[Intel-gfx] [PATCH v7 0/2] drm/i915: Hugepage manager and test for MTL

[Cavitt, Jonathan]

-----Original Message-----
From: Hajda, Andrzej <andrzej.hajda at intel.com> 
Sent: Wednesday, April 26, 2023 8:14 AM
To: intel-gfx at lists.freedesktop.org; linux-kernel at vger.kernel.org; Cavitt, Jonathan <jonathan.cavitt at intel.com>; Hajda, Andrzej <andrzej.hajda at intel.com>; Auld, Matthew <matthew.auld at intel.com>
Subject: [PATCH v7 0/2] drm/i915: Hugepage manager and test for MTL
>
>This patchset patches sent by Jonathan and Andi, with 
>addressed CI failures:
>1. Fixed checking alignment of 64K pages on both Pre-Gen12 and Gen12.
>2. Fixed start alignment of 2M pages.
>
>Regards
>Andrzej
>
>Jonathan Cavitt (2):
>  drm/i915: Migrate platform-dependent mock hugepage selftests to live
>  drm/i915: Use correct huge page manager for MTL
>
>.../gpu/drm/i915/gem/selftests/huge_pages.c   | 88 +++++++++++++++----
> drivers/gpu/drm/i915/gt/gen8_ppgtt.c          |  3 +-
> 2 files changed, 71 insertions(+), 20 deletions(-)
>
>Cc: intel-gfx at lists.freedesktop.org
>Cc: linux-kernel at vger.kernel.org
>Cc: Jonathan Cavitt <jonathan.cavitt at intel.com>
>Cc: Andrzej Hajda <andrzej.hajda at intel.com>
>Cc: Matthew Auld <matthew.auld at intel.com>
>-- 
>2.39.2
>
>---
>Jonathan Cavitt (2):
>      drm/i915: Migrate platform-dependent mock hugepage selftests to live
>      drm/i915: Use correct huge page manager for MTL
>
> drivers/gpu/drm/i915/gem/selftests/huge_pages.c | 105 ++++++++++++++++++------
> drivers/gpu/drm/i915/gt/gen8_ppgtt.c            |   3 +-
> 2 files changed, 82 insertions(+), 26 deletions(-)
>---
>base-commit: 4d0066a1c0763d50b6fb017e27d12b081ce21b57
>change-id: 20230425-hugepage-migrate-1869aaf31a6d
>
>Best regards,
>-- 
>Andrzej Hajda <andrzej.hajda at intel.com>


Just reviewed the changes proper.  It's been a while, so I don't remember everything
about the prior version, but I think I recognized what was changed:

- I wasn't aware a 21 bit alignment was required for 2M page sizes.  I'm glad you caught that.
- The extra debugging/error information will be helpful in the case of a failure.
- Grabbing the per-context VM instead of the ppgtt vm sounds good to me.

Everything here looks amenable.
Acked-by: Jonathan Cavitt <jonathan.cavitt at intel.com>
Though, given I'm also one of the signed-off-bys, I don't know if me acking it is exactly above-board.
-Jonathan Cavitt


>