drm: pre allocate node for create_block
Subsystems (3)
PATCH 1/6 drm: pre allocate node for create_block
For an upcoming patch where we introduce the i915 VMA, it's ideal to have the drm_mm_node as part of the VMA struct (ie. it's pre-allocated). Part of the conversion to VMAs is to kill off obj->gtt_space. Doing this will break a bunch of code, but amongst them are 2 callers of drm_mm_create_block(), both related to stolen memory.
It also allows us to embed the drm_mm_node into the object currently which provides a nice transition over to the new code.
v2: Reordered to do before ripping out obj->gtt_offset. Some minor cleanups made available because of reordering.
--- drivers/gpu/drm/drm_mm.c | 16 +++++---------- drivers/gpu/drm/i915/i915_gem_gtt.c | 18 +++++++++++++---- drivers/gpu/drm/i915/i915_gem_stolen.c | 36 +++++++++++++++++++++++----------- include/drm/drm_mm.h | 9 +++++---- 4 files changed, 49 insertions(+), 30 deletions(-)
drivers/gpu/drm/drm_mm.c +5 -11
@@ -147,12 +147,10 @@ static void drm_mm_insert_helper(struct drm_mm_node *hole_node, } } -struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic) +int drm_mm_create_block (struct drm_mm *mm,struct drm_mm_node *node, +unsigned long start,unsigned long size) { -struct drm_mm_node *hole, *node; +struct drm_mm_node *hole;unsigned long end = start + size;unsigned long hole_start;unsigned long hole_end;
@@ -161,10 +159,6 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm,if (hole_start > start || hole_end < end)continue ; - node =drm_mm_kmalloc (mm, atomic); -if (unlikely (node ==NULL )) -return NULL ; - node->start = start; node->size = size; node->mm = mm;
@@ -184,11 +178,11 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm, node->hole_follows =1 ; } -return node; +return 0 ; }WARN (1 ,"no hole found for block 0x%lx + 0x%lx\n" , start, size); -return NULL ; +return -ENOSPC ; }EXPORT_SYMBOL (drm_mm_create_block);
drivers/gpu/drm/i915/i915_gem_gtt.c +14 -4
@@ -629,14 +629,24 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,/* Mark any preallocated objects as occupied */ list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) { +int ret;DRM_DEBUG_KMS ("reserving preallocated space: %x + %zx\n" , obj->gtt_offset, obj->base.size);BUG_ON (obj->gtt_space !=I915_GTT_RESERVED ); - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_offset, - obj->base.size, - false); + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_ERROR ("Failed to preserve all objects\n" ); +break ; + } + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + obj->gtt_offset, + obj->base.size); +if (ret) { +DRM_DEBUG_KMS ("Reservation failed\n" ); +kfree (obj->gtt_space); + } obj->has_global_gtt_mapping =1 ; }
drivers/gpu/drm/i915/i915_gem_stolen.c +25 -11
@@ -330,6 +330,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,struct drm_i915_private *dev_priv = dev->dev_private;struct drm_i915_gem_object *obj;struct drm_mm_node *stolen; +int ret;if (dev_priv->mm.stolen_base ==0 )return NULL ;
@@ -344,11 +345,15 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,if (WARN_ON (size ==0 ))return NULL ; - stolen =drm_mm_create_block (&dev_priv->mm.stolen, - stolen_offset, size, - false); -if (stolen ==NULL ) { + stolen =kzalloc (sizeof (*stolen),GFP_KERNEL ); +if (!stolen) +return NULL ; + + ret =drm_mm_create_block (&dev_priv->mm.stolen, stolen, stolen_offset, + size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen space\n" ); +kfree (stolen);return NULL ; }
@@ -369,13 +374,18 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - gtt_offset, size, - false); -if (obj->gtt_space ==NULL ) { + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); + goto unref_out; + } + + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + gtt_offset, size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); -drm_gem_object_unreference (&obj->base); -return NULL ; + goto unref_out; } }else obj->gtt_space =I915_GTT_RESERVED ;
@@ -385,8 +395,12 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,list_add_tail (&obj->global_list, &dev_priv->mm.bound_list);list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list); -return obj; + +unref_out: +drm_gem_object_unreference (&obj->base); +drm_mm_put_block (stolen); +return NULL ; }void
include/drm/drm_mm.h +5 -4
@@ -138,10 +138,10 @@ static inline unsigned long drm_mm_hole_node_end(struct drm_mm_node *hole_node)/* * Basic range manager support (drm_mm.c) */ -extern struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic); +extern int drm_mm_create_block (struct drm_mm *mm, +struct drm_mm_node *node, +unsigned long start, +unsigned long size); extern struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *node, unsigned long size, unsigned alignment,
@@ -155,6 +155,7 @@ extern struct drm_mm_node *drm_mm_get_block_range_generic(unsigned long start, unsigned long end, int atomic); +static inline struct drm_mm_node *drm_mm_get_block (struct drm_mm_node *parent,unsigned long size,unsigned alignment)
Signature · 1 line
1.8.3.2
PATCH 2/6 drm/i915: Getter/setter for object attributes
Soon we want to gut a lot of our existing assumptions how many address spaces an object can live in, and in doing so, embed the drm_mm_node in the object (and later the VMA).
It's possible in the future we'll want to add more getter/setter methods, but for now this is enough to enable the VMAs.
v2: Reworked commit message (Ben)
Added comments to the main functions (Ben)
sed -i "s/i915_gem_obj_set_color/i915_gem_obj_ggtt_set_color/" drivers/gpu/drm/i915/*.[ch] sed -i "s/i915_gem_obj_bound/i915_gem_obj_ggtt_bound/" drivers/gpu/drm/i915/*.[ch] sed -i "s/i915_gem_obj_size/i915_gem_obj_ggtt_size/" drivers/gpu/drm/i915/*.[ch] sed -i "s/i915_gem_obj_offset/i915_gem_obj_ggtt_offset/" drivers/gpu/drm/i915/*.[ch] (Daniel)
--- drivers/gpu/drm/i915/i915_debugfs.c | 26 ++++---- drivers/gpu/drm/i915/i915_drv.h | 35 +++++++++- drivers/gpu/drm/i915/i915_gem.c | 101 +++++++++++++++-------------- drivers/gpu/drm/i915/i915_gem_context.c | 2 +- drivers/gpu/drm/i915/i915_gem_evict.c | 6 +- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 19 +++--- drivers/gpu/drm/i915/i915_gem_gtt.c | 24 +++---- drivers/gpu/drm/i915/i915_gem_stolen.c | 10 +-- drivers/gpu/drm/i915/i915_gem_tiling.c | 14 ++-- drivers/gpu/drm/i915/i915_irq.c | 15 ++--- drivers/gpu/drm/i915/i915_trace.h | 8 +-- drivers/gpu/drm/i915/intel_display.c | 28 ++++---- drivers/gpu/drm/i915/intel_fb.c | 8 +-- drivers/gpu/drm/i915/intel_overlay.c | 14 ++-- drivers/gpu/drm/i915/intel_pm.c | 8 +-- drivers/gpu/drm/i915/intel_ringbuffer.c | 12 ++-- drivers/gpu/drm/i915/intel_sprite.c | 8 ++- 17 files changed, 188 insertions(+), 150 deletions(-)
drivers/gpu/drm/i915/i915_debugfs.c +14 -12
@@ -122,9 +122,9 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)seq_printf (m," (pinned x %d)" , obj->pin_count);if (obj->fence_reg !=I915_FENCE_REG_NONE )seq_printf (m," (fence: %d)" , obj->fence_reg); -if (obj->gtt_space !=NULL ) -seq_printf (m," (gtt offset: %08x, size: %08x)" , - obj->gtt_offset, (unsigned int )obj->gtt_space->size); +if (i915_gem_obj_ggtt_bound (obj)) +seq_printf (m," (gtt offset: %08lx, size: %08x)" , +i915_gem_obj_ggtt_offset (obj), (unsigned int )i915_gem_obj_ggtt_size (obj));if (obj->stolen)seq_printf (m," (stolen: %08lx)" , obj->stolen->start);if (obj->pin_mappable || obj->fault_mappable) {
@@ -175,7 +175,7 @@ static int i915_gem_object_list_info(struct seq_file *m, void *data)describe_obj (m, obj);seq_putc (m,'\n' ); total_obj_size += obj->base.size; - total_gtt_size += obj->gtt_space->size; + total_gtt_size +=i915_gem_obj_ggtt_size (obj); count++; }mutex_unlock (&dev->struct_mutex);
@@ -187,10 +187,10 @@ static int i915_gem_object_list_info(struct seq_file *m, void *data)#define count_objects (list, member) do { \ list_for_each_entry(obj, list, member) { \ - size += obj->gtt_space->size; \ + size += i915_gem_obj_ggtt_size(obj); \ ++count; \ if (obj->map_and_fenceable) { \ - mappable_size += obj->gtt_space->size; \ + mappable_size += i915_gem_obj_ggtt_size(obj); \ ++mappable_count; \ } \ } \
@@ -209,7 +209,7 @@ static int per_file_stats(int id, void *ptr, void *data) stats->count++; stats->total += obj->base.size; -if (obj->gtt_space) { +if (i915_gem_obj_ggtt_bound (obj)) {if (!list_empty (&obj->ring_list)) stats->active += obj->base.size;else
@@ -267,11 +267,11 @@ static int i915_gem_object_info(struct seq_file *m, void *data) size = count = mappable_size = mappable_count =0 ;list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) {if (obj->fault_mappable) { - size += obj->gtt_space->size; + size +=i915_gem_obj_ggtt_size (obj); ++count; }if (obj->pin_mappable) { - mappable_size += obj->gtt_space->size; + mappable_size +=i915_gem_obj_ggtt_size (obj); ++mappable_count; }if (obj->madv ==I915_MADV_DONTNEED ) {
@@ -333,7 +333,7 @@ static int i915_gem_gtt_info(struct seq_file *m, void *data)describe_obj (m, obj);seq_putc (m,'\n' ); total_obj_size += obj->base.size; - total_gtt_size += obj->gtt_space->size; + total_gtt_size +=i915_gem_obj_ggtt_size (obj); count++; }
@@ -379,12 +379,14 @@ static int i915_gem_pageflip_info(struct seq_file *m, void *data)if (work->old_fb_obj) {struct drm_i915_gem_object *obj = work->old_fb_obj;if (obj) -seq_printf (m,"Old framebuffer gtt_offset 0x%08x\n" , obj->gtt_offset); +seq_printf (m,"Old framebuffer gtt_offset 0x%08lx\n" , +i915_gem_obj_ggtt_offset (obj)); }if (work->pending_flip_obj) {struct drm_i915_gem_object *obj = work->pending_flip_obj;if (obj) -seq_printf (m,"New framebuffer gtt_offset 0x%08x\n" , obj->gtt_offset); +seq_printf (m,"New framebuffer gtt_offset 0x%08lx\n" , +i915_gem_obj_ggtt_offset (obj)); } }spin_unlock_irqrestore (&dev->event_lock, flags);
drivers/gpu/drm/i915/i915_drv.h +33 -2
@@ -1227,7 +1227,7 @@ struct drm_i915_gem_object {const struct drm_i915_gem_object_ops *ops;/** Current space allocated to this object in the GTT, if any. */ -struct drm_mm_node *gtt_space; +struct drm_mm_node *ggtt_space;/** Stolen memory for this object, instead of being backed by shmem. */ struct drm_mm_node *stolen;struct list_head global_list;
@@ -1333,7 +1333,7 @@ struct drm_i915_gem_object {* * This is the same as gtt_space->start */ -uint32_t gtt_offset; +uint32_t ggtt_offset;struct intel_ring_buffer *ring;
@@ -1360,6 +1360,37 @@ struct drm_i915_gem_object {#define to_intel_bo (x) container_of(x, struct drm_i915_gem_object, base) +/* Offset of the first PTE pointing to this object */ +static inline unsigned long +i915_gem_obj_ggtt_offset (struct drm_i915_gem_object *o) +{ +return o->ggtt_space->start; +} + +/* Whether or not this object is currently mapped by the translation tables */ +static inline bool +i915_gem_obj_ggtt_bound (struct drm_i915_gem_object *o) +{ +return o->ggtt_space !=NULL ; +} + +/* The size used in the translation tables may be larger than the actual size of +* the object on GEN2/GEN3 because of the way tiling is handled. See +* i915_gem_get_gtt_size() for more details. +*/ +static inline unsigned long +i915_gem_obj_ggtt_size (struct drm_i915_gem_object *o) +{ +return o->ggtt_space->size; +} + +static inline void +i915_gem_obj_ggtt_set_color (struct drm_i915_gem_object *o, +enum i915_cache_level color) +{ + o->ggtt_space->color = color; +} +/** * Request queue structure. *
drivers/gpu/drm/i915/i915_gem.c +51 -50
281 lines not on this page · raw message
drivers/gpu/drm/i915/i915_gem_context.c +1 -1
@@ -377,7 +377,7 @@ mi_set_context(struct intel_ring_buffer *ring,intel_ring_emit (ring,MI_NOOP );intel_ring_emit (ring,MI_SET_CONTEXT ); - intel_ring_emit(ring, new_context->obj->gtt_offset | + intel_ring_emit(ring, i915_gem_obj_ggtt_offset(new_context->obj) | MI_MM_SPACE_GTT | MI_SAVE_EXT_STATE_EN | MI_RESTORE_EXT_STATE_EN |
drivers/gpu/drm/i915/i915_gem_evict.c +3 -3
@@ -38,7 +38,7 @@ mark_free(struct drm_i915_gem_object *obj, struct list_head *unwind)return false;list_add (&obj->exec_list, unwind); -return drm_mm_scan_add_block (obj->gtt_space); +return drm_mm_scan_add_block (obj->ggtt_space); }int
@@ -107,7 +107,7 @@ none: struct drm_i915_gem_object, exec_list); - ret =drm_mm_scan_remove_block (obj->gtt_space); + ret =drm_mm_scan_remove_block (obj->ggtt_space);BUG_ON (ret);list_del_init (&obj->exec_list);
@@ -127,7 +127,7 @@ found: obj =list_first_entry (&unwind_list, struct drm_i915_gem_object, exec_list); -if (drm_mm_scan_remove_block (obj->gtt_space)) { +if (drm_mm_scan_remove_block (obj->ggtt_space)) {list_move (&obj->exec_list, &eviction_list);drm_gem_object_reference (&obj->base);continue ;
drivers/gpu/drm/i915/i915_gem_execbuffer.c +10 -9
@@ -188,7 +188,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,return -ENOENT ; target_i915_obj =to_intel_bo (target_obj); - target_offset = target_i915_obj->gtt_offset; + target_offset =i915_gem_obj_ggtt_offset (target_i915_obj);/* Sandybridge PPGTT errata: We need a global gtt mapping for MI and * pipe_control writes because the gpu doesn't properly redirect them
@@ -280,7 +280,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,return ret;/* Map the page containing the relocation we're going to perform. */ - reloc->offset += obj->gtt_offset; + reloc->offset +=i915_gem_obj_ggtt_offset (obj); reloc_page =io_mapping_map_atomic_wc (dev_priv->gtt.mappable, reloc->offset &PAGE_MASK ); reloc_entry = (uint32_t __iomem *)
@@ -436,8 +436,8 @@ i915_gem_execbuffer_reserve_object(struct drm_i915_gem_object *obj, obj->has_aliasing_ppgtt_mapping =1 ; } -if (entry->offset != obj->gtt_offset) { - entry->offset = obj->gtt_offset; +if (entry->offset !=i915_gem_obj_ggtt_offset (obj)) { + entry->offset =i915_gem_obj_ggtt_offset (obj); *need_reloc = true; }
@@ -458,7 +458,7 @@ i915_gem_execbuffer_unreserve_object(struct drm_i915_gem_object *obj) {struct drm_i915_gem_exec_object2 *entry; -if (!obj->gtt_space) +if (!i915_gem_obj_ggtt_bound (obj))return ; entry = obj->exec_entry;
@@ -530,7 +530,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,struct drm_i915_gem_exec_object2 *entry = obj->exec_entry;bool need_fence, need_mappable; -if (!obj->gtt_space) +if (!i915_gem_obj_ggtt_bound (obj))continue ; need_fence =
@@ -539,7 +539,8 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring, obj->tiling_mode !=I915_TILING_NONE ; need_mappable = need_fence ||need_reloc_mappable (obj); -if ((entry->alignment && obj->gtt_offset & (entry->alignment -1 )) || +if ((entry->alignment && +i915_gem_obj_ggtt_offset (obj) & (entry->alignment -1 )) || (need_mappable && !obj->map_and_fenceable)) ret =i915_gem_object_unbind (obj);else
@@ -550,7 +551,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,/* Bind fresh objects */ list_for_each_entry (obj, objects, exec_list) { -if (obj->gtt_space) +if (i915_gem_obj_ggtt_bound (obj))continue ; ret =i915_gem_execbuffer_reserve_object (obj, ring, need_relocs);
@@ -1058,7 +1059,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, goto err; } - exec_start = batch_obj->gtt_offset + args->batch_start_offset; + exec_start =i915_gem_obj_ggtt_offset (batch_obj) + args->batch_start_offset; exec_len = args->batch_len;if (cliprects) {for (i =0 ; i < args->num_cliprects; i++) {
drivers/gpu/drm/i915/i915_gem_gtt.c +12 -12
@@ -378,7 +378,7 @@ void i915_ppgtt_bind_object(struct i915_hw_ppgtt *ppgtt, enum i915_cache_level cache_level) { ppgtt->insert_entries (ppgtt, obj->pages, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , cache_level); }
@@ -386,7 +386,7 @@ void i915_ppgtt_unbind_object(struct i915_hw_ppgtt *ppgtt, struct drm_i915_gem_object *obj) { ppgtt->clear_range (ppgtt, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , obj->base.size >>PAGE_SHIFT ); }
@@ -551,7 +551,7 @@ void i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj,struct drm_i915_private *dev_priv = dev->dev_private; dev_priv->gtt.gtt_insert_entries (dev, obj->pages, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , cache_level); obj->has_global_gtt_mapping =1 ;
@@ -563,7 +563,7 @@ void i915_gem_gtt_unbind_object(struct drm_i915_gem_object *obj)struct drm_i915_private *dev_priv = dev->dev_private; dev_priv->gtt.gtt_clear_range (obj->base.dev, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , obj->base.size >>PAGE_SHIFT ); obj->has_global_gtt_mapping =0 ;
@@ -630,22 +630,22 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,/* Mark any preallocated objects as occupied */ list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) {int ret; -DRM_DEBUG_KMS ("reserving preallocated space: %x + %zx\n" , - obj->gtt_offset, obj->base.size); +DRM_DEBUG_KMS ("reserving preallocated space: %lx + %zx\n" , +i915_gem_obj_ggtt_offset (obj), obj->base.size); -BUG_ON (obj->gtt_space !=I915_GTT_RESERVED ); - obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); -if (!obj->gtt_space) { +BUG_ON (obj->ggtt_space !=I915_GTT_RESERVED ); + obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL ); +if (!obj->ggtt_space) {DRM_ERROR ("Failed to preserve all objects\n" );break ; } ret =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_space, - obj->gtt_offset, + obj->ggtt_space, +i915_gem_obj_ggtt_offset (obj), obj->base.size);if (ret) {DRM_DEBUG_KMS ("Reservation failed\n" ); -kfree (obj->gtt_space); +kfree (obj->ggtt_space); } obj->has_global_gtt_mapping =1 ; }
drivers/gpu/drm/i915/i915_gem_stolen.c +5 -5
@@ -374,23 +374,23 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); -if (!obj->gtt_space) { + obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL ); +if (!obj->ggtt_space) {DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); goto unref_out; } ret =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_space, + obj->ggtt_space, gtt_offset, size);if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); goto unref_out; } }else - obj->gtt_space =I915_GTT_RESERVED ; + obj->ggtt_space =I915_GTT_RESERVED ; - obj->gtt_offset = gtt_offset; + obj->ggtt_offset = gtt_offset; obj->has_global_gtt_mapping =1 ;list_add_tail (&obj->global_list, &dev_priv->mm.bound_list);
drivers/gpu/drm/i915/i915_gem_tiling.c +7 -7
@@ -268,18 +268,18 @@ i915_gem_object_fence_ok(struct drm_i915_gem_object *obj, int tiling_mode)return true;if (INTEL_INFO (obj->base.dev)->gen ==3 ) { -if (obj->gtt_offset & ~I915_FENCE_START_MASK ) +if (i915_gem_obj_ggtt_offset (obj) & ~I915_FENCE_START_MASK )return false; }else { -if (obj->gtt_offset & ~I830_FENCE_START_MASK ) +if (i915_gem_obj_ggtt_offset (obj) & ~I830_FENCE_START_MASK )return false; } size =i915_gem_get_gtt_size (obj->base.dev, obj->base.size, tiling_mode); -if (obj->gtt_space->size != size) +if (i915_gem_obj_ggtt_size (obj) != size)return false; -if (obj->gtt_offset & (size -1 )) +if (i915_gem_obj_ggtt_offset (obj) & (size -1 ))return false;return true;
@@ -359,8 +359,8 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,*/ obj->map_and_fenceable = - obj->gtt_space ==NULL || - (obj->gtt_offset + obj->base.size <= dev_priv->gtt.mappable_end && + !i915_gem_obj_ggtt_bound (obj) || + (i915_gem_obj_ggtt_offset (obj) + obj->base.size <= dev_priv->gtt.mappable_end &&i915_gem_object_fence_ok (obj, args->tiling_mode));/* Rebind if we need a change of alignment */
@@ -369,7 +369,7 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,i915_gem_get_gtt_alignment (dev, obj->base.size, args->tiling_mode, false); -if (obj->gtt_offset & (unfenced_alignment -1 )) +if (i915_gem_obj_ggtt_offset (obj) & (unfenced_alignment -1 )) ret =i915_gem_object_unbind (obj); }
drivers/gpu/drm/i915/i915_irq.c +7 -8
@@ -1516,7 +1516,7 @@ i915_error_object_create_sized(struct drm_i915_private *dev_priv,if (dst ==NULL )return NULL ; - reloc_offset = src->gtt_offset; + reloc_offset = dst->gtt_offset =i915_gem_obj_ggtt_offset (src);for (i =0 ; i < num_pages; i++) {unsigned long flags;void *d;
@@ -1568,7 +1568,6 @@ i915_error_object_create_sized(struct drm_i915_private *dev_priv, reloc_offset +=PAGE_SIZE ; } dst->page_count = num_pages; - dst->gtt_offset = src->gtt_offset;return dst;
@@ -1622,7 +1621,7 @@ static void capture_bo(struct drm_i915_error_buffer *err, err->name = obj->base.name; err->rseqno = obj->last_read_seqno; err->wseqno = obj->last_write_seqno; - err->gtt_offset = obj->gtt_offset; + err->gtt_offset =i915_gem_obj_ggtt_offset (obj); err->read_domains = obj->base.read_domains; err->write_domain = obj->base.write_domain; err->fence_reg = obj->fence_reg;
@@ -1720,8 +1719,8 @@ i915_error_first_batchbuffer(struct drm_i915_private *dev_priv,return NULL ; obj = ring->private; -if (acthd >= obj->gtt_offset && - acthd < obj->gtt_offset + obj->base.size) +if (acthd >=i915_gem_obj_ggtt_offset (obj) && + acthd <i915_gem_obj_ggtt_offset (obj) + obj->base.size)return i915_error_object_create (dev_priv, obj); }
@@ -1802,7 +1801,7 @@ static void i915_gem_record_active_context(struct intel_ring_buffer *ring,return ;list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) { -if ((error->ccid &PAGE_MASK ) == obj->gtt_offset) { +if ((error->ccid &PAGE_MASK ) ==i915_gem_obj_ggtt_offset (obj)) { ering->ctx =i915_error_object_create_sized (dev_priv, obj,1 );break ;
