With multiple rings generating requests independently, the outstanding
requests must also be track independently.
Reported-by: Wang Jinjin <jinjin.wang@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=30380Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Introduced by 48b956c5, I had thought I had already fixed this. Oh well.
Reported-by: Sitsofe Wheeler <sitsofe@yahoo.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

There is no equivalent to mutex_destroy() for spinlocks so just delete
the code.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Daniel Vetter pointed out that in this case is would be clearer and
cleaner to use a spinlock instead of a mutex to protect the per-file
request list manipulation. Make it so.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

It's the same code, essentially, so kill all copies safe one unified
version.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

All functions are extremely similar, so fold them into one generic
implementation.

This function isn't used anyway, because there's not yet a bsd ring
error state dumper.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Two macros that use a base address for HWS_PGA were missing, add them.
Also switch the remaining users of *_ACTHD to the ring-base one.
Kill the other ring-specific macros because they're now unused.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
[ickle: And silence checkpatch whilst in the vicinity]
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

This was mixed up in the following patch:

commit a6c45cf0
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Sep 17 00:32:17 2010 +0100

    drm/i915: INTEL_INFO->gen supercedes i8xx, i9xx, i965g
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Everything is now handled in intel-gtt.h so these defines
are only confusing.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

... and combine it with the wedged completion handler.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Owain Ainsworth reported an issue between the interaction of the
hangcheck and userspace immediately (and permanently) falling back to
s/w rasterisation. In order to break the mutex and begin resetting the
GPU, we must abort the current operation (usually within the wait) and
climb sufficiently far back up the call chain to drop the mutex. In his
implementation, Owain has a loop within the ioctl handler to detect the
hang and then sleep until the error handler has run. I've chosen to
return to userspace and report an EAGAIN which should trigger the
userspace ioctl handler to repeat the call (simply because it felt less
invasive...). Before hitting a wedged GPU, we then wait upon completion
of the error handler.
Reported-by: Owain G. Ainsworth <zerooa@googlemail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Avoid cause latencies in other clients by not taking the global struct
mutex and moving the per-client request manipulation a local per-client
mutex. For example, this allows a compositor to schedule a page-flip
(through X) whilst an OpenGL application is monopolising the GPU.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

This patch fixes the black screen bug on Dell e6510, by
adding two delays to give the eDP panel time to turn on before we
continue with the next write.

300ms is rather arbitray and a rather long sleep, we need to find a way
of refining this value.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=29278Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>

In commit e517a5e9 the call to
map_page_into_agp() got removed from intel_i830_setup_flush(), but the
counterpart call from intel_i830_fini_flush() to unmap_page_from_agp()
was left in place.

Additionally, the page allocated here never gets its physical address
used for sending to hardware, so there's no need to allocate it with
GFP_DMA32. Nor is __GFP_ZERO really necessary, as the page is used
only to store data to force flushing of some internal processor state.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Cc: Eric Anholt <eric@anholt.net>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

First step, lets have a look at the values for troublesome panels and
see if they may be used to improve our link training.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

We need to drain the pending flips prior to disabling the pipe during
modeset, and these need to be done in an uninterruptible fashion.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

This is now private to the DVO connector, remove it from the main device
private.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

As we presume space is signed when computing and looking for wrap along,
make it so.
Reported-by: Owain G. Ainsworth <zerooa@googlemail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

These have served their purpose and are now just noise in the debug
stream.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Try to validate the panel's connection by writing to address 0xA0.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=18072Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

We need to wait for the PLLs to settle prior to detecting the state
changes. The BIOS writers guide suggests waiting for the next vblank.
Reported-by: Carlos R. Mafra <crmafra2@gmail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

This is already performed with the pipelined flush, so by the time we
schedule the flush in the page-flip, the ring is NULL and we OOPs
instead.
Reported-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

The BIOS writer's guide suggests that a VGA connection will ACK a write
to address 0xA0 and that this should be used before doing legacy
load-detection. Considering the extreme cost of load-detection,
performing an extra DDC seems a risk worth taking.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

We are not currently using it as intended, so remove the complication.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Previously when converting the GMBUS pin to the GPIO reg, we would
offset the pin by one and then use the look-up table. Now that we first
try to use the GMBUS pin, we no longer need the offset and can use the
value from the VBIOS directly.
Reported-by: Carlos R. Mafra <crmafra2@gmail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Otherwise we will hit a list handling assertion when moving the object
to the inactive list.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

During i915_gem_create_mmap_offset() if the subsystem reports an error
code, use it.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

As we do not wait for the panel to turn off when we need to adjust the
panel-fitting registers we also need to unlock the PLLs as with the
non-pfit update path.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

In order to handle disable_functions() where the framebuffer is
decoupled from the crtc we need to unpin the fb in order to prevent a
leak.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=29857Reported-by: Sitsofe Wheeler <sitsofe@yahoo.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Suspending (especially hibernating) may take a finite amount of time,
during which a hotplug event may trigger and we will attempt to handle
it with inconsistent state. Disable hotplug polling around suspend and
resume.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=30070Reported-by: Rui Tiago Matos <tiagomatos@gmail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

The old code didn't clean up the i830 chipset flush page. And it
looks nicer.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Storing this explicitly makes for clearer code and hopefully
less further confusion.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Consolidate everything in intel-gtt.c and also kill the export
of intel_max_stolen.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

DMA remapping was only used by the intel-gtt driver. With that
code now folded into the driver, kill the agp generic support for
it.

Cc: Dave Airlie <airlied@linux.ie>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

They're now all the same.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

This is the last differentiator between the different fake agp drivers.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

That indirection mess can now go. Add a dummy i81x gtt_driver to
avoid a NULL pointer check.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>