[0/2] drm/i915/gt: Fix recent kCFI violations

Message

Nathan Chancellor May 30, 2023, 6:24 p.m. UTC

Hi all,

This series fixes a few clang kernel Control Flow Integrity (kCFI)
violations that appear after commit 9275277d5324 ("drm/i915: use
pat_index instead of cache_level"). They were found between run time
testing on real hardware and compile time testing with
-Wincompatible-function-pointer-types-strict (which is not yet enabled
for the kernel but I build with it locally to catch new instances).

If there are any problems or questions, please let me know.

---
Nathan Chancellor (2):
      drm/i915/gt: Fix second parameter type of pre-gen8 pte_encode callbacks
      drm/i915/gt: Fix parameter in gmch_ggtt_insert_{entries,page}()

 drivers/gpu/drm/i915/gt/intel_ggtt.c      | 26 +++++++++++++-------------
 drivers/gpu/drm/i915/gt/intel_ggtt_gmch.c |  8 ++++----
 2 files changed, 17 insertions(+), 17 deletions(-)
---
base-commit: 08264f85c5c05ecc38d409c84d48cfb00ccd3bc4
change-id: 20230530-i915-gt-cache_level-wincompatible-function-pointer-types-strict-32a5c65249a5

Best regards,

Comments

Andi Shyti June 2, 2023, 1:03 a.m. UTC | #1

Hi Nathan,

On Tue, May 30, 2023 at 11:24:37AM -0700, Nathan Chancellor wrote:
> Hi all,
> 
> This series fixes a few clang kernel Control Flow Integrity (kCFI)
> violations that appear after commit 9275277d5324 ("drm/i915: use
> pat_index instead of cache_level"). They were found between run time
> testing on real hardware and compile time testing with
> -Wincompatible-function-pointer-types-strict (which is not yet enabled
> for the kernel but I build with it locally to catch new instances).
> 
> If there are any problems or questions, please let me know.
> 
> ---
> Nathan Chancellor (2):
>       drm/i915/gt: Fix second parameter type of pre-gen8 pte_encode callbacks
>       drm/i915/gt: Fix parameter in gmch_ggtt_insert_{entries,page}()

pushed in drm-intel-gt-next.

Thank you,
Andi