The hw is buggy and it's not currently used, but it's
currently still initialized by the driver.  Skip the init.
Skipping init also seems to improve stability with dpm on
some r6xx asics.

bug:
https://bugs.freedesktop.org/show_bug.cgi?id=66963Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Reviewed-by: NChristian König <christian.koenig@amd.com>

Prevent runtime suspend of non-PX GPUs.  Runtime suspend is
not what we want in those cases.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

Signed-off-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Signed-off-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Otherwise we allocate a new VMID on nearly every submit.
Signed-off-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

It seems this got dropped when we merged UVD support
last year.  Add this back now.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

Copy/paste typos from the ni code. Should not
have any functional change.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Needed to properly flush the read caches for fences.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

We need to set the engine bit to select the ME and
also set the full cache bit.  Should help stability
on TN and cayman.

V2: fix up surface sync in ib execute as well
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

Mclk switching doesn't seem to work reliably on these
cards.  Most RV770 boards specify the same mclk for all
performance levels anyway so in most cases, this has
no affect.

Bug:
https://bugs.freedesktop.org/show_bug.cgi?id=73067Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

Need to swap the data for big endian.
Notcied by sylware in IRC.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

For consistency.  These buffers aren't used when pg is
disabled.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

If the ss percentage is 0 or we are using external ss,
just bail when enabling ss.  We disable it explicitly
earlier in the modeset already.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

It's either 100 or 1000 depending on the flags in the
table.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This is needed for reporting the max GPU engine clock
in OpenCL.  This just reports the max possible engine
clock, it does not take into account current conditions
that may limit that clock.

v2: fix query number for merge with 3.13
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

According to the DP 1.1 spec, the sink must power
up within 1ms.  Noticed while reviewing Thierry's
drm/dp patches.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Based on common dp code proposed by Thierry Reding.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

As per the DP1.2 spec.  Noticed while reviewing
Thierry's drm/dp patches. Also bump native aux
retries to 7 for consistency.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Seems to cause problems with certain DP monitors.

Bug:
https://bugs.freedesktop.org/show_bug.cgi?id=40699Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

Use the hw power up values rather than 0.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Still unstable on some boards.

Bugs:
https://bugs.freedesktop.org/show_bug.cgi?id=73053
https://bugzilla.kernel.org/show_bug.cgi?id=68571Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: 3.13 <stable@vger.kernel.org> # 3.13

This is the preferred flushing method on CIK.

Note, this only works on the PFP so the engine bit must be
set.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This is the preferred flushing method on CIK.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

It's used in several places so move to a common shared
function.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

It's used in several places so move to a common shared
function.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Signed-off-by: NChristian König <christian.koenig@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This reverts commit 99b4f251.

Semaphores work fine after further review and testing.
Cc: 3.13 <stable@vger.kernel.org> # 3.13

i915 doesn't need this kludge for most platforms. Although we do
appear to need something similar on certain platforms, but we can
be more accurate when we apply the adjustment since we know exactly
why the scanline counter doesn't always quite match the vblank
status.

Also the current code doesn't handle interlaced modes correctly,
and we already deal with interlaced modes in i915 code.

So let's just move the current code to radeon_get_crtc_scanoutpos()
since that's why it was added. For i915 we'll add a more finely
targeted variant.

v2: Fix vpos vs. *vpos bug (Mario)

Reviewed-by: mario.kleiner.de@gmail.com
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>

Preparation for moving the early vblank IRQ logic into
radeon_get_crtc_scanoutpos().

v2: Fix radeon_drv.c compile warning (Mario)

Reviewed-by: mario.kleiner.de@gmail.com
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>

crtc_clock is now supposed to be the actual pixel clock corresponding to
the other crtc_ timing values. Populate crtc_clock appropriately in
radeon_atom_get_tv_timings().

This was the only obvious place where we frob with the crtc_ timigns
directly instead of calling drm_mode_set_crtcinfo() which would also
update crtc_clock.

Reviewed-by: mario.kleiner.de@gmail.com
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>

Rather than using crtc->hwmode, just pass the relevant mode to
drm_calc_vbltimestamp_from_scanoutpos(). This removes the last hwmode
usage from core drm.

Reviewed-by: mario.kleiner.de@gmail.com
Signed-off-by: NVille Syrjälä <ville.syrjala@linux.intel.com>

pci config reset is a low level reset that resets
the entire chip from the bus interface.  It can
be more reliable if soft reset fails.

v2: fix rebase
v3: hide behind module parameter
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

pci config reset is a low level reset that resets
the entire chip from the bus interface.  It can
be more reliable if soft reset fails.

v2: hide behind module parameter
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

pci config reset is a low level reset that resets
the entire chip from the bus interface.  It can
be more reliable if soft reset fails.

v2: put behind module parameter
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

pci config reset is a low level reset that resets
the entire chip from the bus interface.  It can
be more reliable if soft reset fails.

There's not much information still available on
r6xx, so r6xx is based on guess-work.

v2: put behind module parameter
v3: add IGP check
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This is used to hard reset the asic.  If a soft
reset is not able to reset things, a hard reset
can be used.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

Enabling this parameter enables pci config reset,
aka hard reset, which is a bus level chip reset.
In some cases this works more reliably than a soft
reset.  Disabled by default.
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

This fixes a bug which was causing rejections of valid GPU commands
from userspace.
Signed-off-by: NMarek Olšák <marek.olsak@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>

It's never allocated on systems without an ATOMBIOS or COMBIOS ROM.

Should fix an oops I encountered while resetting the GPU after a lockup
on my PowerBook with an RV350.
Signed-off-by: NMichel Dänzer <michel.daenzer@amd.com>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

drivers: gpu: Move prototype declarations to header file radeon_mode.h from radeon_atombios.c and radeon_combios.c

Move prototype declarations of functions radeon_get_encoder_enum() and
radeon_link_encoder_connector() to header file drm/radeon/radeon_mode.h
because they are used by more than one file.

This eliminates the following warnings in drm/radeon/radeon_encoders.c:
drivers/gpu/drm/radeon/radeon_encoders.c:86:1: warning: no previous prototype for ‘radeon_get_encoder_enum’ [-Wmissing-prototypes]
drivers/gpu/drm/radeon/radeon_encoders.c:162:1: warning: no previous prototype for ‘radeon_link_encoder_connector’ [-Wmissing-prototypes]
Signed-off-by: NRashika Kheria <rashika.kheria@gmail.com>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>
Signed-off-by: NAlex Deucher <alexander.deucher@amd.com>