bybrooklyn/linux
1
0
Fork 0
mirror of https://github.com/torvalds/linux.git synced 2026-04-21 08:13:56 -04:00
Code Issues Packages Projects Releases Wiki Activity
602,514 Commits 1 Branch 927 Tags
7025776ed1ebdfa1959932e7a4662c2f88607df0
Commit Graph

88 Commits

Author SHA1 Message Date
Ben Skeggs
0cdc3fdfb7 drm/nouveau/fifo/gm107-: remove engines from mmu engine mapping array 
These are specified by PTOP on Maxwell GPUs.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-05-20 14:43:04 +10:00
Ben Skeggs
289e082706 drm/nouveau/fifo/gk104-: identify mmu engine ids for host faults 
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: Ben Skeggs <bskeggs@redhat.com>
 2016-05-20 14:43:04 +10:00
Ben Skeggs
e50d0237fc drm/nouveau/fifo/gk104-: implement support for PTOP fault info 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-05-20 14:43:04 +10:00
Ben Skeggs
91419acf78 drm/nouveau/fifo/gk104-: abstract mmu fault data structures 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-05-20 14:43:04 +10:00
Ben Skeggs
98ac3f061a drm/nouveau/fifo/gk104-: subclass func 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-05-20 14:43:04 +10:00
Ben Skeggs
e93e198d46 drm/nouveau/fifo/gk104-: use device info from top subdev 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-05-20 14:43:04 +10:00
Ben Skeggs
56d06fa29e drm/nouveau/core: remove pmc_enable argument from subdev ctor 
These are now specified directly in the MC subdev.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-05-20 14:43:04 +10:00
Ben Skeggs
7c4f87c9e5 drm/nouveau/fifo/gm107: KeplerChannelGpfifoB, and 2048 channels 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:48 +10:00
Ben Skeggs
63f8c9b7f6 drm/nouveau/fifo/gk110: expose KeplerChannelGpfifoB 
This class supports a WFI method (0x0078) that's not present on the
KeplerChannelGpfifoA class.

The binary driver exposes both classes on these GPUs for some reason,
though there doesn't appear to be any difference in the setup that's
done for each (ie. even if you allocate GpfifoA, the WFI method will
still work).

We shall just expose GpfifoB, as I don't see a good reason to report
the presence of both.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:48 +10:00
Ben Skeggs
b4c5fc4b85 drm/nouveau/fifo/gk104: submit NOP after all PBDMA_INTR_0, not just DEVICE 
Prevents the same interrupt from re-triggering forever.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:47 +10:00
Ben Skeggs
4a3f63f808 drm/nouveau/fifo/gk104: add vic plumbing 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:47 +10:00
Ben Skeggs
a8b005fd52 drm/nouveau/fifo/gk104: add sec plumbing 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:46 +10:00
Ben Skeggs
608fd040b7 drm/nouveau/fifo/gk104: add nvdec plumbing 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:46 +10:00
Ben Skeggs
9e4fff3205 drm/nouveau/fifo/gk104: add nvenc plumbing 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:46 +10:00
Ben Skeggs
5d7fa4de46 drm/nouveau/fifo/gk104: add msenc plumbing 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:45 +10:00
Ben Skeggs
1f5ff7f52b drm/nouveau/fifo/gk104: make use of topology info during gpfifo construction 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:42 +10:00
Ben Skeggs
19f89279fa drm/nouveau/fifo/gk104: make use of topology info during fault recovery 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:42 +10:00
Ben Skeggs
af83a67779 drm/nouveau/fifo/gk104: make use of topology info when handling ctxsw timeout 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:41 +10:00
Ben Skeggs
41e5171ba8 drm/nouveau/fifo/gk104: read device topology information from hw 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:41 +10:00
Ben Skeggs
69aa40e276 drm/nouveau/fifo/gk104: cosmetic engine->runlist changes 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:40 +10:00
Ben Skeggs
acdf7d4f7e drm/nouveau/fifo/gk104: don't attempt recovery of unknown mmu engines 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:40 +10:00
Ben Skeggs
55252da161 drm/nouveau/fifo/gk104: identify fault-recovery members more clearly 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:39 +10:00
Ben Skeggs
6d39b83f13 drm/nouveau/fifo/gk104: rename spoon to pbdma, and move detection to oneinit 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:39 +10:00
Ben Skeggs
1015d81122 drm/nouveau/fifo/gf100: fix certain engines not being recovered after a fault 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:38 +10:00
Ben Skeggs
f22d7d45fa drm/nouveau/fifo/gf100: don't attempt recovery of unknown mmu engines 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:38 +10:00
Ben Skeggs
792662439c drm/nouveau/fifo/gf100: identify fault-recovery members more clearly 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:38 +10:00
Ben Skeggs
adbe24a21e drm/nouveau/fifo/gf100: rename spooon to pbdma, and move detection to oneinit 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:37 +10:00
Alexandre Courbot
2bf1833e51 drm/nouveau/fifo/gk104: kick channel upon removal 
A channel may still be processed by the PBDMA even after removal, unless
it is properly kicked. Some chips are more sensible to this than others,
with GM20B triggering the issue very easily (the PBDMA will try to fetch
methods from the previously-removed channel after a new one is added).

