DPDK patches and discussions
 help / color / mirror / Atom feed
From: Phil Yang <phil.yang@arm.com>
To: erik.g.carrillo@intel.com, rsanford@akamai.com, dev@dpdk.org
Cc: thomas@monjalon.net, david.marchand@redhat.com,
	konstantin.ananyev@intel.com, jerinj@marvell.com,
	hemant.agrawal@nxp.com, Honnappa.Nagarahalli@arm.com,
	gavin.hu@arm.com, nd@arm.com
Subject: [dpdk-dev] [PATCH v2] lib/timer: relax barrier for status update
Date: Tue, 21 Apr 2020 00:05:52 +0800
Message-ID: <1587398752-9345-1-git-send-email-phil.yang@arm.com> (raw)
In-Reply-To: <1582526539-14360-2-git-send-email-phil.yang@arm.com>

Volatile has no ordering semantics. The rte_timer structure defines
timer status as a volatile variable and uses the rte_r/wmb barrier
to guarantee inter-thread visibility.

This patch optimized the volatile operation with c11 atomic operations
and one-way barrier to save the performance penalty. According to the
timer_perf_autotest benchmarking results, this patch can uplift 10%~16%
timer appending performance, 3%~20% timer resetting performance and 45%
timer callbacks scheduling performance on aarch64 and no loss in
performance for x86.

Suggested-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Signed-off-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>

---
This patch depends on patch:
http://patchwork.dpdk.org/patch/65997/

v2:
1. Changed the memory ordering comment in timer_set_config_state.
2. It is still using built-ins as the wrapper functions for C11 built-ins
are not defined yet.

 lib/librte_timer/rte_timer.c | 85 ++++++++++++++++++++++++++++++--------------
 lib/librte_timer/rte_timer.h |  2 +-
 2 files changed, 60 insertions(+), 27 deletions(-)

diff --git a/lib/librte_timer/rte_timer.c b/lib/librte_timer/rte_timer.c
index 269e921..ba17216 100644
--- a/lib/librte_timer/rte_timer.c
+++ b/lib/librte_timer/rte_timer.c
@@ -10,7 +10,6 @@
 #include <assert.h>
 #include <sys/queue.h>
 
-#include <rte_atomic.h>
 #include <rte_common.h>
 #include <rte_cycles.h>
 #include <rte_eal_memconfig.h>
@@ -218,7 +217,7 @@ rte_timer_init(struct rte_timer *tim)
 
 	status.state = RTE_TIMER_STOP;
 	status.owner = RTE_TIMER_NO_OWNER;
-	tim->status.u32 = status.u32;
+	__atomic_store_n(&tim->status.u32, status.u32, __ATOMIC_RELAXED);
 }
 
 /*
@@ -239,9 +238,9 @@ timer_set_config_state(struct rte_timer *tim,
 
 	/* wait that the timer is in correct status before update,
 	 * and mark it as being configured */
-	while (success == 0) {
-		prev_status.u32 = tim->status.u32;
+	prev_status.u32 = __atomic_load_n(&tim->status.u32, __ATOMIC_RELAXED);
 
+	while (success == 0) {
 		/* timer is running on another core
 		 * or ready to run on local core, exit
 		 */
@@ -258,9 +257,15 @@ timer_set_config_state(struct rte_timer *tim,
 		 * mark it atomically as being configured */
 		status.state = RTE_TIMER_CONFIG;
 		status.owner = (int16_t)lcore_id;
-		success = rte_atomic32_cmpset(&tim->status.u32,
-					      prev_status.u32,
-					      status.u32);
+		/* CONFIG states are acting as locked states. If the
+		 * timer is in CONFIG state, the state cannot be changed
+		 * by other threads. So, we should use ACQUIRE here.
+		 */
+		success = __atomic_compare_exchange_n(&tim->status.u32,
+					      &prev_status.u32,
+					      status.u32, 0,
+					      __ATOMIC_ACQUIRE,
+					      __ATOMIC_RELAXED);
 	}
 
 	ret_prev_status->u32 = prev_status.u32;
@@ -279,20 +284,27 @@ timer_set_running_state(struct rte_timer *tim)
 
 	/* wait that the timer is in correct status before update,
 	 * and mark it as running */
-	while (success == 0) {
-		prev_status.u32 = tim->status.u32;
+	prev_status.u32 = __atomic_load_n(&tim->status.u32, __ATOMIC_RELAXED);
 
+	while (success == 0) {
 		/* timer is not pending anymore */
 		if (prev_status.state != RTE_TIMER_PENDING)
 			return -1;
 
