On Thu, Nov 14, 2024 at 10:30 AM Christian König
<ckoenig.leichtzumerken@xxxxxxxxx> wrote:
>
> Just a straightforward conversion without any optimization.
>
> Smoke tested on actual hardware.
>
> v2: rebase
>
> Signed-off-by: Christian König <christian.koenig@xxxxxxx>
Acked-by: Alex Deucher <alexander.deucher@xxxxxxx>
> ---
> drivers/gpu/drm/radeon/Kconfig | 1 +
> drivers/gpu/drm/radeon/radeon.h | 7 ++--
> drivers/gpu/drm/radeon/radeon_cs.c | 45 +++++++++++++-------------
> drivers/gpu/drm/radeon/radeon_gem.c | 39 ++++++++++++----------
> drivers/gpu/drm/radeon/radeon_object.c | 25 +++++++-------
> drivers/gpu/drm/radeon/radeon_object.h | 2 +-
> drivers/gpu/drm/radeon/radeon_vm.c | 10 +++---
> 7 files changed, 66 insertions(+), 63 deletions(-)
>
> diff --git a/drivers/gpu/drm/radeon/Kconfig b/drivers/gpu/drm/radeon/Kconfig
> index 9c6c74a75778..f51bace9555d 100644
> --- a/drivers/gpu/drm/radeon/Kconfig
> +++ b/drivers/gpu/drm/radeon/Kconfig
> @@ -13,6 +13,7 @@ config DRM_RADEON
> select DRM_TTM
> select DRM_TTM_HELPER
> select FB_IOMEM_HELPERS if DRM_FBDEV_EMULATION
> + select DRM_EXEC
> select SND_HDA_COMPONENT if SND_HDA_CORE
> select POWER_SUPPLY
> select HWMON
> diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
> index fd8a4513025f..8605c074d9f7 100644
> --- a/drivers/gpu/drm/radeon/radeon.h
> +++ b/drivers/gpu/drm/radeon/radeon.h
> @@ -75,8 +75,8 @@
>
> #include <drm/ttm/ttm_bo.h>
> #include <drm/ttm/ttm_placement.h>
> -#include <drm/ttm/ttm_execbuf_util.h>
>
> +#include <drm/drm_exec.h>
> #include <drm/drm_gem.h>
> #include <drm/drm_audio_component.h>
> #include <drm/drm_suballoc.h>
> @@ -457,7 +457,8 @@ struct radeon_mman {
>
> struct radeon_bo_list {
> struct radeon_bo *robj;
> - struct ttm_validate_buffer tv;
> + struct list_head list;
> + bool shared;
> uint64_t gpu_offset;
> unsigned preferred_domains;
> unsigned allowed_domains;
> @@ -1030,6 +1031,7 @@ struct radeon_cs_parser {
> struct radeon_bo_list *vm_bos;
> struct list_head validated;
> unsigned dma_reloc_idx;
> + struct drm_exec exec;
> /* indices of various chunks */
> struct radeon_cs_chunk *chunk_ib;
> struct radeon_cs_chunk *chunk_relocs;
> @@ -1043,7 +1045,6 @@ struct radeon_cs_parser {
> u32 cs_flags;
> u32 ring;
> s32 priority;
> - struct ww_acquire_ctx ticket;
> };
>
> static inline u32 radeon_get_ib_value(struct radeon_cs_parser *p, int idx)
> diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
> index a6700d7278bf..64b26bfeafc9 100644
> --- a/drivers/gpu/drm/radeon/radeon_cs.c
> +++ b/drivers/gpu/drm/radeon/radeon_cs.c
> @@ -182,11 +182,8 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
> }
> }
>
> - p->relocs[i].tv.bo = &p->relocs[i].robj->tbo;
> - p->relocs[i].tv.num_shared = !r->write_domain;
> -
> - radeon_cs_buckets_add(&buckets, &p->relocs[i].tv.head,
> - priority);
> + p->relocs[i].shared = !r->write_domain;
> + radeon_cs_buckets_add(&buckets, &p->relocs[i].list, priority);
