From 702adba22433c175e8429a47760f35ca16caf1cd Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Wed, 17 Nov 2010 12:28:29 +0000 Subject: drm/ttm/radeon/nouveau: Kill the bo lock in favour of a bo device fence_lock The bo lock used only to protect the bo sync object members, and since it is a per bo lock, fencing a buffer list will see a lot of locks and unlocks. Replace it with a per-device lock that protects the sync object members on *all* bos. Reading and setting these members will always be very quick, so the risc of heavy lock contention is microscopic. Note that waiting for sync objects will always take place outside of this lock. The bo device fence lock will eventually be replaced with a seqlock / rcu mechanism so we can determine that a bo is idle under a rcu / read seqlock. However this change will allow us to batch fencing and unreserving of buffers with a minimal amount of locking. Signed-off-by: Thomas Hellstrom Reviewed-by: Jerome Glisse Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_object.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/gpu/drm/radeon/radeon_object.h') diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h index d143702b244a..fd536751f865 100644 --- a/drivers/gpu/drm/radeon/radeon_object.h +++ b/drivers/gpu/drm/radeon/radeon_object.h @@ -126,12 +126,12 @@ static inline int radeon_bo_wait(struct radeon_bo *bo, u32 *mem_type, r = ttm_bo_reserve(&bo->tbo, true, no_wait, false, 0); if (unlikely(r != 0)) return r; - spin_lock(&bo->tbo.lock); + spin_lock(&bo->tbo.bdev->fence_lock); if (mem_type) *mem_type = bo->tbo.mem.mem_type; if (bo->tbo.sync_obj) r = ttm_bo_wait(&bo->tbo, true, true, no_wait); - spin_unlock(&bo->tbo.lock); + spin_unlock(&bo->tbo.bdev->fence_lock); ttm_bo_unreserve(&bo->tbo); return r; } -- cgit v1.2.3-59-g8ed1b From 147666fb3b93b8c484f562da33a37f886ddff768 Mon Sep 17 00:00:00 2001 From: Thomas Hellstrom Date: Wed, 17 Nov 2010 12:38:32 +0000 Subject: drm/radeon: Use the ttm execbuf utilities Rather than re-implementing in the Radeon driver, Use the execbuf / cs / pushbuf utilities that comes with TTM. This comes with an even greater benefit now that many spinlocks have been optimized away... Signed-off-by: Thomas Hellstrom Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon.h | 4 +-- drivers/gpu/drm/radeon/radeon_cs.c | 17 ++++++----- drivers/gpu/drm/radeon/radeon_object.c | 55 +++------------------------------- drivers/gpu/drm/radeon/radeon_object.h | 3 -- 4 files changed, 16 insertions(+), 63 deletions(-) (limited to 'drivers/gpu/drm/radeon/radeon_object.h') diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 3a7095743d44..b1e073b7381f 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -69,6 +69,7 @@ #include #include #include +#include #include "radeon_family.h" #include "radeon_mode.h" @@ -259,13 +260,12 @@ struct radeon_bo { }; struct radeon_bo_list { - struct list_head list; + struct ttm_validate_buffer tv; struct radeon_bo *bo; uint64_t gpu_offset; unsigned rdomain; unsigned wdomain; u32 tiling_flags; - bool reserved; }; /* diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index 6d64a2705f12..35b5eb8fbe2a 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -77,13 +77,13 @@ int radeon_cs_parser_relocs(struct radeon_cs_parser *p) p->relocs_ptr[i] = &p->relocs[i]; p->relocs[i].robj = p->relocs[i].gobj->driver_private; p->relocs[i].lobj.bo = p->relocs[i].robj; - p->relocs[i].lobj.rdomain = r->read_domains; p->relocs[i].lobj.wdomain = r->write_domain; + p->relocs[i].lobj.rdomain = r->read_domains; + p->relocs[i].lobj.tv.bo = &p->relocs[i].robj->tbo; p->relocs[i].handle = r->handle; p->relocs[i].flags = r->flags; - INIT_LIST_HEAD(&p->relocs[i].lobj.list); radeon_bo_list_add_object(&p->relocs[i].lobj, - &p->validated); + &p->validated); } } return radeon_bo_list_validate(&p->validated); @@ -189,10 +189,13 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error) { unsigned i; - if (!error && parser->ib) { - radeon_bo_list_fence(&parser->validated, parser->ib->fence); - } - radeon_bo_list_unreserve(&parser->validated); + + if (!error && parser->ib) + ttm_eu_fence_buffer_objects(&parser->validated, + parser->ib->fence); + else + ttm_eu_backoff_reservation(&parser->validated); + if (parser->relocs != NULL) { for (i = 0; i < parser->nrelocs; i++) { if (parser->relocs[i].gobj) diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c index e939cb6a91cc..a8594d289bcf 100644 --- a/drivers/gpu/drm/radeon/radeon_object.c +++ b/drivers/gpu/drm/radeon/radeon_object.c @@ -293,34 +293,9 @@ void radeon_bo_list_add_object(struct radeon_bo_list *lobj, struct list_head *head) { if (lobj->wdomain) { - list_add(&lobj->list, head); + list_add(&lobj->tv.head, head); } else { - list_add_tail(&lobj->list, head); - } -} - -int radeon_bo_list_reserve(struct list_head *head) -{ - struct radeon_bo_list *lobj; - int r; - - list_for_each_entry(lobj, head, list){ - r = radeon_bo_reserve(lobj->bo, false); - if (unlikely(r != 0)) - return r; - lobj->reserved = true; - } - return 0; -} - -void radeon_bo_list_unreserve(struct list_head *head) -{ - struct radeon_bo_list *lobj; - - list_for_each_entry(lobj, head, list) { - /* only unreserve object we successfully reserved */ - if (lobj->reserved && radeon_bo_is_reserved(lobj->bo)) - radeon_bo_unreserve(lobj->bo); + list_add_tail(&lobj->tv.head, head); } } @@ -331,14 +306,11 @@ int radeon_bo_list_validate(struct list_head *head) u32 domain; int r; - list_for_each_entry(lobj, head, list) { - lobj->reserved = false; - } - r = radeon_bo_list_reserve(head); + r = ttm_eu_reserve_buffers(head); if (unlikely(r != 0)) { return r; } - list_for_each_entry(lobj, head, list) { + list_for_each_entry(lobj, head, tv.head) { bo = lobj->bo; if (!bo->pin_count) { domain = lobj->wdomain ? lobj->wdomain : lobj->rdomain; @@ -361,25 +333,6 @@ int radeon_bo_list_validate(struct list_head *head) return 0; } -void radeon_bo_list_fence(struct list_head *head, void *fence) -{ - struct radeon_bo_list *lobj; - struct radeon_bo *bo; - struct radeon_fence *old_fence = NULL; - - list_for_each_entry(lobj, head, list) { - bo = lobj->bo; - spin_lock(&bo->tbo.bdev->fence_lock); - old_fence = (struct radeon_fence *)bo->tbo.sync_obj; - bo->tbo.sync_obj = radeon_fence_ref(fence); - bo->tbo.sync_obj_arg = NULL; - spin_unlock(&bo->tbo.bdev->fence_lock); - if (old_fence) { - radeon_fence_unref(&old_fence); - } - } -} - int radeon_bo_fbdev_mmap(struct radeon_bo *bo, struct vm_area_struct *vma) { diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h index fd536751f865..22d4c237dea5 100644 --- a/drivers/gpu/drm/radeon/radeon_object.h +++ b/drivers/gpu/drm/radeon/radeon_object.h @@ -152,10 +152,7 @@ extern int radeon_bo_init(struct radeon_device *rdev); extern void radeon_bo_fini(struct radeon_device *rdev); extern void radeon_bo_list_add_object(struct radeon_bo_list *lobj, struct list_head *head); -extern int radeon_bo_list_reserve(struct list_head *head); -extern void radeon_bo_list_unreserve(struct list_head *head); extern int radeon_bo_list_validate(struct list_head *head); -extern void radeon_bo_list_fence(struct list_head *head, void *fence); extern int radeon_bo_fbdev_mmap(struct radeon_bo *bo, struct vm_area_struct *vma); extern int radeon_bo_set_tiling_flags(struct radeon_bo *bo, -- cgit v1.2.3-59-g8ed1b