk));
> + if (rc == 0)
> + rc = check_updt_elem(obj, num, &def_elm,
> &loc_elm);
> + if (rc != 0)
> + break;
> +
> + /* enqueue num elems */
> + rte_compiler_barrier();
> + rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
> + if (rc != 0)
> + break;
> +
> + n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
> +
> + /* check return value */
> + rc = check_ring_op(num, n, lc, __func__,
> + RTE_STR(_st_ring_enqueue_bulk));
> + if (rc != 0)
> + break;
> +
> + lcore_stat_update(&la->stats, 1, num, 0, 0);
> +
> + } while (wrk_cmd == WRK_CMD_RUN);
> +
> + /* final stats update */
> + cl = rte_rdtsc_precise() - cl;
> + lcore_stat_update(&la->stats, 0, 0, cl, 0);
> + la->stats.nb_cycle = cl;
> +
> + return rc;
> +}
Just wondering about the need of 2 tests which run the same functionality. The difference is the way in which numbers are collected.
Does 'test_worker_avg' adding any value? IMO, we can remove 'test_worker_avg'.
> +
> +static void
> +mt1_fini(struct rte_ring *rng, void *data) {
> + rte_free(rng);
> + rte_free(data);
> +}
> +
> +static int
> +mt1_init(struct rte_ring **rng, void **data, uint32_t num) {
> + int32_t rc;
> + size_t sz;
> + uint32_t i, nr;
> + struct rte_ring *r;
> + struct ring_elem *elm;
> + void *p;
> +
> + *rng = NULL;
> + *data = NULL;
> +
> + sz = num * sizeof(*elm);
> + elm = rte_zmalloc(NULL, sz, __alignof__(*elm));
> + if (elm == NULL) {
> + printf("%s: alloc(%zu) for %u elems data failed",
> + __func__, sz, num);
> + return -ENOMEM;
> + }
> +
> + *data = elm;
> +
> + /* alloc ring */
> + nr = 2 * num;
> + sz = rte_ring_get_memsize(nr);
> + r = rte_zmalloc(NULL, sz, __alignof__(*r));
> + if (r == NULL) {
> + printf("%s: alloc(%zu) for FIFO with %u elems failed",
> + __func__, sz, nr);
> + return -ENOMEM;
> + }
> +
> + *rng = r;
> +
> + rc = _st_ring_init(r, RING_NAME, nr);
> + if (rc != 0) {
> + printf("%s: _st_ring_init(%p, %u) failed, error: %d(%s)\n",
> + __func__, r, nr, rc, strerror(-rc));
> + return rc;
> + }
> +
> + for (i = 0; i != num; i++) {
> + fill_ring_elm(elm + i, UINT32_MAX);
> + p = elm + i;
> + if (_st_ring_enqueue_bulk(r, &p, 1, NULL) != 1)
> + break;
> + }
> +
> + if (i != num) {
> + printf("%s: _st_ring_enqueue(%p, %u) returned %u\n",
> + __func__, r, num, i);
> + return -ENOSPC;
> + }
> +
> + return 0;
> +}
> +
> +static int
> +test_mt1(int (*test)(void *))
> +{
> + int32_t rc;
> + uint32_t lc, mc;
> + struct rte_ring *r;
> + void *data;
> + struct lcore_arg arg[RTE_MAX_LCORE];
> +
> + static const struct lcore_stat init_stat = {
> + .op.min_cycle = UINT64_MAX,
> + };
> +
> + rc = mt1_init(&r, &data, RING_SIZE);
> + if (rc != 0) {
> + mt1_fini(r, data);
> + return rc;
> + }
> +
> + memset(arg, 0, sizeof(arg));
> +
> + /* launch on all slaves */
> + RTE_LCORE_FOREACH_SLAVE(lc) {
> + arg[lc].rng = r;
> + arg[lc].stats = init_stat;
> + rte_eal_remote_launch(test, &arg[lc], lc);
> + }
> +
> + /* signal worker to start test */
> + wrk_cmd = WRK_CMD_RUN;
> + rte_smp_wmb();
> +
> + usleep(run_time * US_PER_S);
> +
> + /* signal worker to start test */
> + wrk_cmd = WRK_CMD_STOP;
> + rte_smp_wmb();
> +
> + /* wait for slaves and collect stats. */
> + mc = rte_lcore_id();
> + arg[mc].stats = init_stat;
> +
> + rc = 0;
> + RTE_LCORE_FOREACH_SLAVE(lc) {
> + rc |= rte_eal_wait_lcore(lc);
> + lcore_stat_aggr(&arg[mc].stats, &arg[lc].stats);
> + if (verbose != 0)
> + lcore_stat_dump(stdout, lc, &arg[lc].stats);
> + }
> +
> + lcore_stat_dump(stdout, UINT32_MAX, &arg[mc].stats);
> + mt1_fini(r, data);
> + return rc;
> +}
> +
> +static const struct test_case tests[] = {
> + {
> + .name = "MT-WRK_ENQ_DEQ-MST_NONE-PRCS",
> + .func = test_mt1,
> + .wfunc = test_worker_prcs,
> + },
> + {
> + .name = "MT-WRK_ENQ_DEQ-MST_NONE-AVG",
> + .func = test_mt1,
> + .wfunc = test_worker_avg,
> + },
> +};
> --
> 2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [PATCH v3 2/9] ring: prepare ring to allow new sync schemes
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 2/9] ring: prepare ring to allow new sync schemes Konstantin Ananyev
@ 2020-04-08 4:59 ` Honnappa Nagarahalli
2020-04-09 13:39 ` Ananyev, Konstantin
0 siblings, 1 reply; 146+ messages in thread
From: Honnappa Nagarahalli @ 2020-04-08 4:59 UTC (permalink / raw)
To: Konstantin Ananyev, dev
Cc: david.marchand, jielong.zjl, nd, Honnappa Nagarahalli, nd
<snip>
> Subject: [PATCH v3 2/9] ring: prepare ring to allow new sync schemes
>
> Change from *single* to *sync_type* to allow different synchronisation
> schemes to be applied.
> Mark *single* as deprecated in comments.
> Add new functions to allow user to query ring sync types.
> Replace direct access to *single* with appopriate function call.
>
> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> ---
> app/test/test_pdump.c | 6 +-
> lib/librte_pdump/rte_pdump.c | 2 +-
> lib/librte_port/rte_port_ring.c | 12 ++--
> lib/librte_ring/rte_ring.c | 6 +-
> lib/librte_ring/rte_ring.h | 113 ++++++++++++++++++++++++++------
> lib/librte_ring/rte_ring_elem.h | 8 +--
> 6 files changed, 108 insertions(+), 39 deletions(-)
>
> diff --git a/app/test/test_pdump.c b/app/test/test_pdump.c index
> ad183184c..6a1180bcb 100644
> --- a/app/test/test_pdump.c
> +++ b/app/test/test_pdump.c
> @@ -57,8 +57,7 @@ run_pdump_client_tests(void)
> if (ret < 0)
> return -1;
> mp->flags = 0x0000;
> - ring_client = rte_ring_create("SR0", RING_SIZE, rte_socket_id(),
> - RING_F_SP_ENQ | RING_F_SC_DEQ);
> + ring_client = rte_ring_create("SR0", RING_SIZE, rte_socket_id(), 0);
Are you saying to get SP and SC behavior we now have to set the flags to 0? Isn't that a ABI break?
> if (ring_client == NULL) {
> printf("rte_ring_create SR0 failed");
> return -1;
> @@ -71,9 +70,6 @@ run_pdump_client_tests(void)
> }
> rte_eth_dev_probing_finish(eth_dev);
>
> - ring_client->prod.single = 0;
> - ring_client->cons.single = 0;
Just wondering if users outside of DPDK have done the same. I hope not, otherwise, we have an API break?
> -
> printf("\n***** flags = RTE_PDUMP_FLAG_TX *****\n");
>
> for (itr = 0; itr < NUM_ITR; itr++) {
> diff --git a/lib/librte_pdump/rte_pdump.c b/lib/librte_pdump/rte_pdump.c
> index 8a01ac510..65364f2c5 100644
> --- a/lib/librte_pdump/rte_pdump.c
> +++ b/lib/librte_pdump/rte_pdump.c
> @@ -380,7 +380,7 @@ pdump_validate_ring_mp(struct rte_ring *ring, struct
> rte_mempool *mp)
> rte_errno = EINVAL;
> return -1;
> }
> - if (ring->prod.single || ring->cons.single) {
> + if (rte_ring_prod_single(ring) || rte_ring_cons_single(ring)) {
> PDUMP_LOG(ERR, "ring with either SP or SC settings"
> " is not valid for pdump, should have MP and MC settings\n");
> rte_errno = EINVAL;
> diff --git a/lib/librte_port/rte_port_ring.c b/lib/librte_port/rte_port_ring.c
> index 47fcdd06a..2f6c050fa 100644
> --- a/lib/librte_port/rte_port_ring.c
> +++ b/lib/librte_port/rte_port_ring.c
> @@ -44,8 +44,8 @@ rte_port_ring_reader_create_internal(void *params, int
> socket_id,
> /* Check input parameters */
> if ((conf == NULL) ||
> (conf->ring == NULL) ||
> - (conf->ring->cons.single && is_multi) ||
> - (!(conf->ring->cons.single) && !is_multi)) {
> + (rte_ring_cons_single(conf->ring) && is_multi) ||
> + (!rte_ring_cons_single(conf->ring) && !is_multi)) {
> RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
> return NULL;
> }
> @@ -171,8 +171,8 @@ rte_port_ring_writer_create_internal(void *params,
> int socket_id,
> /* Check input parameters */
> if ((conf == NULL) ||
> (conf->ring == NULL) ||
> - (conf->ring->prod.single && is_multi) ||
> - (!(conf->ring->prod.single) && !is_multi) ||
> + (rte_ring_prod_single(conf->ring) && is_multi) ||
> + (!rte_ring_prod_single(conf->ring) && !is_multi) ||
> (conf->tx_burst_sz > RTE_PORT_IN_BURST_SIZE_MAX)) {
> RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
> return NULL;
> @@ -440,8 +440,8 @@ rte_port_ring_writer_nodrop_create_internal(void
> *params, int socket_id,
> /* Check input parameters */
> if ((conf == NULL) ||
> (conf->ring == NULL) ||
> - (conf->ring->prod.single && is_multi) ||
> - (!(conf->ring->prod.single) && !is_multi) ||
> + (rte_ring_prod_single(conf->ring) && is_multi) ||
> + (!rte_ring_prod_single(conf->ring) && !is_multi) ||
> (conf->tx_burst_sz > RTE_PORT_IN_BURST_SIZE_MAX)) {
> RTE_LOG(ERR, PORT, "%s: Invalid Parameters\n", __func__);
> return NULL;
> diff --git a/lib/librte_ring/rte_ring.c b/lib/librte_ring/rte_ring.c index
> 77e5de099..fa5733907 100644
> --- a/lib/librte_ring/rte_ring.c
> +++ b/lib/librte_ring/rte_ring.c
> @@ -106,8 +106,10 @@ rte_ring_init(struct rte_ring *r, const char *name,
> unsigned count,
> if (ret < 0 || ret >= (int)sizeof(r->name))
> return -ENAMETOOLONG;
> r->flags = flags;
> - r->prod.single = (flags & RING_F_SP_ENQ) ? __IS_SP : __IS_MP;
> - r->cons.single = (flags & RING_F_SC_DEQ) ? __IS_SC : __IS_MC;
> + r->prod.sync_type = (flags & RING_F_SP_ENQ) ?
> + RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
> + r->cons.sync_type = (flags & RING_F_SC_DEQ) ?
> + RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
>
> if (flags & RING_F_EXACT_SZ) {
> r->size = rte_align32pow2(count + 1); diff --git
> a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h index
> 18fc5d845..d4775a063 100644
> --- a/lib/librte_ring/rte_ring.h
> +++ b/lib/librte_ring/rte_ring.h
> @@ -61,11 +61,27 @@ enum rte_ring_queue_behavior { #define
> RTE_RING_NAMESIZE (RTE_MEMZONE_NAMESIZE - \
> sizeof(RTE_RING_MZ_PREFIX) + 1)
>
> -/* structure to hold a pair of head/tail values and other metadata */
> +/** prod/cons sync types */
> +enum rte_ring_sync_type {
> + RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
> + RTE_RING_SYNC_ST, /**< single thread only */
> +};
> +
> +/**
> + * structure to hold a pair of head/tail values and other metadata.
> + * Depending on sync_type format of that structure might be different,
> + * but offset for *sync_type* and *tail* values should remain the same.
> + */
> struct rte_ring_headtail {
> - volatile uint32_t head; /**< Prod/consumer head. */
> - volatile uint32_t tail; /**< Prod/consumer tail. */
> - uint32_t single; /**< True if single prod/cons */
> + volatile uint32_t head; /**< prod/consumer head. */
> + volatile uint32_t tail; /**< prod/consumer tail. */
> + RTE_STD_C11
> + union {
> + /** sync type of prod/cons */
> + enum rte_ring_sync_type sync_type;
> + /** deprecated - True if single prod/cons */
> + uint32_t single;
> + };
> };
>
> /**
> @@ -116,11 +132,10 @@ struct rte_ring {
> #define RING_F_EXACT_SZ 0x0004
> #define RTE_RING_SZ_MASK (0x7fffffffU) /**< Ring size mask */
>
> -/* @internal defines for passing to the enqueue dequeue worker functions
> */ -#define __IS_SP 1 -#define __IS_MP 0 -#define __IS_SC 1 -#define
> __IS_MC 0
> +#define __IS_SP RTE_RING_SYNC_ST
> +#define __IS_MP RTE_RING_SYNC_MT
> +#define __IS_SC RTE_RING_SYNC_ST
> +#define __IS_MC RTE_RING_SYNC_MT
I think we can remove these #defines and use the new SYNC types
>
> /**
> * Calculate the memory size needed for a ring @@ -420,7 +435,7 @@
> rte_ring_mp_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
> unsigned int n, unsigned int *free_space) {
> return __rte_ring_do_enqueue(r, obj_table, n,
> RTE_RING_QUEUE_FIXED,
> - __IS_MP, free_space);
> + RTE_RING_SYNC_MT, free_space);
> }
>
> /**
> @@ -443,7 +458,7 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void *
> const *obj_table,
> unsigned int n, unsigned int *free_space) {
> return __rte_ring_do_enqueue(r, obj_table, n,
> RTE_RING_QUEUE_FIXED,
> - __IS_SP, free_space);
> + RTE_RING_SYNC_ST, free_space);
> }
>
> /**
> @@ -470,7 +485,7 @@ rte_ring_enqueue_bulk(struct rte_ring *r, void *
> const *obj_table,
> unsigned int n, unsigned int *free_space) {
> return __rte_ring_do_enqueue(r, obj_table, n,
> RTE_RING_QUEUE_FIXED,
> - r->prod.single, free_space);
> + r->prod.sync_type, free_space);
> }
>
> /**
> @@ -554,7 +569,7 @@ rte_ring_mc_dequeue_bulk(struct rte_ring *r, void
> **obj_table,
> unsigned int n, unsigned int *available) {
> return __rte_ring_do_dequeue(r, obj_table, n,
> RTE_RING_QUEUE_FIXED,
> - __IS_MC, available);
> + RTE_RING_SYNC_MT, available);
> }
>
> /**
> @@ -578,7 +593,7 @@ rte_ring_sc_dequeue_bulk(struct rte_ring *r, void
> **obj_table,
> unsigned int n, unsigned int *available) {
> return __rte_ring_do_dequeue(r, obj_table, n,
> RTE_RING_QUEUE_FIXED,
> - __IS_SC, available);
> + RTE_RING_SYNC_ST, available);
> }
>
> /**
> @@ -605,7 +620,7 @@ rte_ring_dequeue_bulk(struct rte_ring *r, void
> **obj_table, unsigned int n,
> unsigned int *available)
> {
> return __rte_ring_do_dequeue(r, obj_table, n,
> RTE_RING_QUEUE_FIXED,
> - r->cons.single, available);
> + r->cons.sync_type, available);
> }
>
> /**
> @@ -777,6 +792,62 @@ rte_ring_get_capacity(const struct rte_ring *r)
> return r->capacity;
> }
>
> +/**
> + * Return sync type used by producer in the ring.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @return
> + * Producer sync type value.
> + */
> +static inline enum rte_ring_sync_type
> +rte_ring_get_prod_sync_type(const struct rte_ring *r) {
> + return r->prod.sync_type;
> +}
> +
> +/**
> + * Check is the ring for single producer.
^^ if
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @return
> + * true if ring is SP, zero otherwise.
> + */
> +static inline int
> +rte_ring_prod_single(const struct rte_ring *r) {
would rte_ring_is_prod_single better?
> + return (rte_ring_get_prod_sync_type(r) == RTE_RING_SYNC_ST); }
> +
> +/**
> + * Return sync type used by consumer in the ring.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @return
> + * Consumer sync type value.
> + */
> +static inline enum rte_ring_sync_type
> +rte_ring_get_cons_sync_type(const struct rte_ring *r) {
> + return r->cons.sync_type;
> +}
> +
> +/**
> + * Check is the ring for single consumer.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @return
> + * true if ring is SC, zero otherwise.
> + */
> +static inline int
> +rte_ring_cons_single(const struct rte_ring *r) {
> + return (rte_ring_get_cons_sync_type(r) == RTE_RING_SYNC_ST); }
> +
All these new functions are not required to be called in the data path. They can be made non-inline.
> /**
> * Dump the status of all rings on the console
> *
> @@ -820,7 +891,7 @@ rte_ring_mp_enqueue_burst(struct rte_ring *r, void *
> const *obj_table,
> unsigned int n, unsigned int *free_space) {
> return __rte_ring_do_enqueue(r, obj_table, n,
> - RTE_RING_QUEUE_VARIABLE, __IS_MP, free_space);
> + RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_MT,
> free_space);
> }
>
> /**
> @@ -843,7 +914,7 @@ rte_ring_sp_enqueue_burst(struct rte_ring *r, void *
> const *obj_table,
> unsigned int n, unsigned int *free_space) {
> return __rte_ring_do_enqueue(r, obj_table, n,
> - RTE_RING_QUEUE_VARIABLE, __IS_SP, free_space);
> + RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST,
> free_space);
> }
>
> /**
> @@ -870,7 +941,7 @@ rte_ring_enqueue_burst(struct rte_ring *r, void *
> const *obj_table,
> unsigned int n, unsigned int *free_space) {
> return __rte_ring_do_enqueue(r, obj_table, n,
> RTE_RING_QUEUE_VARIABLE,
> - r->prod.single, free_space);
> + r->prod.sync_type, free_space);
> }
>
> /**
> @@ -898,7 +969,7 @@ rte_ring_mc_dequeue_burst(struct rte_ring *r, void
> **obj_table,
> unsigned int n, unsigned int *available) {
> return __rte_ring_do_dequeue(r, obj_table, n,
> - RTE_RING_QUEUE_VARIABLE, __IS_MC, available);
> + RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_MT,
> available);
> }
>
> /**
> @@ -923,7 +994,7 @@ rte_ring_sc_dequeue_burst(struct rte_ring *r, void
> **obj_table,
> unsigned int n, unsigned int *available) {
> return __rte_ring_do_dequeue(r, obj_table, n,
> - RTE_RING_QUEUE_VARIABLE, __IS_SC, available);
> + RTE_RING_QUEUE_VARIABLE, RTE_RING_SYNC_ST,
> available);
> }
>
> /**
> @@ -951,7 +1022,7 @@ rte_ring_dequeue_burst(struct rte_ring *r, void
> **obj_table, {
> return __rte_ring_do_dequeue(r, obj_table, n,
> RTE_RING_QUEUE_VARIABLE,
> - r->cons.single, available);
> + r->cons.sync_type, available);
> }
>
> #ifdef __cplusplus
> diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
> index 663addc73..28f9836e6 100644
> --- a/lib/librte_ring/rte_ring_elem.h
> +++ b/lib/librte_ring/rte_ring_elem.h
> @@ -570,7 +570,7 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r, const
> void *obj_table,
> unsigned int esize, unsigned int n, unsigned int *free_space) {
> return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
> - RTE_RING_QUEUE_FIXED, r->prod.single, free_space);
> + RTE_RING_QUEUE_FIXED, r->prod.sync_type,
> free_space);
> }
>
> /**
> @@ -734,7 +734,7 @@ rte_ring_dequeue_bulk_elem(struct rte_ring *r, void
> *obj_table,
> unsigned int esize, unsigned int n, unsigned int *available) {
> return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
> - RTE_RING_QUEUE_FIXED, r->cons.single, available);
> + RTE_RING_QUEUE_FIXED, r->cons.sync_type,
> available);
> }
>
> /**
> @@ -902,7 +902,7 @@ rte_ring_enqueue_burst_elem(struct rte_ring *r,
> const void *obj_table,
> unsigned int esize, unsigned int n, unsigned int *free_space) {
> return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
> - RTE_RING_QUEUE_VARIABLE, r->prod.single,
> free_space);
> + RTE_RING_QUEUE_VARIABLE, r->prod.sync_type,
> free_space);
> }
>
> /**
> @@ -995,7 +995,7 @@ rte_ring_dequeue_burst_elem(struct rte_ring *r, void
> *obj_table, {
> return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
> RTE_RING_QUEUE_VARIABLE,
> - r->cons.single, available);
> + r->cons.sync_type, available);
> }
>
> #ifdef __cplusplus
> --
> 2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [PATCH v3 3/9] ring: introduce RTS ring mode
2020-04-03 17:42 ` [dpdk-dev] [PATCH v3 3/9] ring: introduce RTS ring mode Konstantin Ananyev
2020-04-04 17:27 ` Wang, Haiyue
@ 2020-04-08 5:00 ` Honnappa Nagarahalli
2020-04-09 14:52 ` Ananyev, Konstantin
1 sibling, 1 reply; 146+ messages in thread
From: Honnappa Nagarahalli @ 2020-04-08 5:00 UTC (permalink / raw)
To: Konstantin Ananyev, dev
Cc: david.marchand, jielong.zjl, Honnappa Nagarahalli, nd, nd
<snip>
>
> Introduce relaxed tail sync (RTS) mode for MT ring synchronization.
> Aim to reduce stall times in case when ring is used on overcommited cpus
> (multiple active threads on the same cpu).
> The main difference from original MP/MC algorithm is that tail value is
> increased not by every thread that finished enqueue/dequeue, but only by the
> last one.
> That allows threads to avoid spinning on ring tail value, leaving actual tail
> value change to the last thread in the update queue.
