On Fri, 19 Dec 2025, Lin, Shuicheng wrote:
On Fri, Dec 19, 2025 11:33 AM Michał Grzelak wrote:
During initialization of DRM buddy memory manager at drm_buddy_init,
mm->free_trees array is allocated for both clear and dirty RB trees.
During cleanup happening at drm_buddy_fini it is never freed, leading to
following memory leaks observed on xe module load & unload cycles:
kmemleak_alloc+0x4a/0x90
__kmalloc_cache_noprof+0x488/0x800
drm_buddy_init+0xc2/0x330 [drm_buddy]
__xe_ttm_vram_mgr_init+0xc3/0x190 [xe]
xe_ttm_stolen_mgr_init+0xf5/0x9d0 [xe]
xe_device_probe+0x326/0x9e0 [xe]
xe_pci_probe+0x39a/0x610 [xe]
local_pci_probe+0x47/0xb0
pci_device_probe+0xf3/0x260
really_probe+0xf1/0x3c0
__driver_probe_device+0x8c/0x180
driver_probe_device+0x24/0xd0
__driver_attach+0x10f/0x220
bus_for_each_dev+0x7f/0xe0
driver_attach+0x1e/0x30
bus_add_driver+0x151/0x290
Deallocate array for free trees when cleaning up buddy memory manager in
the same way as if going through out_free_tree label.
Fixes: d4cd665c98c1 ("drm/buddy: Separate clear and dirty free block trees")
Signed-off-by: Michał Grzelak <[email protected]>
Reviewed-by: Lucas De Marchi <[email protected]>
Reviewed-by: Matthew Auld <[email protected]>
I also meet this issue. And the fix LGTM. It has the same logic as the failure
path of drm_buddy_init().
Reviewed-by: Shuicheng Lin <[email protected]>
Thank you Shuicheng for the review and confirmation. Will update the
commit message along your R-B in next round of resend.
BTW, do you have any recommendation regarding the tree via which this
change should be merged? Asking since I don't have the commit bit to any
of drm-* trees.
BR,
Michał