@@ -2156,10 +2155,10 @@ static void __always_unused i915_pageflip_stall_check(struct drm_device *dev, inif (INTEL_INFO (dev)->gen >=4 ) {int dspsurf =DSPSURF (intel_crtc->plane); stall_detected =I915_HI_DISPBASE (I915_READ (dspsurf)) == - obj->gtt_offset; +i915_gem_obj_ggtt_offset (obj); }else {int dspaddr =DSPADDR (intel_crtc->plane); - stall_detected =I915_READ (dspaddr) == (obj->gtt_offset + + stall_detected =I915_READ (dspaddr) == (i915_gem_obj_ggtt_offset (obj) + crtc->y * crtc->fb->pitches[0 ] + crtc->x * crtc->fb->bits_per_pixel/8 ); }
drivers/gpu/drm/i915/i915_trace.h +4 -4
@@ -46,8 +46,8 @@ TRACE_EVENT(i915_gem_object_bind,TP_fast_assign ( __entry->obj = obj; - __entry->offset = obj->gtt_space->start; - __entry->size = obj->gtt_space->size; + __entry->offset =i915_gem_obj_ggtt_offset (obj); + __entry->size =i915_gem_obj_ggtt_size (obj); __entry->mappable = mappable; ),
@@ -68,8 +68,8 @@ TRACE_EVENT(i915_gem_object_unbind,TP_fast_assign ( __entry->obj = obj; - __entry->offset = obj->gtt_space->start; - __entry->size = obj->gtt_space->size; + __entry->offset =i915_gem_obj_ggtt_offset (obj); + __entry->size =i915_gem_obj_ggtt_size (obj); ), TP_printk("obj=%p, offset=%08x size=%x",
drivers/gpu/drm/i915/intel_display.c +15 -13
@@ -1980,16 +1980,17 @@ static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb, intel_crtc->dspaddr_offset = linear_offset; } -DRM_DEBUG_KMS ("Writing base %08X %08lX %d %d %d\n" , - obj->gtt_offset, linear_offset, x, y, fb->pitches[0 ]); +DRM_DEBUG_KMS ("Writing base %08lX %08lX %d %d %d\n" , +i915_gem_obj_ggtt_offset (obj), linear_offset, x, y, + fb->pitches[0 ]);I915_WRITE (DSPSTRIDE (plane), fb->pitches[0 ]);if (INTEL_INFO (dev)->gen >=4 ) {I915_MODIFY_DISPBASE (DSPSURF (plane), - obj->gtt_offset + intel_crtc->dspaddr_offset); +i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);I915_WRITE (DSPTILEOFF (plane), (y <<16 ) | x);I915_WRITE (DSPLINOFF (plane), linear_offset); }else -I915_WRITE (DSPADDR (plane), obj->gtt_offset + linear_offset); +I915_WRITE (DSPADDR (plane),i915_gem_obj_ggtt_offset (obj) + linear_offset);POSTING_READ (reg);return 0 ;
@@ -2069,11 +2070,12 @@ static int ironlake_update_plane(struct drm_crtc *crtc, fb->pitches[0]); linear_offset -= intel_crtc->dspaddr_offset; -DRM_DEBUG_KMS ("Writing base %08X %08lX %d %d %d\n" , - obj->gtt_offset, linear_offset, x, y, fb->pitches[0 ]); +DRM_DEBUG_KMS ("Writing base %08lX %08lX %d %d %d\n" , +i915_gem_obj_ggtt_offset (obj), linear_offset, x, y, + fb->pitches[0 ]);I915_WRITE (DSPSTRIDE (plane), fb->pitches[0 ]);I915_MODIFY_DISPBASE (DSPSURF (plane), - obj->gtt_offset + intel_crtc->dspaddr_offset); +i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);if (IS_HASWELL (dev)) {I915_WRITE (DSPOFFSET (plane), (y <<16 ) | x); }else {
@@ -6566,7 +6568,7 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc, goto fail_unpin; } - addr = obj->gtt_offset; + addr =i915_gem_obj_ggtt_offset (obj); }else {int align =IS_I830 (dev) ?16 *1024 :256 ; ret = i915_gem_attach_phys_object(dev, obj,
@@ -7338,7 +7340,7 @@ static int intel_gen2_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP |MI_DISPLAY_FLIP_PLANE (intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ]); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);intel_ring_emit (ring,0 );/* aux display base address, unused */ intel_mark_page_flip_active (intel_crtc);
@@ -7379,7 +7381,7 @@ static int intel_gen3_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP_I915 |MI_DISPLAY_FLIP_PLANE (intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ]); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);intel_ring_emit (ring,MI_NOOP );intel_mark_page_flip_active (intel_crtc);
@@ -7419,7 +7421,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev, MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ]);intel_ring_emit (ring, - (obj->gtt_offset + intel_crtc->dspaddr_offset) | + (i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset) | obj->tiling_mode);/* XXX Enabling the panel-fitter across page-flip is so far
@@ -7462,7 +7464,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP |MI_DISPLAY_FLIP_PLANE (intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ] | obj->tiling_mode); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);/* Contrary to the suggestions in the documentation, * "Enable Panel Fitter" does not seem to be required when page
@@ -7527,7 +7529,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP_I915 | plane_bit);intel_ring_emit (ring, (fb->pitches[0 ] | obj->tiling_mode)); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);intel_ring_emit (ring, (MI_NOOP ));intel_mark_page_flip_active (intel_crtc);
drivers/gpu/drm/i915/intel_fb.c +4 -4
@@ -139,11 +139,11 @@ static int intelfb_create(struct drm_fb_helper *helper, info->apertures->ranges[0 ].base = dev->mode_config.fb_base; info->apertures->ranges[0 ].size = dev_priv->gtt.mappable_end; - info->fix.smem_start = dev->mode_config.fb_base + obj->gtt_offset; + info->fix.smem_start = dev->mode_config.fb_base +i915_gem_obj_ggtt_offset (obj); info->fix.smem_len = size; info->screen_base = -ioremap_wc (dev_priv->gtt.mappable_base + obj->gtt_offset, +ioremap_wc (dev_priv->gtt.mappable_base +i915_gem_obj_ggtt_offset (obj), size);if (!info->screen_base) { ret = -ENOSPC ;
@@ -166,9 +166,9 @@ static int intelfb_create(struct drm_fb_helper *helper,/* Use default scratch pixmap (info->pixmap.flags = FB_PIXMAP_SYSTEM) */ -DRM_DEBUG_KMS ("allocated %dx%d fb: 0x%08x, bo %p\n" , +DRM_DEBUG_KMS ("allocated %dx%d fb: 0x%08lx, bo %p\n" , fb->width, fb->height, - obj->gtt_offset, obj); +i915_gem_obj_ggtt_offset (obj), obj);mutex_unlock (&dev->struct_mutex);
drivers/gpu/drm/i915/intel_overlay.c +7 -7
@@ -196,7 +196,7 @@ intel_overlay_map_regs(struct intel_overlay *overlay) regs = (struct overlay_registers __iomem *)overlay->reg_bo->phys_obj->handle->vaddr; else regs =io_mapping_map_wc (dev_priv->gtt.mappable, - overlay->reg_bo->gtt_offset); +i915_gem_obj_ggtt_offset (overlay->reg_bo));return regs; }
@@ -740,7 +740,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay, swidth = params->src_w; swidthsw =calc_swidthsw (overlay->dev, params->offset_Y, tmp_width); sheight = params->src_h; -iowrite32 (new_bo->gtt_offset + params->offset_Y, ®s->OBUF_0Y); +iowrite32 (i915_gem_obj_ggtt_offset (new_bo) + params->offset_Y, ®s->OBUF_0Y); ostride = params->stride_Y;if (params->format &I915_OVERLAY_YUV_PLANAR ) {
@@ -754,8 +754,8 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay, params->src_w/uv_hscale); swidthsw |=max_t (u32, tmp_U, tmp_V) <<16 ; sheight |= (params->src_h/uv_vscale) <<16 ; -iowrite32 (new_bo->gtt_offset + params->offset_U, ®s->OBUF_0U); -iowrite32 (new_bo->gtt_offset + params->offset_V, ®s->OBUF_0V); +iowrite32 (i915_gem_obj_ggtt_offset (new_bo) + params->offset_U, ®s->OBUF_0U); +iowrite32 (i915_gem_obj_ggtt_offset (new_bo) + params->offset_V, ®s->OBUF_0V); ostride |= params->stride_UV <<16 ; }
@@ -1355,7 +1355,7 @@ void intel_setup_overlay(struct drm_device *dev)DRM_ERROR ("failed to pin overlay register bo\n" ); goto out_free_bo; } - overlay->flip_addr = reg_bo->gtt_offset; + overlay->flip_addr =i915_gem_obj_ggtt_offset (reg_bo); ret =i915_gem_object_set_to_gtt_domain (reg_bo, true);if (ret) {
@@ -1435,7 +1435,7 @@ intel_overlay_map_regs_atomic(struct intel_overlay *overlay) overlay->reg_bo->phys_obj->handle->vaddr; else regs =io_mapping_map_atomic_wc (dev_priv->gtt.mappable, - overlay->reg_bo->gtt_offset); +i915_gem_obj_ggtt_offset (overlay->reg_bo));return regs; }
@@ -1468,7 +1468,7 @@ intel_overlay_capture_error_state(struct drm_device *dev)if (OVERLAY_NEEDS_PHYSICAL (overlay->dev)) error->base = (__force long )overlay->reg_bo->phys_obj->handle->vaddr;else - error->base = overlay->reg_bo->gtt_offset; + error->base =i915_gem_obj_ggtt_offset (overlay->reg_bo); regs =intel_overlay_map_regs_atomic (overlay);if (!regs)
drivers/gpu/drm/i915/intel_pm.c +4 -4
@@ -218,7 +218,7 @@ static void ironlake_enable_fbc(struct drm_crtc *crtc, unsigned long interval) (stall_watermark <<DPFC_RECOMP_STALL_WM_SHIFT ) | (interval << DPFC_RECOMP_TIMER_COUNT_SHIFT));I915_WRITE (ILK_DPFC_FENCE_YOFF , crtc->y); -I915_WRITE (ILK_FBC_RT_BASE , obj->gtt_offset |ILK_FBC_RT_VALID ); +I915_WRITE (ILK_FBC_RT_BASE ,i915_gem_obj_ggtt_offset (obj) |ILK_FBC_RT_VALID );/* enable it... */ I915_WRITE (ILK_DPFC_CONTROL , dpfc_ctl |DPFC_CTL_EN );
@@ -275,7 +275,7 @@ static void gen7_enable_fbc(struct drm_crtc *crtc, unsigned long interval)struct drm_i915_gem_object *obj = intel_fb->obj;struct intel_crtc *intel_crtc =to_intel_crtc (crtc); -I915_WRITE (IVB_FBC_RT_BASE , obj->gtt_offset); +I915_WRITE (IVB_FBC_RT_BASE ,i915_gem_obj_ggtt_offset (obj)); I915_WRITE(ILK_DPFC_CONTROL, DPFC_CTL_EN | DPFC_CTL_LIMIT_1X | IVB_DPFC_CTL_FENCE_EN |
@@ -3707,7 +3707,7 @@ static void ironlake_enable_rc6(struct drm_device *dev)intel_ring_emit (ring,MI_SUSPEND_FLUSH |MI_SUSPEND_FLUSH_EN );intel_ring_emit (ring,MI_SET_CONTEXT ); - intel_ring_emit(ring, dev_priv->ips.renderctx->gtt_offset | + intel_ring_emit(ring, i915_gem_obj_ggtt_offset(dev_priv->ips.renderctx) | MI_MM_SPACE_GTT | MI_SAVE_EXT_STATE_EN | MI_RESTORE_EXT_STATE_EN |
@@ -3730,7 +3730,7 @@ static void ironlake_enable_rc6(struct drm_device *dev)return ; } -I915_WRITE (PWRCTXA , dev_priv->ips.pwrctx->gtt_offset |PWRCTX_EN ); +I915_WRITE (PWRCTXA ,i915_gem_obj_ggtt_offset (dev_priv->ips.pwrctx) |PWRCTX_EN );I915_WRITE (RSTDBYCTL ,I915_READ (RSTDBYCTL ) & ~RCX_SW_EXIT ); }
drivers/gpu/drm/i915/intel_ringbuffer.c +6 -6
@@ -424,14 +424,14 @@ static int init_ring_common(struct intel_ring_buffer *ring)* registers with the above sequence (the readback of the HEAD registers * also enforces ordering), otherwise the hw might lose the new ring * register values. */ -I915_WRITE_START (ring, obj->gtt_offset); +I915_WRITE_START (ring,i915_gem_obj_ggtt_offset (obj));I915_WRITE_CTL (ring, ((ring->size -PAGE_SIZE ) &RING_NR_PAGES ) |RING_VALID );/* If the head is still not zero, the ring is dead */ if (wait_for ((I915_READ_CTL (ring) &RING_VALID ) !=0 && -I915_READ_START (ring) == obj->gtt_offset && +I915_READ_START (ring) ==i915_gem_obj_ggtt_offset (obj) && (I915_READ_HEAD (ring) &HEAD_ADDR ) ==0 ,50 )) { DRM_ERROR("%s initialization failed " "ctl %08x head %08x tail %08x start %08x\n",
@@ -489,7 +489,7 @@ init_pipe_control(struct intel_ring_buffer *ring)if (ret) goto err_unref; - pc->gtt_offset = obj->gtt_offset; + pc->gtt_offset =i915_gem_obj_ggtt_offset (obj); pc->cpu_page =kmap (sg_page (obj->pages->sgl));if (pc->cpu_page ==NULL ) { ret = -ENOMEM ;
@@ -1129,7 +1129,7 @@ i830_dispatch_execbuffer(struct intel_ring_buffer *ring,intel_ring_advance (ring); }else {struct drm_i915_gem_object *obj = ring->private; -u32 cs_offset = obj->gtt_offset; +u32 cs_offset =i915_gem_obj_ggtt_offset (obj);if (len >I830_BATCH_LIMIT )return -ENOSPC ;
@@ -1214,7 +1214,7 @@ static int init_status_page(struct intel_ring_buffer *ring) goto err_unref; } - ring->status_page.gfx_addr = obj->gtt_offset; + ring->status_page.gfx_addr =i915_gem_obj_ggtt_offset (obj); ring->status_page.page_addr =kmap (sg_page (obj->pages->sgl));if (ring->status_page.page_addr ==NULL ) { ret = -ENOMEM ;
@@ -1308,7 +1308,7 @@ static int intel_init_ring_buffer(struct drm_device *dev, goto err_unpin; ring->virtual_start = -ioremap_wc (dev_priv->gtt.mappable_base + obj->gtt_offset, +ioremap_wc (dev_priv->gtt.mappable_base +i915_gem_obj_ggtt_offset (obj), ring->size);if (ring->virtual_start ==NULL ) {DRM_ERROR ("Failed to map ringbuffer.\n" );
drivers/gpu/drm/i915/intel_sprite.c +5 -3
@@ -133,7 +133,7 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_framebuffer *fb,I915_WRITE (SPSIZE (pipe, plane), (crtc_h <<16 ) | crtc_w);I915_WRITE (SPCNTR (pipe, plane), sprctl); -I915_MODIFY_DISPBASE (SPSURF (pipe, plane), obj->gtt_offset + +I915_MODIFY_DISPBASE (SPSURF (pipe, plane),i915_gem_obj_ggtt_offset (obj) + sprsurf_offset);POSTING_READ (SPSURF (pipe, plane)); }
@@ -308,7 +308,8 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,if (intel_plane->can_scale)I915_WRITE (SPRSCALE (pipe), sprscale);I915_WRITE (SPRCTL (pipe), sprctl); -I915_MODIFY_DISPBASE (SPRSURF (pipe), obj->gtt_offset + sprsurf_offset); +I915_MODIFY_DISPBASE (SPRSURF (pipe), +i915_gem_obj_ggtt_offset (obj) + sprsurf_offset);POSTING_READ (SPRSURF (pipe));/* potentially re-enable LP watermarks */
@@ -478,7 +479,8 @@ ilk_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,I915_WRITE (DVSSIZE (pipe), (crtc_h <<16 ) | crtc_w);I915_WRITE (DVSSCALE (pipe), dvsscale);I915_WRITE (DVSCNTR (pipe), dvscntr); -I915_MODIFY_DISPBASE (DVSSURF (pipe), obj->gtt_offset + dvssurf_offset); +I915_MODIFY_DISPBASE (DVSSURF (pipe), +i915_gem_obj_ggtt_offset (obj) + dvssurf_offset);POSTING_READ (DVSSURF (pipe)); }
Signature · 1 line
1.8.3.2
Re: [PATCH 2/6] drm/i915: Getter/setter for object attributes
Quoted 16 lines · Ben Widawsky
Soon we want to gut a lot of our existing assumptions how many address spaces an object can live in, and in doing so, embed the
drm_mm_nodein the object (and later the VMA).It's possible in the future we'll want to add more getter/setter methods, but for now this is enough to enable the VMAs.
v2: Reworked commit message (Ben)
Added comments to the main functions (Ben)
sed -i "s/i915_gem_obj_set_color/i915_gem_obj_ggtt_set_color/" drivers/gpu/drm/i915/*.[ch] sed -i "s/i915_gem_obj_bound/i915_gem_obj_ggtt_bound/" drivers/gpu/drm/i915/*.[ch] sed -i "s/i915_gem_obj_size/i915_gem_obj_ggtt_size/" drivers/gpu/drm/i915/*.[ch] sed -i "s/i915_gem_obj_offset/i915_gem_obj_ggtt_offset/" drivers/gpu/drm/i915/*.[ch] (Daniel)Signed-off-by Ben Widawsky <[email protected]>
[snip]
drivers/gpu/drm/i915/i915_drv.h struct drm_i915_gem_object { · line 1230
Quoted 10 lines
drivers/gpu/drm/i915/i915_drv.h +1
-1@@ -1227,7 +1227,7 @@ struct drm_i915_gem_object {const struct drm_i915_gem_object_ops *ops;/** Current space allocated to this object in the GTT, if any. */ -struct drm_mm_node *gtt_space; +struct drm_mm_node *ggtt_space;
Is this ...
drivers/gpu/drm/i915/i915_drv.h struct drm_i915_gem_object { · line 1336
Quoted 9 lines
drivers/gpu/drm/i915/i915_drv.h +1
-1/** Stolen memory for this object, instead of being backed by shmem. */ struct drm_mm_node *stolen;struct list_head global_list;@@ -1333,7 +1333,7 @@ struct drm_i915_gem_object {* * This is the same as gtt_space->start */ -uint32_t gtt_offset; +uint32_t ggtt_offset;
... and this intentional? Feels a bit like needless churn if we move the entire thing into the vma rsn anyway. I can bikeshed while applying if you're ok, need a solid testcase to improve my patch apply scripts anyway ;-)
Cheers, Daniel
Quoted 1,037 lines
drivers/gpu/drm/i915/i915_drv.h +31
-0struct intel_ring_buffer *ring;@@ -1360,6 +1360,37 @@ struct drm_i915_gem_object {#define to_intel_bo (x) container_of(x, struct drm_i915_gem_object, base) +/* Offset of the first PTE pointing to this object */ +static inline unsigned long +i915_gem_obj_ggtt_offset (struct drm_i915_gem_object *o) +{ +return o->ggtt_space->start; +} + +/* Whether or not this object is currently mapped by the translation tables */ +static inline bool +i915_gem_obj_ggtt_bound (struct drm_i915_gem_object *o) +{ +return o->ggtt_space !=NULL ; +} + +/* The size used in the translation tables may be larger than the actual size of +* the object on GEN2/GEN3 because of the way tiling is handled. See +* i915_gem_get_gtt_size() for more details. +*/ +static inline unsigned long +i915_gem_obj_ggtt_size (struct drm_i915_gem_object *o) +{ +return o->ggtt_space->size; +} + +static inline void +i915_gem_obj_ggtt_set_color (struct drm_i915_gem_object *o, +enum i915_cache_level color) +{ + o->ggtt_space->color = color; +} +/** * Request queue structure. * drivers/gpu/drm/i915/i915_gem.c +51
-50@@ -135,7 +135,7 @@ int i915_mutex_lock_interruptible(struct drm_device *dev)static inline bool i915_gem_object_is_inactive (struct drm_i915_gem_object *obj) { -return obj->gtt_space && !obj->active; +return i915_gem_obj_ggtt_bound (obj) && !obj->active; }int @@ -178,7 +178,7 @@ i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,mutex_lock (&dev->struct_mutex);list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list)if (obj->pin_count) - pinned += obj->gtt_space->size; + pinned +=i915_gem_obj_ggtt_size (obj);mutex_unlock (&dev->struct_mutex); args->aper_size = dev_priv->gtt.total;@@ -422,7 +422,7 @@ i915_gem_shmem_pread(struct drm_device *dev,* anyway again before the next pread happens. */ if (obj->cache_level ==I915_CACHE_NONE ) needs_clflush =1 ; -if (obj->gtt_space) { +if (i915_gem_obj_ggtt_bound (obj)) { ret =i915_gem_object_set_to_gtt_domain (obj, false);if (ret)return ret;@@ -609,7 +609,7 @@ i915_gem_gtt_pwrite_fast(struct drm_device *dev, user_data =to_user_ptr (args->data_ptr); remain = args->size; - offset = obj->gtt_offset + args->offset; + offset =i915_gem_obj_ggtt_offset (obj) + args->offset;while (remain >0 ) {/* Operation in this page @@ -739,7 +739,7 @@ i915_gem_shmem_pwrite(struct drm_device *dev,* right away and we therefore have to clflush anyway. */ if (obj->cache_level ==I915_CACHE_NONE ) needs_clflush_after =1 ; -if (obj->gtt_space) { +if (i915_gem_obj_ggtt_bound (obj)) { ret =i915_gem_object_set_to_gtt_domain (obj, true);if (ret)return ret;@@ -1360,8 +1360,9 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) obj->fault_mappable = true; - pfn = ((dev_priv->gtt.mappable_base + obj->gtt_offset) >>PAGE_SHIFT ) + - page_offset; + pfn = dev_priv->gtt.mappable_base +i915_gem_obj_ggtt_offset (obj); + pfn >>=PAGE_SHIFT ; + pfn += page_offset;/* Finally, remap it using the new GTT offset */ ret =vm_insert_pfn (vma, (unsigned long )vmf->virtual_address, pfn);@@ -1667,7 +1668,7 @@ i915_gem_object_put_pages(struct drm_i915_gem_object *obj)if (obj->pages ==NULL )return 0 ; -BUG_ON (obj->gtt_space); +BUG_ON (i915_gem_obj_ggtt_bound (obj));if (obj->pages_pin_count)return -EBUSY ;@@ -2117,8 +2118,8 @@ i915_gem_request_remove_from_client(struct drm_i915_gem_request *request)static bool i915_head_inside_object (u32 acthd,struct drm_i915_gem_object *obj) { -if (acthd >= obj->gtt_offset && - acthd < obj->gtt_offset + obj->base.size) +if (acthd >=i915_gem_obj_ggtt_offset (obj) && + acthd <i915_gem_obj_ggtt_offset (obj) + obj->base.size)return true;return false;@@ -2176,11 +2177,11 @@ static void i915_set_reset_status(struct intel_ring_buffer *ring,if (ring->hangcheck.action != wait &&i915_request_guilty (request, acthd, &inside)) { -DRM_ERROR ("%s hung %s bo (0x%x ctx %d) at 0x%x\n" , +DRM_ERROR ("%s hung %s bo (0x%lx ctx %d) at 0x%x\n" , ring->name, inside ?"inside" :"flushing" , request->batch_obj ? - request->batch_obj->gtt_offset :0 , +i915_gem_obj_ggtt_offset (request->batch_obj) :0 , request->ctx ? request->ctx->id :0 , acthd);@@ -2581,7 +2582,7 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj)drm_i915_private_t *dev_priv = obj->base.dev->dev_private;int ret; -if (obj->gtt_space ==NULL ) +if (!i915_gem_obj_ggtt_bound (obj))return 0 ;if (obj->pin_count)@@ -2620,9 +2621,9 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj)/* Avoid an unnecessary call to unbind on rebind. */ obj->map_and_fenceable = true; -drm_mm_put_block (obj->gtt_space); - obj->gtt_space =NULL ; - obj->gtt_offset =0 ; +drm_mm_put_block (obj->ggtt_space); + obj->ggtt_space =NULL ; + obj->ggtt_offset =0 ;return 0 ; }@@ -2664,11 +2665,11 @@ static void i965_write_fence_reg(struct drm_device *dev, int reg, }if (obj) { -u32 size = obj->gtt_space->size; +u32 size =i915_gem_obj_ggtt_size (obj); - val = (uint64_t )((obj->gtt_offset + size -4096 ) & + val = (uint64_t )((i915_gem_obj_ggtt_offset (obj) + size -4096 ) &0xfffff000 ) <<32 ; - val |= obj->gtt_offset &0xfffff000 ; + val |=i915_gem_obj_ggtt_offset (obj) &0xfffff000 ; val |= (uint64_t )((obj->stride /128 ) -1 ) << fence_pitch_shift;if (obj->tiling_mode ==I915_TILING_Y ) val |=1 <<I965_FENCE_TILING_Y_SHIFT ;@@ -2688,15 +2689,15 @@ static void i915_write_fence_reg(struct drm_device *dev, int reg,u32 val;if (obj) { -u32 size = obj->gtt_space->size; +u32 size =i915_gem_obj_ggtt_size (obj);int pitch_val;int tile_width; -WARN ((obj->gtt_offset & ~I915_FENCE_START_MASK ) || +WARN ((i915_gem_obj_ggtt_offset (obj) & ~I915_FENCE_START_MASK ) || (size & -size) != size || - (obj->gtt_offset & (size -1 )), -"object 0x%08x [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n" , - obj->gtt_offset, obj->map_and_fenceable, size); + (i915_gem_obj_ggtt_offset (obj) & (size -1 )), +"object 0x%08lx [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n" , +i915_gem_obj_ggtt_offset (obj), obj->map_and_fenceable, size);if (obj->tiling_mode ==I915_TILING_Y &&HAS_128_BYTE_Y_TILING (dev)) tile_width =128 ;@@ -2707,7 +2708,7 @@ static void i915_write_fence_reg(struct drm_device *dev, int reg, pitch_val = obj->stride / tile_width; pitch_val =ffs (pitch_val) -1 ; - val = obj->gtt_offset; + val =i915_gem_obj_ggtt_offset (obj);if (obj->tiling_mode ==I915_TILING_Y ) val |=1 <<I830_FENCE_TILING_Y_SHIFT ; val |=I915_FENCE_SIZE_BITS (size);@@ -2732,19 +2733,19 @@ static void i830_write_fence_reg(struct drm_device *dev, int reg,uint32_t val;if (obj) { -u32 size = obj->gtt_space->size; +u32 size =i915_gem_obj_ggtt_size (obj);uint32_t pitch_val; -WARN ((obj->gtt_offset & ~I830_FENCE_START_MASK ) || +WARN ((i915_gem_obj_ggtt_offset (obj) & ~I830_FENCE_START_MASK ) || (size & -size) != size || - (obj->gtt_offset & (size -1 )), -"object 0x%08x not 512K or pot-size 0x%08x aligned\n" , - obj->gtt_offset, size); + (i915_gem_obj_ggtt_offset (obj) & (size -1 )), +"object 0x%08lx not 512K or pot-size 0x%08x aligned\n" , +i915_gem_obj_ggtt_offset (obj), size); pitch_val = obj->stride /128 ; pitch_val =ffs (pitch_val) -1 ; - val = obj->gtt_offset; + val =i915_gem_obj_ggtt_offset (obj);if (obj->tiling_mode ==I915_TILING_Y ) val |=1 <<I830_FENCE_TILING_Y_SHIFT ; val |=I830_FENCE_SIZE_BITS (size);@@ -3033,8 +3034,8 @@ static void i915_gem_verify_gtt(struct drm_device *dev)if (obj->cache_level != obj->gtt_space->color) {printk (KERN_ERR "object reserved space [%08lx, %08lx] with wrong color, cache_level=%x, color=%lx\n" , - obj->gtt_space->start, - obj->gtt_space->start + obj->gtt_space->size, +i915_gem_obj_ggtt_offset (obj), +i915_gem_obj_ggtt_offset (obj) +i915_gem_obj_ggtt_size (obj), obj->cache_level, obj->gtt_space->color); err++;@@ -3045,8 +3046,8 @@ static void i915_gem_verify_gtt(struct drm_device *dev) obj->gtt_space, obj->cache_level)) {printk (KERN_ERR "invalid GTT space found at [%08lx, %08lx] - color=%x\n" , - obj->gtt_space->start, - obj->gtt_space->start + obj->gtt_space->size, +i915_gem_obj_ggtt_offset (obj), +i915_gem_obj_ggtt_offset (obj) +i915_gem_obj_ggtt_size (obj), obj->cache_level); err++;continue ;@@ -3151,15 +3152,15 @@ search_free:list_move_tail (&obj->global_list, &dev_priv->mm.bound_list);list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list); - obj->gtt_space = node; - obj->gtt_offset = node->start; + obj->ggtt_space = node; + obj->ggtt_offset = node->start; fenceable = node->size == fence_size && (node->start & (fence_alignment -1 )) ==0 ; - mappable = - obj->gtt_offset + obj->base.size <= dev_priv->gtt.mappable_end; + mappable =i915_gem_obj_ggtt_offset (obj) + obj->base.size <= + dev_priv->gtt.mappable_end; obj->map_and_fenceable = mappable && fenceable;@@ -3261,7 +3262,7 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)int ret;/* Not valid to be called on unbound objects. */ -if (obj->gtt_space ==NULL ) +if (!i915_gem_obj_ggtt_bound (obj))return -EINVAL ;if (obj->base.write_domain ==I915_GEM_DOMAIN_GTT )@@ -3320,13 +3321,13 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,return -EBUSY ; } -if (!i915_gem_valid_gtt_space (dev, obj->gtt_space, cache_level)) { +if (!i915_gem_valid_gtt_space (dev, obj->ggtt_space, cache_level)) { ret =i915_gem_object_unbind (obj);if (ret)return ret; } -if (obj->gtt_space) { +if (i915_gem_obj_ggtt_bound (obj)) { ret =i915_gem_object_finish_gpu (obj);if (ret)return ret;@@ -3349,7 +3350,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,i915_ppgtt_bind_object (dev_priv->mm.aliasing_ppgtt, obj, cache_level); - obj->gtt_space->color = cache_level; +i915_gem_obj_ggtt_set_color (obj, cache_level); }if (cache_level ==I915_CACHE_NONE ) {@@ -3630,14 +3631,14 @@ i915_gem_object_pin(struct drm_i915_gem_object *obj,if (WARN_ON (obj->pin_count ==DRM_I915_GEM_OBJECT_MAX_PIN_COUNT ))return -EBUSY ; -if (obj->gtt_space !=NULL ) { -if ((alignment && obj->gtt_offset & (alignment -1 )) || +if (i915_gem_obj_ggtt_bound (obj)) { +if ((alignment &&i915_gem_obj_ggtt_offset (obj) & (alignment -1 )) || (map_and_fenceable && !obj->map_and_fenceable)) {WARN (obj->pin_count,"bo is already pinned with incorrect alignment:" -" offset=%x, req.alignment=%x, req.map_and_fenceable=%d," +" offset=%lx, req.alignment=%x, req.map_and_fenceable=%d," " obj->map_and_fenceable=%d\n" , - obj->gtt_offset, alignment, +i915_gem_obj_ggtt_offset (obj), alignment, map_and_fenceable, obj->map_and_fenceable); ret =i915_gem_object_unbind (obj);@@ -3646,7 +3647,7 @@ i915_gem_object_pin(struct drm_i915_gem_object *obj, } } -if (obj->gtt_space ==NULL ) { +if (!i915_gem_obj_ggtt_bound (obj)) {struct drm_i915_private *dev_priv = obj->base.dev->dev_private; ret = i915_gem_object_bind_to_gtt(obj, alignment,@@ -3672,7 +3673,7 @@ void i915_gem_object_unpin(struct drm_i915_gem_object *obj) {BUG_ON (obj->pin_count ==0 ); -BUG_ON (obj->gtt_space ==NULL ); +BUG_ON (!i915_gem_obj_ggtt_bound (obj));if (--obj->pin_count ==0 ) obj->pin_mappable = false;@@ -3722,7 +3723,7 @@ i915_gem_pin_ioctl(struct drm_device *dev, void *data,* as the X server doesn't manage domains yet */ i915_gem_object_flush_cpu_write_domain (obj); - args->offset = obj->gtt_offset; + args->offset =i915_gem_obj_ggtt_offset (obj); out:drm_gem_object_unreference (&obj->base); unlock:drivers/gpu/drm/i915/i915_gem_context.c +1
-1@@ -377,7 +377,7 @@ mi_set_context(struct intel_ring_buffer *ring,intel_ring_emit (ring,MI_NOOP );intel_ring_emit (ring,MI_SET_CONTEXT ); - intel_ring_emit(ring, new_context->obj->gtt_offset | + intel_ring_emit(ring, i915_gem_obj_ggtt_offset(new_context->obj) | MI_MM_SPACE_GTT | MI_SAVE_EXT_STATE_EN | MI_RESTORE_EXT_STATE_EN |drivers/gpu/drm/i915/i915_gem_evict.c +3
-3@@ -38,7 +38,7 @@ mark_free(struct drm_i915_gem_object *obj, struct list_head *unwind)return false;list_add (&obj->exec_list, unwind); -return drm_mm_scan_add_block (obj->gtt_space); +return drm_mm_scan_add_block (obj->ggtt_space); }int @@ -107,7 +107,7 @@ none: struct drm_i915_gem_object, exec_list); - ret =drm_mm_scan_remove_block (obj->gtt_space); + ret =drm_mm_scan_remove_block (obj->ggtt_space);BUG_ON (ret);list_del_init (&obj->exec_list);@@ -127,7 +127,7 @@ found: obj =list_first_entry (&unwind_list, struct drm_i915_gem_object, exec_list); -if (drm_mm_scan_remove_block (obj->gtt_space)) { +if (drm_mm_scan_remove_block (obj->ggtt_space)) {list_move (&obj->exec_list, &eviction_list);drm_gem_object_reference (&obj->base);continue ;drivers/gpu/drm/i915/i915_gem_execbuffer.c +10
-9@@ -188,7 +188,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,return -ENOENT ; target_i915_obj =to_intel_bo (target_obj); - target_offset = target_i915_obj->gtt_offset; + target_offset =i915_gem_obj_ggtt_offset (target_i915_obj);/* Sandybridge PPGTT errata: We need a global gtt mapping for MI and * pipe_control writes because the gpu doesn't properly redirect them @@ -280,7 +280,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,return ret;/* Map the page containing the relocation we're going to perform. */ - reloc->offset += obj->gtt_offset; + reloc->offset +=i915_gem_obj_ggtt_offset (obj); reloc_page =io_mapping_map_atomic_wc (dev_priv->gtt.mappable, reloc->offset &PAGE_MASK ); reloc_entry = (uint32_t __iomem *)@@ -436,8 +436,8 @@ i915_gem_execbuffer_reserve_object(struct drm_i915_gem_object *obj, obj->has_aliasing_ppgtt_mapping =1 ; } -if (entry->offset != obj->gtt_offset) { - entry->offset = obj->gtt_offset; +if (entry->offset !=i915_gem_obj_ggtt_offset (obj)) { + entry->offset =i915_gem_obj_ggtt_offset (obj); *need_reloc = true; }@@ -458,7 +458,7 @@ i915_gem_execbuffer_unreserve_object(struct drm_i915_gem_object *obj) {struct drm_i915_gem_exec_object2 *entry; -if (!obj->gtt_space) +if (!i915_gem_obj_ggtt_bound (obj))return ; entry = obj->exec_entry;@@ -530,7 +530,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,struct drm_i915_gem_exec_object2 *entry = obj->exec_entry;bool need_fence, need_mappable; -if (!obj->gtt_space) +if (!i915_gem_obj_ggtt_bound (obj))continue ; need_fence =@@ -539,7 +539,8 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring, obj->tiling_mode !=I915_TILING_NONE ; need_mappable = need_fence ||need_reloc_mappable (obj); -if ((entry->alignment && obj->gtt_offset & (entry->alignment -1 )) || +if ((entry->alignment && +i915_gem_obj_ggtt_offset (obj) & (entry->alignment -1 )) || (need_mappable && !obj->map_and_fenceable)) ret =i915_gem_object_unbind (obj);else @@ -550,7 +551,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,/* Bind fresh objects */ list_for_each_entry (obj, objects, exec_list) { -if (obj->gtt_space) +if (i915_gem_obj_ggtt_bound (obj))continue ; ret =i915_gem_execbuffer_reserve_object (obj, ring, need_relocs);@@ -1058,7 +1059,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, goto err; } - exec_start = batch_obj->gtt_offset + args->batch_start_offset; + exec_start =i915_gem_obj_ggtt_offset (batch_obj) + args->batch_start_offset; exec_len = args->batch_len;if (cliprects) {for (i =0 ; i < args->num_cliprects; i++) {drivers/gpu/drm/i915/i915_gem_gtt.c +12
-12@@ -378,7 +378,7 @@ void i915_ppgtt_bind_object(struct i915_hw_ppgtt *ppgtt, enum i915_cache_level cache_level) { ppgtt->insert_entries (ppgtt, obj->pages, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , cache_level); }@@ -386,7 +386,7 @@ void i915_ppgtt_unbind_object(struct i915_hw_ppgtt *ppgtt, struct drm_i915_gem_object *obj) { ppgtt->clear_range (ppgtt, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , obj->base.size >>PAGE_SHIFT ); }@@ -551,7 +551,7 @@ void i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj,struct drm_i915_private *dev_priv = dev->dev_private; dev_priv->gtt.gtt_insert_entries (dev, obj->pages, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , cache_level); obj->has_global_gtt_mapping =1 ;@@ -563,7 +563,7 @@ void i915_gem_gtt_unbind_object(struct drm_i915_gem_object *obj)struct drm_i915_private *dev_priv = dev->dev_private; dev_priv->gtt.gtt_clear_range (obj->base.dev, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , obj->base.size >>PAGE_SHIFT ); obj->has_global_gtt_mapping =0 ;@@ -630,22 +630,22 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,/* Mark any preallocated objects as occupied */ list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) {int ret; -DRM_DEBUG_KMS ("reserving preallocated space: %x + %zx\n" , - obj->gtt_offset, obj->base.size); +DRM_DEBUG_KMS ("reserving preallocated space: %lx + %zx\n" , +i915_gem_obj_ggtt_offset (obj), obj->base.size); -BUG_ON (obj->gtt_space !=I915_GTT_RESERVED ); - obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); -if (!obj->gtt_space) { +BUG_ON (obj->ggtt_space !=I915_GTT_RESERVED ); + obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL ); +if (!obj->ggtt_space) {DRM_ERROR ("Failed to preserve all objects\n" );break ; } ret =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_space, - obj->gtt_offset, + obj->ggtt_space, +i915_gem_obj_ggtt_offset (obj), obj->base.size);if (ret) {DRM_DEBUG_KMS ("Reservation failed\n" ); -kfree (obj->gtt_space); +kfree (obj->ggtt_space); } obj->has_global_gtt_mapping =1 ; }drivers/gpu/drm/i915/i915_gem_stolen.c +5
-5@@ -374,23 +374,23 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); -if (!obj->gtt_space) { + obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL ); +if (!obj->ggtt_space) {DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); goto unref_out; } ret =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_space, + obj->ggtt_space, gtt_offset, size);if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); goto unref_out; } }else - obj->gtt_space =I915_GTT_RESERVED ; + obj->ggtt_space =I915_GTT_RESERVED ; - obj->gtt_offset = gtt_offset; + obj->ggtt_offset = gtt_offset; obj->has_global_gtt_mapping =1 ;list_add_tail (&obj->global_list, &dev_priv->mm.bound_list);drivers/gpu/drm/i915/i915_gem_tiling.c +7
-7@@ -268,18 +268,18 @@ i915_gem_object_fence_ok(struct drm_i915_gem_object *obj, int tiling_mode)return true;if (INTEL_INFO (obj->base.dev)->gen ==3 ) { -if (obj->gtt_offset & ~I915_FENCE_START_MASK ) +if (i915_gem_obj_ggtt_offset (obj) & ~I915_FENCE_START_MASK )return false; }else { -if (obj->gtt_offset & ~I830_FENCE_START_MASK ) +if (i915_gem_obj_ggtt_offset (obj) & ~I830_FENCE_START_MASK )return false; } size =i915_gem_get_gtt_size (obj->base.dev, obj->base.size, tiling_mode); -if (obj->gtt_space->size != size) +if (i915_gem_obj_ggtt_size (obj) != size)return false; -if (obj->gtt_offset & (size -1 )) +if (i915_gem_obj_ggtt_offset (obj) & (size -1 ))return false;return true;@@ -359,8 +359,8 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,*/ obj->map_and_fenceable = - obj->gtt_space ==NULL || - (obj->gtt_offset + obj->base.size <= dev_priv->gtt.mappable_end && + !i915_gem_obj_ggtt_bound (obj) || + (i915_gem_obj_ggtt_offset (obj) + obj->base.size <= dev_priv->gtt.mappable_end &&i915_gem_object_fence_ok (obj, args->tiling_mode));/* Rebind if we need a change of alignment */ @@ -369,7 +369,7 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,i915_gem_get_gtt_alignment (dev, obj->base.size, args->tiling_mode, false); -if (obj->gtt_offset & (unfenced_alignment -1 )) +if (i915_gem_obj_ggtt_offset (obj) & (unfenced_alignment -1 )) ret =i915_gem_object_unbind (obj); }drivers/gpu/drm/i915/i915_irq.c +7
-8@@ -1516,7 +1516,7 @@ i915_error_object_create_sized(struct drm_i915_private *dev_priv,if (dst ==NULL )return NULL ; - reloc_offset = src->gtt_offset; + reloc_offset = dst->gtt_offset =i915_gem_obj_ggtt_offset (src);for (i =0 ; i < num_pages; i++) {unsigned long flags;void *d;@@ -1568,7 +1568,6 @@ i915_error_object_create_sized(struct drm_i915_private *dev_priv, reloc_offset +=PAGE_SIZE ; } dst->page_count = num_pages; - dst->gtt_offset = src->gtt_offset;return dst;@@ -1622,7 +1621,7 @@ static void capture_bo(struct drm_i915_error_buffer *err, err->name = obj->base.name; err->rseqno = obj->last_read_seqno; err->wseqno = obj->last_write_seqno; - err->gtt_offset = obj->gtt_offset; + err->gtt_offset =i915_gem_obj_ggtt_offset (obj); err->read_domains = obj->base.read_domains; err->write_domain = obj->base.write_domain; err->fence_reg = obj->fence_reg;@@ -1720,8 +1719,8 @@ i915_error_first_batchbuffer(struct drm_i915_private *dev_priv,return NULL ; obj = ring->private; -if (acthd >= obj->gtt_offset && - acthd < obj->gtt_offset + obj->base.size) +if (acthd >=i915_gem_obj_ggtt_offset (obj) && + acthd <i915_gem_obj_ggtt_offset (obj) + obj->base.size)return i915_error_object_create (dev_priv, obj); }@@ -1802,7 +1801,7 @@ static void i915_gem_record_active_context(struct intel_ring_buffer *ring,return ;list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) { -if ((error->ccid &PAGE_MASK ) == obj->gtt_offset) { +if ((error->ccid &PAGE_MASK ) ==i915_gem_obj_ggtt_offset (obj)) { ering->ctx =i915_error_object_create_sized (dev_priv, obj,1 );break ;@@ -2156,10 +2155,10 @@ static void __always_unused i915_pageflip_stall_check(struct drm_device *dev, inif (INTEL_INFO (dev)->gen >=4 ) {int dspsurf =DSPSURF (intel_crtc->plane); stall_detected =I915_HI_DISPBASE (I915_READ (dspsurf)) == - obj->gtt_offset; +i915_gem_obj_ggtt_offset (obj); }else {int dspaddr =DSPADDR (intel_crtc->plane); - stall_detected =I915_READ (dspaddr) == (obj->gtt_offset + + stall_detected =I915_READ (dspaddr) == (i915_gem_obj_ggtt_offset (obj) + crtc->y * crtc->fb->pitches[0 ] + crtc->x * crtc->fb->bits_per_pixel/8 ); }drivers/gpu/drm/i915/i915_trace.h +4
-4@@ -46,8 +46,8 @@ TRACE_EVENT(i915_gem_object_bind,TP_fast_assign ( __entry->obj = obj; - __entry->offset = obj->gtt_space->start; - __entry->size = obj->gtt_space->size; + __entry->offset =i915_gem_obj_ggtt_offset (obj); + __entry->size =i915_gem_obj_ggtt_size (obj); __entry->mappable = mappable; ),@@ -68,8 +68,8 @@ TRACE_EVENT(i915_gem_object_unbind,TP_fast_assign ( __entry->obj = obj; - __entry->offset = obj->gtt_space->start; - __entry->size = obj->gtt_space->size; + __entry->offset =i915_gem_obj_ggtt_offset (obj); + __entry->size =i915_gem_obj_ggtt_size (obj); ), TP_printk("obj=%p, offset=%08x size=%x",drivers/gpu/drm/i915/intel_display.c +15
-13@@ -1980,16 +1980,17 @@ static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb, intel_crtc->dspaddr_offset = linear_offset; } -DRM_DEBUG_KMS ("Writing base %08X %08lX %d %d %d\n" , - obj->gtt_offset, linear_offset, x, y, fb->pitches[0 ]); +DRM_DEBUG_KMS ("Writing base %08lX %08lX %d %d %d\n" , +i915_gem_obj_ggtt_offset (obj), linear_offset, x, y, + fb->pitches[0 ]);I915_WRITE (DSPSTRIDE (plane), fb->pitches[0 ]);if (INTEL_INFO (dev)->gen >=4 ) {I915_MODIFY_DISPBASE (DSPSURF (plane), - obj->gtt_offset + intel_crtc->dspaddr_offset); +i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);I915_WRITE (DSPTILEOFF (plane), (y <<16 ) | x);I915_WRITE (DSPLINOFF (plane), linear_offset); }else -I915_WRITE (DSPADDR (plane), obj->gtt_offset + linear_offset); +I915_WRITE (DSPADDR (plane),i915_gem_obj_ggtt_offset (obj) + linear_offset);POSTING_READ (reg);return 0 ;@@ -2069,11 +2070,12 @@ static int ironlake_update_plane(struct drm_crtc *crtc, fb->pitches[0]); linear_offset -= intel_crtc->dspaddr_offset; -DRM_DEBUG_KMS ("Writing base %08X %08lX %d %d %d\n" , - obj->gtt_offset, linear_offset, x, y, fb->pitches[0 ]); +DRM_DEBUG_KMS ("Writing base %08lX %08lX %d %d %d\n" , +i915_gem_obj_ggtt_offset (obj), linear_offset, x, y, + fb->pitches[0 ]);I915_WRITE (DSPSTRIDE (plane), fb->pitches[0 ]);I915_MODIFY_DISPBASE (DSPSURF (plane), - obj->gtt_offset + intel_crtc->dspaddr_offset); +i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);if (IS_HASWELL (dev)) {I915_WRITE (DSPOFFSET (plane), (y <<16 ) | x); }else {@@ -6566,7 +6568,7 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc, goto fail_unpin; } - addr = obj->gtt_offset; + addr =i915_gem_obj_ggtt_offset (obj); }else {int align =IS_I830 (dev) ?16 *1024 :256 ; ret = i915_gem_attach_phys_object(dev, obj,@@ -7338,7 +7340,7 @@ static int intel_gen2_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP |MI_DISPLAY_FLIP_PLANE (intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ]); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);intel_ring_emit (ring,0 );/* aux display base address, unused */ intel_mark_page_flip_active (intel_crtc);@@ -7379,7 +7381,7 @@ static int intel_gen3_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP_I915 |MI_DISPLAY_FLIP_PLANE (intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ]); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);intel_ring_emit (ring,MI_NOOP );intel_mark_page_flip_active (intel_crtc);@@ -7419,7 +7421,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev, MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ]);intel_ring_emit (ring, - (obj->gtt_offset + intel_crtc->dspaddr_offset) | + (i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset) | obj->tiling_mode);/* XXX Enabling the panel-fitter across page-flip is so far @@ -7462,7 +7464,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP |MI_DISPLAY_FLIP_PLANE (intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ] | obj->tiling_mode); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);/* Contrary to the suggestions in the documentation, * "Enable Panel Fitter" does not seem to be required when page @@ -7527,7 +7529,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP_I915 | plane_bit);intel_ring_emit (ring, (fb->pitches[0 ] | obj->tiling_mode)); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);intel_ring_emit (ring, (MI_NOOP ));intel_mark_page_flip_active (intel_crtc);drivers/gpu/drm/i915/intel_fb.c +4
-4@@ -139,11 +139,11 @@ static int intelfb_create(struct drm_fb_helper *helper, info->apertures->ranges[0 ].base = dev->mode_config.fb_base; info->apertures->ranges[0 ].size = dev_priv->gtt.mappable_end; - info->fix.smem_start = dev->mode_config.fb_base + obj->gtt_offset; + info->fix.smem_start = dev->mode_config.fb_base +i915_gem_obj_ggtt_offset (obj); info->fix.smem_len = size; info->screen_base = -ioremap_wc (dev_priv->gtt.mappable_base + obj->gtt_offset, +ioremap_wc (dev_priv->gtt.mappable_base +i915_gem_obj_ggtt_offset (obj), size);if (!info->screen_base) { ret = -ENOSPC ;@@ -166,9 +166,9 @@ static int intelfb_create(struct drm_fb_helper *helper,/* Use default scratch pixmap (info->pixmap.flags = FB_PIXMAP_SYSTEM) */ -DRM_DEBUG_KMS ("allocated %dx%d fb: 0x%08x, bo %p\n" , +DRM_DEBUG_KMS ("allocated %dx%d fb: 0x%08lx, bo %p\n" , fb->width, fb->height, - obj->gtt_offset, obj); +i915_gem_obj_ggtt_offset (obj), obj);mutex_unlock (&dev->struct_mutex);drivers/gpu/drm/i915/intel_overlay.c +7
-7@@ -196,7 +196,7 @@ intel_overlay_map_regs(struct intel_overlay *overlay) regs = (struct overlay_registers __iomem *)overlay->reg_bo->phys_obj->handle->vaddr; else regs =io_mapping_map_wc (dev_priv->gtt.mappable, - overlay->reg_bo->gtt_offset); +i915_gem_obj_ggtt_offset (overlay->reg_bo));return regs; }@@ -740,7 +740,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay, swidth = params->src_w; swidthsw =calc_swidthsw (overlay->dev, params->offset_Y, tmp_width); sheight = params->src_h; -iowrite32 (new_bo->gtt_offset + params->offset_Y, ®s->OBUF_0Y); +iowrite32 (i915_gem_obj_ggtt_offset (new_bo) + params->offset_Y, ®s->OBUF_0Y); ostride = params->stride_Y;if (params->format &I915_OVERLAY_YUV_PLANAR ) {@@ -754,8 +754,8 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay, params->src_w/uv_hscale); swidthsw |=max_t (u32, tmp_U, tmp_V) <<16 ; sheight |= (params->src_h/uv_vscale) <<16 ; -iowrite32 (new_bo->gtt_offset + params->offset_U, ®s->OBUF_0U); -iowrite32 (new_bo->gtt_offset + params->offset_V, ®s->OBUF_0V); +iowrite32 (i915_gem_obj_ggtt_offset (new_bo) + params->offset_U, ®s->OBUF_0U); +iowrite32 (i915_gem_obj_ggtt_offset (new_bo) + params->offset_V, ®s->OBUF_0V); ostride |= params->stride_UV <<16 ; }@@ -1355,7 +1355,7 @@ void intel_setup_overlay(struct drm_device *dev)DRM_ERROR ("failed to pin overlay register bo\n" ); goto out_free_bo; } - overlay->flip_addr = reg_bo->gtt_offset; + overlay->flip_addr =i915_gem_obj_ggtt_offset (reg_bo); ret =i915_gem_object_set_to_gtt_domain (reg_bo, true);if (ret) {@@ -1435,7 +1435,7 @@ intel_overlay_map_regs_atomic(struct intel_overlay *overlay) overlay->reg_bo->phys_obj->handle->vaddr; else regs =io_mapping_map_atomic_wc (dev_priv->gtt.mappable, - overlay->reg_bo->gtt_offset); +i915_gem_obj_ggtt_offset (overlay->reg_bo));return regs; }@@ -1468,7 +1468,7 @@ intel_overlay_capture_error_state(struct drm_device *dev)if (OVERLAY_NEEDS_PHYSICAL (overlay->dev)) error->base = (__force long )overlay->reg_bo->phys_obj->handle->vaddr;else - error->base = overlay->reg_bo->gtt_offset; + error->base =i915_gem_obj_ggtt_offset (overlay->reg_bo); regs =intel_overlay_map_regs_atomic (overlay);if (!regs)drivers/gpu/drm/i915/intel_pm.c +4
