Commit Graph

7003 Commits

Author SHA1 Message Date
Ben Skeggs
1a46098e91 drm/nvc0/ttm: use copy engines for async buffer moves
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:23 +10:00
Ben Skeggs
fdf53241c1 drm/nva3/ttm: use copy engine for async buffer moves
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:21 +10:00
Ben Skeggs
5490e5dfb9 drm/nv98/ttm: add in a (disabled) crypto engine buffer copy method
Disabled for the moment until some performance issues are sorted out, code
committed as a reference point.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:19 +10:00
Ben Skeggs
4c193d254e drm/nv84/ttm: use crypto engine for async buffer copies
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:18 +10:00
Ben Skeggs
d1b167e168 drm/nouveau/ttm: untangle code to support accelerated buffer moves
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:16 +10:00
Ben Skeggs
78df3a1c58 drm/nouveau/fbcon: use fence for sync, rather than notifier
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:15 +10:00
Ben Skeggs
b355096992 drm/nv98/crypt: non-stub implementation of the engine hooks
fuc is from pscnv driver.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:13 +10:00
Ben Skeggs
c420b2dc8d drm/nouveau/fifo: turn all fifo modules into engine modules
Been tested on each major revision that's relevant here, but I'm sure there
are still bugs waiting to be ironed out.

This is a *very* invasive change.

There's a couple of pieces left that I don't like much (eg. other engines
using fifo_priv for the channel count), but that's an artefact of there
being a master channel list still.  This is changing, slowly.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:11 +10:00
Ben Skeggs
a226c32a38 drm/nv50/graph: remove ability to do interrupt-driven context switching
We never turn this on, no point maintaining the code for it..

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:10 +10:00
Ben Skeggs
5511d490da drm/nv50: remove manual context unload on context destruction
PFIFO context destruction triggers this automagically now.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:08 +10:00
Ben Skeggs
7f2062e9de drm/nv50: remove execution engine context saves on suspend
Now triggered automagically by the GPU on PFIFO takedown.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:07 +10:00
Ben Skeggs
03bd6efa14 drm/nv50/fifo: use hardware channel kickoff functionality
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:05 +10:00
Ben Skeggs
71af5e62db drm/nv50/gr: make sure NEXT_TO_CURRENT is executed even if nothing done
PFIFO channel kickoff will hang sometimes otherwise.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:03 +10:00
Ben Skeggs
694931d20f drm/nv50/fifo: construct playlist from hw context table state
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:02 +10:00
Ben Skeggs
67b342efc7 drm/nouveau/fifo: remove all the "special" engine hooks
All the places this stuff is actually needed tends to be chipset-specific
anyway, so we're able to just inline the register bashing instead.

The parts of the common code that still directly touch PFIFO temporarily
have conditionals, these will be removed in subsequent commits that will
refactor the fifo modules into engine modules like graph/mpeg etc.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:56:00 +10:00
Ben Skeggs
906c033e27 drm/nouveau/fence: fix a race where fence->channel can disappear
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:55:58 +10:00
Ben Skeggs
299bee10fb drm/nouveau/bios: fix some shadowing issues, particularly acpi
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:55:56 +10:00
Ben Skeggs
f51ee65c75 drm/nouveau: fix engine context destructor ordering
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:55:55 +10:00
Ben Skeggs
5e120f6e4b drm/nouveau/fence: convert to exec engine, and improve channel sync
Now have a somewhat simpler semaphore sync implementation for nv17:nv84,
and a switched to using semaphores as fences on nv84+ and making use of
the hardware's >= acquire operation.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:55:53 +10:00
Ben Skeggs
d375e7d56d drm/nouveau/fence: minor api changes for an upcoming rework
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:55:46 +10:00
Ben Skeggs
875ac34aad drm/nouveau/fence: make ttm interfaces wrap ours, not the other way around
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:55:44 +10:00
Ben Skeggs
35bcf5d555 drm/nouveau: move flip-related channel setup to software engine
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:55:43 +10:00
Ben Skeggs
20abd1634a drm/nouveau: create real execution engine for software object class
Just a cleanup more or less, and to remove the need for special handling of
software objects.

This removes a heap of documentation on dma/graph object formats.  The info
is very out of date with our current understanding, and is far better
documented in rnndb in envytools git.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:55:41 +10:00
Ben Skeggs
2cda7f4c5e drm/nvd0/disp: remove unnecessary sync from flip_next
This shouldn't be necessary, I believe this is just a bit of missed debug
code that got left over somehow.

Causes flips to be always synced to vblank, regardless of swap interval,
which we don't want..

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:32:03 +10:00
Ben Skeggs
afada5e0bb drm/nv04/disp: disable vblank interrupts when disabling display
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:32:01 +10:00
Marcin Slusarz
695b95b810 drm/nouveau: base fence timeout on time of emission
Wait loop can be interrupted by signal, so if signals are raised
periodically (e.g. SIGALRM) this loop may never finish. Use
emission time as a base for fence timeout.

Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:59 +10:00
Ben Skeggs
d58086deaa drm/nv40-50/gr: restructure grctx/prog generation
The conditional definition of the generation helper functions apparently
confuses some IDEs....

