drm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING_INFO
Subsystems (3)
PATCH 1/1 drm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING_INFO
The AMDGPU_GEM_OP_GET_MAPPING_INFO branch of amdgpu_gem_op_ioctl() holds three cleanup-tracked resources before calling kvcalloc():
the drm_gem_object reference from drm_gem_object_lookup(), the drm_exec lock on the looked-up GEM via drm_exec_lock_obj(), and the drm_exec lock on the per-process VM root page directory via amdgpu_vm_lock_pd(). All three are released by the out_exec label that every other error path in this function jumps to. The kvcalloc() failure path returns -ENOMEM directly, skipping out_exec and leaking all three.
The leaked per-process VM root PD dma_resv lock is the load-bearing leak: any subsequent operation on the same VM (further GEM ops, command-submission, eviction, TTM shrinker callbacks) blocks on the held lock. DRM_IOCTL_AMDGPU_GEM_OP is DRM_AUTH | DRM_RENDER_ALLOW, so this is an unprivileged-local denial of service against the caller's GPU context, reachable by any process with /dev/dri/renderD* access.
Route the failure through out_exec so drm_exec_fini() and drm_gem_object_put() run.
Reproduced on stock 7.0.0-10, Ryzen 7 5700U / Radeon Vega (Lucienne): the failing ioctl returns -ENOMEM and a second GET_MAPPING_INFO on the same fd then blocks in drm_exec_lock_obj() on the leaked dma_resv. SIGKILL on the caller does not reap the task; the fd-release path during process exit goes through amdgpu_gem_object_close() -> drm_exec_prepare_obj() on the same lock, leaving the task in D state until the box is rebooted. The patched kernel was not rebuilt and re-tested on this hardware; the fix is mechanical. Tested on a single Lucienne / Vega box only.
Ziyi Guo posted an independent INT_MAX-bound check for args->num_entries in the same branch [1]; the two patches are complementary and can land in either order.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +4 -2
@@ -1091,8 +1091,10 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,* be retried. */ vm_entries =kvcalloc (args->num_entries,sizeof (*vm_entries),GFP_KERNEL ); -if (!vm_entries) -return -ENOMEM ; +if (!vm_entries) { + r = -ENOMEM ; + goto out_exec; + }amdgpu_vm_bo_va_for_each_valid_mapping (bo_va, mapping) {if (num_mappings < args->num_entries) {
Signature · 1 line
2.53.0
Alex Deucher
AppliedApplied. Thanks!
Alex
Quoted 66 lines · Michael Bommarito
The
AMDGPU_GEM_OP_GET_MAPPING_INFObranch ofamdgpu_gem_op_ioctl()holds three cleanup-tracked resources before callingkvcalloc():
thedrm_gem_objectreference fromdrm_gem_object_lookup(), thedrm_execlock on the looked-up GEM viadrm_exec_lock_obj(), and thedrm_execlock on the per-process VM root page directory viaamdgpu_vm_lock_pd(). All three are released by theout_execlabel that every other error path in this function jumps to. Thekvcalloc()failure path returns-ENOMEMdirectly, skippingout_execand leaking all three.The leaked per-process VM root PD
dma_resvlock is the load-bearing leak: any subsequent operation on the same VM (further GEM ops, command-submission, eviction, TTM shrinker callbacks) blocks on the held lock.DRM_IOCTL_AMDGPU_GEM_OPisDRM_AUTH|DRM_RENDER_ALLOW, so this is an unprivileged-local denial of service against the caller's GPU context, reachable by any process with/dev/dri/renderD*access.Route the failure through
out_execsodrm_exec_fini()anddrm_gem_object_put()run.Reproduced on stock 7.0.0-10, Ryzen 7 5700U / Radeon Vega (Lucienne): the failing ioctl returns
-ENOMEMand a secondGET_MAPPING_INFOon the same fd then blocks indrm_exec_lock_obj()on the leakeddma_resv. SIGKILL on the caller does not reap the task; the fd-release path during process exit goes throughamdgpu_gem_object_close()->drm_exec_prepare_obj()on the same lock, leaving the task in D state until the box is rebooted. The patched kernel was not rebuilt and re-tested on this hardware; the fix is mechanical. Tested on a single Lucienne / Vega box only.Ziyi Guo posted an independent INT_MAX-bound check for
args->num_entriesin the same branch [1]; the two patches are complementary and can land in either order.Fixes 4d82724f7f2b ("drm/amdgpu: Add mapping info option for GEM_OP ioctl")Signed-off-by Michael Bommarito <[email protected]>Assisted-by Claude:claude-opus-4-7--- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +4
-3@@ -1091,8 +1091,10 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,* be retried. */ vm_entries =kvcalloc (args->num_entries,sizeof (*vm_entries),GFP_KERNEL ); -if (!vm_entries) -return -ENOMEM ; +if (!vm_entries) { + r = -ENOMEM ; + goto out_exec; + }amdgpu_vm_bo_va_for_each_valid_mapping (bo_va, mapping) {if (num_mappings < args->num_entries) { --2.53.0