Hi Andi, [...] > > diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c > > b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c > > index 78734c404a6d..6d87356e928b 100644 > > --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c > > +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c > > @@ -498,13 +498,15 @@ static int igt_smoke_tiling(void *arg) > > tile.stride = 0; > > tile.swizzle = I915_BIT_6_SWIZZLE_NONE; > > break; > > - > > case I915_TILING_X: > > tile.swizzle = to_gt(i915)->ggtt->bit_6_swizzle_x; > > break; > > case I915_TILING_Y: > > tile.swizzle = to_gt(i915)->ggtt->bit_6_swizzle_y; > > break; > > + default: > > + err = -EINVAL; > > + break; > > You are breaking from the switch/case and ending up where? Should > this be continue? > > If you really want to make static analysers happy (which we don't > care much) you could do something like: > > case I915_TILING_NONE: > default: > > But the C specification is not obliging anyone to use the default > case, so that we can happily ignore the static analyser issue if > there is no need for a default. Hmm, if we indeed do not care about static analyzers, then I agree that this is pushing it a bit. I may skip this patch in the next iteration of this series then.
In any case thanks for review! -- Best Regards, Krzysztof