Reported-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:58 +10:00
Ben Skeggs
a8f81837c5 drm/nv50/disp: fixup error paths in crtc object creation
Reported-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:56 +10:00
Marcin Slusarz
5ace2c9d6f drm/nouveau: cleanup after display init failure
Depending on exact point of failure, not cleaning would lead to
BUG_ONs/oopses in various distant places.

Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:54 +10:00
Marcin Slusarz
d37f60c87f drm/nv50: fix ramin heap size for kernel channel too
Port change from "drm/nouveau: Keep RAMIN heap within the channel"
to kernel channel, which has its own ramin heap initialisation.

Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Cc: Younes Manton <younes.m@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:52 +10:00
Ben Skeggs
d8b6624549 drm/nve0/graph: bump hub2gpc buffer size
Reported-by: Christoph Bumiller <e0425955@student.tuwien.ac.at>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:50 +10:00
Ben Skeggs
6d59702775 drm/nouveau: use the same packet header macros as userspace
Cosmetic cleanup only.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:49 +10:00
Ben Skeggs
78339fb75c drm/nouveau/bios: allow loading alternate vbios image as firmware
Useful for debugging different VBIOS versions.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:47 +10:00
Ben Skeggs
c6b7e89582 drm/nve0/ttm: implement buffer moves with weirdo pcopy-on-pgraph methods
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:45 +10:00
Ben Skeggs
f1c65e7c7f drm/nv50-/fbcon: move 2d class to subchannel 3
Kepler GRAPH has (well, sorta) fixed subchannel<->class assignments, make
this match up to keep it happy without trapping.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:43 +10:00
Ben Skeggs
ab394543dd drm/nve0/gr: initial implementation
This may, perhaps, get re-merged with nvc0_graph.c at some point.  It's
still unclear as to how great an idea that'd be.  Stay tuned...

Completely dependent on firmware blobs from NVIDIA binary driver currently.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:41 +10:00
Ben Skeggs
5132f37700 drm/nve0/fifo: initial implementation
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:39 +10:00
Ben Skeggs
d0f3c7e41d drm/nouveau: give a slightly larger pci(e)gart aperture on all chipsets
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:38 +10:00
Ben Skeggs
78c2018658 drm/nouveau/pm: some more delays for ddr3 reclocking
These numbers from the binary driver's daemon scripts, and fix the transition
to perflvl 3 on my T510.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:36 +10:00
Ben Skeggs
9d6ba0b58c drm/nvc0/pm: very initial mclk freq change
Loads of magic missing, this will probably blow up if you try it.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:34 +10:00
Ben Skeggs
a94ba1fcac drm/nvd9/pm: oops, fix timing calc
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:32 +10:00
Ben Skeggs
6b91d6b056 drm/nvc0/pm: enable mpll src pll, and calc mpll coefficients
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:30 +10:00
Ben Skeggs
a1da205f42 drm/nvc0/pm: start filling in memory reclocking stubs 2012-05-24 16:31:29 +10:00
Ben Skeggs
19a1e47799 drm/nva3/pm: another few magic regs, and slightly better 0x004018 handling
Not entirely convinced 0x004018 transitions are correct yet, but, it's
an improvement.

The 750MHz value comes from fiddling with the binary driver + coolbits on
two different DDR3 NVA8 chipsets (T510 NVS3100M, and NVS300), not a clue
where this number comes from.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:27 +10:00
Ben Skeggs
2b20fd0ab4 drm/nva3/pm: initial attempt at handling 111100/111104
Probably not quite right, but this is enough now to make NVS300 reclock
between all 3 of its perflvls correctly.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:25 +10:00
Ben Skeggs
5f54d29ee9 drm/nva3/pm: make pll->pll mode work
This probably wants a cleanup, but I'm holding off until I know for sure
how the rest of the things that need doing fit together.

Tested on NVS300 by hacking up perflvl 1 to require PLL mode, and switching
between perflvl 3 and 1.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:23 +10:00
Ben Skeggs
001a3990f6 drm/nva3/pm: attempt to bash a few 0x100200 bits correctly
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:21 +10:00
Ben Skeggs
4719b55be5 drm/nva3/pm: begin to restructure memory clock changes + another magic
The binary driver appears to do various bits and pieces of the memory
clock frequency change at different times, depending on the particular
transition that's occuring.  I've attempted to replicate this here
for div->pll, pll->div and div->div transitions.

With some additional (patches upcoming) magic regs being bashed, this
allows me to correctly transition between all 3 perflvls on NVS300.

pll->pll transitions will *not* work correctly at the moment, pending
me tricking the binary driver into doing one and seeing how to correctly
handle it.

This patch also handles (hopefully) 0x1110e0, which appears to need
changing depending on whether in PLL or divider mode.. Maybe.  We'll
see.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:20 +10:00
Ben Skeggs
30e533900e drm/nva3/pm: more random unknown PFB regs
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:18 +10:00
Ben Skeggs
27740383dd drm/nva3/pm: initial attempt at more magic PFB regs
The reg calculation may get moved elsewhere at some point, but lets
figure out what exactly we need to do first.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2012-05-24 16:31:16 +10:00