[Bug debug/11500] New: [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[Bug debug/11500] New: [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[mec at shout dot net]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500

           Summary: [3.4 regression] gcc HEAD emits an extraneous "line 0"
                    debug line at the start of each function
           Product: gcc
           Version: 3.4
            Status: UNCONFIRMED
          Severity: critical
          Priority: P2
         Component: debug
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: mec at shout dot net
                CC: gcc-bugs at gcc dot gnu dot org
 GCC build triplet: i686-pc-linux-gnu
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: i686-pc-linux-gnu

gcc HEAD emits an extraneous "line 0" debug line at the start of each function.
 This definitely happens with -gstabs+ and I believe it happens with gcc
-gdwarf-2 as well.

John Parkes originally reported this:

  http://gcc.gnu.org/ml/gcc/2003-07/msg00827.html

The problem started with this patch:

  Sun Jun  8 21:27:41 CEST 2003  Jan Hubicka  <jh@suse.cz>

    ...
    * final.c (notice_source_line): Use locators.
    (final_start_function): Set initial source file and line.
    (final_scan_insn): Use locators.
    ...

Before this patch, final_start_function called begin_prologue with the line
number and filename of the function.  Now final_start_function calls
begin_prologue with line=0 and filename=0.

See the attachment for more details.

[Bug debug/11500] [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[mec at shout dot net]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500



------- Additional Comments From mec at shout dot net  2003-07-11 16:15 -------
Created an attachment (id=4388)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=4388&action=view)
Sample source file, generated assembly files, analysis text

[Bug debug/11500] [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[mec at shout dot net]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500


mec at shout dot net changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jh at suse dot cz

[Bug debug/11500] [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[pinskia at physics dot uc dot edu]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500


pinskia at physics dot uc dot edu changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
           Priority|P2                          |P1
   Last reconfirmed|0000-00-00 00:00:00         |2003-07-11 19:59:00
               date|                            |


------- Additional Comments From pinskia at physics dot uc dot edu  2003-07-11 19:59 -------
I can confirm this on the mainline (20030711) but I do not know how this confuses gdb (I 
use gdb 2003-07-11-cvs).  This is a regression over 3.3.1 (20030707) so bumping the 
priority.

[Bug debug/11500] [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[mec at shout dot net]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500



------- Additional Comments From mec at shout dot net  2003-07-11 20:20 -------
Subject: Re:  [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

Sorry, I should specify the regressions.

The regressions in the gdb test suite are:

  # gdb 5.3, gcc -gdwarf-2
  gdb.base/annota1.exp: send SIGUSR1
  gdb.base/condbreak.exp: run until breakpoint at marker1
  gdb.base/condbreak.exp: run until breakpoint at marker2
  gdb.base/ena-dis-br.exp: continue to auto-disabled break marker2
  gdb.base/ena-dis-br.exp: continue to break marker1
  gdb.base/ena-dis-br.exp: continue to break marker1, 2nd time
  gdb.base/funcargs.exp: continue to call6k
  gdb.base/funcargs.exp: print st
  gdb.base/funcargs.exp: run to call5a
  gdb.base/funcargs.exp: run to hitbottom
  gdb.base/watchpoint.exp: run to marker1 in test_disabling_watchpoints
  gdb.base/watchpoint.exp: run to marker1 in test_simple_watchpoint
  gdb.base/watchpoint.exp: watchpoints found in watchpoint/breakpoint table
  gdb.c++/classes.exp: print obj_with_enum (2)
  gdb.c++/classes.exp: print obj_with_enum.priv_enum
  gdb.c++/classes.exp: ptype obj_with_enum.priv_enum
  gdb.c++/ovldbreak.exp: continue to bp overloaded : double
  gdb.c++/ovldbreak.exp: continue to bp overloaded : short
  gdb.c++/ovldbreak.exp: continue to bp overloaded : signed char
  gdb.c++/ovldbreak.exp: continue to bp overloaded : unsigned char
  gdb.c++/ovldbreak.exp: continue to bp overloaded : unsigned short
  gdb.mi/mi-until.exp: until after current function
  gdb.mi1/mi-until.exp: until after current function
  gdb.trace/packetlen.exp: run trace experiment

I haven't analyzed what happens inside gdb.  I think most of it is this:

  test script sets a breakpoint at beginning of function like 'marker1'
  gdb takes the breakpoint
  gdb prints where it is
  test script expects to see 'marker1'
  because of the bogus 'line 0' at the beginning of each function,
    gdb always prints '0x12345678 in marker1', because it's never
    at the beginning of a line

The bogus 'line 0' probably causes some other ill effects but the above
happens in gdb.base/condbreak.exp at least.  It would take more analysis
to trace them down.

[Bug debug/11500] [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[neroden at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500



------- Additional Comments From neroden at gcc dot gnu dot org  2003-07-11 23:04 -------
Could this bug also cause wildly incorrect line number reports in GDB, which I've been seeing quite recently?

[Bug debug/11500] [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[jh at suse dot cz]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500



------- Additional Comments From jh at suse dot cz  2003-07-11 23:05 -------
Subject: Re:  [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

> PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.
> 
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500
> 
> 
> pinskia at physics dot uc dot edu changed:
> 
>            What    |Removed                     |Added
> ----------------------------------------------------------------------------
>              Status|UNCONFIRMED                 |NEW
>      Ever Confirmed|                            |1
>            Priority|P2                          |P1
>    Last reconfirmed|0000-00-00 00:00:00         |2003-07-11 19:59:00
>                date|                            |
> 
> 
> ------- Additional Comments From pinskia at physics dot uc dot edu  2003-07-11 19:59 -------
> I can confirm this on the mainline (20030711) but I do not know how this confuses gdb (I 
> use gdb 2003-07-11-cvs).  This is a regression over 3.3.1 (20030707) so bumping the 
> priority.
I can not surf right now.  If you can assign the bug to me it would be
cool.  I will look into it soon (probably next week) as I am still on
the trip.

Honza
> 
> 
> 
> ------- You are receiving this mail because: -------
> You are on the CC list for the bug, or are watching someone who is.

[Bug debug/11500] [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[neroden at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500


neroden at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |jh at suse dot cz
                   |dot org                     |
             Status|NEW                         |ASSIGNED

[Bug debug/11500] [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[pinskia at physics dot uc dot edu]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500


pinskia at physics dot uc dot edu changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|jh at suse dot cz           |hubicka at gcc dot gnu dot
                   |                            |org


------- Additional Comments From pinskia at physics dot uc dot edu  2003-07-11 23:31 -------
Reassigning to Honza's gcc account.

[Bug debug/11500] [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[hubicka at gcc dot gnu dot org]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500


hubicka at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED


------- Additional Comments From hubicka at gcc dot gnu dot org  2003-07-27 20:57 -------
fixed by my earlier patch

[Bug debug/11500] [3.4 regression] gcc HEAD emits an extraneous "line 0" debug line at the start of each function

[mec at shout dot net]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11500



------- Additional Comments From mec at shout dot net  2003-07-28 16:32 -------
Confirmed fixed with gcc HEAD, 2003-07-27 20:07:42 UTC, native
i686-pc-linux-gnu, -gstabs+.  Thanks!