 		/* here, we know that timer is stopped or pending,
-		 * mark it atomically as being configured */
+		 * mark it atomically as being running
+		 */
 		status.state = RTE_TIMER_RUNNING;
 		status.owner = (int16_t)lcore_id;
-		success = rte_atomic32_cmpset(&tim->status.u32,
-					      prev_status.u32,
-					      status.u32);
+		/* RUNNING states are acting as locked states. If the
+		 * timer is in RUNNING state, the state cannot be changed
+		 * by other threads. So, we should use ACQUIRE here.
+		 */
+		success = __atomic_compare_exchange_n(&tim->status.u32,
+					      &prev_status.u32,
+					      status.u32, 0,
+					      __ATOMIC_ACQUIRE,
+					      __ATOMIC_RELAXED);
 	}
 
 	return 0;
@@ -520,10 +532,12 @@ __rte_timer_reset(struct rte_timer *tim, uint64_t expire,
 
 	/* update state: as we are in CONFIG state, only us can modify
 	 * the state so we don't need to use cmpset() here */
-	rte_wmb();
 	status.state = RTE_TIMER_PENDING;
 	status.owner = (int16_t)tim_lcore;
-	tim->status.u32 = status.u32;
+	/* The "RELEASE" ordering guarantees the memory operations above
+	 * the status update are observed before the update by all threads
+	 */
+	__atomic_store_n(&tim->status.u32, status.u32, __ATOMIC_RELEASE);
 
 	if (tim_lcore != lcore_id || !local_is_locked)
 		rte_spinlock_unlock(&priv_timer[tim_lcore].list_lock);
@@ -600,10 +614,12 @@ __rte_timer_stop(struct rte_timer *tim, int local_is_locked,
 	}
 
 	/* mark timer as stopped */
-	rte_wmb();
 	status.state = RTE_TIMER_STOP;
 	status.owner = RTE_TIMER_NO_OWNER;
-	tim->status.u32 = status.u32;
+	/* The "RELEASE" ordering guarantees the memory operations above
+	 * the status update are observed before the update by all threads
+	 */
+	__atomic_store_n(&tim->status.u32, status.u32, __ATOMIC_RELEASE);
 
 	return 0;
 }
@@ -637,7 +653,8 @@ rte_timer_stop_sync(struct rte_timer *tim)
 int
 rte_timer_pending(struct rte_timer *tim)
 {
-	return tim->status.state == RTE_TIMER_PENDING;
+	return __atomic_load_n(&tim->status.state,
+				__ATOMIC_RELAXED) == RTE_TIMER_PENDING;
 }
 
 /* must be called periodically, run all timer that expired */
@@ -739,8 +756,12 @@ __rte_timer_manage(struct rte_timer_data *timer_data)
 			/* remove from done list and mark timer as stopped */
 			status.state = RTE_TIMER_STOP;
 			status.owner = RTE_TIMER_NO_OWNER;
-			rte_wmb();
-			tim->status.u32 = status.u32;
+			/* The "RELEASE" ordering guarantees the memory
+			 * operations above the status update are observed
+			 * before the update by all threads
+			 */
+			__atomic_store_n(&tim->status.u32, status.u32,
+				__ATOMIC_RELEASE);
 		}
 		else {
 			/* keep it in list and mark timer as pending */
@@ -748,8 +769,12 @@ __rte_timer_manage(struct rte_timer_data *timer_data)
 			status.state = RTE_TIMER_PENDING;
 			__TIMER_STAT_ADD(priv_timer, pending, 1);
 			status.owner = (int16_t)lcore_id;
-			rte_wmb();
-			tim->status.u32 = status.u32;
+			/* The "RELEASE" ordering guarantees the memory
+			 * operations above the status update are observed
+			 * before the update by all threads
+			 */
+			__atomic_store_n(&tim->status.u32, status.u32,
+				__ATOMIC_RELEASE);
 			__rte_timer_reset(tim, tim->expire + tim->period,
 				tim->period, lcore_id, tim->f, tim->arg, 1,
 				timer_data);
@@ -919,8 +944,12 @@ rte_timer_alt_manage(uint32_t timer_data_id,
 			/* remove from done list and mark timer as stopped */
 			status.state = RTE_TIMER_STOP;
 			status.owner = RTE_TIMER_NO_OWNER;
-			rte_wmb();
-			tim->status.u32 = status.u32;
+			/* The "RELEASE" ordering guarantees the memory
+			 * operations above the status update are observed
+			 * before the update by all threads
+			 */
+			__atomic_store_n(&tim->status.u32, status.u32,
+				__ATOMIC_RELEASE);
 		} else {
 			/* keep it in list and mark timer as pending */
 			rte_spinlock_lock(
@@ -928,8 +957,12 @@ rte_timer_alt_manage(uint32_t timer_data_id,
 			status.state = RTE_TIMER_PENDING;
 			__TIMER_STAT_ADD(data->priv_timer, pending, 1);
 			status.owner = (int16_t)this_lcore;
-			rte_wmb();
-			tim->status.u32 = status.u32;
+			/* The "RELEASE" ordering guarantees the memory
+			 * operations above the status update are observed
+			 * before the update by all threads
+			 */
+			__atomic_store_n(&tim->status.u32, status.u32,
+				__ATOMIC_RELEASE);
 			__rte_timer_reset(tim, tim->expire + tim->period,
 				tim->period, this_lcore, tim->f, tim->arg, 1,
 				data);
diff --git a/lib/librte_timer/rte_timer.h b/lib/librte_timer/rte_timer.h
index c6b3d45..df533fa 100644
--- a/lib/librte_timer/rte_timer.h
+++ b/lib/librte_timer/rte_timer.h
@@ -101,7 +101,7 @@ struct rte_timer
 {
 	uint64_t expire;       /**< Time when timer expire. */
 	struct rte_timer *sl_next[MAX_SKIPLIST_DEPTH];
-	volatile union rte_timer_status status; /**< Status of timer. */
+	union rte_timer_status status; /**< Status of timer. */
 	uint64_t period;       /**< Period of timer (0 if not periodic). */
 	rte_timer_cb_t f;      /**< Callback function. */
 	void *arg;             /**< Argument to callback function. */
-- 
2.7.4


