Re: system() call causes gcc-x64 profiling to crash

1 msg	problem in compiling bootstrap gcc 4.2.2

system() call causes gcc-x64 profiling to crash
\ Vincent Diepeveen (20 Dec 2007)
. \ Andrew Haley (20 Dec 2007)
. . \ Vincent Diepeveen (21 Dec 2007)

1 msg	file not found when compiling gcc under AIX 5.3
7 msg	More questions on sysroots
1 msg	Bizarre behavior with various __attribute__s
1 msg	-pg profiling instrumentation question
3 msg	Problem building gcc 4.2.2 on SunOS 5.9
8 msg	Problem building gcc-4.2.2 64-bit on hp-ux 11.11
12 msg	Problem building gcc-4.2.2 on 64-bit ubuntu linux
1 msg	Vote for Hungary!
1 msg	Fwd: gcc build fails on Mac Intel
1 msg	failure to build gcc
7 msg	Binary built on different platforms with exact ...
1 msg	gFortran advice
5 msg	(bad) disassemblie
4 msg	Two questions of gcc testing
4 msg	posix threads error
4 msg	Documentation about debugging formats
2 msg	major linker errors \| gcc (gfortran)
2 msg	gcc build fails with undefined __gthread_active_p

Subject:	Re: system() call causes gcc-x64 profiling to crash
Group:	Gcc-help
From:	Vincent Diepeveen
Date:	21 Dec 2007

Thanks Andrew,

We figured out the problem in the meantime: setting a signal at start
of program causes that profiler gives error on system() function.

On Dec 20, 2007, at 11:28 AM, Andrew Haley wrote:

> Vincent Diepeveen writes:
>> Good morning,
>>
>> When we try to profile (-pg and with optimizations turned on)
>> at SUSE x64 gcc 4.2.2, then the software gets hung at the
>> point of calling the function system() in c++ code.
>>
>> We are using both intel as well as amd hardware it gets hung at
>> them all.
>>
>> What is the best way for us to get around this and proceed?
>
> Produce a simple test case and then we'll be able to talk about it.
> As it stands we don't know if this is a kernel problem, libc, or gcc.
> Maybe you're just using it wrong.
>
> Andrew.
>
> --
> Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor,
> Berkshire, SL4 1TE, UK
> Registered in England and Wales No. 3798903
>