By selecting the cache level (essentially whether or not the CPU snoops
any updates to the bo, and on more recent machines whether it resides
inside the CPU's last-level-cache) a userspace driver is able to then
manage all of its memory within buffer objects, if it so desires. This
enables the userspace driver to accelerate uploads and more importantly
downloads from the GPU and to able to mix CPU and GPU rendering/activity
efficiently.
Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
---
drivers/gpu/drm/i915/i915_dma.c | 2 ++
drivers/gpu/drm/i915/i915_drv.h | 8 +++--
drivers/gpu/drm/i915/i915_gem.c | 62 +++++++++++++++++++++++++++++++++++++++
include/drm/i915_drm.h | 15 ++++++++++
4 files changed, 85 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index f64ef4b..ed462fe 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1829,6 +1829,8 @@ struct drm_ioctl_desc i915_ioctls[] = {
DRM_IOCTL_DEF_DRV(I915_GEM_PIN, i915_gem_pin_ioctl, DRM_AUTH|DRM_ROOT_ONLY|DRM_UNLOCKED),
DRM_IOCTL_DEF_DRV(I915_GEM_UNPIN, i915_gem_unpin_ioctl, DRM_AUTH|DRM_ROOT_ONLY|DRM_UNLOCKED),
DRM_IOCTL_DEF_DRV(I915_GEM_BUSY, i915_gem_busy_ioctl, DRM_AUTH|DRM_UNLOCKED),
+ DRM_IOCTL_DEF_DRV(I915_GEM_SET_CACHEING, i915_gem_set_cacheing_ioctl, DRM_UNLOCKED),
+ DRM_IOCTL_DEF_DRV(I915_GEM_GET_CACHEING, i915_gem_get_cacheing_ioctl, DRM_UNLOCKED),
DRM_IOCTL_DEF_DRV(I915_GEM_THROTTLE, i915_gem_throttle_ioctl, DRM_AUTH|DRM_UNLOCKED),
DRM_IOCTL_DEF_DRV(I915_GEM_ENTERVT, i915_gem_entervt_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY|DRM_UNLOCKED),
DRM_IOCTL_DEF_DRV(I915_GEM_LEAVEVT, i915_gem_leavevt_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY|DRM_UNLOCKED),
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4fb358e..038c29c 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -854,9 +854,9 @@ enum hdmi_force_audio {
};
enum i915_cache_level {
- I915_CACHE_NONE,
+ I915_CACHE_NONE = 0,
I915_CACHE_LLC,
- I915_CACHE_LLC_MLC, /* gen6+ */
+ I915_CACHE_LLC_MLC, /* gen6+, in docs at least! */
};
struct drm_i915_gem_object {
@@ -1249,6 +1249,10 @@ int i915_gem_unpin_ioctl(struct drm_device *dev, void *data,
struct drm_file *file_priv);
int i915_gem_busy_ioctl(struct drm_device *dev, void *data,
struct drm_file *file_priv);
+int i915_gem_get_cacheing_ioctl(struct drm_device *dev, void *data,
+ struct drm_file *file);
+int i915_gem_set_cacheing_ioctl(struct drm_device *dev, void *data,
+ struct drm_file *file);
int i915_gem_throttle_ioctl(struct drm_device *dev, void *data,
struct drm_file *file_priv);
int i915_gem_madvise_ioctl(struct drm_device *dev, void *data,
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index da89f13..b8f5c4f 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3102,6 +3102,68 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
return 0;
}
+int i915_gem_get_cacheing_ioctl(struct drm_device *dev, void *data,
+ struct drm_file *file)
+{
+ struct drm_i915_gem_cacheing *args = data;
+ struct drm_i915_gem_object *obj;
+ int ret;
+
+ ret = i915_mutex_lock_interruptible(dev);
+ if (ret)
+ return ret;
+
+ obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+ if (&obj->base == NULL) {
+ ret = -ENOENT;
+ goto unlock;
+ }
+
+ args->cacheing = obj->cache_level != I915_CACHE_NONE;
+
+ drm_gem_object_unreference(&obj->base);
+unlock:
+ mutex_unlock(&dev->struct_mutex);
+ return ret;
+}
+
+int i915_gem_set_cacheing_ioctl(struct drm_device *dev, void *data,
+ struct drm_file *file)
+{
+ struct drm_i915_gem_cacheing *args = data;
+ struct drm_i915_gem_object *obj;
+ enum i915_cache_level level;
+ int ret;
+
+ ret = i915_mutex_lock_interruptible(dev);
+ if (ret)
+ return ret;
+
+ switch (args->cacheing) {
+ case I915_CACHEING_NONE:
+ level = I915_CACHE_NONE;
+ break;
+ case I915_CACHEING_CACHED:
+ level = I915_CACHE_LLC;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
+ if (&obj->base == NULL) {
+ ret = -ENOENT;
+ goto unlock;
+ }
+
+ ret = i915_gem_object_set_cache_level(obj, level);
+
+ drm_gem_object_unreference(&obj->base);
+unlock:
+ mutex_unlock(&dev->struct_mutex);
+ return ret;
+}
+
/*
* Prepare buffer for display plane (scanout, cursors, etc).
