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: replace PASID IDR with XArray
Date: Tue, 31 Mar 2026 16:57:38 +1000
Message-ID: <review-patch1-20260330191120.105065-1-mikhail.v.gavrilov@gmail.com>
In-Reply-To: <20260330191120.105065-1-mikhail.v.gavrilov@gmail.com>
References: <20260330191120.105065-1-mikhail.v.gavrilov@gmail.com> <20260330191120.105065-1-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

**Commit message**: Excellent. The bug description is thorough, with clear =
lockdep-style annotation and a concrete reproduction scenario (RX 7900 XTX,=
 Vulkan game under Proton). The changelog across v1-v5 is well documented.

**Bug =E2=80=94 xa_alloc_cyclic() is not IRQ-safe** (line 70-72 of patched =
file):

```c
r =3D xa_alloc_cyclic(&amdgpu_pasid_xa, &pasid, xa_mk_value(0),
		    XA_LIMIT(1, (1U << bits) - 1),
		    &amdgpu_pasid_xa_next, GFP_KERNEL);
```

`xa_alloc_cyclic()` uses plain `xa_lock()` (see `include/linux/xarray.h:982=
`):
```c
xa_lock(xa);
err =3D __xa_alloc_cyclic(xa, id, entry, limit, next, gfp);
xa_unlock(xa);
```

Since `amdgpu_pasid_free()` can be called from hardirq context (the entire =
reason for this patch), an IRQ arriving while this lock is held will deadlo=
ck. This must be `xa_alloc_cyclic_irq()` which disables IRQs around the loc=
k (`include/linux/xarray.h:1054`).

**Return value change** (lines 74-79): The original `idr_alloc_cyclic()` re=
turned the allocated ID on success (positive) or negative errno. `xa_alloc_=
cyclic()` returns 0 on success and writes the ID to `*pasid`. The patch cor=
rectly handles this =E2=80=94 checking `r >=3D 0` and returning `pasid` rat=
her than `r`. Note that `xa_alloc_cyclic()` actually swallows the wrap-arou=
nd indicator (returns 0 even on wrap), but `__xa_alloc_cyclic()` returns 1 =
on wrap. Since this code doesn't care about wrap notification, this is fine.

**Return value for -EBUSY** (line 79): The original IDR code returned `-ENO=
SPC` on exhaustion (per IDR semantics), while `xa_alloc_cyclic()` returns `=
-EBUSY`. The docstring at line 59 still says "Returns %-ENOSPC if no PASID =
was available" =E2=80=94 this should be updated to `-EBUSY` to match realit=
y. Minor documentation nit.

**amdgpu_pasid_free() locking** (lines 92-94): Correct. Using `xa_lock_irqs=
ave()`/`__xa_erase()`/`xa_unlock_irqrestore()` is the right approach for ha=
rdirq context.

**xa_mk_value(0) as stored entry** (line 70): Storing `xa_mk_value(0)` work=
s =E2=80=94 XArray treats NULL as "empty", so an explicit non-NULL value is=
 needed to mark the slot as occupied. `xa_mk_value(0)` is the lightest-weig=
ht way to do this. Fine.

**Cleanup** (line 261): `xa_destroy()` without locking is correct =E2=80=94=
 this is called at module teardown when no concurrent access is possible.

**Recommendation**: NAK as-is due to the `xa_alloc_cyclic()` deadlock. The =
one-line fix is to change it to `xa_alloc_cyclic_irq()`. With that change, =
the patch would be correct and ready.

---
Generated by Claude Code Patch Reviewer