>
> check-abi.sh reports what I believe is a false-positive about ring cons/prod
> changes. As a workaround, devtools/libabigail.abignore is updated to suppress
> *struct ring* related errors.
This can be removed from the commit message.
>
> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> ---
> devtools/libabigail.abignore | 7 +
> lib/librte_ring/Makefile | 5 +-
> lib/librte_ring/meson.build | 5 +-
> lib/librte_ring/rte_ring.c | 100 +++++++-
> lib/librte_ring/rte_ring.h | 110 ++++++++-
> lib/librte_ring/rte_ring_elem.h | 86 ++++++-
> lib/librte_ring/rte_ring_rts.h | 316 +++++++++++++++++++++++++
> lib/librte_ring/rte_ring_rts_elem.h | 205 ++++++++++++++++
> lib/librte_ring/rte_ring_rts_generic.h | 210 ++++++++++++++++
> 9 files changed, 1015 insertions(+), 29 deletions(-) create mode 100644
> lib/librte_ring/rte_ring_rts.h create mode 100644
> lib/librte_ring/rte_ring_rts_elem.h
> create mode 100644 lib/librte_ring/rte_ring_rts_generic.h
>
> diff --git a/devtools/libabigail.abignore b/devtools/libabigail.abignore index
> a59df8f13..cd86d89ca 100644
> --- a/devtools/libabigail.abignore
> +++ b/devtools/libabigail.abignore
> @@ -11,3 +11,10 @@
> type_kind = enum
> name = rte_crypto_asym_xform_type
> changed_enumerators = RTE_CRYPTO_ASYM_XFORM_TYPE_LIST_END
> +; Ignore updates of ring prod/cons
> +[suppress_type]
> + type_kind = struct
> + name = rte_ring
> +[suppress_type]
> + type_kind = struct
> + name = rte_event_ring
Does this block the reporting of these structures forever?
> diff --git a/lib/librte_ring/Makefile b/lib/librte_ring/Makefile index
> 917c560ad..8f5c284cc 100644
> --- a/lib/librte_ring/Makefile
> +++ b/lib/librte_ring/Makefile
> @@ -18,6 +18,9 @@ SRCS-$(CONFIG_RTE_LIBRTE_RING) := rte_ring.c
> SYMLINK-$(CONFIG_RTE_LIBRTE_RING)-include := rte_ring.h \
> rte_ring_elem.h \
> rte_ring_generic.h \
> - rte_ring_c11_mem.h
> + rte_ring_c11_mem.h \
> + rte_ring_rts.h \
> + rte_ring_rts_elem.h \
> + rte_ring_rts_generic.h
>
> include $(RTE_SDK)/mk/rte.lib.mk
> diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build index
> f2f3ccc88..612936afb 100644
> --- a/lib/librte_ring/meson.build
> +++ b/lib/librte_ring/meson.build
> @@ -5,7 +5,10 @@ sources = files('rte_ring.c') headers = files('rte_ring.h',
> 'rte_ring_elem.h',
> 'rte_ring_c11_mem.h',
> - 'rte_ring_generic.h')
> + 'rte_ring_generic.h',
> + 'rte_ring_rts.h',
> + 'rte_ring_rts_elem.h',
> + 'rte_ring_rts_generic.h')
>
> # rte_ring_create_elem and rte_ring_get_memsize_elem are experimental
> allow_experimental_apis = true diff --git a/lib/librte_ring/rte_ring.c
> b/lib/librte_ring/rte_ring.c index fa5733907..222eec0fb 100644
> --- a/lib/librte_ring/rte_ring.c
> +++ b/lib/librte_ring/rte_ring.c
> @@ -45,6 +45,9 @@ EAL_REGISTER_TAILQ(rte_ring_tailq)
> /* true if x is a power of 2 */
> #define POWEROF2(x) ((((x)-1) & (x)) == 0)
>
> +/* by default set head/tail distance as 1/8 of ring capacity */
> +#define HTD_MAX_DEF 8
> +
> /* return the size of memory occupied by a ring */ ssize_t
> rte_ring_get_memsize_elem(unsigned int esize, unsigned int count) @@ -
> 79,11 +82,84 @@ rte_ring_get_memsize(unsigned int count)
> return rte_ring_get_memsize_elem(sizeof(void *), count); }
>
> +/*
> + * internal helper function to reset prod/cons head-tail values.
> + */
> +static void
> +reset_headtail(void *p)
> +{
> + struct rte_ring_headtail *ht;
> + struct rte_ring_rts_headtail *ht_rts;
> +
> + ht = p;
> + ht_rts = p;
> +
> + switch (ht->sync_type) {
> + case RTE_RING_SYNC_MT:
> + case RTE_RING_SYNC_ST:
> + ht->head = 0;
> + ht->tail = 0;
> + break;
> + case RTE_RING_SYNC_MT_RTS:
> + ht_rts->head.raw = 0;
> + ht_rts->tail.raw = 0;
> + break;
> + default:
> + /* unknown sync mode */
> + RTE_ASSERT(0);
> + }
> +}
> +
> void
> rte_ring_reset(struct rte_ring *r)
> {
> - r->prod.head = r->cons.head = 0;
> - r->prod.tail = r->cons.tail = 0;
> + reset_headtail(&r->prod);
> + reset_headtail(&r->cons);
> +}
> +
> +/*
> + * helper function, calculates sync_type values for prod and cons
> + * based on input flags. Returns zero at success or negative
> + * errno value otherwise.
> + */
> +static int
> +get_sync_type(uint32_t flags, enum rte_ring_sync_type *prod_st,
> + enum rte_ring_sync_type *cons_st)
> +{
> + static const uint32_t prod_st_flags =
> + (RING_F_SP_ENQ | RING_F_MP_RTS_ENQ);
> + static const uint32_t cons_st_flags =
> + (RING_F_SC_DEQ | RING_F_MC_RTS_DEQ);
> +
> + switch (flags & prod_st_flags) {
> + case 0:
> + *prod_st = RTE_RING_SYNC_MT;
> + break;
> + case RING_F_SP_ENQ:
> + *prod_st = RTE_RING_SYNC_ST;
> + break;
> + case RING_F_MP_RTS_ENQ:
> + *prod_st = RTE_RING_SYNC_MT_RTS;
> + break;
> + default:
> + return -EINVAL;
> + }
> +
> + switch (flags & cons_st_flags) {
> + case 0:
> + *cons_st = RTE_RING_SYNC_MT;
> + break;
> + case RING_F_SC_DEQ:
> + *cons_st = RTE_RING_SYNC_ST;
> + break;
> + case RING_F_MC_RTS_DEQ:
> + *cons_st = RTE_RING_SYNC_MT_RTS;
> + break;
> + default:
> + return -EINVAL;
> + }
> +
> + return 0;
> }
>
> int
> @@ -100,16 +176,20 @@ rte_ring_init(struct rte_ring *r, const char *name,
> unsigned count,
> RTE_BUILD_BUG_ON((offsetof(struct rte_ring, prod) &
> RTE_CACHE_LINE_MASK) != 0);
>
> + RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, sync_type) !=
> + offsetof(struct rte_ring_rts_headtail, sync_type));
> + RTE_BUILD_BUG_ON(offsetof(struct rte_ring_headtail, tail) !=
> + offsetof(struct rte_ring_rts_headtail, tail.val.pos));
> +
> /* init the ring structure */
> memset(r, 0, sizeof(*r));
> ret = strlcpy(r->name, name, sizeof(r->name));
> if (ret < 0 || ret >= (int)sizeof(r->name))
> return -ENAMETOOLONG;
> r->flags = flags;
> - r->prod.sync_type = (flags & RING_F_SP_ENQ) ?
> - RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
> - r->cons.sync_type = (flags & RING_F_SC_DEQ) ?
> - RTE_RING_SYNC_ST : RTE_RING_SYNC_MT;
> + ret = get_sync_type(flags, &r->prod.sync_type, &r->cons.sync_type);
> + if (ret != 0)
> + return ret;
>
> if (flags & RING_F_EXACT_SZ) {
> r->size = rte_align32pow2(count + 1); @@ -126,8 +206,12
> @@ rte_ring_init(struct rte_ring *r, const char *name, unsigned count,
> r->mask = count - 1;
> r->capacity = r->mask;
> }
> - r->prod.head = r->cons.head = 0;
> - r->prod.tail = r->cons.tail = 0;
> +
> + /* set default values for head-tail distance */
> + if (flags & RING_F_MP_RTS_ENQ)
> + rte_ring_set_prod_htd_max(r, r->capacity / HTD_MAX_DEF);
> + if (flags & RING_F_MC_RTS_DEQ)
> + rte_ring_set_cons_htd_max(r, r->capacity / HTD_MAX_DEF);
>
> return 0;
> }
> diff --git a/lib/librte_ring/rte_ring.h b/lib/librte_ring/rte_ring.h index
> d4775a063..f6f084d79 100644
> --- a/lib/librte_ring/rte_ring.h
> +++ b/lib/librte_ring/rte_ring.h
> @@ -48,6 +48,7 @@ extern "C" {
> #include <rte_branch_prediction.h>
> #include <rte_memzone.h>
> #include <rte_pause.h>
> +#include <rte_debug.h>
>
> #define RTE_TAILQ_RING_NAME "RTE_RING"
>
> @@ -65,10 +66,13 @@ enum rte_ring_queue_behavior { enum
> rte_ring_sync_type {
> RTE_RING_SYNC_MT, /**< multi-thread safe (default mode) */
> RTE_RING_SYNC_ST, /**< single thread only */
> +#ifdef ALLOW_EXPERIMENTAL_API
> + RTE_RING_SYNC_MT_RTS, /**< multi-thread relaxed tail sync */
> #endif
> };
>
> /**
> - * structure to hold a pair of head/tail values and other metadata.
> + * structures to hold a pair of head/tail values and other metadata.
> * Depending on sync_type format of that structure might be different,
> * but offset for *sync_type* and *tail* values should remain the same.
> */
> @@ -84,6 +88,21 @@ struct rte_ring_headtail {
> };
> };
>
> +union rte_ring_ht_poscnt {
nit, this is specific to RTS, may be change this to rte_ring_rts_ht_poscnt?
> + uint64_t raw;
> + struct {
> + uint32_t cnt; /**< head/tail reference counter */
> + uint32_t pos; /**< head/tail position */
> + } val;
> +};
> +
> +struct rte_ring_rts_headtail {
> + volatile union rte_ring_ht_poscnt tail;
> + enum rte_ring_sync_type sync_type; /**< sync type of prod/cons */
> + uint32_t htd_max; /**< max allowed distance between head/tail */
> + volatile union rte_ring_ht_poscnt head; };
> +
> /**
> * An RTE ring structure.
> *
> @@ -111,11 +130,21 @@ struct rte_ring {
> char pad0 __rte_cache_aligned; /**< empty cache line */
>
> /** Ring producer status. */
> - struct rte_ring_headtail prod __rte_cache_aligned;
> + RTE_STD_C11
> + union {
> + struct rte_ring_headtail prod;
> + struct rte_ring_rts_headtail rts_prod;
> + } __rte_cache_aligned;
> +
> char pad1 __rte_cache_aligned; /**< empty cache line */
>
> /** Ring consumer status. */
> - struct rte_ring_headtail cons __rte_cache_aligned;
> + RTE_STD_C11
> + union {
> + struct rte_ring_headtail cons;
> + struct rte_ring_rts_headtail rts_cons;
> + } __rte_cache_aligned;
> +
> char pad2 __rte_cache_aligned; /**< empty cache line */ };
>
> @@ -132,6 +161,9 @@ struct rte_ring {
> #define RING_F_EXACT_SZ 0x0004
> #define RTE_RING_SZ_MASK (0x7fffffffU) /**< Ring size mask */
>
> +#define RING_F_MP_RTS_ENQ 0x0008 /**< The default enqueue is "MP RTS".
