Claude review: dma-fence: Clarify external lock use case in dma_fence_init() docs

[Claude Code Review Bot <claude-review@example.com>]

Overall Series Review

Subject: dma-fence: Clarify external lock use case in dma_fence_init() docs
Author: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Patches: 2
Reviewed: 2026-06-04T14:48:33.627035

---

This is a single documentation-only patch (v3) that improves the kerneldoc comments for `dma_fence_init()` and `dma_fence_init64()` in `drivers/dma-buf/dma-fence.c`. The motivation is clear and correct: the existing wording ("prevent multiple fences from signaling out of order") is misleading because a shared spinlock cannot actually enforce signaling order — that's the driver's responsibility. The patch rewrites the comment to explain the *actual* legacy rationale (serializing signaling when no out-of-order signaling was possible via `dma_fence_ops.signaled`) and adds a strong "MUST NOT" prohibition for new users.

The patch is well-scoped, addresses only the misleading documentation, and the de-duplication approach for `dma_fence_init64()` (pointing to `dma_fence_init()` docs rather than repeating the explanation) is sensible.

**Verdict: Looks good.** No functional changes, documentation improvement is accurate and clearly worded. Minor observations below.

---
Generated by Claude Code Patch Reviewer