On Mon, 14 Apr 2025 at 14:51, Thomas Zimmermann <tzimmerm...@suse.de> wrote:
>
> Instead of testing import_attach for imported GEM buffers, invoke
> drm_gem_is_imported() to do the test. The helper tests the dma_buf
> itself while import_attach is just an artifact of the import. Prepares
> to make import_attach optional.
>
> Signed-off-by: Thomas Zimmermann <tzimmerm...@suse.de>
> Cc: Maxime Ripard <mrip...@kernel.org>
> Cc: Dave Stevenson <dave.steven...@raspberrypi.com>
> Cc: "Maíra Canal" <mca...@igalia.com>
> Cc: Raspberry Pi Kernel Maintenance <kernel-l...@raspberrypi.com>
Acked-by: Dave Stevenson <dave.steven...@raspberrypi.com>
> ---
> drivers/gpu/drm/vc4/vc4_bo.c | 2 +-
> drivers/gpu/drm/vc4/vc4_gem.c | 2 +-
> 2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c
> index 7125773889f1..b503fd0488b0 100644
> --- a/drivers/gpu/drm/vc4/vc4_bo.c
> +++ b/drivers/gpu/drm/vc4/vc4_bo.c
> @@ -556,7 +556,7 @@ static void vc4_free_object(struct drm_gem_object *gem_bo)
> mutex_lock(&vc4->bo_lock);
> /* If the object references someone else's memory, we can't cache it.
> */
> - if (gem_bo->import_attach) {
> + if (drm_gem_is_imported(gem_bo)) {
> vc4_bo_destroy(bo);
> goto out;
> }
> diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c
> index 8125f87edc60..3e79ad69250a 100644
> --- a/drivers/gpu/drm/vc4/vc4_gem.c
> +++ b/drivers/gpu/drm/vc4/vc4_gem.c
> @@ -1249,7 +1249,7 @@ int vc4_gem_madvise_ioctl(struct drm_device *dev, void
> *data,
> /* Not sure it's safe to purge imported BOs. Let's just assume it's
> * not until proven otherwise.
> */
> - if (gem_obj->import_attach) {
> + if (drm_gem_is_imported(gem_obj)) {
> DRM_DEBUG("madvise not supported on imported BOs\n");
> ret = -EINVAL;
> goto out_put_gem;
> --
> 2.49.0
>
