patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Kevin Traynor <ktraynor@redhat.com>
To: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Cc: Phil Yang <phil.yang@arm.com>,
	Ruifeng Wang <ruifeng.wang@arm.com>,
	Ferruh Yigit <ferruh.yigit@intel.com>,
	Ali Alnubani <alialnu@mellanox.com>,
	dpdk stable <stable@dpdk.org>
Subject: [dpdk-stable] patch 'app/testpmd: use clock time in throughput calculation' has been queued to LTS release 18.11.10
Date: Thu, 20 Aug 2020 16:33:20 +0100	[thread overview]
Message-ID: <20200820153341.171927-15-ktraynor@redhat.com> (raw)
In-Reply-To: <20200820153341.171927-1-ktraynor@redhat.com>

Hi,

FYI, your patch has been queued to LTS release 18.11.10

Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
It will be pushed if I get no objections before 08/25/20. So please
shout if anyone has objections.

Also note that after the patch there's a diff of the upstream commit vs the
patch applied to the branch. This will indicate if there was any rebasing
needed to apply to the stable branch. If there were code changes for rebasing
(ie: not only metadata diffs), please double check that the rebase was
correctly done.

Queued patches are on a temporary branch at:
https://github.com/kevintraynor/dpdk-stable-queue

This queued commit can be viewed at:
https://github.com/kevintraynor/dpdk-stable-queue/commit/c6b6392a369508349351c1b9407a80515be14361

Thanks.

Kevin.

---
From c6b6392a369508349351c1b9407a80515be14361 Mon Sep 17 00:00:00 2001
From: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Date: Mon, 6 Jul 2020 18:32:29 -0500
Subject: [PATCH] app/testpmd: use clock time in throughput calculation

[ upstream commit aac2b6a78d7728ee39aed8822cba102b3a99a6bf ]

The throughput calculation requires a counter that measures
passing of time. However, the kernel saves and restores the PMU
state when a thread is unscheduled and scheduled. This ensures
that the PMU cycles are not counted towards a thread that is
not scheduled. Hence, when RTE_ARM_EAL_RDTSC_USE_PMU is enabled,
the PMU cycles do not represent the passing of time.
This results in incorrect calculation of throughput numbers.
Use clock_gettime system call to calculate the time passed since
last call.

Bugzilla ID: 450
Fixes: 0e106980301d ("app/testpmd: show throughput in port stats")

Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
Tested-by: Phil Yang <phil.yang@arm.com>
Tested-by: Ali Alnubani <alialnu@mellanox.com>
---
 app/test-pmd/config.c | 37 +++++++++++++++++++++++++++----------
 1 file changed, 27 insertions(+), 10 deletions(-)

diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index 0e5d77159d..b51e821a45 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -55,4 +55,12 @@
 #include "testpmd.h"
 
+#ifdef CLOCK_MONOTONIC_RAW /* Defined in glibc bits/time.h */
+#define CLOCK_TYPE_ID CLOCK_MONOTONIC_RAW
+#else
+#define CLOCK_TYPE_ID CLOCK_MONOTONIC
+#endif
+
+#define NS_PER_SEC 1E9
+
 static char *flowtype_to_str(uint16_t flow_type);
 
@@ -121,6 +129,7 @@ nic_stats_display(portid_t port_id)
 	static uint64_t prev_pkts_rx[RTE_MAX_ETHPORTS];
 	static uint64_t prev_pkts_tx[RTE_MAX_ETHPORTS];
-	static uint64_t prev_cycles[RTE_MAX_ETHPORTS];
-	uint64_t diff_pkts_rx, diff_pkts_tx, diff_cycles;
+	static uint64_t prev_ns[RTE_MAX_ETHPORTS];
+	struct timespec cur_time;
+	uint64_t diff_pkts_rx, diff_pkts_tx, diff_ns;
 	uint64_t mpps_rx, mpps_tx;
 	struct rte_eth_stats stats;
@@ -179,8 +188,15 @@ nic_stats_display(portid_t port_id)
 	}
 