> +*/ #define RING_F_MC_RTS_DEQ 0x0010 /**< The default dequeue is "MC
> +RTS". */
> +
> #define __IS_SP RTE_RING_SYNC_ST
> #define __IS_MP RTE_RING_SYNC_MT
> #define __IS_SC RTE_RING_SYNC_ST
> @@ -461,6 +493,10 @@ rte_ring_sp_enqueue_bulk(struct rte_ring *r, void *
> const *obj_table,
> RTE_RING_SYNC_ST, free_space);
> }
>
> +#ifdef ALLOW_EXPERIMENTAL_API
> +#include <rte_ring_rts.h>
> +#endif
> +
> /**
> * Enqueue several objects on a ring.
> *
> @@ -484,8 +520,21 @@ static __rte_always_inline unsigned int
> rte_ring_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
> unsigned int n, unsigned int *free_space) {
> - return __rte_ring_do_enqueue(r, obj_table, n,
> RTE_RING_QUEUE_FIXED,
> - r->prod.sync_type, free_space);
> + switch (r->prod.sync_type) {
> + case RTE_RING_SYNC_MT:
> + return rte_ring_mp_enqueue_bulk(r, obj_table, n, free_space);
> + case RTE_RING_SYNC_ST:
> + return rte_ring_sp_enqueue_bulk(r, obj_table, n, free_space);
Have you validated if these affect the performance for the existing APIs?
I am also wondering why should we support these new modes in the legacy APIs?
I think users should move to use rte_ring_xxx_elem APIs. If users want to use RTS/HTS it will be a good time for them to move to new APIs. They anyway have to test their code for RTS/HTS, might as well make the change to new APIs and test both.
It will be less code to maintain for the community as well.
> #ifdef
> +ALLOW_EXPERIMENTAL_API
> + case RTE_RING_SYNC_MT_RTS:
> + return rte_ring_mp_rts_enqueue_bulk(r, obj_table, n,
> + free_space);
> +#endif
> + }
> +
> + /* valid ring should never reach this point */
> + RTE_ASSERT(0);
> + return 0;
> }
>
> /**
> @@ -619,8 +668,20 @@ static __rte_always_inline unsigned int
> rte_ring_dequeue_bulk(struct rte_ring *r, void **obj_table, unsigned int n,
> unsigned int *available)
> {
> - return __rte_ring_do_dequeue(r, obj_table, n,
> RTE_RING_QUEUE_FIXED,
> - r->cons.sync_type, available);
> + switch (r->cons.sync_type) {
> + case RTE_RING_SYNC_MT:
> + return rte_ring_mc_dequeue_bulk(r, obj_table, n, available);
> + case RTE_RING_SYNC_ST:
> + return rte_ring_sc_dequeue_bulk(r, obj_table, n, available);
> #ifdef
> +ALLOW_EXPERIMENTAL_API
> + case RTE_RING_SYNC_MT_RTS:
> + return rte_ring_mc_rts_dequeue_bulk(r, obj_table, n,
> available);
> +#endif
> + }
> +
> + /* valid ring should never reach this point */
> + RTE_ASSERT(0);
> + return 0;
> }
>
> /**
> @@ -940,8 +1001,21 @@ static __rte_always_inline unsigned
> rte_ring_enqueue_burst(struct rte_ring *r, void * const *obj_table,
> unsigned int n, unsigned int *free_space) {
> - return __rte_ring_do_enqueue(r, obj_table, n,
> RTE_RING_QUEUE_VARIABLE,
> - r->prod.sync_type, free_space);
> + switch (r->prod.sync_type) {
> + case RTE_RING_SYNC_MT:
> + return rte_ring_mp_enqueue_burst(r, obj_table, n,
> free_space);
> + case RTE_RING_SYNC_ST:
> + return rte_ring_sp_enqueue_burst(r, obj_table, n, free_space);
> #ifdef
> +ALLOW_EXPERIMENTAL_API
> + case RTE_RING_SYNC_MT_RTS:
> + return rte_ring_mp_rts_enqueue_burst(r, obj_table, n,
> + free_space);
> +#endif
> + }
> +
> + /* valid ring should never reach this point */
> + RTE_ASSERT(0);
> + return 0;
> }
>
> /**
> @@ -1020,9 +1094,21 @@ static __rte_always_inline unsigned
> rte_ring_dequeue_burst(struct rte_ring *r, void **obj_table,
> unsigned int n, unsigned int *available) {
> - return __rte_ring_do_dequeue(r, obj_table, n,
> - RTE_RING_QUEUE_VARIABLE,
> - r->cons.sync_type, available);
> + switch (r->cons.sync_type) {
> + case RTE_RING_SYNC_MT:
> + return rte_ring_mc_dequeue_burst(r, obj_table, n, available);
> + case RTE_RING_SYNC_ST:
> + return rte_ring_sc_dequeue_burst(r, obj_table, n, available);
> #ifdef
> +ALLOW_EXPERIMENTAL_API
> + case RTE_RING_SYNC_MT_RTS:
> + return rte_ring_mc_rts_dequeue_burst(r, obj_table, n,
> + available);
> +#endif
> + }
> +
> + /* valid ring should never reach this point */
> + RTE_ASSERT(0);
> + return 0;
> }
>
> #ifdef __cplusplus
> diff --git a/lib/librte_ring/rte_ring_elem.h b/lib/librte_ring/rte_ring_elem.h
> index 28f9836e6..5de0850dc 100644
> --- a/lib/librte_ring/rte_ring_elem.h
> +++ b/lib/librte_ring/rte_ring_elem.h
> @@ -542,6 +542,8 @@ rte_ring_sp_enqueue_bulk_elem(struct rte_ring *r,
> const void *obj_table,
> RTE_RING_QUEUE_FIXED, __IS_SP, free_space); }
>
> +#include <rte_ring_rts_elem.h>
> +
> /**
> * Enqueue several objects on a ring.
> *
> @@ -571,6 +573,26 @@ rte_ring_enqueue_bulk_elem(struct rte_ring *r,
> const void *obj_table, {
> return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
> RTE_RING_QUEUE_FIXED, r->prod.sync_type,
> free_space);
> +
> + switch (r->prod.sync_type) {
> + case RTE_RING_SYNC_MT:
> + return rte_ring_mp_enqueue_bulk_elem(r, obj_table, esize, n,
> + free_space);
> + case RTE_RING_SYNC_ST:
> + return rte_ring_sp_enqueue_bulk_elem(r, obj_table, esize, n,
> + free_space);
> +#ifdef ALLOW_EXPERIMENTAL_API
> + case RTE_RING_SYNC_MT_RTS:
> + return rte_ring_mp_rts_enqueue_bulk_elem(r, obj_table,
> esize, n,
> + free_space);
> +#endif
> + }
> +
> + /* valid ring should never reach this point */
> + RTE_ASSERT(0);
> + if (free_space != NULL)
> + *free_space = 0;
> + return 0;
> }
>
> /**
> @@ -733,8 +755,25 @@ static __rte_always_inline unsigned int
> rte_ring_dequeue_bulk_elem(struct rte_ring *r, void *obj_table,
> unsigned int esize, unsigned int n, unsigned int *available) {
> - return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
> - RTE_RING_QUEUE_FIXED, r->cons.sync_type,
> available);
> + switch (r->cons.sync_type) {
> + case RTE_RING_SYNC_MT:
> + return rte_ring_mc_dequeue_bulk_elem(r, obj_table, esize, n,
> + available);
> + case RTE_RING_SYNC_ST:
> + return rte_ring_sc_dequeue_bulk_elem(r, obj_table, esize, n,
> + available);
> +#ifdef ALLOW_EXPERIMENTAL_API
> + case RTE_RING_SYNC_MT_RTS:
> + return rte_ring_mc_rts_dequeue_bulk_elem(r, obj_table,
> esize,
> + n, available);
> +#endif
> + }
> +
> + /* valid ring should never reach this point */
> + RTE_ASSERT(0);
> + if (available != NULL)
> + *available = 0;
> + return 0;
> }
>
> /**
> @@ -901,8 +940,25 @@ static __rte_always_inline unsigned
> rte_ring_enqueue_burst_elem(struct rte_ring *r, const void *obj_table,
> unsigned int esize, unsigned int n, unsigned int *free_space) {
> - return __rte_ring_do_enqueue_elem(r, obj_table, esize, n,
> - RTE_RING_QUEUE_VARIABLE, r->prod.sync_type,
> free_space);
> + switch (r->prod.sync_type) {
> + case RTE_RING_SYNC_MT:
> + return rte_ring_mp_enqueue_burst_elem(r, obj_table, esize, n,
> + free_space);
> + case RTE_RING_SYNC_ST:
> + return rte_ring_sp_enqueue_burst_elem(r, obj_table, esize, n,
> + free_space);
> +#ifdef ALLOW_EXPERIMENTAL_API
> + case RTE_RING_SYNC_MT_RTS:
> + return rte_ring_mp_rts_enqueue_burst_elem(r, obj_table,
> esize,
> + n, free_space);
> +#endif
> + }
> +
> + /* valid ring should never reach this point */
> + RTE_ASSERT(0);
> + if (free_space != NULL)
> + *free_space = 0;
> + return 0;
> }
>
> /**
> @@ -993,9 +1049,25 @@ static __rte_always_inline unsigned int
> rte_ring_dequeue_burst_elem(struct rte_ring *r, void *obj_table,
> unsigned int esize, unsigned int n, unsigned int *available) {
> - return __rte_ring_do_dequeue_elem(r, obj_table, esize, n,
> - RTE_RING_QUEUE_VARIABLE,
> - r->cons.sync_type, available);
> + switch (r->cons.sync_type) {
> + case RTE_RING_SYNC_MT:
> + return rte_ring_mc_dequeue_burst_elem(r, obj_table, esize, n,
> + available);
> + case RTE_RING_SYNC_ST:
> + return rte_ring_sc_dequeue_burst_elem(r, obj_table, esize, n,
> + available);
> +#ifdef ALLOW_EXPERIMENTAL_API
> + case RTE_RING_SYNC_MT_RTS:
> + return rte_ring_mc_rts_dequeue_burst_elem(r, obj_table,
> esize,
> + n, available);
> +#endif
> + }
> +
> + /* valid ring should never reach this point */
> + RTE_ASSERT(0);
> + if (available != NULL)
> + *available = 0;
> + return 0;
> }
>
> #ifdef __cplusplus
> diff --git a/lib/librte_ring/rte_ring_rts.h b/lib/librte_ring/rte_ring_rts.h new
> file mode 100644 index 000000000..18404fe48
> --- /dev/null
> +++ b/lib/librte_ring/rte_ring_rts.h
IMO, we should not provide these APIs.
> @@ -0,0 +1,316 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + *
> + * Copyright (c) 2010-2017 Intel Corporation
nit, the year should change to 2020? Look at others too.
> + * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
> + * All rights reserved.
> + * Derived from FreeBSD's bufring.h
> + * Used as BSD-3 Licensed with permission from Kip Macy.
