patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Yuanhan Liu <yliu@fridaylinux.org>
To: Andriy Berestovskyy <aber@semihalf.com>
Cc: Remy Horton <remy.horton@intel.com>, dpdk stable <stable@dpdk.org>
Subject: [dpdk-stable] patch 'keepalive: fix state alignment' has been queued to LTS release 17.11.1
Date: Thu,  1 Feb 2018 17:47:19 +0800	[thread overview]
Message-ID: <1517478479-12417-5-git-send-email-yliu@fridaylinux.org> (raw)
In-Reply-To: <1517478479-12417-1-git-send-email-yliu@fridaylinux.org>

Hi,

FYI, your patch has been queued to LTS release 17.11.1

Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
It will be pushed if I get no objections before 02/03/18. So please
shout if anyone has objections.

Thanks.

	--yliu

---
>From 0197da770df2a30d6753634f00db8b39132a64ae Mon Sep 17 00:00:00 2001
From: Andriy Berestovskyy <aber@semihalf.com>
Date: Tue, 23 Jan 2018 16:43:16 +0100
Subject: [PATCH] keepalive: fix state alignment

[ upstream commit 462d9189374f582166ff77b8c301b89947d016cc ]

The __rte_cache_aligned was applied to the whole array,
not the array elements. This leads to a false sharing between
the monitored cores.

Fixes: e70a61ad50ab ("keepalive: export states")

Signed-off-by: Andriy Berestovskyy <aber@semihalf.com>
Acked-by: Remy Horton <remy.horton@intel.com>
---
 doc/guides/sample_app_ug/keep_alive.rst |  2 +-
 lib/librte_eal/common/rte_keepalive.c   | 28 ++++++++++++++++++----------
 2 files changed, 19 insertions(+), 11 deletions(-)

diff --git a/doc/guides/sample_app_ug/keep_alive.rst b/doc/guides/sample_app_ug/keep_alive.rst
index 9b8be48..6c131a8 100644
--- a/doc/guides/sample_app_ug/keep_alive.rst
+++ b/doc/guides/sample_app_ug/keep_alive.rst
@@ -168,5 +168,5 @@ The rte_keepalive_mark_alive function simply sets the core state to alive.
     static inline void
     rte_keepalive_mark_alive(struct rte_keepalive *keepcfg)
     {
-        keepcfg->state_flags[rte_lcore_id()] = ALIVE;
+        keepcfg->live_data[rte_lcore_id()].core_state = RTE_KA_STATE_ALIVE;
     }
diff --git a/lib/librte_eal/common/rte_keepalive.c b/lib/librte_eal/common/rte_keepalive.c
index cdd6956..0c7ecba 100644
--- a/lib/librte_eal/common/rte_keepalive.c
+++ b/lib/librte_eal/common/rte_keepalive.c
@@ -41,8 +41,12 @@
 
 struct rte_keepalive {
 	/** Core Liveness. */
-	enum rte_keepalive_state __rte_cache_aligned state_flags[
-		RTE_KEEPALIVE_MAXCORES];
+	struct {
+		/*
+		 * Each element must be cache aligned to prevent false sharing.
+		 */
+		enum rte_keepalive_state core_state __rte_cache_aligned;
+	} live_data[RTE_KEEPALIVE_MAXCORES];
 
 	/** Last-seen-alive timestamps */
 	uint64_t last_alive[RTE_KEEPALIVE_MAXCORES];
@@ -95,19 +99,22 @@ rte_keepalive_dispatch_pings(__rte_unused void *ptr_timer,
 		if (keepcfg->active_cores[idx_core] == 0)
 			continue;
 