-4@@ -218,7 +218,7 @@ static void ironlake_enable_fbc(struct drm_crtc *crtc, unsigned long interval) (stall_watermark <<DPFC_RECOMP_STALL_WM_SHIFT ) | (interval << DPFC_RECOMP_TIMER_COUNT_SHIFT));I915_WRITE (ILK_DPFC_FENCE_YOFF , crtc->y); -I915_WRITE (ILK_FBC_RT_BASE , obj->gtt_offset |ILK_FBC_RT_VALID ); +I915_WRITE (ILK_FBC_RT_BASE ,i915_gem_obj_ggtt_offset (obj) |ILK_FBC_RT_VALID );/* enable it... */ I915_WRITE (ILK_DPFC_CONTROL , dpfc_ctl |DPFC_CTL_EN );@@ -275,7 +275,7 @@ static void gen7_enable_fbc(struct drm_crtc *crtc, unsigned long interval)struct drm_i915_gem_object *obj = intel_fb->obj;struct intel_crtc *intel_crtc =to_intel_crtc (crtc); -I915_WRITE (IVB_FBC_RT_BASE , obj->gtt_offset); +I915_WRITE (IVB_FBC_RT_BASE ,i915_gem_obj_ggtt_offset (obj)); I915_WRITE(ILK_DPFC_CONTROL, DPFC_CTL_EN | DPFC_CTL_LIMIT_1X | IVB_DPFC_CTL_FENCE_EN |@@ -3707,7 +3707,7 @@ static void ironlake_enable_rc6(struct drm_device *dev)intel_ring_emit (ring,MI_SUSPEND_FLUSH |MI_SUSPEND_FLUSH_EN );intel_ring_emit (ring,MI_SET_CONTEXT ); - intel_ring_emit(ring, dev_priv->ips.renderctx->gtt_offset | + intel_ring_emit(ring, i915_gem_obj_ggtt_offset(dev_priv->ips.renderctx) | MI_MM_SPACE_GTT | MI_SAVE_EXT_STATE_EN | MI_RESTORE_EXT_STATE_EN |@@ -3730,7 +3730,7 @@ static void ironlake_enable_rc6(struct drm_device *dev)return ; } -I915_WRITE (PWRCTXA , dev_priv->ips.pwrctx->gtt_offset |PWRCTX_EN ); +I915_WRITE (PWRCTXA ,i915_gem_obj_ggtt_offset (dev_priv->ips.pwrctx) |PWRCTX_EN );I915_WRITE (RSTDBYCTL ,I915_READ (RSTDBYCTL ) & ~RCX_SW_EXIT ); }drivers/gpu/drm/i915/intel_ringbuffer.c +6
-6@@ -424,14 +424,14 @@ static int init_ring_common(struct intel_ring_buffer *ring)* registers with the above sequence (the readback of the HEAD registers * also enforces ordering), otherwise the hw might lose the new ring * register values. */ -I915_WRITE_START (ring, obj->gtt_offset); +I915_WRITE_START (ring,i915_gem_obj_ggtt_offset (obj));I915_WRITE_CTL (ring, ((ring->size -PAGE_SIZE ) &RING_NR_PAGES ) |RING_VALID );/* If the head is still not zero, the ring is dead */ if (wait_for ((I915_READ_CTL (ring) &RING_VALID ) !=0 && -I915_READ_START (ring) == obj->gtt_offset && +I915_READ_START (ring) ==i915_gem_obj_ggtt_offset (obj) && (I915_READ_HEAD (ring) &HEAD_ADDR ) ==0 ,50 )) { DRM_ERROR("%s initialization failed " "ctl %08x head %08x tail %08x start %08x\n",@@ -489,7 +489,7 @@ init_pipe_control(struct intel_ring_buffer *ring)if (ret) goto err_unref; - pc->gtt_offset = obj->gtt_offset; + pc->gtt_offset =i915_gem_obj_ggtt_offset (obj); pc->cpu_page =kmap (sg_page (obj->pages->sgl));if (pc->cpu_page ==NULL ) { ret = -ENOMEM ;@@ -1129,7 +1129,7 @@ i830_dispatch_execbuffer(struct intel_ring_buffer *ring,intel_ring_advance (ring); }else {struct drm_i915_gem_object *obj = ring->private; -u32 cs_offset = obj->gtt_offset; +u32 cs_offset =i915_gem_obj_ggtt_offset (obj);if (len >I830_BATCH_LIMIT )return -ENOSPC ;@@ -1214,7 +1214,7 @@ static int init_status_page(struct intel_ring_buffer *ring) goto err_unref; } - ring->status_page.gfx_addr = obj->gtt_offset; + ring->status_page.gfx_addr =i915_gem_obj_ggtt_offset (obj); ring->status_page.page_addr =kmap (sg_page (obj->pages->sgl));if (ring->status_page.page_addr ==NULL ) { ret = -ENOMEM ;@@ -1308,7 +1308,7 @@ static int intel_init_ring_buffer(struct drm_device *dev, goto err_unpin; ring->virtual_start = -ioremap_wc (dev_priv->gtt.mappable_base + obj->gtt_offset, +ioremap_wc (dev_priv->gtt.mappable_base +i915_gem_obj_ggtt_offset (obj), ring->size);if (ring->virtual_start ==NULL ) {DRM_ERROR ("Failed to map ringbuffer.\n" );drivers/gpu/drm/i915/intel_sprite.c +5
-3@@ -133,7 +133,7 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_framebuffer *fb,I915_WRITE (SPSIZE (pipe, plane), (crtc_h <<16 ) | crtc_w);I915_WRITE (SPCNTR (pipe, plane), sprctl); -I915_MODIFY_DISPBASE (SPSURF (pipe, plane), obj->gtt_offset + +I915_MODIFY_DISPBASE (SPSURF (pipe, plane),i915_gem_obj_ggtt_offset (obj) + sprsurf_offset);POSTING_READ (SPSURF (pipe, plane)); }@@ -308,7 +308,8 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,if (intel_plane->can_scale)I915_WRITE (SPRSCALE (pipe), sprscale);I915_WRITE (SPRCTL (pipe), sprctl); -I915_MODIFY_DISPBASE (SPRSURF (pipe), obj->gtt_offset + sprsurf_offset); +I915_MODIFY_DISPBASE (SPRSURF (pipe), +i915_gem_obj_ggtt_offset (obj) + sprsurf_offset);POSTING_READ (SPRSURF (pipe));/* potentially re-enable LP watermarks */ @@ -478,7 +479,8 @@ ilk_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,I915_WRITE (DVSSIZE (pipe), (crtc_h <<16 ) | crtc_w);I915_WRITE (DVSSCALE (pipe), dvsscale);I915_WRITE (DVSCNTR (pipe), dvscntr); -I915_MODIFY_DISPBASE (DVSSURF (pipe), obj->gtt_offset + dvssurf_offset); +I915_MODIFY_DISPBASE (DVSSURF (pipe), +i915_gem_obj_ggtt_offset (obj) + dvssurf_offset);POSTING_READ (DVSSURF (pipe)); }Signature · 6 lines
1.8.3.2 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
Re: [PATCH 2/6] drm/i915: Getter/setter for object attributes
Quoted 33 lines · Daniel Vetter
Quoted 16 lines · Ben Widawsky
Soon we want to gut a lot of our existing assumptions how many address spaces an object can live in, and in doing so, embed the
drm_mm_nodein the object (and later the VMA).It's possible in the future we'll want to add more getter/setter methods, but for now this is enough to enable the VMAs.
v2: Reworked commit message (Ben)
Added comments to the main functions (Ben)
sed -i "s/i915_gem_obj_set_color/i915_gem_obj_ggtt_set_color/" drivers/gpu/drm/i915/*.[ch] sed -i "s/i915_gem_obj_bound/i915_gem_obj_ggtt_bound/" drivers/gpu/drm/i915/*.[ch] sed -i "s/i915_gem_obj_size/i915_gem_obj_ggtt_size/" drivers/gpu/drm/i915/*.[ch] sed -i "s/i915_gem_obj_offset/i915_gem_obj_ggtt_offset/" drivers/gpu/drm/i915/*.[ch] (Daniel)Signed-off-by Ben Widawsky <[email protected]>[snip]
drivers/gpu/drm/i915/i915_drv.h struct drm_i915_gem_object { · line 1230
Quoted 10 lines
drivers/gpu/drm/i915/i915_drv.h +1
-1@@ -1227,7 +1227,7 @@ struct drm_i915_gem_object {const struct drm_i915_gem_object_ops *ops;/** Current space allocated to this object in the GTT, if any. */ -struct drm_mm_node *gtt_space; +struct drm_mm_node *ggtt_space;Is this ...
Quoted 15 lines
drivers/gpu/drm/i915/i915_drv.h struct drm_i915_gem_object { · line 1336
Quoted 9 lines
drivers/gpu/drm/i915/i915_drv.h +1
-1/** Stolen memory for this object, instead of being backed by shmem. */ struct drm_mm_node *stolen;struct list_head global_list;@@ -1333,7 +1333,7 @@ struct drm_i915_gem_object {* * This is the same as gtt_space->start */ -uint32_t gtt_offset; +uint32_t ggtt_offset;... and this intentional? Feels a bit like needless churn if we move the entire thing into the vma rsn anyway. I can bikeshed while applying if you're ok, need a solid testcase to improve my patch apply scripts anyway ;-)
Yeah. With everything else being converted to ggtt, I liked naming the node ggtt. As for the offset, I used that to find all the places that needing fixing up, and then decided not to change it back since I killed it in a future patch anyway.
If you want to fix it up on merge, I am fine with that - though I guess after the fact, I do prefer ggtt.
Quoted 1,044 lines
Cheers, Daniel
Quoted 1,037 lines
drivers/gpu/drm/i915/i915_drv.h +31
-0struct intel_ring_buffer *ring;@@ -1360,6 +1360,37 @@ struct drm_i915_gem_object {#define to_intel_bo (x) container_of(x, struct drm_i915_gem_object, base) +/* Offset of the first PTE pointing to this object */ +static inline unsigned long +i915_gem_obj_ggtt_offset (struct drm_i915_gem_object *o) +{ +return o->ggtt_space->start; +} + +/* Whether or not this object is currently mapped by the translation tables */ +static inline bool +i915_gem_obj_ggtt_bound (struct drm_i915_gem_object *o) +{ +return o->ggtt_space !=NULL ; +} + +/* The size used in the translation tables may be larger than the actual size of +* the object on GEN2/GEN3 because of the way tiling is handled. See +* i915_gem_get_gtt_size() for more details. +*/ +static inline unsigned long +i915_gem_obj_ggtt_size (struct drm_i915_gem_object *o) +{ +return o->ggtt_space->size; +} + +static inline void +i915_gem_obj_ggtt_set_color (struct drm_i915_gem_object *o, +enum i915_cache_level color) +{ + o->ggtt_space->color = color; +} +/** * Request queue structure. * drivers/gpu/drm/i915/i915_gem.c +51
-50@@ -135,7 +135,7 @@ int i915_mutex_lock_interruptible(struct drm_device *dev)static inline bool i915_gem_object_is_inactive (struct drm_i915_gem_object *obj) { -return obj->gtt_space && !obj->active; +return i915_gem_obj_ggtt_bound (obj) && !obj->active; }int @@ -178,7 +178,7 @@ i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,mutex_lock (&dev->struct_mutex);list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list)if (obj->pin_count) - pinned += obj->gtt_space->size; + pinned +=i915_gem_obj_ggtt_size (obj);mutex_unlock (&dev->struct_mutex); args->aper_size = dev_priv->gtt.total;@@ -422,7 +422,7 @@ i915_gem_shmem_pread(struct drm_device *dev,* anyway again before the next pread happens. */ if (obj->cache_level ==I915_CACHE_NONE ) needs_clflush =1 ; -if (obj->gtt_space) { +if (i915_gem_obj_ggtt_bound (obj)) { ret =i915_gem_object_set_to_gtt_domain (obj, false);if (ret)return ret;@@ -609,7 +609,7 @@ i915_gem_gtt_pwrite_fast(struct drm_device *dev, user_data =to_user_ptr (args->data_ptr); remain = args->size; - offset = obj->gtt_offset + args->offset; + offset =i915_gem_obj_ggtt_offset (obj) + args->offset;while (remain >0 ) {/* Operation in this page @@ -739,7 +739,7 @@ i915_gem_shmem_pwrite(struct drm_device *dev,* right away and we therefore have to clflush anyway. */ if (obj->cache_level ==I915_CACHE_NONE ) needs_clflush_after =1 ; -if (obj->gtt_space) { +if (i915_gem_obj_ggtt_bound (obj)) { ret =i915_gem_object_set_to_gtt_domain (obj, true);if (ret)return ret;@@ -1360,8 +1360,9 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) obj->fault_mappable = true; - pfn = ((dev_priv->gtt.mappable_base + obj->gtt_offset) >>PAGE_SHIFT ) + - page_offset; + pfn = dev_priv->gtt.mappable_base +i915_gem_obj_ggtt_offset (obj); + pfn >>=PAGE_SHIFT ; + pfn += page_offset;/* Finally, remap it using the new GTT offset */ ret =vm_insert_pfn (vma, (unsigned long )vmf->virtual_address, pfn);@@ -1667,7 +1668,7 @@ i915_gem_object_put_pages(struct drm_i915_gem_object *obj)if (obj->pages ==NULL )return 0 ; -BUG_ON (obj->gtt_space); +BUG_ON (i915_gem_obj_ggtt_bound (obj));if (obj->pages_pin_count)return -EBUSY ;@@ -2117,8 +2118,8 @@ i915_gem_request_remove_from_client(struct drm_i915_gem_request *request)static bool i915_head_inside_object (u32 acthd,struct drm_i915_gem_object *obj) { -if (acthd >= obj->gtt_offset && - acthd < obj->gtt_offset + obj->base.size) +if (acthd >=i915_gem_obj_ggtt_offset (obj) && + acthd <i915_gem_obj_ggtt_offset (obj) + obj->base.size)return true;return false;@@ -2176,11 +2177,11 @@ static void i915_set_reset_status(struct intel_ring_buffer *ring,if (ring->hangcheck.action != wait &&i915_request_guilty (request, acthd, &inside)) { -DRM_ERROR ("%s hung %s bo (0x%x ctx %d) at 0x%x\n" , +DRM_ERROR ("%s hung %s bo (0x%lx ctx %d) at 0x%x\n" , ring->name, inside ?"inside" :"flushing" , request->batch_obj ? - request->batch_obj->gtt_offset :0 , +i915_gem_obj_ggtt_offset (request->batch_obj) :0 , request->ctx ? request->ctx->id :0 , acthd);@@ -2581,7 +2582,7 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj)drm_i915_private_t *dev_priv = obj->base.dev->dev_private;int ret; -if (obj->gtt_space ==NULL ) +if (!i915_gem_obj_ggtt_bound (obj))return 0 ;if (obj->pin_count)@@ -2620,9 +2621,9 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj)/* Avoid an unnecessary call to unbind on rebind. */ obj->map_and_fenceable = true; -drm_mm_put_block (obj->gtt_space); - obj->gtt_space =NULL ; - obj->gtt_offset =0 ; +drm_mm_put_block (obj->ggtt_space); + obj->ggtt_space =NULL ; + obj->ggtt_offset =0 ;return 0 ; }@@ -2664,11 +2665,11 @@ static void i965_write_fence_reg(struct drm_device *dev, int reg, }if (obj) { -u32 size = obj->gtt_space->size; +u32 size =i915_gem_obj_ggtt_size (obj); - val = (uint64_t )((obj->gtt_offset + size -4096 ) & + val = (uint64_t )((i915_gem_obj_ggtt_offset (obj) + size -4096 ) &0xfffff000 ) <<32 ; - val |= obj->gtt_offset &0xfffff000 ; + val |=i915_gem_obj_ggtt_offset (obj) &0xfffff000 ; val |= (uint64_t )((obj->stride /128 ) -1 ) << fence_pitch_shift;if (obj->tiling_mode ==I915_TILING_Y ) val |=1 <<I965_FENCE_TILING_Y_SHIFT ;@@ -2688,15 +2689,15 @@ static void i915_write_fence_reg(struct drm_device *dev, int reg,u32 val;if (obj) { -u32 size = obj->gtt_space->size; +u32 size =i915_gem_obj_ggtt_size (obj);int pitch_val;int tile_width; -WARN ((obj->gtt_offset & ~I915_FENCE_START_MASK ) || +WARN ((i915_gem_obj_ggtt_offset (obj) & ~I915_FENCE_START_MASK ) || (size & -size) != size || - (obj->gtt_offset & (size -1 )), -"object 0x%08x [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n" , - obj->gtt_offset, obj->map_and_fenceable, size); + (i915_gem_obj_ggtt_offset (obj) & (size -1 )), +"object 0x%08lx [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n" , +i915_gem_obj_ggtt_offset (obj), obj->map_and_fenceable, size);if (obj->tiling_mode ==I915_TILING_Y &&HAS_128_BYTE_Y_TILING (dev)) tile_width =128 ;@@ -2707,7 +2708,7 @@ static void i915_write_fence_reg(struct drm_device *dev, int reg, pitch_val = obj->stride / tile_width; pitch_val =ffs (pitch_val) -1 ; - val = obj->gtt_offset; + val =i915_gem_obj_ggtt_offset (obj);if (obj->tiling_mode ==I915_TILING_Y ) val |=1 <<I830_FENCE_TILING_Y_SHIFT ; val |=I915_FENCE_SIZE_BITS (size);@@ -2732,19 +2733,19 @@ static void i830_write_fence_reg(struct drm_device *dev, int reg,uint32_t val;if (obj) { -u32 size = obj->gtt_space->size; +u32 size =i915_gem_obj_ggtt_size (obj);uint32_t pitch_val; -WARN ((obj->gtt_offset & ~I830_FENCE_START_MASK ) || +WARN ((i915_gem_obj_ggtt_offset (obj) & ~I830_FENCE_START_MASK ) || (size & -size) != size || - (obj->gtt_offset & (size -1 )), -"object 0x%08x not 512K or pot-size 0x%08x aligned\n" , - obj->gtt_offset, size); + (i915_gem_obj_ggtt_offset (obj) & (size -1 )), +"object 0x%08lx not 512K or pot-size 0x%08x aligned\n" , +i915_gem_obj_ggtt_offset (obj), size); pitch_val = obj->stride /128 ; pitch_val =ffs (pitch_val) -1 ; - val = obj->gtt_offset; + val =i915_gem_obj_ggtt_offset (obj);if (obj->tiling_mode ==I915_TILING_Y ) val |=1 <<I830_FENCE_TILING_Y_SHIFT ; val |=I830_FENCE_SIZE_BITS (size);@@ -3033,8 +3034,8 @@ static void i915_gem_verify_gtt(struct drm_device *dev)if (obj->cache_level != obj->gtt_space->color) {printk (KERN_ERR "object reserved space [%08lx, %08lx] with wrong color, cache_level=%x, color=%lx\n" , - obj->gtt_space->start, - obj->gtt_space->start + obj->gtt_space->size, +i915_gem_obj_ggtt_offset (obj), +i915_gem_obj_ggtt_offset (obj) +i915_gem_obj_ggtt_size (obj), obj->cache_level, obj->gtt_space->color); err++;@@ -3045,8 +3046,8 @@ static void i915_gem_verify_gtt(struct drm_device *dev) obj->gtt_space, obj->cache_level)) {printk (KERN_ERR "invalid GTT space found at [%08lx, %08lx] - color=%x\n" , - obj->gtt_space->start, - obj->gtt_space->start + obj->gtt_space->size, +i915_gem_obj_ggtt_offset (obj), +i915_gem_obj_ggtt_offset (obj) +i915_gem_obj_ggtt_size (obj), obj->cache_level); err++;continue ;@@ -3151,15 +3152,15 @@ search_free:list_move_tail (&obj->global_list, &dev_priv->mm.bound_list);list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list); - obj->gtt_space = node; - obj->gtt_offset = node->start; + obj->ggtt_space = node; + obj->ggtt_offset = node->start; fenceable = node->size == fence_size && (node->start & (fence_alignment -1 )) ==0 ; - mappable = - obj->gtt_offset + obj->base.size <= dev_priv->gtt.mappable_end; + mappable =i915_gem_obj_ggtt_offset (obj) + obj->base.size <= + dev_priv->gtt.mappable_end; obj->map_and_fenceable = mappable && fenceable;@@ -3261,7 +3262,7 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)int ret;/* Not valid to be called on unbound objects. */ -if (obj->gtt_space ==NULL ) +if (!i915_gem_obj_ggtt_bound (obj))return -EINVAL ;if (obj->base.write_domain ==I915_GEM_DOMAIN_GTT )@@ -3320,13 +3321,13 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,return -EBUSY ; } -if (!i915_gem_valid_gtt_space (dev, obj->gtt_space, cache_level)) { +if (!i915_gem_valid_gtt_space (dev, obj->ggtt_space, cache_level)) { ret =i915_gem_object_unbind (obj);if (ret)return ret; } -if (obj->gtt_space) { +if (i915_gem_obj_ggtt_bound (obj)) { ret =i915_gem_object_finish_gpu (obj);if (ret)return ret;@@ -3349,7 +3350,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,i915_ppgtt_bind_object (dev_priv->mm.aliasing_ppgtt, obj, cache_level); - obj->gtt_space->color = cache_level; +i915_gem_obj_ggtt_set_color (obj, cache_level); }if (cache_level ==I915_CACHE_NONE ) {@@ -3630,14 +3631,14 @@ i915_gem_object_pin(struct drm_i915_gem_object *obj,if (WARN_ON (obj->pin_count ==DRM_I915_GEM_OBJECT_MAX_PIN_COUNT ))return -EBUSY ; -if (obj->gtt_space !=NULL ) { -if ((alignment && obj->gtt_offset & (alignment -1 )) || +if (i915_gem_obj_ggtt_bound (obj)) { +if ((alignment &&i915_gem_obj_ggtt_offset (obj) & (alignment -1 )) || (map_and_fenceable && !obj->map_and_fenceable)) {WARN (obj->pin_count,"bo is already pinned with incorrect alignment:" -" offset=%x, req.alignment=%x, req.map_and_fenceable=%d," +" offset=%lx, req.alignment=%x, req.map_and_fenceable=%d," " obj->map_and_fenceable=%d\n" , - obj->gtt_offset, alignment, +i915_gem_obj_ggtt_offset (obj), alignment, map_and_fenceable, obj->map_and_fenceable); ret =i915_gem_object_unbind (obj);@@ -3646,7 +3647,7 @@ i915_gem_object_pin(struct drm_i915_gem_object *obj, } } -if (obj->gtt_space ==NULL ) { +if (!i915_gem_obj_ggtt_bound (obj)) {struct drm_i915_private *dev_priv = obj->base.dev->dev_private; ret = i915_gem_object_bind_to_gtt(obj, alignment,@@ -3672,7 +3673,7 @@ void i915_gem_object_unpin(struct drm_i915_gem_object *obj) {BUG_ON (obj->pin_count ==0 ); -BUG_ON (obj->gtt_space ==NULL ); +BUG_ON (!i915_gem_obj_ggtt_bound (obj));if (--obj->pin_count ==0 ) obj->pin_mappable = false;@@ -3722,7 +3723,7 @@ i915_gem_pin_ioctl(struct drm_device *dev, void *data,* as the X server doesn't manage domains yet */ i915_gem_object_flush_cpu_write_domain (obj); - args->offset = obj->gtt_offset; + args->offset =i915_gem_obj_ggtt_offset (obj); out:drm_gem_object_unreference (&obj->base); unlock:drivers/gpu/drm/i915/i915_gem_context.c +1
-1@@ -377,7 +377,7 @@ mi_set_context(struct intel_ring_buffer *ring,intel_ring_emit (ring,MI_NOOP );intel_ring_emit (ring,MI_SET_CONTEXT ); - intel_ring_emit(ring, new_context->obj->gtt_offset | + intel_ring_emit(ring, i915_gem_obj_ggtt_offset(new_context->obj) | MI_MM_SPACE_GTT | MI_SAVE_EXT_STATE_EN | MI_RESTORE_EXT_STATE_EN |drivers/gpu/drm/i915/i915_gem_evict.c +3
-3@@ -38,7 +38,7 @@ mark_free(struct drm_i915_gem_object *obj, struct list_head *unwind)return false;list_add (&obj->exec_list, unwind); -return drm_mm_scan_add_block (obj->gtt_space); +return drm_mm_scan_add_block (obj->ggtt_space); }int @@ -107,7 +107,7 @@ none: struct drm_i915_gem_object, exec_list); - ret =drm_mm_scan_remove_block (obj->gtt_space); + ret =drm_mm_scan_remove_block (obj->ggtt_space);BUG_ON (ret);list_del_init (&obj->exec_list);@@ -127,7 +127,7 @@ found: obj =list_first_entry (&unwind_list, struct drm_i915_gem_object, exec_list); -if (drm_mm_scan_remove_block (obj->gtt_space)) { +if (drm_mm_scan_remove_block (obj->ggtt_space)) {list_move (&obj->exec_list, &eviction_list);drm_gem_object_reference (&obj->base);continue ;drivers/gpu/drm/i915/i915_gem_execbuffer.c +10
