From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id D31AA3858C2F; Wed, 3 Jan 2024 10:59:00 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org D31AA3858C2F DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1704279540; bh=B1ZsaO9nskwW6nf5GDi9TaUl+JJJJPmYYtbN3dy7vjY=; h=From:To:Subject:Date:In-Reply-To:References:From; b=s+d8QTswHL5+WVEbAZWRntdRNf+Wy4GnlILILVoeyOQC9usNFsiqS0pmMTDHA8v95 oDN6OisToIty5LUWPKANdieUg/JOGFIssV0gqzL3wx3tLyBw3TA9WXBoZ3eEUB6WaO Niznov17CQGLJs++TPavRXqAM6BczQzMr9Z4qWlw= From: "vries at gcc dot gnu.org" To: gdb-prs@sourceware.org Subject: [Bug gdb/31203] [gdb] FAIL: gdb.base/kill-during-detach.exp: exit_p=true: checkpoint_p=true: python kill_and_detach() Date: Wed, 03 Jan 2024 10:58:58 +0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: gdb X-Bugzilla-Component: gdb X-Bugzilla-Version: HEAD X-Bugzilla-Keywords: X-Bugzilla-Severity: normal X-Bugzilla-Who: vries at gcc dot gnu.org X-Bugzilla-Status: NEW X-Bugzilla-Resolution: X-Bugzilla-Priority: P2 X-Bugzilla-Assigned-To: unassigned at sourceware dot org 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=3D31203 --- Comment #6 from Tom de Vries --- I've investigated why this doesn't trigger on x86_64. I found that in the x86_64 case, we've only got fds 0, 1 and 2, and none of them get=20 their file pointer restored, so inferior lseek is not called during linux_fork_detach. In contrast, in the m1 case, we've got fds 42 and 52, whose file pointers a= re to be restored to 0, so inferior lseek is called during linux_fork_detach. File descriptors in more detail: ... $ ls -la /proc/3487043/fd total 0 dr-x------. 2 vries vries 5 3 jan 11:41 . dr-xr-xr-x. 9 vries vries 0 3 jan 11:41 .. lrwx------. 1 vries vries 64 3 jan 11:41 0 -> /dev/pts/1 lrwx------. 1 vries vries 64 3 jan 11:41 1 -> /dev/pts/1 lrwx------. 1 vries vries 64 3 jan 11:41 2 -> /dev/pts/1 lr-x------. 1 vries vries 64 3 jan 11:41 42 -> anon_inode:inotify lrwx------. 1 vries vries 64 3 jan 11:41 52 -> '/memfd:pulseaudio (deleted= )' ... So, my guess is that opening a file before the checkpoint and reading somet= hing from the file after the checkpoint will also trigger this on x86_64. --=20 You are receiving this mail because: You are on the CC list for the bug.=