* Can be called from an uninterruptible phase (modesetting) and allows
diff --git a/include/drm/i915_drm.h b/include/drm/i915_drm.h
index 564005e..87e46b4 100644
--- a/include/drm/i915_drm.h
+++ b/include/drm/i915_drm.h
@@ -203,6 +203,8 @@ typedef struct _drm_i915_sarea {
#define DRM_I915_GEM_WAIT 0x2c
#define DRM_I915_GEM_CONTEXT_CREATE 0x2d
#define DRM_I915_GEM_CONTEXT_DESTROY 0x2e
+#define DRM_I915_GEM_SET_CACHEING 0x2f
+#define DRM_I915_GEM_GET_CACHEING 0x30
#define DRM_IOCTL_I915_INIT DRM_IOW( DRM_COMMAND_BASE + DRM_I915_INIT, drm_i915_init_t)
#define DRM_IOCTL_I915_FLUSH DRM_IO ( DRM_COMMAND_BASE + DRM_I915_FLUSH)
@@ -227,6 +229,8 @@ typedef struct _drm_i915_sarea {
#define DRM_IOCTL_I915_GEM_PIN DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_PIN, struct drm_i915_gem_pin)
#define DRM_IOCTL_I915_GEM_UNPIN DRM_IOW(DRM_COMMAND_BASE + DRM_I915_GEM_UNPIN, struct drm_i915_gem_unpin)
#define DRM_IOCTL_I915_GEM_BUSY DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_BUSY, struct drm_i915_gem_busy)
+#define DRM_IOCTL_I915_GEM_SET_CACHEING DRM_IOW(DRM_COMMAND_BASE + DRM_I915_GEM_SET_CACHEING, struct drm_i915_gem_cacheing)
+#define DRM_IOCTL_I915_GEM_GET_CACHEING DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_GET_CACHEING, struct drm_i915_gem_cacheing)
#define DRM_IOCTL_I915_GEM_THROTTLE DRM_IO ( DRM_COMMAND_BASE + DRM_I915_GEM_THROTTLE)
#define DRM_IOCTL_I915_GEM_ENTERVT DRM_IO(DRM_COMMAND_BASE + DRM_I915_GEM_ENTERVT)
#define DRM_IOCTL_I915_GEM_LEAVEVT DRM_IO(DRM_COMMAND_BASE + DRM_I915_GEM_LEAVEVT)
@@ -707,6 +711,17 @@ struct drm_i915_gem_busy {
__u32 busy;
};
+#define I915_CACHEING_NONE 0
+#define I915_CACHEING_CACHED 1
+
+struct drm_i915_gem_cacheing {
+ /** Handle of the buffer to check for busy */
+ __u32 handle;
+
+ /** Cacheing level to apply or return value */
+ __u32 cacheing;
+};
+
#define I915_TILING_NONE 0
#define I915_TILING_X 1
#define I915_TILING_Y 2
--
1.7.10.4
