提交 · 24f246ac10ae6a6ae873045387d4501498869f74 · openanolis / cloud-kernel

23 6月, 2011 40 次提交
- B
  drm/nouveau: rework vram init/fini ordering a little · 24f246ac
  由 Ben Skeggs 提交于 6月 10, 2011
```
Commit "drm/nouveau: add some debug output if nouveau_mm busy at destroy time"
revealed an issue where vram mm takedown would actually fail due to there
still being nodes present, causing nouveau to leak a small amount of memory
on module unload.

This splits TTM/nouveau_mm a bit more cleanly and ensures nouveau_mm fini
isn't done until all gpuobjs are also destroyed.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  24f246ac
- B
  drm/nouveau: shut lockdep up if last vm ref needs to destroy pgd · 15ba79ad
  由 Ben Skeggs 提交于 6月 08, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  15ba79ad
- B
  drm/nouveau: fix display takedown order to match reverse init order · 06b75e35
  由 Ben Skeggs 提交于 6月 08, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  06b75e35
- B
  drm/nvc0: enable per-client address spaces · 5de8037a
  由 Ben Skeggs 提交于 6月 08, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  5de8037a
- B
  drm/nouveau: add some debug output if nouveau_mm busy at destroy time · ad9ac437
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  ad9ac437
- B
  drm/nv50: enable use of per-client gpu address space · e41f26e7
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  e41f26e7
- B
  drm/nouveau: remove implicit mapping of every bo into chan_vm · 2fd3db6f
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  2fd3db6f
- B
  drm/nouveau: remove 'chan' argument from nouveau_bo_new · 7375c95b
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  7375c95b
- B
  drm/nouveau: fixup gem_info ioctl to return client-specific bo virtual · e758a311
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  e758a311
- B
  drm/nvc0: explicitly map PDISP semaphore buffer into each channel's vm · 3d483d57
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  3d483d57
- B
  drm/nv50-nvc0: lookup pushbuf virtual address on dma_push · 9f9f51fc
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  9f9f51fc
- B
  drm/nv84-nvc0: explicitly map semaphore buffer into channel vm · d02836b4
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  d02836b4
- B
  drm/nv50-nvc0: explicitly map pushbuf bo into channel vm · ce163f69
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  ce163f69
- B
  drm/nv50-nvc0: explicitly map notifier bo into channel vm · 0b718733
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  0b718733
- B
  drm/nv50-nvc0: explicitly map fbcon fb into channel vm · 45143cb5
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  45143cb5
- B
  drm/nouveau: initial changes to support multiple VMAs per buffer object · fd2871af
  由 Ben Skeggs 提交于 6月 06, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  fd2871af
- B
  drm/nv50-nvc0: completely disable relocs · a3fcd0a9
  由 Ben Skeggs 提交于 6月 07, 2011
```
GPU virtual addresses are constant now so this should never be getting hit
anyway and userspace shouldn't break from them being ignored.

This is being done in preference to teaching the code how to deal with BOs
that exist at different virtual addresses within separate VMs.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  a3fcd0a9
- B
  drm/nouveau: convert bo.mem.start usage to bo.offset · 180cc306
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  180cc306
- B
  drm/nouveau: convert some bo.offset use to vma.offset · 07533ea5
  由 Ben Skeggs 提交于 6月 07, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  07533ea5
- B
  drm/nouveau: create temp vmas for both src and dst of bo moves · d2f96666
  由 Ben Skeggs 提交于 6月 06, 2011
```
Greatly simplifies a number of things, particularly once per-client GPU
address spaces are involved.

May add this back later once I know what things'll look like.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  d2f96666
- B
  drm/nouveau: store bo's page size in nouveau_bo · f91bac5b
  由 Ben Skeggs 提交于 6月 06, 2011
```
Was previously assuming a page size of 4KiB unless a VMA was present to
override it.  Eventually, a buffer won't necessarily have a VMA at all at
some stages of its life, so we need to store this info elsewhere.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  f91bac5b
- B
  drm/nouveau: skip move_notify() if bo does not have a vma attached · 111af5c1
  由 Ben Skeggs 提交于 6月 03, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  111af5c1
- B
  drm/nouveau: will need to specify channel for vm-ful gpuobj allocations · 6e32fedc
  由 Ben Skeggs 提交于 6月 03, 2011
```
Abuses existing gpuobj_new() chan argument for this, which in turn forces
all NVOBJ_FLAG_VM allocations to be done from the global heap, not
suballocated from the channel's private heap.  Not a problem though in
practise.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  6e32fedc
- B
  drm/nouveau: initialise any vm for a channel before pushbuf/ntfy · dd6a46cc
  由 Ben Skeggs 提交于 6月 03, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  dd6a46cc
- B
  drm/nouveau: modify gpuobj/ntfy takedown ordering · b7cb6c01
  由 Ben Skeggs 提交于 6月 06, 2011
```
gpuobj really needs splitting into channel/gpuobj code instead...
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  b7cb6c01
- B
  drm/nouveau/gem: implement stub hooks for GEM object open/close · 639212d0
  由 Ben Skeggs 提交于 6月 03, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  639212d0
- B
  drm/nouveau: remove 'chan' argument from nouveau_gem_new · f6d4e621
  由 Ben Skeggs 提交于 6月 07, 2011
```
Userspace hasn't passed us a channel_hint for a long long time now, and
there isn't actually a need to do so anymore anyway.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  f6d4e621
- B
  drm/nv50-nvc0/chan: inherit vm from fpriv, rather than chan_vm directly · 0320d791
  由 Ben Skeggs 提交于 6月 03, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  0320d791
- B
  drm/nv50-nvc0/vm: take client reference on shared channel vm · fe32b16e
  由 Ben Skeggs 提交于 6月 03, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  fe32b16e
- B
  drm/nv50-nvc0/vm: don't touch chan_vm · b79181cb
  由 Ben Skeggs 提交于 6月 03, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  b79181cb
- B
  drm/nouveau: no need to update bo.offset from vma after validate · d359d51c
  由 Ben Skeggs 提交于 6月 03, 2011
```
On chipsets using nouveau_vm, the virtual address stays constant, so
the value set at bo creation time is fine.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  d359d51c
- B
  drm/nouveau: store a per-client channel list · e8a863c1
  由 Ben Skeggs 提交于 6月 01, 2011
```
Removes the need to disable IRQs to lookup channel struct on every pushbuf
ioctl, among others.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  e8a863c1
- B
  drm/nouveau: use NULL file_priv for DRM-created channels · f8656f0b
  由 Ben Skeggs 提交于 5月 31, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  f8656f0b
- B
  drm/nouveau: allocate structure to store per-client data · 3f0a68d8
  由 Ben Skeggs 提交于 5月 31, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  3f0a68d8
- B
  drm/nouveau: silence error for missing dac loadval table · 1562ffde
  由 Ben Skeggs 提交于 6月 01, 2011
```
There's lots of boards (all recent ones) that don't have this anymore, so
punt the message to debug loglevel.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  1562ffde
- B
  drm/nv50: DCB table quirks for another busted XFX board · e540afc3
  由 Ben Skeggs 提交于 5月 30, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  e540afc3
- B
  drm/nouveau: log if accel is disabled by default on a chipset · ad830d23
  由 Ben Skeggs 提交于 5月 27, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  ad830d23
- B
  drm/nvc0/gr: fill in some more data for 0xc1/0xc8/0xce · 752ab0a0
  由 Ben Skeggs 提交于 5月 27, 2011
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  752ab0a0
- B
  drm/nvc0/gr: fix typo in class9197 init · 2b6f1c5f
  由 Ben Skeggs 提交于 5月 27, 2011
```
Reported-by: NChristoph Bumiller <e0425955@student.tuwien.ac.at>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  2b6f1c5f
- B
  drm/nvc0/gr: calculate magicgpc918 ourselves · 066d65db
  由 Ben Skeggs 提交于 5月 26, 2011
```
Not a clue what it is yet, but we get the same numbers as NVIDIA now.

My 465 didn't seem to care to greatly *what* I bashed into these registers..
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  066d65db

openanolis / cloud-kernel 1 年多 前同步成功

openanolis / cloud-kernel
1 年多前同步成功