From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 1503448AEF for ; Wed, 12 Nov 2025 17:55:45 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 0E7DD40DF5; Wed, 12 Nov 2025 17:55:45 +0100 (CET) Received: from mail-wm1-f41.google.com (mail-wm1-f41.google.com [209.85.128.41]) by mails.dpdk.org (Postfix) with ESMTP id BB8FD40DFD for ; Wed, 12 Nov 2025 17:55:43 +0100 (CET) Received: by mail-wm1-f41.google.com with SMTP id 5b1f17b1804b1-47758595eecso5772695e9.0 for ; Wed, 12 Nov 2025 08:55:43 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1762966543; x=1763571343; darn=dpdk.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=vLyrrquqietDszbzmXPanU2lfesyj/mrdk+xw65dydA=; b=f9swCQOES0Aj5dtM8Fm4wkrUcIgQAy+ZXJ3LJf+KL1UUpjzstRNASxZyA8V1BJrTbJ iFpei1BxRCk790D6Aqie3a3pbBTQU/MWfAlUbeAZfLvBPWSumshIP1qPG8Ze4N8Z2ndn zN45DcWV5RmX0EKeQA1JUri1MR276aCMEz+66REZC4CUa4lVglf5jjZMIPpWy17FXXIp kZsUD54N0MP5ZNdJ1LGEdc79KEq9noYoGMNpOq835zFIyZf/2fN7UDPnLTrS57v0wj4g eCp9SuMyNEciy6H1147D545fcMealktVdvedahGpTvoUstvk3X487dObs/DyWl3jZqp7 JFxw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1762966543; x=1763571343; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to; bh=vLyrrquqietDszbzmXPanU2lfesyj/mrdk+xw65dydA=; b=VM0M+eLtA6exfDd9OhuEPUwdZtk0//BTSODbBhuryS5rPLCTUrlkRE69K4FeTgNXEy iPSPvwkBl9pRCAlxa/a9UOK6OxtkRCH7Upgyxbz0tgB4VtSUJssmZ8FCDgN503q7gpnp JaAudMBVmBIqakePypIhyfyA98oZVaZNmyhsYl6Q98G7P1sT1o+9gRzr3ki3kuJwPIgn hrOU9pA1nWE5qAKMC1y5yJVtt+YIsxtD+Ek5ClsVNzlNqvcpU882u+K3NGj6o6TA809e Kk2V/OZUN7KAqnB5inmM6I06ets42X8a1YV0av7xAZpJMF2CpyBHlpL03xvK/+HbxAZ8 0Rkg== X-Forwarded-Encrypted: i=1; AJvYcCVGz+AxgboA7Rpm0Pa6QwxNfY9CIw8jBtJVKDRY7HRGylp/7GoRdQo7jnjh8Jrx8ZM3UqQITkg=@dpdk.org X-Gm-Message-State: AOJu0YywXKZtIU5xAvG+Gqqejo52a5LeixIQq3vCoPoS5M/FZSCTeC+S YhvDIhHZav/129m1hg387dZRhy/Emwa3BKlX8E/90YnC+7x3Mn89fXt1 X-Gm-Gg: ASbGncvw/H7YMFgnGYfTiz2otFTmeVWnOwzyGaYFYvYzuvgCGH7TQvtKUGRTIj4a/on w7tf+pCFM+lR6Wt4WCyICSt20cuL/ZbNzuwFvdT9JDGr+2r97VPDdEFk+wIeOlQaVX+WmseEf3N gJ4kMHWK4v5FevjRF2dk8LG1ECVmu5d8wESKEalOFukSucI8pjHFJIuXuEag6Ri0858m1qG1p6f KxWWi+A7TOP2gOS+f5EIppUcwazSlZiMd/ninmkEZlYaObtrIfWKLtS+oeEJLZkD+xSfsaV6cW4 OtpUXiKAGR7XjYcl7LRbPP7zSlCaRtSo8e4QJOq+haQVP3PD6tKMUpAs/nzaogKvUyAAc+PSZtj GUke0CctXWArfu2nEi5MqP7sAJIldEMy0XDPzDVPYOwt5wvDP5AP9HAJiMK90G+tOUZRCT0ON6z hpvVxRFQ== X-Google-Smtp-Source: AGHT+IGHmSS3yubrjban6+o6hvwtRQPW6xrvpMcaWwHm5MZuE7ncEsjY3QQ7u0wngFoido1Hw26uYA== X-Received: by 2002:a05:600c:3b24:b0:471:9da:5252 with SMTP id 5b1f17b1804b1-477870c321emr34498135e9.29.1762966543173; Wed, 12 Nov 2025 08:55:43 -0800 (PST) Received: from localhost ([2a01:4b00:d036:ae00:a397:14bc:5982:5745]) by smtp.gmail.com with UTF8SMTPSA id ffacd0b85a97d-42b303386f1sm26294888f8f.3.2025.11.12.08.55.42 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 12 Nov 2025 08:55:42 -0800 (PST) From: luca.boccassi@gmail.com To: Wathsala Vithanage Cc: Ola Liljedahl , Honnappa Nagarahalli , Dhruv Tripathi , Konstantin Ananyev , dpdk stable Subject: patch 'ring: establish safe partial order in default mode' has been queued to stable release 22.11.11 Date: Wed, 12 Nov 2025 16:53:02 +0000 Message-ID: <20251112165308.1618107-48-luca.boccassi@gmail.com> X-Mailer: git-send-email 2.47.3 In-Reply-To: <20251112165308.1618107-1-luca.boccassi@gmail.com> References: <20251027162001.3710450-79-luca.boccassi@gmail.com> <20251112165308.1618107-1-luca.boccassi@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-BeenThere: stable@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: patches for DPDK stable branches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: stable-bounces@dpdk.org Hi, FYI, your patch has been queued to stable release 22.11.11 Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet. It will be pushed if I get no objections before 11/14/25. 