-9@@ -188,7 +188,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,return -ENOENT ; target_i915_obj =to_intel_bo (target_obj); - target_offset = target_i915_obj->gtt_offset; + target_offset =i915_gem_obj_ggtt_offset (target_i915_obj);/* Sandybridge PPGTT errata: We need a global gtt mapping for MI and * pipe_control writes because the gpu doesn't properly redirect them @@ -280,7 +280,7 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj,return ret;/* Map the page containing the relocation we're going to perform. */ - reloc->offset += obj->gtt_offset; + reloc->offset +=i915_gem_obj_ggtt_offset (obj); reloc_page =io_mapping_map_atomic_wc (dev_priv->gtt.mappable, reloc->offset &PAGE_MASK ); reloc_entry = (uint32_t __iomem *)@@ -436,8 +436,8 @@ i915_gem_execbuffer_reserve_object(struct drm_i915_gem_object *obj, obj->has_aliasing_ppgtt_mapping =1 ; } -if (entry->offset != obj->gtt_offset) { - entry->offset = obj->gtt_offset; +if (entry->offset !=i915_gem_obj_ggtt_offset (obj)) { + entry->offset =i915_gem_obj_ggtt_offset (obj); *need_reloc = true; }@@ -458,7 +458,7 @@ i915_gem_execbuffer_unreserve_object(struct drm_i915_gem_object *obj) {struct drm_i915_gem_exec_object2 *entry; -if (!obj->gtt_space) +if (!i915_gem_obj_ggtt_bound (obj))return ; entry = obj->exec_entry;@@ -530,7 +530,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,struct drm_i915_gem_exec_object2 *entry = obj->exec_entry;bool need_fence, need_mappable; -if (!obj->gtt_space) +if (!i915_gem_obj_ggtt_bound (obj))continue ; need_fence =@@ -539,7 +539,8 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring, obj->tiling_mode !=I915_TILING_NONE ; need_mappable = need_fence ||need_reloc_mappable (obj); -if ((entry->alignment && obj->gtt_offset & (entry->alignment -1 )) || +if ((entry->alignment && +i915_gem_obj_ggtt_offset (obj) & (entry->alignment -1 )) || (need_mappable && !obj->map_and_fenceable)) ret =i915_gem_object_unbind (obj);else @@ -550,7 +551,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring,/* Bind fresh objects */ list_for_each_entry (obj, objects, exec_list) { -if (obj->gtt_space) +if (i915_gem_obj_ggtt_bound (obj))continue ; ret =i915_gem_execbuffer_reserve_object (obj, ring, need_relocs);@@ -1058,7 +1059,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, goto err; } - exec_start = batch_obj->gtt_offset + args->batch_start_offset; + exec_start =i915_gem_obj_ggtt_offset (batch_obj) + args->batch_start_offset; exec_len = args->batch_len;if (cliprects) {for (i =0 ; i < args->num_cliprects; i++) {drivers/gpu/drm/i915/i915_gem_gtt.c +12
-12@@ -378,7 +378,7 @@ void i915_ppgtt_bind_object(struct i915_hw_ppgtt *ppgtt, enum i915_cache_level cache_level) { ppgtt->insert_entries (ppgtt, obj->pages, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , cache_level); }@@ -386,7 +386,7 @@ void i915_ppgtt_unbind_object(struct i915_hw_ppgtt *ppgtt, struct drm_i915_gem_object *obj) { ppgtt->clear_range (ppgtt, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , obj->base.size >>PAGE_SHIFT ); }@@ -551,7 +551,7 @@ void i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj,struct drm_i915_private *dev_priv = dev->dev_private; dev_priv->gtt.gtt_insert_entries (dev, obj->pages, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , cache_level); obj->has_global_gtt_mapping =1 ;@@ -563,7 +563,7 @@ void i915_gem_gtt_unbind_object(struct drm_i915_gem_object *obj)struct drm_i915_private *dev_priv = dev->dev_private; dev_priv->gtt.gtt_clear_range (obj->base.dev, - obj->gtt_space->start >>PAGE_SHIFT , +i915_gem_obj_ggtt_offset (obj) >>PAGE_SHIFT , obj->base.size >>PAGE_SHIFT ); obj->has_global_gtt_mapping =0 ;@@ -630,22 +630,22 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,/* Mark any preallocated objects as occupied */ list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) {int ret; -DRM_DEBUG_KMS ("reserving preallocated space: %x + %zx\n" , - obj->gtt_offset, obj->base.size); +DRM_DEBUG_KMS ("reserving preallocated space: %lx + %zx\n" , +i915_gem_obj_ggtt_offset (obj), obj->base.size); -BUG_ON (obj->gtt_space !=I915_GTT_RESERVED ); - obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); -if (!obj->gtt_space) { +BUG_ON (obj->ggtt_space !=I915_GTT_RESERVED ); + obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL ); +if (!obj->ggtt_space) {DRM_ERROR ("Failed to preserve all objects\n" );break ; } ret =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_space, - obj->gtt_offset, + obj->ggtt_space, +i915_gem_obj_ggtt_offset (obj), obj->base.size);if (ret) {DRM_DEBUG_KMS ("Reservation failed\n" ); -kfree (obj->gtt_space); +kfree (obj->ggtt_space); } obj->has_global_gtt_mapping =1 ; }drivers/gpu/drm/i915/i915_gem_stolen.c +5
-5@@ -374,23 +374,23 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); -if (!obj->gtt_space) { + obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL ); +if (!obj->ggtt_space) {DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); goto unref_out; } ret =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_space, + obj->ggtt_space, gtt_offset, size);if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); goto unref_out; } }else - obj->gtt_space =I915_GTT_RESERVED ; + obj->ggtt_space =I915_GTT_RESERVED ; - obj->gtt_offset = gtt_offset; + obj->ggtt_offset = gtt_offset; obj->has_global_gtt_mapping =1 ;list_add_tail (&obj->global_list, &dev_priv->mm.bound_list);drivers/gpu/drm/i915/i915_gem_tiling.c +7
-7@@ -268,18 +268,18 @@ i915_gem_object_fence_ok(struct drm_i915_gem_object *obj, int tiling_mode)return true;if (INTEL_INFO (obj->base.dev)->gen ==3 ) { -if (obj->gtt_offset & ~I915_FENCE_START_MASK ) +if (i915_gem_obj_ggtt_offset (obj) & ~I915_FENCE_START_MASK )return false; }else { -if (obj->gtt_offset & ~I830_FENCE_START_MASK ) +if (i915_gem_obj_ggtt_offset (obj) & ~I830_FENCE_START_MASK )return false; } size =i915_gem_get_gtt_size (obj->base.dev, obj->base.size, tiling_mode); -if (obj->gtt_space->size != size) +if (i915_gem_obj_ggtt_size (obj) != size)return false; -if (obj->gtt_offset & (size -1 )) +if (i915_gem_obj_ggtt_offset (obj) & (size -1 ))return false;return true;@@ -359,8 +359,8 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,*/ obj->map_and_fenceable = - obj->gtt_space ==NULL || - (obj->gtt_offset + obj->base.size <= dev_priv->gtt.mappable_end && + !i915_gem_obj_ggtt_bound (obj) || + (i915_gem_obj_ggtt_offset (obj) + obj->base.size <= dev_priv->gtt.mappable_end &&i915_gem_object_fence_ok (obj, args->tiling_mode));/* Rebind if we need a change of alignment */ @@ -369,7 +369,7 @@ i915_gem_set_tiling(struct drm_device *dev, void *data,i915_gem_get_gtt_alignment (dev, obj->base.size, args->tiling_mode, false); -if (obj->gtt_offset & (unfenced_alignment -1 )) +if (i915_gem_obj_ggtt_offset (obj) & (unfenced_alignment -1 )) ret =i915_gem_object_unbind (obj); }drivers/gpu/drm/i915/i915_irq.c +7
-8@@ -1516,7 +1516,7 @@ i915_error_object_create_sized(struct drm_i915_private *dev_priv,if (dst ==NULL )return NULL ; - reloc_offset = src->gtt_offset; + reloc_offset = dst->gtt_offset =i915_gem_obj_ggtt_offset (src);for (i =0 ; i < num_pages; i++) {unsigned long flags;void *d;@@ -1568,7 +1568,6 @@ i915_error_object_create_sized(struct drm_i915_private *dev_priv, reloc_offset +=PAGE_SIZE ; } dst->page_count = num_pages; - dst->gtt_offset = src->gtt_offset;return dst;@@ -1622,7 +1621,7 @@ static void capture_bo(struct drm_i915_error_buffer *err, err->name = obj->base.name; err->rseqno = obj->last_read_seqno; err->wseqno = obj->last_write_seqno; - err->gtt_offset = obj->gtt_offset; + err->gtt_offset =i915_gem_obj_ggtt_offset (obj); err->read_domains = obj->base.read_domains; err->write_domain = obj->base.write_domain; err->fence_reg = obj->fence_reg;@@ -1720,8 +1719,8 @@ i915_error_first_batchbuffer(struct drm_i915_private *dev_priv,return NULL ; obj = ring->private; -if (acthd >= obj->gtt_offset && - acthd < obj->gtt_offset + obj->base.size) +if (acthd >=i915_gem_obj_ggtt_offset (obj) && + acthd <i915_gem_obj_ggtt_offset (obj) + obj->base.size)return i915_error_object_create (dev_priv, obj); }@@ -1802,7 +1801,7 @@ static void i915_gem_record_active_context(struct intel_ring_buffer *ring,return ;list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) { -if ((error->ccid &PAGE_MASK ) == obj->gtt_offset) { +if ((error->ccid &PAGE_MASK ) ==i915_gem_obj_ggtt_offset (obj)) { ering->ctx =i915_error_object_create_sized (dev_priv, obj,1 );break ;@@ -2156,10 +2155,10 @@ static void __always_unused i915_pageflip_stall_check(struct drm_device *dev, inif (INTEL_INFO (dev)->gen >=4 ) {int dspsurf =DSPSURF (intel_crtc->plane); stall_detected =I915_HI_DISPBASE (I915_READ (dspsurf)) == - obj->gtt_offset; +i915_gem_obj_ggtt_offset (obj); }else {int dspaddr =DSPADDR (intel_crtc->plane); - stall_detected =I915_READ (dspaddr) == (obj->gtt_offset + + stall_detected =I915_READ (dspaddr) == (i915_gem_obj_ggtt_offset (obj) + crtc->y * crtc->fb->pitches[0 ] + crtc->x * crtc->fb->bits_per_pixel/8 ); }drivers/gpu/drm/i915/i915_trace.h +4
-4@@ -46,8 +46,8 @@ TRACE_EVENT(i915_gem_object_bind,TP_fast_assign ( __entry->obj = obj; - __entry->offset = obj->gtt_space->start; - __entry->size = obj->gtt_space->size; + __entry->offset =i915_gem_obj_ggtt_offset (obj); + __entry->size =i915_gem_obj_ggtt_size (obj); __entry->mappable = mappable; ),@@ -68,8 +68,8 @@ TRACE_EVENT(i915_gem_object_unbind,TP_fast_assign ( __entry->obj = obj; - __entry->offset = obj->gtt_space->start; - __entry->size = obj->gtt_space->size; + __entry->offset =i915_gem_obj_ggtt_offset (obj); + __entry->size =i915_gem_obj_ggtt_size (obj); ), TP_printk("obj=%p, offset=%08x size=%x",drivers/gpu/drm/i915/intel_display.c +15
-13@@ -1980,16 +1980,17 @@ static int i9xx_update_plane(struct drm_crtc *crtc, struct drm_framebuffer *fb, intel_crtc->dspaddr_offset = linear_offset; } -DRM_DEBUG_KMS ("Writing base %08X %08lX %d %d %d\n" , - obj->gtt_offset, linear_offset, x, y, fb->pitches[0 ]); +DRM_DEBUG_KMS ("Writing base %08lX %08lX %d %d %d\n" , +i915_gem_obj_ggtt_offset (obj), linear_offset, x, y, + fb->pitches[0 ]);I915_WRITE (DSPSTRIDE (plane), fb->pitches[0 ]);if (INTEL_INFO (dev)->gen >=4 ) {I915_MODIFY_DISPBASE (DSPSURF (plane), - obj->gtt_offset + intel_crtc->dspaddr_offset); +i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);I915_WRITE (DSPTILEOFF (plane), (y <<16 ) | x);I915_WRITE (DSPLINOFF (plane), linear_offset); }else -I915_WRITE (DSPADDR (plane), obj->gtt_offset + linear_offset); +I915_WRITE (DSPADDR (plane),i915_gem_obj_ggtt_offset (obj) + linear_offset);POSTING_READ (reg);return 0 ;@@ -2069,11 +2070,12 @@ static int ironlake_update_plane(struct drm_crtc *crtc, fb->pitches[0]); linear_offset -= intel_crtc->dspaddr_offset; -DRM_DEBUG_KMS ("Writing base %08X %08lX %d %d %d\n" , - obj->gtt_offset, linear_offset, x, y, fb->pitches[0 ]); +DRM_DEBUG_KMS ("Writing base %08lX %08lX %d %d %d\n" , +i915_gem_obj_ggtt_offset (obj), linear_offset, x, y, + fb->pitches[0 ]);I915_WRITE (DSPSTRIDE (plane), fb->pitches[0 ]);I915_MODIFY_DISPBASE (DSPSURF (plane), - obj->gtt_offset + intel_crtc->dspaddr_offset); +i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);if (IS_HASWELL (dev)) {I915_WRITE (DSPOFFSET (plane), (y <<16 ) | x); }else {@@ -6566,7 +6568,7 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc, goto fail_unpin; } - addr = obj->gtt_offset; + addr =i915_gem_obj_ggtt_offset (obj); }else {int align =IS_I830 (dev) ?16 *1024 :256 ; ret = i915_gem_attach_phys_object(dev, obj,@@ -7338,7 +7340,7 @@ static int intel_gen2_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP |MI_DISPLAY_FLIP_PLANE (intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ]); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);intel_ring_emit (ring,0 );/* aux display base address, unused */ intel_mark_page_flip_active (intel_crtc);@@ -7379,7 +7381,7 @@ static int intel_gen3_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP_I915 |MI_DISPLAY_FLIP_PLANE (intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ]); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);intel_ring_emit (ring,MI_NOOP );intel_mark_page_flip_active (intel_crtc);@@ -7419,7 +7421,7 @@ static int intel_gen4_queue_flip(struct drm_device *dev, MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ]);intel_ring_emit (ring, - (obj->gtt_offset + intel_crtc->dspaddr_offset) | + (i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset) | obj->tiling_mode);/* XXX Enabling the panel-fitter across page-flip is so far @@ -7462,7 +7464,7 @@ static int intel_gen6_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP |MI_DISPLAY_FLIP_PLANE (intel_crtc->plane));intel_ring_emit (ring, fb->pitches[0 ] | obj->tiling_mode); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);/* Contrary to the suggestions in the documentation, * "Enable Panel Fitter" does not seem to be required when page @@ -7527,7 +7529,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev,intel_ring_emit (ring,MI_DISPLAY_FLIP_I915 | plane_bit);intel_ring_emit (ring, (fb->pitches[0 ] | obj->tiling_mode)); -intel_ring_emit (ring, obj->gtt_offset + intel_crtc->dspaddr_offset); +intel_ring_emit (ring,i915_gem_obj_ggtt_offset (obj) + intel_crtc->dspaddr_offset);intel_ring_emit (ring, (MI_NOOP ));intel_mark_page_flip_active (intel_crtc);drivers/gpu/drm/i915/intel_fb.c +4
-4@@ -139,11 +139,11 @@ static int intelfb_create(struct drm_fb_helper *helper, info->apertures->ranges[0 ].base = dev->mode_config.fb_base; info->apertures->ranges[0 ].size = dev_priv->gtt.mappable_end; - info->fix.smem_start = dev->mode_config.fb_base + obj->gtt_offset; + info->fix.smem_start = dev->mode_config.fb_base +i915_gem_obj_ggtt_offset (obj); info->fix.smem_len = size; info->screen_base = -ioremap_wc (dev_priv->gtt.mappable_base + obj->gtt_offset, +ioremap_wc (dev_priv->gtt.mappable_base +i915_gem_obj_ggtt_offset (obj), size);if (!info->screen_base) { ret = -ENOSPC ;@@ -166,9 +166,9 @@ static int intelfb_create(struct drm_fb_helper *helper,/* Use default scratch pixmap (info->pixmap.flags = FB_PIXMAP_SYSTEM) */ -DRM_DEBUG_KMS ("allocated %dx%d fb: 0x%08x, bo %p\n" , +DRM_DEBUG_KMS ("allocated %dx%d fb: 0x%08lx, bo %p\n" , fb->width, fb->height, - obj->gtt_offset, obj); +i915_gem_obj_ggtt_offset (obj), obj);mutex_unlock (&dev->struct_mutex);drivers/gpu/drm/i915/intel_overlay.c +7
-7@@ -196,7 +196,7 @@ intel_overlay_map_regs(struct intel_overlay *overlay) regs = (struct overlay_registers __iomem *)overlay->reg_bo->phys_obj->handle->vaddr; else regs =io_mapping_map_wc (dev_priv->gtt.mappable, - overlay->reg_bo->gtt_offset); +i915_gem_obj_ggtt_offset (overlay->reg_bo));return regs; }@@ -740,7 +740,7 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay, swidth = params->src_w; swidthsw =calc_swidthsw (overlay->dev, params->offset_Y, tmp_width); sheight = params->src_h; -iowrite32 (new_bo->gtt_offset + params->offset_Y, ®s->OBUF_0Y); +iowrite32 (i915_gem_obj_ggtt_offset (new_bo) + params->offset_Y, ®s->OBUF_0Y); ostride = params->stride_Y;if (params->format &I915_OVERLAY_YUV_PLANAR ) {@@ -754,8 +754,8 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay, params->src_w/uv_hscale); swidthsw |=max_t (u32, tmp_U, tmp_V) <<16 ; sheight |= (params->src_h/uv_vscale) <<16 ; -iowrite32 (new_bo->gtt_offset + params->offset_U, ®s->OBUF_0U); -iowrite32 (new_bo->gtt_offset + params->offset_V, ®s->OBUF_0V); +iowrite32 (i915_gem_obj_ggtt_offset (new_bo) + params->offset_U, ®s->OBUF_0U); +iowrite32 (i915_gem_obj_ggtt_offset (new_bo) + params->offset_V, ®s->OBUF_0V); ostride |= params->stride_UV <<16 ; }@@ -1355,7 +1355,7 @@ void intel_setup_overlay(struct drm_device *dev)DRM_ERROR ("failed to pin overlay register bo\n" ); goto out_free_bo; } - overlay->flip_addr = reg_bo->gtt_offset; + overlay->flip_addr =i915_gem_obj_ggtt_offset (reg_bo); ret =i915_gem_object_set_to_gtt_domain (reg_bo, true);if (ret) {@@ -1435,7 +1435,7 @@ intel_overlay_map_regs_atomic(struct intel_overlay *overlay) overlay->reg_bo->phys_obj->handle->vaddr; else regs =io_mapping_map_atomic_wc (dev_priv->gtt.mappable, - overlay->reg_bo->gtt_offset); +i915_gem_obj_ggtt_offset (overlay->reg_bo));return regs; }@@ -1468,7 +1468,7 @@ intel_overlay_capture_error_state(struct drm_device *dev)if (OVERLAY_NEEDS_PHYSICAL (overlay->dev)) error->base = (__force long )overlay->reg_bo->phys_obj->handle->vaddr;else - error->base = overlay->reg_bo->gtt_offset; + error->base =i915_gem_obj_ggtt_offset (overlay->reg_bo); regs =intel_overlay_map_regs_atomic (overlay);if (!regs)drivers/gpu/drm/i915/intel_pm.c +4
-4@@ -218,7 +218,7 @@ static void ironlake_enable_fbc(struct drm_crtc *crtc, unsigned long interval) (stall_watermark <<DPFC_RECOMP_STALL_WM_SHIFT ) | (interval << DPFC_RECOMP_TIMER_COUNT_SHIFT));I915_WRITE (ILK_DPFC_FENCE_YOFF , crtc->y); -I915_WRITE (ILK_FBC_RT_BASE , obj->gtt_offset |ILK_FBC_RT_VALID ); +I915_WRITE (ILK_FBC_RT_BASE ,i915_gem_obj_ggtt_offset (obj) |ILK_FBC_RT_VALID );/* enable it... */ I915_WRITE (ILK_DPFC_CONTROL , dpfc_ctl |DPFC_CTL_EN );@@ -275,7 +275,7 @@ static void gen7_enable_fbc(struct drm_crtc *crtc, unsigned long interval)struct drm_i915_gem_object *obj = intel_fb->obj;struct intel_crtc *intel_crtc =to_intel_crtc (crtc); -I915_WRITE (IVB_FBC_RT_BASE , obj->gtt_offset); +I915_WRITE (IVB_FBC_RT_BASE ,i915_gem_obj_ggtt_offset (obj)); I915_WRITE(ILK_DPFC_CONTROL, DPFC_CTL_EN | DPFC_CTL_LIMIT_1X | IVB_DPFC_CTL_FENCE_EN |@@ -3707,7 +3707,7 @@ static void ironlake_enable_rc6(struct drm_device *dev)intel_ring_emit (ring,MI_SUSPEND_FLUSH |MI_SUSPEND_FLUSH_EN );intel_ring_emit (ring,MI_SET_CONTEXT ); - intel_ring_emit(ring, dev_priv->ips.renderctx->gtt_offset | + intel_ring_emit(ring, i915_gem_obj_ggtt_offset(dev_priv->ips.renderctx) | MI_MM_SPACE_GTT | MI_SAVE_EXT_STATE_EN | MI_RESTORE_EXT_STATE_EN |@@ -3730,7 +3730,7 @@ static void ironlake_enable_rc6(struct drm_device *dev)return ; } -I915_WRITE (PWRCTXA , dev_priv->ips.pwrctx->gtt_offset |PWRCTX_EN ); +I915_WRITE (PWRCTXA ,i915_gem_obj_ggtt_offset (dev_priv->ips.pwrctx) |PWRCTX_EN );I915_WRITE (RSTDBYCTL ,I915_READ (RSTDBYCTL ) & ~RCX_SW_EXIT ); }drivers/gpu/drm/i915/intel_ringbuffer.c +6
-6@@ -424,14 +424,14 @@ static int init_ring_common(struct intel_ring_buffer *ring)* registers with the above sequence (the readback of the HEAD registers * also enforces ordering), otherwise the hw might lose the new ring * register values. */ -I915_WRITE_START (ring, obj->gtt_offset); +I915_WRITE_START (ring,i915_gem_obj_ggtt_offset (obj));I915_WRITE_CTL (ring, ((ring->size -PAGE_SIZE ) &RING_NR_PAGES ) |RING_VALID );/* If the head is still not zero, the ring is dead */ if (wait_for ((I915_READ_CTL (ring) &RING_VALID ) !=0 && -I915_READ_START (ring) == obj->gtt_offset && +I915_READ_START (ring) ==i915_gem_obj_ggtt_offset (obj) && (I915_READ_HEAD (ring) &HEAD_ADDR ) ==0 ,50 )) { DRM_ERROR("%s initialization failed " "ctl %08x head %08x tail %08x start %08x\n",@@ -489,7 +489,7 @@ init_pipe_control(struct intel_ring_buffer *ring)if (ret) goto err_unref; - pc->gtt_offset = obj->gtt_offset; + pc->gtt_offset =i915_gem_obj_ggtt_offset (obj); pc->cpu_page =kmap (sg_page (obj->pages->sgl));if (pc->cpu_page ==NULL ) { ret = -ENOMEM ;@@ -1129,7 +1129,7 @@ i830_dispatch_execbuffer(struct intel_ring_buffer *ring,intel_ring_advance (ring); }else {struct drm_i915_gem_object *obj = ring->private; -u32 cs_offset = obj->gtt_offset; +u32 cs_offset =i915_gem_obj_ggtt_offset (obj);if (len >I830_BATCH_LIMIT )return -ENOSPC ;@@ -1214,7 +1214,7 @@ static int init_status_page(struct intel_ring_buffer *ring) goto err_unref; } - ring->status_page.gfx_addr = obj->gtt_offset; + ring->status_page.gfx_addr =i915_gem_obj_ggtt_offset (obj); ring->status_page.page_addr =kmap (sg_page (obj->pages->sgl));if (ring->status_page.page_addr ==NULL ) { ret = -ENOMEM ;@@ -1308,7 +1308,7 @@ static int intel_init_ring_buffer(struct drm_device *dev, goto err_unpin; ring->virtual_start = -ioremap_wc (dev_priv->gtt.mappable_base + obj->gtt_offset, +ioremap_wc (dev_priv->gtt.mappable_base +i915_gem_obj_ggtt_offset (obj), ring->size);if (ring->virtual_start ==NULL ) {DRM_ERROR ("Failed to map ringbuffer.\n" );drivers/gpu/drm/i915/intel_sprite.c +5
-3@@ -133,7 +133,7 @@ vlv_update_plane(struct drm_plane *dplane, struct drm_framebuffer *fb,I915_WRITE (SPSIZE (pipe, plane), (crtc_h <<16 ) | crtc_w);I915_WRITE (SPCNTR (pipe, plane), sprctl); -I915_MODIFY_DISPBASE (SPSURF (pipe, plane), obj->gtt_offset + +I915_MODIFY_DISPBASE (SPSURF (pipe, plane),i915_gem_obj_ggtt_offset (obj) + sprsurf_offset);POSTING_READ (SPSURF (pipe, plane)); }@@ -308,7 +308,8 @@ ivb_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,if (intel_plane->can_scale)I915_WRITE (SPRSCALE (pipe), sprscale);I915_WRITE (SPRCTL (pipe), sprctl); -I915_MODIFY_DISPBASE (SPRSURF (pipe), obj->gtt_offset + sprsurf_offset); +I915_MODIFY_DISPBASE (SPRSURF (pipe), +i915_gem_obj_ggtt_offset (obj) + sprsurf_offset);POSTING_READ (SPRSURF (pipe));/* potentially re-enable LP watermarks */ @@ -478,7 +479,8 @@ ilk_update_plane(struct drm_plane *plane, struct drm_framebuffer *fb,I915_WRITE (DVSSIZE (pipe), (crtc_h <<16 ) | crtc_w);I915_WRITE (DVSSCALE (pipe), dvsscale);I915_WRITE (DVSCNTR (pipe), dvscntr); -I915_MODIFY_DISPBASE (DVSSURF (pipe), obj->gtt_offset + dvssurf_offset); +I915_MODIFY_DISPBASE (DVSSURF (pipe), +i915_gem_obj_ggtt_offset (obj) + dvssurf_offset);POSTING_READ (DVSSURF (pipe)); }Signature · 6 lines
1.8.3.2 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfxSignature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
Signature · 1 line
Ben Widawsky, Intel Open Source Technology Center
PATCH 3/6 drm/i915: Kill obj->gtt_offset
In reply to Ben WidawskyWith the getters in place from the previous patch this members serves no purpose other than saving one spare pointer chase, which will be killed in the next patch anyway.
