From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id 720193858297; Mon, 26 Sep 2022 14:32:03 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 720193858297 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1664202723; bh=Jyqhq3qajHD0hUYchTg+4bGI5L+kCRnNH2Qc/KffnnY=; h=From:To:Subject:Date:In-Reply-To:References:From; b=ez5ELmM6nyFQlci+uq3bYChzKVP5zCJ+nRPUuxCSAIjfHpIJF6CQ9ZKZ4iosl5ZfI 9DcU6wHIaqhVZ+azAXykg7ZM1Gg+h2NPWsy7JK+5RtkfJCq016pD+d02ASVwSbdjFQ MlDekomGmvyFeRB7j5UJGCa6v4jI8LDGey+/KYJk= From: "ehagberg at janestreet dot com" To: glibc-bugs@sourceware.org Subject: [Bug nptl/25847] pthread_cond_signal failed to wake up pthread_cond_wait due to a bug in undoing stealing Date: Mon, 26 Sep 2022 14:32:03 +0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: glibc X-Bugzilla-Component: nptl X-Bugzilla-Version: 2.27 X-Bugzilla-Keywords: X-Bugzilla-Severity: normal X-Bugzilla-Who: ehagberg at janestreet dot com X-Bugzilla-Status: UNCONFIRMED X-Bugzilla-Resolution: X-Bugzilla-Priority: P2 X-Bugzilla-Assigned-To: carlos at redhat dot com X-Bugzilla-Target-Milestone: --- X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://sourceware.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 List-Id: https://sourceware.org/bugzilla/show_bug.cgi?id=3D25847 --- Comment #49 from Eric Hagberg --- This is a description of how to run the reproducer (https://sourceware.org/bugzilla/attachment.cgi?id=3D14360) =E2=80=9Cpthread_condition_issue.c=E2=80=9D on AWS: Boot a c5.metal instance running the AMI Rocky-8-ec2-8.6-20220515.0.x86_64-d6577ceb-8ea8-4e0e-84c6-f098fc302e82. Build with =E2=80=9Cgcc pthread_condition_issue.c -lpthread=E2=80=9D. Run the following as root: tuned-adm profile latency-performance. This is required to obtain a timely reproduction. Run the resulting executable 80 times concurrently. With the stock glibc the problem will occur fairly quickly - within about an hour, there should be 7 or more aborts. If the Ubuntu patch (https://sourceware.org/bugzilla/attachment.cgi?id=3D12= 484) has been applied to glibc, the following additional changes should be made = to obtain a timely reproduction: sysctl -w kernel.sched_min_granularity_ns=3D300000 sysctl -w kernel.sched_wakeup_granularity_ns=3D400000 Within about an hour, you should expect to see 3 or 4 aborts. --=20 You are receiving this mail because: You are on the CC list for the bug.=