-	diff_cycles = prev_cycles[port_id];
-	prev_cycles[port_id] = rte_rdtsc();
-	if (diff_cycles > 0)
-		diff_cycles = prev_cycles[port_id] - diff_cycles;
+	diff_ns = 0;
+	if (clock_gettime(CLOCK_TYPE_ID, &cur_time) == 0) {
+		uint64_t ns;
+
+		ns = cur_time.tv_sec * NS_PER_SEC;
+		ns += cur_time.tv_nsec;
+
+		if (prev_ns[port_id] != 0)
+			diff_ns = ns - prev_ns[port_id];
+		prev_ns[port_id] = ns;
+	}
 
 	diff_pkts_rx = (stats.ipackets > prev_pkts_rx[port_id]) ?
@@ -190,8 +206,9 @@ nic_stats_display(portid_t port_id)
 	prev_pkts_rx[port_id] = stats.ipackets;
 	prev_pkts_tx[port_id] = stats.opackets;
-	mpps_rx = diff_cycles > 0 ?
-		diff_pkts_rx * rte_get_tsc_hz() / diff_cycles : 0;
-	mpps_tx = diff_cycles > 0 ?
-		diff_pkts_tx * rte_get_tsc_hz() / diff_cycles : 0;
+	mpps_rx = diff_ns > 0 ?
+		(double)diff_pkts_rx / diff_ns * NS_PER_SEC : 0;
+	mpps_tx = diff_ns > 0 ?
+		(double)diff_pkts_tx / diff_ns * NS_PER_SEC : 0;
+
 	printf("\n  Throughput (since last show)\n");
 	printf("  Rx-pps: %12"PRIu64"\n  Tx-pps: %12"PRIu64"\n",
-- 
2.26.2

---
  Diff of the applied patch vs upstream commit (please double-check if non-empty:
---
--- -	2020-08-20 16:26:16.437389024 +0100
+++ 0015-app-testpmd-use-clock-time-in-throughput-calculation.patch	2020-08-20 16:26:15.799324097 +0100
@@ -1 +1 @@
-From aac2b6a78d7728ee39aed8822cba102b3a99a6bf Mon Sep 17 00:00:00 2001
+From c6b6392a369508349351c1b9407a80515be14361 Mon Sep 17 00:00:00 2001
@@ -5,0 +6,2 @@
+[ upstream commit aac2b6a78d7728ee39aed8822cba102b3a99a6bf ]
+
@@ -18 +19,0 @@
-Cc: stable@dpdk.org
@@ -27,2 +28,2 @@
- app/test-pmd/config.c | 44 +++++++++++++++++++++++++++++--------------
- 1 file changed, 30 insertions(+), 14 deletions(-)
+ app/test-pmd/config.c | 37 +++++++++++++++++++++++++++----------
+ 1 file changed, 27 insertions(+), 10 deletions(-)
@@ -31 +32 @@
-index a79019f522..75013100f5 100644
+index 0e5d77159d..b51e821a45 100644
@@ -35 +36 @@
- #define ETHDEV_FWVERS_LEN 32
+ #include "testpmd.h"
@@ -47,3 +48,3 @@
-@@ -140,7 +148,8 @@ nic_stats_display(portid_t port_id)
- 	static uint64_t prev_bytes_rx[RTE_MAX_ETHPORTS];
- 	static uint64_t prev_bytes_tx[RTE_MAX_ETHPORTS];
+@@ -121,6 +129,7 @@ nic_stats_display(portid_t port_id)
+ 	static uint64_t prev_pkts_rx[RTE_MAX_ETHPORTS];
+ 	static uint64_t prev_pkts_tx[RTE_MAX_ETHPORTS];
@@ -50,0 +52 @@
+-	uint64_t diff_pkts_rx, diff_pkts_tx, diff_cycles;
@@ -53,4 +55,2 @@
- 	uint64_t diff_pkts_rx, diff_pkts_tx, diff_bytes_rx, diff_bytes_tx,
--								diff_cycles;
-+								diff_ns;
- 	uint64_t mpps_rx, mpps_tx, mbps_rx, mbps_tx;
++	uint64_t diff_pkts_rx, diff_pkts_tx, diff_ns;
+ 	uint64_t mpps_rx, mpps_tx;
@@ -58 +58 @@
-@@ -199,8 +208,15 @@ nic_stats_display(portid_t port_id)
+@@ -179,8 +188,15 @@ nic_stats_display(portid_t port_id)
@@ -78 +78 @@
-@@ -210,8 +226,8 @@ nic_stats_display(portid_t port_id)
+@@ -190,8 +206,9 @@ nic_stats_display(portid_t port_id)
@@ -89,14 +89 @@
- 
- 	diff_bytes_rx = (stats.ibytes > prev_bytes_rx[port_id]) ?
-@@ -221,8 +237,8 @@ nic_stats_display(portid_t port_id)
- 	prev_bytes_rx[port_id] = stats.ibytes;
- 	prev_bytes_tx[port_id] = stats.obytes;
--	mbps_rx = diff_cycles > 0 ?
--		diff_bytes_rx * rte_get_tsc_hz() / diff_cycles : 0;
--	mbps_tx = diff_cycles > 0 ?
--		diff_bytes_tx * rte_get_tsc_hz() / diff_cycles : 0;
-+	mbps_rx = diff_ns > 0 ?
-+		(double)diff_bytes_rx / diff_ns * NS_PER_SEC : 0;
-+	mbps_tx = diff_ns > 0 ?
-+		(double)diff_bytes_tx / diff_ns * NS_PER_SEC : 0;
- 
++
@@ -103,0 +91 @@
+ 	printf("  Rx-pps: %12"PRIu64"\n  Tx-pps: %12"PRIu64"\n",


