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: fix sleeping allocation under spinlock in PASID IDR
Date: Tue, 31 Mar 2026 17:50:57 +1000
Message-ID: <review-patch1-20260328213900.19255-1-mikhail.v.gavrilov@gmail.com>
In-Reply-To: <20260328213900.19255-1-mikhail.v.gavrilov@gmail.com>
References: <20260328213900.19255-1-mikhail.v.gavrilov@gmail.com> <20260328213900.19255-1-mikhail.v.gavrilov@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

**Analysis of the bug:** Confirmed by reading the current tree at `amdgpu_i=
ds.c:70-73` =E2=80=94 the code currently calls `idr_alloc_cyclic()` with `G=
FP_KERNEL` while holding `amdgpu_pasid_idr_lock` (a spinlock). This is inde=
ed a sleeping-in-atomic-context bug.

**Analysis of the fix:**

```c
+	idr_preload(GFP_KERNEL);
 	spin_lock(&amdgpu_pasid_idr_lock);
 	pasid =3D idr_alloc_cyclic(&amdgpu_pasid_idr, NULL, 1,
-				 1U << bits, GFP_KERNEL);
+				 1U << bits, GFP_NOWAIT);
 	spin_unlock(&amdgpu_pasid_idr_lock);
+	idr_preload_end();
```

This is textbook correct:
1. `idr_preload(GFP_KERNEL)` pre-allocates radix tree nodes (can sleep, cal=
led outside spinlock) and disables preemption.
2. `GFP_NOWAIT` inside the spinlock ensures the allocator uses the preloade=
d pool and never sleeps.
3. `idr_preload_end()` re-enables preemption after the spinlock is released.

**One minor ordering note:** `idr_preload_end()` is called after `spin_unlo=
ck()`, which is correct =E2=80=94 `idr_preload()` disables preemption, so y=
ou must hold the preload context across the allocation. The unlock-before-p=
reload_end ordering is the standard pattern (see e.g., `drivers/infiniband/=
`, `net/core/`).

**Fixes tag and Cc: stable** are both appropriate since this is a straightf=
orward bugfix for a commit that introduced the regression.

**No issues found.**=20

Reviewed-by recommendation: This patch is correct and follows established k=
ernel patterns.

---
Generated by Claude Code Patch Reviewer