| Message ID | 20230726155356.1652979-1-jonathan.cavitt@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [1/3] drm/i915/gt: Simplify shmem_create_from_object map_type selection | expand |
On 26/07/2023 16:53, Jonathan Cavitt wrote: > The object pin created for shmem_create_from_object is just a > single use mapping with the sole purpose of reading the contents > of the whole object in bulk. And the whole source object is also > even a throw-away. Ergo, the additional logic required by > i915_coherent_map_type can be safely dropped and simplified. > > Signed-off-by: Jonathan Cavitt <jonathan.cavitt@intel.com> > Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> > --- > drivers/gpu/drm/i915/gt/shmem_utils.c | 3 +-- > 1 file changed, 1 insertion(+), 2 deletions(-) > > diff --git a/drivers/gpu/drm/i915/gt/shmem_utils.c b/drivers/gpu/drm/i915/gt/shmem_utils.c > index 449c9ed44382..bccc3a1200bc 100644 > --- a/drivers/gpu/drm/i915/gt/shmem_utils.c > +++ b/drivers/gpu/drm/i915/gt/shmem_utils.c > @@ -33,7 +33,6 @@ struct file *shmem_create_from_data(const char *name, void *data, size_t len) > > struct file *shmem_create_from_object(struct drm_i915_gem_object *obj) > { > - struct drm_i915_private *i915 = to_i915(obj->base.dev); > enum i915_map_type map_type; > struct file *file; > void *ptr; > @@ -44,7 +43,7 @@ struct file *shmem_create_from_object(struct drm_i915_gem_object *obj) > return file; > } > > - map_type = i915_coherent_map_type(i915, obj, true); > + map_type = i915_gem_object_is_lmem(obj) ? I915_MAP_WC : I915_MAP_WB; > ptr = i915_gem_object_pin_map_unlocked(obj, map_type); > if (IS_ERR(ptr)) > return ERR_CAST(ptr); Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Regards, Tvrtko
Hi Jonathan, On Wed, Jul 26, 2023 at 08:53:54AM -0700, Jonathan Cavitt wrote: > The object pin created for shmem_create_from_object is just a > single use mapping with the sole purpose of reading the contents > of the whole object in bulk. And the whole source object is also > even a throw-away. Ergo, the additional logic required by > i915_coherent_map_type can be safely dropped and simplified. > > Signed-off-by: Jonathan Cavitt <jonathan.cavitt@intel.com> > Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> > --- > drivers/gpu/drm/i915/gt/shmem_utils.c | 3 +-- > 1 file changed, 1 insertion(+), 2 deletions(-) > > diff --git a/drivers/gpu/drm/i915/gt/shmem_utils.c b/drivers/gpu/drm/i915/gt/shmem_utils.c > index 449c9ed44382..bccc3a1200bc 100644 > --- a/drivers/gpu/drm/i915/gt/shmem_utils.c > +++ b/drivers/gpu/drm/i915/gt/shmem_utils.c > @@ -33,7 +33,6 @@ struct file *shmem_create_from_data(const char *name, void *data, size_t len) > > struct file *shmem_create_from_object(struct drm_i915_gem_object *obj) > { > - struct drm_i915_private *i915 = to_i915(obj->base.dev); > enum i915_map_type map_type; > struct file *file; > void *ptr; > @@ -44,7 +43,7 @@ struct file *shmem_create_from_object(struct drm_i915_gem_object *obj) > return file; > } > > - map_type = i915_coherent_map_type(i915, obj, true); > + map_type = i915_gem_object_is_lmem(obj) ? I915_MAP_WC : I915_MAP_WB; > ptr = i915_gem_object_pin_map_unlocked(obj, map_type); > if (IS_ERR(ptr)) > return ERR_CAST(ptr); Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com> Thanks, Andi [ Nitpick section ] Next time, please: 1. Number your patches. 2. Add a changelog (either in the cover letter (which is best for multi-patch series) or after the '---' section (for single patches)). 3. Don't forget the previous tags, e.g. Acked-by Fei.
diff --git a/drivers/gpu/drm/i915/gt/shmem_utils.c b/drivers/gpu/drm/i915/gt/shmem_utils.c index 449c9ed44382..bccc3a1200bc 100644 --- a/drivers/gpu/drm/i915/gt/shmem_utils.c +++ b/drivers/gpu/drm/i915/gt/shmem_utils.c @@ -33,7 +33,6 @@ struct file *shmem_create_from_data(const char *name, void *data, size_t len) struct file *shmem_create_from_object(struct drm_i915_gem_object *obj) { - struct drm_i915_private *i915 = to_i915(obj->base.dev); enum i915_map_type map_type; struct file *file; void *ptr; @@ -44,7 +43,7 @@ struct file *shmem_create_from_object(struct drm_i915_gem_object *obj) return file; } - map_type = i915_coherent_map_type(i915, obj, true); + map_type = i915_gem_object_is_lmem(obj) ? I915_MAP_WC : I915_MAP_WB; ptr = i915_gem_object_pin_map_unlocked(obj, map_type); if (IS_ERR(ptr)) return ERR_CAST(ptr);
The object pin created for shmem_create_from_object is just a single use mapping with the sole purpose of reading the contents of the whole object in bulk. And the whole source object is also even a throw-away. Ergo, the additional logic required by i915_coherent_map_type can be safely dropped and simplified. Signed-off-by: Jonathan Cavitt <jonathan.cavitt@intel.com> Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> --- drivers/gpu/drm/i915/gt/shmem_utils.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)