From: Matthew Brost <matthew.brost@intel.com>
To: <intel-xe@lists.freedesktop.org>
Cc: <thomas.hellstrom@linux.intel.com>
Subject: Re: [PATCH v3 0/2] Fix serialization on burst of unbinds - v2
Date: Sun, 26 Oct 2025 19:13:14 -0700 [thread overview]
Message-ID: <aP7VOn81hrFqnhN4@lstrano-desk.jf.intel.com> (raw)
In-Reply-To: <20251024222047.1481039-1-matthew.brost@intel.com>
On Fri, Oct 24, 2025 at 03:20:45PM -0700, Matthew Brost wrote:
> Attempting to resolve part of [1]; this solution differs than v1 [2] by
> changing last fence semantics detailed in the patches.
>
> Overview of issue in [1]:
>
> When a burst of unbind jobs is issued, a dependency chain can form
> between the TLB invalidation of a previous unbind job and the current
> one. This leads to undesirable serialization, causing current jobs to
> wait unnecessarily for prior TLB invalidations, execute on the GPU when
> not needed, and significantly slow down the unbind burst—resulting in up
> to a 4× slowdown.
>
> Matt
>
> [1] https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/6047
> [2] https://patchwork.freedesktop.org/series/156144/
>
Don't review this one, still issues here.
Matt
> Matthew Brost (2):
> drm/xe: Add last fence attachment to TLB invalidation job queues
> drm/xe: Decouple bind queue last fence from TLB invalidations
>
> drivers/gpu/drm/xe/xe_exec_queue.c | 105 ++++++++++++++++++++++-
> drivers/gpu/drm/xe/xe_exec_queue.h | 18 ++++
> drivers/gpu/drm/xe/xe_exec_queue_types.h | 5 ++
> drivers/gpu/drm/xe/xe_pt.c | 64 +++++---------
> drivers/gpu/drm/xe/xe_sync.c | 58 +++++++++++--
> drivers/gpu/drm/xe/xe_vm.c | 60 +++++++------
> 6 files changed, 234 insertions(+), 76 deletions(-)
>
> --
> 2.34.1
>
prev parent reply other threads:[~2025-10-27 2:13 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-24 22:20 [PATCH v3 0/2] Fix serialization on burst of unbinds - v2 Matthew Brost
2025-10-24 22:20 ` [PATCH v3 1/2] drm/xe: Add last fence attachment to TLB invalidation job queues Matthew Brost
2025-10-24 22:20 ` [PATCH v3 2/2] drm/xe: Decouple bind queue last fence from TLB invalidations Matthew Brost
2025-10-24 22:27 ` ✗ CI.checkpatch: warning for Fix serialization on burst of unbinds - v2 Patchwork
2025-10-24 22:28 ` ✓ CI.KUnit: success " Patchwork
2025-10-24 23:15 ` ✗ Xe.CI.BAT: failure " Patchwork
2025-10-25 11:11 ` ✗ Xe.CI.Full: " Patchwork
2025-10-27 2:13 ` Matthew Brost [this message]
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=aP7VOn81hrFqnhN4@lstrano-desk.jf.intel.com \
--to=matthew.brost@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--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