[igt-dev] [PATCH] tests/i915/i915_pipe_stress: Add plane pipes stress test

[Petri Latvala]

On Mon, Sep 12, 2022 at 08:41:31AM +0300, Stanislav Lisovskiy wrote:
> This test attempts to utilize all connected
> outputs at the same time, using maximum possible
> resolution and amount of planes, to check whether
> we are hiting any kind of bandwidth, watermark or
> other limitations.
> 
> v2: Use igt_skip_on to skip unsupported formats in
>     fixture properly.
> 
> v3: - Changed to SPDX license
>     - Removed execution engine usage, which seems to
>       be not needed yet here
>     - Fixed some obvious things found, like missing
>       dot and new line after declarations
>     - Added igt_skip_on to igt_fixture to properly
>       skip formats/modifiers, not supported on
>       current platform.
> 
> v4: - Added mutex, as gpu_fill seems to cause some
>       crashes if run from multiple threads.
>     - Some additional info addedy
> 
> v5: - Removed igt_require_f as it is being called from
>       non-main thread(Petri Latvala)
> 
> v6: - Added igt_assert for fill_fn
>     - Don't do gpu fill for gens < 8, as there are seem
>       to be some issues, when it is done from multiple
>       threads.
> 
> v7: - Apparently need to exclude ApolloLake as well.
>       Anyways we are not much interested in stress-testing
>       those.
> 
> Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy at intel.com>


Recap of what we discussed about this offline is:
Acked-by: Petri Latvala <petri.latvala at intel.com>

But someone with more kms experience could be useful for having a
sanity checking look over this.