ttm mem manager refactoring.
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest49 patches 49 repliesIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/49 | drm/qxl/ttm: call ttm manager debugdrivers/ | +4 | 0 replies | |
| 2/49 | drm/vram-helper: call the ttm manager debug functiondrivers/ | +2 | 1 reply | |
| 3/49 | drm/ttm: split the mm manager init codedrivers/ include/ include/ | +30 | 5 replies | |
| 4/49 | drm/ttm: provide a driver-led init path for generic mm manager.drivers/ include/ | +23 | 4 replies | |
| 5/49 | drm/amdgpu/ttm: init managers from the driver side.drivers/ drivers/ drivers/ +1 more | +33 | 0 replies | |
| 6/49 | drm/radeon: use new ttm man init pathdrivers/ | +3 | 0 replies | |
| 7/49 | drm/qxl/ttm: use new init path for managerdrivers/ | +1 | 0 replies | |
| 8/49 | drm/vram_helper: use new ttm manager init functiondrivers/ | +1 | 0 replies | |
| 9/49 | drm/nouveau: use new memory manager init pathsdrivers/ | +19 | 0 replies | |
| 10/49 | drm/vmwgfx/ttm: convert vram mm init to new code pathsdrivers/ drivers/ drivers/ | +28 | 0 replies | |
| 11/49 | drm/vmwgfx/ttm: switch gmrid allocator to new init paths.drivers/ drivers/ drivers/ | +17 | 0 replies | |
| 12/49 | drm/ttm: convert system manager init to new code.drivers/ | +17 | 1 reply | |
| 13/49 | drm/ttm: purge old manager init path.drivers/ drivers/ include/ +1 more | +9 | 1 reply | |
| 14/49 | drm/ttm: pass man around instead of mem_type in some placesdrivers/ | +6 | 1 reply | |
| 15/49 | drm/ttm: make some inline helper functions for cleanup paths.drivers/ include/ | +14 | 2 replies | |
| 16/49 | drm/ttm: start allowing drivers to use new takedown pathdrivers/ drivers/ include/ | +28 | 1 reply | |
| 17/49 | drm/amdgpu/ttm: use new takedown pathdrivers/ drivers/ drivers/ +1 more | +29 | 1 reply | |
| 18/49 | drm/vmwgfx: takedown vram managerdrivers/ drivers/ drivers/ | +26 | 0 replies | |
| 19/49 | drm/vram_helper: call explicit mm takedowndrivers/ | +1 | 0 replies | |
| 20/49 | drm/nouveau: use new cleanup pathsdrivers/ | +30 | 0 replies | |
| 21/49 | drm/radeon/ttm: use new takedown pathsdrivers/ | +2 | 1 reply | |
| 22/49 | drm/qxl/ttm: use new takedown pathdrivers/ | +2 | 0 replies | |
| 23/49 | drm/vmwgfx: fix gmrid takedown paths to new interfacedrivers/ drivers/ drivers/ | +14 | 0 replies | |
| 24/49 | drm/ttm: remove range manager legacy takedown pathdrivers/ | +9 | 1 reply | |
| 25/49 | drm/ttm: make TTM responsible for cleaning system only.drivers/ include/ include/ | +4 | 1 reply | |
| 26/49 | drm/ttm: add wrapper to get manager from bdev.drivers/ drivers/ drivers/ +1 more | +34 | 5 replies | |
| 27/49 | drm/amdgfx/ttm: use wrapper to get ttm memory managersdrivers/ drivers/ drivers/ +5 more | +39 | 0 replies | |
| 28/49 | drm/vram-helper: use wrapper to access memory managersdrivers/ | +3 | 0 replies | |
| 29/49 | drm/nouveau/ttm: use wrapper to access memory managersdrivers/ | +4 | 0 replies | |
| 30/49 | drm/qxl/ttm: use wrapper to access memory managerdrivers/ | +5 | 0 replies | |
| 31/49 | drm/radeon/ttm: use wrapper to access memory managerdrivers/ drivers/ | +7 | 0 replies | |
| 32/49 | drm/vmwgfx/ttm: use wrapper to access memory managerdrivers/ drivers/ drivers/ | +17 | 0 replies | |
| 33/49 | drm/ttm: rename manager variable to make sure wrapper is used.drivers/ include/ | +6 | 0 replies | |
| 34/49 | drm/ttm: make manager debug function optionaldrivers/ | +1 | 3 replies | |
| 35/49 | drm/nouveau/ttm: don't fill in blank ttm debug callbackdrivers/ | +0 | 1 reply | |
| 36/49 | drm/vmwgfx/gmrid: don't provide pointless ttm debug callbackdrivers/ | +0 | 1 reply | |
