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/amdgpu: remove dead empty checks in ring_mux ib_mark_offset and end_ib
Date: Mon, 25 May 2026 20:07:45 +1000
Message-ID: <review-patch1-20260521133537.3438468-1-maoyixie.tju@gmail.com>
In-Reply-To: <20260521133537.3438468-1-maoyixie.tju@gmail.com>
References: <20260521133537.3438468-1-maoyixie.tju@gmail.com> <20260521133537.3438468-1-maoyixie.tju@gmail.com>
X-Mailer: Claude Code Patch Reviewer
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0

Patch Review

**Correctness: Good.** The two removed blocks are identical in structure:

```c
chunk =3D list_last_entry(&e->list, struct amdgpu_mux_chunk, entry);
-	if (!chunk) {
-		DRM_ERROR("cannot find chunk!\n");
-		return;
-	}
```

`list_last_entry` expands to `container_of((ptr)->prev, type, member)` (`in=
clude/linux/list.h:630`), which performs pointer arithmetic and can never y=
ield NULL. The check was always false =E2=80=94 removing it is correct.

**Precondition argument: Sound.** The commit message claims `e->list` is gu=
aranteed non-empty at both call sites. Looking at the post-patch code at `a=
mdgpu_ring_mux.c:497` and `:526`, both functions first look up the mux entr=
y via `amdgpu_ring_mux_sw_entry()` and bail on failure. The `list_last_entr=
y` call only executes after that guard, on a ring that has gone through `st=
art_ib` which links a chunk. This is a valid invariant.

**No functional change:** The removal only deletes unreachable code paths. =
The `list_last_entry()` call and all subsequent logic are unchanged.

**No issues found.** Clean, minimal patch with an accurate commit message. =
No concerns.

---
Generated by Claude Code Patch Reviewer