bug-binutils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Bug gold/15341] debug_msg.sh make check fails


From: ccoutant at google dot com
Subject: [Bug gold/15341] debug_msg.sh make check fails
Date: Mon, 15 Apr 2013 22:46:59 +0000

http://sourceware.org/bugzilla/show_bug.cgi?id=15341

Cary Coutant <ccoutant at google dot com> changed:

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

--- Comment #11 from Cary Coutant <ccoutant at google dot com> 2013-04-15 
22:46:59 UTC ---
Disassembly from odr_violation1.o:

0000000000000000 <_ZN10OdrDerivedD0Ev>:
~OdrDerived():
/usr/local/src/xxx-devel/binutils-2.23.2-gold/binutils-build/gold/testsuite/../../../gold/testsuite/odr_header1.h:4
   0:   55                      push   %rbp
   1:   48 89 e5                mov    %rsp,%rbp
   4:   48 83 ec 10             sub    $0x10,%rsp
   8:   48 89 7d f8             mov    %rdi,-0x8(%rbp)
   c:   ba 00 00 00 00          mov    $0x0,%edx
  11:   48 8b 45 f8             mov    -0x8(%rbp),%rax
  15:   48 89 10                mov    %rdx,(%rax)

Disassembly from odr_violation2.o:

0000000000000000 <_ZN10OdrDerivedD0Ev>:
~OdrBase():
/usr/local/src/xxx-devel/binutils-2.23.2-gold/binutils-build/gold/testsuite/../../../gold/testsuite/odr_header2.h:2
   0:   48 c7 07 00 00 00 00    movq   $0x0,(%rdi)

Gold is comparing "odr_header1.h:4" and "odr_header2.h:2" and concluding that
these are two different functions. We've got a bunch of heuristics to try to
cope with inconsistent line number information, but none of them work here.

This was built, from what I saw in your config.log with GCC 4.1, so I'm
inclined to close this as a compiler problem that has been fixed in more recent
versions.

It would be nice if we could mark this test as "unsupported" when we're using a
compiler that's known not to generate good enough line numbers, but I'm not
really sure what kind of check to run. I guess I could just run objdump -dl on
these two .o files and scan for expected output before running the test, but
there are still a lot of variations that should be accepted.

I'm open to suggestions.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]