Moving to VMAs, this members adds unnecessary confusion since an object may exist at different offsets in different VMs.
--- drivers/gpu/drm/i915/i915_drv.h | 7 ------- drivers/gpu/drm/i915/i915_gem.c | 2 -- drivers/gpu/drm/i915/i915_gem_stolen.c | 1 - 3 files changed, 10 deletions(-)
drivers/gpu/drm/i915/i915_drv.h +0 -7
@@ -1328,13 +1328,6 @@ struct drm_i915_gem_object {unsigned long exec_handle;struct drm_i915_gem_exec_object2 *exec_entry; -/** -* Current offset of the object in GTT space. -* -* This is the same as gtt_space->start -*/ -uint32_t ggtt_offset; -struct intel_ring_buffer *ring;/** Breadcrumb of last rendering to the buffer. */
drivers/gpu/drm/i915/i915_gem.c +0 -2
@@ -2623,7 +2623,6 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj)drm_mm_put_block (obj->ggtt_space); obj->ggtt_space =NULL ; - obj->ggtt_offset =0 ;return 0 ; }
@@ -3153,7 +3152,6 @@ search_free:list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list); obj->ggtt_space = node; - obj->ggtt_offset = node->start; fenceable = node->size == fence_size &&
drivers/gpu/drm/i915/i915_gem_stolen.c +0 -1
@@ -390,7 +390,6 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev, }else obj->ggtt_space =I915_GTT_RESERVED ; - obj->ggtt_offset = gtt_offset; obj->has_global_gtt_mapping =1 ;list_add_tail (&obj->global_list, &dev_priv->mm.bound_list);
Signature · 1 line
1.8.3.2
Re: [PATCH 3/6] drm/i915: Kill obj->gtt_offset
Quoted 8 lines · Ben Widawsky
With the getters in place from the previous patch this members serves no purpose other than saving one spare pointer chase, which will be killed in the next patch anyway.
Moving to VMAs, this members adds unnecessary confusion since an object may exist at different offsets in different VMs.
Signed-off-by Ben Widawsky <[email protected]>
Ok, I've merged this and the previous patch to dinq (I'm stalling on the first one until David's comments are addressed).
-Daniel
Quoted 63 lines
--- drivers/gpu/drm/i915/i915_drv.h | 7 ------- drivers/gpu/drm/i915/i915_gem.c | 2 -- drivers/gpu/drm/i915/i915_gem_stolen.c | 1 - 3 files changed, 10 deletions(-)drivers/gpu/drm/i915/i915_drv.h +0
-7@@ -1328,13 +1328,6 @@ struct drm_i915_gem_object {unsigned long exec_handle;struct drm_i915_gem_exec_object2 *exec_entry; -/** -* Current offset of the object in GTT space. -* -* This is the same as gtt_space->start -*/ -uint32_t ggtt_offset; -struct intel_ring_buffer *ring;/** Breadcrumb of last rendering to the buffer. */ drivers/gpu/drm/i915/i915_gem.c +0
-2@@ -2623,7 +2623,6 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj)drm_mm_put_block (obj->ggtt_space); obj->ggtt_space =NULL ; - obj->ggtt_offset =0 ;return 0 ; }@@ -3153,7 +3152,6 @@ search_free:list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list); obj->ggtt_space = node; - obj->ggtt_offset = node->start; fenceable = node->size == fence_size &&drivers/gpu/drm/i915/i915_gem_stolen.c +0
-1@@ -390,7 +390,6 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev, }else obj->ggtt_space =I915_GTT_RESERVED ; - obj->ggtt_offset = gtt_offset; obj->has_global_gtt_mapping =1 ;list_add_tail (&obj->global_list, &dev_priv->mm.bound_list);Signature · 6 lines
1.8.3.2 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
PATCH 4/6 drm/i915: Use gtt_space->start for stolen reservation
In reply to Ben WidawskyShortly we'll want to switch to an embedded drm_mm_node per object, and therefore using pointers as we're doing is unfeasible.
I've chosen an impossible gtt start offset as the way to denote a reservation.
--- drivers/gpu/drm/i915/i915_drv.h | 2 +- drivers/gpu/drm/i915/i915_gem_gtt.c | 2 +- drivers/gpu/drm/i915/i915_gem_stolen.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/i915_drv.h +1 -1
@@ -1201,7 +1201,7 @@ enum hdmi_force_audio { HDMI_AUDIO_ON, /* force turn on HDMI audio */ }; -#define I915_GTT_RESERVED ((struct drm_mm_node *)0x1) +#define I915_GTT_RESERVED (dev_priv->gtt.total + 1)struct drm_i915_gem_object_ops {/* Interface between the GEM object and its backing storage.
drivers/gpu/drm/i915/i915_gem_gtt.c +1 -1
@@ -633,7 +633,7 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,DRM_DEBUG_KMS ("reserving preallocated space: %lx + %zx\n" ,i915_gem_obj_ggtt_offset (obj), obj->base.size); -BUG_ON (obj->ggtt_space !=I915_GTT_RESERVED ); +BUG_ON (obj->ggtt_space->start !=I915_GTT_RESERVED ); obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL );if (!obj->ggtt_space) {DRM_ERROR ("Failed to preserve all objects\n" );
drivers/gpu/drm/i915/i915_gem_stolen.c +1 -1
@@ -388,7 +388,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev, goto unref_out; } }else - obj->ggtt_space =I915_GTT_RESERVED ; + obj->ggtt_space->start =I915_GTT_RESERVED ; obj->has_global_gtt_mapping =1 ;
Signature · 1 line
1.8.3.2
Re: [PATCH 4/6] drm/i915: Use gtt_space->start for stolen reservation
Quoted 19 lines · Ben Widawsky
Shortly we'll want to switch to an embedded
drm_mm_nodeper object, and therefore using pointers as we're doing is unfeasible.I've chosen an impossible gtt start offset as the way to denote a reservation.
CC Chris Wilson <[email protected]>Signed-off-by Ben Widawsky <[email protected]>--- drivers/gpu/drm/i915/i915_drv.h | 2 +- drivers/gpu/drm/i915/i915_gem_gtt.c | 2 +- drivers/gpu/drm/i915/i915_gem_stolen.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-)drivers/gpu/drm/i915/i915_drv.h +0
-0
drivers/gpu/drm/i915/i915_drv.h enum hdmi_force_audio { · line 1204
Quoted 5 lines
drivers/gpu/drm/i915/i915_drv.h +1
-1@@ -1201,7 +1201,7 @@ enum hdmi_force_audio { HDMI_AUDIO_ON, /* force turn on HDMI audio */ }; -#define I915_GTT_RESERVED ((struct drm_mm_node *)0x1) +#define I915_GTT_RESERVED (dev_priv->gtt.total + 1)
This is _really_ fragile since if we every change gtt.total in the init sequence this will blow up. I've switched over to an appropriate real constant (ULONG_MAX).
-Daniel
Quoted 36 lines
drivers/gpu/drm/i915/i915_drv.h +0
-0struct drm_i915_gem_object_ops {/* Interface between the GEM object and its backing storage. drivers/gpu/drm/i915/i915_gem_gtt.c +1
-1@@ -633,7 +633,7 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,DRM_DEBUG_KMS ("reserving preallocated space: %lx + %zx\n" ,i915_gem_obj_ggtt_offset (obj), obj->base.size); -BUG_ON (obj->ggtt_space !=I915_GTT_RESERVED ); +BUG_ON (obj->ggtt_space->start !=I915_GTT_RESERVED ); obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL );if (!obj->ggtt_space) {DRM_ERROR ("Failed to preserve all objects\n" );drivers/gpu/drm/i915/i915_gem_stolen.c +1
-1@@ -388,7 +388,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev, goto unref_out; } }else - obj->ggtt_space =I915_GTT_RESERVED ; + obj->ggtt_space->start =I915_GTT_RESERVED ; obj->has_global_gtt_mapping =1 ;Signature · 6 lines
1.8.3.2 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
PATCH 5/6 drm/i915: Embed drm_mm_node in i915 gem obj
In reply to Ben WidawskyEmbedding the node in the obj is more natural in the transition to VMAs which will also have embedded nodes. This change also helps transition away from put_block to remove node.
Though it's quite an uncommon occurrence, it's somewhat convenient to not fail at bind time because we cannot allocate the node. Though in practice there are other allocations (like the request structure) which would probably make this point not terribly useful.
Quoting Daniel:
Note that the only difference between put_block and remove_node is that the former fills up the preallocation cache. Which we don't need anyway and hence is just wasted space.
--- drivers/gpu/drm/i915/i915_drv.h | 10 +++++----- drivers/gpu/drm/i915/i915_gem.c | 31 +++++++++++-------------------- drivers/gpu/drm/i915/i915_gem_evict.c | 6 +++--- drivers/gpu/drm/i915/i915_gem_gtt.c | 13 +++---------- drivers/gpu/drm/i915/i915_gem_stolen.c | 10 ++-------- 5 files changed, 24 insertions(+), 46 deletions(-)
drivers/gpu/drm/i915/i915_drv.h +5 -5
@@ -1227,7 +1227,7 @@ struct drm_i915_gem_object {const struct drm_i915_gem_object_ops *ops;/** Current space allocated to this object in the GTT, if any. */ -struct drm_mm_node *ggtt_space; +struct drm_mm_node ggtt_space;/** Stolen memory for this object, instead of being backed by shmem. */ struct drm_mm_node *stolen;struct list_head global_list;
@@ -1357,14 +1357,14 @@ struct drm_i915_gem_object {static inline unsigned long i915_gem_obj_ggtt_offset (struct drm_i915_gem_object *o) { -return o->ggtt_space->start; +return o->ggtt_space.start; }/* Whether or not this object is currently mapped by the translation tables */ static inline bool i915_gem_obj_ggtt_bound (struct drm_i915_gem_object *o) { -return o->ggtt_space !=NULL ; +return drm_mm_node_allocated (&o->ggtt_space); }/* The size used in the translation tables may be larger than the actual size of
@@ -1374,14 +1374,14 @@ i915_gem_obj_ggtt_bound(struct drm_i915_gem_object *o)static inline unsigned long i915_gem_obj_ggtt_size (struct drm_i915_gem_object *o) { -return o->ggtt_space->size; +return o->ggtt_space.size; }static inline void i915_gem_obj_ggtt_set_color (struct drm_i915_gem_object *o,enum i915_cache_level color) { - o->ggtt_space->color = color; + o->ggtt_space.color = color; }/**
drivers/gpu/drm/i915/i915_gem.c +11 -20
@@ -2621,8 +2621,7 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj)/* Avoid an unnecessary call to unbind on rebind. */ obj->map_and_fenceable = true; -drm_mm_put_block (obj->ggtt_space); - obj->ggtt_space =NULL ; +drm_mm_remove_node (&obj->ggtt_space);return 0 ; }
@@ -3000,7 +2999,7 @@ static bool i915_gem_valid_gtt_space(struct drm_device *dev,if (HAS_LLC (dev))return true; -if (gtt_space ==NULL ) +if (!drm_mm_node_allocated (gtt_space))return true;if (list_empty (>t_space->node_list))
@@ -3068,7 +3067,6 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj, {struct drm_device *dev = obj->base.dev;drm_i915_private_t *dev_priv = dev->dev_private; -struct drm_mm_node *node;u32 size, fence_size, fence_alignment, unfenced_alignment;bool mappable, fenceable; size_t gtt_max = map_and_fenceable ?
@@ -3113,14 +3111,9 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj,i915_gem_object_pin_pages (obj); - node =kzalloc (sizeof (*node),GFP_KERNEL ); -if (node ==NULL ) { -i915_gem_object_unpin_pages (obj); -return -ENOMEM ; - } - search_free: - ret =drm_mm_insert_node_in_range_generic (&dev_priv->mm.gtt_space, node, + ret =drm_mm_insert_node_in_range_generic (&dev_priv->mm.gtt_space, + &obj->ggtt_space, size, alignment, obj->cache_level,0 , gtt_max);if (ret) {
@@ -3132,30 +3125,28 @@ search_free: goto search_free;i915_gem_object_unpin_pages (obj); -kfree (node);return ret; } -if (WARN_ON (!i915_gem_valid_gtt_space (dev, node, obj->cache_level))) { +if (WARN_ON (!i915_gem_valid_gtt_space (dev, &obj->ggtt_space, + obj->cache_level))) {i915_gem_object_unpin_pages (obj); -drm_mm_put_block (node); +drm_mm_remove_node (&obj->ggtt_space);return -EINVAL ; } ret =i915_gem_gtt_prepare_object (obj);if (ret) {i915_gem_object_unpin_pages (obj); -drm_mm_put_block (node); +drm_mm_remove_node (&obj->ggtt_space);return ret; }list_move_tail (&obj->global_list, &dev_priv->mm.bound_list);list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list); - obj->ggtt_space = node; - fenceable = - node->size == fence_size && - (node->start & (fence_alignment -1 )) ==0 ; +i915_gem_obj_ggtt_size (obj) == fence_size && + (i915_gem_obj_ggtt_offset (obj) & (fence_alignment -1 )) ==0 ; mappable =i915_gem_obj_ggtt_offset (obj) + obj->base.size <= dev_priv->gtt.mappable_end;
@@ -3319,7 +3310,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,return -EBUSY ; } -if (!i915_gem_valid_gtt_space (dev, obj->ggtt_space, cache_level)) { +if (!i915_gem_valid_gtt_space (dev, &obj->ggtt_space, cache_level)) { ret =i915_gem_object_unbind (obj);if (ret)return ret;
drivers/gpu/drm/i915/i915_gem_evict.c +3 -3
@@ -38,7 +38,7 @@ mark_free(struct drm_i915_gem_object *obj, struct list_head *unwind)return false;list_add (&obj->exec_list, unwind); -return drm_mm_scan_add_block (obj->ggtt_space); +return drm_mm_scan_add_block (&obj->ggtt_space); }int
@@ -107,7 +107,7 @@ none: struct drm_i915_gem_object, exec_list); - ret =drm_mm_scan_remove_block (obj->ggtt_space); + ret =drm_mm_scan_remove_block (&obj->ggtt_space);BUG_ON (ret);list_del_init (&obj->exec_list);
@@ -127,7 +127,7 @@ found: obj =list_first_entry (&unwind_list, struct drm_i915_gem_object, exec_list); -if (drm_mm_scan_remove_block (obj->ggtt_space)) { +if (drm_mm_scan_remove_block (&obj->ggtt_space)) {list_move (&obj->exec_list, &eviction_list);drm_gem_object_reference (&obj->base);continue ;
drivers/gpu/drm/i915/i915_gem_gtt.c +3 -10
@@ -633,20 +633,13 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,DRM_DEBUG_KMS ("reserving preallocated space: %lx + %zx\n" ,i915_gem_obj_ggtt_offset (obj), obj->base.size); -BUG_ON (obj->ggtt_space->start !=I915_GTT_RESERVED ); - obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL ); -if (!obj->ggtt_space) { -DRM_ERROR ("Failed to preserve all objects\n" ); -break ; - } +BUG_ON (i915_gem_obj_ggtt_offset (obj) !=I915_GTT_RESERVED ); ret =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->ggtt_space, + &obj->ggtt_space,i915_gem_obj_ggtt_offset (obj), obj->base.size); -if (ret) { +if (ret)DRM_DEBUG_KMS ("Reservation failed\n" ); -kfree (obj->ggtt_space); - } obj->has_global_gtt_mapping =1 ; }
drivers/gpu/drm/i915/i915_gem_stolen.c +2 -8
@@ -374,21 +374,15 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL ); -if (!obj->ggtt_space) { -DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); - goto unref_out; - } - ret =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->ggtt_space, + &obj->ggtt_space, gtt_offset, size);if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); goto unref_out; } }else - obj->ggtt_space->start =I915_GTT_RESERVED ; + obj->ggtt_space.start =I915_GTT_RESERVED ; obj->has_global_gtt_mapping =1 ;
Signature · 1 line
1.8.3.2
Re: [PATCH 5/6] drm/i915: Embed drm_mm_node in i915 gem obj
Quoted 9 lines · Ben Widawsky
Embedding the node in the obj is more natural in the transition to VMAs which will also have embedded nodes. This change also helps transition away from
put_blockto remove node.Though it's quite an uncommon occurrence, it's somewhat convenient to not fail at bind time because we cannot allocate the node. Though in practice there are other allocations (like the request structure) which would probably make this point not terribly useful.
Quoted 6 lines
Quoting Daniel:
Note that the only difference betweenput_blockandremove_nodeis that the former fills up the preallocation cache. Which we don't need anyway and hence is just wasted space.Signed-off-by Ben Widawsky <[email protected]>
This one is blocked for now on getting the create_block drm_mm changes in. -Daniel
Quoted 243 lines
--- drivers/gpu/drm/i915/i915_drv.h | 10 +++++----- drivers/gpu/drm/i915/i915_gem.c | 31 +++++++++++-------------------- drivers/gpu/drm/i915/i915_gem_evict.c | 6 +++--- drivers/gpu/drm/i915/i915_gem_gtt.c | 13 +++---------- drivers/gpu/drm/i915/i915_gem_stolen.c | 10 ++-------- 5 files changed, 24 insertions(+), 46 deletions(-)drivers/gpu/drm/i915/i915_drv.h +5
-5@@ -1227,7 +1227,7 @@ struct drm_i915_gem_object {const struct drm_i915_gem_object_ops *ops;/** Current space allocated to this object in the GTT, if any. */ -struct drm_mm_node *ggtt_space; +struct drm_mm_node ggtt_space;/** Stolen memory for this object, instead of being backed by shmem. */ struct drm_mm_node *stolen;struct list_head global_list;@@ -1357,14 +1357,14 @@ struct drm_i915_gem_object {static inline unsigned long i915_gem_obj_ggtt_offset (struct drm_i915_gem_object *o) { -return o->ggtt_space->start; +return o->ggtt_space.start; }/* Whether or not this object is currently mapped by the translation tables */ static inline bool i915_gem_obj_ggtt_bound (struct drm_i915_gem_object *o) { -return o->ggtt_space !=NULL ; +return drm_mm_node_allocated (&o->ggtt_space); }/* The size used in the translation tables may be larger than the actual size of @@ -1374,14 +1374,14 @@ i915_gem_obj_ggtt_bound(struct drm_i915_gem_object *o)static inline unsigned long i915_gem_obj_ggtt_size (struct drm_i915_gem_object *o) { -return o->ggtt_space->size; +return o->ggtt_space.size; }static inline void i915_gem_obj_ggtt_set_color (struct drm_i915_gem_object *o,enum i915_cache_level color) { - o->ggtt_space->color = color; + o->ggtt_space.color = color; }/** drivers/gpu/drm/i915/i915_gem.c +11
-20@@ -2621,8 +2621,7 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj)/* Avoid an unnecessary call to unbind on rebind. */ obj->map_and_fenceable = true; -drm_mm_put_block (obj->ggtt_space); - obj->ggtt_space =NULL ; +drm_mm_remove_node (&obj->ggtt_space);return 0 ; }@@ -3000,7 +2999,7 @@ static bool i915_gem_valid_gtt_space(struct drm_device *dev,if (HAS_LLC (dev))return true; -if (gtt_space ==NULL ) +if (!drm_mm_node_allocated (gtt_space))return true;if (list_empty (>t_space->node_list))@@ -3068,7 +3067,6 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj, {struct drm_device *dev = obj->base.dev;drm_i915_private_t *dev_priv = dev->dev_private; -struct drm_mm_node *node;u32 size, fence_size, fence_alignment, unfenced_alignment;bool mappable, fenceable; size_t gtt_max = map_and_fenceable ?@@ -3113,14 +3111,9 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj,i915_gem_object_pin_pages (obj); - node =kzalloc (sizeof (*node),GFP_KERNEL ); -if (node ==NULL ) { -i915_gem_object_unpin_pages (obj); -return -ENOMEM ; - } - search_free: - ret =drm_mm_insert_node_in_range_generic (&dev_priv->mm.gtt_space, node, + ret =drm_mm_insert_node_in_range_generic (&dev_priv->mm.gtt_space, + &obj->ggtt_space, size, alignment, obj->cache_level,0 , gtt_max);if (ret) {@@ -3132,30 +3125,28 @@ search_free: goto search_free;i915_gem_object_unpin_pages (obj); -kfree (node);return ret; } -if (WARN_ON (!i915_gem_valid_gtt_space (dev, node, obj->cache_level))) { +if (WARN_ON (!i915_gem_valid_gtt_space (dev, &obj->ggtt_space, + obj->cache_level))) {i915_gem_object_unpin_pages (obj); -drm_mm_put_block (node); +drm_mm_remove_node (&obj->ggtt_space);return -EINVAL ; } ret =i915_gem_gtt_prepare_object (obj);if (ret) {i915_gem_object_unpin_pages (obj); -drm_mm_put_block (node); +drm_mm_remove_node (&obj->ggtt_space);return ret; }list_move_tail (&obj->global_list, &dev_priv->mm.bound_list);list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list); - obj->ggtt_space = node; - fenceable = - node->size == fence_size && - (node->start & (fence_alignment -1 )) ==0 ; +i915_gem_obj_ggtt_size (obj) == fence_size && + (i915_gem_obj_ggtt_offset (obj) & (fence_alignment -1 )) ==0 ; mappable =i915_gem_obj_ggtt_offset (obj) + obj->base.size <= dev_priv->gtt.mappable_end;@@ -3319,7 +3310,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,return -EBUSY ; } -if (!i915_gem_valid_gtt_space (dev, obj->ggtt_space, cache_level)) { +if (!i915_gem_valid_gtt_space (dev, &obj->ggtt_space, cache_level)) { ret =i915_gem_object_unbind (obj);if (ret)return ret;drivers/gpu/drm/i915/i915_gem_evict.c +3
-3@@ -38,7 +38,7 @@ mark_free(struct drm_i915_gem_object *obj, struct list_head *unwind)return false;list_add (&obj->exec_list, unwind); -return drm_mm_scan_add_block (obj->ggtt_space); +return drm_mm_scan_add_block (&obj->ggtt_space); }int @@ -107,7 +107,7 @@ none: struct drm_i915_gem_object, exec_list); - ret =drm_mm_scan_remove_block (obj->ggtt_space); + ret =drm_mm_scan_remove_block (&obj->ggtt_space);BUG_ON (ret);list_del_init (&obj->exec_list);@@ -127,7 +127,7 @@ found: obj =list_first_entry (&unwind_list, struct drm_i915_gem_object, exec_list); -if (drm_mm_scan_remove_block (obj->ggtt_space)) { +if (drm_mm_scan_remove_block (&obj->ggtt_space)) {list_move (&obj->exec_list, &eviction_list);drm_gem_object_reference (&obj->base);continue ;drivers/gpu/drm/i915/i915_gem_gtt.c +3
-10@@ -633,20 +633,13 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,DRM_DEBUG_KMS ("reserving preallocated space: %lx + %zx\n" ,i915_gem_obj_ggtt_offset (obj), obj->base.size); -BUG_ON (obj->ggtt_space->start !=I915_GTT_RESERVED ); - obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL ); -if (!obj->ggtt_space) { -DRM_ERROR ("Failed to preserve all objects\n" ); -break ; - } +BUG_ON (i915_gem_obj_ggtt_offset (obj) !=I915_GTT_RESERVED ); ret =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->ggtt_space, + &obj->ggtt_space,i915_gem_obj_ggtt_offset (obj), obj->base.size); -if (ret) { +if (ret)DRM_DEBUG_KMS ("Reservation failed\n" ); -kfree (obj->ggtt_space); - } obj->has_global_gtt_mapping =1 ; }drivers/gpu/drm/i915/i915_gem_stolen.c +2
-8@@ -374,21 +374,15 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->ggtt_space =kzalloc (sizeof (*obj->ggtt_space),GFP_KERNEL ); -if (!obj->ggtt_space) { -DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); - goto unref_out; - } - ret =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->ggtt_space, + &obj->ggtt_space, gtt_offset, size);if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); goto unref_out; } }else - obj->ggtt_space->start =I915_GTT_RESERVED ; + obj->ggtt_space.start =I915_GTT_RESERVED ; obj->has_global_gtt_mapping =1 ;Signature · 6 lines
1.8.3.2 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
PATCH 6/6 drm: Optionally create mm blocks from top-to-bottom
In reply to Ben WidawskyFrom: Chris Wilson <[email protected]>
Clients like i915 needs to segregate cache domains within the GTT which can lead to small amounts of fragmentation. By allocating the uncached buffers from the bottom and the cacheable buffers from the top, we can reduce the amount of wasted space and also optimize allocation of the mappable portion of the GTT to only those buffers that require CPU access through the GTT.
