gfx: drv: Move flip trace commands into psb_fb

The overlay code also manipulates sync object counters these days.
Instead of duplicating the same tracing code in two places, move
the code to the place where the sync counters are actually
manipulated.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: overlay: Synchronize overlay updates with CRTC page flips

Tie the overlay updates to the CRTC page flips. Whenever the overlay is
updated, we simply "arm" the overlay, and when the next page flip happens
on the associated CRTC, we push the page flip and the pending overlay
updates to the hardware simultaneosly.

This will guarantee that the page flip and overlay updates happen on the
same vblank, and are thus perfectly synchronized with each other.

Issue: ANDROID-1574
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

HACK: gfx: display: Double the tc35876x pixel clock

The pixel clock and vertical refresh values the driver reports for the
internal display are wrong. The page flip logic needs accurate clock
information to convert time intervals to scanline intervals.

Double the pixel clock in the tc35876x driver so that the values are
at least a little closer to reality.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Avoid page flipping while too close to vblank start

The hardware doesn't have proper facilities to indicate when a specific
page flip has occured. We can deduce that information from the frame
counter and scanline counter, but only if we know on which side of the
vblank start the flip was programmed. So if we're too close to the
vblank start to be able make that determination, wait until the vblank
start has passed before programming the flip into the hardware.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Rewrite the CRTC page flipping code to use the new drm_flip mechanism

The drm_flip mechanism allows us to track page flip progess, and thus
maintain proper reference counts on the backing memory and GTT mappings
while the buffers are being scanned out.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Introduce drm_flip helper class

The drm_flip mechanism can be used to implement robust page flipping
support, and also to synchronize the flips on multiple hardware scanout
engines (eg. CRTCs and overlays).

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Add some helper function to manipulate fb refcounts

Upcoming page flip logic will need to maintain extra references
to the pvr mem_info, and the GTT mapping. Add some functions to
make that easier.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Move read ops sync counter functions into psb_fb.c

The overlay code will want to manipulate sync counters as well, so move
the code to a more appropriate place.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Check fb bpp before doing irreversible changes to state

When the fb uses an unsupported bpp layout, the current code messes with
the hardware state before checking whether it can actually use the
provided framebuffer. This could leave the hardware in some weird state.
So check the framebuffer bpp before touching the hardware state.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: gtt: Support unmapping from arbitrary context

Pass the 'tgid' down from the caller of psb_gtt_unmap_meminfo(). This
will allow unmapping from eg. workqueues. This will be used to keep
references to the GTT mapping possibly after the drm_framebuffer has
been destroyed.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: ossync: Indicate whether sync callback is called from MISR

The callee may want to do some extra locking when the sync callback is
called from MISR (as opposed to being called directly). Add a boolean
parameter to the sync callback which indicates this.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Add PIPE DSL regs

Add defines for display scanline counter registers.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: Lock gPVRSRVLock in PVRSRVMISR()

It's totally unclear how various data are protected in the driver.
Currently the MISR work is exectuted without holding the gPVRSRVLock
mutex, which is held across bridge ioctls. To play things safe,
take the same lock in MISR work.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Lock gPVRSRVLock mutex before touching the meminfo/bm ref counts

The pvr mem ref counts are not modified using atomic operations, so we
need to protect them against simultaneous RMW access. Additionally it's
not at all clear how the related data structures are protected in the
pvr code. Use gPVRSRVLock as that's always locked around bridge ioctls.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: Remove most of the display class code

The display class swap chain code calls into the psb_fb driver
which causes KMS mode_config mutex to be locked after gPVRSRVLock.
To avoid problems with non-atomic reference counts, a later patch
will take the gPVRSRVLock inside code sections where mode_config
mutex is already locked. That could lead to dead locks, so the simple
solution is to eliminate the DC swap chain code altogether. We don't
want to use it anyway.

Unfortunately the DC driver can't be removed completely, since the
current userspace code requires a DC device to be registered. So
leave a minimal DC driver in place.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: update meminfo reference count when creating/destroying fbs

The fbs may outlive the mem info if the client frees the memory before
removing the fb so make sure the memory backing the fb lives as long
as the fb itself.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: add an interface for inc/dec meminfo ref count

The drm part of the pvr driver stores pointers to mem infos but those
may be freed by the clients. If the reference counts aren't increased
properly, those mem infos may be freed while the kernel still
references them.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: support creating fb's from mapped memory

If a client passes a handle to a mapped memory to drmModeAddFB, the
ioctl will fail since the gtt mapping code will fail to get its page
list. Fix that by using a reference to the original meminfo instead
of the mapped one.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: add a way of getting the src meminfo backing a mapped one

This will be used to implement on the drm side support for creating
a drm fb from a mapped memory handle.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: unifdef RES_MAN_EXTEND

The PVRSRVGetResData function will be used in a follow up patches to
implement support for creating drm fb's from mapped exported memory.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: don't store meminfo handle on psbfb

The handle has no reference count and might be freed by the client
before the psbfb is destroyed.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: use psbfb->pvrBO instead of hKernelMemInfo on page flip code

The meminfo is already stored on psbfb. Remove the extra indirection
through psb_get_meminfo_by_handle.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: pass meminfo instead of handle to psb_gtt_{unmap, map}_meminfo

The handles are process specific and may be freed while the kernel is
still using the meminfo.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: gtt: refcount gtt mappings

Currently psb_gtt_map_* can handle multiple calls with the same
handle but will remove the pages from gtt as soon as one call to
psb_gtt_unmap_* is made. Since multiple drm_fb's can be added with
the same handle and the driver will cal psb_gtt_map_meminfo for each
one of them, this could lead to unmapping memory that is still being
used.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: gtt: refactor gtt mapping code

Make room for inserting reference counting in the psb_gtt_mem_mapping
objects by refactoring the map and unmap code so that lower level
objects are allocated in lower level functions.

There is a hash table per process that holds a psb_gtt_mem_mapping
object which in turn holds a drm_mm_node object. The current
implementation allocs a drm_mm_node object in the psb_gtt_map_*
functions and then uses helper functions to wrap this node inside
a mapping object. When freeing this object, the drm_mm_node has to
be passed to the psb_gtt_unmap_* functions so that it can be freed.

This patch changes this logic so that the psb_gtt_map_* functions
allocate the mapping object through a helper function
(psb_gtt_add_mapping) that in turns allocates the node object. When
freeing, the psb_gtt_unmap_* functions will call psb_gtt_remove_mapping
which in will in turn free the node object.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: fix resource leak in psb_gtt_map_meminfo

If psb_gtt_map_meminfo was called with an already mapped handle a node
would be allocated with drm_mm_get_block_atomic in psb_gtt_mm_alloc_mem.
Later, on psb_gtt_mm_alloc_insert_mem_mapping (called by
psb_gtt_add_node), the reference to the allocated node would be
replaced by the node previously allocated for the handle, so the former
was never free'd.

This commit changes psb_gtt_map_meminfo to check for an existing node
before allocating a new one.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv/psb_gtt.c: add spaces around comments

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv/psb_gtt.c: replace printk with DRM_DEBUG

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv/psb_gtt.c: improve readability by using ERR_PTR

A lot of functions in this file return a value through a parameter
and use the return value for returning an error value. Using ERR_PTR
makes the code a bit more readable.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv/psb_gtt.c: make function calls take as few lines as possible

Comestic fix. No code changes.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: change psb_gtt_insert_* to remove casting

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv/psb_gtt.c: remove unnecessary casting of psb_get_tgid() to u32

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: fix psb_gtt_{map, unmap}_pvr_memory interface

Change data types to avoid unnecessary casting.

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv/psb_gtt.h: fix indentation

Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Do not kfree() pvr buffer's page list

psb_get_pages_by_mem_handle() doesn't allocate a new page list. Instead
it simply returns a reference to the pvr buffer's page list. Do not
kfree() said list in the caller.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

drv: psb: check crtc pointer before calling in DPMS and mode set.

Avoid the possibility of oops-ing the kernel by running the
test-opengl-fillrate test (and others) which bypass DRM_MASTER
authentication.

