Re: [PATCH v9 2/6] tests/functional/test_aarch64_rme: update image

[Pierrick Bouvier <pierrick.bouvier@linaro.org>]

On 8/15/25 9:29 AM, Alex Bennée wrote:
> Peter Maydell <peter.maydell@linaro.org> writes:
> 
>> On Sun, 27 Jul 2025 at 08:42, Richard Henderson
>> <richard.henderson@linaro.org> wrote:
>>>
>>> From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
>>>
>>> TF-A needs to be patched to enable support for FEAT_TCR2 and
>>> FEAT_SCTLR2. This new image contains updated firmware.
>>>
>>> Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
>>> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
>>> Message-ID: <20250719035838.2284029-3-pierrick.bouvier@linaro.org>
>>
>> I see this test failing in local runs of 'make check-functional'
>> sometimes, complaining that it can't create the scratch/out/EFI
>> directory because it already exists:
>>
>> stderr:
>> Traceback (most recent call last):
>>    File "/mnt/nvmedisk/linaro/qemu-from-laptop/qemu/tests/functional/test_aarch64_rme_sbsaref.py",
>> line 49, in test_aarch64_rme_sbsaref
>>      os.mkdir(efi)
> 
>    os.makedirs(efi, exist_ok=True)
> 
> would solve that.
>

Yes, probably safer to add that if we can't guarantee correctly that 
scratch folder gets deleted everytime. I wrote the test assuming this 
behaviour.

>> FileExistsError: [Errno 17] File exists:
>> '/mnt/nvmedisk/linaro/qemu-from-laptop/qemu/build/arm-clang/tests/functional/aarch64/test_aarch64_rme_sbsaref.Aarch64RMESbsaRefMachine.test_aarch64_rme_sbsaref/scratch/out/EFI'
>>
>> More information on
>> test_aarch64_rme_sbsaref.Aarch64RMESbsaRefMachine.test_aarch64_rme_sbsaref
>> could be found here:
>>   /mnt/nvmedisk/linaro/qemu-from-laptop/qemu/build/arm-clang/tests/functional/aarch64/test_aarch64_rme_sbsaref.Aarch64RMESbsaRefMachine.test_aarch64_rme_sbsaref/base.log
>>   /mnt/nvmedisk/linaro/qemu-from-laptop/qemu/build/arm-clang/tests/functional/aarch64/test_aarch64_rme_sbsaref.Aarch64RMESbsaRefMachine.test_aarch64_rme_sbsaref/console.log
>>
>> (test program exited with status code 1)
>>
>>
>> Is this something where we need to blow away any old output
>> directory for robustness ?
> 
> Generally scratch is cleaned out once a test finishes (unless you
> preserve it via QEMU_TEST_KEEP_SCRATCH=1). But I guess there may be
> cases when it isn't cleaned out properly - does Ctrl-c get caught?
> 
>>
>> thanks
>> -- PMM
>