v2 by Ben:
Update callers in i915_gem_object_bind_to_gtt() Turn search flags and allocation flags into separate enums Make checkpatch happy where logical/easy
v3: by Ben:
Rebased on create_block removal
Conflicts:
drivers/gpu/drm/drm_mm.c
include/drm/drm_mm.h
--- drivers/gpu/drm/drm_mm.c | 121 +++++++++++++++------------ drivers/gpu/drm/i915/i915_gem.c | 4 +- drivers/gpu/drm/i915/i915_gem_gtt.c | 3 +- drivers/gpu/drm/i915/i915_gem_stolen.c | 5 +- include/drm/drm_mm.h | 148 ++++++++++++++++++++------------- 5 files changed, 166 insertions(+), 115 deletions(-)
drivers/gpu/drm/drm_mm.c +68 -53
@@ -49,7 +49,7 @@#define MM_UNUSED_TARGET 4 -static struct drm_mm_node *drm_mm_kmalloc (struct drm_mm *mm,int atomic) +static struct drm_mm_node *drm_mm_kmalloc (struct drm_mm *mm,bool atomic) {struct drm_mm_node *child;
@@ -105,7 +105,8 @@ EXPORT_SYMBOL(drm_mm_pre_get);static void drm_mm_insert_helper (struct drm_mm_node *hole_node,struct drm_mm_node *node,unsigned long size,unsigned alignment, -unsigned long color) +unsigned long color, +enum drm_mm_allocator_flags flags) {struct drm_mm *mm = hole_node->mm;unsigned long hole_start =drm_mm_hole_node_start (hole_node);
@@ -118,12 +119,22 @@ static void drm_mm_insert_helper(struct drm_mm_node *hole_node,if (mm->color_adjust) mm->color_adjust (hole_node, color, &adj_start, &adj_end); +if (flags &DRM_MM_CREATE_TOP ) + adj_start = adj_end - size; +if (alignment) {unsigned tmp = adj_start % alignment; -if (tmp) - adj_start += alignment - tmp; +if (tmp) { +if (flags &DRM_MM_CREATE_TOP ) + adj_start -= tmp; +else + adj_start += alignment - tmp; + } } +BUG_ON (adj_start < hole_start); +BUG_ON (adj_end > hole_end); +if (adj_start == hole_start) { hole_node->hole_follows =0 ;list_del (&hole_node->hole_stack);
@@ -148,7 +159,8 @@ static void drm_mm_insert_helper(struct drm_mm_node *hole_node, }int drm_mm_create_block (struct drm_mm *mm,struct drm_mm_node *node, -unsigned long start,unsigned long size) +unsigned long start,unsigned long size, +enum drm_mm_allocator_flags flags) {struct drm_mm_node *hole;unsigned long end = start + size;
@@ -190,15 +202,15 @@ struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *hole_node,unsigned long size, unsigned alignment, unsigned long color, - int atomic) + enum drm_mm_allocator_flags flags) { struct drm_mm_node *node; - node =drm_mm_kmalloc (hole_node->mm, atomic); + node =drm_mm_kmalloc (hole_node->mm, flags &DRM_MM_CREATE_ATOMIC );if (unlikely (node ==NULL ))return NULL ; -drm_mm_insert_helper (hole_node, node, size, alignment, color); +drm_mm_insert_helper (hole_node, node, size, alignment, color, flags);return node; }
@@ -211,32 +223,28 @@ EXPORT_SYMBOL(drm_mm_get_block_generic);*/ int drm_mm_insert_node_generic (struct drm_mm *mm,struct drm_mm_node *node,unsigned long size,unsigned alignment, -unsigned long color) +unsigned long color, +enum drm_mm_allocator_flags aflags, +enum drm_mm_search_flags sflags) {struct drm_mm_node *hole_node; hole_node =drm_mm_search_free_generic (mm, size, alignment, - color,0 ); + color, sflags);if (!hole_node)return -ENOSPC ; -drm_mm_insert_helper (hole_node, node, size, alignment, color); +drm_mm_insert_helper (hole_node, node, size, alignment, color, aflags);return 0 ; }EXPORT_SYMBOL (drm_mm_insert_node_generic); -int drm_mm_insert_node (struct drm_mm *mm,struct drm_mm_node *node, -unsigned long size,unsigned alignment) -{ -return drm_mm_insert_node_generic (mm, node, size, alignment,0 ); -} -EXPORT_SYMBOL (drm_mm_insert_node); -static void drm_mm_insert_helper_range (struct drm_mm_node *hole_node,struct drm_mm_node *node,unsigned long size,unsigned alignment,unsigned long color, -unsigned long start,unsigned long end) +unsigned long start,unsigned long end, +enum drm_mm_search_flags flags) {struct drm_mm *mm = hole_node->mm;unsigned long hole_start =drm_mm_hole_node_start (hole_node);
@@ -251,13 +259,20 @@ static void drm_mm_insert_helper_range(struct drm_mm_node *hole_node,if (adj_end > end) adj_end = end; +if (flags &DRM_MM_CREATE_TOP ) + adj_start = adj_end - size; +if (mm->color_adjust) mm->color_adjust (hole_node, color, &adj_start, &adj_end);if (alignment) {unsigned tmp = adj_start % alignment; -if (tmp) - adj_start += alignment - tmp; +if (tmp) { +if (flags &DRM_MM_CREATE_TOP ) + adj_start -= tmp; +else + adj_start += alignment - tmp; + } }if (adj_start == hole_start) {
@@ -274,6 +289,8 @@ static void drm_mm_insert_helper_range(struct drm_mm_node *hole_node,INIT_LIST_HEAD (&node->hole_stack);list_add (&node->node_list, &hole_node->node_list); +BUG_ON (node->start < start); +BUG_ON (node->start < adj_start);BUG_ON (node->start + node->size > adj_end);BUG_ON (node->start + node->size > end);
@@ -284,22 +301,23 @@ static void drm_mm_insert_helper_range(struct drm_mm_node *hole_node, } } -struct drm_mm_node *drm_mm_get_block_range_generic (struct drm_mm_node *hole_node, -unsigned long size, -unsigned alignment, -unsigned long color, -unsigned long start, -unsigned long end, -int atomic) +struct drm_mm_node * +drm_mm_get_block_range_generic (struct drm_mm_node *hole_node, +unsigned long size, +unsigned alignment, +unsigned long color, +unsigned long start, +unsigned long end, +enum drm_mm_allocator_flags flags) {struct drm_mm_node *node; - node =drm_mm_kmalloc (hole_node->mm, atomic); + node =drm_mm_kmalloc (hole_node->mm, flags &DRM_MM_CREATE_ATOMIC );if (unlikely (node ==NULL ))return NULL ;drm_mm_insert_helper_range (hole_node, node, size, alignment, color, - start, end); + start, end, flags);return node; }
@@ -312,31 +330,25 @@ EXPORT_SYMBOL(drm_mm_get_block_range_generic);*/ int drm_mm_insert_node_in_range_generic (struct drm_mm *mm,struct drm_mm_node *node,unsigned long size,unsigned alignment,unsigned long color, -unsigned long start,unsigned long end) +unsigned long start,unsigned long end, +enum drm_mm_allocator_flags aflags, +enum drm_mm_search_flags sflags) {struct drm_mm_node *hole_node; hole_node =drm_mm_search_free_in_range_generic (mm, size, alignment, color, - start, end,0 ); + start, end, sflags);if (!hole_node)return -ENOSPC ;drm_mm_insert_helper_range (hole_node, node, size, alignment, color, - start, end); + start, end, aflags);return 0 ; }EXPORT_SYMBOL (drm_mm_insert_node_in_range_generic); -int drm_mm_insert_node_in_range (struct drm_mm *mm,struct drm_mm_node *node, -unsigned long size,unsigned alignment, -unsigned long start,unsigned long end) -{ -return drm_mm_insert_node_in_range_generic (mm, node, size, alignment,0 , start, end); -} -EXPORT_SYMBOL (drm_mm_insert_node_in_range); -/** * Remove a memory node from the allocator. */
@@ -412,7 +424,7 @@ struct drm_mm_node *drm_mm_search_free_generic(const struct drm_mm *mm,unsigned long size, unsigned alignment, unsigned long color, - bool best_match) + enum drm_mm_search_flags flags) { struct drm_mm_node *entry;struct drm_mm_node *best;
@@ -425,7 +437,8 @@ struct drm_mm_node *drm_mm_search_free_generic(const struct drm_mm *mm, best =NULL ; best_size = ~0UL ; -drm_mm_for_each_hole (entry, mm, adj_start, adj_end) { +__drm_mm_for_each_hole (entry, mm, adj_start, adj_end, + flags &DRM_MM_SEARCH_BELOW ) {if (mm->color_adjust) { mm->color_adjust (entry, color, &adj_start, &adj_end);if (adj_end <= adj_start)
@@ -435,7 +448,7 @@ struct drm_mm_node *drm_mm_search_free_generic(const struct drm_mm *mm,if (!check_free_hole (adj_start, adj_end, size, alignment))continue ; -if (!best_match) +if ((flags &DRM_MM_SEARCH_BEST ) ==0 )return entry;if (entry->size < best_size) {
@@ -448,13 +461,14 @@ struct drm_mm_node *drm_mm_search_free_generic(const struct drm_mm *mm, }EXPORT_SYMBOL (drm_mm_search_free_generic); -struct drm_mm_node *drm_mm_search_free_in_range_generic (const struct drm_mm *mm, -unsigned long size, -unsigned alignment, -unsigned long color, -unsigned long start, -unsigned long end, -bool best_match) +struct drm_mm_node * +drm_mm_search_free_in_range_generic (const struct drm_mm *mm, +unsigned long size, +unsigned alignment, +unsigned long color, +unsigned long start, +unsigned long end, +enum drm_mm_search_flags flags) {struct drm_mm_node *entry;struct drm_mm_node *best;
@@ -467,7 +481,8 @@ struct drm_mm_node *drm_mm_search_free_in_range_generic(const struct drm_mm *mm, best =NULL ; best_size = ~0UL ; -drm_mm_for_each_hole (entry, mm, adj_start, adj_end) { +__drm_mm_for_each_hole (entry, mm, adj_start, adj_end, + flags &DRM_MM_SEARCH_BELOW ) {if (adj_start < start) adj_start = start;if (adj_end > end)
@@ -482,7 +497,7 @@ struct drm_mm_node *drm_mm_search_free_in_range_generic(const struct drm_mm *mm,if (!check_free_hole (adj_start, adj_end, size, alignment))continue ; -if (!best_match) +if ((flags &DRM_MM_SEARCH_BEST ) ==0 )return entry;if (entry->size < best_size) {
drivers/gpu/drm/i915/i915_gem.c +3 -1
@@ -3115,7 +3115,9 @@ search_free: ret =drm_mm_insert_node_in_range_generic (&dev_priv->mm.gtt_space, &obj->ggtt_space, size, alignment, - obj->cache_level,0 , gtt_max); + obj->cache_level,0 , gtt_max, +DRM_MM_CREATE_DEFAULT , +DRM_MM_SEARCH_DEFAULT );if (ret) { ret = i915_gem_evict_something(dev, size, alignment, obj->cache_level,
drivers/gpu/drm/i915/i915_gem_gtt.c +2 -1
@@ -637,7 +637,8 @@ void i915_gem_setup_global_gtt(struct drm_device *dev, ret =drm_mm_create_block (&dev_priv->mm.gtt_space, &obj->ggtt_space,i915_gem_obj_ggtt_offset (obj), - obj->base.size); + obj->base.size, +DRM_MM_CREATE_DEFAULT );if (ret)DRM_DEBUG_KMS ("Reservation failed\n" ); obj->has_global_gtt_mapping =1 ;
drivers/gpu/drm/i915/i915_gem_stolen.c +3 -2
@@ -350,7 +350,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,return NULL ; ret =drm_mm_create_block (&dev_priv->mm.stolen, stolen, stolen_offset, - size); + size,DRM_MM_CREATE_DEFAULT );if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen space\n" );kfree (stolen);
@@ -376,7 +376,8 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { ret =drm_mm_create_block (&dev_priv->mm.gtt_space, &obj->ggtt_space, - gtt_offset, size); + gtt_offset, size, +DRM_MM_CREATE_DEFAULT );if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); goto unref_out;
include/drm/drm_mm.h +90 -58
213 lines not on this page · raw message
Signature · 1 line
1.8.3.2
Hi
Quoted 85 lines · Ben Widawsky
For an upcoming patch where we introduce the i915 VMA, it's ideal to have the
drm_mm_nodeas part of the VMA struct (ie. it's pre-allocated). Part of the conversion to VMAs is to kill offobj->gtt_space. Doing this will break a bunch of code, but amongst them are 2 callers ofdrm_mm_create_block(), both related to stolen memory.It also allows us to embed the
drm_mm_nodeinto the object currently which provides a nice transition over to the new code.v2: Reordered to do before ripping out
obj->gtt_offset. Some minor cleanups made available because of reordering.CC <[email protected]>Signed-off-by Ben Widawsky <[email protected]>--- drivers/gpu/drm/drm_mm.c | 16 +++++---------- drivers/gpu/drm/i915/i915_gem_gtt.c | 18 +++++++++++++---- drivers/gpu/drm/i915/i915_gem_stolen.c | 36 +++++++++++++++++++++++----------- include/drm/drm_mm.h | 9 +++++---- 4 files changed, 49 insertions(+), 30 deletions(-)drivers/gpu/drm/drm_mm.c +5
-11@@ -147,12 +147,10 @@ static void drm_mm_insert_helper(struct drm_mm_node *hole_node, } } -struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic) +int drm_mm_create_block (struct drm_mm *mm,struct drm_mm_node *node, +unsigned long start,unsigned long size) { -struct drm_mm_node *hole, *node; +struct drm_mm_node *hole;unsigned long end = start + size;unsigned long hole_start;unsigned long hole_end;@@ -161,10 +159,6 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm,if (hole_start > start || hole_end < end)continue ; - node =drm_mm_kmalloc (mm, atomic); -if (unlikely (node ==NULL )) -return NULL ; - node->start = start; node->size = size; node->mm = mm;@@ -184,11 +178,11 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm, node->hole_follows =1 ; } -return node; +return 0 ; }WARN (1 ,"no hole found for block 0x%lx + 0x%lx\n" , start, size); -return NULL ; +return -ENOSPC ; }EXPORT_SYMBOL (drm_mm_create_block);drivers/gpu/drm/i915/i915_gem_gtt.c +3
-4@@ -629,14 +629,24 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,/* Mark any preallocated objects as occupied */ list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) { +int ret;DRM_DEBUG_KMS ("reserving preallocated space: %x + %zx\n" , obj->gtt_offset, obj->base.size);BUG_ON (obj->gtt_space !=I915_GTT_RESERVED ); - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_offset, - obj->base.size, - false); + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) {
drivers/gpu/drm/i915/i915_gem_gtt.c void i915_gem_setup_global_gtt(struct drm_device *dev, · line 648
Quoted 10 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +10
-0+DRM_ERROR ("Failed to preserve all objects\n" ); +break ; + } + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + obj->gtt_offset, + obj->base.size); +if (ret) { +DRM_DEBUG_KMS ("Reservation failed\n" ); +kfree (obj->gtt_space);
Are you sure you don't need:
obj->gtt_space = NULL;
here?
I am no expert in i915 gem handling, but looking at i915_gem.c I think you might run into bugs if not.
Also, why did you add the "break;" above, but not here? I am confused.
Quoted 48 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +1
-0+ } obj->has_global_gtt_mapping =1 ; }drivers/gpu/drm/i915/i915_gem_stolen.c +13
-8@@ -330,6 +330,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,struct drm_i915_private *dev_priv = dev->dev_private;struct drm_i915_gem_object *obj;struct drm_mm_node *stolen; +int ret;if (dev_priv->mm.stolen_base ==0 )return NULL ;@@ -344,11 +345,15 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,if (WARN_ON (size ==0 ))return NULL ; - stolen =drm_mm_create_block (&dev_priv->mm.stolen, - stolen_offset, size, - false); -if (stolen ==NULL ) { + stolen =kzalloc (sizeof (*stolen),GFP_KERNEL ); +if (!stolen) +return NULL ; + + ret =drm_mm_create_block (&dev_priv->mm.stolen, stolen, stolen_offset, + size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen space\n" ); +kfree (stolen);return NULL ; }@@ -369,13 +374,18 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - gtt_offset, size, - false); -if (obj->gtt_space ==NULL ) { + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); + goto unref_out;
drivers/gpu/drm/i915/i915_gem_stolen.c i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev, · line 388
Quoted 10 lines
drivers/gpu/drm/i915/i915_gem_stolen.c +7
-2+ } + + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + gtt_offset, size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); -drm_gem_object_unreference (&obj->base); -return NULL ; + goto unref_out;
Again:
kfree(obj->gtt_space);
obj->gtt_space = NULL;
Otherwise, if gem-cleanup calls drm_mm_put_block() on an already removed node, you end up with NULL-derefs in drm_mm.c
drivers/gpu/drm/i915/i915_gem_stolen.c i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev, · line 402
Quoted 13 lines
drivers/gpu/drm/i915/i915_gem_stolen.c +4
-1} }else obj->gtt_space =I915_GTT_RESERVED ;@@ -385,8 +395,12 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,list_add_tail (&obj->global_list, &dev_priv->mm.bound_list);list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list); -return obj; + +unref_out: +drm_gem_object_unreference (&obj->base); +drm_mm_put_block (stolen);
"stolen" is already cleared by drm_gem_object_unreference(). So that's a double-free here.
The drm_mm_create_block() change looks good.
Cheers
David
Quoted 38 lines
drivers/gpu/drm/i915/i915_gem_stolen.c +1
-0+return NULL ; }void include/drm/drm_mm.h +5
-5@@ -138,10 +138,10 @@ static inline unsigned long drm_mm_hole_node_end(struct drm_mm_node *hole_node)/* * Basic range manager support (drm_mm.c) */ -extern struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic); +extern int drm_mm_create_block (struct drm_mm *mm, +struct drm_mm_node *node, +unsigned long start, +unsigned long size); extern struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *node, unsigned long size, unsigned alignment,@@ -155,6 +155,7 @@ extern struct drm_mm_node *drm_mm_get_block_range_generic(unsigned long start, unsigned long end, int atomic); +static inline struct drm_mm_node *drm_mm_get_block (struct drm_mm_node *parent,unsigned long size,unsigned alignment) --1.8.3.2
_______________________________________________
dri-devel mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/dri-devel
Quoted 96 lines · David Herrmann
Hi
Quoted 93 lines · Ben Widawsky
For an upcoming patch where we introduce the i915 VMA, it's ideal to have the
drm_mm_nodeas part of the VMA struct (ie. it's pre-allocated). Part of the conversion to VMAs is to kill offobj->gtt_space. Doing this will break a bunch of code, but amongst them are 2 callers ofdrm_mm_create_block(), both related to stolen memory.It also allows us to embed the
drm_mm_nodeinto the object currently which provides a nice transition over to the new code.v2: Reordered to do before ripping out
obj->gtt_offset. Some minor cleanups made available because of reordering.CC <[email protected]>Signed-off-by Ben Widawsky <[email protected]>--- drivers/gpu/drm/drm_mm.c | 16 +++++---------- drivers/gpu/drm/i915/i915_gem_gtt.c | 18 +++++++++++++---- drivers/gpu/drm/i915/i915_gem_stolen.c | 36 +++++++++++++++++++++++----------- include/drm/drm_mm.h | 9 +++++---- 4 files changed, 49 insertions(+), 30 deletions(-)drivers/gpu/drm/drm_mm.c +5
-11@@ -147,12 +147,10 @@ static void drm_mm_insert_helper(struct drm_mm_node *hole_node, } } -struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic) +int drm_mm_create_block (struct drm_mm *mm,struct drm_mm_node *node, +unsigned long start,unsigned long size) { -struct drm_mm_node *hole, *node; +struct drm_mm_node *hole;unsigned long end = start + size;unsigned long hole_start;unsigned long hole_end;@@ -161,10 +159,6 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm,if (hole_start > start || hole_end < end)continue ; - node =drm_mm_kmalloc (mm, atomic); -if (unlikely (node ==NULL )) -return NULL ; - node->start = start; node->size = size; node->mm = mm;@@ -184,11 +178,11 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm, node->hole_follows =1 ; } -return node; +return 0 ; }WARN (1 ,"no hole found for block 0x%lx + 0x%lx\n" , start, size); -return NULL ; +return -ENOSPC ; }EXPORT_SYMBOL (drm_mm_create_block);drivers/gpu/drm/i915/i915_gem_gtt.c +11
-4@@ -629,14 +629,24 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,/* Mark any preallocated objects as occupied */ list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) { +int ret;DRM_DEBUG_KMS ("reserving preallocated space: %x + %zx\n" , obj->gtt_offset, obj->base.size);BUG_ON (obj->gtt_space !=I915_GTT_RESERVED ); - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_offset, - obj->base.size, - false); + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_ERROR ("Failed to preserve all objects\n" ); +break ; + } + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + obj->gtt_offset, + obj->base.size); +if (ret) {
Quoted 8 lines
drivers/gpu/drm/i915/i915_gem_gtt.c void i915_gem_setup_global_gtt(struct drm_device *dev, · line 648
Quoted 2 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +2
-0+DRM_DEBUG_KMS ("Reservation failed\n" ); +kfree (obj->gtt_space);Are you sure you don't need:
obj->gtt_space= NULL;
here?
