From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756773AbaHHUl5 (ORCPT ); Fri, 8 Aug 2014 16:41:57 -0400 Received: from youngberry.canonical.com ([91.189.89.112]:40137 "EHLO youngberry.canonical.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756491AbaHHUlt (ORCPT ); Fri, 8 Aug 2014 16:41:49 -0400 From: Kamal Mostafa To: linux-kernel@vger.kernel.org, stable@vger.kernel.org, kernel-team@lists.ubuntu.com Cc: Benjamin LaHaise , Jens Axboe , Kamal Mostafa Subject: [PATCH 3.13 131/259] aio: protect reqs_available updates from changes in interrupt handlers Date: Fri, 8 Aug 2014 13:39:14 -0700 Message-Id: <1407530482-4483-132-git-send-email-kamal@canonical.com> X-Mailer: git-send-email 1.9.1 In-Reply-To: <1407530482-4483-1-git-send-email-kamal@canonical.com> References: <1407530482-4483-1-git-send-email-kamal@canonical.com> X-Extended-Stable: 3.13 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 3.13.11.6 -stable review patch. If anyone has any objections, please let me know. ------------------ From: Benjamin LaHaise commit 263782c1c95bbddbb022dc092fd89a36bb8d5577 upstream. As of commit f8567a3845ac05bb28f3c1b478ef752762bd39ef it is now possible to have put_reqs_available() called from irq context. While put_reqs_available() is per cpu, it did not protect itself from interrupts on the same CPU. This lead to aio_complete() corrupting the available io requests count when run under a heavy O_DIRECT workloads as reported by Robert Elliott. Fix this by disabling irq updates around the per cpu batch updates of reqs_available. Many thanks to Robert and folks for testing and tracking this down. Reported-by: Robert Elliot Tested-by: Robert Elliot Signed-off-by: Benjamin LaHaise Cc: Jens Axboe , Christoph Hellwig Signed-off-by: Kamal Mostafa --- fs/aio.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/fs/aio.c b/fs/aio.c index 19e7d95..a0d9e43 100644 --- a/fs/aio.c +++ b/fs/aio.c @@ -816,16 +816,20 @@ void exit_aio(struct mm_struct *mm) static void put_reqs_available(struct kioctx *ctx, unsigned nr) { struct kioctx_cpu *kcpu; + unsigned long flags; preempt_disable(); kcpu = this_cpu_ptr(ctx->cpu); + local_irq_save(flags); kcpu->reqs_available += nr; + while (kcpu->reqs_available >= ctx->req_batch * 2) { kcpu->reqs_available -= ctx->req_batch; atomic_add(ctx->req_batch, &ctx->reqs_available); } + local_irq_restore(flags); preempt_enable(); } @@ -833,10 +837,12 @@ static bool get_reqs_available(struct kioctx *ctx) { struct kioctx_cpu *kcpu; bool ret = false; + unsigned long flags; preempt_disable(); kcpu = this_cpu_ptr(ctx->cpu); + local_irq_save(flags); if (!kcpu->reqs_available) { int old, avail = atomic_read(&ctx->reqs_available); @@ -855,6 +861,7 @@ static bool get_reqs_available(struct kioctx *ctx) ret = true; kcpu->reqs_available--; out: + local_irq_restore(flags); preempt_enable(); return ret; } -- 1.9.1