> + */
> +
> +#ifndef _RTE_RING_RTS_H_
> +#define _RTE_RING_RTS_H_
> +
> +/**
> + * @file rte_ring_rts.h
> + * @b EXPERIMENTAL: this API may change without prior notice
> + * It is not recommended to include this file directly.
> + * Please include <rte_ring.h> instead.
> + *
> + * Contains functions for Relaxed Tail Sync (RTS) ring mode.
> + * The main idea remains the same as for our original MP/MC
^^^ the
> +synchronization
> + * mechanism.
> + * The main difference is that tail value is increased not
> + * by every thread that finished enqueue/dequeue,
> + * but only by the last one doing enqueue/dequeue.
should we say 'current last' or 'last thread at a given instance'?
> + * That allows threads to skip spinning on tail value,
> + * leaving actual tail value change to last thread in the update queue.
nit, I understand what you mean by 'update queue' here. IMO, we should remove it as it might confuse some.
> + * RTS requires 2 64-bit CAS for each enqueue(/dequeue) operation:
> + * one for head update, second for tail update.
> + * As a gain it allows thread to avoid spinning/waiting on tail value.
> + * In comparision original MP/MC algorithm requires one 32-bit CAS
> + * for head update and waiting/spinning on tail value.
> + *
> + * Brief outline:
> + * - introduce refcnt for both head and tail.
Suggesting using the same names as used in the structures.
> + * - increment head.refcnt for each head.value update
> + * - write head:value and head:refcnt atomically (64-bit CAS)
> + * - move tail.value ahead only when tail.refcnt + 1 == head.refcnt
May be add '(indicating that this is the last thread updating the tail)'
> + * - increment tail.refcnt when each enqueue/dequeue op finishes
May be add 'otherwise' at the beginning.
> + * (no matter is tail:value going to change or not)
nit ^^ if
> + * - write tail.value and tail.recnt atomically (64-bit CAS)
> + *
> + * To avoid producer/consumer starvation:
> + * - limit max allowed distance between head and tail value (HTD_MAX).
> + * I.E. thread is allowed to proceed with changing head.value,
> + * only when: head.value - tail.value <= HTD_MAX
> + * HTD_MAX is an optional parameter.
> + * With HTD_MAX == 0 we'll have fully serialized ring -
> + * i.e. only one thread at a time will be able to enqueue/dequeue
> + * to/from the ring.
> + * With HTD_MAX >= ring.capacity - no limitation.
> + * By default HTD_MAX == ring.capacity / 8.
> + */
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +#include <rte_ring_rts_generic.h>
> +
> +/**
> + * @internal Enqueue several objects on the RTS ring.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to add in the ring from the obj_table.
> + * @param behavior
> + * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
> + * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from
> ring
> + * @param free_space
> + * returns the amount of space after the enqueue operation has finished
> + * @return
> + * Actual number of objects enqueued.
> + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> + */
> +static __rte_always_inline unsigned int
> +__rte_ring_do_rts_enqueue(struct rte_ring *r, void * const *obj_table,
> + uint32_t n, enum rte_ring_queue_behavior behavior,
> + uint32_t *free_space)
> +{
> + uint32_t free, head;
> +
> + n = __rte_ring_rts_move_prod_head(r, n, behavior, &head, &free);
> +
> + if (n != 0) {
> + ENQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
> + __rte_ring_rts_update_tail(&r->rts_prod);
> + }
> +
> + if (free_space != NULL)
> + *free_space = free - n;
> + return n;
> +}
> +
> +/**
> + * @internal Dequeue several objects from the RTS ring.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to pull from the ring.
> + * @param behavior
> + * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
> + * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from
> ring
> + * @param available
> + * returns the number of remaining ring entries after the dequeue has
> finished
> + * @return
> + * - Actual number of objects dequeued.
> + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> + */
> +static __rte_always_inline unsigned int
> +__rte_ring_do_rts_dequeue(struct rte_ring *r, void **obj_table,
> + uint32_t n, enum rte_ring_queue_behavior behavior,
> + uint32_t *available)
> +{
> + uint32_t entries, head;
> +
> + n = __rte_ring_rts_move_cons_head(r, n, behavior, &head, &entries);
> +
> + if (n != 0) {
> + DEQUEUE_PTRS(r, &r[1], head, obj_table, n, void *);
> + __rte_ring_rts_update_tail(&r->rts_cons);
> + }
> +
> + if (available != NULL)
> + *available = entries - n;
> + return n;
> +}
> +
> +/**
> + * Enqueue several objects on the RTS ring (multi-producers safe).
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to add in the ring from the obj_table.
> + * @param free_space
> + * if non-NULL, returns the amount of space in the ring after the
> + * enqueue operation has finished.
> + * @return
> + * The number of objects enqueued, either 0 or n
> + */
> +__rte_experimental
> +static __rte_always_inline unsigned int
> +rte_ring_mp_rts_enqueue_bulk(struct rte_ring *r, void * const *obj_table,
> + unsigned int n, unsigned int *free_space) {
> + return __rte_ring_do_rts_enqueue(r, obj_table, n,
> RTE_RING_QUEUE_FIXED,
> + free_space);
> +}
> +
> +/**
> + * Dequeue several objects from an RTS ring (multi-consumers safe).
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects) that will be filled.
> + * @param n
> + * The number of objects to dequeue from the ring to the obj_table.
> + * @param available
> + * If non-NULL, returns the number of remaining ring entries after the
> + * dequeue has finished.
> + * @return
> + * The number of objects dequeued, either 0 or n
> + */
> +__rte_experimental
> +static __rte_always_inline unsigned int
> +rte_ring_mc_rts_dequeue_bulk(struct rte_ring *r, void **obj_table,
> + unsigned int n, unsigned int *available) {
> + return __rte_ring_do_rts_dequeue(r, obj_table, n,
> RTE_RING_QUEUE_FIXED,
> + available);
> +}
> +
> +/**
> + * Return producer max Head-Tail-Distance (HTD).
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @return
> + * Producer HTD value, if producer is set in appropriate sync mode,
> + * or UINT32_MAX otherwise.
> + */
> +__rte_experimental
> +static inline uint32_t
> +rte_ring_get_prod_htd_max(const struct rte_ring *r) {
> + if (r->prod.sync_type == RTE_RING_SYNC_MT_RTS)
> + return r->rts_prod.htd_max;
> + return UINT32_MAX;
> +}
> +
> +/**
> + * Set producer max Head-Tail-Distance (HTD).
> + * Note that producer has to use appropriate sync mode (RTS).
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param v
> + * new HTD value to setup.
> + * @return
> + * Zero on success, or negative error code otherwise.
> + */
> +__rte_experimental
> +static inline int
> +rte_ring_set_prod_htd_max(struct rte_ring *r, uint32_t v) {
> + if (r->prod.sync_type != RTE_RING_SYNC_MT_RTS)
> + return -ENOTSUP;
> +
> + r->rts_prod.htd_max = v;
> + return 0;
> +}
> +
> +/**
> + * Return consumer max Head-Tail-Distance (HTD).
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @return
> + * Consumer HTD value, if consumer is set in appropriate sync mode,
> + * or UINT32_MAX otherwise.
> + */
> +__rte_experimental
> +static inline uint32_t
> +rte_ring_get_cons_htd_max(const struct rte_ring *r) {
> + if (r->cons.sync_type == RTE_RING_SYNC_MT_RTS)
> + return r->rts_cons.htd_max;
> + return UINT32_MAX;
> +}
> +
> +/**
> + * Set consumer max Head-Tail-Distance (HTD).
> + * Note that consumer has to use appropriate sync mode (RTS).
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param v
> + * new HTD value to setup.
> + * @return
> + * Zero on success, or negative error code otherwise.
> + */
> +__rte_experimental
> +static inline int
> +rte_ring_set_cons_htd_max(struct rte_ring *r, uint32_t v) {
> + if (r->cons.sync_type != RTE_RING_SYNC_MT_RTS)
> + return -ENOTSUP;
> +
> + r->rts_cons.htd_max = v;
> + return 0;
> +}
> +
> +/**
> + * Enqueue several objects on the RTS ring (multi-producers safe).
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to add in the ring from the obj_table.
> + * @param free_space
> + * if non-NULL, returns the amount of space in the ring after the
> + * enqueue operation has finished.
> + * @return
> + * - n: Actual number of objects enqueued.
> + */
> +__rte_experimental
> +static __rte_always_inline unsigned
> +rte_ring_mp_rts_enqueue_burst(struct rte_ring *r, void * const *obj_table,
> + unsigned int n, unsigned int *free_space) {
> + return __rte_ring_do_rts_enqueue(r, obj_table, n,
> + RTE_RING_QUEUE_VARIABLE, free_space); }
> +
> +/**
> + * Dequeue several objects from an RTS ring (multi-consumers safe).
> + * When the requested objects are more than the available objects,
> + * only dequeue the actual number of objects.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects) that will be filled.
> + * @param n
> + * The number of objects to dequeue from the ring to the obj_table.
> + * @param available
> + * If non-NULL, returns the number of remaining ring entries after the
> + * dequeue has finished.
> + * @return
> + * - n: Actual number of objects dequeued, 0 if ring is empty
> + */
> +__rte_experimental
> +static __rte_always_inline unsigned
> +rte_ring_mc_rts_dequeue_burst(struct rte_ring *r, void **obj_table,
> + unsigned int n, unsigned int *available) {
> + return __rte_ring_do_rts_dequeue(r, obj_table, n,
> + RTE_RING_QUEUE_VARIABLE, available); }
> +
> +#ifdef __cplusplus
> +}
> +#endif
> +
> +#endif /* _RTE_RING_RTS_H_ */
> diff --git a/lib/librte_ring/rte_ring_rts_elem.h
> b/lib/librte_ring/rte_ring_rts_elem.h
> new file mode 100644
> index 000000000..71a331b23
> --- /dev/null
> +++ b/lib/librte_ring/rte_ring_rts_elem.h
> @@ -0,0 +1,205 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + *
> + * Copyright (c) 2010-2017 Intel Corporation
> + * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
> + * All rights reserved.
> + * Derived from FreeBSD's bufring.h
> + * Used as BSD-3 Licensed with permission from Kip Macy.
> + */
> +
> +#ifndef _RTE_RING_RTS_ELEM_H_
> +#define _RTE_RING_RTS_ELEM_H_
> +
> +/**
> + * @file rte_ring_rts_elem.h
> + * @b EXPERIMENTAL: this API may change without prior notice
> + *
> + * It is not recommended to include this file directly.
> + * Please include <rte_ring_elem.h> instead.
> + * Contains *ring_elem* functions for Relaxed Tail Sync (RTS) ring mode.
> + * for more details please refer to <rte_ring_rts.h>.
> + */
> +
> +#ifdef __cplusplus
> +extern "C" {
> +#endif
> +
> +#include <rte_ring_rts_generic.h>
> +
> +/**
> + * @internal Enqueue several objects on the RTS ring.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to add in the ring from the obj_table.
> + * @param behavior
> + * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
> + * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from
> ring
> + * @param free_space
> + * returns the amount of space after the enqueue operation has finished
> + * @return
> + * Actual number of objects enqueued.
> + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> + */
> +static __rte_always_inline unsigned int
> +__rte_ring_do_rts_enqueue_elem(struct rte_ring *r, void * const *obj_table,
obj_table should be of type 'const void * obj_table' (looks like copy paste error). Please check the other APIs below too.
> + uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
'esize' is not documented in the comments above the function. You can copy the header from rte_ring_elem.h file. Please check other APIs as well.
