Testing entire GGTT privatization preparation series.
Subsystems (3)
cover Testing entire GGTT privatization preparation series.
Do not review, last version had some CI failures, but too old results so need a fresh run to debug.
Maarten Lankhorst (12):
drm/xe: Start using ggtt->start in preparation of balloon removal drm/xe: Convert xe_fb_pin to use a callback for insertion into GGTT drm/xe: Add xe_ggtt_node_addr() to avoid dereferencing xe_ggtt_node drm/xe/display: Avoid dereferencing xe_ggtt_node
drm/xe: Do not dereference ggtt_node in xe_bo.c
drm/xe: Improve xe_gt_sriov_pf_config GGTT handling drm/xe: Privatize xe_ggtt_node
drm/xe: Make xe_ggtt_node offset relative to starting offset drm/xe: Rewrite GGTT VF initialisation
drm/xe: Move struct xe_ggtt to xe_ggtt.c
drm/xe: Make xe_ggtt_node_insert return a node
drm/xe: Remove xe_ggtt_node_allocated
.../gpu/drm/xe/compat-i915-headers/i915_vma.h | 4 +- drivers/gpu/drm/xe/display/xe_fb_pin.c | 105 ++-- drivers/gpu/drm/xe/display/xe_stolen.c | 2 +- drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c | 8 +- drivers/gpu/drm/xe/xe_bo.c | 6 +- drivers/gpu/drm/xe/xe_bo.h | 8 +- drivers/gpu/drm/xe/xe_device_types.h | 2 - drivers/gpu/drm/xe/xe_ggtt.c | 490 ++++++++++-------- drivers/gpu/drm/xe/xe_ggtt.h | 28 +- drivers/gpu/drm/xe/xe_ggtt_types.h | 80 +-- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 53 +- drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 5 +- drivers/gpu/drm/xe/xe_pci.c | 7 + drivers/gpu/drm/xe/xe_tile_sriov_vf.c | 197 +------ drivers/gpu/drm/xe/xe_tile_sriov_vf.h | 4 +- drivers/gpu/drm/xe/xe_tile_sriov_vf_types.h | 4 + 16 files changed, 399 insertions(+), 604 deletions(-)
Signature · 1 line
2.51.0
PATCH 1/12 drm/xe: Start using ggtt->start in preparation of balloon removal
Instead of having ggtt->size point to the end of ggtt, have ggtt->size be the actual size of the GGTT, and introduce ggtt->start to point to the beginning of GGTT.
This will allow a massive cleanup of GGTT in case of SRIOV-VF.
--- drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c | 2 +- drivers/gpu/drm/xe/xe_ggtt.c | 82 +++++++++++++-------- drivers/gpu/drm/xe/xe_ggtt.h | 2 + drivers/gpu/drm/xe/xe_ggtt_types.h | 4 +- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 4 +- 5 files changed, 60 insertions(+), 34 deletions(-)
drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c +1 -1
@@ -67,7 +67,7 @@ static int guc_buf_test_init(struct kunit *test)KUNIT_ASSERT_EQ (test,0 ,xe_ggtt_init_kunit (ggtt,DUT_GGTT_START , -DUT_GGTT_START +DUT_GGTT_SIZE )); +DUT_GGTT_SIZE ));kunit_activate_static_stub (test, xe_managed_bo_create_pin_map, replacement_xe_managed_bo_create_pin_map);
drivers/gpu/drm/xe/xe_ggtt.c +52 -30
151 lines not on this page · raw message
drivers/gpu/drm/xe/xe_ggtt.h +2 -0
@@ -23,6 +23,8 @@ int xe_ggtt_node_insert_balloon_locked(struct xe_ggtt_node *node, u64 start, u64 size);void xe_ggtt_node_remove_balloon_locked (struct xe_ggtt_node *node);void xe_ggtt_shift_nodes_locked (struct xe_ggtt *ggtt,s64 shift); +u64 xe_ggtt_start (struct xe_ggtt *ggtt); +u64 xe_ggtt_size (struct xe_ggtt *ggtt);int xe_ggtt_node_insert (struct xe_ggtt_node *node,u32 size,u32 align); int xe_ggtt_node_insert_locked(struct xe_ggtt_node *node,
drivers/gpu/drm/xe/xe_ggtt_types.h +3 -1
@@ -22,7 +22,9 @@ struct xe_gt;struct xe_ggtt {/** @tile: Back pointer to tile where this GGTT belongs */ struct xe_tile *tile; -/** @size: Total size of this GGTT */ +/** @start: Start offset of GGTT */ +u64 start; +/** @size: Total usable size of this GGTT */ u64 size;#define XE_GGTT_FLAGS_64K BIT(0)
drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +2 -2
@@ -393,8 +393,8 @@ static int pf_push_full_vf_config(struct xe_gt *gt, unsigned int vfid)xe_gt_assert (gt, num_dwords <= max_cfg_dwords);if (vfid ==PFID ) { -u64 ggtt_start =xe_wopcm_size (gt_to_xe (gt)); -u64 ggtt_size =gt_to_tile (gt)->mem.ggtt->size - ggtt_start; +u64 ggtt_start =xe_ggtt_start (gt_to_tile (gt)->mem.ggtt); +u64 ggtt_size =xe_ggtt_size (gt_to_tile (gt)->mem.ggtt);/* plain PF config data will never include a real GGTT region */ xe_gt_assert (gt, !encode_config_ggtt (cfg + num_dwords, config, true));
Signature · 1 line
2.51.0
PATCH 2/12 drm/xe: Convert xe_fb_pin to use a callback for insertion into GGTT
In reply to Maarten LankhorstThe rotation details belong in xe_fb_pin.c, while the operations involving GGTT belong to xe_ggtt.c. As directly locking xe_ggtt etc results in exposing all of xe_ggtt details anyway, create a special function that allocates a ggtt_node, and allow display to populate it using a callback as a compromise.