> }
>
> radeon_cs_buckets_get_list(&buckets, &p->validated);
> @@ -197,7 +194,7 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p)
> if (need_mmap_lock)
> mmap_read_lock(current->mm);
>
> - r = radeon_bo_list_validate(p->rdev, &p->ticket, &p->validated, p->ring);
> + r = radeon_bo_list_validate(p->rdev, &p->exec, &p->validated, p->ring);
>
> if (need_mmap_lock)
> mmap_read_unlock(current->mm);
> @@ -253,12 +250,11 @@ static int radeon_cs_sync_rings(struct radeon_cs_parser *p)
> struct radeon_bo_list *reloc;
> int r;
>
> - list_for_each_entry(reloc, &p->validated, tv.head) {
> + list_for_each_entry(reloc, &p->validated, list) {
> struct dma_resv *resv;
>
> resv = reloc->robj->tbo.base.resv;
> - r = radeon_sync_resv(p->rdev, &p->ib.sync, resv,
> - reloc->tv.num_shared);
> + r = radeon_sync_resv(p->rdev, &p->ib.sync, resv, reloc->shared);
> if (r)
> return r;
> }
> @@ -276,6 +272,7 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
> s32 priority = 0;
>
> INIT_LIST_HEAD(&p->validated);
> + drm_exec_init(&p->exec, DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
>
> if (!cs->num_chunks) {
> return 0;
> @@ -397,8 +394,8 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
> static int cmp_size_smaller_first(void *priv, const struct list_head *a,
> const struct list_head *b)
> {
> - struct radeon_bo_list *la = list_entry(a, struct radeon_bo_list, tv.head);
> - struct radeon_bo_list *lb = list_entry(b, struct radeon_bo_list, tv.head);
> + struct radeon_bo_list *la = list_entry(a, struct radeon_bo_list, list);
> + struct radeon_bo_list *lb = list_entry(b, struct radeon_bo_list, list);
>
> /* Sort A before B if A is smaller. */
> if (la->robj->tbo.base.size > lb->robj->tbo.base.size)
> @@ -417,11 +414,13 @@ static int cmp_size_smaller_first(void *priv, const struct list_head *a,
> * If error is set than unvalidate buffer, otherwise just free memory
> * used by parsing context.
> **/
> -static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error, bool backoff)
> +static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error)
> {
> unsigned i;
>
> if (!error) {
> + struct radeon_bo_list *reloc;
> +
> /* Sort the buffer list from the smallest to largest buffer,
> * which affects the order of buffers in the LRU list.
> * This assures that the smallest buffers are added first
> @@ -433,15 +432,17 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error, bo
> * per frame under memory pressure.
> */
> list_sort(NULL, &parser->validated, cmp_size_smaller_first);
> -
> - ttm_eu_fence_buffer_objects(&parser->ticket,
> - &parser->validated,
> - &parser->ib.fence->base);
> - } else if (backoff) {
> - ttm_eu_backoff_reservation(&parser->ticket,
> - &parser->validated);
> + list_for_each_entry(reloc, &parser->validated, list) {
> + dma_resv_add_fence(reloc->robj->tbo.base.resv,
> + &parser->ib.fence->base,
> + reloc->shared ?
