drm/gpusvm: let the drm_gpusvm core context purely MM level

Browse source

The core mechanism of drm_gpusvm is HMM, which is fundamentally an
MM side subsystem. A drm_device, enters the picture on the device side at
DMA mapping / GPU bind.

So move struct drm_device from struct drm_gpusvm in drm_gpusvm. Let
drm_gpusvm keep its core neutral and leave device side decisions to
the driver. Make drm_gpusvm a pure MM level object.

  - Move the drm_device from struct drm_gpusvm. drm_device now stored in
    drm_gpusvm_pages.
  - Drop the drm parameter from drm_gpusvm_init()
  - Update the xe call sites in xe_svm_init() and other callers.

drm_device does not disappear from the framework, it is
relocated onto each drm_gpusvm_pages where DMA actually happens.

Suggested-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Honglei Huang <honghuan@amd.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://patch.msgid.link/20260630102127.392396-6-honghuan@amd.com

...

This commit is contained in:

Honglei Huang 2026-06-30 18:21:27 +08:00 • committed by Matthew Brost

parent e19d32c2b3

commit 808db118eb

4 changed files with 8 additions and 10 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

8

drivers/gpu/drm/drm_gpusvm.c

View file

4

drivers/gpu/drm/xe/xe_svm.c

View file

2

drivers/gpu/drm/xe/xe_svm.h

View file

4

include/drm/drm_gpusvm.h

View file