Re: [PATCH v2 12/16] testsuite, fortran: fix info-types for intel compilers

[Tom de Vries <tdevries@suse.de>]

On 6/1/22 11:21, Kempke, Nils-Christian wrote:
> 
> 
>> -----Original Message-----
>> From: Tom de Vries <tdevries@suse.de>
>> Sent: Wednesday, June 1, 2022 7:25 AM
>> To: Andrew Burgess <aburgess@redhat.com>; Kempke, Nils-Christian <nils-
>> christian.kempke@intel.com>; gdb-patches@sourceware.org
>> Cc: JiniSusan.George@amd.com
>> Subject: Re: [PATCH v2 12/16] testsuite, fortran: fix info-types for intel
>> compilers
>>
>> On 5/31/22 18:06, Andrew Burgess via Gdb-patches wrote:
>>> Nils-Christian Kempke <nils-christian.kempke@intel.com> writes:
>>>
>>>> This info-types.exp test case had a few issues that this patch fixes.
>>>>
>>
>> I'm seeing this regression:
>> ...
>> Running
>> /home/vries/gdb_versions/devel/src/gdb/testsuite/gdb.fortran/info-
>> types.exp
>> ...
>> gdb compile failed,
>> /home/vries/gdb_versions/devel/src/gdb/testsuite/gdb.fortran/info-
>> types.f90:47:13:
>>
>>      type (m1t1) :: var_b
>>                1
>> Error: Derived type 'm1t1' at (1) is being used before it is defined
>> /home/vries/gdb_versions/devel/src/gdb/testsuite/gdb.fortran/info-
>> types.f90:50:8:
>>
>>      var_b%b = 2
>>           1
>> Error: Symbol 'var_b' at (1) has no IMPLICIT type
>> UNTESTED: gdb.fortran/info-types.exp: failed to prepare
>> ...
>>
>> Likewise for a few other test-cases that use the same source.
>>
>> Thanks,
>> - Tom
> 
> Hi Tom,
> 
> Sorry for this, too. I did not change the executable - nor the actual
> compilation of it I think but this must be the source of the regressions.
> 
> I made one of the checks in the info-types.exp file optional, but that does
> not really explain the problems you are seeing..
> 
> I cannot reproduce this currently. I assume this is from a native run? Does
> this also happen in the tests when ran individually? Or is this failure related
> to the regressions seen because of the compiler_info caching?
> 

Hi Nils,

I've not been able to reproduce this anymore.

I'm gonna assume that this was due to an incorrect temporary fix I was 
trying out.

I currently just have a patch in my master-local branch that disables 
compiler_info caching, and that seems to work fine.

Sorry for the noise.

Thanks,
- Tom