From mboxrd@z Thu Jan  1 00:00:00 1970
From: Claude Code Review Bot <claude-review@example.com>
To: dri-devel-reviews@example.com
Subject: Claude review: drm/vblank: Add some debugging trace events
Date: Fri, 27 Feb 2026 14:40:53 +1000
Message-ID: <review-patch5-20260224212639.390768-6-sunpeng.li@amd.com>
In-Reply-To: <20260224212639.390768-6-sunpeng.li@amd.com>
References: <20260224212639.390768-1-sunpeng.li@amd.com> <20260224212639.390768-6-sunpeng.li@amd.com>
X-Mailer: Claude Code Patch Reviewer
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
MIME-Version: 1.0

Patch Review

Straightforward tracing additions. Useful for debugging the new deferred path.

**Minor issues:**

1. Format string mismatch in `drm_vblank_get_put_template`:
```c
TP_STRUCT__entry(
    __field(int, refcount)
),
TP_printk(
    "crtc=%d, refcount=%u",    // %u for signed int
```
`refcount` is declared as `int` but printed with `%u`. Use `%d` for consistency.

2. The `drm_vblank_put` trace point reads `atomic_read(&vblank->refcount)` **before** `atomic_dec_and_test()`:
```c
trace_drm_vblank_put(pipe, atomic_read(&vblank->refcount));

/* Last user schedules interrupt disable */
if (!atomic_dec_and_test(&vblank->refcount))
    return;
```
The comment "put's refcount not sync'd using vbl_lock, use for debugging purposes only" is appropriate, but the trace shows the refcount *before* the put, not after. This is fine for debugging but could confuse readers expecting the post-decrement value.

3. The `trace_drm_vblank_get` placement is before the `atomic_add_return`, so it also shows the pre-increment refcount, which is at least consistent with the put trace.

---
Generated by Claude Code Patch Reviewer