| 37/49 | drm/ttm: allow drivers to provide their own manager subclassesinclude/ | +10 | 0 replies | |
| 38/49 | drm/amdgpu/ttm: use bo manager subclassing for vram/gtt mgrsdrivers/ drivers/ | +44 | 1 reply | |
| 39/49 | drm/ttm: make ttm_bo_man_init/takedown take type + argsdrivers/ drivers/ drivers/ +5 more | +67 | 5 replies | |
| 40/49 | drm/ttm: move range manager to subclassed driver allocationdrivers/ | +20 | 1 reply | |
| 41/49 | drm/vmwgfx/ttm: move thp to driver manageddrivers/ | +22 | 0 replies | |
| 42/49 | drm/vmwgfx/gmrid: convert to driver controlled allocation.drivers/ | +20 | 0 replies | |
| 43/49 | drm/nouveau/ttm: move to driver allocated managerdrivers/ | +28 | 0 replies | |
| 44/49 | drm/ttm: drop priv pointer in memory managerinclude/ | +0 | 1 reply | |
| 45/49 | drm/amdgpu/ttm: remove man->bdev references.drivers/ | +5 | 1 reply | |
| 46/49 | drm/ttm: drop man->bdev link.drivers/ drivers/ drivers/ +6 more | +10 | 1 reply | |
| 47/49 | drm/ttm: drop list of memory managers from device.drivers/ include/ | +4 | 2 replies | |
| 48/49 | drm/ttm: drop type manager has_typedrivers/ include/ | +2 | 0 replies | |
| 49/49 | drm/ttm: consolidate manager used apis into a set and get.drivers/ drivers/ drivers/ +6 more | +28 | 2 replies |
Reviews trailers per patch
16 of 49 patches have a Reviewed-by or Acked-by.
| Patch | Thomas Zimmermann | Christian König |
|---|---|---|
| 1/49 | ||
| 2/49 | R-b | |
| 3/49 | ||
| 4/49 | R-b | |
| 5/49 | ||
| 6/49 | ||
| 7/49 | ||
| 8/49 | ||
| 9/49 | ||
| 10/49 | ||
| 11/49 | ||
| 12/49 | R-b | |
| 13/49 | R-b | |
| 14/49 | R-b | |
| 15/49 | ||
| 16/49 | ||
| 17/49 | R-b | |
| 18/49 | ||
| 19/49 | ||
| 20/49 | ||
| 21/49 | R-b | |
| 22/49 | ||
| 23/49 | ||
| 24/49 | ||
| 25/49 | R-b | |
| 26/49 | ||
| 27/49 | ||
| 28/49 | ||
| 29/49 | ||
| 30/49 | ||
| 31/49 | ||
| 32/49 | ||
| 33/49 | ||
| 34/49 | R-b | |
| 35/49 | A-b | |
| 36/49 | A-b | |
| 37/49 | ||
| 38/49 | R-b | |
| 39/49 | ||
| 40/49 | R-b | |
| 41/49 | ||
| 42/49 | ||
| 43/49 | ||
| 44/49 | R-b | |
| 45/49 | R-b | |
| 46/49 | R-b | |
| 47/49 | ||
| 48/49 | ||
| 49/49 |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [PATCH 00/49] ttm mem manager refactoring.
I started pulling on a thread, and it led me down a hole.
This series refactors the ttm ttm_mem_type_manager object into a driver owned, allocated, subclassaed object.
It starts with two minor fixes for some bad assumptions in two drivers.
Enables a new init path, ports all the drivers to the new init path, and cleans up the old init path.
Enables a new takedown path, ports all the drivers to the new takedown path, and cleans up the old takedown path
Wraps all access to the memory managers in the bo_device in a wrapper across all drivers.
Make debug callback optional
Enables driver to provide their own mem manager objects Subclasses the objects in all drivers and makes them into driver owned object Drops the bo_device arrays of pointers, and some unneeded links and struct members
Cleans up one api.
I think I'd probably want to merge all this via drm-misc, so if I can collect acks/r-b from driver maintainers that would be good.
This is also based on Chrisitan's work to remove init_mem_type, so it won't apply until he's finished getting all of that into drm-misc.
https://cgit.freedesktop.org/~airlied/linux/log/?h=ttm-refactor-mem-manager
is the tree I've built this on top off, so it's probably going to get rebased but the code should stay mostly the same.
I've done some boot testing on nouveau, and I hope to test it on vmwgfx and amdgpu soon.
Dave.
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel