DPDK patches and discussions
 help / color / mirror / Atom feed
From: Feifei Wang <feifei.wang2@arm.com>
To: Matan Azrad <matan@nvidia.com>,
	Shahaf Shuler <shahafs@nvidia.com>,
	Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Cc: dev@dpdk.org, nd@arm.com, Feifei Wang <feifei.wang2@arm.com>,
	Ruifeng Wang <ruifeng.wang@arm.com>
Subject: [dpdk-dev] [PATCH v3 2/2] net/mlx5: remove unnecessary wmb for Memory Region cache
Date: Tue, 18 May 2021 16:50:58 +0800
Message-ID: <20210518085058.630072-3-feifei.wang2@arm.com> (raw)
In-Reply-To: <20210518085058.630072-1-feifei.wang2@arm.com>

'dev_gen' is a variable to trigger all cores to flush their local caches
once the global MR cache has been rebuilt.

This is due to MR cache's R/W lock can maintain synchronization between
threads:

1. dev_gen and global cache updating ordering inside the lock protected
section does not matter. Because other threads cannot take the lock
until global cache has been updated. Thus, in out of order platform,
even if other agents firstly observe updated dev_gen but global does
not update, they also have to wait the lock. As a result, it is
unnecessary to add a wmb between global cache rebuilding and updating
the dev_gen to keep the memory store order.

2. Store-Release of unlock provides the implicit wmb at the level
visible by software. This makes 'rebuilding global cache' and 'updating
dev_gen' be observed before local_cache starts to be updated by other
agents. Thus, wmb after 'updating dev_gen' can be removed.

Suggested-by: Ruifeng Wang <ruifeng.wang@arm.com>
Signed-off-by: Feifei Wang <feifei.wang2@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
---
 drivers/net/mlx5/mlx5_mr.c | 22 ++++++----------------
 1 file changed, 6 insertions(+), 16 deletions(-)

diff --git a/drivers/net/mlx5/mlx5_mr.c b/drivers/net/mlx5/mlx5_mr.c
index e791b6338d..0c5403e493 100644
--- a/drivers/net/mlx5/mlx5_mr.c
+++ b/drivers/net/mlx5/mlx5_mr.c
@@ -107,18 +107,13 @@ mlx5_mr_mem_event_free_cb(struct mlx5_dev_ctx_shared *sh,
 	if (rebuild) {
 		mlx5_mr_rebuild_cache(&sh->share_cache);
 		/*
-		 * Flush local caches by propagating invalidation across cores.
-		 * rte_smp_wmb() is enough to synchronize this event. If one of
-		 * freed memsegs is seen by other core, that means the memseg
-		 * has been allocated by allocator, which will come after this
-		 * free call. Therefore, this store instruction (incrementing
-		 * generation below) will be guaranteed to be seen by other core
-		 * before the core sees the newly allocated memory.
+		 * No explicit wmb is needed after updating dev_gen due to
+		 * store-release ordering in unlock that provides the
+		 * implicit barrier at the software visible level.
 		 */
 		++sh->share_cache.dev_gen;
 		DRV_LOG(DEBUG, "broadcasting local cache flush, gen=%d",
 		      sh->share_cache.dev_gen);
-		rte_smp_wmb();
 	}
 	rte_rwlock_write_unlock(&sh->share_cache.rwlock);
 }
@@ -411,18 +406,13 @@ mlx5_dma_unmap(struct rte_pci_device *pdev, void *addr,
 	      (void *)mr);
 	mlx5_mr_rebuild_cache(&sh->share_cache);
 	/*
-	 * Flush local caches by propagating invalidation across cores.
-	 * rte_smp_wmb() is enough to synchronize this event. If one of
-	 * freed memsegs is seen by other core, that means the memseg
-	 * has been allocated by allocator, which will come after this
-	 * free call. Therefore, this store instruction (incrementing
-	 * generation below) will be guaranteed to be seen by other core
-	 * before the core sees the newly allocated memory.
+	 * No explicit wmb is needed after updating dev_gen due to
+	 * store-release ordering in unlock that provides the
+	 * implicit barrier at the software visible level.
 	 */
 	++sh->share_cache.dev_gen;
 	DRV_LOG(DEBUG, "broadcasting local cache flush, gen=%d",
 	      sh->share_cache.dev_gen);
-	rte_smp_wmb();
 	rte_rwlock_read_unlock(&sh->share_cache.rwlock);
 	return 0;
 }
-- 
2.25.1