> + DMA_RESV_USAGE_READ :
> + DMA_RESV_USAGE_WRITE);
> + }
> }
>
> + drm_exec_fini(&parser->exec);
> +
> if (parser->relocs != NULL) {
> for (i = 0; i < parser->nrelocs; i++) {
> struct radeon_bo *bo = parser->relocs[i].robj;
> @@ -693,7 +694,7 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
> r = radeon_cs_parser_init(&parser, data);
> if (r) {
> DRM_ERROR("Failed to initialize parser !\n");
> - radeon_cs_parser_fini(&parser, r, false);
> + radeon_cs_parser_fini(&parser, r);
> up_read(&rdev->exclusive_lock);
> r = radeon_cs_handle_lockup(rdev, r);
> return r;
> @@ -707,7 +708,7 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
> }
>
> if (r) {
> - radeon_cs_parser_fini(&parser, r, false);
> + radeon_cs_parser_fini(&parser, r);
> up_read(&rdev->exclusive_lock);
> r = radeon_cs_handle_lockup(rdev, r);
> return r;
> @@ -724,7 +725,7 @@ int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
> goto out;
> }
> out:
> - radeon_cs_parser_fini(&parser, r, true);
> + radeon_cs_parser_fini(&parser, r);
> up_read(&rdev->exclusive_lock);
> r = radeon_cs_handle_lockup(rdev, r);
> return r;
> diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c
> index bf2d4b16dc2a..f86773f3db20 100644
> --- a/drivers/gpu/drm/radeon/radeon_gem.c
> +++ b/drivers/gpu/drm/radeon/radeon_gem.c
> @@ -605,33 +605,40 @@ int radeon_gem_get_tiling_ioctl(struct drm_device *dev, void *data,
> static void radeon_gem_va_update_vm(struct radeon_device *rdev,
> struct radeon_bo_va *bo_va)
> {
> - struct ttm_validate_buffer tv, *entry;
> - struct radeon_bo_list *vm_bos;
> - struct ww_acquire_ctx ticket;
> + struct radeon_bo_list *vm_bos, *entry;
> struct list_head list;
> + struct drm_exec exec;
> unsigned domain;
> int r;
>
> INIT_LIST_HEAD(&list);
>
> - tv.bo = &bo_va->bo->tbo;
> - tv.num_shared = 1;
> - list_add(&tv.head, &list);
> -
> vm_bos = radeon_vm_get_bos(rdev, bo_va->vm, &list);
> if (!vm_bos)
> return;
>
> - r = ttm_eu_reserve_buffers(&ticket, &list, true, NULL);
> - if (r)
> - goto error_free;
> + drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
> + drm_exec_until_all_locked(&exec) {
> + list_for_each_entry(entry, &list, list) {
> + r = drm_exec_prepare_obj(&exec, &entry->robj->tbo.base,
> + 1);
> + drm_exec_retry_on_contention(&exec);
> + if (unlikely(r))
> + goto error_cleanup;
> + }
>
> - list_for_each_entry(entry, &list, head) {
> - domain = radeon_mem_type_to_domain(entry->bo->resource->mem_type);
> + r = drm_exec_prepare_obj(&exec, &bo_va->bo->tbo.base, 1);
> + drm_exec_retry_on_contention(&exec);
> + if (unlikely(r))
> + goto error_cleanup;
> + }
> +
> + list_for_each_entry(entry, &list, list) {
> + domain = radeon_mem_type_to_domain(entry->robj->tbo.resource->mem_type);
> /* if anything is swapped out don't swap it in here,
> just abort and wait for the next CS */
> if (domain == RADEON_GEM_DOMAIN_CPU)
> - goto error_unreserve;
> + goto error_cleanup;
> }
>
> mutex_lock(&bo_va->vm->mutex);
> @@ -645,10 +652,8 @@ static void radeon_gem_va_update_vm(struct radeon_device *rdev,
> error_unlock:
> mutex_unlock(&bo_va->vm->mutex);
>
> -error_unreserve:
> - ttm_eu_backoff_reservation(&ticket, &list);
> -
> -error_free:
> +error_cleanup:
> + drm_exec_fini(&exec);
> kvfree(vm_bos);
>
> if (r && r != -ERESTARTSYS)
> diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
> index 7672404fdb29..a0fc0801abb0 100644
> --- a/drivers/gpu/drm/radeon/radeon_object.c
> +++ b/drivers/gpu/drm/radeon/radeon_object.c
