summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2007-02-14Rename drm_ttm.h to drm_objects.httm-vram-0-1-branchThomas Hellstrom3-9/+10
Fix up some header incompatibilities in drm_fence.c caused by the previous commit.
2007-02-14Move fence- and buffer-object related header stuff to drm_ttm.hThomas Hellstrom2-371/+357
2007-02-14Remove an intel-specific hack and replace it with a fence driver callback.Thomas Hellstrom5-50/+79
2007-02-14Set the drm bus map type for each buffer object memory type.Thomas Hellstrom4-3/+9
2007-02-14Rework buffer object vm code to use nopfn() for kernels >= 2.6.19.Thomas Hellstrom3-67/+145
2007-02-13More bugfixes.Thomas Hellstrom4-61/+139
Fixed memory, pinned buffers and unmappable memory now seems fully functional.
2007-02-12Update flags and comments.Thomas Hellstrom1-35/+60
2007-02-12Lindent.Thomas Hellstrom5-305/+244
2007-02-12More bugfixes.Thomas Hellstrom2-33/+34
2007-02-12Cleanup and fix support for pinned buffers.Thomas Hellstrom4-185/+263
2007-02-10Various bugfixes.Thomas Hellstrom4-33/+103
2007-02-09I915 accelerated blit copy functional.Thomas Hellstrom6-17/+77
Fixed - to System memory copies are implemented by flipping in a cache-coherent TTM, blitting to it, and then flipping it out.
2007-02-09Reinstate some LRU handling.Thomas Hellstrom1-0/+21
2007-02-09Remove some code that should have gone inThomas Hellstrom1-19/+0
commit 6a49d9a8abd9f168211017c2d585d0d64e89c530
2007-02-09Fix copyright statements.Thomas Hellstrom6-31/+32
2007-02-09Fix evict_mutex locking range.Thomas Hellstrom3-22/+38
Implement unmappable buffers. (fault moves them to mappable when needed). Various bugfixes.
2007-02-08Add an accelerated buffer copy cleanup helper.Thomas Hellstrom4-54/+177
Export helper functions and make some important buffer-object functions non-static. Add an i915 accelerated blit buffer move for pci memory buffers.
2007-02-08Don't create a ttm just to copy from.Thomas Hellstrom2-2/+3
2007-02-08Fix mm_block leak.Thomas Hellstrom3-4/+6
Some other minor fixes.
2007-02-08A minor function interface change and some memcpy bugfixing.Thomas Hellstrom4-29/+31
Hooray!! it sort of works with a fixed AGP area as faked VRAM.
2007-02-08Simplify external ttm page allocation.Thomas Hellstrom7-107/+210
Implement a memcpy fallback for copying between buffers.
2007-02-08Update memory compatibility tests.Thomas Hellstrom1-11/+78
Now only pinned buffers are broken.
2007-02-07Checkpoint commit.Thomas Hellstrom9-207/+142
Flag handling and memory type selection cleanup. glxgears won't start.
2007-02-07Fix a stray unlock_kernel() in drm_vm.cThomas Hellstrom6-203/+264
Add a file for memory move helpers, drm_bo_move.c Implement generic memory move. Cached, no_move and unmapped memory temporarily broken.
2007-02-06Simplify pci map vs no pci map choice.Thomas Hellstrom7-57/+58
2007-02-06Implement a drm_mem_reg_t substructure in the buffer object type.Thomas Hellstrom2-92/+89
2007-02-06Implement a policy for selecting memory types.Thomas Hellstrom6-56/+205
2007-02-05i915: Add copy-blit operation.Thomas Hellstrom1-0/+44
2007-02-02Make also later kernels work with buffer object vmThomas Hellstrom8-185/+166
and clean up some function names.
2007-02-02Make vm handle buffer objects instead of ttm objects.Thomas Hellstrom9-423/+467
Remove ttm objects. Make vm aware of PCI memory type buffer objects. (Only works for pre 2.6.16 kernels for now).
2007-02-01Fix missing ttm_open_vma call from previous commit.Thomas Hellstrom6-14/+21
Honour the ttm backend cant-use-aperture flag.
2007-02-01Prepare for removal of the ttm_object type.Thomas Hellstrom1-8/+26
2007-02-01Protect drm_mmap against disappearing maps.Thomas Hellstrom1-26/+39
The map lists and hash tables are protected using dev->struct_mutex, but drm_mmap strangely never locked this mutex.
2007-01-31memory manager: Make device driver aware of different memory types.Thomas Hellstrom6-50/+127
Memory types are either fixed (on-card or pre-bound AGP) or not fixed (dynamically bound) to an aperture. They also carry information about: 1) Whether they can be mapped cached. 2) Whether they are at all mappable. 3) Whether they need an ioremap to be accessible from kernel space. In this way VRAM memory and, for example, pre-bound AGP appear identical to the memory manager. This also makes support for unmappable VRAM simple to implement.
2007-01-31Make the utility runnable also for normal users.Thomas Hellstrom1-1/+1
2007-01-31Fix an error-path oops.Thomas Hellstrom1-2/+4
2007-01-30Add a buffer object transfer function.Thomas Hellstrom1-0/+43
Creates a placeholder for the old buffer contents when it is transfered to / from static memory like VRAM.
2007-01-30Indent according to xorg rules.Thomas Hellstrom1-70/+54
2007-01-30Add license header.Thomas Hellstrom1-0/+31
2007-01-30Add some relevant tests for the new buffer object interface.Thomas Hellstrom1-1/+220
2007-01-30Add the ttmtest test utility.Thomas Hellstrom12-0/+1372
2007-01-30Clean up buffer object destruction somewhat.Thomas Hellstrom1-134/+152
2007-01-29Use pre-defined list_splice function.Thomas Hellstrom1-3/+3
2007-01-29s/buf/bo/ for consistency.Thomas Hellstrom1-35/+35
2007-01-29Some cleanup. A buffer object should only have one active memory type.Thomas Hellstrom2-119/+51
2007-01-28nouveau: determine chipset type at startup, instead of every time we use it.Ben Skeggs5-23/+24
2007-01-26make works ctx switch on nv10.Matthieu Castet1-4/+7
2007-01-26nouveau: oops, wrong indexing in nv17 regsPatrice Mandin1-5/+5
2007-01-26nouveau: read gpu type oncePatrice Mandin1-5/+8
2007-01-26nouveau: only save/restore nv17 regs on nv17,18 hwPatrice Mandin1-15/+31