Re: [PATCH v8 1/3] drm/buddy: Implement tracking clear page feature

2024-03-07 Thread Matthew Auld

On 07/03/2024 12:25, Paneer Selvam, Arunpravin wrote:

Hi Matthew,

On 3/6/2024 11:19 PM, Matthew Auld wrote:

On 04/03/2024 16:32, Arunpravin Paneer Selvam wrote:

- Add tracking clear page feature.

- Driver should enable the DRM_BUDDY_CLEARED flag if it
   successfully clears the blocks in the free path. On the otherhand,
   DRM buddy marks each block as cleared.

- Track the available cleared pages size

- If driver requests cleared memory we prefer cleared memory
   but fallback to uncleared if we can't find the cleared blocks.
   when driver requests uncleared memory we try to use uncleared but
   fallback to cleared memory if necessary.

- When a block gets freed we clear it and mark the freed block as 
cleared,

   when there are buddies which are cleared as well we can merge them.
   Otherwise, we prefer to keep the blocks as separated.

- Add a function to support defragmentation.

v1:
   - Depends on the flag check DRM_BUDDY_CLEARED, enable the block as
 cleared. Else, reset the clear flag for each block in the 
list(Christian)

   - For merging the 2 cleared blocks compare as below,
 drm_buddy_is_clear(block) != drm_buddy_is_clear(buddy)(Christian)
   - Defragment the memory beginning from min_order
 till the required memory space is available.

v2: (Matthew)
   - Add a wrapper drm_buddy_free_list_internal for the freeing of 
blocks

 operation within drm buddy.
   - Write a macro block_incompatible() to allocate the required blocks.
   - Update the xe driver for the drm_buddy_free_list change in 
arguments.

   - add a warning if the two blocks are incompatible on
 defragmentation
   - call full defragmentation in the fini() function
   - place a condition to test if min_order is equal to 0
   - replace the list with safe_reverse() variant as we might
 remove the block from the list.

v3:
   - fix Gitlab user reported lockup issue.
   - Keep DRM_BUDDY_HEADER_CLEAR define sorted(Matthew)
   - modify to pass the root order instead max_order in fini()
 function(Matthew)
   - change bool 1 to true(Matthew)
   - add check if min_block_size is power of 2(Matthew)
   - modify the min_block_size datatype to u64(Matthew)

Signed-off-by: Arunpravin Paneer Selvam 


Signed-off-by: Matthew Auld 
Suggested-by: Christian König 
Suggested-by: Matthew Auld 


Is there a unit test for this? What about maybe something roughly like:

- Pick small random mm_size which is not always power-of-two.
- Allocate and free some random portion of the address space, 
returning those pages as non-dirty. Then do another cycle and randomly 
alternate between requesting dirty and non-dirty pages (not going over 
the limit you freed as non-dirty), putting that into two separate 
lists. Loop over both lists at the end checking that the dirty list is 
indeed all dirty pages and vice versa. Free it all again, keeping the 
dirty/clear status.
- Also try to go over the clear limit for some allocation. The 
allocation should never fail with reasonable page-size.
- Test the defrag/force_merge interface. Clean the mm or create new 
one. Intentionally fragment the address space, by creating two 
alternating lists. Free both lists, one as dirty the other as clean. 
Try to allocate double the previous size with matching min_page_size. 
Should fail. Call force_merge. Should now succeed. Also check that the 
page is always dirty after force_merge. Free the page as dirty, then 
repeat the whole thing, doubling the page-size until you hit max_order.
- Make sure we also call fini() with some part of the address space 
left as non-dirty. Should not trigger any warnings.


I think would be good to consider, but not a blocker or anything. Some 
comments below, otherwise I think looks good.
Yes. It is good to have a unit test for this feature. I will send the 
patch.