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/bluca/dpdk-stable This queued commit can be viewed at: https://github.com/bluca/dpdk-stable/commit/8e64e64659fe628f6b7ce903b67a6c8d271da524 Thanks. Luca Boccassi --- >From 8e64e64659fe628f6b7ce903b67a6c8d271da524 Mon Sep 17 00:00:00 2001 From: Wathsala Vithanage Date: Tue, 11 Nov 2025 18:37:17 +0000 Subject: [PATCH] ring: establish safe partial order in default mode MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [ upstream commit a4ad0eba9def1d1d071da8afe5e96eb2a2e0d71f ] The function __rte_ring_headtail_move_head() assumes that the barrier (fence) between the load of the head and the load-acquire of the opposing tail guarantees the following: if a first thread reads tail and then writes head and a second thread reads the new value of head and then reads tail, then it should observe the same (or a later) value of tail. This assumption is incorrect under the C11 memory model. If the barrier (fence) is intended to establish a total ordering of ring operations, it fails to do so. Instead, the current implementation only enforces a partial ordering, which can lead to unsafe interleavings. In particular, some partial orders can cause underflows in free slot or available element computations, potentially resulting in data corruption. The issue manifests when a CPU first acts as a producer and later as a consumer. In this scenario, the barrier assumption may fail when another core takes the consumer role. A Herd7 litmus test in C11 can demonstrate this violation. The problem has not been widely observed so far because: (a) on strong memory models (e.g., x86-64) the assumption holds, and (b) on relaxed models with RCsc semantics the ordering is still strong enough to prevent hazards. The problem becomes visible only on weaker models, when load-acquire is implemented with RCpc semantics (e.g. some AArch64 CPUs which support the LDAPR and LDAPUR instructions). Three possible solutions exist: 1. Strengthen ordering by upgrading release/acquire semantics to sequential consistency. This requires using seq-cst for stores, loads, and CAS operations. However, this approach introduces a significant performance penalty on relaxed-memory architectures. 2. Establish a safe partial order by enforcing a pair-wise happens-before relationship between thread of same role by changing the CAS and the preceding load of the head by converting them to release and acquire respectively. This approach makes the original barrier assumption unnecessary and allows its removal. 3. Retain partial ordering but ensure only safe partial orders are committed. This can be done by detecting underflow conditions (producer < consumer) and quashing the update in such cases. This approach makes the original barrier assumption unnecessary and allows its removal. This patch implements solution (2) to preserve the “enqueue always succeeds” contract expected by dependent libraries (e.g., mempool). While solution (3) offers higher performance, adopting it now would break that assumption. Fixes: 49594a63147a9 ("ring/c11: relax ordering for load and store of the head") Signed-off-by: Wathsala Vithanage Signed-off-by: Ola Liljedahl Reviewed-by: Honnappa Nagarahalli Reviewed-by: Dhruv Tripathi Acked-by: Konstantin Ananyev Tested-by: Konstantin Ananyev --- lib/ring/rte_ring_c11_pvt.h | 37 +++++++++++++++++++++++++++++-------- 1 file changed, 29 insertions(+), 8 deletions(-) diff --git a/lib/ring/rte_ring_c11_pvt.h b/lib/ring/rte_ring_c11_pvt.h index f895950df4..5c04a001e1 100644 --- a/lib/ring/rte_ring_c11_pvt.h +++ b/lib/ring/rte_ring_c11_pvt.h @@ -24,6 +24,11 @@ __rte_ring_update_tail(struct rte_ring_headtail *ht, uint32_t old_val, if (!single) rte_wait_until_equal_32(&ht->tail, old_val, __ATOMIC_RELAXED); + /* + * R0: Establishes a synchronizing edge with load-acquire of tail at A1. + * Ensures that memory effects by this thread on ring elements array + * is observed by a different thread of the other type. + */ __atomic_store_n(&ht->tail, new_val, __ATOMIC_RELEASE); } @@ -61,16 +66,23 @@ __rte_ring_move_prod_head(struct rte_ring *r, unsigned int is_sp, unsigned int max = n; int success; - *old_head = __atomic_load_n(&r->prod.head, __ATOMIC_RELAXED); + /* + * A0: Establishes