Commit Graph

415545 Commits

Author SHA1 Message Date
Ben Skeggs 46bf1c389f drm/nve0/fb: multi-stage reclock is required for certain transitions
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:39:00 +10:00
Ben Skeggs 1789cab4ef drm/nouveau/clk: allow fb to signal it needs to do a multi-stage reclock
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:59 +10:00
Ben Skeggs b655f2bb77 drm/nve0/fb/gddr5: parse bios data into struct rather than using directly
Still essentially a struct of magic values with magic names and unknown
purposes.  But, we will shortly need to be able to mix and match bits of
the previous and next configurations to do a transition reclock, as such,
we can no longer directly use the vbios data with any ease.

This is probably nicer anyway in the long run, for a few reasons.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:58 +10:00
Ben Skeggs ea8b4a380d drm/nve0/fb/gddr5: found LP3 setting
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:57 +10:00
Ben Skeggs 971372eac1 drm/nve0/fb: note the memory voltage toggle, not using it yet
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:56 +10:00
Ben Skeggs db6735cab2 drm/nve0/fb/gddr5: somewhat better attempt at 100770/10f604/610/614
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>

fb/gddr5/nve0: 100770 is like 10f604

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:56 +10:00
Ben Skeggs f4aa2c6677 drm/nve0/fb/gddr5: fixup delays a bit
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:55 +10:00
Ben Skeggs 1522ecae65 drm/nouveau/bios: timing 2.0 entries can have subentries
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:54 +10:00
Ben Skeggs 09692e5b4e drm/nve0/fb/gddr5: note another semi-unknown
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:53 +10:00
Ben Skeggs 1e1d6b4c53 drm/nouveau/fb/gddr5: modify mr8 with high bits of CL/WR
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:52 +10:00
Ben Skeggs e7084c669b drm/nve0/fb/gddr5: fix calculation of RDQS setting
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:52 +10:00
Ben Skeggs 334565abfe drm/nve0/fb/gddr5: switch off some other random bit at some point
As seen when comparing us vs nv on my GTX660

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:51 +10:00
Ben Skeggs 01891690e8 drm/nve0/fb/gddr5: punt all 10f910/914 accesses through ram_train
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:50 +10:00
Ben Skeggs d394fb12ec drm/nve0/fb/gddr5: not all memory partitions are created equal
As seen when comparing us vs nv on my GTX660.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:49 +10:00
Ben Skeggs dd95c8f782 drm/nve0/fb: typo in register name
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:48 +10:00
Ben Skeggs 0a0dc8f564 drm/nouveau/bios: make common code to handle ramcfg strap etc
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:48 +10:00
Ben Skeggs 5905439224 drm/nve0/fb/gddr5: fix an assumption of sane memory controller layout
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:47 +10:00
Ben Skeggs 2daaf5b0e4 drm/nve0/fb/gddr5: fix behaviour of lp3 setting
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:46 +10:00
Ben Skeggs cb1567c286 drm/nve0/fifo: recover from mmu faults on bar1/bar3
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:45 +10:00
Ben Skeggs 649ec925ae drm/nve0/fifo: keep mmu fault interrupts enabled at all times
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:44 +10:00
Ben Skeggs e1b6b14ad5 drm/nve0/fifo: update human-readable mmu fault descriptions
Ordering from Android GK20A driver, names from binary driver strings.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:44 +10:00
Ben Skeggs e9fb9805ad drm/nve0/fifo: document more intr status bits
As per Android GK20A driver.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:43 +10:00
Ben Skeggs 9f8459cf06 drm/nve0/fifo: populate PBDMA status bitfield with more definitions
As per Android GK20A driver.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:42 +10:00
Ben Skeggs 39b055427e drm/nve0/fifo: s/subfifo/PBDMA/
As per Android GK20A driver.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:41 +10:00
Ben Skeggs f82c44a78f drm/nve0/fifo: s/playlist/runlist/
As per Android GK20A driver.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:40 +10:00
Ben Skeggs f76dd80f76 drm/nvf0/gr: enable acceleration with our chsw ucode
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:40 +10:00
Ben Skeggs aa97cd31b6 drm/nv108/gr: enable acceleration with our chsw ucode
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:39 +10:00
Ben Skeggs 5d91e1916d drm/nvc0-/gr: handle fwmthd interrupts in ucode
Compute code in mesa triggers one of these, hanging the engine.  Let's
at least ack the request for now to avoid the hang.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:38 +10:00
Ben Skeggs e1b22bc148 drm/nvc0-/gr: fiddle some magic around strand init
Fixes HUB_INIT timeout on GK110/GK208 when not using NVIDIA's ucode.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:37 +10:00
Ben Skeggs 96616b4caf drm/nv108/gr: initial support (need external fuc)
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:36 +10:00
Ben Skeggs daa9ab5821 drm/nv108/ce: enable copy engines
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:35 +10:00
Ben Skeggs a763951a86 drm/nv108/fifo: initial support
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:35 +10:00
Ben Skeggs a0f95f11a3 drm/nvf0/gr: remove a copy+pasto in ctx reglist
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:34 +10:00
Ben Skeggs 67af60f0aa drm/nvc0-/gr: bring in some macros to abstract falcon isa differences
Need. A. Compiler...