I am no expert in i915 gem handling, but looking ati915_gem.cI think you might run into bugs if not.
I'm too lazy to actually check, but I believe you're probably right. It's fixed in a later patch where I added the getters and use node_allocated so I don't check obj->gtt_space != NULL anymore; but it would potentially be a painful bisect point.
Thanks for catching it (and the following ones).
Quoted 2 lines
Also, why did you add the "break;" above, but not here? I am confused.
The thought at the time was if kzalloc fails at this point, subsequent kzallocs are really likely to fail also. drm_mm_create_block OTOH is something I won't pretend to inquire about failure recurrence. I agree it looks funny though, so I'll change the break to continue.
Quoted 83 lines
Quoted 49 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +1
-0+ } obj->has_global_gtt_mapping =1 ; }drivers/gpu/drm/i915/i915_gem_stolen.c +13
-8@@ -330,6 +330,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,struct drm_i915_private *dev_priv = dev->dev_private;struct drm_i915_gem_object *obj;struct drm_mm_node *stolen; +int ret;if (dev_priv->mm.stolen_base ==0 )return NULL ;@@ -344,11 +345,15 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,if (WARN_ON (size ==0 ))return NULL ; - stolen =drm_mm_create_block (&dev_priv->mm.stolen, - stolen_offset, size, - false); -if (stolen ==NULL ) { + stolen =kzalloc (sizeof (*stolen),GFP_KERNEL ); +if (!stolen) +return NULL ; + + ret =drm_mm_create_block (&dev_priv->mm.stolen, stolen, stolen_offset, + size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen space\n" ); +kfree (stolen);return NULL ; }@@ -369,13 +374,18 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - gtt_offset, size, - false); -if (obj->gtt_space ==NULL ) { + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); + goto unref_out;drivers/gpu/drm/i915/i915_gem_stolen.c i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev, · line 388
Quoted 10 lines
drivers/gpu/drm/i915/i915_gem_stolen.c +7
-2+ } + + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + gtt_offset, size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); -drm_gem_object_unreference (&obj->base); -return NULL ; + goto unref_out;Again:
kfree(obj->gtt_space);
obj->gtt_space= NULL;
Otherwise, if gem-cleanup callsdrm_mm_put_block()on an already removed node, you end up with NULL-derefs indrm_mm.cdrivers/gpu/drm/i915/i915_gem_stolen.c i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev, · line 402
Quoted 13 lines
drivers/gpu/drm/i915/i915_gem_stolen.c +4
-1} }else obj->gtt_space =I915_GTT_RESERVED ;@@ -385,8 +395,12 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,list_add_tail (&obj->global_list, &dev_priv->mm.bound_list);list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list); -return obj; + +unref_out: +drm_gem_object_unreference (&obj->base); +drm_mm_put_block (stolen);"stolen" is already cleared by
drm_gem_object_unreference(). So that's a double-free here.
Quoted 4 lines
The
drm_mm_create_block()change looks good.
Cheers
David
Thanks for reviewing the i915 parts so thoroughly :D
Quoted 39 lines
Quoted 39 lines
drivers/gpu/drm/i915/i915_gem_stolen.c +1
-0+return NULL ; }void include/drm/drm_mm.h +5
-5@@ -138,10 +138,10 @@ static inline unsigned long drm_mm_hole_node_end(struct drm_mm_node *hole_node)/* * Basic range manager support (drm_mm.c) */ -extern struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic); +extern int drm_mm_create_block (struct drm_mm *mm, +struct drm_mm_node *node, +unsigned long start, +unsigned long size); extern struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *node, unsigned long size, unsigned alignment,@@ -155,6 +155,7 @@ extern struct drm_mm_node *drm_mm_get_block_range_generic(unsigned long start, unsigned long end, int atomic); +static inline struct drm_mm_node *drm_mm_get_block (struct drm_mm_node *parent,unsigned long size,unsigned alignment) --1.8.3.2
_______________________________________________
dri-devel mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/dri-devel
Signature · 1 line
Ben Widawsky, Intel Open Source Technology Center
Hi
Quoted 190 lines · Ben Widawsky
For an upcoming patch where we introduce the i915 VMA, it's ideal to have the
drm_mm_nodeas part of the VMA struct (ie. it's pre-allocated). Part of the conversion to VMAs is to kill offobj->gtt_space. Doing this will break a bunch of code, but amongst them are 2 callers ofdrm_mm_create_block(), both related to stolen memory.It also allows us to embed the
drm_mm_nodeinto the object currently which provides a nice transition over to the new code.v2: Reordered to do before ripping out
obj->gtt_offset. Some minor cleanups made available because of reordering.CC <[email protected]>Signed-off-by Ben Widawsky <[email protected]>--- drivers/gpu/drm/drm_mm.c | 16 +++++---------- drivers/gpu/drm/i915/i915_gem_gtt.c | 18 +++++++++++++---- drivers/gpu/drm/i915/i915_gem_stolen.c | 36 +++++++++++++++++++++++----------- include/drm/drm_mm.h | 9 +++++---- 4 files changed, 49 insertions(+), 30 deletions(-)drivers/gpu/drm/drm_mm.c +5
-11@@ -147,12 +147,10 @@ static void drm_mm_insert_helper(struct drm_mm_node *hole_node, } } -struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic) +int drm_mm_create_block (struct drm_mm *mm,struct drm_mm_node *node, +unsigned long start,unsigned long size) { -struct drm_mm_node *hole, *node; +struct drm_mm_node *hole;unsigned long end = start + size;unsigned long hole_start;unsigned long hole_end;@@ -161,10 +159,6 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm,if (hole_start > start || hole_end < end)continue ; - node =drm_mm_kmalloc (mm, atomic); -if (unlikely (node ==NULL )) -return NULL ; - node->start = start; node->size = size; node->mm = mm;@@ -184,11 +178,11 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm, node->hole_follows =1 ; } -return node; +return 0 ; }WARN (1 ,"no hole found for block 0x%lx + 0x%lx\n" , start, size); -return NULL ; +return -ENOSPC ; }EXPORT_SYMBOL (drm_mm_create_block);drivers/gpu/drm/i915/i915_gem_gtt.c +14
-4@@ -629,14 +629,24 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,/* Mark any preallocated objects as occupied */ list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) { +int ret;DRM_DEBUG_KMS ("reserving preallocated space: %x + %zx\n" , obj->gtt_offset, obj->base.size);BUG_ON (obj->gtt_space !=I915_GTT_RESERVED ); - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_offset, - obj->base.size, - false); + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_ERROR ("Failed to preserve all objects\n" ); +break ; + } + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + obj->gtt_offset, + obj->base.size); +if (ret) { +DRM_DEBUG_KMS ("Reservation failed\n" ); +kfree (obj->gtt_space); + } obj->has_global_gtt_mapping =1 ; }drivers/gpu/drm/i915/i915_gem_stolen.c +25
-11@@ -330,6 +330,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,struct drm_i915_private *dev_priv = dev->dev_private;struct drm_i915_gem_object *obj;struct drm_mm_node *stolen; +int ret;if (dev_priv->mm.stolen_base ==0 )return NULL ;@@ -344,11 +345,15 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,if (WARN_ON (size ==0 ))return NULL ; - stolen =drm_mm_create_block (&dev_priv->mm.stolen, - stolen_offset, size, - false); -if (stolen ==NULL ) { + stolen =kzalloc (sizeof (*stolen),GFP_KERNEL ); +if (!stolen) +return NULL ; + + ret =drm_mm_create_block (&dev_priv->mm.stolen, stolen, stolen_offset, + size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen space\n" ); +kfree (stolen);return NULL ; }@@ -369,13 +374,18 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - gtt_offset, size, - false); -if (obj->gtt_space ==NULL ) { + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); + goto unref_out; + } + + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + gtt_offset, size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); -drm_gem_object_unreference (&obj->base); -return NULL ; + goto unref_out; } }else obj->gtt_space =I915_GTT_RESERVED ;@@ -385,8 +395,12 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,list_add_tail (&obj->global_list, &dev_priv->mm.bound_list);list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list); -return obj; + +unref_out: +drm_gem_object_unreference (&obj->base); +drm_mm_put_block (stolen); +return NULL ; }void include/drm/drm_mm.h +4
-4@@ -138,10 +138,10 @@ static inline unsigned long drm_mm_hole_node_end(struct drm_mm_node *hole_node)/* * Basic range manager support (drm_mm.c) */ -extern struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic); +extern int drm_mm_create_block (struct drm_mm *mm, +struct drm_mm_node *node, +unsigned long start, +unsigned long size); extern struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *node, unsigned long size, unsigned alignment,
include/drm/drm_mm.h extern struct drm_mm_node *drm_mm_get_block_range_generic( · line 158
Quoted 4 lines
include/drm/drm_mm.h +1
-0@@ -155,6 +155,7 @@ extern struct drm_mm_node *drm_mm_get_block_range_generic(unsigned long start, unsigned long end, int atomic); +
Nitpick: This newline doesn't belong in this patch.
Cheers
David
Quoted 10 lines
include/drm/drm_mm.h +0
-1static inline struct drm_mm_node *drm_mm_get_block (struct drm_mm_node *parent,unsigned long size,unsigned alignment) --1.8.3.2
_______________________________________________
dri-devel mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/dri-devel
PATCH v3 1/1 drm: pre allocate node for create_block
In reply to Ben WidawskyFor an upcoming patch where we introduce the i915 VMA, it's ideal to have the drm_mm_node as part of the VMA struct (ie. it's pre-allocated). Part of the conversion to VMAs is to kill off obj->gtt_space. Doing this will break a bunch of code, but amongst them are 2 callers of drm_mm_create_block(), both related to stolen memory.
It also allows us to embed the drm_mm_node into the object currently which provides a nice transition over to the new code.
v2: Reordered to do before ripping out obj->gtt_offset. Some minor cleanups made available because of reordering.
v3: s/continue/break on failed stolen node allocation (David) Set obj->gtt_space on failed node allocation (David) Only unref stolen (fix double free) on failed create_stolen (David) Free node, and NULL it in failed create_stolen (David) Add back accidentally removed newline (David)
--- drivers/gpu/drm/drm_mm.c | 16 +++++---------- drivers/gpu/drm/i915/i915_gem_gtt.c | 20 ++++++++++++++---- drivers/gpu/drm/i915/i915_gem_stolen.c | 37 +++++++++++++++++++++++++--------- include/drm/drm_mm.h | 9 +++++---- 4 files changed, 53 insertions(+), 29 deletions(-)
drivers/gpu/drm/drm_mm.c +5 -11
@@ -147,12 +147,10 @@ static void drm_mm_insert_helper(struct drm_mm_node *hole_node, } } -struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic) +int drm_mm_create_block (struct drm_mm *mm,struct drm_mm_node *node, +unsigned long start,unsigned long size) { -struct drm_mm_node *hole, *node; +struct drm_mm_node *hole;unsigned long end = start + size;unsigned long hole_start;unsigned long hole_end;
@@ -161,10 +159,6 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm,if (hole_start > start || hole_end < end)continue ; - node =drm_mm_kmalloc (mm, atomic); -if (unlikely (node ==NULL )) -return NULL ; - node->start = start; node->size = size; node->mm = mm;
@@ -184,11 +178,11 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm, node->hole_follows =1 ; } -return node; +return 0 ; }WARN (1 ,"no hole found for block 0x%lx + 0x%lx\n" , start, size); -return NULL ; +return -ENOSPC ; }EXPORT_SYMBOL (drm_mm_create_block);
drivers/gpu/drm/i915/i915_gem_gtt.c +16 -4
@@ -629,14 +629,26 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,/* Mark any preallocated objects as occupied */ list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) { +int ret;DRM_DEBUG_KMS ("reserving preallocated space: %x + %zx\n" , obj->gtt_offset, obj->base.size);BUG_ON (obj->gtt_space !=I915_GTT_RESERVED ); - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_offset, - obj->base.size, - false); + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_ERROR ("Failed to preserve object at offset %x\n" , + obj->gtt_offset); +continue ; + } + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + obj->gtt_offset, + obj->base.size); +if (ret) { +DRM_DEBUG_KMS ("Reservation failed\n" ); +kfree (obj->gtt_space); + obj->gtt_space =NULL ; + } obj->has_global_gtt_mapping =1 ; }
drivers/gpu/drm/i915/i915_gem_stolen.c +27 -10
@@ -330,6 +330,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,struct drm_i915_private *dev_priv = dev->dev_private;struct drm_i915_gem_object *obj;struct drm_mm_node *stolen; +int ret;if (dev_priv->mm.stolen_base ==0 )return NULL ;
@@ -344,11 +345,15 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,if (WARN_ON (size ==0 ))return NULL ; - stolen =drm_mm_create_block (&dev_priv->mm.stolen, - stolen_offset, size, - false); -if (stolen ==NULL ) { + stolen =kzalloc (sizeof (*stolen),GFP_KERNEL ); +if (!stolen) +return NULL ; + + ret =drm_mm_create_block (&dev_priv->mm.stolen, stolen, stolen_offset, + size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen space\n" ); +kfree (stolen);return NULL ; }
@@ -369,13 +374,18 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - gtt_offset, size, - false); -if (obj->gtt_space ==NULL ) { + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); + goto unref_out; + } + + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + gtt_offset, size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); -drm_gem_object_unreference (&obj->base); -return NULL ; + goto free_out; } }else obj->gtt_space =I915_GTT_RESERVED ;
@@ -387,6 +397,13 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list);return obj; + +free_out: +kfree (obj->gtt_space); + obj->gtt_space =NULL ; +unref_out: +drm_gem_object_unreference (&obj->base); +return NULL ; }void
include/drm/drm_mm.h +5 -4
@@ -138,10 +138,10 @@ static inline unsigned long drm_mm_hole_node_end(struct drm_mm_node *hole_node)/* * Basic range manager support (drm_mm.c) */ -extern struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic); +extern int drm_mm_create_block (struct drm_mm *mm, +struct drm_mm_node *node, +unsigned long start, +unsigned long size); extern struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *node, unsigned long size, unsigned alignment,
@@ -155,6 +155,7 @@ extern struct drm_mm_node *drm_mm_get_block_range_generic(unsigned long start, unsigned long end, int atomic); +static inline struct drm_mm_node *drm_mm_get_block (struct drm_mm_node *parent,unsigned long size,unsigned alignment)
Signature · 1 line
1.8.3.2
Hi
Quoted 18 lines · Ben Widawsky
For an upcoming patch where we introduce the i915 VMA, it's ideal to have the
drm_mm_nodeas part of the VMA struct (ie. it's pre-allocated). Part of the conversion to VMAs is to kill offobj->gtt_space. Doing this will break a bunch of code, but amongst them are 2 callers ofdrm_mm_create_block(), both related to stolen memory.It also allows us to embed the
drm_mm_nodeinto the object currently which provides a nice transition over to the new code.v2: Reordered to do before ripping out
obj->gtt_offset. Some minor cleanups made available because of reordering.v3: s/continue/break on failed stolen node allocation (David) Set
obj->gtt_spaceon failed node allocation (David) Only unref stolen (fix double free) on failedcreate_stolen(David) Free node, and NULL it in failedcreate_stolen(David) Add back accidentally removed newline (David)
Quoted 3 lines
CC <[email protected]>CC David Herrmann <[email protected]>Signed-off-by Ben Widawsky <[email protected]>
I already suspected that you'd embed drm_mm_node in a follow-up patch but I am not subscribed to intel-gfx so I didn't get the other patches. Looks good now.
Cheers
David
Quoted 188 lines
--- drivers/gpu/drm/drm_mm.c | 16 +++++---------- drivers/gpu/drm/i915/i915_gem_gtt.c | 20 ++++++++++++++---- drivers/gpu/drm/i915/i915_gem_stolen.c | 37 +++++++++++++++++++++++++--------- include/drm/drm_mm.h | 9 +++++---- 4 files changed, 53 insertions(+), 29 deletions(-)drivers/gpu/drm/drm_mm.c +5
-11@@ -147,12 +147,10 @@ static void drm_mm_insert_helper(struct drm_mm_node *hole_node, } } -struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic) +int drm_mm_create_block (struct drm_mm *mm,struct drm_mm_node *node, +unsigned long start,unsigned long size) { -struct drm_mm_node *hole, *node; +struct drm_mm_node *hole;unsigned long end = start + size;unsigned long hole_start;unsigned long hole_end;@@ -161,10 +159,6 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm,if (hole_start > start || hole_end < end)continue ; - node =drm_mm_kmalloc (mm, atomic); -if (unlikely (node ==NULL )) -return NULL ; - node->start = start; node->size = size; node->mm = mm;@@ -184,11 +178,11 @@ struct drm_mm_node *drm_mm_create_block(struct drm_mm *mm, node->hole_follows =1 ; } -return node; +return 0 ; }WARN (1 ,"no hole found for block 0x%lx + 0x%lx\n" , start, size); -return NULL ; +return -ENOSPC ; }EXPORT_SYMBOL (drm_mm_create_block);drivers/gpu/drm/i915/i915_gem_gtt.c +16
-4@@ -629,14 +629,26 @@ void i915_gem_setup_global_gtt(struct drm_device *dev,/* Mark any preallocated objects as occupied */ list_for_each_entry (obj, &dev_priv->mm.bound_list, global_list) { +int ret;DRM_DEBUG_KMS ("reserving preallocated space: %x + %zx\n" , obj->gtt_offset, obj->base.size);BUG_ON (obj->gtt_space !=I915_GTT_RESERVED ); - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - obj->gtt_offset, - obj->base.size, - false); + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_ERROR ("Failed to preserve object at offset %x\n" , + obj->gtt_offset); +continue ; + } + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + obj->gtt_offset, + obj->base.size); +if (ret) { +DRM_DEBUG_KMS ("Reservation failed\n" ); +kfree (obj->gtt_space); + obj->gtt_space =NULL ; + } obj->has_global_gtt_mapping =1 ; }drivers/gpu/drm/i915/i915_gem_stolen.c +27
-10@@ -330,6 +330,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,struct drm_i915_private *dev_priv = dev->dev_private;struct drm_i915_gem_object *obj;struct drm_mm_node *stolen; +int ret;if (dev_priv->mm.stolen_base ==0 )return NULL ;@@ -344,11 +345,15 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,if (WARN_ON (size ==0 ))return NULL ; - stolen =drm_mm_create_block (&dev_priv->mm.stolen, - stolen_offset, size, - false); -if (stolen ==NULL ) { + stolen =kzalloc (sizeof (*stolen),GFP_KERNEL ); +if (!stolen) +return NULL ; + + ret =drm_mm_create_block (&dev_priv->mm.stolen, stolen, stolen_offset, + size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen space\n" ); +kfree (stolen);return NULL ; }@@ -369,13 +374,18 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,* later. */ if (drm_mm_initialized (&dev_priv->mm.gtt_space)) { - obj->gtt_space =drm_mm_create_block (&dev_priv->mm.gtt_space, - gtt_offset, size, - false); -if (obj->gtt_space ==NULL ) { + obj->gtt_space =kzalloc (sizeof (*obj->gtt_space),GFP_KERNEL ); +if (!obj->gtt_space) { +DRM_DEBUG_KMS ("-ENOMEM stolen GTT space\n" ); + goto unref_out; + } + + ret =drm_mm_create_block (&dev_priv->mm.gtt_space, + obj->gtt_space, + gtt_offset, size); +if (ret) {DRM_DEBUG_KMS ("failed to allocate stolen GTT space\n" ); -drm_gem_object_unreference (&obj->base); -return NULL ; + goto free_out; } }else obj->gtt_space =I915_GTT_RESERVED ;@@ -387,6 +397,13 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,list_add_tail (&obj->mm_list, &dev_priv->mm.inactive_list);return obj; + +free_out: +kfree (obj->gtt_space); + obj->gtt_space =NULL ; +unref_out: +drm_gem_object_unreference (&obj->base); +return NULL ; }void include/drm/drm_mm.h +5
-5@@ -138,10 +138,10 @@ static inline unsigned long drm_mm_hole_node_end(struct drm_mm_node *hole_node)/* * Basic range manager support (drm_mm.c) */ -extern struct drm_mm_node *drm_mm_create_block (struct drm_mm *mm, -unsigned long start, -unsigned long size, -bool atomic); +extern int drm_mm_create_block (struct drm_mm *mm, +struct drm_mm_node *node, +unsigned long start, +unsigned long size); extern struct drm_mm_node *drm_mm_get_block_generic(struct drm_mm_node *node, unsigned long size, unsigned alignment,@@ -155,6 +155,7 @@ extern struct drm_mm_node *drm_mm_get_block_range_generic(unsigned long start, unsigned long end, int atomic); +static inline struct drm_mm_node *drm_mm_get_block (struct drm_mm_node *parent,unsigned long size,unsigned alignment) --1.8.3.2
Quoted 25 lines · David Herrmann
Hi
Quoted 18 lines · Ben Widawsky
For an upcoming patch where we introduce the i915 VMA, it's ideal to have the
drm_mm_nodeas part of the VMA struct (ie. it's pre-allocated). Part of the conversion to VMAs is to kill offobj->gtt_space. Doing this will break a bunch of code, but amongst them are 2 callers ofdrm_mm_create_block(), both related to stolen memory.It also allows us to embed the
drm_mm_nodeinto the object currently which provides a nice transition over to the new code.v2: Reordered to do before ripping out
obj->gtt_offset. Some minor cleanups made available because of reordering.v3: s/continue/break on failed stolen node allocation (David) Set
obj->gtt_spaceon failed node allocation (David) Only unref stolen (fix double free) on failedcreate_stolen(David) Free node, and NULL it in failedcreate_stolen(David) Add back accidentally removed newline (David)Quoted 3 lines
CC <[email protected]>CC David Herrmann <[email protected]>Signed-off-by Ben Widawsky <[email protected]>
Quoted 5 lines
I already suspected that you'd embed
drm_mm_nodein a follow-up patch but I am not subscribed to intel-gfx so I didn't get the other patches. Looks good now.Reviewed-by David Herrmann <[email protected]>
Ok, I've discussed this a bit with Dave on irc and he's a bit unhappy with the creat_block name. After a bit of chatting with Ben I think we should go with
int drm_mm_reserve_node(struct drm_mm *mm, struct drm_mm_node *node);
The start/size/color arguments would then be passed in through the drm_mm_node argument. Ben asked whether that's not too much poking around in drm_mm_node internals, but imo those three pieces of it are part of the public interface to drm_mm users.
Also, the patch as-is conflicts a bit too badly with my current tree. So can you please apply the little bikeshed, rebase, and then rebase the other patches in this series that I haven't merged on top of this?
Thanks, Daniel
Signature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
Quoted 36 lines · Daniel Vetter
Quoted 25 lines · David Herrmann
Hi
Quoted 18 lines · Ben Widawsky
For an upcoming patch where we introduce the i915 VMA, it's ideal to have the
drm_mm_nodeas part of the VMA struct (ie. it's pre-allocated). Part of the conversion to VMAs is to kill offobj->gtt_space. Doing this will break a bunch of code, but amongst them are 2 callers ofdrm_mm_create_block(), both related to stolen memory.It also allows us to embed the
drm_mm_nodeinto the object currently which provides a nice transition over to the new code.v2: Reordered to do before ripping out
obj->gtt_offset. Some minor cleanups made available because of reordering.v3: s/continue/break on failed stolen node allocation (David) Set
obj->gtt_spaceon failed node allocation (David) Only unref stolen (fix double free) on failedcreate_stolen(David) Free node, and NULL it in failedcreate_stolen(David) Add back accidentally removed newline (David)Quoted 3 lines
CC <[email protected]>CC David Herrmann <[email protected]>Signed-off-by Ben Widawsky <[email protected]>Quoted 5 lines
I already suspected that you'd embed
drm_mm_nodein a follow-up patch but I am not subscribed to intel-gfx so I didn't get the other patches. Looks good now.Reviewed-by David Herrmann <[email protected]>Ok, I've discussed this a bit with Dave on irc and he's a bit unhappy with the
creat_blockname. After a bit of chatting with Ben I think we should go with
Quoted 6 lines
int
drm_mm_reserve_node(struct drm_mm *mm, struct drm_mm_node *node);The start/size/color arguments would then be passed in through the
drm_mm_nodeargument. Ben asked whether that's not too much poking around indrm_mm_nodeinternals, but imo those three pieces of it are part of the public interface todrm_mmusers.
Do you mind if I leave the patch as is, since it's reviewed, and put the rename patch on top? I have a long history of screwing up simple rebases.
Quoted 10 lines
Also, the patch as-is conflicts a bit too badly with my current tree. So can you please apply the little bikeshed, rebase, and then rebase the other patches in this series that I haven't merged on top of this?
Thanks, Daniel
Signature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
Signature · 1 line
Ben Widawsky, Intel Open Source Technology Center