a synchronizing edge with R1. + * Ensure that this thread observes same values + * to stail observed by the thread that updated + * d->head. + * If not, an unsafe partial order may ensue. + */ + *old_head = __atomic_load_n(&r->prod.head, __ATOMIC_ACQUIRE); do { /* Reset n to the initial burst count */ n = max; - /* Ensure the head is read before tail */ - __atomic_thread_fence(__ATOMIC_ACQUIRE); - - /* load-acquire synchronize with store-release of ht->tail - * in update_tail. + /* + * A1: Establishes a synchronizing edge with R0. + * Ensures that other thread's memory effects on + * ring elements array is observed by the time + * this thread observes its tail update. */ cons_tail = __atomic_load_n(&r->cons.tail, __ATOMIC_ACQUIRE); @@ -170,10 +182,19 @@ __rte_ring_move_cons_head(struct rte_ring *r, int is_sc, r->cons.head = *new_head, success = 1; else /* on failure, *old_head will be updated */ + /* + * R1/A2. + * R1: Establishes a synchronizing edge with A0 of a + * different thread. + * A2: Establishes a synchronizing edge with R1 of a + * different thread to observe same value for stail + * observed by that thread on CAS failure (to retry + * with an updated *old_head). + */ success = __atomic_compare_exchange_n(&r->cons.head, old_head, *new_head, - 0, __ATOMIC_RELAXED, - __ATOMIC_RELAXED); + 0, __ATOMIC_RELEASE, + __ATOMIC_ACQUIRE); } while (unlikely(success == 0)); return n; } -- 2.47.3 --- Diff of the applied patch vs upstream commit (please double-check if non-empty: --- --- - 2025-11-12 16:20:42.793162159 +0000 +++ 0048-ring-establish-safe-partial-order-in-default-mode.patch 2025-11-12 16:20:41.007718917 +0000 @@ -1 +1 @@ -From a4ad0eba9def1d1d071da8afe5e96eb2a2e0d71f Mon Sep 17 00:00:00 2001 +From 8e64e64659fe628f6b7ce903b67a6c8d271da524 Mon Sep 17 00:00:00 2001 @@ -8,0 +9,2 @@ +[ upstream commit a4ad0eba9def1d1d071da8afe5e96eb2a2e0d71f ] + @@ -58 +59,0 @@ -Cc: stable@dpdk.org @@ -71 +72 @@ -index b9388af0da..07b6efc416 100644 +index f895950df4..5c04a001e1 100644 @@ -74,3 +75,3 @@ -@@ -36,6 +36,11 @@ __rte_ring_update_tail(struct rte_ring_headtail *ht, uint32_t old_val, - rte_wait_until_equal_32((uint32_t *)(uintptr_t)&ht->tail, old_val, - rte_memory_order_relaxed); +@@ -24,6 +24,11 @@ __rte_ring_update_tail(struct rte_ring_headtail *ht, uint32_t old_val, + if (!single) + rte_wait_until_equal_32(&ht->tail, old_val, __ATOMIC_RELAXED); @@ -83 +84 @@ - rte_atomic_store_explicit(&ht->tail, new_val, rte_memory_order_release); + __atomic_store_n(&ht->tail, new_val, __ATOMIC_RELEASE); @@ -86,2 +87 @@ -@@ -77,17 +82,24 @@ __rte_ring_headtail_move_head(struct rte_ring_headtail *d, - int success; +@@ -61,16 +66,23 @@ __rte_ring_move_prod_head(struct rte_ring *r, unsigned int is_sp, @@ -88,0 +89 @@ + int success; @@ -89,0 +91 @@ +- *old_head = __atomic_load_n(&r->prod.head, __ATOMIC_RELAXED); @@ -97,3 +99 @@ - *old_head = rte_atomic_load_explicit(&d->head, -- rte_memory_order_relaxed); -+ rte_memory_order_acquire); ++ *old_head = __atomic_load_n(&r->prod.head, __ATOMIC_ACQUIRE); @@ -105 +105 @@ -- rte_atomic_thread_fence(rte_memory_order_acquire); +- __atomic_thread_fence(__ATOMIC_ACQUIRE); @@ -115,6 +115,6 @@ - stail = rte_atomic_load_explicit(&s->tail, - rte_memory_order_acquire); -@@ -113,10 +125,19 @@ __rte_ring_headtail_move_head(struct rte_ring_headtail *d, - success = 1; - } else - /* on failure, *old_head is updated */ + cons_tail = __atomic_load_n(&r->cons.tail, + __ATOMIC_ACQUIRE); +@@ -170,10 +182,19 @@ __rte_ring_move_cons_head(struct rte_ring *r, int is_sc, + r->cons.head = *new_head, success = 1; + else + /* on failure, *old_head will be updated */ @@ -130,6 +130,6 @@ - success = rte_atomic_compare_exchange_strong_explicit( - &d->head, old_head, *new_head, -- rte_memory_order_relaxed, -- rte_memory_order_relaxed); -+ rte_memory_order_release, -+ rte_memory_order_acquire); + success = __atomic_compare_exchange_n(&r->cons.head, + old_head, *new_head, +- 0, __ATOMIC_RELAXED, +- __ATOMIC_RELAXED); ++ 0, __ATOMIC_RELEASE, ++ __ATOMIC_ACQUIRE);