> + uint32_t *free_space)
> +{
> + uint32_t free, head;
> +
> + n = __rte_ring_rts_move_prod_head(r, n, behavior, &head, &free);
> +
> + if (n != 0) {
> + __rte_ring_enqueue_elems(r, head, obj_table, esize, n);
> + __rte_ring_rts_update_tail(&r->rts_prod);
> + }
> +
> + if (free_space != NULL)
> + *free_space = free - n;
> + return n;
> +}
> +
> +/**
> + * @internal Dequeue several objects from the RTS ring.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to pull from the ring.
> + * @param behavior
> + * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
> + * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from
> ring
> + * @param available
> + * returns the number of remaining ring entries after the dequeue has
> finished
> + * @return
> + * - Actual number of objects dequeued.
> + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> + */
> +static __rte_always_inline unsigned int
> +__rte_ring_do_rts_dequeue_elem(struct rte_ring *r, void **obj_table,
> + uint32_t esize, uint32_t n, enum rte_ring_queue_behavior behavior,
> + uint32_t *available)
> +{
> + uint32_t entries, head;
> +
> + n = __rte_ring_rts_move_cons_head(r, n, behavior, &head, &entries);
> +
> + if (n != 0) {
> + __rte_ring_dequeue_elems(r, head, obj_table, esize, n);
> + __rte_ring_rts_update_tail(&r->rts_cons);
> + }
> +
> + if (available != NULL)
> + *available = entries - n;
> + return n;
> +}
> +
> +/**
> + * Enqueue several objects on the RTS ring (multi-producers safe).
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to add in the ring from the obj_table.
> + * @param free_space
> + * if non-NULL, returns the amount of space in the ring after the
> + * enqueue operation has finished.
> + * @return
> + * The number of objects enqueued, either 0 or n
> + */
> +__rte_experimental
> +static __rte_always_inline unsigned int
> +rte_ring_mp_rts_enqueue_bulk_elem(struct rte_ring *r, void * const
> *obj_table,
> + unsigned int esize, unsigned int n, unsigned int *free_space) {
> + return __rte_ring_do_rts_enqueue_elem(r, obj_table, esize, n,
> + RTE_RING_QUEUE_FIXED, free_space);
> +}
> +
> +/**
> + * Dequeue several objects from an RTS ring (multi-consumers safe).
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects) that will be filled.
> + * @param n
> + * The number of objects to dequeue from the ring to the obj_table.
> + * @param available
> + * If non-NULL, returns the number of remaining ring entries after the
> + * dequeue has finished.
> + * @return
> + * The number of objects dequeued, either 0 or n
> + */
> +__rte_experimental
> +static __rte_always_inline unsigned int
> +rte_ring_mc_rts_dequeue_bulk_elem(struct rte_ring *r, void **obj_table,
> + unsigned int esize, unsigned int n, unsigned int *available) {
> + return __rte_ring_do_rts_dequeue_elem(r, obj_table, esize, n,
> + RTE_RING_QUEUE_FIXED, available);
> +}
> +
> +/**
> + * Enqueue several objects on the RTS ring (multi-producers safe).
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects).
> + * @param n
> + * The number of objects to add in the ring from the obj_table.
> + * @param free_space
> + * if non-NULL, returns the amount of space in the ring after the
> + * enqueue operation has finished.
> + * @return
> + * - n: Actual number of objects enqueued.
> + */
> +__rte_experimental
> +static __rte_always_inline unsigned
> +rte_ring_mp_rts_enqueue_burst_elem(struct rte_ring *r, void * const
> *obj_table,
> + unsigned int esize, unsigned int n, unsigned int *free_space) {
> + return __rte_ring_do_rts_enqueue_elem(r, obj_table, esize, n,
> + RTE_RING_QUEUE_VARIABLE, free_space); }
> +
> +/**
> + * Dequeue several objects from an RTS ring (multi-consumers safe).
> + * When the requested objects are more than the available objects,
> + * only dequeue the actual number of objects.
> + *
> + * @param r
> + * A pointer to the ring structure.
> + * @param obj_table
> + * A pointer to a table of void * pointers (objects) that will be filled.
> + * @param n
> + * The number of objects to dequeue from the ring to the obj_table.
> + * @param available
> + * If non-NULL, returns the number of remaining ring entries after the
> + * dequeue has finished.
> + * @return
> + * - n: Actual number of objects dequeued, 0 if ring is empty
> + */
> +__rte_experimental
> +static __rte_always_inline unsigned
> +rte_ring_mc_rts_dequeue_burst_elem(struct rte_ring *r, void **obj_table,
> + unsigned int esize, unsigned int n, unsigned int *available) {
> + return __rte_ring_do_rts_dequeue_elem(r, obj_table, esize, n,
> + RTE_RING_QUEUE_VARIABLE, available); }
> +
> +#ifdef __cplusplus
> +}
> +#endif
> +
> +#endif /* _RTE_RING_RTS_ELEM_H_ */
> diff --git a/lib/librte_ring/rte_ring_rts_generic.h
> b/lib/librte_ring/rte_ring_rts_generic.h
> new file mode 100644
> index 000000000..f88460d47
> --- /dev/null
> +++ b/lib/librte_ring/rte_ring_rts_generic.h
I do not know the benefit to providing the generic version. Do you know why this was done in the legacy APIs?
If there is no performance difference between generic and C11 versions, should we just skip the generic version?
The oldest compiler in CI are GCC 4.8.5 and Clang 3.4.2 and C11 built-ins are supported earlier than these compiler versions.
I feel the code is growing exponentially in rte_ring library and we should try to cut non-value-ass code/APIs aggressively.
> @@ -0,0 +1,210 @@
> +/* SPDX-License-Identifier: BSD-3-Clause
> + *
> + * Copyright (c) 2010-2017 Intel Corporation
> + * Copyright (c) 2007-2009 Kip Macy kmacy@freebsd.org
> + * All rights reserved.
> + * Derived from FreeBSD's bufring.h
> + * Used as BSD-3 Licensed with permission from Kip Macy.
> + */
> +
> +#ifndef _RTE_RING_RTS_GENERIC_H_
> +#define _RTE_RING_RTS_GENERIC_H_
> +
> +/**
> + * @file rte_ring_rts_generic.h
> + * It is not recommended to include this file directly,
> + * include <rte_ring.h> instead.
> + * Contains internal helper functions for Relaxed Tail Sync (RTS) ring mode.
> + * For more information please refer to <rte_ring_rts.h>.
> + */
> +
> +/**
> + * @internal This function updates tail values.
> + */
> +static __rte_always_inline void
> +__rte_ring_rts_update_tail(struct rte_ring_rts_headtail *ht) {
> + union rte_ring_ht_poscnt h, ot, nt;
> +
> + /*
> + * If there are other enqueues/dequeues in progress that
> + * might preceded us, then don't update tail with new value.
> + */
> +
> + do {
> + ot.raw = ht->tail.raw;
> + rte_smp_rmb();
> +
> + /* on 32-bit systems we have to do atomic read here */
> + h.raw = rte_atomic64_read((rte_atomic64_t *)
> + (uintptr_t)&ht->head.raw);
> +
> + nt.raw = ot.raw;
> + if (++nt.val.cnt == h.val.cnt)
> + nt.val.pos = h.val.pos;
> +
> + } while (rte_atomic64_cmpset(&ht->tail.raw, ot.raw, nt.raw) == 0); }
> +
> +/**
> + * @internal This function waits till head/tail distance wouldn't
> + * exceed pre-defined max value.
> + */
> +static __rte_always_inline void
> +__rte_ring_rts_head_wait(const struct rte_ring_rts_headtail *ht,
> + union rte_ring_ht_poscnt *h)
> +{
> + uint32_t max;
> +
> + max = ht->htd_max;
> + h->raw = ht->head.raw;
> + rte_smp_rmb();
> +
> + while (h->val.pos - ht->tail.val.pos > max) {
> + rte_pause();
> + h->raw = ht->head.raw;
> + rte_smp_rmb();
> + }
> +}
> +
> +/**
> + * @internal This function updates the producer head for enqueue.
> + *
> + * @param r
> + * A pointer to the ring structure
> + * @param is_sp
> + * Indicates whether multi-producer path is needed or not
> + * @param n
> + * The number of elements we will want to enqueue, i.e. how far should the
> + * head be moved
> + * @param behavior
> + * RTE_RING_QUEUE_FIXED: Enqueue a fixed number of items from a ring
> + * RTE_RING_QUEUE_VARIABLE: Enqueue as many items as possible from
> ring
> + * @param old_head
> + * Returns head value as it was before the move, i.e. where enqueue starts
> + * @param new_head
> + * Returns the current/new head value i.e. where enqueue finishes
> + * @param free_entries
> + * Returns the amount of free space in the ring BEFORE head was moved
> + * @return
> + * Actual number of objects enqueued.
> + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> + */
> +static __rte_always_inline uint32_t
> +__rte_ring_rts_move_prod_head(struct rte_ring *r, uint32_t num,
> + enum rte_ring_queue_behavior behavior, uint32_t *old_head,
> + uint32_t *free_entries)
> +{
> + uint32_t n;
> + union rte_ring_ht_poscnt nh, oh;
> +
> + const uint32_t capacity = r->capacity;
> +
> + do {
> + /* Reset n to the initial burst count */
> + n = num;
> +
> + /* read prod head (may spin on prod tail) */
> + __rte_ring_rts_head_wait(&r->rts_prod, &oh);
> +
> + /* add rmb barrier to avoid load/load reorder in weak
> + * memory model. It is noop on x86
> + */
> + rte_smp_rmb();
> +
> + /*
> + * The subtraction is done between two unsigned 32bits value
> + * (the result is always modulo 32 bits even if we have
> + * *old_head > cons_tail). So 'free_entries' is always between
> 0
> + * and capacity (which is < size).
> + */
> + *free_entries = capacity + r->cons.tail - oh.val.pos;
> +
> + /* check that we have enough room in ring */
> + if (unlikely(n > *free_entries))
> + n = (behavior == RTE_RING_QUEUE_FIXED) ?
> + 0 : *free_entries;
> +
> + if (n == 0)
> + break;
> +
> + nh.val.pos = oh.val.pos + n;
> + nh.val.cnt = oh.val.cnt + 1;
> +
> + } while (rte_atomic64_cmpset(&r->rts_prod.head.raw,
> + oh.raw, nh.raw) == 0);
> +
> + *old_head = oh.val.pos;
> + return n;
> +}
> +
> +/**
> + * @internal This function updates the consumer head for dequeue
> + *
> + * @param r
> + * A pointer to the ring structure
> + * @param is_sc
> + * Indicates whether multi-consumer path is needed or not
> + * @param n
> + * The number of elements we will want to enqueue, i.e. how far should the
> + * head be moved
> + * @param behavior
> + * RTE_RING_QUEUE_FIXED: Dequeue a fixed number of items from a ring
> + * RTE_RING_QUEUE_VARIABLE: Dequeue as many items as possible from
> ring
> + * @param old_head
> + * Returns head value as it was before the move, i.e. where dequeue starts
> + * @param new_head
> + * Returns the current/new head value i.e. where dequeue finishes
> + * @param entries
> + * Returns the number of entries in the ring BEFORE head was moved
> + * @return
> + * - Actual number of objects dequeued.
> + * If behavior == RTE_RING_QUEUE_FIXED, this will be 0 or n only.