  parent reply	other threads:[~2021-05-18  8:51 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-18  7:18 [dpdk-dev] [PATCH v1 0/4] refactor SMP barriers for net/mlx Feifei Wang
2021-03-18  7:18 ` [dpdk-dev] [PATCH v1 1/4] net/mlx4: fix rebuild bug for Memory Region cache Feifei Wang
2021-03-18  7:18 ` [dpdk-dev] [PATCH v1 2/4] net/mlx4: replace SMP barrier with C11 barriers Feifei Wang
2021-03-18  7:18 ` [dpdk-dev] [PATCH v1 3/4] net/mlx5: fix rebuild bug for Memory Region cache Feifei Wang
2021-04-12  8:27   ` Slava Ovsiienko
2021-04-13  5:20     ` [dpdk-dev] 回复: " Feifei Wang
2021-04-19 18:50       ` [dpdk-dev] " Slava Ovsiienko
2021-04-20  5:53         ` [dpdk-dev] 回复: " Feifei Wang
2021-04-20  7:29           ` Feifei Wang
2021-04-20  7:53             ` [dpdk-dev] " Slava Ovsiienko
2021-04-20  8:42               ` [dpdk-dev] 回复: " Feifei Wang
2021-05-06  2:52                 ` Feifei Wang
2021-05-06 11:21                   ` [dpdk-dev] " Slava Ovsiienko
2021-05-07  6:36                     ` [dpdk-dev] 回复: " Feifei Wang
2021-05-07 10:14                       ` [dpdk-dev] " Slava Ovsiienko
2021-05-08  3:13                         ` [dpdk-dev] 回复: " Feifei Wang
2021-05-11  8:18                           ` [dpdk-dev] " Slava Ovsiienko
2021-05-12  5:34                             ` [dpdk-dev] 回复: " Feifei Wang
2021-05-12 11:07                               ` [dpdk-dev] " Slava Ovsiienko
2021-05-13  5:49                                 ` [dpdk-dev] 回复: " Feifei Wang
2021-05-13 10:49                                   ` [dpdk-dev] " Slava Ovsiienko
2021-05-14  5:18                                     ` [dpdk-dev] 回复: " Feifei Wang
2021-03-18  7:18 ` [dpdk-dev] [PATCH v1 4/4] net/mlx5: replace SMP barriers with C11 barriers Feifei Wang
2021-04-07  1:45 ` [dpdk-dev] [PATCH v1 0/4] refactor SMP barriers for net/mlx Alexander Kozyrev
2021-05-17 10:00 ` [dpdk-dev] [PATCH v2 0/2] remove wmb " Feifei Wang
2021-05-17 10:00   ` [dpdk-dev] [PATCH v2 1/2] net/mlx4: remove unnecessary wmb for Memory Region cache Feifei Wang
2021-05-17 10:00   ` [dpdk-dev] [PATCH v2 2/2] net/mlx5: " Feifei Wang
2021-05-17 14:15     ` Slava Ovsiienko
2021-05-18  8:52       ` [dpdk-dev] 回复: " Feifei Wang
2021-05-18  8:50 ` [dpdk-dev] [PATCH v3 0/2] remove wmb for net/mlx Feifei Wang
2021-05-18  8:50   ` [dpdk-dev] [PATCH v3 1/2] net/mlx4: remove unnecessary wmb for Memory Region cache Feifei Wang
2021-05-18 12:13     ` Slava Ovsiienko
2021-05-18  8:50   ` Feifei Wang [this message]
2021-05-18 10:17     ` [dpdk-dev] [PATCH v3 2/2] net/mlx5: " Slava Ovsiienko
2021-05-19  1:54       ` [dpdk-dev] 回复: " Feifei Wang
2021-05-27  8:37   ` [dpdk-dev] [PATCH v3 0/2] remove wmb for net/mlx Raslan Darawsheh

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=20210518085058.630072-3-feifei.wang2@arm.com \
    --to=feifei.wang2@arm.com \
    --cc=dev@dpdk.org \
    --cc=matan@nvidia.com \
    --cc=nd@arm.com \
    --cc=ruifeng.wang@arm.com \
    --cc=shahafs@nvidia.com \
    --cc=viacheslavo@nvidia.com \
    /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