From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=100.103.45.18 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1794679740; b=Vizvx2qZo7aAEOof4a215uqpsKEuc7bGLS90tu8fFnbTYJrd/1TNSXtwvR/++QqO6IZD GcXiv8pBLGSzPXqOmGBic+H3OJlrDcI5mJHuaR6oyBMZW8/xBX8c1ZjNjEh20cv9BQslE CuGRw+pX1gH/StGFOqdCFXWN+0JKXjwgKzDXXvaOnyECgRLp2IvhLNfIB4MnyJ56c90s5 aRLtD+Xr8mN/FzKgdCupXO0FyC2bPQbOK359fPdYmalD8Z8xoVzYnc4E05U0SS0/zAZtU 0jZtUXqRAdw3Ekq/auIu+j3mCfbdYAOXHmrMx4hS4JW4WmvfZmJjxaxtArhh5tBBPvQ== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1794679740; h=Received:DKIM-Signature:From:To:Cc:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:Content-Transfer-Encoding; bh=mDVfGMNKuKmMgPDUO0i3tvTJlQmoOmD1MigQIiDSGH0=; b=tr8gabXT7w62j5h7he38mgQyCZmGn2gSr4+l6xOdWR2xPw7KKJxX6XsJ/kI79rD4aRbL AM3Vfu3EQtdH3GRQ1IwGHa2dYoSxrgBBKMkxbd05pVzj/mrgp/XNOtFZQ8JYETGguiES9 CVBTLg0GAifoI5zMyLCgWLKSN0iZVQSlaH9Dct9VFuZvrEuXW4ekOYEIhWaBJJqKvK30C SBMwSs18Y+koYFuMPUOR2IT5SO8q3v94ads/oVjNJbLcyk0NdCx1eEK7OHPMW6mxQy7aQ lJ54oMQ3vwA/SlgMCHaky0Ypz6YV3QTTbh7ZDCsPob1Duu/ZHMdjvHsoww3xxVO1kAQ== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=VOg1rcOH; arc=none smtp.remote-ip=100.103.45.18 Received: from smtp.kernel.org (quasi.space.kernel.org [100.103.45.18]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id CEA51606A327 for ; Sat, 14 Nov 2026 18:09:00 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=VOg1rcOH Received: by smtp.kernel.org (Postfix) with ESMTPSA id C14241F000C4; Sat, 14 Nov 2026 18:08:59 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1794679740; bh=mDVfGMNKuKmMgPDUO0i3tvTJlQmoOmD1MigQIiDSGH0=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=VOg1rcOHyW5zeZOtEIUGCXSTr8DcRBu/qydONyAEDKkJVy1hsMzK+VBrGU0JWVFz0 JEMjHHmGdnnuiONsHsmbBK7XAlPpDHjsw1GzlG7/2JPeEVNggUvfV4KC1pzvlW+t+Y 3CAB0QkGFXbNUupCRI3JCoOFZlrugI9BMCJS6IY5jwfpqwGUNjZCu5YfRi7usrEWp1 tyohK8eFSTW5MdmoxMoIaiZ4ZX92IAwHxNZ0D0khesLsWPCGrleN+7bt9HKwc8akni FFAI8205nWw9eQH1wB+LBeVniHZ4NbIl84uFaFLNZ8R3oRGipE5A2K7ibXvD0+lTJY 3rUwzTtGBFmxg== From: Sasha Levin To: stable@vger.kernel.org Cc: Jesse Zhang , Felix Kuehling , Alex Deucher , Sasha Levin Subject: [PATCH 6.1.y] drm/amdkfd: use the queue's process info for the WPTR BO eviction fence Date: Sat, 14 Nov 2026 13:08:58 -0500 Message-ID: <20261114180840.359293-1-sashal@kernel.org> X-Mailer: git-send-email 2.53.0 In-Reply-To: <20261109201925-failed-6.1-da2897ee8b81@kernel.org> References: <20261109201925-failed-6.1-da2897ee8b81@kernel.org> X-Mailing-List: stable@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list MIME-Version: 1.0 Content-Transfer-Encoding: 8bit From: Jesse Zhang [ Upstream commit da2897ee8b811945fd40ecc4aaf03291c14ce61c ] amdgpu_amdkfd_map_gtt_bo_to_gart() removes the eviction fence of bo->vm_bo->vm->process_info from the WPTR BO of a new queue. bo->vm_bo is the VM which opened the BO last, not necessarily the KFD process' VM, and a render node VM has no process info. On navi31, an unprivileged process which exports its WPTR BO with AMDKFD_IOC_EXPORT_DMABUF and imports it on a second render node fd oopses in AMDKFD_IOC_CREATE_QUEUE: BUG: kernel NULL pointer dereference, address: 0000000000000060 RIP: 0010:amdgpu_amdkfd_map_gtt_bo_to_gart+0x86/0x1a0 [amdgpu] init_user_queue.part.0+0x180/0x250 [amdgpu] pqm_create_queue+0x32d/0x730 [amdgpu] kfd_ioctl_create_queue+0x583/0x9f0 [amdgpu] Pass the process info of the queue instead. With the fix, the queue is created and destroyed normally. Fixes: 7f347e3f8217 ("drm/amdkfd: Fix NULL pointer error for GC 11.0.1 on mGPU") Reviewed-by: Felix Kuehling Signed-off-by: Jesse Zhang Signed-off-by: Alex Deucher (cherry picked from commit 6b4157498a176bd1c6083f6bb5aaf8a257a2027e) Cc: stable@vger.kernel.org [ sashal: adapted the process_info argument to the existing adev-based mapping API in kfd_ioctl_create_queue() ] Signed-off-by: Sasha Levin --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 3 ++- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 8 +++++--- drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 3 ++- 3 files changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index ba065f7887aad..ed0f5e76324e0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -294,7 +294,8 @@ int amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel(struct kgd_mem *mem, void **kptr, uint64_t *size); void amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel(struct kgd_mem *mem); -int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_device *adev, struct amdgpu_bo *bo); +int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_device *adev, struct amdgpu_bo *bo, + struct amdkfd_process_info *process_info); int amdgpu_amdkfd_gpuvm_restore_process_bos(void *process_info, struct dma_fence **ef); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index eb4d652afbb60..49b9e0033be9e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -2163,11 +2163,13 @@ int amdgpu_amdkfd_gpuvm_sync_memory( * amdgpu_amdkfd_map_gtt_bo_to_gart - Map BO to GART and increment reference count * @adev: Device to which allocated BO belongs * @bo: Buffer object to be mapped + * @process_info: KFD process info of the queue the BO is for * * Before return, bo reference count is incremented. To release the reference and unpin/ * unmap the BO, call amdgpu_amdkfd_free_gtt_mem. */ -int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_device *adev, struct amdgpu_bo *bo) +int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_device *adev, struct amdgpu_bo *bo, + struct amdkfd_process_info *process_info) { int ret; @@ -2189,8 +2191,8 @@ int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_device *adev, struct amdgpu_b goto err_map_bo_gart_failed; } - amdgpu_amdkfd_remove_eviction_fence( - bo, bo->vm_bo->vm->process_info->eviction_fence); + /* bo->vm_bo can be the VM of any file which opened the BO */ + amdgpu_amdkfd_remove_eviction_fence(bo, process_info->eviction_fence); amdgpu_bo_unreserve(bo); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index a08c84556ace4..4212d8c77de3b 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -368,7 +368,8 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, goto err_wptr_map_gart; } - err = amdgpu_amdkfd_map_gtt_bo_to_gart(dev->adev, wptr_bo); + err = amdgpu_amdkfd_map_gtt_bo_to_gart(dev->adev, wptr_bo, + p->kgd_process_info); if (err) { pr_err("Failed to map wptr bo to GART\n"); goto err_wptr_map_gart; -- 2.53.0