From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: pvr: acquire vm_ctx->lock before mapping memory to GPU VM Date: Thu, 23 Apr 2026 08:14:32 +1000 Message-ID: In-Reply-To: <20260421175228.1928742-1-zhengxingda@iscas.ac.cn> References: <20260421175228.1928742-1-zhengxingda@iscas.ac.cn> X-Mailer: Claude Code Patch Reviewer Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Overall Series Review Subject: pvr: acquire vm_ctx->lock before mapping memory to GPU VM Author: Icenowy Zheng Patches: 2 Reviewed: 2026-04-23T08:14:32.334692 --- This is a single-patch series that adds `vm_ctx->lock` around the `drm_gpuv= m_exec_lock()`/`drm_gpuvm_exec_unlock()` + `pvr_vm_bind_op_exec()` sequence= in `pvr_vm_map()`. The stated goal =E2=80=94 protecting `drm_gpuva_find*()= ` from concurrent map operations =E2=80=94 is reasonable and consistent wit= h the locking pattern already used in `pvr_vm_unmap()` (line 865) and `pvr_= vm_unmap_all()` (line 891). **However, the patch has a critical bug: it double-unlocks `vm_ctx->lock` o= n the success path.** --- Generated by Claude Code Patch Reviewer