[Intel-gfx] [PATCH] drm/i915: Add soft-pinning API for execbuffer
Chris Wilson
chris at chris-wilson.co.uk
Fri Mar 6 01:44:07 PST 2015
Userspace can pass in an offset that it presumes the object is located
at. The kernel will then do its utmost to fit the object into that
location. The assumption is that userspace is handling its own object
locations (for example along with full-ppgtt) and that the kernel will
rarely have to make space for the user's requests.
Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
---
drivers/gpu/drm/i915/i915_drv.h | 5 +++
drivers/gpu/drm/i915/i915_gem.c | 53 ++++++++++++++++++++++--------
drivers/gpu/drm/i915/i915_gem_evict.c | 52 +++++++++++++++++++++++++++++
drivers/gpu/drm/i915/i915_gem_execbuffer.c | 9 ++++-
include/uapi/drm/i915_drm.h | 3 +-
5 files changed, 106 insertions(+), 16 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 38ee0811f53f..5ac96e2adc6c 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -2590,7 +2590,9 @@ void i915_gem_vma_destroy(struct i915_vma *vma);
#define PIN_NONBLOCK 0x2
#define PIN_GLOBAL 0x4
#define PIN_OFFSET_BIAS 0x8
+#define PIN_OFFSET_FIXED 0x10
#define PIN_OFFSET_MASK (~4095)
+
int __must_check i915_gem_object_pin_view(struct drm_i915_gem_object *obj,
struct i915_address_space *vm,
uint32_t size,
@@ -2942,6 +2944,9 @@ int __must_check i915_gem_evict_something(struct drm_device *dev,
unsigned long start,
unsigned long end,
unsigned flags);
+int __must_check
+i915_gem_evict_range(struct drm_device *dev, struct i915_address_space *vm,
+ unsigned long start, unsigned long end);
int i915_gem_evict_vm(struct i915_address_space *vm, bool do_idle);
int i915_gem_evict_everything(struct drm_device *dev);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 9d0df4d85693..b266b31690e4 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3592,22 +3592,43 @@ i915_gem_object_bind_to_vm(struct drm_i915_gem_object *obj,
if (IS_ERR(vma))
goto err_unpin;
+ if (flags & PIN_OFFSET_FIXED) {
+ uint64_t offset = flags & PIN_OFFSET_MASK;
+ if (offset & (alignment - 1)) {
+ vma = ERR_PTR(-EINVAL);
+ goto err_free_vma;
+ }
+ vma->node.start = offset;
+ vma->node.size = size;
+ vma->node.color = obj->cache_level;
+ ret = drm_mm_reserve_node(&vm->mm, &vma->node);
+ if (ret) {
+ ret = i915_gem_evict_range(dev, vm, start, end);
+ if (ret == 0)
+ ret = drm_mm_reserve_node(&vm->mm, &vma->node);
+ }
+ if (ret) {
+ vma = ERR_PTR(ret);
+ goto err_free_vma;
+ }
+ } else {
search_free:
- ret = drm_mm_insert_node_in_range_generic(&vm->mm, &vma->node,
- size, alignment,
- obj->cache_level,
- start, end,
- DRM_MM_SEARCH_DEFAULT,
- DRM_MM_CREATE_DEFAULT);
- if (ret) {
- ret = i915_gem_evict_something(dev, vm, size, alignment,
- obj->cache_level,
- start, end,
- flags);
- if (ret == 0)
- goto search_free;
+ ret = drm_mm_insert_node_in_range_generic(&vm->mm, &vma->node,
+ size, alignment,
+ obj->cache_level,
+ start, end,
+ DRM_MM_SEARCH_DEFAULT,
+ DRM_MM_CREATE_DEFAULT);
+ if (ret) {
+ ret = i915_gem_evict_something(dev, vm, size, alignment,
+ obj->cache_level,
+ start, end,
+ flags);
+ if (ret == 0)
+ goto search_free;
- goto err_free_vma;
+ goto err_free_vma;
+ }
}
if (WARN_ON(!i915_gem_valid_gtt_space(vma, obj->cache_level))) {
ret = -EINVAL;
