Quoting Matthew Auld (2020-03-23 11:03:01) > Stolen memory is allocated at creation, returning -ENOSPC if we run out > space. Sigh. Eviction was always a planned feature for stolen, for when we exposed it to userspace. We should have just done that to flesh out the driver APIs and uAPI. > Closes: https://gitlab.freedesktop.org/drm/intel/issues/1424 > Signed-off-by: Matthew Auld <matthew.auld@xxxxxxxxx> > --- > drivers/gpu/drm/i915/selftests/intel_memory_region.c | 5 ++++- > 1 file changed, 4 insertions(+), 1 deletion(-) > > diff --git a/drivers/gpu/drm/i915/selftests/intel_memory_region.c b/drivers/gpu/drm/i915/selftests/intel_memory_region.c > index 2a1d4ba1f9f3..6e80d99048e4 100644 > --- a/drivers/gpu/drm/i915/selftests/intel_memory_region.c > +++ b/drivers/gpu/drm/i915/selftests/intel_memory_region.c > @@ -594,8 +594,11 @@ create_region_for_mapping(struct intel_memory_region *mr, u64 size, u32 type, > void *addr; > > obj = i915_gem_object_create_region(mr, size, 0); > - if (IS_ERR(obj)) > + if (IS_ERR(obj)) { > + if (PTR_ERR(obj) == -ENOSPC) /* Stolen memory */ > + return ERR_PTR(-ENODEV); Ok, we can live with this for now. Reviewed-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> -Chris _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
