PT-2026-98768 · Linux · Linux

CVE-2026-98106

·

Published

2026-09-25

·

Updated

2026-09-25

None

No severity ratings or metrics are available. When they are, we'll update the corresponding info on the page.
In the Linux kernel, the following vulnerability has been resolved:
drm/pagemap: Prevent double migration of device pages
A device-private folio migrated to system memory by a CPU fault can remain reachable through the raw-PFN eviction path until migration finalization drops the source reference.
If eviction selects the same device-private folio during this window, it can attempt to migrate the folio again. The second migration can leave an uncharged folio on an LRU list, causing folio lruvec lock irqsave() to retry indefinitely and resulting in a soft lockup and RCU stall.
Mark successfully migrated device-private folios using a low bit of their zone device data before migration finalization. Make both CPU-fault and raw-PFN migration paths skip device-private folios carrying this flag.
Mask the flag when retrieving the drm pagemap zdd pointer and preserve it when a device-private folio is split. Keeping the state on the physical folio also avoids depending on a virtual address that may change before a fault occurs.
v2:
  • Replace the retired-PFN XArray with an embedded bitmap. (Matthew Brost)
  • Mark every base page covered by a migrated folio so retirement remains valid if the folio is later split.
v3:
  • Store the migrated state in a low bit of zone device data instead of adding virtual-range and bitmap tracking to the ZDD. (Matthew Brost)
  • Mask the flag when retrieving the ZDD and preserve it when splitting a folio.
  • Drop the pre-existing fixes already covered by Matthew Brost's series: https://patchwork.freedesktop.org/series/171651/
v4:
  • Advance by the folio size only for migration entries marked with MIGRATE PFN COMPOUND. (Sashiko)
v5:
  • Simplify ZDD flag updates and folio iteration. (Matthew Brost)
  • Skip retired device-private folios in the CPU-fault path. (Matthew Brost)
  • Preserve flag bits while taking a new ZDD reference for split folios.
v6:
The lockup was observed as: [10109.860465] watchdog: BUG: soft lockup - CPU#9 stuck for 26s! [kworker/u65:5:6557] [10109.860524] Tainted: [S]=CPU OUT OF SPEC, [O]=OOT MODULE [10109.860524] Hardware name: ASUS System Product Name/PRIME Z790-P WIFI, BIOS 0812 02/24/2023 [10109.860525] Workqueue: xe page fault work queue xe pagefault queue work [xe] [10109.860644] RIP: 0010: raw spin unlock irqrestore+0x57/0x80 [10109.860655] Call Trace: [10109.860655] [10109.860657] folio lruvec lock irqsave+0x216/0x220 [10109.860661] ? pfx lru add+0x10/0x10 [10109.860665] folio batch move lru+0xc8/0x450 [10109.860670] ? lock acquire+0xc4/0x2d0 [10109.860674] ? folio batch add and move+0x60/0x2e0 [10109.860677] ? folio migrate mapping+0xa6/0x110 [10109.860679] ? folio migrate flags+0x13b/0x1b0 [10109.860681] ? pfx lru add+0x10/0x10 [10109.860683] folio batch add and move+0xe7/0x2e0 [10109.860685] ? dma iova try alloc+0xb0/0x140 [10109.860689] folio add lru+0x64/0x80 [10109.860691] migrate device finalize+0x12c/0x270 [10109.860695] migrate device finalize+0x10/0x20 [10109.860698] drm pagemap evict to ram+0x185/0x370 [drm gpusvm helper] [10109.860704] ? drm pagemap evict to ram+0x96/0x370 [drm gpusvm helper] [10109.860709] xe svm bo evict+0x15/0x20 [xe] [10109.860819] ? xe svm bo evict+0x15/0x20 [xe] [10109.860921] xe bo move+0x107e/0x1570 [xe] [10109.860992] ? xe ttm tt create+0x168/0x340 [xe] [10109.861059] ? up read+0x98/0x2b0 [10109.861061] ? lock is held type+0xa3/0x130 [10109.861067] ttm bo handle move mem+0xe8/0x1e0 [ttm] [10109.861075] ttm bo evict+0x141/0x1c0 [ttm] [10109.861081] ttm bo evict cb+0x9f/0x100 [ttm] [10109.861086] ttm lru walk for evict+0x84/0x190 [ttm] [10109.861091] ? xe ttm vram mgr new+0x258/0x3a0 [xe] [10109.861198] ttm bo alloc resource+0x219/0 ---truncated---
Found an issue in the description? Have something to add? Feel free to write us 👾
dbugs@ptsecurity.com

Related Identifiers

CVE-2026-98106

Affected Products

Linux