  parent reply	other threads:[~2020-04-20 16:06 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-24  6:42 [dpdk-dev] [PATCH 1/2] lib/timer: protect timer subsystem initialized with lock Phil Yang
2020-02-24  6:42 ` [dpdk-dev] [PATCH 2/2] lib/timer: relax barrier for status update Phil Yang
2020-04-08 10:23   ` Phil Yang
2020-04-08 21:10   ` Carrillo, Erik G
2020-04-08 21:16     ` Honnappa Nagarahalli
2020-04-08 21:26       ` Carrillo, Erik G
2020-04-08 21:56         ` Honnappa Nagarahalli
2020-04-09 19:29           ` Carrillo, Erik G
2020-04-10  4:39             ` Phil Yang
2020-04-20 16:05   ` Phil Yang [this message]
2020-04-23 20:06     ` [dpdk-dev] [PATCH v2] " Honnappa Nagarahalli
2020-04-24  1:26       ` Carrillo, Erik G
2020-04-24  7:27         ` Phil Yang
2020-04-24  7:24     ` [dpdk-dev] [PATCH v3] " Phil Yang
2020-04-25 17:17       ` Thomas Monjalon
2020-04-26  7:36         ` Phil Yang
2020-04-26 12:18           ` Carrillo, Erik G
2020-04-26 14:20             ` Phil Yang
2020-04-26 19:30               ` Carrillo, Erik G
2020-04-26 14:45       ` [dpdk-dev] [PATCH v4] " Phil Yang
2020-04-26 20:06         ` Thomas Monjalon
2020-04-25 14:36     ` [dpdk-dev] [PATCH v2] " Thomas Monjalon
2020-04-25 15:51       ` Phil Yang
2020-04-25 16:07         ` Thomas Monjalon
2020-02-25 22:26 ` [dpdk-dev] [PATCH 1/2] lib/timer: protect timer subsystem initialized with lock Carrillo, Erik G
2020-04-25 17:21   ` [dpdk-dev] [dpdk-stable] " Thomas Monjalon

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1587398752-9345-1-git-send-email-phil.yang@arm.com \
    --to=phil.yang@arm.com \
    --cc=Honnappa.Nagarahalli@arm.com \
    --cc=david.marchand@redhat.com \
    --cc=dev@dpdk.org \
    --cc=erik.g.carrillo@intel.com \
    --cc=gavin.hu@arm.com \
    --cc=hemant.agrawal@nxp.com \
    --cc=jerinj@marvell.com \
    --cc=konstantin.ananyev@intel.com \
    --cc=nd@arm.com \
    --cc=rsanford@akamai.com \
    --cc=thomas@monjalon.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

DPDK patches and discussions

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://inbox.dpdk.org/dev/0 dev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 dev dev/ https://inbox.dpdk.org/dev \
		dev@dpdk.org
	public-inbox-index dev

Example config snippet for mirrors.
Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.dev


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git