From: Matthew Brost <matthew.brost@intel.com>
To: intel-xe@lists.freedesktop.org
Cc: thomas.hellstrom@linux.intel.com
Subject: [PATCH v5 0/6] Fix serialization on burst of unbinds - v2
Date: Wed, 29 Oct 2025 13:57:13 -0700 [thread overview]
Message-ID: <20251029205719.2746501-1-matthew.brost@intel.com> (raw)
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/
Matthew Brost (6):
drm/xe: Enforce correct user fence signaling order using drm_syncobjs
drm/xe: Attach last fence to TLB invalidation job queues
drm/xe: Decouple bind queue last fence from TLB invalidations
drm/xe: Skip TLB invalidation waits in page fault binds
drm/xe: Disallow input fences on zero batch execs and zero binds
drm/xe: Remove last fence dependency check from binds
drivers/gpu/drm/xe/xe_exec.c | 3 +-
drivers/gpu/drm/xe/xe_exec_queue.c | 119 ++++++++++++++++++++---
drivers/gpu/drm/xe/xe_exec_queue.h | 23 ++++-
drivers/gpu/drm/xe/xe_exec_queue_types.h | 12 +++
drivers/gpu/drm/xe/xe_migrate.c | 14 +++
drivers/gpu/drm/xe/xe_migrate.h | 8 ++
drivers/gpu/drm/xe/xe_oa.c | 45 ++++++---
drivers/gpu/drm/xe/xe_oa_types.h | 8 ++
drivers/gpu/drm/xe/xe_pt.c | 80 +++++----------
drivers/gpu/drm/xe/xe_sync.c | 91 ++++++++++-------
drivers/gpu/drm/xe/xe_sync.h | 3 +
drivers/gpu/drm/xe/xe_sync_types.h | 3 +
drivers/gpu/drm/xe/xe_tlb_inval_job.c | 31 ++++--
drivers/gpu/drm/xe/xe_tlb_inval_job.h | 5 +-
drivers/gpu/drm/xe/xe_vm.c | 97 ++++++++++--------
drivers/gpu/drm/xe/xe_vm_types.h | 6 +-
16 files changed, 376 insertions(+), 172 deletions(-)
--
2.34.1
next reply other threads:[~2025-10-29 20:57 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-29 20:57 Matthew Brost [this message]
2025-10-29 20:57 ` [PATCH v5 1/6] drm/xe: Enforce correct user fence signaling order using drm_syncobjs Matthew Brost
2025-10-30 7:58 ` Thomas Hellström
2025-10-30 12:54 ` Matthew Brost
2025-10-29 20:57 ` [PATCH v5 2/6] drm/xe: Attach last fence to TLB invalidation job queues Matthew Brost
2025-10-30 8:24 ` Thomas Hellström
2025-10-29 20:57 ` [PATCH v5 3/6] drm/xe: Decouple bind queue last fence from TLB invalidations Matthew Brost
2025-10-30 9:52 ` Thomas Hellström
2025-10-29 20:57 ` [PATCH v5 4/6] drm/xe: Skip TLB invalidation waits in page fault binds Matthew Brost
2025-11-03 15:19 ` Thomas Hellström
2025-10-29 20:57 ` [PATCH v5 5/6] drm/xe: Disallow input fences on zero batch execs and zero binds Matthew Brost
2025-11-03 15:21 ` Thomas Hellström
2025-11-03 15:22 ` Thomas Hellström
2025-10-29 20:57 ` [PATCH v5 6/6] drm/xe: Remove last fence dependency check from binds Matthew Brost
2025-10-30 8:43 ` Thomas Hellström
2025-11-03 15:24 ` Thomas Hellström
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=20251029205719.2746501-1-matthew.brost@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