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:33 +10:00
Ilia Mirkin 90d6db1635 drm/nouveau/falcon: use vmalloc to create firwmare copies
Some firmware images may be large (64K), so using kmalloc memory is
inappropriate for them. Use vmalloc instead, to avoid high-order
allocation failures.

Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Cc: stable@vger.kernel.org
2014-01-23 13:38:32 +10:00
Ben Skeggs d96bf437b6 drm/nouveau/gem: remove (now) unneeded pre-validate fence sync
Now that nouveau_bo.c can handle sync when it actually needs to, we can
remove this and avoid a double semaphore acquire when syncing in the
command submission path.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:31 +10:00
Ben Skeggs cef9e99e1e drm/nouveau/ttm: explicitly wait for bo idle before memcpy buffer move
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:30 +10:00
Ben Skeggs 35b8141b82 drm/nouveau/ttm: explicity sync with kernel channel before moving buffer
The GEM code handles this currently, but that'll be removed.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:30 +10:00
Ben Skeggs 3c57d85d7d drm/nouveau/ttm: tidy up creation of temporary buffer move vmas
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:29 +10:00
Ilia Mirkin ab9b18a6a1 drm/nv04/plane: add support for nv04/nv05 video overlay
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:28 +10:00
Ilia Mirkin 7ffb078172 drm/nv10/plane: add YUYV support
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:27 +10:00
Maarten Lankhorst a554090664 drm/nv50-: map TTM_PL_SYSTEM through a BAR for CPU access
Moves bo's to TTM_PL_TT for BAR mapping, to hide tiling from user.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:26 +10:00
Maarten Lankhorst ce8f7699f2 drm/nouveau: fix m2mf copy to tiled gart
Commit de7b7d59d5 introduced tiled GART, but a linear copy is
still performed. This may result in errors on eviction, fix it by
checking tiling from memtype.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Cc: stable@vger.kernel.org #3.10+
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:25 +10:00
Ben Skeggs 2e2cfbe61b drm/nouveau/vm: reduce number of entry-points to vm_map()
Pretty much everywhere had to make the decision which to use, so it
makes a lot more sense to just have one entrypoint decide the path
to take instead.

Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2014-01-23 13:38:25 +10:00
Takashi Iwai 2510538fa0 drm/cirrus: correct register values for 16bpp
When the mode is set with 16bpp on QEMU, the output gets totally broken.
The culprit is the bogus register values set for 16bpp, which was likely
copied from from a wrong place.

Addresses https://bugzilla.novell.com/show_bug.cgi?id=799216

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: David Airlie <airlied@linux.ie>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2014-01-23 11:01:40 +10:00
Jeff Mahoney d0ce7b8567 drm/nouveau: make vga_switcheroo code depend on VGA_SWITCHEROO
Commit 8116188fde ("nouveau/acpi: hook up to the MXM method for mux
switching.") broke the build on non-x86 architectures due to the new
dependency on MXM and MXM being an x86 platform driver.

It built previously since the vga switcheroo registration routines were
zereod out on !X86.  The code was built in but unused.

This patch makes all of the DSM code depend on CONFIG_VGA_SWITCHEROO,
allowing it to build on non-x86 and shrinking the module size as well.

[rdunlap@infradead.org: fix build eror when VGA_SWITCHEROO is not enabled]
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: David Airlie <airlied@linux.ie>
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
2014-01-23 11:01:40 +10:00
Dave Airlie 918be888d6 drm/mgag200: on cards with < 2MB VRAM default to 16-bit
This aligns with what the userspace -mga driver does in
the same situation.

Signed-off-by: Dave Airlie <airlied@redhat.com>
2014-01-23 11:01:35 +10:00
Kenneth Graunke 4318101121 drm/i915: Allow reading the TIMESTAMP register on Gen8.
Nothing's changed here; we just need to bump the generation check.

Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-01-22 10:48:09 +01:00
Chris Wilson 74e21ac2cc drm/i915: Repeat evictions whilst pageflip completions are outstanding
Since an old pageflip will keep its scanout buffer object pinned until
it has executed its unpin task on the common workqueue, we can clog up
our GGTT with stale pinned objects. As we cannot flush those workqueues
without dropping our locks, we have to resort to falling back to
userspace and telling them to repeat the operation in order to have a
chance to run our workqueues and free up the required memory. If we
fail, then we are forced to report ENOSPC back to userspace causing the
operation to fail and best-case scenario is that it introduces temporary
corruption.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Jon Bloomfield <jon.bloomfield@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-01-22 10:34:41 +01:00
Chris Wilson 5dce5b9387 drm/i915: Wait for completion of pending flips when starved of fences
On older generations (gen2, gen3) the GPU requires fences for many
operations, such as blits. The display hardware also requires fences for
scanouts and this leads to a situation where an arbitrary number of
fences may be pinned by old scanouts following a pageflip but before we
have executed the unpin workqueue. This is unpredictable by userspace
and leads to random EDEADLK when submitting an otherwise benign
execbuffer. However, we can detect when we have an outstanding flip and
so cause userspace to wait upon their completion before finally
declaring that the system is starved of fences. This is really no worse
than forcing the GPU to stall waiting for older execbuffer to retire and
release their fences before we can reallocate them for the next
execbuffer.

v2: move the test for a pending fb unpin to a common routine for
later reuse during eviction

Reported-and-tested-by: dimon@gmx.net
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=73696
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Jon Bloomfield <jon.bloomfield@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-01-22 10:34:40 +01:00