Hi!

On 2026-07-30T17:34:12+0200, Peter0x44 <[email protected]> wrote:
> On 2026-07-30 15:15, Thomas Schwinge wrote:
>> I'm confirming that broken behavior with your
>> <https://github.com/Peter0x44/w64devkit/blob/22a780033053c041af05a6ff9c3012454ea628fe/src/crossgcc-libgomp-gfortran-path.patch>.
>> 
>> But: that 'crossgcc-libgomp-gfortran-path.patch' isn't the patch I've
>> sent, but is some mangled variant of it, and applying that one doesn't
>> change 'libgomp/configure.ac', 'libgomp/configure' in the expected way.
>> Things certainly work fine if you correctly apply the patch I've sent:
>> 
>>     $ find ~/W/ -name omp_lib\*
>>     [...]/W/lib/gcc/x86_64-w64-mingw32/16.1.0/finclude/omp_lib.mod
>>     
>> [...]/W/lib/gcc/x86_64-w64-mingw32/16.1.0/finclude/omp_lib_kinds.mod
>>     [...]/W/lib/gcc/x86_64-w64-mingw32/16.1.0/finclude/omp_lib.f90
>>     [...]/W/lib/gcc/x86_64-w64-mingw32/16.1.0/finclude/omp_lib.h
>>     $ WINEPATH=$HOME/W/bin wine64 gfortran.exe -fopenmp o.f90
>>     0138:fixme:ntdll:NtQuerySystemInformation info_class 
>> SYSTEM_PERFORMANCE_INFORMATION
>>     $ OMP_NUM_THREADS=5 wine64 ./a.exe
>>      Hello from thread           3
>>      Hello from thread           2
>>      Hello from thread           1
>>      Hello from thread           4
>>      Hello from thread           0
>
> Indeed. Somehow, when I gave the agent the link and asked it to update 
> the patch it decided it made more sense to recreate the hunks manually 
> instead of just downloading the patch exactly. I thought the AIs weren't 
> so dumb.

In my opinion, humans are still responsible (and still need to be
capable) to oversee and understand what any "AI" is doing.

And, I personally don't understand why you need "AI" to copy a patch
file?

> That is my bad and sorry for wasting your time.

Thanks.  You assumed correctly that I was angry when I discovered what
was going on here, after, indeed, having spent quite some time on this.

The good thing that I got out of this is that I learned to use your
'Dockerfile' locally with Podman, and learned to build Windows-hosted
GCC, and run under Wine that one as well as the resulting multi-threaded
GCC/OpenMP executable.

So: we're good again.  :-)

> Confirmed solves the 
> issue and approved!

Thanks.  As there were no further comments by anyone else, I've now
pushed this to trunk in commit e5efa9eff99da897f0bbd69455d8f7856985b2db
"libgomp: Revise libgomp/Fortran support conditionals".


Grüße
 Thomas

Reply via email to