drm/xe: Fine grained page fault locking

Browse source

Enable page faults to be serviced while holding vm->lock in read mode.

Introduce additional locks to:
 - Ensure only one page fault thread services a given range or VMA
 - Serialize SVM garbage collection
 - Protect SVM range insertion and removal

While these locks may contend during page faults, expensive operations
like migration can now run in parallel within a single VM.

In addition to new locking, ranges must be reference-counted after
lookup, as another thread could immediately remove them from the GPU SVM
tree, potentially dropping the last reference.

Lastly, decouple the VM’s ASID from the page fault queue selection to
allow parallel page fault handling within the same VM.

Lays the groundwork for prefetch IOCTLs to use threaded migration too.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Gwan-gyeong Mun <gwan-gyeong.mun@intel.com>
Link: https://patch.msgid.link/20260806185202.3922432-2-matthew.brost@intel.com

...

This commit is contained in:

Matthew Brost 2026-08-06 11:51:51 -07:00

parent 52d223c05e

commit 941bcc1a4b

7 changed files with 300 additions and 131 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

2

drivers/gpu/drm/xe/xe_device_types.h

View file

106

drivers/gpu/drm/xe/xe_pagefault.c

View file

122

drivers/gpu/drm/xe/xe_svm.c

View file

44

drivers/gpu/drm/xe/xe_svm.h

View file

19

drivers/gpu/drm/xe/xe_userptr.c

View file

113

drivers/gpu/drm/xe/xe_vm.c

View file

29

drivers/gpu/drm/xe/xe_vm_types.h

View file