Inspired by the recent ppgtt regression report, where switching of
dmar only for the gpu seems to fix things completely, I've looked
again at the semaphores+vt-d situation.

Contrary to my earlier testing a few months back my system is now
stable with dmar disabled for the igd, and not only when disabling
dmar completely.

So I'm rather hopeful that all our recent fixes for snb have changed
things for code and it's time to try enabling semaphores again. We've
also had issues with enabling semaphores which are not vt-d related,
but I guess these are all fixed by the autoreport-disabling and lazy
request fix. And there's only one way to find out whether there are
still other issues ...

When I've tried to apply this patch I've noticed that semaphores on
gen6 have already silently been enabled in

commit 2911a35b
Author: Ben Widawsky <ben@bwidawsk.net>
Date:   Thu Apr 5 14:47:36 2012 -0700

    drm/i915: use semaphores for the display plane

Fix this up by only checking whether dmar is enabled on the gfx (not
on the entire system).
Reviewed-by: NBen Widawsky <ben@bwidawsk.net>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

While trying to fix up gen4 gpu reset in

commit f49f0586
Author: Kenneth Graunke <kenneth@whitecape.org>
Date:   Sat Sep 11 01:19:14 2010 -0700

    drm/i915: Actually set the reset bit in i965_reset

a little confusion about when wait_for times out has been introduced -
wait for loops _until_ the condition is true.

This fixes gpu reset on my gm45, testing with my hangman code shows
that it's now fairly reliable - it only died after well over 100 reset
cycles.

Cc: Eric Anholt <eric@anholt.net>
Reviewed-by: NKenneth Graunke <kenneth@whitecape.org>
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

On gen4+ we don't reset the display unit, so resetting the complete
modeset state should not be necessary.

We can't do reset on gen3 anyway, which leaves us with gen2 reset:
According to Chris Wilson, that doesn't work so great, so he suggested
we just ignore that. If the need ever arrises, we can re-add it later
on.
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

... we actually use it.

Unfortunately we can't reset both at the same time without also
resetting the display unit, so do render and media separately.

Also replace magic constants with proper #defines.
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Only half of them even cared, and it's always the same one.
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

- reset the stop_rings infrastructure while resetting the hw to
  avoid angering the hangcheck right away (and potentially declaring
  the gpu permanently wedged).

- ignore reset failures when hanging due to the hangman - we don't
  have reset code for all generations.

v2: Ensure that we only ignore reset failures when the hw reset is not
implemented and not when it failed.
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Slightly cleans up the code and could be useful for e.g. Ben
Widawsky's hw context patches.

v2: New colours!

Cc: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

- need_display is always true, scrap it.
- don't reacquire the mutex to do nothing after having restored the
  gem state.
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

gpu reset is a very important piece of our infrastructure.
Unfortunately we only really it test by actually hanging the gpu,
which often has bad side-effects for the entire system. And the gpu
hang handling code is one of the rather complicated pieces of code we
have, consisting of
- hang detection
- error capture
- actual gpu reset
- reset of all the gem bookkeeping
- reinitialition of the entire gpu

This patch adds a debugfs to selectively stopping rings by ceasing to
update the hw tail pointer, which will result in the gpu no longer
updating it's head pointer and eventually to the hangcheck firing.
This way we can exercise the gpu hang code under controlled conditions
without a dying gpu taking down the entire systems.

Patch motivated by me forgetting to properly reinitialize ppgtt after
a gpu reset.

Usage:

echo $((1 << $ringnum)) > i915_ring_stop # stops one ring

echo 0xffffffff > i915_ring_stop # stops all, future-proof version

then run whatever testload is desired. i915_ring_stop automatically
resets after a gpu hang is detected to avoid hanging the gpu to fast
and declaring it wedged.

v2: Incorporate feedback from Chris Wilson.

v3: Add the missing cleanup.

v4: Fix up inconsistent size of ring_stop_read vs _write, noticed by
Eugeni Dodonov.
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

... and put them to so good use.

Note that there's functional change in vlv clock gating code, we now
no longer spuriously read back the current value of the bit. According
to Bspec the high bits should always read zero, so ORing this in
should have no effect.
Reviewed-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

PCH PLLs aren't required for outputs on the CPU, so we shouldn't just
treat them as part of the pipe.

