From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-8.4 required=3.0 tests=DKIMWL_WL_HIGH,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH, MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id C506FC33CB1 for ; Fri, 17 Jan 2020 16:53:55 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 9A3222072E for ; Fri, 17 Jan 2020 16:53:55 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="QfVrvNR8" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726901AbgAQQxy (ORCPT ); Fri, 17 Jan 2020 11:53:54 -0500 Received: from us-smtp-delivery-1.mimecast.com ([207.211.31.120]:52326 "EHLO us-smtp-1.mimecast.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1726559AbgAQQxy (ORCPT ); Fri, 17 Jan 2020 11:53:54 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1579280032; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding; bh=+D+2i5u8m+6VdqENi+Zlil9tmBeW+a+J8CirwEbN+O0=; b=QfVrvNR887ItjCdiQ3pkxdn0cW5b1wfCA5g/2Dk+hnAUzXhfdDwPfYCgOc1NZicilAs3k5 x58tfuQ+ORNNenMHqhOpvgugZWbTWj404MJCNgiD4vn2gzXyczta0Hu0hWzxRoJG7Zzd/A cej8DV/53J4brqR3wB03O3WXGyxpr9w= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-153-ketbYkMvMuyd-WQZaOp4qw-1; Fri, 17 Jan 2020 11:53:51 -0500 X-MC-Unique: ketbYkMvMuyd-WQZaOp4qw-1 Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id F2233190B2CD; Fri, 17 Jan 2020 16:53:48 +0000 (UTC) Received: from llong.remote.csb (dhcp-17-59.bos.redhat.com [10.18.17.59]) by smtp.corp.redhat.com (Postfix) with ESMTP id 7055B81201; Fri, 17 Jan 2020 16:53:48 +0000 (UTC) To: David Howells Cc: Alexander Viro , "linux-fsdevel@vger.kernel.org" , Linux List Kernel Mailing , Linus Torvalds From: Waiman Long Subject: Performance regression introduced by commit b667b8673443 ("pipe: Advance tail pointer inside of wait spinlock in pipe_read()") Organization: Red Hat Message-ID: Date: Fri, 17 Jan 2020 11:53:48 -0500 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.7.2 MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Language: en-US X-Scanned-By: MIMEDefang 2.79 on 10.5.11.11 Content-Transfer-Encoding: quoted-printable Sender: linux-fsdevel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-fsdevel@vger.kernel.org David, I had found that parallel kernel build became much slower when a 5.5-based kernel is used. On a 2-socket 96-thread x86-64 system, the "make -j88" time increased from less than 3 minutes with the 5.4 kernel to more than double with the 5.5 kernel. So I used bisection to try to find the culprit: b667b867344301e24f21d4a4c844675ff61d89e1 is the first bad commit commit b667b867344301e24f21d4a4c844675ff61d89e1 Author: David Howells Date:=C2=A0=C2=A0 Tue Sep 24 16:09:04 2019 +0100 =C2=A0=C2=A0=C2=A0 pipe: Advance tail pointer inside of wait spinlock in = pipe_read() =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 Advance the pipe ring tail pointer inside of wait spin= lock in pipe_read() =C2=A0=C2=A0=C2=A0 so that the pipe can be written into with kernel notif= ications from =C2=A0=C2=A0=C2=A0 contexts where pipe->mutex cannot be taken. =C2=A0=C2=A0=C2=A0 =C2=A0=C2=A0=C2=A0 Signed-off-by: David Howells diff --git a/fs/pipe.c b/fs/pipe.c index 69afeab8a73a..ea134f69a292 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -325,9 +325,14 @@ pipe_read(struct kiocb *iocb, struct iov_iter *to) =C2=A0 =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 if (!buf-= >len) { =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 pipe_buf_release(pipe, buf); +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 spin_lock_irq(&pipe->wait.lock); =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 tail++; =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 pipe->tail =3D tail; -=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 do_wakeup =3D 1; +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 do_wakeup =3D 0; +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 wake_up_interruptible_sync_poll_locked( +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0 &pipe->wait, EPOLLOUT | EPOLLWRNORM); +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 spin_unlock_irq(&pipe->wait.lock); +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0= =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 kill_fasync(&pipe->fasync_writers, SIGIO, POLL_O =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2= =A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 } I guess the make command may make heavy use of pipe. The adding of spinlock code in your patch may probably over-serialize the pipe operation. Could you achieve the same functionality without adding a lock= ? Cheers, Longman