--- drivers/gpu/drm/xe/display/xe_fb_pin.c | 102 ++++++++++++------------- drivers/gpu/drm/xe/xe_ggtt.c | 90 ++++++++++++++++------ drivers/gpu/drm/xe/xe_ggtt.h | 9 ++- drivers/gpu/drm/xe/xe_ggtt_types.h | 9 ++- 4 files changed, 130 insertions(+), 80 deletions(-)
drivers/gpu/drm/xe/display/xe_fb_pin.c +51 -51
@@ -171,12 +171,13 @@ static int __xe_pin_fb_vma_dpt(const struct intel_framebuffer *fb, }static void -write_ggtt_rotated (struct xe_bo *bo,struct xe_ggtt *ggtt,u32 *ggtt_ofs,u32 bo_ofs, +write_ggtt_rotated (struct xe_ggtt *ggtt,u32 *ggtt_ofs, +u64 pte_flags, +xe_ggtt_set_pte_fn write_pte, +struct xe_bo *bo,u32 bo_ofs,u32 width,u32 height,u32 src_stride,u32 dst_stride) { -struct xe_device *xe =xe_bo_device (bo);u32 column, row; -u64 pte = ggtt->pt_ops->pte_encode_flags (bo, xe->pat.idx[XE_CACHE_NONE ]);for (column =0 ; column < width; column++) {u32 src_idx = src_stride * (height -1 ) + column + bo_ofs;
@@ -184,7 +185,7 @@ write_ggtt_rotated(struct xe_bo *bo, struct xe_ggtt *ggtt, u32 *ggtt_ofs, u32 bofor (row =0 ; row < height; row++) {u64 addr =xe_bo_addr (bo, src_idx *XE_PAGE_SIZE ,XE_PAGE_SIZE ); - ggtt->pt_ops->ggtt_set_pte (ggtt, *ggtt_ofs, pte | addr); +write_pte (ggtt, *ggtt_ofs, pte_flags | addr); *ggtt_ofs +=XE_PAGE_SIZE ; src_idx -= src_stride; }
@@ -194,6 +195,28 @@ write_ggtt_rotated(struct xe_bo *bo, struct xe_ggtt *ggtt, u32 *ggtt_ofs, u32 bo } } +struct fb_rotate_args { +const struct i915_gtt_view *view; +struct xe_bo *bo; +}; + +static void write_ggtt_rotated_node (struct xe_ggtt *ggtt,struct xe_ggtt_node *node, +u64 pte_flags,xe_ggtt_set_pte_fn write_pte,void *data) +{ +struct fb_rotate_args *args = data; +struct xe_bo *bo = args->bo; +const struct intel_rotation_info *rot_info = &args->view->rotated; +u32 ggtt_ofs = node->base.start; + +for (u32 i =0 ; i <ARRAY_SIZE (rot_info->plane); i++) +write_ggtt_rotated (ggtt, &ggtt_ofs, pte_flags, write_pte, + bo, rot_info->plane[i].offset, + rot_info->plane[i].width, + rot_info->plane[i].height, + rot_info->plane[i].src_stride, + rot_info->plane[i].dst_stride); +} + static int __xe_pin_fb_vma_ggtt(const struct intel_framebuffer *fb, const struct i915_gtt_view *view, struct i915_vma *vma,
@@ -204,66 +227,43 @@ static int __xe_pin_fb_vma_ggtt(const struct intel_framebuffer *fb,struct xe_device *xe =to_xe_device (fb->base.dev);struct xe_tile *tile0 =xe_device_get_root_tile (xe);struct xe_ggtt *ggtt = tile0->mem.ggtt; +u64 pte, size;u32 align; -int ret; +int ret =0 ;/* TODO: Consider sharing framebuffer mapping? * embed i915_vma inside intel_framebuffer */ guard (xe_pm_runtime_noresume)(xe); -ACQUIRE (mutex_intr, lock)(&ggtt->lock); - ret =ACQUIRE_ERR (mutex_intr, &lock); -if (ret) -return ret; align =XE_PAGE_SIZE ; -if (xe_bo_is_vram (bo) && ggtt->flags &XE_GGTT_FLAGS_64K ) - align =max_t (u32, align,SZ_64K ); +if (xe_bo_is_vram (bo) && xe->info.vram_flags &XE_VRAM_FLAGS_NEED64K ) + align =max (align,SZ_64K ); +/* Fast case, preallocated GGTT view? */ if (bo->ggtt_node[tile0->id] && view->type ==I915_GTT_VIEW_NORMAL ) { vma->node = bo->ggtt_node[tile0->id]; - }else if (view->type ==I915_GTT_VIEW_NORMAL ) { - vma->node =xe_ggtt_node_init (ggtt); -if (IS_ERR (vma->node)) -return PTR_ERR (vma->node); - - ret =xe_ggtt_node_insert_locked (vma->node,xe_bo_size (bo), align,0 ); -if (ret) { -xe_ggtt_node_fini (vma->node); -return ret; - } - -xe_ggtt_map_bo (ggtt, vma->node, bo, xe->pat.idx[XE_CACHE_NONE ]); - }else { -u32 i, ggtt_ofs; -const struct intel_rotation_info *rot_info = &view->rotated; - -/* display seems to use tiles instead of bytes here, so convert it back.. */ -u32 size =intel_rotation_info_size (rot_info) *XE_PAGE_SIZE ; - - vma->node =xe_ggtt_node_init (ggtt); -if (IS_ERR (vma->node)) { - ret =PTR_ERR (vma->node); -return ret; - } - - ret =xe_ggtt_node_insert_locked (vma->node, size, align,0 ); -if (ret) { -xe_ggtt_node_fini (vma->node); -return ret; - } - - ggtt_ofs = vma->node->base.start; - -for (i =0 ; i <ARRAY_SIZE (rot_info->plane); i++) -write_ggtt_rotated (bo, ggtt, &ggtt_ofs, - rot_info->plane[i].offset, - rot_info->plane[i].width, - rot_info->plane[i].height, - rot_info->plane[i].src_stride, - rot_info->plane[i].dst_stride); +return 0 ; } +/* TODO: Consider sharing framebuffer mapping? +* embed i915_vma inside intel_framebuffer +*/ +if (view->type ==I915_GTT_VIEW_NORMAL ) + size =xe_bo_size (bo); +else +/* display uses tiles instead of bytes here, so convert it back.. */ + size =intel_rotation_info_size (&view->rotated) *XE_PAGE_SIZE ; + + pte =xe_ggtt_encode_pte_flags (ggtt, bo, xe->pat.idx[XE_CACHE_NONE ]); + vma->node =xe_ggtt_node_insert_transform (ggtt, bo, pte, +ALIGN (size, align), align, + view->type ==I915_GTT_VIEW_NORMAL ? +NULL : write_ggtt_rotated_node, + &(struct fb_rotate_args ){view, bo}); +if (IS_ERR (vma->node)) + ret =PTR_ERR (vma->node); +return ret; }
drivers/gpu/drm/xe/xe_ggtt.c +66 -24
132 lines not on this page · raw message
drivers/gpu/drm/xe/xe_ggtt.h +5 -4
@@ -27,13 +27,14 @@ u64 xe_ggtt_start(struct xe_ggtt *ggtt);u64 xe_ggtt_size (struct xe_ggtt *ggtt);int xe_ggtt_node_insert (struct xe_ggtt_node *node,u32 size,u32 align); -int xe_ggtt_node_insert_locked (struct xe_ggtt_node *node, -u32 size,u32 align,u32 mm_flags); +struct xe_ggtt_node * +xe_ggtt_node_insert_transform (struct xe_ggtt *ggtt, +struct xe_bo *bo,u64 pte, +u64 size,u32 align, +xe_ggtt_transform_cb transform,void *arg);void xe_ggtt_node_remove (struct xe_ggtt_node *node,bool invalidate);bool xe_ggtt_node_allocated (const struct xe_ggtt_node *node);size_t xe_ggtt_node_pt_size (const struct xe_ggtt_node *node); -void xe_ggtt_map_bo (struct xe_ggtt *ggtt,struct xe_ggtt_node *node, -struct xe_bo *bo,u16 pat_index);void xe_ggtt_map_bo_unlocked (struct xe_ggtt *ggtt,struct xe_bo *bo);int xe_ggtt_insert_bo (struct xe_ggtt *ggtt,struct xe_bo *bo,struct drm_exec *exec); int xe_ggtt_insert_bo_at(struct xe_ggtt *ggtt, struct xe_bo *bo,
drivers/gpu/drm/xe/xe_ggtt_types.h +8 -1
@@ -71,6 +71,11 @@ struct xe_ggtt_node {bool invalidate_on_remove; }; +typedef void (*xe_ggtt_set_pte_fn )(struct xe_ggtt *ggtt,u64 addr,u64 pte); +typedef void (*xe_ggtt_transform_cb )(struct xe_ggtt *ggtt, +struct xe_ggtt_node *node, +u64 pte_flags, +xe_ggtt_set_pte_fn set_pte,void *arg);/** * struct xe_ggtt_pt_ops - GGTT Page table operations * Which can vary from platform to platform.
@@ -78,8 +83,10 @@ struct xe_ggtt_node {struct xe_ggtt_pt_ops {/** @pte_encode_flags: Encode PTE flags for a given BO */ u64 (*pte_encode_flags)(struct xe_bo *bo,u16 pat_index); +/** @ggtt_set_pte: Directly write into GGTT's PTE */ -void (*ggtt_set_pte)(struct xe_ggtt *ggtt,u64 addr,u64 pte); +xe_ggtt_set_pte_fn ggtt_set_pte; +/** @ggtt_get_pte: Directly read from GGTT's PTE */ u64 (*ggtt_get_pte)(struct xe_ggtt *ggtt,u64 addr); };
Signature · 1 line
2.51.0
PATCH 3/12 drm/xe: Add xe_ggtt_node_addr() to avoid dereferencing xe_ggtt_node
In reply to Maarten LankhorstThis function makes it possible to add an offset that is applied to all xe_ggtt_node's, and hides the internals from all its users.
--- drivers/gpu/drm/xe/xe_bo.h | 8 +++++--- drivers/gpu/drm/xe/xe_ggtt.c | 11 +++++++++++ drivers/gpu/drm/xe/xe_ggtt.h | 2 ++ 3 files changed, 18 insertions(+), 3 deletions(-)
drivers/gpu/drm/xe/xe_bo.h +5 -3
@@ -9,6 +9,7 @@#include <drm/ttm/ttm_tt.h> #include "xe_bo_types.h" +#include "xe_ggtt.h" #include "xe_macros.h" #include "xe_validation.h" #include "xe_vm_types.h"
@@ -252,13 +253,14 @@ static inline u32 __xe_bo_ggtt_addr(struct xe_bo *bo, u8 tile_id) {struct xe_ggtt_node *ggtt_node = bo->ggtt_node[tile_id]; +u64 offset;if (XE_WARN_ON (!ggtt_node))return 0 ; -XE_WARN_ON (ggtt_node->base.size >xe_bo_size (bo)); -XE_WARN_ON (ggtt_node->base.start + ggtt_node->base.size > (1ull <<32 )); -return ggtt_node->base.start; + offset =xe_ggtt_node_addr (ggtt_node); +XE_WARN_ON (offset +xe_bo_size (bo) > (1ull <<32 )); +return offset; } static inline u32
drivers/gpu/drm/xe/xe_ggtt.c +11 -0
@@ -1185,3 +1185,14 @@ u64 xe_ggtt_read_pte(struct xe_ggtt *ggtt, u64 offset) {return ioread64 (ggtt->gsm + (offset /XE_PAGE_SIZE )); } + +/** +* xe_ggtt_node_addr - Get @node offset in GGTT. +* @node: &xe_ggtt_node +* +* Get the GGTT offset for allocated node. +*/ +u64 xe_ggtt_node_addr (const struct xe_ggtt_node *node) +{ +return node->base.start; +}
drivers/gpu/drm/xe/xe_ggtt.h +2 -0
@@ -61,4 +61,6 @@ void xe_ggtt_might_lock(struct xe_ggtt *ggtt);u64 xe_ggtt_encode_pte_flags (struct xe_ggtt *ggtt,struct xe_bo *bo,u16 pat_index);u64 xe_ggtt_read_pte (struct xe_ggtt *ggtt,u64 offset); +u64 xe_ggtt_node_addr (const struct xe_ggtt_node *node); +#endif
Signature · 1 line
2.51.0
PATCH 4/12 drm/xe/display: Avoid dereferencing xe_ggtt_node
In reply to Maarten LankhorstStart using xe_ggtt_node_addr, and avoid comparing the base offset as vma->node is dynamically allocated.
Also sneak in a xe_bo_size() for stolen, too small to put as separate commit.
--- drivers/gpu/drm/xe/compat-i915-headers/i915_vma.h | 4 ++-- drivers/gpu/drm/xe/display/xe_fb_pin.c | 4 ++-- drivers/gpu/drm/xe/display/xe_stolen.c | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-)
drivers/gpu/drm/xe/compat-i915-headers/i915_vma.h +2 -2
@@ -8,7 +8,7 @@#include <uapi/drm/i915_drm.h> -#include "xe_ggtt_types.h" +#include "xe_ggtt.h" #include <linux/refcount.h>
@@ -30,7 +30,7 @@ struct i915_vma {static inline u32 i915_ggtt_offset (const struct i915_vma *vma) { -return vma->node->base.start; +return xe_ggtt_node_addr (vma->node); }#endif
drivers/gpu/drm/xe/display/xe_fb_pin.c +2 -2
@@ -206,7 +206,7 @@ static void write_ggtt_rotated_node(struct xe_ggtt *ggtt, struct xe_ggtt_node *nstruct fb_rotate_args *args = data;struct xe_bo *bo = args->bo;const struct intel_rotation_info *rot_info = &args->view->rotated; -u32 ggtt_ofs = node->base.start; +u32 ggtt_ofs =xe_ggtt_node_addr (node);for (u32 i =0 ; i <ARRAY_SIZE (rot_info->plane); i++) write_ggtt_rotated(ggtt, &ggtt_ofs, pte_flags, write_pte,
@@ -353,7 +353,7 @@ static void __xe_unpin_fb_vma(struct i915_vma *vma)if (vma->dpt)xe_bo_unpin_map_no_vm (vma->dpt);else if (!xe_ggtt_node_allocated (vma->bo->ggtt_node[tile_id]) || - vma->bo->ggtt_node[tile_id]->base.start != vma->node->base.start) + vma->bo->ggtt_node[tile_id] != vma->node)xe_ggtt_node_remove (vma->node, false);ttm_bo_reserve (&vma->bo->ttm, false, false,NULL );
drivers/gpu/drm/xe/display/xe_stolen.c +1 -1
@@ -78,7 +78,7 @@ static u64 xe_stolen_node_address(const struct intel_stolen_node *node)static u64 xe_stolen_node_size (const struct intel_stolen_node *node) { -return node->bo->ttm.base.size; +return xe_bo_size (node->bo); }static struct intel_stolen_node *xe_stolen_node_alloc (struct drm_device *drm)
Signature · 1 line
2.51.0
PATCH 5/12 drm/xe: Do not dereference ggtt_node in xe_bo.c
In reply to Maarten LankhorstA careful inspection of __xe_ggtt_insert_bo_at() shows that the ggtt_node can always be seen as inserted from xe_bo.c due to the way error handling is performed.
The checks are also a little bit too paranoid, since we never create a bo with ggtt_node[id] initialised but not inserted into the GGTT, which can be seen by looking at __xe_ggtt_insert_bo_at()
Additionally, the size of the GGTT is never bigger than 4 GB, so adding a check at that level is incorrect.
--- drivers/gpu/drm/xe/xe_bo.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/gpu/drm/xe/xe_bo.c +3 -3
@@ -1709,7 +1709,7 @@ static void xe_ttm_bo_destroy(struct ttm_buffer_object *ttm_bo)xe_assert (xe,list_empty (&ttm_bo->base.gpuva.list));for_each_tile (tile, xe, id) -if (bo->ggtt_node[id] && bo->ggtt_node[id]->base.size) +if (bo->ggtt_node[id])xe_ggtt_remove_bo (tile->mem.ggtt, bo);#ifdef CONFIG_PROC_FS
@@ -3603,8 +3603,8 @@ void xe_bo_put(struct xe_bo *bo)might_lock (&bo->client->bos_lock);#endif for_each_tile (tile,xe_bo_device (bo), id) -if (bo->ggtt_node[id] && bo->ggtt_node[id]->ggtt) -xe_ggtt_might_lock (bo->ggtt_node[id]->ggtt); +if (bo->ggtt_node[id]) +xe_ggtt_might_lock (tile->mem.ggtt);drm_gem_object_put (&bo->ttm.base); } }
Signature · 1 line
2.51.0
PATCH 6/12 drm/xe: Improve xe_gt_sriov_pf_config GGTT handling
In reply to Maarten LankhorstDo not directly dereference xe_ggtt_node, and add
a function to retrieve the allocated GGTT size.
--- drivers/gpu/drm/xe/xe_ggtt.c | 11 +++++++++++ drivers/gpu/drm/xe/xe_ggtt.h | 1 + drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 15 ++++++++------- 3 files changed, 20 insertions(+), 7 deletions(-)
drivers/gpu/drm/xe/xe_ggtt.c +11 -0
@@ -1196,3 +1196,14 @@ u64 xe_ggtt_node_addr(const struct xe_ggtt_node *node) {return node->base.start; } + +/** +* xe_ggtt_node_size - Get @node allocation size. +* @node: &xe_ggtt_node +* +* Get the allocated node's size. +*/ +u64 xe_ggtt_node_size (const struct xe_ggtt_node *node) +{ +return node->base.size; +}
drivers/gpu/drm/xe/xe_ggtt.h +1 -0
@@ -62,5 +62,6 @@ u64 xe_ggtt_encode_pte_flags(struct xe_ggtt *ggtt, struct xe_bo *bo, u16 pat_indu64 xe_ggtt_read_pte (struct xe_ggtt *ggtt,u64 offset);u64 xe_ggtt_node_addr (const struct xe_ggtt_node *node); +u64 xe_ggtt_node_size (const struct xe_ggtt_node *node);#endif
drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +8 -7
@@ -284,7 +284,7 @@ static u32 encode_config_ggtt(u32 *cfg, const struct xe_gt_sriov_config *config,if (!xe_ggtt_node_allocated (node))return 0 ; -return encode_ggtt (cfg, node->base.start, node->base.size, details); +return encode_ggtt (cfg,xe_ggtt_node_addr (node),xe_ggtt_node_size (node), details); } static u32 encode_config_sched(struct xe_gt *gt, u32 *cfg, u32 n,
@@ -545,9 +545,9 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size)xe_ggtt_assign (node, vfid);xe_gt_sriov_dbg_verbose (gt,"VF%u assigned GGTT %llx-%llx\n" , - vfid, node->base.start, node->base.start + node->base.size -1 ); + vfid,xe_ggtt_node_addr (node),xe_ggtt_node_addr (node) + size -1 ); - err =pf_distribute_config_ggtt (gt->tile, vfid, node->base.start, node->base.size); + err =pf_distribute_config_ggtt (gt->tile, vfid,xe_ggtt_node_addr (node), size);if (unlikely (err)) goto err;
@@ -564,7 +564,7 @@ static u64 pf_get_vf_config_ggtt(struct xe_gt *gt, unsigned int vfid)struct xe_ggtt_node *node = config->ggtt_region;xe_gt_assert (gt,xe_gt_is_main_type (gt)); -return xe_ggtt_node_allocated (node) ? node->base.size :0 ; +return xe_ggtt_node_allocated (node) ?xe_ggtt_node_size (node) :0 ; }/**
@@ -3040,11 +3040,12 @@ int xe_gt_sriov_pf_config_print_ggtt(struct xe_gt *gt, struct drm_printer *p)if (!xe_ggtt_node_allocated (config->ggtt_region))continue ; -string_get_size (config->ggtt_region->base.size,1 ,STRING_UNITS_2 , +string_get_size (xe_ggtt_node_size (config->ggtt_region),1 ,STRING_UNITS_2 , buf,sizeof (buf));drm_printf (p,"VF%u:\t%#0llx-%#llx\t(%s)\n" , - n, config->ggtt_region->base.start, - config->ggtt_region->base.start + config->ggtt_region->base.size -1 , + n,xe_ggtt_node_addr (config->ggtt_region), +xe_ggtt_node_addr (config->ggtt_region) + +xe_ggtt_node_size (config->ggtt_region) -1 , buf); }
Signature · 1 line
2.51.0
PATCH 7/12 drm/xe: Privatize xe_ggtt_node
In reply to Maarten LankhorstNothing requires it any more, make the member private.
--- drivers/gpu/drm/xe/xe_ggtt.c | 18 ++++++++++++++++++ drivers/gpu/drm/xe/xe_ggtt_types.h | 19 +------------------ 2 files changed, 19 insertions(+), 18 deletions(-)
drivers/gpu/drm/xe/xe_ggtt.c +18 -0
@@ -67,6 +67,24 @@* give us the correct placement for free. */ +/** +* struct xe_ggtt_node - A node in GGTT. +* +* This struct needs to be initialized (only-once) with xe_ggtt_node_init() before any node +* insertion, reservation, or 'ballooning'. +* It will, then, be finalized by either xe_ggtt_node_remove() or xe_ggtt_node_deballoon(). +*/ +struct xe_ggtt_node { +/** @ggtt: Back pointer to xe_ggtt where this region will be inserted at */ +struct xe_ggtt *ggtt; +/** @base: A drm_mm_node */ +struct drm_mm_node base; +/** @delayed_removal_work: The work struct for the delayed removal */ +struct work_struct delayed_removal_work; +/** @invalidate_on_remove: If it needs invalidation upon removal */ +bool invalidate_on_remove; +}; +static u64 xelp_ggtt_pte_flags (struct xe_bo *bo,u16 pat_index) {u64 pte =XE_PAGE_PRESENT ;
drivers/gpu/drm/xe/xe_ggtt_types.h +1 -18
@@ -11,6 +11,7 @@#include "xe_pt_types.h" struct xe_bo ; +struct xe_ggtt_node ;struct xe_gt ;/**
@@ -53,24 +54,6 @@ struct xe_ggtt {struct workqueue_struct *wq; }; -/** -* struct xe_ggtt_node - A node in GGTT. -* -* This struct needs to be initialized (only-once) with xe_ggtt_node_init() before any node -* insertion, reservation, or 'ballooning'. -* It will, then, be finalized by either xe_ggtt_node_remove() or xe_ggtt_node_deballoon(). -*/ -struct xe_ggtt_node { -/** @ggtt: Back pointer to xe_ggtt where this region will be inserted at */ -struct xe_ggtt *ggtt; -/** @base: A drm_mm_node */ -struct drm_mm_node base; -/** @delayed_removal_work: The work struct for the delayed removal */ -struct work_struct delayed_removal_work; -/** @invalidate_on_remove: If it needs invalidation upon removal */ -bool invalidate_on_remove; -}; -typedef void (*xe_ggtt_set_pte_fn )(struct xe_ggtt *ggtt,u64 addr,u64 pte); typedef void (*xe_ggtt_transform_cb)(struct xe_ggtt *ggtt, struct xe_ggtt_node *node,
Signature · 1 line
2.51.0
PATCH 8/12 drm/xe: Make xe_ggtt_node offset relative to starting offset
In reply to Maarten LankhorstThis will make node shifting in the next commit a one-liner.
--- drivers/gpu/drm/xe/xe_ggtt.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-)
drivers/gpu/drm/xe/xe_ggtt.c +14 -7
@@ -300,7 +300,7 @@ static void __xe_ggtt_init_early(struct xe_ggtt *ggtt, u64 start, u64 size) { ggtt->start = start; ggtt->size = size; -drm_mm_init (&ggtt->mm, start, size); +drm_mm_init (&ggtt->mm,0 , size - start); }int xe_ggtt_init_kunit (struct xe_ggtt *ggtt,u32 start,u32 size)
@@ -402,7 +402,7 @@ static void xe_ggtt_initial_clear(struct xe_ggtt *ggtt)/* Display may have allocated inside ggtt, so be careful with clearing here */ mutex_lock (&ggtt->lock);drm_mm_for_each_hole (hole, &ggtt->mm, start, end) -xe_ggtt_clear (ggtt, start, end - start); +xe_ggtt_clear (ggtt, ggtt->start + start, end - start);xe_ggtt_invalidate (ggtt);mutex_unlock (&ggtt->lock);
@@ -419,7 +419,7 @@ static void ggtt_node_remove(struct xe_ggtt_node *node)mutex_lock (&ggtt->lock);if (bound) -xe_ggtt_clear (ggtt, node->base.start, node->base.size); +xe_ggtt_clear (ggtt,xe_ggtt_node_addr (node), node->base.size);drm_mm_remove_node (&node->base); node->base.size =0 ;mutex_unlock (&ggtt->lock);
@@ -574,13 +574,13 @@ int xe_ggtt_node_insert_balloon_locked(struct xe_ggtt_node *node, u64 start, u64lockdep_assert_held (&ggtt->lock); node->base.color =0 ; - node->base.start = start; + node->base.start = start - ggtt->start; node->base.size = end - start; err =drm_mm_reserve_node (&ggtt->mm, &node->base);if (xe_tile_WARN (ggtt->tile, err,"Failed to balloon GGTT %#llx-%#llx (%pe)\n" , - node->base.start, node->base.start + node->base.size,ERR_PTR (err))) +xe_ggtt_node_addr (node),xe_ggtt_node_addr (node) + node->base.size,ERR_PTR (err)))return err;xe_ggtt_dump_node (ggtt, &node->base,"balloon" );
@@ -771,7 +771,7 @@ static void xe_ggtt_map_bo(struct xe_ggtt *ggtt, struct xe_ggtt_node *node,if (XE_WARN_ON (!node))return ; - start = node->base.start; + start =xe_ggtt_node_addr (node); end = start +xe_bo_size (bo);if (!xe_bo_is_vram (bo) && !xe_bo_is_stolen (bo)) {
@@ -892,6 +892,13 @@ static int __xe_ggtt_insert_bo_at(struct xe_ggtt *ggtt, struct xe_bo *bo, }mutex_lock (&ggtt->lock); +if (start >= ggtt->start) + start -= ggtt->start; +else + start =0 ; + + end -= ggtt->start; + err =drm_mm_insert_node_in_range (&ggtt->mm, &bo->ggtt_node[tile_id]->base,xe_bo_size (bo), alignment,0 , start, end,0 );if (err) {
@@ -1212,7 +1219,7 @@ u64 xe_ggtt_read_pte(struct xe_ggtt *ggtt, u64 offset)*/ u64 xe_ggtt_node_addr (const struct xe_ggtt_node *node) { -return node->base.start; +return node->base.start +READ_ONCE (node->ggtt->start); }/**
Signature · 1 line
2.51.0
PATCH 9/12 drm/xe: Rewrite GGTT VF initialisation
In reply to Maarten LankhorstThe previous code was using a complicated system with 2 balloons to set GGTT size and adjust GGTT offset. While it works, it's overly complicated.
A better approach is to set the offset and size when initialising GGTT, this removes the need for adding balloons. The resize function only needs readjust ggtt->start to have GGTT at the new offset.
This removes the need to manipulate the internals of xe_ggtt outside of xe_ggtt, and cleans up a lot of now unneeded code.
--- drivers/gpu/drm/xe/xe_device_types.h | 2 - drivers/gpu/drm/xe/xe_ggtt.c | 158 +++------------- drivers/gpu/drm/xe/xe_ggtt.h | 5 +- drivers/gpu/drm/xe/xe_ggtt_types.h | 1 + drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 5 +- drivers/gpu/drm/xe/xe_pci.c | 7 + drivers/gpu/drm/xe/xe_tile_sriov_vf.c | 197 ++------------------ drivers/gpu/drm/xe/xe_tile_sriov_vf.h | 4 +- drivers/gpu/drm/xe/xe_tile_sriov_vf_types.h | 4 + 9 files changed, 59 insertions(+), 324 deletions(-)
drivers/gpu/drm/xe/xe_device_types.h +0 -2
@@ -202,8 +202,6 @@ struct xe_tile {struct xe_lmtt lmtt; } pf;struct { -/** @sriov.vf.ggtt_balloon: GGTT regions excluded from use. */ -struct xe_ggtt_node *ggtt_balloon[2 ];/** @sriov.vf.self_config: VF configuration data */ struct xe_tile_sriov_vf_selfconfig self_config; } vf;
drivers/gpu/drm/xe/xe_ggtt.c +30 -128
204 lines not on this page · raw message
drivers/gpu/drm/xe/xe_ggtt.h +1 -4
@@ -19,10 +19,7 @@ int xe_ggtt_init(struct xe_ggtt *ggtt);struct xe_ggtt_node *xe_ggtt_node_init (struct xe_ggtt *ggtt);void xe_ggtt_node_fini (struct xe_ggtt_node *node); -int xe_ggtt_node_insert_balloon_locked (struct xe_ggtt_node *node, -u64 start,u64 size); -void xe_ggtt_node_remove_balloon_locked (struct xe_ggtt_node *node); -void xe_ggtt_shift_nodes_locked (struct xe_ggtt *ggtt,s64 shift); +void xe_ggtt_shift_nodes (struct xe_ggtt *ggtt,s64 shift);u64 xe_ggtt_start (struct xe_ggtt *ggtt);u64 xe_ggtt_size (struct xe_ggtt *ggtt);
drivers/gpu/drm/xe/xe_ggtt_types.h +1 -0
@@ -59,6 +59,7 @@ typedef void (*xe_ggtt_transform_cb)(struct xe_ggtt *ggtt,struct xe_ggtt_node *node, u64 pte_flags, xe_ggtt_set_pte_fn set_pte, void *arg); +/** * struct xe_ggtt_pt_ops - GGTT Page table operations * Which can vary from platform to platform.
drivers/gpu/drm/xe/xe_gt_sriov_vf.c +2 -3
@@ -489,7 +489,6 @@ u32 xe_gt_sriov_vf_gmdid(struct xe_gt *gt)static int vf_get_ggtt_info (struct xe_gt *gt) {struct xe_tile *tile =gt_to_tile (gt); -struct xe_ggtt *ggtt = tile->mem.ggtt;struct xe_guc *guc = >->uc.guc;u64 start, size, ggtt_size;s64 shift;
@@ -497,7 +496,7 @@ static int vf_get_ggtt_info(struct xe_gt *gt)xe_gt_assert (gt,IS_SRIOV_VF (gt_to_xe (gt))); -guard (mutex)(&ggtt->lock); +guard (mutex)(&tile->sriov.vf.self_config.ggtt_move_mutex); err =guc_action_query_single_klv64 (guc,GUC_KLV_VF_CFG_GGTT_START_KEY , &start);if (unlikely (err))
@@ -527,7 +526,7 @@ static int vf_get_ggtt_info(struct xe_gt *gt)if (shift && shift != start) {xe_gt_sriov_info (gt,"Shifting GGTT base by %lld to 0x%016llx\n" , shift, start); -xe_tile_sriov_vf_fixup_ggtt_nodes_locked (gt_to_tile (gt), shift); +xe_ggtt_shift_nodes (tile->mem.ggtt, shift); }if (xe_sriov_vf_migration_supported (gt_to_xe (gt))) {
drivers/gpu/drm/xe/xe_pci.c +7 -0
@@ -33,6 +33,7 @@#include "xe_pm.h" #include "xe_printk.h" #include "xe_sriov.h" +#include "xe_tile_sriov_vf.h" #include "xe_step.h" #include "xe_survivability_mode.h" #include "xe_tile.h"
@@ -906,6 +907,12 @@ static int xe_info_init(struct xe_device *xe,for_each_tile (tile, xe, id) {int err; +if (IS_SRIOV_VF (xe)) { + err =xe_tile_sriov_vf_init (tile); +if (err) +return err; + } + err =xe_tile_alloc_vram (tile);if (err)return err;
drivers/gpu/drm/xe/xe_tile_sriov_vf.c +13 -184
243 lines not on this page · raw message
drivers/gpu/drm/xe/xe_tile_sriov_vf.h +1 -3
@@ -10,9 +10,7 @@struct xe_tile ; -int xe_tile_sriov_vf_prepare_ggtt (struct xe_tile *tile); -void xe_tile_sriov_vf_deballoon_ggtt_locked (struct xe_tile *tile); -void xe_tile_sriov_vf_fixup_ggtt_nodes_locked (struct xe_tile *tile,s64 shift); +int xe_tile_sriov_vf_init (struct xe_tile *tile);u64 xe_tile_sriov_vf_ggtt (struct xe_tile *tile);void xe_tile_sriov_vf_ggtt_store (struct xe_tile *tile,u64 ggtt_size);u64 xe_tile_sriov_vf_ggtt_base (struct xe_tile *tile);
drivers/gpu/drm/xe/xe_tile_sriov_vf_types.h +4 -0
@@ -7,11 +7,15 @@#define _XE_TILE_SRIOV_VF_TYPES_H_#include <linux/types.h> +#include <linux/mutex.h> /** * struct xe_tile_sriov_vf_selfconfig - VF configuration data. */ struct xe_tile_sriov_vf_selfconfig { +/** @ggtt_move_mutex: Prevents multiple movements from happening in parallel */ +struct mutex ggtt_move_mutex; +/** @ggtt_base: assigned base offset of the GGTT region. */ u64 ggtt_base;/** @ggtt_size: assigned size of the GGTT region. */
Signature · 1 line
2.51.0
PATCH 10/12 drm/xe: Move struct xe_ggtt to xe_ggtt.c
In reply to Maarten LankhorstFrom: Maarten Lankhorst <[email protected]>
No users left outside of xe_ggtt.c, so we can make the struct private.
This prevents us from accidentally touching it before init.
--- drivers/gpu/drm/xe/xe_ggtt.c | 55 +++++++++++++++++++++++++++ drivers/gpu/drm/xe/xe_ggtt.h | 1 + drivers/gpu/drm/xe/xe_ggtt_types.h | 61 +----------------------------- 3 files changed, 58 insertions(+), 59 deletions(-)
drivers/gpu/drm/xe/xe_ggtt.c +55 -0
@@ -85,6 +85,61 @@ struct xe_ggtt_node {bool invalidate_on_remove; }; +/** +* struct xe_ggtt_pt_ops - GGTT Page table operations +* Which can vary from platform to platform. +*/ +struct xe_ggtt_pt_ops { +/** @pte_encode_flags: Encode PTE flags for a given BO */ +u64 (*pte_encode_flags)(struct xe_bo *bo,u16 pat_index); + +/** @ggtt_set_pte: Directly write into GGTT's PTE */ +xe_ggtt_set_pte_fn ggtt_set_pte; + +/** @ggtt_get_pte: Directly read from GGTT's PTE */ +u64 (*ggtt_get_pte)(struct xe_ggtt *ggtt,u64 addr); +}; + +/** +* struct xe_ggtt - Main GGTT struct +* +* In general, each tile can contains its own Global Graphics Translation Table +* (GGTT) instance. +*/ +struct xe_ggtt { +/** @tile: Back pointer to tile where this GGTT belongs */ +struct xe_tile *tile; +/** @start: Start offset of GGTT */ +u64 start; +/** @size: Total usable size of this GGTT */ +u64 size; + +#define XE_GGTT_FLAGS_64K BIT(0) +/** +* @flags: Flags for this GGTT +* Acceptable flags: +* - %XE_GGTT_FLAGS_64K - if PTE size is 64K. Otherwise, regular is 4K. +*/ +unsigned int flags; +/** @scratch: Internal object allocation used as a scratch page */ +struct xe_bo *scratch; +/** @lock: Mutex lock to protect GGTT data */ +struct mutex lock; +/** +* @gsm: The iomem pointer to the actual location of the translation +* table located in the GSM for easy PTE manipulation +*/ +u64 __iomem *gsm; +/** @pt_ops: Page Table operations per platform */ +const struct xe_ggtt_pt_ops *pt_ops; +/** @mm: The memory manager used to manage individual GGTT allocations */ +struct drm_mm mm; +/** @access_count: counts GGTT writes */ +unsigned int access_count; +/** @wq: Dedicated unordered work queue to process node removals */ +struct workqueue_struct *wq; +}; +static u64 xelp_ggtt_pte_flags (struct xe_bo *bo,u16 pat_index) {u64 pte =XE_PAGE_PRESENT ;
drivers/gpu/drm/xe/xe_ggtt.h +1 -0
@@ -9,6 +9,7 @@#include "xe_ggtt_types.h" struct drm_printer ; +struct xe_bo ;struct xe_tile ;struct drm_exec ;
drivers/gpu/drm/xe/xe_ggtt_types.h +2 -59
74 lines not on this page · raw message
Signature · 1 line
2.51.0
PATCH 11/12 drm/xe: Make xe_ggtt_node_insert return a node
This extra step is easier to handle inside xe_ggtt.c and makes xe_ggtt_node_allocated a simple null check instead, as the intermediate state 'allocated but not inserted' is no longer used.
Privatize xe_ggtt_node_fini() as it's no longer used outside of xe_ggtt.c
--- drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c | 6 +- drivers/gpu/drm/xe/xe_ggtt.c | 76 +++++++++------------ drivers/gpu/drm/xe/xe_ggtt.h | 5 +- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 24 +------ 4 files changed, 38 insertions(+), 73 deletions(-)
drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c +1 -5
@@ -38,12 +38,8 @@ static struct xe_bo *replacement_xe_managed_bo_create_pin_map(struct xe_device *if (flags &XE_BO_FLAG_GGTT ) {struct xe_ggtt *ggtt = tile->mem.ggtt; - bo->ggtt_node[tile->id] =xe_ggtt_node_init (ggtt); + bo->ggtt_node[tile->id] =xe_ggtt_node_insert (ggtt,xe_bo_size (bo),SZ_4K );KUNIT_ASSERT_NOT_ERR_OR_NULL (test, bo->ggtt_node[tile->id]); - -KUNIT_ASSERT_EQ (test,0 , -xe_ggtt_node_insert (bo->ggtt_node[tile->id], -xe_bo_size (bo),SZ_4K )); }return bo;
drivers/gpu/drm/xe/xe_ggtt.c +32 -44
106 lines not on this page · raw message
drivers/gpu/drm/xe/xe_ggtt.h +2 -3
@@ -18,13 +18,12 @@ int xe_ggtt_init_early(struct xe_ggtt *ggtt);int xe_ggtt_init_kunit (struct xe_ggtt *ggtt,u32 reserved,u32 size);int xe_ggtt_init (struct xe_ggtt *ggtt); -struct xe_ggtt_node *xe_ggtt_node_init (struct xe_ggtt *ggtt); -void xe_ggtt_node_fini (struct xe_ggtt_node *node);void xe_ggtt_shift_nodes (struct xe_ggtt *ggtt,s64 shift);u64 xe_ggtt_start (struct xe_ggtt *ggtt);u64 xe_ggtt_size (struct xe_ggtt *ggtt); -int xe_ggtt_node_insert (struct xe_ggtt_node *node,u32 size,u32 align); +struct xe_ggtt_node * +xe_ggtt_node_insert (struct xe_ggtt *ggtt,u32 size,u32 align);struct xe_ggtt_node * xe_ggtt_node_insert_transform(struct xe_ggtt *ggtt, struct xe_bo *bo, u64 pte,
drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +3 -21
@@ -484,23 +484,9 @@ static int pf_distribute_config_ggtt(struct xe_tile *tile, unsigned int vfid, u6return err ?: err2; } -static void pf_release_ggtt (struct xe_tile *tile,struct xe_ggtt_node *node) -{ -if (xe_ggtt_node_allocated (node)) { -/* -* explicit GGTT PTE assignment to the PF using xe_ggtt_assign() -* is redundant, as PTE will be implicitly re-assigned to PF by -* the xe_ggtt_clear() called by below xe_ggtt_remove_node(). -*/ -xe_ggtt_node_remove (node, false); - }else { -xe_ggtt_node_fini (node); - } -} -static void pf_release_vf_config_ggtt (struct xe_gt *gt,struct xe_gt_sriov_config *config) { -pf_release_ggtt (gt_to_tile (gt), config->ggtt_region); +xe_ggtt_node_remove (config->ggtt_region, false); config->ggtt_region =NULL ; }
@@ -535,14 +521,10 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size)if (!size)return 0 ; - node =xe_ggtt_node_init (ggtt); + node =xe_ggtt_node_insert (ggtt, size, alignment);if (IS_ERR (node))return PTR_ERR (node); - err =xe_ggtt_node_insert (node, size, alignment); -if (unlikely (err)) - goto err; -xe_ggtt_assign (node, vfid);xe_gt_sriov_dbg_verbose (gt,"VF%u assigned GGTT %llx-%llx\n" , vfid,xe_ggtt_node_addr (node),xe_ggtt_node_addr (node) + size -1 );
@@ -554,7 +536,7 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size) config->ggtt_region = node;return 0 ; err: -pf_release_ggtt (tile, node); +xe_ggtt_node_remove (node, false);return err; }
Signature · 1 line
2.51.0
PATCH 12/12 drm/xe: Remove xe_ggtt_node_allocated
In reply to Maarten LankhorstWith the intermediate state gone, no longer useful. Just check against NULL.
--- drivers/gpu/drm/xe/display/xe_fb_pin.c | 3 +-- drivers/gpu/drm/xe/xe_ggtt.c | 14 -------------- drivers/gpu/drm/xe/xe_ggtt.h | 1 - drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 12 ++++++------ 4 files changed, 7 insertions(+), 23 deletions(-)
drivers/gpu/drm/xe/display/xe_fb_pin.c +1 -2
@@ -352,8 +352,7 @@ static void __xe_unpin_fb_vma(struct i915_vma *vma)if (vma->dpt)xe_bo_unpin_map_no_vm (vma->dpt); -else if (!xe_ggtt_node_allocated (vma->bo->ggtt_node[tile_id]) || - vma->bo->ggtt_node[tile_id] != vma->node) +else if (vma->bo->ggtt_node[tile_id] != vma->node)xe_ggtt_node_remove (vma->node, false);ttm_bo_reserve (&vma->bo->ttm, false, false,NULL );
drivers/gpu/drm/xe/xe_ggtt.c +0 -14
@@ -672,20 +672,6 @@ struct xe_ggtt_node *xe_ggtt_node_insert(struct xe_ggtt *ggtt, u32 size, u32 alireturn node; } -/** -* xe_ggtt_node_allocated - Check if node is allocated in GGTT -* @node: the &xe_ggtt_node to be inspected -* -* Return: True if allocated, False otherwise. -*/ -bool xe_ggtt_node_allocated (const struct xe_ggtt_node *node) -{ -if (!node || !node->ggtt) -return false; - -return drm_mm_node_allocated (&node->base); -} -/** * xe_ggtt_node_pt_size() - Get the size of page table entries needed to map a GGTT node. * @node: the &xe_ggtt_node
drivers/gpu/drm/xe/xe_ggtt.h +0 -1
@@ -30,7 +30,6 @@ xe_ggtt_node_insert_transform(struct xe_ggtt *ggtt, u64 size, u32 align, xe_ggtt_transform_cb transform, void *arg);void xe_ggtt_node_remove (struct xe_ggtt_node *node,bool invalidate); -bool xe_ggtt_node_allocated (const struct xe_ggtt_node *node);size_t xe_ggtt_node_pt_size (const struct xe_ggtt_node *node);void xe_ggtt_map_bo_unlocked (struct xe_ggtt *ggtt,struct xe_bo *bo);int xe_ggtt_insert_bo (struct xe_ggtt *ggtt,struct xe_bo *bo,struct drm_exec *exec);
drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +6 -6
@@ -281,7 +281,7 @@ static u32 encode_config_ggtt(u32 *cfg, const struct xe_gt_sriov_config *config, {struct xe_ggtt_node *node = config->ggtt_region; -if (!xe_ggtt_node_allocated (node)) +if (!node)return 0 ;return encode_ggtt (cfg,xe_ggtt_node_addr (node),xe_ggtt_node_size (node), details);
@@ -505,7 +505,7 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size) size =round_up (size, alignment); -if (xe_ggtt_node_allocated (config->ggtt_region)) { +if (config->ggtt_region) { err =pf_distribute_config_ggtt (tile, vfid,0 ,0 );if (unlikely (err))return err;
@@ -516,7 +516,7 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size)if (unlikely (err))return err; } -xe_gt_assert (gt, !xe_ggtt_node_allocated (config->ggtt_region)); +xe_gt_assert (gt, !config->ggtt_region);if (!size)return 0 ;
@@ -546,7 +546,7 @@ static u64 pf_get_vf_config_ggtt(struct xe_gt *gt, unsigned int vfid)struct xe_ggtt_node *node = config->ggtt_region;xe_gt_assert (gt,xe_gt_is_main_type (gt)); -return xe_ggtt_node_allocated (node) ?xe_ggtt_node_size (node) :0 ; +return node ?xe_ggtt_node_size (node) :0 ; }/**
@@ -2560,7 +2560,7 @@ int xe_gt_sriov_pf_config_release(struct xe_gt *gt, unsigned int vfid, bool forcstatic void pf_sanitize_ggtt (struct xe_ggtt_node *ggtt_region,unsigned int vfid) { -if (xe_ggtt_node_allocated (ggtt_region)) +if (ggtt_region)xe_ggtt_assign (ggtt_region, vfid); }
@@ -3019,7 +3019,7 @@ int xe_gt_sriov_pf_config_print_ggtt(struct xe_gt *gt, struct drm_printer *p)for (n =1 ; n <= total_vfs; n++) { config = >->sriov.pf.vfs[n].config; -if (!xe_ggtt_node_allocated (config->ggtt_region)) +if (!config->ggtt_region)continue ; string_get_size(xe_ggtt_node_size(config->ggtt_region), 1, STRING_UNITS_2,
Signature · 1 line
2.51.0
PATCH 11/12 drm/xe: Make xe_ggtt_node_insert return a node
In reply to Maarten LankhorstThis extra step is easier to handle inside xe_ggtt.c and makes xe_ggtt_node_allocated a simple null check instead, as the intermediate state 'allocated but not inserted' is no longer used.
Privatize xe_ggtt_node_fini() as it's no longer used outside of xe_ggtt.c
--- drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c | 6 +- drivers/gpu/drm/xe/xe_ggtt.c | 76 +++++++++------------ drivers/gpu/drm/xe/xe_ggtt.h | 5 +- drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 24 +------ 4 files changed, 38 insertions(+), 73 deletions(-)
drivers/gpu/drm/xe/tests/xe_guc_buf_kunit.c +1 -5
@@ -38,12 +38,8 @@ static struct xe_bo *replacement_xe_managed_bo_create_pin_map(struct xe_device *if (flags &XE_BO_FLAG_GGTT ) {struct xe_ggtt *ggtt = tile->mem.ggtt; - bo->ggtt_node[tile->id] =xe_ggtt_node_init (ggtt); + bo->ggtt_node[tile->id] =xe_ggtt_node_insert (ggtt,xe_bo_size (bo),SZ_4K );KUNIT_ASSERT_NOT_ERR_OR_NULL (test, bo->ggtt_node[tile->id]); - -KUNIT_ASSERT_EQ (test,0 , -xe_ggtt_node_insert (bo->ggtt_node[tile->id], -xe_bo_size (bo),SZ_4K )); }return bo;
drivers/gpu/drm/xe/xe_ggtt.c +32 -44
106 lines not on this page · raw message
drivers/gpu/drm/xe/xe_ggtt.h +2 -3
@@ -18,13 +18,12 @@ int xe_ggtt_init_early(struct xe_ggtt *ggtt);int xe_ggtt_init_kunit (struct xe_ggtt *ggtt,u32 reserved,u32 size);int xe_ggtt_init (struct xe_ggtt *ggtt); -struct xe_ggtt_node *xe_ggtt_node_init (struct xe_ggtt *ggtt); -void xe_ggtt_node_fini (struct xe_ggtt_node *node);void xe_ggtt_shift_nodes (struct xe_ggtt *ggtt,s64 shift);u64 xe_ggtt_start (struct xe_ggtt *ggtt);u64 xe_ggtt_size (struct xe_ggtt *ggtt); -int xe_ggtt_node_insert (struct xe_ggtt_node *node,u32 size,u32 align); +struct xe_ggtt_node * +xe_ggtt_node_insert (struct xe_ggtt *ggtt,u32 size,u32 align);struct xe_ggtt_node * xe_ggtt_node_insert_transform(struct xe_ggtt *ggtt, struct xe_bo *bo, u64 pte,
drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +3 -21
@@ -484,23 +484,9 @@ static int pf_distribute_config_ggtt(struct xe_tile *tile, unsigned int vfid, u6return err ?: err2; } -static void pf_release_ggtt (struct xe_tile *tile,struct xe_ggtt_node *node) -{ -if (xe_ggtt_node_allocated (node)) { -/* -* explicit GGTT PTE assignment to the PF using xe_ggtt_assign() -* is redundant, as PTE will be implicitly re-assigned to PF by -* the xe_ggtt_clear() called by below xe_ggtt_remove_node(). -*/ -xe_ggtt_node_remove (node, false); - }else { -xe_ggtt_node_fini (node); - } -} -static void pf_release_vf_config_ggtt (struct xe_gt *gt,struct xe_gt_sriov_config *config) { -pf_release_ggtt (gt_to_tile (gt), config->ggtt_region); +xe_ggtt_node_remove (config->ggtt_region, false); config->ggtt_region =NULL ; }
@@ -535,14 +521,10 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size)if (!size)return 0 ; - node =xe_ggtt_node_init (ggtt); + node =xe_ggtt_node_insert (ggtt, size, alignment);if (IS_ERR (node))return PTR_ERR (node); - err =xe_ggtt_node_insert (node, size, alignment); -if (unlikely (err)) - goto err; -xe_ggtt_assign (node, vfid);xe_gt_sriov_dbg_verbose (gt,"VF%u assigned GGTT %llx-%llx\n" , vfid,xe_ggtt_node_addr (node),xe_ggtt_node_addr (node) + size -1 );
@@ -554,7 +536,7 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size) config->ggtt_region = node;return 0 ; err: -pf_release_ggtt (tile, node); +xe_ggtt_node_remove (node, false);return err; }
Signature · 1 line
2.51.0
PATCH 12/12 drm/xe: Remove xe_ggtt_node_allocated
In reply to Maarten LankhorstWith the intermediate state gone, no longer useful. Just check against NULL.
--- drivers/gpu/drm/xe/display/xe_fb_pin.c | 3 +-- drivers/gpu/drm/xe/xe_ggtt.c | 14 -------------- drivers/gpu/drm/xe/xe_ggtt.h | 1 - drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 12 ++++++------ 4 files changed, 7 insertions(+), 23 deletions(-)
drivers/gpu/drm/xe/display/xe_fb_pin.c +1 -2
@@ -352,8 +352,7 @@ static void __xe_unpin_fb_vma(struct i915_vma *vma)if (vma->dpt)xe_bo_unpin_map_no_vm (vma->dpt); -else if (!xe_ggtt_node_allocated (vma->bo->ggtt_node[tile_id]) || - vma->bo->ggtt_node[tile_id] != vma->node) +else if (vma->bo->ggtt_node[tile_id] != vma->node)xe_ggtt_node_remove (vma->node, false);ttm_bo_reserve (&vma->bo->ttm, false, false,NULL );
drivers/gpu/drm/xe/xe_ggtt.c +0 -14
@@ -672,20 +672,6 @@ struct xe_ggtt_node *xe_ggtt_node_insert(struct xe_ggtt *ggtt, u32 size, u32 alireturn node; } -/** -* xe_ggtt_node_allocated - Check if node is allocated in GGTT -* @node: the &xe_ggtt_node to be inspected -* -* Return: True if allocated, False otherwise. -*/ -bool xe_ggtt_node_allocated (const struct xe_ggtt_node *node) -{ -if (!node || !node->ggtt) -return false; - -return drm_mm_node_allocated (&node->base); -} -/** * xe_ggtt_node_pt_size() - Get the size of page table entries needed to map a GGTT node. * @node: the &xe_ggtt_node
drivers/gpu/drm/xe/xe_ggtt.h +0 -1
@@ -30,7 +30,6 @@ xe_ggtt_node_insert_transform(struct xe_ggtt *ggtt, u64 size, u32 align, xe_ggtt_transform_cb transform, void *arg);void xe_ggtt_node_remove (struct xe_ggtt_node *node,bool invalidate); -bool xe_ggtt_node_allocated (const struct xe_ggtt_node *node);size_t xe_ggtt_node_pt_size (const struct xe_ggtt_node *node);void xe_ggtt_map_bo_unlocked (struct xe_ggtt *ggtt,struct xe_bo *bo);int xe_ggtt_insert_bo (struct xe_ggtt *ggtt,struct xe_bo *bo,struct drm_exec *exec);
drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c +6 -6
@@ -281,7 +281,7 @@ static u32 encode_config_ggtt(u32 *cfg, const struct xe_gt_sriov_config *config, {struct xe_ggtt_node *node = config->ggtt_region; -if (!xe_ggtt_node_allocated (node)) +if (!node)return 0 ;return encode_ggtt (cfg,xe_ggtt_node_addr (node),xe_ggtt_node_size (node), details);
@@ -505,7 +505,7 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size) size =round_up (size, alignment); -if (xe_ggtt_node_allocated (config->ggtt_region)) { +if (config->ggtt_region) { err =pf_distribute_config_ggtt (tile, vfid,0 ,0 );if (unlikely (err))return err;
@@ -516,7 +516,7 @@ static int pf_provision_vf_ggtt(struct xe_gt *gt, unsigned int vfid, u64 size)if (unlikely (err))return err; } -xe_gt_assert (gt, !xe_ggtt_node_allocated (config->ggtt_region)); +xe_gt_assert (gt, !config->ggtt_region);if (!size)return 0 ;
@@ -546,7 +546,7 @@ static u64 pf_get_vf_config_ggtt(struct xe_gt *gt, unsigned int vfid)struct xe_ggtt_node *node = config->ggtt_region;xe_gt_assert (gt,xe_gt_is_main_type (gt)); -return xe_ggtt_node_allocated (node) ?xe_ggtt_node_size (node) :0 ; +return node ?xe_ggtt_node_size (node) :0 ; }/**
@@ -2560,7 +2560,7 @@ int xe_gt_sriov_pf_config_release(struct xe_gt *gt, unsigned int vfid, bool forcstatic void pf_sanitize_ggtt (struct xe_ggtt_node *ggtt_region,unsigned int vfid) { -if (xe_ggtt_node_allocated (ggtt_region)) +if (ggtt_region)xe_ggtt_assign (ggtt_region, vfid); }
@@ -3019,7 +3019,7 @@ int xe_gt_sriov_pf_config_print_ggtt(struct xe_gt *gt, struct drm_printer *p)for (n =1 ; n <= total_vfs; n++) { config = >->sriov.pf.vfs[n].config; -if (!xe_ggtt_node_allocated (config->ggtt_region)) +if (!config->ggtt_region)continue ; string_get_size(xe_ggtt_node_size(config->ggtt_region), 1, STRING_UNITS_2,
Signature · 1 line
2.51.0