So split the code out and manage PCH PLLs separately, allocating them
when needed or trying to re-use existing PCH PLL setups when the timings
match.

v2: add num_pch_pll field to dev_priv (Daniel)
    don't NULL the pch_pll pointer in disable or DPMS will fail (Jesse)
    put register offsets in pll struct (Chris)

v3: Decouple enable/disable of PLLs from get/put.
v4: Track temporary PLL disabling during modeset
v5: Tidy PLL initialisation by only checking for num_pch_pll == 0 (Eugeni)
v6: Avoid mishandling allocation failure by embedding the small array of
    PLLs into the device struct

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=44309
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> (up to v2)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> (v3+)
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Tested-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Reviewed-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

After a gpu reset we need to re-init some of the hw state we only
initialize when modeset is enabled, like rc6, hw contexts or render/GT
core clock gating and workaround register settings.

Note that this patch has a small change in the resume code:
- rc6 on gen6+ is only restored for the modeset case (for more
  consistency with other callsites). This is no problem because recent
  kernels refuse to load drm/i915 without kms on gen6+
- rc6/emon on ilk is only restored for the modeset case. This is no
  problem because rc6 is disabled by default on ilk, and ums on ilk
  has never really been a supported option outside of horrible rhel
  backports.

v2: Chris Wilson noticed that we not only fail to restore the clock
gating settings after gpu reset.

v3: Move the call to modeset_init_hw in _reset out of the
struct_mutext protected area - other callers don't hold it, too.
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

In theory this will have performance and power improvements. Performance
because we don't need to stall when the scanout BO is busy, and power
because we don't have to stall when the BO is busy (and the ring can
even go to sleep if the HW supports it).

v2:
squash 2 patches into 1 (me)
un-inline the enable_semaphores function (Daniel)
remove comment about SNB hangs from i915_gem_object_sync (Chris)
rename intel_enable_semaphores to i915_semaphore_is_enabled (me)
removed page flip comment; "no why" (Chris)

To address other comments from Daniel (irc):
update the comment to say 'vt-d is crap, don't enable semaphores'
  - I think you misinterpreted Chris' comment, it already exists.
checking out whether we can pageflip on the render ring on ivb (didn't
work on early silicon)
  - We don't want to enable workarounds for early silicon unless we have
    to.
  - I can't find any references in the docs about this.
optionally use it if the fb is already busy on the render ring
  - This should be how the code already worked, unless I am
    misunderstanding your meaning.
Signed-off-by: NBen Widawsky <ben@bwidawsk.net>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

People have been getting confused and thinking this is a runtime control.

Cc: stable@vger.kernel.org
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Reviewed-by: NRodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This adds product definitions for desktop, mobile and server boards.

v2: split into a separate patch, add .has_pch_split feature.
Signed-off-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The macro is becoming too complex and with VLV upon us it can lead to
confusion. So transforming this into a feature check instead.
Signed-off-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
[danvet: fixed conflict with is_valleyview addition.]
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

i915_drm_thaw was not locking the mode_config lock when calling
drm_helper_resume_force_mode. When there were multiple wake sources,
this caused FDI training failure on SNB which in turn corrupted the
display.
Signed-off-by: NSean Paul <seanpaul@chromium.org>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Totally unexpected that this regressed. Luckily it sounds like we just
need to have dmar disable on the igfx, not the entire system. At least
that's what a few days of testing between Tony Vroon and me indicates.
Reported-by: NTony Vroon <tony@linx.net>
Cc: Tony Vroon <tony@linx.net>
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=43024Acked-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This adds PCI ID for IVB GT2 server variant which we were missing.
Signed-off-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
[danvet: fix up conflict because the patch has been diffed against next. tsk.]
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This allows to select which rc6 modes are to be used via kernel parameter,
via a bitmask parameter. E.g.:

- to enable rc6, i915_enable_rc6=1
- to enable rc6 and deep rc6, i915_enable_rc6=3
- to enable rc6 and deepest rc6, use i915_enable_rc6=5
- to enable rc6, deep and deepest rc6, use i915_enable_rc6=7

Please keep in mind that the deepest RC6 state really should NOT be used
by default, as it could potentially worsen the issues with deep RC6. So do
enable it only when you know what you are doing. However, having it around
could help solving possible future rc6-related issues and their debugging
on user machines.