---
  drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c  |   6 +-
  drivers/gpu/drm/drm_buddy.c   | 294 +++---
  drivers/gpu/drm/i915/i915_ttm_buddy_manager.c |   6 +-
  drivers/gpu/drm/tests/drm_buddy_test.c    |  18 +-
  drivers/gpu/drm/xe/xe_ttm_vram_mgr.c  |   4 +-
  include/drm/drm_buddy.h   |  22 +-
  6 files changed, 290 insertions(+), 60 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c

index 8db880244324..c0c851409241 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -571,7 +571,7 @@ static int amdgpu_vram_mgr_new(struct 
ttm_resource_manager *man,

  return 0;
    error_free_blocks:
-    drm_buddy_free_list(mm, >blocks);
+    drm_buddy_free_list(mm, >blocks, 0);
  mutex_unlock(>lock);
  error_fini:
  ttm_resource_fini(man, >base);
@@ -604,7 +604,7 @@ static void amdgpu_vram_mgr_del(struct 
ttm_resource_manager *man,

    amdgpu_vram_mgr_do_reserve(man);
  -    drm_buddy_free_list(mm, >blocks);
+    drm_buddy_free_list(mm, >blocks, 0);
  

Re: [PATCH v8 1/3] drm/buddy: Implement tracking clear page feature

2024-03-07 Thread Paneer Selvam, Arunpravin

Hi Matthew,

On 3/6/2024 11:19 PM, Matthew Auld wrote:

On 04/03/2024 16:32, Arunpravin Paneer Selvam wrote:

- Add tracking clear page feature.

- Driver should enable the DRM_BUDDY_CLEARED flag if it
   successfully clears the blocks in the free path. On the otherhand,
   DRM buddy marks each block as cleared.

- Track the available cleared pages size

- If driver requests cleared memory we prefer cleared memory
   but fallback to uncleared if we can't find the cleared blocks.
   when driver requests uncleared memory we try to use uncleared but
   fallback to cleared memory if necessary.

- When a block gets freed we clear it and mark the freed block as 
cleared,

   when there are buddies which are cleared as well we can merge them.
   Otherwise, we prefer to keep the blocks as separated.

- Add a function to support defragmentation.

v1:
   - Depends on the flag check DRM_BUDDY_CLEARED, enable the block as
 cleared. Else, reset the clear flag for each block in the 
list(Christian)

   - For merging the 2 cleared blocks compare as below,
 drm_buddy_is_clear(block) != drm_buddy_is_clear(buddy)(Christian)
   - Defragment the memory beginning from min_order
 till the required memory space is available.

v2: (Matthew)
   - Add a wrapper drm_buddy_free_list_internal for the freeing of 
blocks

 operation within drm buddy.
   - Write a macro block_incompatible() to allocate the required blocks.
   - Update the xe driver for the drm_buddy_free_list change in 
arguments.

   - add a warning if the two blocks are incompatible on
 defragmentation
   - call full defragmentation in the fini() function
   - place a condition to test if min_order is equal to 0
   - replace the list with safe_reverse() variant as we might
 remove the block from the list.

v3:
   - fix Gitlab user reported lockup issue.
   - Keep DRM_BUDDY_HEADER_CLEAR define sorted(Matthew)
   - modify to pass the root order instead max_order in fini()
 function(Matthew)
   - change bool 1 to true(Matthew)
   - add check if min_block_size is power of 2(Matthew)
   - modify the min_block_size datatype to u64(Matthew)

Signed-off-by: Arunpravin Paneer Selvam 


Signed-off-by: Matthew Auld 
Suggested-by: Christian König 
Suggested-by: Matthew Auld 


Is there a unit test for this? What about maybe something roughly like:

- Pick small random mm_size which is not always power-of-two.
- Allocate and free some random portion of the address space, 
returning those pages as non-dirty. Then do another cycle and randomly 
alternate between requesting dirty and non-dirty pages (not going over 
the limit you freed as non-dirty), putting that into two separate 
lists. Loop over both lists at the end checking that the dirty list is 
indeed all dirty pages and vice versa. Free it all again, keeping the 
dirty/clear status.
- Also try to go over the clear limit for some allocation. The 
allocation should never fail with reasonable page-size.
- Test the defrag/force_merge interface. Clean the mm or create new 
one. Intentionally fragment the address space, by creating two 
alternating lists. Free both lists, one as dirty the other as clean. 
Try to allocate double the previous size with matching min_page_size. 
Should fail. Call force_merge. Should now succeed. Also check that the 
page is always dirty after force_merge. Free the page as dirty, then 
repeat the whole thing, doubling the page-size until you hit max_order.
- Make sure we also call fini() with some part of the address space 
left as non-dirty. Should not trigger any warnings.


I think would be good to consider, but not a blocker or anything. Some 
comments below, otherwise I think looks good.

Yes. It is good to have a unit test for this feature. I will send the patch.



---
  drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c  |   6 +-
  drivers/gpu/drm/drm_buddy.c   | 294 +++---
  drivers/gpu/drm/i915/i915_ttm_buddy_manager.c |   6 +-
  drivers/gpu/drm/tests/drm_buddy_test.c    |  18 +-
  drivers/gpu/drm/xe/xe_ttm_vram_mgr.c  |   4 +-
  include/drm/drm_buddy.h   |  22 +-
  6 files changed, 290 insertions(+), 60 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c

index 8db880244324..c0c851409241 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -571,7 +571,7 @@ static int amdgpu_vram_mgr_new(struct 
ttm_resource_manager *man,

  return 0;
    error_free_blocks:
-    drm_buddy_free_list(mm, >blocks);
+    drm_buddy_free_list(mm, >blocks, 0);
  mutex_unlock(>lock);
  error_fini:
  ttm_resource_fini(man, >base);
@@ -604,7 +604,7 @@ static void amdgpu_vram_mgr_del(struct 
ttm_resource_manager *man,

    amdgpu_vram_mgr_do_reserve(man);
  -    drm_buddy_free_list(mm, >blocks);
+    drm_buddy_free_list(mm, >blocks, 0);
  mutex_unlock(>lock);
    atomic64_sub(vis_usage, 

Re: [PATCH v8 1/3] drm/buddy: Implement tracking clear page feature

2024-03-06 Thread Matthew Auld

On 04/03/2024 16:32, Arunpravin Paneer Selvam wrote:

- Add tracking clear page feature.

- Driver should enable the DRM_BUDDY_CLEARED flag if it
   successfully clears the blocks in the free path. On the otherhand,
   DRM buddy marks each block as cleared.

- Track the available cleared pages size

- If driver requests cleared memory we prefer cleared memory
   but fallback to uncleared if we can't find the cleared blocks.
   when driver requests uncleared memory we try to use uncleared but
   fallback to cleared memory if necessary.

- When a block gets freed we clear it and mark the freed block as cleared,
   when there are buddies which are cleared as well we can merge them.
   Otherwise, we prefer to keep the blocks as separated.

- Add a function to support defragmentation.

v1:
   - Depends on the flag check DRM_BUDDY_CLEARED, enable the block as
 cleared. Else, reset the clear flag for each block in the list(Christian)
   - For merging the 2 cleared blocks compare as below,
 drm_buddy_is_clear(block) != drm_buddy_is_clear(buddy)(Christian)
   - Defragment the memory beginning from min_order
 till the required memory space is available.

v2: (Matthew)
   - Add a wrapper drm_buddy_free_list_internal for the freeing of blocks
 operation within drm buddy.
   - Write a macro block_incompatible() to allocate the required blocks.
   - Update the xe driver for the drm_buddy_free_list change in arguments.
   - add a warning if the two blocks are incompatible on
 defragmentation
   - call full defragmentation in the fini() function
   - place a condition to test if min_order is equal to 0
   - replace the list with safe_reverse() variant as we might
 remove the block from the list.

v3:
   - fix Gitlab user reported lockup issue.
   - Keep DRM_BUDDY_HEADER_CLEAR define sorted(Matthew)
   - modify to pass the root order instead max_order in fini()
 function(Matthew)
   - change bool 1 to true(Matthew)
   - add check if min_block_size is power of 2(Matthew)
   - modify the min_block_size datatype to u64(Matthew)

Signed-off-by: Arunpravin Paneer Selvam 
Signed-off-by: Matthew Auld 
Suggested-by: Christian König 
Suggested-by: Matthew Auld 


Is there a unit test for this? What about maybe something roughly like:

- Pick small random mm_size which is not always power-of-two.
- Allocate and free some random portion of the address space, returning 
those pages as non-dirty. Then do another cycle and randomly alternate 
between requesting dirty and non-dirty pages (not going over the limit 
you freed as non-dirty), putting that into two separate lists. Loop over 
both lists at the end checking that the dirty list is indeed all dirty 
pages and vice versa. Free it all again, keeping the dirty/clear status.
- Also try to go over the clear limit for some allocation. The 
allocation should never fail with reasonable page-size.
- Test the defrag/force_merge interface. Clean the mm or create new one. 
Intentionally fragment the address space, by creating two alternating 
lists. Free both lists, one as dirty the other as clean. Try to allocate 
double the previous size with matching min_page_size. Should fail. Call 
force_merge. Should now succeed. Also check that the page is always 
dirty after force_merge. Free the page as dirty, then repeat the whole 
thing, doubling the page-size until you hit max_order.
- Make sure we also call fini() with some part of the address space left 
as non-dirty. Should not trigger any warnings.


I think would be good to consider, but not a blocker or anything. Some 
comments below, otherwise I think looks good.



---
  drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c  |   6 +-
  drivers/gpu/drm/drm_buddy.c   | 294 +++---
  drivers/gpu/drm/i915/i915_ttm_buddy_manager.c |   6 +-
  drivers/gpu/drm/tests/drm_buddy_test.c|  18 +-
  drivers/gpu/drm/xe/xe_ttm_vram_mgr.c  |   4 +-
  include/drm/drm_buddy.h   |  22 +-
  6 files changed, 290 insertions(+), 60 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 8db880244324..c0c851409241 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -571,7 +571,7 @@ static int amdgpu_vram_mgr_new(struct ttm_resource_manager 
*man,
return 0;
  
  error_free_blocks:

-   drm_buddy_free_list(mm, >blocks);
+   drm_buddy_free_list(mm, >blocks, 0);
mutex_unlock(>lock);
  error_fini:
ttm_resource_fini(man, >base);
@@ -604,7 +604,7 @@ static void amdgpu_vram_mgr_del(struct ttm_resource_manager 
*man,
  
  	amdgpu_vram_mgr_do_reserve(man);
  
-	drm_buddy_free_list(mm, >blocks);

+   drm_buddy_free_list(mm, >blocks, 0);
mutex_unlock(>lock);
  
  	atomic64_sub(vis_usage, >vis_usage);

@@ -912,7 +912,7 @@ void amdgpu_vram_mgr_fini(struct amdgpu_device *adev)
kfree(rsv);
  
  	

Re: [PATCH v8 1/3] drm/buddy: Implement tracking clear page feature

2024-03-06 Thread Paneer Selvam, Arunpravin

Hi Matthew,
Ping?

Thanks,
Arun.

On 3/4/2024 10:02 PM, Arunpravin Paneer Selvam wrote:

- Add tracking clear page feature.

- Driver should enable the DRM_BUDDY_CLEARED flag if it
   successfully clears the blocks in the free path. On the otherhand,
   DRM buddy marks each block as cleared.

- Track the available cleared pages size

- If driver requests cleared memory we prefer cleared memory
   but fallback to uncleared if we can't find the cleared blocks.
   when driver requests uncleared memory we try to use uncleared but
   fallback to cleared memory if necessary.

- When a block gets freed we clear it and mark the freed block as cleared,
   when there are buddies which are cleared as well we can merge them.
   Otherwise, we prefer to keep the blocks as separated.

- Add a function to support defragmentation.

v1:
   - Depends on the flag check DRM_BUDDY_CLEARED, enable the block as
 cleared. Else, reset the clear flag for each block in the list(Christian)
   - For merging the 2 cleared blocks compare as below,
 drm_buddy_is_clear(block) != drm_buddy_is_clear(buddy)(Christian)
   - Defragment the memory beginning from min_order
 till the required memory space is available.

v2: (Matthew)
   - Add a wrapper drm_buddy_free_list_internal for the freeing of blocks
 operation within drm buddy.
   - Write a macro block_incompatible() to allocate the required blocks.
   - Update the xe driver for the drm_buddy_free_list change in arguments.
   - add a warning if the two blocks are incompatible on
 defragmentation
   - call full defragmentation in the fini() function
   - place a condition to test if min_order is equal to 0
   - replace the list with safe_reverse() variant as we might
 remove the block from the list.

v3:
   - fix Gitlab user reported lockup issue.
   - Keep DRM_BUDDY_HEADER_CLEAR define sorted(Matthew)
   - modify to pass the root order instead max_order in fini()
 function(Matthew)
   - change bool 1 to true(Matthew)
   - add check if min_block_size is power of 2(Matthew)
   - modify the min_block_size datatype to u64(Matthew)

Signed-off-by: Arunpravin Paneer Selvam 
Signed-off-by: Matthew Auld 
Suggested-by: Christian König 
Suggested-by: Matthew Auld 
---
  drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c  |   6 +-
  drivers/gpu/drm/drm_buddy.c   | 294 +++---
  drivers/gpu/drm/i915/i915_ttm_buddy_manager.c |   6 +-
  drivers/gpu/drm/tests/drm_buddy_test.c|  18 +-
  drivers/gpu/drm/xe/xe_ttm_vram_mgr.c  |   4 +-
  include/drm/drm_buddy.h   |  22 +-
  6 files changed, 290 insertions(+), 60 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 8db880244324..c0c851409241 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -571,7 +571,7 @@ static int amdgpu_vram_mgr_new(struct ttm_resource_manager 
*man,
return 0;
  
  error_free_blocks:

-   drm_buddy_free_list(mm, >blocks);
+   drm_buddy_free_list(mm, >blocks, 0);
mutex_unlock(>lock);
  error_fini:
ttm_resource_fini(man, >base);
@@ -604,7 +604,7 @@ static void amdgpu_vram_mgr_del(struct ttm_resource_manager 
*man,
  
  	amdgpu_vram_mgr_do_reserve(man);
  
-	drm_buddy_free_list(mm, >blocks);

+   drm_buddy_free_list(mm, >blocks, 0);
mutex_unlock(>lock);
  
  	atomic64_sub(vis_usage, >vis_usage);

@@ -912,7 +912,7 @@ void amdgpu_vram_mgr_fini(struct amdgpu_device *adev)
kfree(rsv);
  
  	list_for_each_entry_safe(rsv, temp, >reserved_pages, blocks) {

-   drm_buddy_free_list(>mm, >allocated);
+   drm_buddy_free_list(>mm, >allocated, 0);
kfree(rsv);
}
if (!adev->gmc.is_app_apu)
diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c
index c4222b886db7..40131ed9b0cd 100644
--- a/drivers/gpu/drm/drm_buddy.c
+++ b/drivers/gpu/drm/drm_buddy.c
@@ -57,6 +57,16 @@ static void list_insert_sorted(struct drm_buddy *mm,
__list_add(>link, node->link.prev, >link);
  }
  
+static void clear_reset(struct drm_buddy_block *block)

+{
+   block->header &= ~DRM_BUDDY_HEADER_CLEAR;
+}
+
+static void mark_cleared(struct drm_buddy_block *block)
+{
+   block->header |= DRM_BUDDY_HEADER_CLEAR;
+}
+
  static void mark_allocated(struct drm_buddy_block *block)
  {
block->header &= ~DRM_BUDDY_HEADER_STATE;
@@ -186,11 +196,21 @@ EXPORT_SYMBOL(drm_buddy_init);
   */
  void drm_buddy_fini(struct drm_buddy *mm)
  {
+   u64 root_size, size;
+   unsigned int order;
int i;
  
+	size = mm->size;

+
for (i = 0; i < mm->n_roots; ++i) {
+   order = ilog2(size) - ilog2(mm->chunk_size);
+   root_size = mm->chunk_size << order;
+   drm_buddy_defrag(mm, root_size);
+
WARN_ON(!drm_buddy_block_is_free(mm->roots[i]));

[PATCH v8 1/3] drm/buddy: Implement tracking clear page feature

2024-03-04 Thread Arunpravin Paneer Selvam
- Add tracking clear page feature.

- Driver should enable the DRM_BUDDY_CLEARED flag if it
  successfully clears the blocks in the free path. On the otherhand,
  DRM buddy marks each block as cleared.

- Track the available cleared pages size

- If driver requests cleared memory we prefer cleared memory
  but fallback to uncleared if we can't find the cleared blocks.
  when driver requests uncleared memory we try to use uncleared but
  fallback to cleared memory if necessary.

- When a block gets freed we clear it and mark the freed block as cleared,
  when there are buddies which are cleared as well we can merge them.
  Otherwise, we prefer to keep the blocks as separated.

- Add a function to support defragmentation.

v1:
  - Depends on the flag check DRM_BUDDY_CLEARED, enable the block as
cleared. Else, reset the clear flag for each block in the list(Christian)
  - For merging the 2 cleared blocks compare as below,
drm_buddy_is_clear(block) != drm_buddy_is_clear(buddy)(Christian)
  - Defragment the memory beginning from min_order
till the required memory space is available.

v2: (Matthew)
  - Add a wrapper drm_buddy_free_list_internal for the freeing of blocks
operation within drm buddy.
  - Write a macro block_incompatible() to allocate the required blocks.
  - Update the xe driver for the drm_buddy_free_list change in arguments.
  - add a warning if the two blocks are incompatible on
defragmentation
  - call full defragmentation in the fini() function
  - place a condition to test if min_order is equal to 0
  - replace the list with safe_reverse() variant as we might
remove the block from the list.

v3:
  - fix Gitlab user reported lockup issue.
  - Keep DRM_BUDDY_HEADER_CLEAR define sorted(Matthew)
  - modify to pass the root order instead max_order in fini()
function(Matthew)
  - change bool 1 to true(Matthew)
  - add check if min_block_size is power of 2(Matthew)
  - modify the min_block_size datatype to u64(Matthew)

Signed-off-by: Arunpravin Paneer Selvam 
Signed-off-by: Matthew Auld 
Suggested-by: Christian König 
Suggested-by: Matthew Auld 
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c  |   6 +-
 drivers/gpu/drm/drm_buddy.c   | 294 +++---
 drivers/gpu/drm/i915/i915_ttm_buddy_manager.c |   6 +-
 drivers/gpu/drm/tests/drm_buddy_test.c|  18 +-
 drivers/gpu/drm/xe/xe_ttm_vram_mgr.c  |   4 +-
 include/drm/drm_buddy.h   |  22 +-
 6 files changed, 290 insertions(+), 60 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 8db880244324..c0c851409241 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -571,7 +571,7 @@ static int amdgpu_vram_mgr_new(struct ttm_resource_manager 
*man,
return 0;
 
 error_free_blocks:
-   drm_buddy_free_list(mm, >blocks);
+   drm_buddy_free_list(mm, >blocks, 0);
mutex_unlock(>lock);
 error_fini:
ttm_resource_fini(man, >base);
@@ -604,7 +604,7 @@ static void amdgpu_vram_mgr_del(struct ttm_resource_manager 
*man,
 
amdgpu_vram_mgr_do_reserve(man);
 
-   drm_buddy_free_list(mm, >blocks);
+   drm_buddy_free_list(mm, >blocks, 0);
mutex_unlock(>lock);
 
atomic64_sub(vis_usage, >vis_usage);
@@ -912,7 +912,7 @@ void amdgpu_vram_mgr_fini(struct amdgpu_device *adev)
kfree(rsv);
 
list_for_each_entry_safe(rsv, temp, >reserved_pages, blocks) {
-   drm_buddy_free_list(>mm, >allocated);
+   drm_buddy_free_list(>mm, >allocated, 0);
kfree(rsv);
}
if (!adev->gmc.is_app_apu)
diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c
index c4222b886db7..40131ed9b0cd 100644
--- a/drivers/gpu/drm/drm_buddy.c
+++ b/drivers/gpu/drm/drm_buddy.c
@@ -57,6 +57,16 @@ static void list_insert_sorted(struct drm_buddy *mm,
__list_add(>link, node->link.prev, >link);
 }
 
+static void clear_reset(struct drm_buddy_block *block)
+{
+   block->header &= ~DRM_BUDDY_HEADER_CLEAR;
+}
+
+static void mark_cleared(struct drm_buddy_block *block)
+{
+   block->header |= DRM_BUDDY_HEADER_CLEAR;
+}
+
 static void mark_allocated(struct drm_buddy_block *block)
 {
block->header &= ~DRM_BUDDY_HEADER_STATE;
@@ -186,11 +196,21 @@ EXPORT_SYMBOL(drm_buddy_init);
  */
 void drm_buddy_fini(struct drm_buddy *mm)
 {
+   u64 root_size, size;
+   unsigned int order;
int i;
 
+   size = mm->size;
+
for (i = 0; i < mm->n_roots; ++i) {
+   order = ilog2(size) - ilog2(mm->chunk_size);
+   root_size = mm->chunk_size << order;
+   drm_buddy_defrag(mm, root_size);
+
WARN_ON(!drm_buddy_block_is_free(mm->roots[i]));
drm_block_free(mm, mm->roots[i]);
+
+   size -= root_size;
}
 
WARN_ON(mm->avail != mm->size);
@@ -223,6