From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by dpdk.org (Postfix) with ESMTP id 2B71D1B4FA for ; Fri, 23 Nov 2018 11:29:27 +0100 (CET) Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.phx2.redhat.com [10.5.11.14]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 80F60308624A; Fri, 23 Nov 2018 10:29:26 +0000 (UTC) Received: from ktraynor.remote.csb (unknown [10.36.118.7]) by smtp.corp.redhat.com (Postfix) with ESMTP id 068D1179CF; Fri, 23 Nov 2018 10:29:23 +0000 (UTC) From: Kevin Traynor To: Gavin Hu Cc: Honnappa Nagarahalli , Steve Capper , Ola Liljedahl , Jia He , Jerin Jacob , Olivier Matz , dpdk stable Date: Fri, 23 Nov 2018 10:26:30 +0000 Message-Id: <20181123102713.17309-26-ktraynor@redhat.com> In-Reply-To: <20181123102713.17309-1-ktraynor@redhat.com> References: <20181123102713.17309-1-ktraynor@redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Scanned-By: MIMEDefang 2.79 on 10.5.11.14 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.49]); Fri, 23 Nov 2018 10:29:26 +0000 (UTC) Subject: [dpdk-stable] patch 'ring/c11: synchronize load and store of the tail' has been queued to stable release 18.08.1 X-BeenThere: stable@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches for DPDK stable branches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 23 Nov 2018 10:29:27 -0000 Hi, FYI, your patch has been queued to stable release 18.08.1 Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet. It will be pushed if I get no objections before 11/29/18. 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. If the code is different (ie: not only metadata diffs), due for example to a change in context or macro names, please double check it. Thanks. Kevin Traynor --- >>From f57da12160802dfe41ef7b82c7d41316769d9179 Mon Sep 17 00:00:00 2001 From: Gavin Hu Date: Fri, 2 Nov 2018 19:21:27 +0800 Subject: [PATCH] ring/c11: synchronize load and store of the tail [ upstream commit 9ed877062898dd1eb7150bca56ebbd5329e96ee4 ] Synchronize the load-acquire of the tail and the store-release within update_tail, the store release ensures all the ring operations, enqueue or dequeue, are seen by the observers on the other side as soon as they see the updated tail. The load-acquire is needed here as the data dependency is not a reliable way for ordering as the compiler might break it by saving to temporary values to boost performance. When computing the free_entries and avail_entries, use atomic semantics to load the heads and tails instead. The patch was benchmarked with test/ring_perf_autotest and it decreases the enqueue/dequeue latency by 5% ~ 27.6% with two lcores, the real gains are dependent on the number of lcores, depth of the ring, SPSC or MPMC. For 1 lcore, it also improves a little, about 3 ~ 4%. It is a big improvement, in case of MPMC, with two lcores and ring size of 32, it saves latency up to (3.26-2.36)/3.26 = 27.6%. This patch is a bug fix, while the improvement is a bonus. In our analysis the improvement comes from the cacheline pre-filling after hoisting load- acquire from _atomic_compare_exchange_n up above. The test command: $sudo ./test/test/test -l 16-19,44-47,72-75,100-103 -n 4 --socket-mem=\ 1024 -- -i Test result with this patch(two cores): SP/SC bulk enq/dequeue (size: 8): 5.86 MP/MC bulk enq/dequeue (size: 8): 10.15 SP/SC bulk enq/dequeue (size: 32): 1.94 MP/MC bulk enq/dequeue (size: 32): 2.36 In comparison of the test result without this patch: SP/SC bulk enq/dequeue (size: 8): 6.67 MP/MC bulk enq/dequeue (size: 8): 13.12 SP/SC bulk enq/dequeue (size: 32): 2.04 MP/MC bulk enq/dequeue (size: 32): 3.26 Fixes: 39368ebfc6 ("ring: introduce C11 memory model barrier option") Signed-off-by: Gavin Hu Reviewed-by: Honnappa Nagarahalli Reviewed-by: Steve Capper Reviewed-by: Ola Liljedahl Reviewed-by: Jia He Acked-by: Jerin Jacob Tested-by: Jerin Jacob Acked-by: Olivier Matz --- lib/librte_ring/rte_ring_c11_mem.h | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) diff --git a/lib/librte_ring/rte_ring_c11_mem.h b/lib/librte_ring/rte_ring_c11_mem.h index 94df3c4a6..52da95a21 100644 --- a/lib/librte_ring/rte_ring_c11_mem.h +++ b/lib/librte_ring/rte_ring_c11_mem.h @@ -58,4 +58,5 @@ __rte_ring_move_prod_head(struct rte_ring *r, unsigned int is_sp, { const uint32_t capacity = r->capacity; + uint32_t cons_tail; unsigned int max = n; int success; @@ -68,11 +69,16 @@ __rte_ring_move_prod_head(struct rte_ring *r, unsigned int is_sp, __ATOMIC_ACQUIRE); - /* - * The subtraction is done between two unsigned 32bits value + /* load-acquire synchronize with store-release of ht->tail + * in update_tail. + */ + cons_tail = __atomic_load_n(&r->cons.tail, + __ATOMIC_ACQUIRE); + + /* 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 - *old_head); + *free_entries = (capacity + cons_tail - *old_head); /* check that we have enough room in ring */ @@ -126,4 +132,5 @@ __rte_ring_move_cons_head(struct rte_ring *r, int is_sc, { unsigned int max = n; + uint32_t prod_tail; int success; @@ -132,7 +139,14 @@ __rte_ring_move_cons_head(struct rte_ring *r, int is_sc, /* Restore n as it may change every loop */ n = max; + *old_head = __atomic_load_n(&r->cons.head, __ATOMIC_ACQUIRE); + /* this load-acquire synchronize with store-release of ht->tail + * in update_tail. + */ + prod_tail = __atomic_load_n(&r->prod.tail, + __ATOMIC_ACQUIRE); + /* The subtraction is done between two unsigned 32bits value * (the result is always modulo 32 bits even if we have @@ -140,5 +154,5 @@ __rte_ring_move_cons_head(struct rte_ring *r, int is_sc, * and size(ring)-1. */ - *entries = (r->prod.tail - *old_head); + *entries = (prod_tail - *old_head); /* Set the actual entries for dequeue */ -- 2.19.0 --- Diff of the applied patch vs upstream commit (please double-check if non-empty: --- --- - 2018-11-23 10:22:54.936963738 +0000 +++ 0026-ring-c11-synchronize-load-and-store-of-the-tail.patch 2018-11-23 10:22:54.000000000 +0000 @@ -1,8 +1,10 @@ -From 9ed877062898dd1eb7150bca56ebbd5329e96ee4 Mon Sep 17 00:00:00 2001 +From f57da12160802dfe41ef7b82c7d41316769d9179 Mon Sep 17 00:00:00 2001 From: Gavin Hu Date: Fri, 2 Nov 2018 19:21:27 +0800 Subject: [PATCH] ring/c11: synchronize load and store of the tail +[ upstream commit 9ed877062898dd1eb7150bca56ebbd5329e96ee4 ] + Synchronize the load-acquire of the tail and the store-release within update_tail, the store release ensures all the ring operations, enqueue or dequeue, are seen by the observers on the other side as soon @@ -40,7 +42,6 @@ MP/MC bulk enq/dequeue (size: 32): 3.26 Fixes: 39368ebfc6 ("ring: introduce C11 memory model barrier option") -Cc: stable@dpdk.org Signed-off-by: Gavin Hu Reviewed-by: Honnappa Nagarahalli