From: Aleksandar Ristovski <aristovski@qnx.com>
To: gdb-patches@sources.redhat.com
Cc: gdb-patches@sources.redhat.com
Subject: Re: [patch] Assert when 'break' with no arguments
Date: Tue, 14 Feb 2012 18:53:00 -0000 [thread overview]
Message-ID: <4F3AA28A.6090501@qnx.com> (raw)
In-Reply-To: <jhe7kc$d91$1@dough.gmane.org>
Please disregard line
"# This file was written by Rob Savoye. (rob@cygnus.com)"
from the test, I will remove it before commit.
Sorry about that.
On 12-02-14 01:02 PM, Aleksandar Ristovski wrote:
> Hello,
>
> An issue exists where if 'break' command is issued while on a line that
> underwent inlining by the compiler gdb asserts with
>
> Assertion `sal.pspace != NULL'
>
> I narrowed down the issue to frame.c:find_frame_sal, combined with
> stack.c:set_last_displayed_sal and in the view of print_frame_info.
>
> The fix proposed here would be to properly initialize 'sal' in
> find_frmae_sal. Additional check is performed in set_last_displayed_sal
> to make sure we do not set last_displayed_* vars and validate them if
> pspace is NULL as, clearly, the rest of the code expects it to be
> properly set.
>
> I identified the same issue in 7.2, 7.3.1, 7.4 and HEAD. I have not
> checked earlier versions.
>
> Test suite did not show regressions, and new test passes where it would
> fail without the patch.
>
>
> ChangeLog:
> 2012-02-14 Aleksandar Ristovski <aristovski@qnx.com>
>
> * frame.c (find_frame_sal): Initialise sal->pspace field from frame data.
> * stack.c (set_last_displayed_sal): Perform sanity check of the data
> passed in, in particular, validate that PSPACE is not NULL if requesting
> valid last_displayed_* data.
>
>
> Test suite ChangeLOg:
> 2012-02-14 Aleksandar Ristovski <aristovski@qnx.com>
>
> * gdb.base/break-inline.exp: New test.
> * gdb.base/break-inline.c: New test.
>
>
> Attached are patch for the fix and tests.
>
>
>
> Thank you,
>
> Aleksandar Ristovski
> QNX Software Systems
WARNING: multiple messages have this Message-ID
From: Aleksandar Ristovski <aristovski@qnx.com>
Cc: <gdb-patches@sources.redhat.com>
Subject: Re: [patch] Assert when 'break' with no arguments
Date: Tue, 14 Feb 2012 18:06:00 -0000 [thread overview]
Message-ID: <4F3AA28A.6090501@qnx.com> (raw)
Message-ID: <20120214180600.q5EspcLQr3KrgW9DCm4WNYl-CFdnqUs02JEcwFZhptM@z> (raw)
In-Reply-To: <jhe7kc$d91$1@dough.gmane.org>
Please disregard line
"# This file was written by Rob Savoye. (rob@cygnus.com)"
from the test, I will remove it before commit.
Sorry about that.
On 12-02-14 01:02 PM, Aleksandar Ristovski wrote:
> Hello,
>
> An issue exists where if 'break' command is issued while on a line that
> underwent inlining by the compiler gdb asserts with
>
> Assertion `sal.pspace != NULL'
>
> I narrowed down the issue to frame.c:find_frame_sal, combined with
> stack.c:set_last_displayed_sal and in the view of print_frame_info.
>
> The fix proposed here would be to properly initialize 'sal' in
> find_frmae_sal. Additional check is performed in set_last_displayed_sal
> to make sure we do not set last_displayed_* vars and validate them if
> pspace is NULL as, clearly, the rest of the code expects it to be
> properly set.
>
> I identified the same issue in 7.2, 7.3.1, 7.4 and HEAD. I have not
> checked earlier versions.
>
> Test suite did not show regressions, and new test passes where it would
> fail without the patch.
>
>
> ChangeLog:
> 2012-02-14 Aleksandar Ristovski <aristovski@qnx.com>
>
> * frame.c (find_frame_sal): Initialise sal->pspace field from frame data.
> * stack.c (set_last_displayed_sal): Perform sanity check of the data
> passed in, in particular, validate that PSPACE is not NULL if requesting
> valid last_displayed_* data.
>
>
> Test suite ChangeLOg:
> 2012-02-14 Aleksandar Ristovski <aristovski@qnx.com>
>
> * gdb.base/break-inline.exp: New test.
> * gdb.base/break-inline.c: New test.
>
>
> Attached are patch for the fix and tests.
>
>
>
> Thank you,
>
> Aleksandar Ristovski
> QNX Software Systems
next prev parent reply other threads:[~2012-02-14 18:10 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-14 18:03 Aleksandar Ristovski
2012-02-14 18:53 ` Aleksandar Ristovski [this message]
2012-02-14 18:06 ` Aleksandar Ristovski
2012-02-14 19:23 ` Joel Brobecker
2012-02-14 20:09 ` Aleksandar Ristovski
2012-02-14 20:18 ` Alfred M. Szmidt
2012-02-14 22:37 ` Aleksandar Ristovski
2012-02-15 17:15 ` Tom Tromey
2012-02-15 20:06 ` Aleksandar Ristovski
2012-02-19 16:42 ` [commit] testsuite: Fix break-inline.exp with gdbserver Jan Kratochvil
2012-02-24 15:52 ` [patch] Assert when 'break' with no arguments Pedro Alves
2012-02-24 16:00 ` Aleksandar Ristovski
2012-02-24 16:09 ` Pedro Alves
2012-02-24 16:18 ` Aleksandar Ristovski
2012-02-24 17:16 ` Pedro Alves
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4F3AA28A.6090501@qnx.com \
--to=aristovski@qnx.com \
--cc=gdb-patches@sources.redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox