提交 · 570889dc5070e1f98b5898dce426f970c9b9329b · openeuler / Kernel

02 11月, 2017 1 次提交

drm/nouveau/bar: modify interface to bar1 vmm mapping · 570889dc

由 Ben Skeggs 提交于 11月 01, 2017

Upcoming changes will remove the nvkm_vmm pointer from nvkm_vma, instead
requiring it to be explicitly specified on each operation.

It's not currently possible to get this information for BAR1 mappings,
so let's fix that ahead of time.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

570889dc

17 5月, 2017 1 次提交

drm/nouveau/fifo/gk104-: Silence a locking warning · 2579b8b0

由 Dan Carpenter 提交于 4月 27, 2017

Presumably we can never actually hit this return, but static checkers
complain that we should unlock before we return.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

2579b8b0

17 2月, 2017 10 次提交

drm/nouveau/fifo/gk104-: trigger mmu fault before attempting engine recovery · 3ebef76a

由 Ben Skeggs 提交于 1月 18, 2017

Greatly improves the chances of recovering the GPU from a CTXSW_TIMEOUT.

Tested with piglit's arb_shader_image_load_store-atomicity, which causes
GR to hang in such a way that recovery failed (CTXSW_TIMEOUT continually
re-triggers).
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

3ebef76a

B
drm/nouveau/fifo/gk104-: ACK SCHED_ERROR before attempting CTXSW_TIMEOUT recovery · 03f16f5f
由 Ben Skeggs 提交于 1月 18, 2017
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
03f16f5f
B
drm/nouveau/fifo/gk104-: directly use new recovery code for ctxsw timeout · 91b9d659
由 Ben Skeggs 提交于 1月 18, 2017
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
91b9d659
B
drm/nouveau/fifo/gk104-: directly use new recovery code for mmu faults · 3534821d
由 Ben Skeggs 提交于 1月 18, 2017
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
3534821d
B
drm/nouveau/fifo/gk104-: reset all engines a killed channel is still active on · eaa5ed65
由 Ben Skeggs 提交于 1月 18, 2017
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
eaa5ed65

drm/nouveau/fifo/gk104-: refactor recovery code · 0faaa47d

由 Ben Skeggs 提交于 1月 18, 2017

This will serve as a basis for implementing some improvements to how
we recover the GPU from channel errors.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

0faaa47d

drm/nouveau/fifo/gk104-: better detection of chid when parsing engine status · ec5c6bda

由 Ben Skeggs 提交于 1月 16, 2017

The previous commit simply changes the interface, but should result in
the same behaviour as previously. This commit has been split out from
it as it can result in a different channel being selected.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

ec5c6bda

B
drm/nouveau/fifo/gk104-: separate out engine status parsing · b88917fe
由 Ben Skeggs 提交于 1月 16, 2017
```
We'll be wanting to reuse this logic in more places.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
b88917fe

drm/nouveau/fifo/gf100-: provide notification to user if channel is killed · ff9f29ab

由 Ben Skeggs 提交于 11月 05, 2016

There are instances (such as non-recoverable GPU page faults) where
NVKM decides that a channel's context is no longer viable, and will
be removed from the runlist.

This commit notifies the owner of the channel when this happens, so
it has the opportunity to take some kind of recovery action instead
of hanging.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

ff9f29ab

B
drm/nouveau/core/memory: distinguish between coherent/non-coherent targets · d2ee3605
由 Ben Skeggs 提交于 5月 10, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
d2ee3605

13 12月, 2016 1 次提交

drm/nouveau/fifo/gf100-: recover from host mmu faults · ec884f74

由 Ben Skeggs 提交于 12月 09, 2016

This has been on the TODO list for a while now, recovering from things
such as attempting to execute a push buffer or touch a semaphore in an
unmapped memory area.

The only thing required on the HW side here is that the offending
channel is removed from the runlist, and *not* a full reset of PFIFO.

This used to be a bit messier to handle before the rework to make use
of engine topology info, but is apparently now trivial.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

ec884f74

14 7月, 2016 2 次提交
- B
  drm/nouveau/fifo/gk104-: translate engidx into human-readable name in debug output · 1fe8c02f
  由 Ben Skeggs 提交于 6月 22, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  1fe8c02f
- B
  drm/nouveau/top: take nvkm_device as argument to public functions · 952eb819
  由 Ben Skeggs 提交于 5月 30, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  952eb819
20 5月, 2016 5 次提交
- B
  drm/nouveau/fifo/gk104-: identify mmu engine ids for host faults · 289e0827
  由 Ben Skeggs 提交于 4月 08, 2016
```
It appears these don't map to PBDMAs (at least on Kepler, it may or may
be valid for Fermi - this hasn't been checked), but to runlists.

This drops the NVKM_ENGINE_FIFO data from the entries too, as resetting
all of PFIFO is *not* the way to handle such faults.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  289e0827
- B
  drm/nouveau/fifo/gk104-: implement support for PTOP fault info · e50d0237
  由 Ben Skeggs 提交于 4月 08, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  e50d0237
- B
  drm/nouveau/fifo/gk104-: abstract mmu fault data structures · 91419acf
  由 Ben Skeggs 提交于 4月 08, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  91419acf
- B
  drm/nouveau/fifo/gk104-: subclass func · 98ac3f06
  由 Ben Skeggs 提交于 4月 08, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  98ac3f06
- B
  drm/nouveau/fifo/gk104-: use device info from top subdev · e93e198d
  由 Ben Skeggs 提交于 4月 08, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  e93e198d
14 3月, 2016 12 次提交
- B
  drm/nouveau/fifo/gk104: submit NOP after all PBDMA_INTR_0, not just DEVICE · b4c5fc4b
  由 Ben Skeggs 提交于 3月 11, 2016
```
Prevents the same interrupt from re-triggering forever.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  b4c5fc4b
- B
  drm/nouveau/fifo/gk104: add nvdec plumbing · 608fd040
  由 Ben Skeggs 提交于 3月 11, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  608fd040
- B
  drm/nouveau/fifo/gk104: add nvenc plumbing · 9e4fff32
  由 Ben Skeggs 提交于 3月 11, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  9e4fff32
- B
  drm/nouveau/fifo/gk104: make use of topology info during fault recovery · 19f89279
  由 Ben Skeggs 提交于 3月 11, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  19f89279
- B
  drm/nouveau/fifo/gk104: make use of topology info when handling ctxsw timeout · af83a677
  由 Ben Skeggs 提交于 3月 11, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  af83a677
- B
  drm/nouveau/fifo/gk104: read device topology information from hw · 41e5171b
  由 Ben Skeggs 提交于 3月 11, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  41e5171b
- B
  drm/nouveau/fifo/gk104: cosmetic engine->runlist changes · 69aa40e2
  由 Ben Skeggs 提交于 3月 11, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  69aa40e2
- B
  drm/nouveau/fifo/gk104: don't attempt recovery of unknown mmu engines · acdf7d4f
  由 Ben Skeggs 提交于 3月 11, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  acdf7d4f
- B
  drm/nouveau/fifo/gk104: identify fault-recovery members more clearly · 55252da1
  由 Ben Skeggs 提交于 3月 11, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  55252da1
- B
  drm/nouveau/fifo/gk104: rename spoon to pbdma, and move detection to oneinit · 6d39b83f
  由 Ben Skeggs 提交于 3月 11, 2016
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  6d39b83f
- A
  drm/nouveau/fifo/gk104: take runlist target into account · a2e435a1
  由 Alexandre Courbot 提交于 3月 01, 2016
```
Bits 28:29 of RUNLIST_BASE specify the memory target of the runlist. Set
it to 0x3 (SYS_MEM_NONCOHERENT) if the runlist object resides in system
memory.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  a2e435a1
- X
  drm/nouveau/fifo/gk104: fix chid bit mask · 0689aad7
  由 Xia Yang 提交于 2月 25, 2016
```
Fix the channel id bit mask in FIFO schedule timeout error handling.

FIFO_ENGINE_STATUS_NEXT_ID is bit 27:16 thus 0x0fff0000.
FIFO_ENGINE_STATUS_ID      is bit 11:0  thus 0x00000fff.
Signed-off-by: NXia Yang <xiay@nvidia.com>
Reviewed-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  0689aad7
11 1月, 2016 2 次提交

drm/nouveau/fifo/gk104: fix engine status register offset · da4ee13c

由 Vince Hsu 提交于 11月 16, 2015

The offset should be 8 on Kepler and later.
Signed-off-by: NVince Hsu <vinceh@nvidia.com>
Reviewed-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

da4ee13c

drm/nouveau/fifo/gk104: fix race condition when updating engine runlists · 386ffd5e

由 Ben Skeggs 提交于 11月 11, 2015

The CPU-side tracking of engine runlists was not protected by a lock,
leading to list corruption, eventually causing runlist_update() to
overrun the GPU-side runlist, triggering an OOPS.

Fixes some of the issues noticed during parallel piglit runs.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

386ffd5e

28 8月, 2015 6 次提交
- B
  drm/nouveau/object: merge with handle · fbd58ebd
  由 Ben Skeggs 提交于 8月 20, 2015
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  fbd58ebd
- B
  drm/nouveau/core: remove the remainder of the previous style · 68f3f702
  由 Ben Skeggs 提交于 8月 20, 2015
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  68f3f702
- B
  drm/nouveau/fifo: convert to new-style nvkm_engine · 13de7f46
  由 Ben Skeggs 提交于 8月 20, 2015
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  13de7f46
- B
  drm/nouveau/bar: convert to new-style nvkm_subdev · 32932281
  由 Ben Skeggs 提交于 8月 20, 2015
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  32932281
- B
  drm/nouveau/fifo: convert user classes to new-style nvkm_object · 8f0649b5
  由 Ben Skeggs 提交于 8月 20, 2015
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  8f0649b5
- B
  drm/nouveau/fifo: split user classes out from engine implementations · 9a65a38c
  由 Ben Skeggs 提交于 8月 20, 2015
```
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>
```
  9a65a38c

openeuler / Kernel 接近 2 年 前同步成功

openeuler / Kernel
接近 2 年前同步成功