-		switch (keepcfg->state_flags[idx_core]) {
+		switch (keepcfg->live_data[idx_core].core_state) {
 		case RTE_KA_STATE_UNUSED:
 			break;
 		case RTE_KA_STATE_ALIVE: /* Alive */
-			keepcfg->state_flags[idx_core] = RTE_KA_STATE_MISSING;
+			keepcfg->live_data[idx_core].core_state =
+			    RTE_KA_STATE_MISSING;
 			keepcfg->last_alive[idx_core] = rte_rdtsc();
 			break;
 		case RTE_KA_STATE_MISSING: /* MIA */
 			print_trace("Core MIA. ", keepcfg, idx_core);
-			keepcfg->state_flags[idx_core] = RTE_KA_STATE_DEAD;
+			keepcfg->live_data[idx_core].core_state =
+			    RTE_KA_STATE_DEAD;
 			break;
 		case RTE_KA_STATE_DEAD: /* Dead */
-			keepcfg->state_flags[idx_core] = RTE_KA_STATE_GONE;
+			keepcfg->live_data[idx_core].core_state =
+			    RTE_KA_STATE_GONE;
 			print_trace("Core died. ", keepcfg, idx_core);
 			if (keepcfg->callback)
 				keepcfg->callback(
@@ -118,7 +125,8 @@ rte_keepalive_dispatch_pings(__rte_unused void *ptr_timer,
 		case RTE_KA_STATE_GONE: /* Buried */
 			break;
 		case RTE_KA_STATE_DOZING: /* Core going idle */
-			keepcfg->state_flags[idx_core] = RTE_KA_STATE_SLEEP;
+			keepcfg->live_data[idx_core].core_state =
+			    RTE_KA_STATE_SLEEP;
 			keepcfg->last_alive[idx_core] = rte_rdtsc();
 			break;
 		case RTE_KA_STATE_SLEEP: /* Idled core */
@@ -128,7 +136,7 @@ rte_keepalive_dispatch_pings(__rte_unused void *ptr_timer,
 			keepcfg->relay_callback(
 				keepcfg->relay_callback_data,
 				idx_core,
-				keepcfg->state_flags[idx_core],
+				keepcfg->live_data[idx_core].core_state,
 				keepcfg->last_alive[idx_core]
 				);
 	}
@@ -172,11 +180,11 @@ rte_keepalive_register_core(struct rte_keepalive *keepcfg, const int id_core)
 void
 rte_keepalive_mark_alive(struct rte_keepalive *keepcfg)
 {
-	keepcfg->state_flags[rte_lcore_id()] = RTE_KA_STATE_ALIVE;
+	keepcfg->live_data[rte_lcore_id()].core_state = RTE_KA_STATE_ALIVE;
 }
 
 void
 rte_keepalive_mark_sleep(struct rte_keepalive *keepcfg)
 {
-	keepcfg->state_flags[rte_lcore_id()] = RTE_KA_STATE_DOZING;
+	keepcfg->live_data[rte_lcore_id()].core_state = RTE_KA_STATE_DOZING;
 }
-- 
2.7.4

  parent reply	other threads:[~2018-02-01  9:48 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-02-01  9:47 [dpdk-stable] patch 'event/sw: fix debug logging config option' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'service: fix possible mem leak on initialize' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'cmdline: fix dynamic tokens parsing' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'cmdline: avoid garbage in unused fields of parsed result' " Yuanhan Liu
2018-02-01  9:47 ` Yuanhan Liu [this message]
2018-02-01  9:47 ` [dpdk-stable] patch 'log: fix memory leak in regexp level set' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'eal/arm64: remove the braces in memory barrier macros' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'mbuf: fix NULL freeing when debug enabled' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'crypto/qat: fix out-of-bounds access' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'crypto/qat: fix parameter type' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'cryptodev: fix session pointer cast' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'crypto/qat: fix null auth algo overwrite' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/mlx5: fix return value of start operation' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/enic: fix crash due to static max number of queues' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/e1000: fix null pointer check' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/i40e: fix memory leak' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/mlx5: fix missing RSS capability' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/mlx5: fix flow item validation' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/mlx5: fix memory region cache lookup' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/mlx5: fix memory region cache last index' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/mlx5: fix memory region boundary checks' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/ena: do not set Tx L4 offloads in Rx path' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/virtio-user: fix crash as features change' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/virtio: fix Rx and Tx handler selection for ARM32' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/virtio: fix queue flushing with vector Rx enabled' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/virtio: fix memory leak when reinitializing device' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/i40e: fix VF Rx interrupt enabling' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/ixgbe: " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/e1000: " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/bnxt: fix size of Tx ring in HW' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/bnxt: fix number of pools for RSS' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/qede: check tunnel L3 header' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/qede: fix tunnel header size in Tx BD configuration' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/qede: fix MTU set and max Rx length' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'eal/ppc: remove the braces in memory barrier macros' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'crypto/dpaa2_sec: fix enum conversion for GCM' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'mk: support renamed Makefile in external project' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'crypto/scheduler: fix strncpy' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'ethdev: fix port data reset timing' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'ethdev: fix port id allocation' " Yuanhan Liu
2018-02-11  2:46   ` Yuanhan Liu
2018-02-11  6:40     ` Matan Azrad
2018-02-11  6:54       ` Yuanhan Liu
2018-02-11  7:15         ` Matan Azrad
2018-02-01  9:47 ` [dpdk-stable] patch 'app/testpmd: fix port validation' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'net/ixgbe: fix reset error handling' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'examples/bond: fix vdev name' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'crypto/qat: fix allocation check and leak' " Yuanhan Liu
2018-02-01  9:47 ` [dpdk-stable] patch 'examples/bond: check mbuf allocation' " Yuanhan Liu

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=1517478479-12417-5-git-send-email-yliu@fridaylinux.org \
    --to=yliu@fridaylinux.org \
    --cc=aber@semihalf.com \
    --cc=remy.horton@intel.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).