mbox series

[v3,0/3] Check for valid framebuffer's format

Message ID 20230113165919.580210-1-mcanal@igalia.com (mailing list archive)
Headers show
Series Check for valid framebuffer's format | expand

Message

Maíra Canal Jan. 13, 2023, 4:59 p.m. UTC 
This series is a follow-up of the [1] in which I introduced a check for valid
formats on drm_gem_fb_create(). During the discussion, I realized that would be
a better idea to put the check inside framebuffer_check() so that it wouldn't be
needed to hit any driver-specific code path when the check fails.

Therefore, add the valid format check inside framebuffer_check() and remove the
same check from the drivers, except from i915, because this doesn't work for the
legacy tiling->modifier path. Adding the check to framebuffer_check() will
guarantee that the igt@kms_addfb_basic@addfb25-bad-modifier IGT test passes,
showing the correct behavior of the check.

This patchset was tested on amdgpu and vc4 with the IGT tests.

[1] https://lore.kernel.org/dri-devel/20230103125322.855089-1-mcanal@igalia.com/T/

---

v1 -> v2: https://lore.kernel.org/dri-devel/20230109105807.18172-1-mcanal@igalia.com/T/

- Don't remove check from i915 driver (Ville Syrjälä).
- Don't unexport drm_any_plane_has_format().

v2 -> v3: https://lore.kernel.org/dri-devel/20230113112743.188486-1-mcanal@igalia.com/T/

- Check if r->modifier[0] != 0 (Ville Syrjälä).

---

Best Regards,
- Maíra Canal

Maíra Canal (3):
  drm/framebuffer: Check for valid formats
  drm/amdgpu: Remove redundant framebuffer format check
  drm/vmwgfx: Remove redundant framebuffer format check

 Documentation/gpu/todo.rst                  |  9 ++++-----
 drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 10 ----------
 drivers/gpu/drm/drm_framebuffer.c           |  8 ++++++++
 drivers/gpu/drm/vmwgfx/vmwgfx_kms.c         | 18 ------------------
 4 files changed, 12 insertions(+), 33 deletions(-)