@@ -4187,6 +4208,10 @@ i915_vma_misplaced(struct i915_vma *vma,
vma->node.start < (flags & PIN_OFFSET_MASK))
return true;
+ if (flags & PIN_OFFSET_FIXED &&
+ vma->node.start != (flags & PIN_OFFSET_MASK))
+ return true;
+
return false;
}
diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c
index e3a49d94da3a..c4b2ead0d805 100644
--- a/drivers/gpu/drm/i915/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/i915_gem_evict.c
@@ -195,6 +195,58 @@ found:
return ret;
}
+int
+i915_gem_evict_range(struct drm_device *dev, struct i915_address_space *vm,
+ unsigned long start, unsigned long end)
+{
+ struct drm_mm_node *node;
+ struct list_head eviction_list;
+ int ret = 0;
+
+ INIT_LIST_HEAD(&eviction_list);
+ drm_mm_for_each_node(node, &vm->mm) {
+ struct i915_vma *vma;
+
+ if (node->start + node->size <= start)
+ continue;
+ if (node->start >= end)
+ break;
+
+ vma = container_of(node, typeof(*vma), node);
+ if (vma->pin_count) {
+ ret = -EBUSY;
+ break;
+ }
+
+ if (WARN_ON(!list_empty(&vma->exec_list))) {
+ ret = -EINVAL;
+ break;
+ }
+
+ drm_gem_object_reference(&vma->obj->base);
+ list_add(&vma->exec_list, &eviction_list);
+ }
+
+ while (!list_empty(&eviction_list)) {
+ struct i915_vma *vma;
+ struct drm_gem_object *obj;
+
+ vma = list_first_entry(&eviction_list,
+ struct i915_vma,
+ exec_list);
+
+ obj = &vma->obj->base;
+
+ list_del_init(&vma->exec_list);
+ if (ret == 0)
+ ret = i915_vma_unbind(vma);
+
+ drm_gem_object_unreference(obj);
+ }
+
+ return ret;
+}
+
/**
* i915_gem_evict_vm - Evict all idle vmas from a vm
* @vm: Address space to cleanse
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 183f2914486e..8129a8c75a21 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -590,6 +590,8 @@ i915_gem_execbuffer_reserve_vma(struct i915_vma *vma,
flags |= PIN_GLOBAL;
if (entry->flags & __EXEC_OBJECT_NEEDS_BIAS)
flags |= BATCH_OFFSET_BIAS | PIN_OFFSET_BIAS;
+ if (entry->flags & EXEC_OBJECT_PINNED)
+ flags |= entry->offset | PIN_OFFSET_FIXED;
}
ret = i915_gem_object_pin(obj, vma->vm,
@@ -666,6 +668,10 @@ eb_vma_misplaced(struct i915_vma *vma)
if (vma->node.size < entry->pad_to_size)
return true;
+ if (entry->flags & EXEC_OBJECT_PINNED &&
+ vma->node.start != entry->offset)
+ return true;
+
if (entry->flags & __EXEC_OBJECT_NEEDS_BIAS &&
vma->node.start < BATCH_OFFSET_BIAS)
return true;
@@ -1392,7 +1398,8 @@ eb_get_batch(struct eb_vmas *eb)
* Note that actual hangs have only been observed on gen7, but for
* paranoia do it everywhere.
*/
- vma->exec_entry->flags |= __EXEC_OBJECT_NEEDS_BIAS;
+ if ((vma->exec_entry->flags & EXEC_OBJECT_PINNED) == 0)
+ vma->exec_entry->flags |= __EXEC_OBJECT_NEEDS_BIAS;
return vma->obj;
}
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index cb3417ca406f..ddef37e970d3 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -679,7 +679,8 @@ struct drm_i915_gem_exec_object2 {
#define EXEC_OBJECT_NEEDS_GTT (1<<1)
#define EXEC_OBJECT_WRITE (1<<2)
#define EXEC_OBJECT_PAD_TO_SIZE (1<<3)
-#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_PAD_TO_SIZE<<1)
+#define EXEC_OBJECT_PINNED (1<<4)
+#define __EXEC_OBJECT_UNKNOWN_FLAGS -(EXEC_OBJECT_PINNED<<1)
__u64 flags;
__u64 pad_to_size;
--
2.1.4
More information about the Intel-gfx
mailing list