public inbox for igt-dev@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Sobin Thomas <sobin.thomas@intel.com>
To: igt-dev@lists.freedesktop.org
Cc: thomas.hellstrom@intel.com, nishit.sharma@intel.com,
	Sobin Thomas <sobin.thomas@intel.com>
Subject: [PATCH v7 i-g-t 0/3] tests/intel/xe_vm: Add support for overcommit tests
Date: Wed, 25 Mar 2026 05:55:47 +0000	[thread overview]
Message-ID: <20260325055552.2496156-1-sobin.thomas@intel.com> (raw)
In-Reply-To: <20260320100554.2427448-1-sobin.thomas@intel.com>

Current tests focus on VM creation and basic mode selection, but do not
cover overcommit scenarios.

This change adds tests to verify overcommit behavior across different VM
modes.

Non-fault mode tests:
 - vram-lr-defer: DEFER_BACKING rejects overcommit at bind time
 - vram-lr-external-nodefer: Long-running mode with an external BO and
                             no deferred backing
 - vram-no-lr: Non-long-running mode

Fault mode tests:
 - vram-lr-fault: Fault handling allows graceful overcommit via page faults
 - vram-lr-fault-no-overcommit: Verifies NO_VM_OVERCOMMIT blocks same-VM
   BO eviction during VM_BIND, while still allowing eviction during
   page-fault OOM handling

These tests validate that VMs respond correctly to memory pressure based
on their configuration—by rejecting at bind time, failing during execution,
or handling overcommit gracefully via page faults.

Currently, overcommit behavior is tested for VRAM only.

--------------------------------------
Link (historical context):
https://patchwork.freedesktop.org/series/161557/


Sobin Thomas (3):
  drm-uapi/xe: sync with kernel header
  lib/xe: Add failable variant of xe_vm_bind_lr_sync
  tests/intel/xe_vm: Add support for overcommit tests

 include/drm-uapi/xe_drm.h |  22 +-
 lib/xe/xe_ioctl.c         |  36 ++++
 lib/xe/xe_ioctl.h         |   2 +
 tests/intel/xe_vm.c       | 417 +++++++++++++++++++++++++++++++++++++-
 4 files changed, 465 insertions(+), 12 deletions(-)

-- 
2.43.0


  parent reply	other threads:[~2026-03-25  5:56 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-20 10:05 [PATCH v6 i-g-t 0/2] tests/intel/xe_vm: Add support for overcommit tests Sobin Thomas
2026-03-20 10:05 ` [PATCH v6 i-g-t 1/2] drm-uapi/xe: sync with kernel header Sobin Thomas
2026-03-20 10:05 ` [PATCH i-g-t 2/2] tests/intel/xe_vm: Add support for overcommit tests Sobin Thomas
2026-03-20 12:58 ` ✓ Xe.CI.BAT: success for " Patchwork
2026-03-20 13:07 ` ✓ i915.CI.BAT: " Patchwork
2026-03-21  3:42 ` ✓ i915.CI.Full: " Patchwork
2026-03-21 11:14 ` ✗ Xe.CI.FULL: failure " Patchwork
2026-03-25  5:55 ` Sobin Thomas [this message]
2026-03-25  5:55   ` [PATCH v7 i-g-t 1/3] drm-uapi/xe: sync with kernel header Sobin Thomas
2026-03-25  5:55   ` [PATCH v7 i-g-t 2/3] lib/xe: Add failable variant of xe_vm_bind_lr_sync Sobin Thomas
2026-03-25  5:55   ` [PATCH v7 i-g-t 3/3] tests/intel/xe_vm: Add support for overcommit tests Sobin Thomas
  -- strict thread matches above, loose matches on Subject: below --
2026-03-25  6:00 [PATCH v7 i-g-t 0/3] " Sobin Thomas
2026-03-25  6:03 Sobin Thomas

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=20260325055552.2496156-1-sobin.thomas@intel.com \
    --to=sobin.thomas@intel.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=nishit.sharma@intel.com \
    --cc=thomas.hellstrom@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