Extends the user space ioctl for CL submissions so it can include a request to flush the cache once the CL execution has completed. Fixes memory write violation messages reported by the kernel in workloads involving shader memory writes (SSBOs, shader images, scratch, etc) which sometimes also lead to GPU resets during Piglit and CTS workloads. v2: if v3d_job_init() fails we need to kfree() the job instead of v3d_job_put() it (Eric Anholt). v3 (Eric Anholt): - Drop _FLAG suffix from the new flag name. - Add a new param so userspace can tell whether cache flushing is implemented in the kernel.
Showing the changes to include/uapi (1 of 3 files). Show all
Unified · Split
| @@ -48,6 +48,8 @@ extern "C" { · 47 unmodified lines | ||
| 48 | 48 | #define DRM_IOCTL_V3D_SUBMIT_TFU DRM_IOW(DRM_COMMAND_BASE + DRM_V3D_SUBMIT_TFU, struct drm_v3d_submit_tfu) |
| 49 | 49 | #define DRM_IOCTL_V3D_SUBMIT_CSD DRM_IOW(DRM_COMMAND_BASE + DRM_V3D_SUBMIT_CSD, struct drm_v3d_submit_csd) |
| 50 | 50 | |
| 51 | +#define DRM_V3D_SUBMIT_CL_FLUSH_CACHE 0x01 | |
| 52 | + | |
| 51 | 53 | /** |
| 52 | 54 | * struct drm_v3d_submit_cl - ioctl argument for submitting commands to the 3D |
| 53 | 55 | * engine. |
| @@ -61,7 +63,7 @@ extern "C" { · 7 unmodified lines | ||
| 61 | 63 | * flushed by the time the render done IRQ happens, which is the |
| 62 | 64 | * trigger for out_sync. Any dirtying of cachelines by the job (only |
| 63 | 65 | * possible using TMU writes) must be flushed by the caller using the |
| 64 | − * CL's cache flush commands. | |
| 66 | + * DRM_V3D_SUBMIT_CL_FLUSH_CACHE_FLAG flag. | |
| 65 | 67 | */ |
| 66 | 68 | struct drm_v3d_submit_cl { |
| 67 | 69 | /* Pointer to the binner command list. |
| @@ -124,8 +126,7 @@ struct drm_v3d_submit_cl { · 56 unmodified lines | ||
| 124 | 126 | /* Number of BO handles passed in (size is that times 4). */ |
| 125 | 127 | __u32 bo_handle_count; |
| 126 | 128 | |
| 127 | − /* Pad, must be zero-filled. */ | |
| 128 | − __u32 pad; | |
| 129 | + __u32 flags; | |
| 129 | 130 | }; |
| 130 | 131 | |
| 131 | 132 | /** |
| @@ -193,6 +194,7 @@ enum drm_v3d_param { · 61 unmodified lines | ||
| 193 | 194 | DRM_V3D_PARAM_V3D_CORE0_IDENT2, |
| 194 | 195 | DRM_V3D_PARAM_SUPPORTS_TFU, |
| 195 | 196 | DRM_V3D_PARAM_SUPPORTS_CSD, |
| 197 | + DRM_V3D_PARAM_SUPPORTS_CACHE_FLUSH, | |
| 196 | 198 | }; |
| 197 | 199 | |
| 198 | 200 | struct drm_v3d_get_param { |