Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Matthew Brost <matthew.brost@intel.com>
To: intel-xe@lists.freedesktop.org, dri-devel@lists.freedesktop.org
Cc: mika.kuoppala@intel.com, matthew.auld@intel.com,
	thomas.hellstrom@linux.intel.com
Subject: [PATCH v5 0/9] Fix non-contiguous VRAM BO access in Xe
Date: Mon, 21 Oct 2024 14:18:26 -0700	[thread overview]
Message-ID: <20241021211835.1675640-1-matthew.brost@intel.com> (raw)

Mapping a non-contiguous VRAM BO doesn't work, start to fix this.

v2:
 - Include missing local change
v3:
 - Use GPU for non-visible access
 - Take PM ref in xe_ttm_access_memory
 - Disable VMAP for non-contiguous BOs
v4:
 - Fix less than page sized non-visible access
 - Add PM ref to xe_ttm_access_memory
 - Add PM ref to delayed snapshot capture worker
 - Fix a couple of warnings in xe_res_cursor
 - Test all possible code paths
v5:
 - Fix 32 bit build (CI)

Matthew Brost (9):
  drm/xe: Add xe_bo_vm_access
  drm/ttm: Add ttm_bo_access
  drm/xe: Add xe_ttm_access_memory
  drm/xe: Take PM ref in delayed snapshot capture worker
  drm/xe: Update xe_ttm_access_memory to use GPU for non-visible access
  drm/xe/display: Update intel_bo_read_from_page to use ttm_bo_access
  drm/xe: Use ttm_bo_access in xe_vm_snapshot_capture_delayed
  drm/xe: Set XE_BO_FLAG_PINNED in migrate selftest BOs
  drm/xe: Only allow contiguous BOs to use xe_bo_vmap

 drivers/gpu/drm/ttm/ttm_bo_util.c     |  86 +++++++
 drivers/gpu/drm/ttm/ttm_bo_vm.c       |  65 +-----
 drivers/gpu/drm/xe/display/intel_bo.c |  25 +-
 drivers/gpu/drm/xe/tests/xe_migrate.c |  13 +-
 drivers/gpu/drm/xe/xe_bo.c            | 106 +++++++--
 drivers/gpu/drm/xe/xe_devcoredump.c   |   6 +
 drivers/gpu/drm/xe/xe_migrate.c       | 325 ++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_migrate.h       |   4 +
 drivers/gpu/drm/xe/xe_vm.c            |  17 +-
 include/drm/ttm/ttm_bo.h              |   2 +
 10 files changed, 533 insertions(+), 116 deletions(-)

-- 
2.34.1


             reply	other threads:[~2024-10-21 21:18 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-21 21:18 Matthew Brost [this message]
2024-10-21 21:18 ` [PATCH v5 1/9] drm/xe: Add xe_bo_vm_access Matthew Brost
2024-10-24  8:13   ` Matthew Auld
2024-10-21 21:18 ` [PATCH v5 2/9] drm/ttm: Add ttm_bo_access Matthew Brost
2024-10-22  9:18   ` Mika Kuoppala
2024-10-23 22:34     ` Matthew Brost
2024-10-24  7:42       ` Mika Kuoppala
2024-10-24  9:20   ` Matthew Auld
2024-10-21 21:18 ` [PATCH v5 3/9] drm/xe: Add xe_ttm_access_memory Matthew Brost
2024-10-24  8:52   ` Matthew Auld
2024-10-21 21:18 ` [PATCH v5 4/9] drm/xe: Take PM ref in delayed snapshot capture worker Matthew Brost
2024-10-22 14:35   ` Matthew Auld
2024-10-21 21:18 ` [PATCH v5 5/9] drm/xe: Update xe_ttm_access_memory to use GPU for non-visible access Matthew Brost
2024-10-21 21:45   ` Matthew Brost
2024-10-21 21:18 ` [PATCH v5 6/9] drm/xe/display: Update intel_bo_read_from_page to use ttm_bo_access Matthew Brost
2024-10-22  8:43   ` Matthew Auld
2024-10-22 18:02     ` Matthew Brost
2024-10-21 21:18 ` [PATCH v5 7/9] drm/xe: Use ttm_bo_access in xe_vm_snapshot_capture_delayed Matthew Brost
2024-10-22  8:54   ` Matthew Auld
2024-10-21 21:18 ` [PATCH v5 8/9] drm/xe: Set XE_BO_FLAG_PINNED in migrate selftest BOs Matthew Brost
2024-10-21 21:18 ` [PATCH v5 9/9] drm/xe: Only allow contiguous BOs to use xe_bo_vmap Matthew Brost
2024-10-21 21:39 ` ✓ CI.Patch_applied: success for Fix non-contiguous VRAM BO access in Xe (rev5) Patchwork
2024-10-21 21:39 ` ✗ CI.checkpatch: warning " Patchwork
2024-10-21 21:40 ` ✓ CI.KUnit: success " Patchwork
2024-10-21 21:52 ` ✓ CI.Build: " Patchwork
2024-10-21 21:54 ` ✓ CI.Hooks: " Patchwork
2024-10-21 21:56 ` ✗ CI.checksparse: warning " Patchwork
2024-10-21 22:21 ` ✓ CI.BAT: success " Patchwork
2024-10-22  4:51 ` ✗ CI.FULL: failure " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20241021211835.1675640-1-matthew.brost@intel.com \
    --to=matthew.brost@intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=matthew.auld@intel.com \
    --cc=mika.kuoppala@intel.com \
    --cc=thomas.hellstrom@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox