提交 · b1cd916ab96e5fcd16e2009410af2e8edfae2962 · openanolis / cloud-kernel

29 12月, 2010 1 次提交
- B
  drm/nvc0: kill off a couple more magics · b1cd916a
  由 Ben Skeggs 提交于 12月 24, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  b1cd916a
27 12月, 2010 2 次提交

drm/nouveau: Validate channel indices passed from userspace. · 8e91182b

由 Michel Hermier 提交于 12月 25, 2010

When hacking the libdrm for improvements, I triggered a kernel crash
related to the fact that the NOUVEAU_NOTIFIEROBJ_ALLOC ioctl calls
nouveau_channel_get with an unchecked channel index.
The patch ensures that the channel index is an unsigned and validates
its value in nouveau_channel_get.
Signed-off-by: NMichel Hermier <hermier@frugalware.org>
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

8e91182b

drm/nouveau: Only select ACPI_VIDEO if its dependencies are met · c4298801

由 Ben Hutchings 提交于 12月 25, 2010

CONFIG_ACPI_VIDEO depends on more than just CONFIG_ACPI, so add those
dependencies to the Kconfig select condition.  The case where some
dependencies fail to be satisfied should be handled correctly, because
in that case the ACPI_VIDEO symbols we use are converted into
static-inline stubs.
Signed-off-by: NBen Hutchings <ben@decadent.org.uk>
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

c4298801

21 12月, 2010 18 次提交
- B
  drm/nvc0: accelerate ttm buffer moves · 183720b8
  由 Ben Skeggs 提交于 12月 09, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  183720b8
- B
  drm/nvc0: initial support for tiled buffer objects · 6d86951a
  由 Ben Skeggs 提交于 12月 08, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  6d86951a
- B
  drm/nvc0: implement fbcon acceleration · ddbaf79a
  由 Ben Skeggs 提交于 11月 24, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  ddbaf79a
- B
  drm/nvc0: implement pgraph engine hooks · 966a5b7d
  由 Ben Skeggs 提交于 11月 24, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  966a5b7d
- B
  drm/nvc0: implement pfifo engine hooks · b2b09938
  由 Ben Skeggs 提交于 11月 24, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  b2b09938
- B
  drm/nvc0: implement fencing · 529c4959
  由 Ben Skeggs 提交于 11月 24, 2010
```
Just simple REF_CNT fencing for the moment.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  529c4959
- B
  drm/nvc0: fix channel dma init paths · 96545299
  由 Ben Skeggs 提交于 11月 24, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  96545299
- B
  drm/nvc0: skip dma object creation for drm channel · 5216782b
  由 Ben Skeggs 提交于 11月 24, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  5216782b
- B
  drm/nvc0: implement channel structure initialisation · effd6e06
  由 Ben Skeggs 提交于 11月 24, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  effd6e06
- B
  drm/nvc0: gpuobj_new need only check validity and init the relevant engine · 7460d703
  由 Ben Skeggs 提交于 11月 24, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  7460d703
- B
  drm/nvc0: reject the notifier_alloc ioctl · 587107b6
  由 Ben Skeggs 提交于 11月 24, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  587107b6
- B
  drm/nvc0: create shared channel vm · 99805566
  由 Ben Skeggs 提交于 11月 16, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  99805566
- B
  drm/nvc0: initial vm implementation, use for bar1/bar3 management · 8984e046
  由 Ben Skeggs 提交于 11月 15, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  8984e046
- B
  drm/nvc0: import initial vm backend · 4c74eb7f
  由 Ben Skeggs 提交于 11月 10, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  4c74eb7f
- B
  drm/nouveau: modify vm to accomodate dual page tables for nvc0 · 3ee01281
  由 Ben Skeggs 提交于 12月 15, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  3ee01281
- B
  drm/nv50: add missing license header to nv50_fbcon.c · 147dc38e
  由 Ben Skeggs 提交于 12月 16, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  147dc38e
- B
  drm/nv50: fix smatch warning in nv50_vram.c · 910d1b3a
  由 Ben Skeggs 提交于 12月 21, 2010
```
Reported-by: NDan Carpenter <error27@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  910d1b3a
- D
  drm/nouveau: sizeof() vs ARRAY_SIZE() · d095e232
  由 Dan Carpenter 提交于 12月 20, 2010
```
ARRAY_SIZE() was intended here, sizeof() is too large.
Signed-off-by: NDan Carpenter <error27@gmail.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  d095e232
16 12月, 2010 5 次提交

drm/radeon/kms: properly print ontario chip id · b08ebe7e

由 Alex Deucher 提交于 12月 03, 2010

Signed-off-by: NAlex Deucher <alexdeucher@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

b08ebe7e

A
drm/radeon/kms: fix vram start calculation on ontario (v2) · b4183e30
由 Alex Deucher 提交于 12月 15, 2010
```
Signed-off-by: NAlex Deucher <alexdeucher@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>
```
b4183e30

drm/ttm: delay freeing of old node during move_memcpy until after iounmap · b921bae2

由 Ben Skeggs 提交于 12月 16, 2010

Drivers using their own implementation of io_mem_reserve/io_mem_free are
likely to store the tracking information for the map in mem.mm_node, so
it can't be freed while still mapped.

Signed-off-by: Ben Skeggs<bskeggs@redhat.com>
Reviewed-by: NThomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

b921bae2

drm/radeon: add initial tracepoint support. · 99ee7fac

由 Dave Airlie 提交于 11月 23, 2010

this adds a bo create, and fence seq tracking tracepoints.

This is just an initial set to play around with, we should investigate
what others we need would be useful.
Signed-off-by: NDave Airlie <airlied@redhat.com>

99ee7fac

drm/nv50: fix a couple of vm init issues · c45aadab

由 Francisco Jerez 提交于 12月 16, 2010

Fixes overwriting the first page table entry when testing that the PRAMIN
BAR can be correctly read/written, and adds an additional bar flush after
poking the BAR3 control regs.
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

c45aadab

08 12月, 2010 14 次提交

F
drm/nv04-nv40: Fix up PCI(E) GART DMA object bus address calculation. · fd70b6cd
由 Francisco Jerez 提交于 12月 08, 2010
```
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
fd70b6cd

drm/nouveau: kick vram functions out into an "engine" · 60d2a88a

由 Ben Skeggs 提交于 12月 06, 2010

NVC0 will be able to share some of nv50's paths this way.  This also makes
it the card-specific vram code responsible for deciding if a given set
of tile_flags is valid, rather than duplicating the allowed types in
nv50_vram.c and nouveau_gem.c
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

60d2a88a

B
drm/nouveau: allow gpuobj vinst to be a virtual address when necessary · 34cf01bc
由 Ben Skeggs 提交于 11月 22, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
34cf01bc
B
drm/nv50: tidy up PCIEGART implementation · b571fe21
由 Ben Skeggs 提交于 11月 16, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
b571fe21
B
drm/nv50: enable non-contig vram allocations where requested · 5f6fdca5
由 Ben Skeggs 提交于 11月 12, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
5f6fdca5
B
drm/nv50: enable 4KiB pages for small vram allocations · bfd83aca
由 Ben Skeggs 提交于 11月 12, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
bfd83aca

drm/nv50: implement global channel address space on new VM code · 4c136142

由 Ben Skeggs 提交于 11月 15, 2010

As of this commit, it's guaranteed that if an object is in VRAM that its
GPU virtual address will be constant.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

4c136142

B
drm/nv50: implement BAR1/BAR3 management on top of new VM code · f869ef88
由 Ben Skeggs 提交于 11月 15, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
f869ef88
B
drm/nv50: import new vm code · a11c3198
由 Ben Skeggs 提交于 8月 27, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
a11c3198

drm/nv50: implement custom vram mm · 573a2a37

由 Ben Skeggs 提交于 8月 25, 2010

This is required on nv50 as we need to be able to have more precise control
over physical VRAM allocations to avoid buffer corruption when using
buffers of mixed memory types.

This removes some nasty overallocation/alignment that we were previously
using to "control" this problem.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

573a2a37

F
drm/nouveau: Avoid potential race between nouveau_fence_update() and context takedown. · 937c3471
由 Francisco Jerez 提交于 12月 08, 2010
```
Signed-off-by: NFrancisco Jerez <currojerez@riseup.net>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
937c3471

drm/nouveau: fix use of drm_mm_node in semaphore object · a8b214f0

由 Ben Skeggs 提交于 12月 03, 2010

At some point in the future, this bo won't necessarily be backed by
a drm_mm_node, so use the start/size fields of the ttm_mem_reg instead.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

a8b214f0

drm/nouveau: wrap calls to ttm_bo_validate() · 7a45d764

由 Ben Skeggs 提交于 11月 22, 2010

This will be used later to fixup bo.offset with a buffer's fixed GPU
virtual address.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

7a45d764

B
drm/nouveau: no need to zero dma objects, we fill them completely anyway · a0fd9b9f
由 Ben Skeggs 提交于 11月 26, 2010
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
a0fd9b9f

openanolis / cloud-kernel 1 年多 前同步成功

openanolis / cloud-kernel
1 年多前同步成功