On Wed, Jun 17, 2026 at 4:13 PM Marioukhine, Vladimir <[email protected]> wrote:
I'd suggest fixing your git setup to have your name in standard order. E.g., Vladimir Marioukhine <[email protected]> > > When MES is enabled and amdgpu_amdkfd_alloc_kernel_mem() fails during > the first queue creation for a process, pqm_create_queue() returns > early via 'return retval' without going through the err_create_queue > cleanup label. > > This means clear_bit(*qid, pqm->queue_slot_bitmap) is never called, > leaving the reserved QID bit permanently set in queue_slot_bitmap. > Over time this leaks QID slots, potentially exhausting all available > queue slots. > > Fix this by replacing 'return retval' with 'goto err_create_queue' > so that clear_bit() is always called on the error path. > > Fixes: AILIKFD-813 > Reported-by: Deucher, Alexander <[email protected]> > Signed-off-by: Marioukhine, Vladimir <[email protected]> Same here for the Reported-by: and Signed-off-by: lines. E.g., Reported-by: Alexander Deucher <[email protected]> Signed-off-by: Vladimir Marioukhine <[email protected]> You can also drop the AILIKFD-813 line. That's an internal ticket number so it's not really useful for external parties. > --- > drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c > b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c > index 1234567..abcdefg 100644 > --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c > +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c > @@ -378,7 +378,7 @@ > false); > if (retval) { > dev_err(dev->adev->dev, "failed to allocate process > context bo\n"); > - return retval; > + goto err_create_queue; I think you actually want: goto err_allocate_pqn; here. Alex > } > memset(pdd->proc_ctx_cpu_ptr, 0, AMDGPU_MES_PROC_CTX_SIZE); > } > -- > 2.34.1