> + */
> +static __rte_always_inline unsigned int
> +__rte_ring_rts_move_cons_head(struct rte_ring *r, uint32_t num,
> + enum rte_ring_queue_behavior behavior, uint32_t *old_head,
> + uint32_t *entries)
> +{
> + uint32_t n;
> + union rte_ring_ht_poscnt nh, oh;
> +
> + /* move cons.head atomically */
> + do {
> + /* Restore n as it may change every loop */
> + n = num;
> +
> + /* read cons head (may spin on cons tail) */
> + __rte_ring_rts_head_wait(&r->rts_cons, &oh);
> +
> +
> + /* add rmb barrier to avoid load/load reorder in weak
> + * memory model. It is noop on x86
> + */
> + rte_smp_rmb();
> +
> + /* The subtraction is done between two unsigned 32bits value
> + * (the result is always modulo 32 bits even if we have
> + * cons_head > prod_tail). So 'entries' is always between 0
> + * and size(ring)-1.
> + */
> + *entries = r->prod.tail - oh.val.pos;
> +
> + /* Set the actual entries for dequeue */
> + if (n > *entries)
> + n = (behavior == RTE_RING_QUEUE_FIXED) ? 0 :
> *entries;
> +
> + if (unlikely(n == 0))
> + break;
> +
> + nh.val.pos = oh.val.pos + n;
> + nh.val.cnt = oh.val.cnt + 1;
> +
> + } while (rte_atomic64_cmpset(&r->rts_cons.head.raw,
> + oh.raw, nh.raw) == 0);
> +
> + *old_head = oh.val.pos;
> + return n;
> +}
> +
> +#endif /* _RTE_RING_RTS_GENERIC_H_ */
> --
> 2.17.1
^ permalink raw reply [flat|nested] 146+ messages in thread
* Re: [dpdk-dev] [PATCH v3 1/9] test/ring: add contention stress test
2020-04-08 4:59 ` Honnappa Nagarahalli
@ 2020-04-09 12:36 ` Ananyev, Konstantin
2020-04-09 13:00 ` Ananyev, Konstantin
2020-04-10 16:59 ` Honnappa Nagarahalli
0 siblings, 2 replies; 146+ messages in thread
From: Ananyev, Konstantin @ 2020-04-09 12:36 UTC (permalink / raw)
To: Honnappa Nagarahalli, dev; +Cc: david.marchand, jielong.zjl, nd, nd
> <snip>
>
> > Subject: [PATCH v3 1/9] test/ring: add contention stress test
> Minor, would 'add stress test for overcommitted use case' sound better?
I liked to point out that this test-case can be used as contention stress-test
(many threads do enqueue/dequeue to/from the same ring) for both
over-committed and not scenarios...
Will probably try to add few extra explanations in v4.
> >
> > Introduce new test-case to measure ring perfomance under contention
> Minor, 'over committed' seems to the word commonly used from the references you provided. Does it make sense to use that?
>
> > (miltiple producers/consumers).
> ^^^^^^^ multiple
ack.
>
> > Starts dequeue/enqueue loop on all available slave lcores.
> >
> > Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> > ---
> > app/test/Makefile | 2 +
> > app/test/meson.build | 2 +
> > app/test/test_ring_mpmc_stress.c | 31 +++
> > app/test/test_ring_stress.c | 48 ++++
> > app/test/test_ring_stress.h | 35 +++
> > app/test/test_ring_stress_impl.h | 444 +++++++++++++++++++++++++++++++
> Would be good to change the file names to indicate that these tests are for over-committed usecase/configuration.
> These are performance tests, better to have 'perf' or 'performance' in their names.
>
> > 6 files changed, 562 insertions(+)
> > create mode 100644 app/test/test_ring_mpmc_stress.c create mode 100644
> > app/test/test_ring_stress.c create mode 100644 app/test/test_ring_stress.h
> > create mode 100644 app/test/test_ring_stress_impl.h
> >
> > diff --git a/app/test/Makefile b/app/test/Makefile index
> > 1f080d162..4eefaa887 100644
> > --- a/app/test/Makefile
> > +++ b/app/test/Makefile
> > @@ -77,7 +77,9 @@ SRCS-y += test_external_mem.c SRCS-y +=
> > test_rand_perf.c
> >
> > SRCS-y += test_ring.c
> > +SRCS-y += test_ring_mpmc_stress.c
> > SRCS-y += test_ring_perf.c
> > +SRCS-y += test_ring_stress.c
> > SRCS-y += test_pmd_perf.c
> >
> > ifeq ($(CONFIG_RTE_LIBRTE_TABLE),y)
> > diff --git a/app/test/meson.build b/app/test/meson.build index
> > 351d29cb6..827b04886 100644
> > --- a/app/test/meson.build
> > +++ b/app/test/meson.build
> > @@ -100,7 +100,9 @@ test_sources = files('commands.c',
> > 'test_rib.c',
> > 'test_rib6.c',
> > 'test_ring.c',
> > + 'test_ring_mpmc_stress.c',
> > 'test_ring_perf.c',
> > + 'test_ring_stress.c',
> > 'test_rwlock.c',
> > 'test_sched.c',
> > 'test_service_cores.c',
> > diff --git a/app/test/test_ring_mpmc_stress.c
> > b/app/test/test_ring_mpmc_stress.c
> > new file mode 100644
> > index 000000000..1524b0248
> > --- /dev/null
> > +++ b/app/test/test_ring_mpmc_stress.c
> > @@ -0,0 +1,31 @@
> > +/* SPDX-License-Identifier: BSD-3-Clause
> > + * Copyright(c) 2020 Intel Corporation
> > + */
> > +
> > +#include "test_ring_stress_impl.h"
> > +
> > +static inline uint32_t
> > +_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
> > + uint32_t *avail)
> > +{
> > + return rte_ring_mc_dequeue_bulk(r, obj, n, avail); }
> > +
> > +static inline uint32_t
> > +_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
> > + uint32_t *free)
> > +{
> > + return rte_ring_mp_enqueue_bulk(r, obj, n, free); }
> > +
> > +static int
> > +_st_ring_init(struct rte_ring *r, const char *name, uint32_t num) {
> > + return rte_ring_init(r, name, num, 0); }
> > +
> > +const struct test test_ring_mpmc_stress = {
> > + .name = "MP/MC",
> > + .nb_case = RTE_DIM(tests),
> > + .cases = tests,
> > +};
> > diff --git a/app/test/test_ring_stress.c b/app/test/test_ring_stress.c new file
> > mode 100644 index 000000000..60706f799
> > --- /dev/null
> > +++ b/app/test/test_ring_stress.c
> > @@ -0,0 +1,48 @@
> > +/* SPDX-License-Identifier: BSD-3-Clause
> > + * Copyright(c) 2020 Intel Corporation
> > + */
> > +
> > +#include "test_ring_stress.h"
> > +
> > +static int
> > +run_test(const struct test *test)
> > +{
> > + int32_t rc;
> > + uint32_t i, k;
> > +
> > + for (i = 0, k = 0; i != test->nb_case; i++) {
> > +
> > + printf("TEST-CASE %s %s START\n",
> > + test->name, test->cases[i].name);
> > +
> > + rc = test->cases[i].func(test->cases[i].wfunc);
> > + k += (rc == 0);
> > +
> > + if (rc != 0)
> > + printf("TEST-CASE %s %s FAILED\n",
> > + test->name, test->cases[i].name);
> > + else
> > + printf("TEST-CASE %s %s OK\n",
> > + test->name, test->cases[i].name);
> > + }
> > +
> > + return k;
> > +}
> > +
> > +static int
> > +test_ring_stress(void)
> > +{
> > + uint32_t n, k;
> > +
> > + n = 0;
> > + k = 0;
> > +
> > + n += test_ring_mpmc_stress.nb_case;
> > + k += run_test(&test_ring_mpmc_stress);
> > +
> > + printf("Number of tests:\t%u\nSuccess:\t%u\nFailed:\t%u\n",
> > + n, k, n - k);
> > + return (k != n);
> > +}
> > +
> > +REGISTER_TEST_COMMAND(ring_stress_autotest, test_ring_stress);
> > diff --git a/app/test/test_ring_stress.h b/app/test/test_ring_stress.h new file
> > mode 100644 index 000000000..60eac6216
> > --- /dev/null
> > +++ b/app/test/test_ring_stress.h
> > @@ -0,0 +1,35 @@
> > +/* SPDX-License-Identifier: BSD-3-Clause
> > + * Copyright(c) 2020 Intel Corporation
> > + */
> > +
> > +
> > +#include <inttypes.h>
> > +#include <stddef.h>
> > +#include <stdalign.h>
> > +#include <string.h>
> > +#include <stdio.h>
> > +#include <unistd.h>
> > +
> > +#include <rte_ring.h>
> > +#include <rte_cycles.h>
> > +#include <rte_launch.h>
> > +#include <rte_pause.h>
> > +#include <rte_random.h>
> > +#include <rte_malloc.h>
> > +#include <rte_spinlock.h>
> > +
> > +#include "test.h"
> > +
> > +struct test_case {
> > + const char *name;
> > + int (*func)(int (*)(void *));
> > + int (*wfunc)(void *arg);
> > +};
> > +
> > +struct test {
> > + const char *name;
> > + uint32_t nb_case;
> > + const struct test_case *cases;
> > +};
> > +
> > +extern const struct test test_ring_mpmc_stress;
> > diff --git a/app/test/test_ring_stress_impl.h
> > b/app/test/test_ring_stress_impl.h
> > new file mode 100644
> > index 000000000..11476d28c
> > --- /dev/null
> > +++ b/app/test/test_ring_stress_impl.h
> > @@ -0,0 +1,444 @@
> > +/* SPDX-License-Identifier: BSD-3-Clause
> > + * Copyright(c) 2020 Intel Corporation
> > + */
> > +
> > +#include "test_ring_stress.h"
> > +
> > +/*
> > + * Measures performance of ring enqueue/dequeue under high contention
> > +*/
> > +
> > +#define RING_NAME "RING_STRESS"
> > +#define BULK_NUM 32
> > +#define RING_SIZE (2 * BULK_NUM * RTE_MAX_LCORE)
> > +
> > +enum {
> > + WRK_CMD_STOP,
> > + WRK_CMD_RUN,
> > +};
> > +
> > +static volatile uint32_t wrk_cmd __rte_cache_aligned;
> > +
> > +/* test run-time in seconds */
> > +static const uint32_t run_time = 60;
> > +static const uint32_t verbose;
> > +
> > +struct lcore_stat {
> > + uint64_t nb_cycle;
> > + struct {
> > + uint64_t nb_call;
> > + uint64_t nb_obj;
> > + uint64_t nb_cycle;
> > + uint64_t max_cycle;
> > + uint64_t min_cycle;
> > + } op;
> > +};
> > +
> > +struct lcore_arg {
> > + struct rte_ring *rng;
> > + struct lcore_stat stats;
> > +} __rte_cache_aligned;
> > +
> > +struct ring_elem {
> > + uint32_t cnt[RTE_CACHE_LINE_SIZE / sizeof(uint32_t)]; }
> > +__rte_cache_aligned;
> > +
> > +/*
> > + * redefinable functions
> > + */
> > +static uint32_t
> > +_st_ring_dequeue_bulk(struct rte_ring *r, void **obj, uint32_t n,
