As pointed out by Thomas, we likely need to flush the pages here if the
GPU can read the page contents directly from main memory. Underneath we
don't know what the sg_table is pointing to, so just add a
wbinvd_on_all_cpus() here, for now.
Reported-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20211018174508.2137279-4-matthew.auld@intel.com
static int i915_gem_object_get_pages_dmabuf(struct drm_i915_gem_object *obj)
{
+ struct drm_i915_private *i915 = to_i915(obj->base.dev);
struct sg_table *pages;
unsigned int sg_page_sizes;
if (IS_ERR(pages))
return PTR_ERR(pages);
- sg_page_sizes = i915_sg_dma_sizes(pages->sgl);
+ /* XXX: consider doing a vmap flush or something */
+ if (!HAS_LLC(i915) || i915_gem_object_can_bypass_llc(obj))
+ wbinvd_on_all_cpus();
+ sg_page_sizes = i915_sg_dma_sizes(pages->sgl);
__i915_gem_object_set_pages(obj, pages, sg_page_sizes);
return 0;