Note that this changes behavior - previously, value of 1 would enable both
RC6 and deep RC6. Now it should only enable RC6 and deep/deepest RC6
stages must be enabled manually.

v2: address Chris Wilson comments and clean up the code.

References: https://bugs.freedesktop.org/show_bug.cgi?id=42579Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NBen Widawsky <benjamin.widawsky@intel.com>
Signed-off-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Looking at hibernate overwriting I though it looked like a cursor,
so I tracked down this missing piece to stop the cursor blink
timer. I've no idea if this is sufficient to fix the hibernate
problems people are seeing, but please test it.

Both radeon and nouveau have done this for a long time.

I've run this personally all night hib/resume cycles with no fails.
Reviewed-by: NKeith Packard <keithp@keithp.com>
Reported-by: NPetr Tesarik <kernel@tesarici.cz>
Reported-by: NStanislaw Gruszka <sgruszka@redhat.com>
Reported-by: Lots of misc segfaults after hibernate across the world.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=37142Tested-by: NDave Airlie <airlied@redhat.com>
Tested-by: NBojan Smojver <bojan@rexursive.com>
Tested-by: NAndreas Hartmann <andihartmann@01019freenet.de>
Cc: stable@vger.kernel.org
Signed-off-by: NDave Airlie <airlied@redhat.com>

ValleyView handles force wake differently than previous chipsets, so add
a couple of new functions for it.  But leave it disabled by default
until we test it (need a chip with the Punit enabled first).
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

For use by the rest of the ValleyView code.

v2: fix desktop variant to not set is_mobile (Ben)
Acked-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

It's only used by the main read/write functions, so we can keep it with
them.
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Acked-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Add a new module optoin lvds_channel to specify the LVDS channel mode
explicitly instead of probing the LVDS register value set by BIOS.
This will be helpful when VBT is broken or incompatible with the
current code.

Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=42842Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Reviewed-by: NAdam Jackson <ajax@redhat.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This change enables the use of displays where the vbt table just
contains inappropriate values, but either the vesa defaults or
the video=... modes do something sensible with the attached display.

The problem happens with an embedded board that contains vbt bios
tables that do not match the attached display.  Using this change and
the appropriate kernel boot command line they are able to use an
otherwise completely unusable secondary display on that embedded
board.
Reviewed-by: NPaul Menzel <paulepanter@users.sourceforge.net>
Signed-off-by: NMathias Froehlich <Mathias.Froehlich@web.de>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

If we don't have a sufficient number of free entries in the FIFO, we
proceed to do a write anyway. With this check we should have a clue if
that write actually failed or not.

After some discussion with Daniel Vetter regarding his original
complaint, we agreed upon this.
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This is similar to a patch I wrote several months ago. It's been updated
for the new FORCEWAKE_MT. As recommended by Chris Wilson, use WARN()
instead of DRM_ERROR, so we can get a backtrace.

This shouldn't impact performance too much as the extra register read
can replace the POSTING_READ we had previously.
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: NBen Widawsky <ben@bwidawsk.net>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We want to unconditionally enable ppgtt for two reasons:
- Windows uses this on snb and later.
- We need the basic hw support to work before we can think about real
  per-process address spaces and other cool features we want.

But Chris Wilson was complaining all over irc and intel-gfx that this
will blow up if we don't have a module option to disable it. Hence add
one, to prevent this.

ppgtt support seems to slightly change the timings and make crashy
things slightly more or less crashy. Now in my testing and the testing
this got on troublesome snb machines, it seems to have improved things
only. But on ivb it makes quite a few crashes happen much more often,
see

https://bugs.freedesktop.org/show_bug.cgi?id=41353

Luckily Eugeni Dodonov seems to have a set of workarounds that fix
this issue.

v2: Don't try to enable ppgtt on pre-snb.

v3: Pimp commit message and make Chris Wilson less grumpy by adding a
module option.

v4: New try at making Chris Wilson happy.
Reviewed-by: NBen Widawsky <ben@bwidawsk.net>
Acked-by: NChris Wilson <chris@chris-wilson.co.uk>
Tested-by: NChris Wilson <chris@chris-wilson.co.uk>
Tested-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

We have to do this manually. Somebody had a Great Idea.