  parent reply	other threads:[~2020-08-20 15:34 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-20 15:33 [dpdk-stable] patch 'test: fix rpath for drivers with meson' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'rawdev: allow getting info for unknown device' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'rawdev: fill NUMA socket ID in info' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'rawdev: export dump function in map file' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'cfgfile: fix stack buffer underflow' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'drivers/crypto: add missing OOP feature flag' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'test/crypto: fix asymmetric session mempool creation' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'common/cpt: fix encryption offset' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'crypto/dpaax_sec: fix inline query for descriptors' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'service: fix core mapping reset' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/ixgbe: report 10Mbps link speed for x553' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/iavf: fix uninitialized variable' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/ixgbe/base: remove dead code' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/i40e: fix getting EEPROM information' " Kevin Traynor
2020-08-20 15:33 ` Kevin Traynor [this message]
2020-08-20 15:33 ` [dpdk-stable] patch 'app/testpmd: fix burst percentage calculation' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'ethdev: fix log type for some error messages' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'service: fix C++ linkage' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net: fix unneeded replacement of TCP checksum 0' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net: fix checksum on big endian CPUs' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'eal: fix parentheses in alignment macros' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'vhost: fix virtio ready flag check' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'bus/fslmc: fix getting FD error' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/dpaa: fix FD offset data type' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/bonding: fix socket ID check' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/netvsc: fix underflow when Rx external mbuf' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/ixgbe/base: fix host interface shadow RAM read' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/ixgbe/base: fix x550em 10G NIC link status' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/ixgbe/base: fix infinite recursion on PCIe link down' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'ethdev: fix data room size verification in Rx queue setup' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/nfp: fix RSS hash configuration reporting' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'drivers/net: fix exposing internal headers' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/bonding: fix LACP negotiation' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'net/bnxt: remove unused enum declaration' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'examples/fips_validation: fix parsing of TDES vectors' " Kevin Traynor
2020-08-20 15:33 ` [dpdk-stable] patch 'examples/fips_validation: fix count overwrite for TDES' " Kevin Traynor

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=20200820153341.171927-15-ktraynor@redhat.com \
    --to=ktraynor@redhat.com \
    --cc=alialnu@mellanox.com \
    --cc=ferruh.yigit@intel.com \
    --cc=honnappa.nagarahalli@arm.com \
    --cc=phil.yang@arm.com \
    --cc=ruifeng.wang@arm.com \
    --cc=stable@dpdk.org \
    /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
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).