Re: [PATCH i-g-t v5 5/6] tests/intel/xe_pxp: Termination tests

["Teres Alexis, Alan Previn" <alan.previn.teres.alexis@intel.com>]

I see you have added the documentation for existing __xe_bb_exec and xe_exec_sync,
as well as the new xe_exec_sync_failable. Also, u have included <fcntl.h> to address
the undeclared ‘O_RDONLY’. These addresses the earlier comments from Kamil.

I also notice you have added the drmModeResPtr usage to disable all of the
display connectors before running the runtime-pm subtest via setup_rpm()
(and re-enabling after via restore_rpm) since an active display pipeline
could prevent the runtime-pm from kicking in. Good catch on this one.

It would be better to get Kamil's take on whether this version of patch #5 meets
the requests he had about fixtures in v3 especially which the updated changes...
also I am not 100% clear on the usage of fixture WRT zero-impact-of-subtest-enumeration
vs cleanup-after-subtest-assert.

That said, I'll provide an RB here (since Kamil didnt raise it again in V4) and we can
check with him offline:

Reviewed-by: Alan Previn <alan.previn.teres.alexis@intel.com>


On Wed, 2025-03-05 at 15:39 -0800, Ceraolo Spurio, Daniele wrote:
> 
alan:snip