[igt-dev] [PATCH i-g-t v2 2/2] tests/i915_pm_rpm: use device coherent mapping instead of mmap_gtt

[Dixit, Ashutosh]

On Sun, 09 Feb 2020 21:47:55 -0800, Ramalingam C wrote:
>
> On 2020-02-07 at 14:44:36 +0000, Chris Wilson wrote:
> > Quoting Ramalingam C (2020-02-07 13:45:27)
> > > @@ -2016,16 +2016,20 @@ igt_main_args("", long_options, help_str, opt_handler, NULL)
> > >         /* GEM */
> > >         igt_subtest("gem-mmap-cpu")
> > >                 gem_mmap_subtest(false);
> > > -       igt_subtest("gem-mmap-gtt")
> > > +       igt_subtest("gem-mmap-gtt") {
> > > +               gem_require_mappable_ggtt(drm_fd);
> > >                 gem_mmap_subtest(true);
> >
> > Would not the lmem bar also be susceptible to power management
> > artefacts?
> Chris,
>
> I think "gem-mmap-cpu" would do the same test with lmem, but that
> can be done when we have the uAPI for allocating from lmem, through
> another subtest say "gem-mmap-cpu-lmem"
>
> But at present "gem-mmap-gtt" will fail at platforms where mmap_gtt is
> not present. So to avoid that, could you please approve this change?

My reading was Chris is referring to using for_each_mmap_offset_type(), see
973bbbff.