Commit Graph

754348 Commits

Author SHA1 Message Date
Ben Skeggs
2b297b0d6d drm/nouveau/gr/gf100-: virtualise init_40601c
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:23 +10:00
Ben Skeggs
3ac72e98b4 drm/nouveau/gr/gf100-: virtualise init_ds_hww_esr_2
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
2585a1b131 drm/nouveau/gr/gf100-: virtualise init_fecs_exceptions + apply fixes from traces
The value for GF100 has changed here, but it matches RM now.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
7c76ebb65a drm/nouveau/gr/gf100: write 0x400124 during init
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
dff30dbd1d drm/nouveau/gr/gf100-: virtualise init_swdx_pes_mask
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
0f78acc86b drm/nouveau/gr/gf100-: implement another chunk of bios-provided init
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
429412e231 drm/nouveau/gr/gf100-: virtualise init_rop_active_fbps
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
bfd27f39b5 drm/nouveau/gr/gf100-: virtualise init_num_active_ltcs
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
02917aa39d drm/nouveau/gr/gf100-: virtualise init_zcull
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
2fe5ff6371 drm/nouveau/gr/gf100-: virtualise init_vsc_stream_master
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
a37279e94c drm/nouveau/gr/gf100-: virtualise init_bios
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
cd9662f89e drm/nouveau/gr/gf100-: support clkgate_pack everywhere
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
8b058ca518 drm/nouveau/gr/gf100-: virtualise r405a14
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
6f63a5fb1e drm/nouveau/gr/gf100-: support firmware-provided sw_nonctx everywhere
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
1246f1dc22 drm/nouveau/gr/gf100-: virtualise init_gpc_mmu + apply fixes from traces
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
334cc26d4d drm/nouveau/fifo/gp100-: force individual channels into a channel group
RM does this for some reason, and is enforced in HW on Volta.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
eda12417d3 drm/nouveau/fifo/gm107-: write instance address in channel runlist entry
RM does this for some reason.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
79bb4b617f drm/nouveau/fifo/gk208-: write pbdma timeout regs during initialisation
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
8c4e9f9dff drm/nouveau/fifo/gk110-: support writing channel group runlist entries
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:22 +10:00
Ben Skeggs
4f2fc25c0f drm/nouveau/fifo/gk104-: poll for runlist update completion
Newer HW doesn't appear to send this event, which will cause long delays
in runlist updates if they don't complete immediately.

RM doesn't use these events anywhere, and an NVGPU commit message notes
that polling is the preferred method even on HW that supports the event.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
665870837a drm/nouveau/fifo/gk104-: add interfaces to support different runlist layouts
This will be required to support features on newer hardware.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
f9360c3aa6 drm/nouveau/fifo/gk104-: simplify definition of channel classes
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
a7cf01809b drm/nouveau/fifo/gk104-: require explicit runlist selection for channel allocation
We didn't used to be aware that runlist/engine IDs weren't the same thing,
or that there was such variability in configuration between GPUs.

By exposing this information to a client, and giving it explicit control
of which runlist it's allocating a channel on, we're able to make better
choices.

The immediate effect of this is that on GPUs where CE0 is the "GRCE", we
will now be allocating a copy engine running asynchronously to GR for BO
migrations - as intended.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
cc36205085 drm/nouveau/fifo/gk104-: support querying engines available on each runlist
Will be used to improve channel runlist selection.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
ddc669e256 drm/nouveau/fifo/gk104-: allow fault recovery code to be called by other subdevs
This will be required to support Volta.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
55b8e85b0b drm/nouveau/fifo/gk104-: accept engine contexts for CE3 and up
These can exist on GP100 and newer.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
eb47db4f3b drm/nouveau/fifo: support channel count query
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
6eb01aa898 drm/nouveau/device: support querying available engines of a specific type
Will be used for fifo runlist selection.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
c5c9127b25 drm/nouveau/device: implement a generic method to query device-specific properties
We have a need to fetch data from GPU-specific sub-devices that is not
tied to any particular engine object.

This commit provides the framework to support such queries.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
f5650478ab drm/nouveau/disp/nv50-: pass nvkm_memory objects for channel push buffers
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
a9c44a88ca drm/nouveau/disp/nv50-: add channel interfaces to control error interrupts
This will be required to support Volta, but also allows us to remove code
that's duplicated for each channel type already.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
4a8621a24a drm/nouveau/disp/nv50-: add channel interfaces to determine the user area
This will be required to support Volta.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
8531f57027 drm/nouveau/disp/nv50-: merge handling of pio and dma channels
Unnecessarily complicated, and a barrier to cleanly supporting Volta.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
9b096283bf drm/nouveau/disp/nv50-: simplify definiton of core channels
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
6d41a7536f drm/nouveau/disp/nv50-: simplify definition of cursor channels
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
3ceeef9c03 drm/nouveau/disp/nv50-: simplify definition of base channels
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:21 +10:00
Ben Skeggs
c2c3a00310 drm/nouveau/disp/nv50-: simplify definition of overlay immediate channels
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
46f74a8ad7 drm/nouveau/disp/nv50-: simplify definition of overlay channels
Introduces a new method of defining channels available from the display,
common to all channel types, allowing for more flexibility in available
channel types/counts, and reducing the amount of boiler-plate required.

This will be required to support Volta.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
abc1d4379b drm/nouveau/disp/nv50-: replace user object with engine pointer in channels
More simplification.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
bb3b0a4220 drm/nouveau/disp/nv50-: initialise from the engine, rather than the user object
Engines are initialised on an as-needed basis, so this results in the
same behaviour, whilst allowing us to simplify things a bit.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
f5e088d6f0 drm/nouveau/disp/nv50-: fetch mask of available piors during oneinit
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
9fe4e17704 drm/nouveau/disp/nv50-: fetch mask of available sors during oneinit
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
bf5d1a6b6a drm/nouveau/disp/nv50-: fetch mask of available dacs during oneinit
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
f7b2ece37f drm/nouveau/disp/nv50-: fetch mask of available heads during oneinit
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
3b9ba66ab0 drm/nouveau/disp/nv50-: delay subunit construction until oneinit
We should be reading registers to determine which subunits are really
present on a given board, and this needs to be done after DEVINIT.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
85a3b9c839 drm/nouveau/fb/gm200-: fix overwriting of big page setting
Likely a rebase bug.  Should have no impact in default configuration due
to using per-instance setting by default.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
d1ea77ab5f drm/nouveau/fb/gf100-: bump size of mmu debug buffers to match big page size
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
d0e9351e42 drm/nouveau/fault/gp100: implement replayable fault buffer initialisation
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
36780d7eee drm/nouveau/fault: add infrastructure to support fault buffers
GPU-specific support will be added separately.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00
Ben Skeggs
2f68234fb3 drm/nouveau/mc/gp100-: route fault buffer interrupts to FAULT
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2018-05-18 15:01:20 +10:00