public inbox for gdb-prs@sourceware.org
help / color / mirror / Atom feed
From: "aldyh at redhat dot com" <sourceware-bugzilla@sourceware.org>
To: gdb-prs@sourceware.org
Subject: [Bug gdb/18142] New: dprintf to stderr with no stderr DIE will segfault
Date: Wed, 18 Mar 2015 19:36:00 -0000	[thread overview]
Message-ID: <bug-18142-4717@http.sourceware.org/bugzilla/> (raw)

https://sourceware.org/bugzilla/show_bug.cgi?id=18142

            Bug ID: 18142
           Summary: dprintf to stderr with no stderr DIE will segfault
           Product: gdb
           Version: HEAD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: gdb
          Assignee: unassigned at sourceware dot org
          Reporter: aldyh at redhat dot com

I have a faulty gcc which is failing to generate a DIE for stderr in this
sample:

struct _IO_FILE;
extern struct _IO_FILE *stderr;
extern int fprintf (struct _IO_FILE *__restrict __stream,
      const char *__restrict __format, ...);

int main ()
{
  fprintf (stderr, "also to stderr\n");
  int local = 8;
  return local;
}

When using dprintf to stderr, gdb will segfault.  Since I can print "stderr"
from the gdb command line, presumably gdb is using its internal stderr (which
is an int not a FILE *), and dying.

Perhaps this is a case of the user being stupid, but it seems like gdb should
be able to figure out that dprintf should not use the builtin stderr, which
AFAICT, is just complete garbage.

Feel free to ignore this :).  I'm not that attached to this bug, and I'm fixing
gcc anyhow :).

(gdb) b main
Breakpoint 1 at 0x40055e: file x.i, line 8.
(gdb) r
Starting program: /home/build/dearly/gcc/bad.out 

Breakpoint 1, main () at x.i:8
8         fprintf (stderr, "also to stderr\n");
(gdb) p stderr
$1 = -136485120
(gdb) dprintf 10,"At foo entry\n"
Dprintf 2 at 0x400583: file x.i, line 10.
(gdb) set dprintf-style call
(gdb) set dprintf-function fprintf
(gdb) set dprintf-channel stderr
(gdb) run
The program being debugged has been started already.
Start it from the beginning? (y or n) y
Starting program: /home/build/dearly/gcc/bad.out 

Breakpoint 1, main () at x.i:8
8         fprintf (stderr, "also to stderr\n");
(gdb) print stderr
$2 = -136485120
(gdb) c
Continuing.
also to stderr

Program received signal SIGSEGV, Segmentation fault.

-- 
You are receiving this mail because:
You are on the CC list for the bug.


             reply	other threads:[~2015-03-18 19:29 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-18 19:36 aldyh at redhat dot com [this message]
2015-03-18 20:55 ` [Bug gdb/18142] " aldyh at redhat dot com

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=bug-18142-4717@http.sourceware.org/bugzilla/ \
    --to=sourceware-bugzilla@sourceware.org \
    --cc=gdb-prs@sourceware.org \
    /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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).