From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id 0BCCE3858419; Mon, 7 Nov 2022 18:23:10 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 0BCCE3858419 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1667845391; bh=ctyZPA/y6Iv2wqoF34YAClM18JOgJOa586c5x5DnvAE=; h=From:To:Subject:Date:In-Reply-To:References:From; b=xhJgYJ4avbFO2tjs0Tb7tdqw0K2x+MsO7+Gfxoeo4pX5gyYywf8dyCTgOO1Sho6nh x1Mat3502i5HrIFLX0Y9KCNyyfna8D3O0l01A84FIvaKabL5jENjb2zniJlmsBsgYb GG0RUJMRSdIqOl9i+vbb78dAE0LN+3OHcn8L204E= From: "sourceware-bugzilla at djr61 dot uk" 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, 07 Nov 2022 18:23:06 +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: sourceware-bugzilla at djr61 dot uk 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: cc 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 Daniel Richman changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |sourceware-bugzilla at djr= 61 dot u | |k --- Comment #52 from Daniel Richman -= -- Hello, For full disclosure, I work at the same company as Eric and Malte. Eric and I were investigating reproducing this issue after we started regul= arly (~once a day) hitting it in one of our production systems, despite said sys= tem running against a glibc that has the patch in attachment 12484 (i.e., the p= atch that Ubuntu has adopted); it was preventing us from upgrading production to= a more recent version of the linux distribution we use (the previous version = used a glibc older than 2.25). Unfortunately, said system system is large and proprietary so we couldn=E2= =80=99t simply provide a copy of it, hence the reproducer Eric shared. We hope the reproducer (a) is simple enough that you can convince yourself that the C is correct, (b) makes it possible for you to observe the deadlocks for yourself (on that specific hardware and kernel settings that Eric mentions), and (c) exhibits said deadlock in a reasonable amount of time that you don=E2=80=99= t have to wait too long. I just wanted to tell this story to try and explain that this wasn=E2=80=99= t just an academic exercise: we were regularly running into the bug, and would be very happy to see a patch reviewed and merged so that we don=E2=80=99t need to m= aintain a separately patched glibc internally. Thanks, Daniel --=20 You are receiving this mail because: You are on the CC list for the bug.=