From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id 7DD67385828D; Fri, 12 Jan 2024 10:12:18 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 7DD67385828D DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1705054338; bh=ZXq0dcECK4CVpUxfv6XC4NbAUzLirMbSFaWrqI88x9U=; h=From:To:Subject:Date:In-Reply-To:References:From; b=ilY+ixHp968PQYXPV6aFcxpv5CRGWgUaAGrcgBJITGUZbTTZo8E6+Kq/Y8wkM9L9d Q7hu/IwaXmLSFcwNwPgblTBOFAEV5VWDUipMjj4fPb+IHCH6uHlJXmySGTVEFvD75g ycujQLW9b8S6jncz4ROV6qaSTG9z7gJMY3sA0rPo= From: "blarsen at redhat dot com" To: gdb-prs@sourceware.org Subject: [Bug record/22725] "run" seems to reset btrace recording Date: Fri, 12 Jan 2024 10:12:17 +0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: gdb X-Bugzilla-Component: record X-Bugzilla-Version: unknown X-Bugzilla-Keywords: X-Bugzilla-Severity: normal X-Bugzilla-Who: blarsen at redhat dot com 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: 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=3D22725 Guinevere Larsen changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |blarsen at redhat dot com --- Comment #1 from Guinevere Larsen --- I still see this issue on current GDB, so this is still a valid bug Adding "pt" at the end of "record btrace" is not required, that is complete= ly unrelated to your issue. Also, btrace is incidental, this issue also occurs with record full. I think the problem is that when you "run" or "start" the inferior, the tar= get stack gets re-set and there is nothing in GDB that checks if there was a re= cord target that should be added again. A related problem is that the target has to have started running before we = can start recording. It would be nice if you could set a target to record before even running. The issue with solving this comes from the question: when should the record= ing target be added? once we enter main? Once we enter __libc_start_main? I wou= ld argue that most users would want the former, but that is a decision that ne= eds to be made, and properly handled in the run and start codes. --=20 You are receiving this mail because: You are on the CC list for the bug.=