I've measured speed-ups just a few percent above the noise level
(below 5% for the best case), but no slowdows. Chris Wilson measured
quite a bit more (10-20% above the usual snb variance) on a more
recent and better tuned version of sna, but also recorded a few
slow-downs on benchmarks know for uglier amounts of snb-induced
variance.

v2: Incorporate Ben Widawsky's preliminary review comments and
elaborate a bit about the performance impact in the changelog.

v3: Add a comment as to why we don't need to check the 3rd memory
channel.

v4: Fixup whitespace.
Acked-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NBen Widawsky <ben@bwidawsk.net>
Reviewed-by: NEric Anholt <eric@anholt.net>
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Along with the previous patch to make the reset operation protected by
the gt_lock as well, this ensures that all register read operations
will occur with the forcewake hardware enabled. As an added bonus,
this makes read operations more efficient by taking the spinlock only
once per read instead of twice.
Signed-off-by: NKeith Packard <keithp@keithp.com>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This ensures that no register reads occur while the forcewake state of
the hardware is indeterminate during the reset operation.
Signed-off-by: NKeith Packard <keithp@keithp.com>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

No reason to have half of the reset split from the other half.
Signed-off-by: NKeith Packard <keithp@keithp.com>
Reviewed-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

The problem this patch solves is that the forcewake accounting
necessary for register reads is protected by dev->struct_mutex. But the
hangcheck and error_capture code need to access registers without
grabbing this mutex because we hold it while waiting for the gpu.
So a new lock is required. Because currently the error_state capture
is called from the error irq handler and the hangcheck code runs from
a timer, it needs to be an irqsafe spinlock (note that the registers
used by the irq handler (neglecting the error handling part) only uses
registers that don't need the forcewake dance).

We could tune this down to a normal spinlock when we rework the
error_state capture and hangcheck code to run from a workqueue.  But
we don't have any read in a fastpath that needs forcewake, so I've
decided to not care much about overhead.

This prevents tests/gem_hangcheck_forcewake from i-g-t from killing my
snb on recent kernels - something must have slightly changed the
timings. On previous kernels it only trigger a WARN about the broken
locking.

v2: Drop the previous patch for the register writes.

v3: Improve the commit message per Chris Wilson's suggestions.
Signed-Off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: NKeith Packard <keithp@keithp.com>

LLC is not SNB/IVB-specific, so we should check for it in a more generic
way.
Reviewed-by: NChris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: NEric Anholt <eric@anholt.net>
Reviewed-by: NKenneth Graunke <kenneth@whitecape.org>
Signed-off-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

Fix function name in comments, a left-over from when i965_reset was
renamed to i915_reset.
Signed-off-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

After checking the specs and discussing with Jesse, turns out CxSR is not
available on Ironlake and gen5, and its advertisement on the device
description is misleading.
Acked-by: NJesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: NDaniel Vetter <daniel.vetter@ffwll.ch>

This was forgotten in the original multi-threaded forcewake
conversion:

commit 8d715f00
Author: Keith Packard <keithp at keithp.com>
Date:   Fri Nov 18 20:39:01 2011 -0800

    drm/i915: add multi-threaded forcewake support

Signed-Off-by: Daniel Vetter <daniel.vetter at ffwll.ch>
Reviewed-by: NEugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: NKeith Packard <keithp@keithp.com>

We learned that the ECOBUS register was inside the GT power well, and
so *did* need force wake to be read, so it gets removed from the list
of 'doesn't need force wake' registers.

That means the code reading ECOBUS after forcing the mt_force_wake
function to be called needs to use I915_READ_NOTRACE; it doesn't need
to do more force wake fun as it's already done it manually.

This also adds a comment explaining why the MT forcewake testing code
only needs to call mt_forcewake_get/put and not disable RC6 manually
-- the ECOBUS read will return 0 if the device is in RC6 and isn't
using MT forcewake, causing the test to work correctly.
Signed-off-by: NKeith Packard <keithp@keithp.com>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>