> > + uint32_t *avail);
> > +
> > +static uint32_t
> > +_st_ring_enqueue_bulk(struct rte_ring *r, void * const *obj, uint32_t n,
> > + uint32_t *free);
> > +
> > +static int
> > +_st_ring_init(struct rte_ring *r, const char *name, uint32_t num);
> > +
> > +
> > +static void
> > +lcore_stat_update(struct lcore_stat *ls, uint64_t call, uint64_t obj,
> > + uint64_t tm, int32_t prcs)
> > +{
> > + ls->op.nb_call += call;
> > + ls->op.nb_obj += obj;
> > + ls->op.nb_cycle += tm;
> > + if (prcs) {
> > + ls->op.max_cycle = RTE_MAX(ls->op.max_cycle, tm);
> > + ls->op.min_cycle = RTE_MIN(ls->op.min_cycle, tm);
> > + }
> > +}
> > +
> > +static void
> > +lcore_op_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls)
> > +{
> > +
> > + ms->op.nb_call += ls->op.nb_call;
> > + ms->op.nb_obj += ls->op.nb_obj;
> > + ms->op.nb_cycle += ls->op.nb_cycle;
> > + ms->op.max_cycle = RTE_MAX(ms->op.max_cycle, ls->op.max_cycle);
> > + ms->op.min_cycle = RTE_MIN(ms->op.min_cycle, ls->op.min_cycle); }
> > +
> > +static void
> > +lcore_stat_aggr(struct lcore_stat *ms, const struct lcore_stat *ls) {
> > + ms->nb_cycle = RTE_MAX(ms->nb_cycle, ls->nb_cycle);
> > + lcore_op_stat_aggr(ms, ls);
> > +}
> > +
> > +static void
> > +lcore_stat_dump(FILE *f, uint32_t lc, const struct lcore_stat *ls) {
> > + long double st;
> > +
> > + st = (long double)rte_get_timer_hz() / US_PER_S;
> > +
> > + if (lc == UINT32_MAX)
> > + fprintf(f, "%s(AGGREGATE)={\n", __func__);
> > + else
> > + fprintf(f, "%s(lc=%u)={\n", __func__, lc);
> > +
> > + fprintf(f, "\tnb_cycle=%" PRIu64 "(%.2Lf usec),\n",
> > + ls->nb_cycle, (long double)ls->nb_cycle / st);
> > +
> > + fprintf(f, "\tDEQ+ENQ={\n");
> > +
> > + fprintf(f, "\t\tnb_call=%" PRIu64 ",\n", ls->op.nb_call);
> > + fprintf(f, "\t\tnb_obj=%" PRIu64 ",\n", ls->op.nb_obj);
> > + fprintf(f, "\t\tnb_cycle=%" PRIu64 ",\n", ls->op.nb_cycle);
> > + fprintf(f, "\t\tobj/call(avg): %.2Lf\n",
> > + (long double)ls->op.nb_obj / ls->op.nb_call);
> > + fprintf(f, "\t\tcycles/obj(avg): %.2Lf\n",
> > + (long double)ls->op.nb_cycle / ls->op.nb_obj);
> > + fprintf(f, "\t\tcycles/call(avg): %.2Lf\n",
> > + (long double)ls->op.nb_cycle / ls->op.nb_call);
> > +
> > + /* if min/max cycles per call stats was collected */
> > + if (ls->op.min_cycle != UINT64_MAX) {
> > + fprintf(f, "\t\tmax cycles/call=%" PRIu64 "(%.2Lf usec),\n",
> > + ls->op.max_cycle,
> > + (long double)ls->op.max_cycle / st);
> > + fprintf(f, "\t\tmin cycles/call=%" PRIu64 "(%.2Lf usec),\n",
> > + ls->op.min_cycle,
> > + (long double)ls->op.min_cycle / st);
> > + }
> > +
> > + fprintf(f, "\t},\n");
> > + fprintf(f, "};\n");
> > +}
> > +
> > +static void
> > +fill_ring_elm(struct ring_elem *elm, uint32_t fill) {
> > + uint32_t i;
> > +
> > + for (i = 0; i != RTE_DIM(elm->cnt); i++)
> > + elm->cnt[i] = fill;
> > +}
> > +
> > +static int32_t
> > +check_updt_elem(struct ring_elem *elm[], uint32_t num,
> > + const struct ring_elem *check, const struct ring_elem *fill) {
> > + uint32_t i;
> > +
> > + static rte_spinlock_t dump_lock;
> > +
> > + for (i = 0; i != num; i++) {
> > + if (memcmp(check, elm[i], sizeof(*check)) != 0) {
> > + rte_spinlock_lock(&dump_lock);
> > + printf("%s(lc=%u, num=%u) failed at %u-th iter, "
> > + "offending object: %p\n",
> > + __func__, rte_lcore_id(), num, i, elm[i]);
> > + rte_memdump(stdout, "expected", check,
> > sizeof(*check));
> > + rte_memdump(stdout, "result", elm[i], sizeof(elm[i]));
> > + rte_spinlock_unlock(&dump_lock);
> > + return -EINVAL;
> > + }
> > + memcpy(elm[i], fill, sizeof(*elm[i]));
> > + }
> > +
> > + return 0;
> > +}
> > +
> > +static int
> > +check_ring_op(uint32_t exp, uint32_t res, uint32_t lc,
> minor, lcore instead of lc would be better
>
> > + const char *fname, const char *opname) {
> > + if (exp != res) {
> > + printf("%s(lc=%u) failure: %s expected: %u, returned %u\n",
> Suggest using lcore in the printf
>
> > + fname, lc, opname, exp, res);
> > + return -ENOSPC;
> > + }
> > + return 0;
> > +}
> > +
> > +static int
> > +test_worker_prcs(void *arg)
> > +{
> > + int32_t rc;
> > + uint32_t lc, n, num;
> minor, lcore instead of lc would be better
>
> > + uint64_t cl, tm0, tm1;
> > + struct lcore_arg *la;
> > + struct ring_elem def_elm, loc_elm;
> > + struct ring_elem *obj[2 * BULK_NUM];
> > +
> > + la = arg;
> > + lc = rte_lcore_id();
> > +
> > + fill_ring_elm(&def_elm, UINT32_MAX);
> > + fill_ring_elm(&loc_elm, lc);
> > +
> > + while (wrk_cmd != WRK_CMD_RUN) {
> > + rte_smp_rmb();
> > + rte_pause();
> > + }
> > +
> > + cl = rte_rdtsc_precise();
> > +
> > + do {
> > + /* num in interval [7/8, 11/8] of BULK_NUM */
> > + num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
> > +
> > + /* reset all pointer values */
> > + memset(obj, 0, sizeof(obj));
> > +
> > + /* dequeue num elems */
> > + tm0 = rte_rdtsc_precise();
> > + n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
> > + tm0 = rte_rdtsc_precise() - tm0;
> > +
> > + /* check return value and objects */
> > + rc = check_ring_op(num, n, lc, __func__,
> > + RTE_STR(_st_ring_dequeue_bulk));
> > + if (rc == 0)
> > + rc = check_updt_elem(obj, num, &def_elm,
> > &loc_elm);
> > + if (rc != 0)
> > + break;
> Since this seems like a performance test, should we skip validating the objects?
> Did these tests run on Travis CI? I believe Travis CI has trouble running stress/performance tests if they take too much time.
> The RTS and HTS tests should be added to functional tests.
>
> > +
> > + /* enqueue num elems */
> > + rte_compiler_barrier();
> > + rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
> > + if (rc != 0)
> > + break;
> > +
> > + tm1 = rte_rdtsc_precise();
> > + n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
> > + tm1 = rte_rdtsc_precise() - tm1;
> > +
> > + /* check return value */
> > + rc = check_ring_op(num, n, lc, __func__,
> > + RTE_STR(_st_ring_enqueue_bulk));
> > + if (rc != 0)
> > + break;
> > +
> > + lcore_stat_update(&la->stats, 1, num, tm0 + tm1, 1);
> > +
> > + } while (wrk_cmd == WRK_CMD_RUN);
> > +
> > + la->stats.nb_cycle = rte_rdtsc_precise() - cl;
> > + return rc;
> > +}
> > +
> > +static int
> > +test_worker_avg(void *arg)
> > +{
> > + int32_t rc;
> > + uint32_t lc, n, num;
> > + uint64_t cl;
> > + struct lcore_arg *la;
> > + struct ring_elem def_elm, loc_elm;
> > + struct ring_elem *obj[2 * BULK_NUM];
> > +
> > + la = arg;
> > + lc = rte_lcore_id();
> > +
> > + fill_ring_elm(&def_elm, UINT32_MAX);
> > + fill_ring_elm(&loc_elm, lc);
> > +
> > + while (wrk_cmd != WRK_CMD_RUN) {
> > + rte_smp_rmb();
> > + rte_pause();
> > + }
> > +
> > + cl = rte_rdtsc_precise();
> > +
> > + do {
> > + /* num in interval [7/8, 11/8] of BULK_NUM */
> > + num = 7 * BULK_NUM / 8 + rte_rand() % (BULK_NUM / 2);
> > +
> > + /* reset all pointer values */
> > + memset(obj, 0, sizeof(obj));
> > +
> > + /* dequeue num elems */
> > + n = _st_ring_dequeue_bulk(la->rng, (void **)obj, num, NULL);
> > +
> > + /* check return value and objects */
> > + rc = check_ring_op(num, n, lc, __func__,
> > + RTE_STR(_st_ring_dequeue_bulk));
> > + if (rc == 0)
> > + rc = check_updt_elem(obj, num, &def_elm,
> > &loc_elm);
> > + if (rc != 0)
> > + break;
> > +
> > + /* enqueue num elems */
> > + rte_compiler_barrier();
> > + rc = check_updt_elem(obj, num, &loc_elm, &def_elm);
> > + if (rc != 0)
> > + break;
> > +
> > + n = _st_ring_enqueue_bulk(la->rng, (void **)obj, num, NULL);
> > +
> > + /* check return value */
> > + rc = check_ring_op(num, n, lc, __func__,
> > + RTE_STR(_st_ring_enqueue_bulk));
> > + if (rc != 0)
> > + break;
> > +
> > + lcore_stat_update(&la->stats, 1, num, 0, 0);
> > +
> > + } while (wrk_cmd == WRK_CMD_RUN);
> > +
> > + /* final stats update */
> > + cl = rte_rdtsc_precise() - cl;
> > + lcore_stat_update(&la->stats, 0, 0, cl, 0);
> > + la->stats.nb_cycle = cl;
> > +
> > + return rc;
> > +}
> Just wondering about the need of 2 tests which run the same functionality. The difference is the way in which numbers are collected.
> Does 'test_worker_avg' adding any value? IMO, we can remove 'test_worker_avg'.
Yeh, they are quite similar.
I added _average_ version for two reasons:
1. In precise I call rte_rdtsc_precise() straight before/after
enqueue/dequeue op.
At least at IA rte_rdtsc_precise() means mb().
This extra sync point might hide some sync problems in the ring
enqueue/dequeue itself.
So having a separate test without such extra sync points
gives extra confidence that these tests would catch ring sync problems if any.
2. People usually don't do enqueue/dequeue on its own.
One of common patterns: dequeue/read-write data from the dequed objects/enqueue.
So this test measures cycles for dequeue/enqueue plus some reads/writes
to the objects from the ring.
> > +
> > +static void
> > +mt1_fini(struct rte_ring *rng, void *data) {
> > + rte_free(rng);
> > + rte_free(data);
> > +}
> > +
> > +static int
> > +mt1_init(struct rte_ring **rng, void **data, uint32_t num) {
> > + int32_t rc;
> >