> @@ -464,23 +464,26 @@ static u64 radeon_bo_get_threshold_for_moves(struct radeon_device *rdev)
> }
>
> int radeon_bo_list_validate(struct radeon_device *rdev,
> - struct ww_acquire_ctx *ticket,
> + struct drm_exec *exec,
> struct list_head *head, int ring)
> {
> struct ttm_operation_ctx ctx = { true, false };
> struct radeon_bo_list *lobj;
> - struct list_head duplicates;
> - int r;
> u64 bytes_moved = 0, initial_bytes_moved;
> u64 bytes_moved_threshold = radeon_bo_get_threshold_for_moves(rdev);
> + int r;
>
> - INIT_LIST_HEAD(&duplicates);
> - r = ttm_eu_reserve_buffers(ticket, head, true, &duplicates);
> - if (unlikely(r != 0)) {
> - return r;
> + drm_exec_until_all_locked(exec) {
> + list_for_each_entry(lobj, head, list) {
> + r = drm_exec_prepare_obj(exec, &lobj->robj->tbo.base,
> + 1);
> + drm_exec_retry_on_contention(exec);
> + if (unlikely(r && r != -EALREADY))
> + return r;
> + }
> }
>
> - list_for_each_entry(lobj, head, tv.head) {
> + list_for_each_entry(lobj, head, list) {
> struct radeon_bo *bo = lobj->robj;
> if (!bo->tbo.pin_count) {
> u32 domain = lobj->preferred_domains;
> @@ -519,7 +522,6 @@ int radeon_bo_list_validate(struct radeon_device *rdev,
> domain = lobj->allowed_domains;
> goto retry;
> }
> - ttm_eu_backoff_reservation(ticket, head);
> return r;
> }
> }
> @@ -527,11 +529,6 @@ int radeon_bo_list_validate(struct radeon_device *rdev,
> lobj->tiling_flags = bo->tiling_flags;
> }
>
> - list_for_each_entry(lobj, &duplicates, tv.head) {
> - lobj->gpu_offset = radeon_bo_gpu_offset(lobj->robj);
> - lobj->tiling_flags = lobj->robj->tiling_flags;
> - }
> -
> return 0;
> }
>
> diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h
> index 39cc87a59a9a..d7bbb52db546 100644
> --- a/drivers/gpu/drm/radeon/radeon_object.h
> +++ b/drivers/gpu/drm/radeon/radeon_object.h
> @@ -152,7 +152,7 @@ extern void radeon_bo_force_delete(struct radeon_device *rdev);
> extern int radeon_bo_init(struct radeon_device *rdev);
> extern void radeon_bo_fini(struct radeon_device *rdev);
> extern int radeon_bo_list_validate(struct radeon_device *rdev,
> - struct ww_acquire_ctx *ticket,
> + struct drm_exec *exec,
> struct list_head *head, int ring);
> extern int radeon_bo_set_tiling_flags(struct radeon_bo *bo,
> u32 tiling_flags, u32 pitch);
> diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
> index c38b4d5d6a14..21a5340aefdf 100644
> --- a/drivers/gpu/drm/radeon/radeon_vm.c
> +++ b/drivers/gpu/drm/radeon/radeon_vm.c
> @@ -142,10 +142,9 @@ struct radeon_bo_list *radeon_vm_get_bos(struct radeon_device *rdev,
> list[0].robj = vm->page_directory;
> list[0].preferred_domains = RADEON_GEM_DOMAIN_VRAM;
> list[0].allowed_domains = RADEON_GEM_DOMAIN_VRAM;
> - list[0].tv.bo = &vm->page_directory->tbo;
> - list[0].tv.num_shared = 1;
> + list[0].shared = true;
> list[0].tiling_flags = 0;
> - list_add(&list[0].tv.head, head);
> + list_add(&list[0].list, head);
>
> for (i = 0, idx = 1; i <= vm->max_pde_used; i++) {
> if (!vm->page_tables[i].bo)
> @@ -154,10 +153,9 @@ struct radeon_bo_list *radeon_vm_get_bos(struct radeon_device *rdev,
> list[idx].robj = vm->page_tables[i].bo;
> list[idx].preferred_domains = RADEON_GEM_DOMAIN_VRAM;
> list[idx].allowed_domains = RADEON_GEM_DOMAIN_VRAM;
> - list[idx].tv.bo = &list[idx].robj->tbo;
> - list[idx].tv.num_shared = 1;
> + list[idx].shared = true;
> list[idx].tiling_flags = 0;
> - list_add(&list[idx++].tv.head, head);
> + list_add(&list[idx++].list, head);
> }
>
> return list;
> --
> 2.34.1
>
