CVE-2026-52965
Description
In the Linux kernel, the following vulnerability has been resolved:
drm/ttm: Fix ttm_bo_swapout() infinite LRU walk on swapout failure
When ttm_tt_swapout() fails, the current code calls
ttm_resource_add_bulk_move() followed by ttm_resource_move_to_lru_tail()
to restore the resource's bulk_move membership.
However, ttm_resource_move_to_lru_tail() places the resource at the tail
of the LRU list which, relative to the walk cursor's hitch node (placed
immediately after the resource when it was yielded), puts the resource
*in front of the* the hitch. The next list_for_each_entry_continue() from
the hitch finds the same resource again, causing an infinite loop.
Fix by deferring del_bulk_move to the success path only.
On the success path, TTM_TT_FLAG_SWAPPED has just been set by
ttm_tt_swapout() but the resource is still tracked in the bulk_move range,
so ttm_resource_del_bulk_move()'s !ttm_resource_unevictable() guard would
incorrectly skip the removal. Introduce
ttm_resource_del_bulk_move_unevictable() which bypasses that guard.
Summary dbcve.org
In the Linux kernel's TTM (Translation Table Manager) subsystem, a logic bug in ttm_bo_swapout() causes an infinite loop when swapout fails. After ttm_tt_swapout() fails, the code restores the resource's bulk_move membership by calling ttm_resource_move_to_lru_tail(), which places the resource in front of the walk cursor's hitch node. This causes list_for_each_entry_continue() to find the same resource repeatedly, resulting in an infinite LRU walk that hangs the system.
Mitigation
This is a kernel bug requiring a code fix. The fix involves modifying the failure path in ttm_bo_swapout() to defer del_bulk_move to the success path only, and adding a new helper function ttm_resource_del_bulk_move_unevictable() to handle the swapped state correctly.