From: Claude Code Review Bot <claude-review@example.com>
To: dri-devel-reviews@example.com
Subject: Claude review: Two-pass MMU interval notifiers
Date: Tue, 03 Mar 2026 13:05:43 +1000 [thread overview]
Message-ID: <review-overall-20260302163248.105454-1-thomas.hellstrom@linux.intel.com> (raw)
In-Reply-To: <20260302163248.105454-1-thomas.hellstrom@linux.intel.com>
Overall Series Review
Subject: Two-pass MMU interval notifiers
Author: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Patches: 15
Reviewed: 2026-03-03T13:05:43.927349
---
This series adds two-pass MMU interval notifier support to enable overlapping GPU operations (TLB flushes, context preemption) across multiple GPUs in multi-GPU scenarios. The motivation is sound: instead of serializing notify-then-wait per-GPU, the first pass kicks off operations on all GPUs, and the second pass waits for them all. The series is well-structured, with patch 1 providing the core mm infrastructure, patch 2 converting xe userptr to the two-pass model, patch 3 refactoring TLB invalidation into submit/wait halves, and patch 4 wiring up deferred TLB waits through the two-pass mechanism.
Overall the design is reasonable and the code is generally clean. There are a few issues to address, ranging from a real bug (static local variable in patch 2) to design concerns about the three-pass usage in patch 4 abusing the two-pass infrastructure.
---
---
Generated by Claude Code Patch Reviewer
next prev parent reply other threads:[~2026-03-03 3:05 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-02 16:32 [PATCH v2 0/4] Two-pass MMU interval notifiers Thomas Hellström
2026-03-02 16:32 ` [PATCH v2 1/4] mm/mmu_notifier: Allow two-pass struct mmu_interval_notifiers Thomas Hellström
2026-03-02 19:48 ` Matthew Brost
2026-03-02 21:12 ` Thomas Hellström
2026-03-03 3:05 ` Claude review: " Claude Code Review Bot
2026-03-02 16:32 ` [PATCH v2 2/4] drm/xe/userptr: Convert invalidation to two-pass MMU notifier Thomas Hellström
2026-03-02 18:57 ` Matthew Brost
2026-03-02 21:22 ` Thomas Hellström
2026-03-02 21:30 ` Matthew Brost
2026-03-03 3:05 ` Claude review: " Claude Code Review Bot
2026-03-02 16:32 ` [PATCH v2 3/4] drm/xe: Split TLB invalidation into submit and wait steps Thomas Hellström
2026-03-02 19:06 ` Matthew Brost
2026-03-02 21:29 ` Thomas Hellström
2026-03-02 21:31 ` Matthew Brost
2026-03-03 3:05 ` Claude review: " Claude Code Review Bot
2026-03-02 16:32 ` [PATCH v2 4/4] drm/xe/userptr: Defer Waiting for TLB invalidation to the second pass if possible Thomas Hellström
2026-03-02 19:14 ` Matthew Brost
2026-03-02 21:33 ` Thomas Hellström
2026-03-03 3:05 ` Claude review: " Claude Code Review Bot
2026-03-03 3:05 ` Claude Code Review Bot [this message]
-- strict thread matches above, loose matches on Subject: below --
2026-03-03 13:34 [PATCH v3 0/4] Two-pass MMU interval notifiers Thomas Hellström
2026-03-03 21:09 ` Claude review: " Claude Code Review Bot
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=review-overall-20260302163248.105454-1-thomas.hellstrom@linux.intel.com \
--to=claude-review@example.com \
--cc=dri-devel-reviews@example.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