From patchwork Mon Nov 22 04:11:35 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: yangerkun X-Patchwork-Id: 12631209 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 484D2C433EF for ; Mon, 22 Nov 2021 03:59:24 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233997AbhKVEC3 (ORCPT ); Sun, 21 Nov 2021 23:02:29 -0500 Received: from szxga01-in.huawei.com ([45.249.212.187]:31895 "EHLO szxga01-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230492AbhKVEC2 (ORCPT ); Sun, 21 Nov 2021 23:02:28 -0500 Received: from dggemv711-chm.china.huawei.com (unknown [172.30.72.57]) by szxga01-in.huawei.com (SkyGuard) with ESMTP id 4HyD0t6jRrzcZxf; Mon, 22 Nov 2021 11:54:22 +0800 (CST) Received: from kwepemm600019.china.huawei.com (7.193.23.64) by dggemv711-chm.china.huawei.com (10.1.198.66) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2308.20; Mon, 22 Nov 2021 11:59:21 +0800 Received: from localhost.localdomain (10.175.127.227) by kwepemm600019.china.huawei.com (7.193.23.64) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2308.20; Mon, 22 Nov 2021 11:59:21 +0800 From: yangerkun To: , CC: Subject: [PATCH] pipe: remove needless spin_lock in pipe_read/pipe_write Date: Mon, 22 Nov 2021 12:11:35 +0800 Message-ID: <20211122041135.2450220-1-yangerkun@huawei.com> X-Mailer: git-send-email 2.31.1 MIME-Version: 1.0 X-Originating-IP: [10.175.127.227] X-ClientProxiedBy: dggems701-chm.china.huawei.com (10.3.19.178) To kwepemm600019.china.huawei.com (7.193.23.64) X-CFilter-Loop: Reflected Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org Once enable CONFIG_WATCH_QUEUE, we should protect pipe with pipe->rd_wait.lock since post_one_notification may write pipe from contexts where pipe->mutex cannot be token. But nowdays we will try take it for anycase, it seems needless. Besides, pipe_write will break once it's pipe with O_NOTIFICATION_PIPE, pipe->rd_wait.lock seems no need at all. We make some change base on that, and it can help improve performance for some case like pipe_pst1 in libMicro. ARMv7 for our scene, before this patch: 5483 nsecs/call After this patch: 4854 nsecs/call Signed-off-by: yangerkun --- fs/pipe.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/fs/pipe.c b/fs/pipe.c index 6d4342bad9f1..e8ced0c50824 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -320,14 +320,18 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to) if (!buf->len) { pipe_buf_release(pipe, buf); - spin_lock_irq(&pipe->rd_wait.lock); #ifdef CONFIG_WATCH_QUEUE + if (pipe->watch_queue) + spin_lock_irq(&pipe->rd_wait.lock); if (buf->flags & PIPE_BUF_FLAG_LOSS) pipe->note_loss = true; #endif tail++; pipe->tail = tail; - spin_unlock_irq(&pipe->rd_wait.lock); +#ifdef CONFIG_WATCH_QUEUE + if (pipe->watch_queue) + spin_unlock_irq(&pipe->rd_wait.lock); +#endif } total_len -= chars; if (!total_len) @@ -504,16 +508,11 @@ pipe_write(struct kiocb *iocb, struct iov_iter *from) * it, either the reader will consume it or it'll still * be there for the next write. */ - spin_lock_irq(&pipe->rd_wait.lock); - head = pipe->head; - if (pipe_full(head, pipe->tail, pipe->max_usage)) { - spin_unlock_irq(&pipe->rd_wait.lock); + if (pipe_full(head, pipe->tail, pipe->max_usage)) continue; - } pipe->head = head + 1; - spin_unlock_irq(&pipe->rd_wait.lock); /* Insert it into the buffer array */ buf = &pipe->bufs[head & mask];