On Mon, Jul 14, 2025 at 4:08 PM Maíra Canal <mca...@igalia.com> wrote:
>
> The global fault counter is no longer used since commit 12578c075f89
> ("drm/msm/gpu: Skip retired submits in recover worker"). Additionally,
> with commit eab7766c79fd ("drm/msm: Remove vram carveout support"), all
> supported devices now have a defined virtual memory address space, which
> eliminates the need for a global fault counter.
Hmm, I think commit 12578c075f89 ("drm/msm/gpu: Skip retired submits
in recover worker") was wrong, now that you point it out. We could
still have situations where we get a fault after the faulting process
has ended. In that case get_pid_task() would return NULL.
BR,
-R
>
> Hence, remove the global faults counter. While here, s/unusuable/unusable.
>
> Signed-off-by: Maíra Canal <mca...@igalia.com>
> ---
> drivers/gpu/drm/msm/adreno/adreno_gpu.c | 5 +----
> drivers/gpu/drm/msm/msm_gpu.c | 23 +++++++++++------------
> drivers/gpu/drm/msm/msm_gpu.h | 6 ------
> 3 files changed, 12 insertions(+), 22 deletions(-)
>
> diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
> b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
> index f1230465bf0d..5bb00b9c998a 100644
> --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
> +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
> @@ -416,10 +416,7 @@ int adreno_get_param(struct msm_gpu *gpu, struct
> msm_context *ctx,
> *value = 0;
> return 0;
> case MSM_PARAM_FAULTS:
> - if (vm)
> - *value = gpu->global_faults + to_msm_vm(vm)->faults;
> - else
> - *value = gpu->global_faults;
> + *value = vm ? to_msm_vm(vm)->faults : 0;
> return 0;
> case MSM_PARAM_SUSPENDS:
> *value = gpu->suspend_count;
> diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
> index c317b25a8162..3d5a76d2901c 100644
> --- a/drivers/gpu/drm/msm/msm_gpu.c
> +++ b/drivers/gpu/drm/msm/msm_gpu.c
> @@ -463,6 +463,7 @@ static void recover_worker(struct kthread_work *work)
> struct drm_device *dev = gpu->dev;
> struct msm_drm_private *priv = dev->dev_private;
> struct msm_gem_submit *submit;
> + struct msm_gem_vm *vm;
> struct msm_ringbuffer *cur_ring = gpu->funcs->active_ring(gpu);
> char *comm = NULL, *cmd = NULL;
> int i;
> @@ -482,20 +483,18 @@ static void recover_worker(struct kthread_work *work)
>
> /* Increment the fault counts */
> submit->queue->faults++;
> - if (submit->vm) {
> - struct msm_gem_vm *vm = to_msm_vm(submit->vm);
>
> - vm->faults++;
> + vm = to_msm_vm(submit->vm);
> + vm->faults++;
>
> - /*
> - * If userspace has opted-in to VM_BIND (and therefore
> userspace
> - * management of the VM), faults mark the VM as unusuable.
> This
> - * matches vulkan expectations (vulkan is the main target for
> - * VM_BIND)
> - */
> - if (!vm->managed)
> - msm_gem_vm_unusable(submit->vm);
> - }
> + /*
> + * If userspace has opted-in to VM_BIND (and therefore userspace
> + * management of the VM), faults mark the VM as unusable. This
> + * matches vulkan expectations (vulkan is the main target for
> + * VM_BIND)
> + */
> + if (!vm->managed)
> + msm_gem_vm_unusable(submit->vm);
>
> get_comm_cmdline(submit, &comm, &cmd);
>
> diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h
> index b2a96544f92a..88d8ddef616b 100644
> --- a/drivers/gpu/drm/msm/msm_gpu.h
> +++ b/drivers/gpu/drm/msm/msm_gpu.h
> @@ -226,12 +226,6 @@ struct msm_gpu {
> /* does gpu need hw_init? */
> bool needs_hw_init;
>
> - /**
> - * global_faults: number of GPU hangs not attributed to a particular
> - * address space
> - */
> - int global_faults;
> -
> void __iomem *mmio;
> int irq;
>
> --
> 2.50.0
>