Make sure this cannot happen by kicking the channel right after it is
disabled, and before the new runlist is submitted.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:34 +10:00
Alexandre Courbot
a2e435a1b0 drm/nouveau/fifo/gk104: take runlist target into account 
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: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:31 +10:00
Alexandre Courbot
c694ecad9d drm/nouveau/fifo/gf100: take runlist target into account 
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: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:30 +10:00
Xia Yang
0689aad70d drm/nouveau/fifo/gk104: fix chid bit mask 
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: Xia Yang <xiay@nvidia.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:30 +10:00
Ben Skeggs
db1eb52846 drm/nouveau: s/gm204/gm200/ in a number of places 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-03-14 10:13:12 +10:00
Ben Skeggs
9402aec544 drm/nouveau/fifo/gf100: remove references to "daemon" 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-01-11 11:28:23 +10:00
Ben Skeggs
f01c4e682c drm/nouveau/nvif: modify nvif_unvers/nvif_unpack macros to be more obvious 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-01-11 11:17:40 +10:00
Ben Skeggs
8ed1730ccd drm/nouveau/nvif: split out fifo interface definitions 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-01-11 11:17:40 +10:00
Vince Hsu
da4ee13cd5 drm/nouveau/fifo/gk104: fix engine status register offset 
The offset should be 8 on Kepler and later.

Signed-off-by: Vince Hsu <vinceh@nvidia.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-01-11 11:17:40 +10:00
Ben Skeggs
d40d0fd487 drm/nouveau/fifo/gf100: fix race condition when updating engine runlists 
Similar in spirit to the gk104 fix with a similar title.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2016-01-11 11:17:40 +10:00
Ben Skeggs
386ffd5e80 drm/nouveau/fifo/gk104: fix race condition when updating engine runlists 
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: Ben Skeggs <bskeggs@redhat.com>
 2016-01-11 11:17:40 +10:00
Ben Skeggs
7e8820fed7 drm/nouveau/device: cleaner abstraction for device resource functions 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:48 +10:00
Ben Skeggs
fbd58ebda9 drm/nouveau/object: merge with handle 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:47 +10:00
Ben Skeggs
68f3f702b6 drm/nouveau/core: remove the remainder of the previous style 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:47 +10:00
Ben Skeggs
13de7f4629 drm/nouveau/fifo: convert to new-style nvkm_engine 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:46 +10:00
Ben Skeggs
3293228174 drm/nouveau/bar: convert to new-style nvkm_subdev 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:42 +10:00
Ben Skeggs
ef8bc5760b drm/nouveau/core: kill some (now) dead code 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:42 +10:00
Ben Skeggs
8f0649b5c6 drm/nouveau/fifo: convert user classes to new-style nvkm_object 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:40 +10:00
Ben Skeggs
9a65a38c45 drm/nouveau/fifo: split user classes out from engine implementations 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:40 +10:00
Ben Skeggs
0710cc3148 drm/nouveau/dma: convert user classes to new-style nvkm_object 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:40 +10:00
Ben Skeggs
19fef52d93 drm/nouveau/dma: split user classes out from engine implementations 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:39 +10:00
Ben Skeggs
6157091177 drm/nouveau/sw: remove dependence on namedb/engctx lookup 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:39 +10:00
Ben Skeggs
6ca307b0c9 drm/nouveau/fifo: remove dependence on namedb/engctx lookup 
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
 2015-08-28 12:40:38 +10:00