Typically the test will execute succesfully, however when
unblanking/un-suspending we NULL deref in gfx_late_resume()

Signed-off-by: Oliver McFadden <oliver.mcfadden@intel.com>
Reviewed-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: remove unused panel_desc field and defines

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Tested-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: cleanup ospm_power_suspend()

No functional changes, just a stylistic cleanup to make the function easier
to follow.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Tested-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: remove useless gb{Suspend,Resume}InProgress flags

The gbSuspendInProgress and gbResumeInProgress flags are only ever true
when the g_ospm_mutex is held. The locking should be in order without
these.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Tested-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: remove redundant code in ospm_power_using_hw_begin()

The display island should already be on at this point, and the power up
status for video enc/dec islands has already been checked some lines
earlier.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Tested-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: drop redundant ospm_resume_pci() in ospm_power_using_hw_begin()

The runtime resume through pm_runtime_get_sync() should already have taken
care of ospm_resume_pci(). No need to repeat here.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Tested-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: ospm_power_using_hw_begin() rewrite

ospm_power_using_hw_begin() had gotten overly complicated with a plethora
of variables and code paths. The code can be simplified considerably by
making the observation that the function is never called with force_on in
atomic context, i.e. if the power island is off, it will not be switched on
in atomic context.

Split the function into ospm_power_using_hw_begin() and
ospm_power_using_hw_begin_atomic(), of which only the latter may be called
in an atomic context. Clean up and simplify the code with the new
assumptions that can be made.

There is still plenty of strangeness left, but this is a big step forward
in making the code possible to understand.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Tested-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: remove no-op assignments in ospm_power_using_hw_begin()

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Tested-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: switch to the power island management code in the PMU driver

Both the GFX and ISP drivers need to switch on and off the relevant power
islands through PUnit. If a centralized driver with proper locking is not
used, they may stomp on each others' toes accessing the PUnit registers.

Obviously, this also reduces code duplication, and the power island
management code is of better quality in the PMU driver anyway.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Tested-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: use a bool parameter for force in ospm_power_using_hw_begin()

Remove an unnecessary level of abstraction from what is a simple thing.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Tested-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv/pvr: cmd trace: show both old and new flip req buffers

For flip request commands the trace output showed only information on
one buffer: the DC flip code showed the new buffer while the DRM
specific flip code showed the old buffer. This in itself is
inconsistent, besides we want to get information on both buffers. So
show both buffers.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: cmd trace: rename flip request syncobj names

The flip request command has an old and a new buffer (and the
corresponding sync objects) as a parameter. Rename the relevant
entries in the trace output to make this clearer.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: check the size of SGX_READ_HWPERF IOCTL parameters

We need to add a separate size check for this IOCTL's parameters, since
they are passed by a pointer in the IOCTL's input structure. We do check
the size for all IOCTL's input structure already, but we need to cover
such derefenced objects separately.

Increase the driver's minor version too. This will still let the current
user space libraries work, except for this IOCTL, which was broken
before anyway. A related user space library fix will be submitted to
make the IOCTL work.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Tested-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: fix SGX_READ_HWPERF IOCTL according to v1.7 ABI

There has been a change in the parameters for this IOCTL in v1.7 of the
driver. While the userspace library was updated accordingly we didn't do
this on the kernel side. We missed the difference since the size of the
structure wasn't checked against the user space version. The next patch
will add this check to avoid similar problems in the future.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Tested-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

pvr_debugfs: Fix inline function declaration

This fixes a compilation error if CONFIG_DEBUG_FS isn't enabled.

Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Add support for CABC

CABC (Content Adaptive Backlight Control) is a power saving technique
that is supported by the mfld tablet LCD panel.

Signed-off-by: Antti Koskipaa <antti.koskipaa@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Fix panel poweron sleep value

Panel spec says t1+t2+t5 = 260 ms. It's more complicated than that
but the driver is not flexible enough to handle all the cases. This
should be good enough, and is the absolute minimum.

Signed-off-by: Antti Koskipaa <antti.koskipaa@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Move stuff around

Move a function to where it belongs so that panel register definitions
need not to be declared globally.

Signed-off-by: Antti Koskipaa <antti.koskipaa@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Add register definitions for Chimei Innolux panel

Signed-off-by: Antti Koskipaa <antti.koskipaa@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

Fix intermittent hotplug lost issue because of page flip.

Since the DRM page flip mechanism keeps updatign the crtc->fb member
with the current fb to be flipped, this gets out-of-sync with the
psb_fb_helper->fb member. In the event of a HDMI hotplug, this mismatch
causes the HDMI hotplug event to be deferred and never handled. Ensure
that we are keeping both the members in sync always during flip.

Signed-off-by: Arun Kannan <arun.kannan@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

mode management changes between local and external display.

- Defining our policy of setting the FB and surface dimensions
  of frame buffer equal to local display dimensions and choosing
  panel fitting for external display when needed. This is required since
  Medfield hardware supports panel fitting only for HDMI pipe.
- Also, remove the modes selection limitations at drm during hdmi hotplug.
  During drm_fb_helper_hotplug_event, the max_width and max_height are
  reseted to the size of available framebuffer (which is equal to local
  display), and thus removes all the hdmi modes that are greater than this
  limit thus not allowing modes like 1080P to get selected for HDMI.
  Hence changed the reset logic.

Signed-off-by: Arun Kannan <arun.kannan@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

Enable suspend/resume support for HDMI.

Enable the suspend/resume hooks for OTM HDMI.

Signed-off-by: Arun Kannan <arun.kannan@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

MUST_REVERT: Avoid calling mode set for MIPI during HDMI hotplug.

Currently, during HDMI hotplug/unplug, the MIPI mode set is invoked along
with HDMI mode set. The MIPI mode set during HDMI hotplug causes the system to hang
due to an i2c transaction failure for MIPI. Till the local display issue is fixed, we
avoid calling the MIPI mode set on every HDMI hotplug event.

Signed-off-by: Arun Kannan <arun.kannan@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

Enable hotplug kernel handler.

Enable the OTM HDMI hotplug handler for handling HDMI
connection/disconnection events.

Signed-off-by: Arun Kannan <arun.kannan@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

Enable the OTM HDMI driver on ICS.

Enable the OTM HDMI driver to be called from display driver and remove
older HDMI related code.

Signed-off-by: Arun Kannan <arun.kannan@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

Addition of the OTM HDMI driver for Medfield.

This is a modular, reusable driver for HDMI used across platforms.
This driver has already achieved HDMI certification on the Greenridge platform,
and was also reused on the Medfield honeycomb stack.

Signed-off-by: Arun Kannan <arun.kannan@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

imgv: Fix video bind page management

Shared pages has to be reference count and free by kernel if userspace
fails to free them.

To have bind ioctl manage memory correctly use existing fiex pages
infrastructure to bind and unbind pages. Fixed pages already include ttm
backed functions to free pages correctly when application process
terminates.

Based on Robin Mathew's "Fix TTM Leaking on SIGKILL" patch.

Issue: Android-926
Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Acked-by: Imre Deak <imre.deak@intel.com>
Acked-by: Sean V Kelley <sean.v.kelley@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

pvr: increase source sync object limit

Issue: ANDROID-372
Signed-off-by: Haitao Huang <haitao.huang@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: do not spam tc35876x_brightness_control notifications.

These are very annoying and often get in the way when trying to read a
backtrace. If you really need them, you may turn on PSB debugging.

Signed-off-by: Oliver McFadden <oliver.mcfadden@intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

Make psb driver interface files not world writable

Androids CTS test doesn't allow there to be world writable /sys files.

Signed-off-by: ken lierman <ken.lierman@windriver.com>
Signed-off-by: Kwanhong Lee <kwanhong.lee@windriver.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

HACK: gfx: drv: when resuming make sure power is on

The PVR driver expects ospm_power_using_hw_begin to return only when
power is already on. But at the moment that function will only call
pm_runtime_get which is asynchronous. This can lead to the PVR driver
accessing its HW registers before the HW is powered and cause a CPU
fabric error hung.

To work around this call the sync version of pm_runtime_get if we are
not in atomic context. In fact I can't see any reason to call it
atomically, since then power on can't be guaranteed on return anyway.
But to clean up that we should check all call-sites for correctness.

Jani Nikula is currently reworking ospm_power_using_hw_begin() and
related parts that will give us a clean solution. Until that's done we'd
like this hack applied to let us resolve other PM related issues that
are currently hidden by this bug.

Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: optimize clearing sync counter trace info struct

Clearing this struct is necessary to suppress tracing for a given sync
counter. But for this we don't need to bzero the whole sync counter info
struct it's enough to set the address to zero.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Tested-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: fix corrupted command trace for SGX transfer command

When the transfer command is called with the SGXMKIF_TQFLAGS_KEEPPENDING
source and destination sync counters were not zeroed in the
corresponding trace record, leading to a corrupted trace output.

Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Tested-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Add trace events for powermanagement

SGX may lockup because of incorrectly timed or missing power state
trasition for SGX or display controller.

To make future debugging simpler for these cases add command trace
entries for realeted power management changes.

Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Add command trace points to flip

To make easier to debug SGX lockups with page flip code we need command
trace to have page flip events. There already exists API to add trace
events for flip chain. We can use same API for drm mode page flips.

Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: overlay: Set CC_OUT bit in OCONFIG

Unfortunately the docs aren't exactly clear on when the bit should be set.
However all user space solutions I've seen set the bit, so let's follow suit.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: overlay: More thorough fix for SWIDTHSW issues

The current code can still misprogram SWIDTHSW if fb->offsets[] is
suitably misaligned. Simply limit SWIDTHSW to 2k pixels in such cases.

Change the SWIDTHSW alignment code to be more readable, and align the
width to 32 bytes (SWORD), instead of 64 bytes.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

MUST_REVERT: drm: psb: Added gralloc buffer support for video

Gralloc buffers are created and bound to
TTM buffers independent of the BCD module.

This patch has received many change requests from Pauli and Jani.
Jani Nacked it. However, Sean believes the requests are not critical.
Pierre said that we have to take this patch because it is urgent,
and improve it later. This is why it is marked as MUST_REVERT. Sean is
supposed to send the new version which should be taken instead of this
one, and this commit message should be removed.

Signed-off-by: Naeem M Afzal <naeem.m.afzal@intel.com>
Signed-off-by: Sean V Kelley <sean.v.kelley@intel.com>

MUST_REVERT: gfx: drm: explicitly authenticate for Android

Added as an interim measure to proper client authentication using
DRM_AUTH IOCTLs in user space.  Applied to drm_fops.c:drm_open_helper
where there is already a similar hack.  Shall be removed upon
userspace implementaiton.

Signed-off-by: Sean V Kelley <sean.v.kelley@intel.com>

gfx: drv: overlay: Fix NV12 chroma SWIDTHSW

NV12 chroma SWIDTHSW must be calculated in a somewhat bizarre way.

Also fix an off-by-one error for other pixel formats.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: remove the outdated README

The README is outdated, and this information is better shared outside of
the kernel tree anyway.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: remove unused uopt (user options)

Remove unused code for uopt, user options.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: remove unused code in intel-linux directories

The directories pvr/services4/{include,srvkm}/env/linux-intel are not used
for anything. Remove.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

Revert "drm: Protect drm/drm_os_linux.h inclusion with #ifdef"

This reverts commit "drm: Protect drm/drm_os_linux.h inclusion with #ifdef"
to minimize gfx related diff outside drivers/staging/mrst. The header file
<drm/drm_os_linux.h> is not supposed to be included directly anyway.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: imgv: remove redundant #include <drm/drm_os_linux.h>

The header file <drm/drm_os_linux.h> is not supposed to be included
directly, it does not have #ifdef protection for multiple inclusion, and
it's included through <drm/drmP.h> anyway. Remove the #include.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: remove obsolete .gitignore

The .gitignore is no longer relevant with Medfield only build.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: make tc35876x independent of tmd vid driver

The TC35876X driver used to be bolted on the TMD panel driver with some
ifs. Make it a more independent driver instead.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: remove redundant dpi_panel_on setting

dpi_panel_on is false at init time, no need to set it again.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

pvr: unifdef pvr power management code for improved readability

The PVR power management code is difficult to read due to excess #ifdef 1
and #ifdef 0. Remove them.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: allow powering down GL3 cache regardless of GL3 config

The commit "psb_video: remove OSPM_GL3_CACHE_ISLAND when not enabled" made
it impossible to power down the GL3 cache island when
CONFIG_MDFD_GL3=n. It's powered on at boot (likely by firmware) so allow
powering down regardless of config.

This fixes the following situation in /sys/kernel/debug/mid_pmu_states when
display is off:

pci 8086 0130 0000:00:02.0             pvrsrvkm:
                      |------->             GFX                   D0i3
                      |------->   Video Decoder                   D0i3
                      |------->   Video Encoder                   D0i3
                      |------->       GL3 Cache                   D0
                      |------->       Display A                   D0i3
                      |------->       Display B                   D0i3
                      |------->       Display C                   D0i3
                      |------->        MIPI-DSI                   D0i3

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: remove another redundant panel state variable gbdispstatus

The dev_priv->dpi_panel_on{,2} fields already track panel state. Use those
instead of gbdispstatus.

The old implementation could not have worked any better for DBI panels than
this one, but leave a comment to revisit this.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: refactor psb_runtime_{idle,suspend}

Reuse psb_runtime_idle for checking whether runtime suspend is possible in
psb_runtime_suspend, making the conditions the same for both. Do HDMI audio
suspend in suspend, not idle.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: tc35876x: remove redundant switching of dpi_panel_on state

The dev_priv->dpi_panel_on{,2} state is flipped accordingly, and on the
correct pipe, by the caller of mdfld_dsi_configure_{up,down}, so don't do
it again.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: tc35876x: make mdfld_dsi_configure_{up,down} more pipe aware

This does not fix everything about pipes in mdfld_dsi_configure_{up,down},
but is a step in the right direction.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: display: tc35876x: remove extra dsi_device_ready state variable

The driver is full of display state variables. Remove dsi_device_ready used
by the tc35876x, and use the existing dev_priv->dpi_panel_on instead.

Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Clear links when freeing head of list

Linked list is freed in psb_cleanup_pending_events is called just before
head of linked list is freed. All list entries needs to be initialized
to be empty lists to avoid accessing freed memory if logic changes in
future.

Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: Fix SGX failing to complete queued rendering

SGXCommandComplete wakeups SGX to render when blocking external
operation completes. If blocking operation completes in IRQ handler
driver only sets flag to wakeup hardware but doesn't schedule the work
to handle wakeup operation.

Issue: ANDROID-529
Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Clean pending page flip events when device is closed

If userspace closes the device node with pending page flips driver keeps
a pointer to file private that is already freed. Following backtrace
shows the oops happening.

[ 643.828884] BUG: unable to handle kernel paging request at 6b6b6b6b
[ 643.835082] IP: [<c1224b6e>] __list_add+0x2a/0x5c
[ 643.888465] EIP is at __list_add+0x2a/0x5c
[ 643.943931] Call Trace:
[ 643.946376] [<c139c0d8>] psb_intel_flip_complete+0x71/0xcb
[ 643.951937] [<c139c1c8>] sync_callback+0x96/0x9f
[ 643.956631] [<c1371310>] PVRSRVCallbackOnSync+0x37/0xb7
[ 643.961934] [<c139c132>] ? psb_intel_flip_complete+0xcb/0xcb
[ 643.967673] [<c139c049>] psb_intel_crtc_page_flip+0xa1/0xbf
[ 643.973324] [<c1282f5d>] drm_mode_page_flip_ioctl+0x13f/0x190
[ 643.979151] [<c12778bf>] drm_ioctl+0x276/0x332

To cleanup pending events driver has to keep linked list of pending
flips in file private. When file is closed all events in pending flips
for the file are cleared.

To avoid race conditions dev->event_lock is used to protect pending
flips lists from parallel access.

Issue: KER-47
Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
CC: Rahul Verman <rahul.verma@intel.com>
Acked-by: Sean V Kelley <sean.v.kelley@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Fix page flip lockup when requesting vblank fails

drm_vblank_get may fail if specified pipe is off. In that case we won't
get vblank even and can't call drm_vblank_put. But driver still has to
complete page flip to send required completion event to userspace.

Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Remove useless global variable

Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

drm: ttm: Fix clearing of highmem pages

ttm might allocate highmem pages for GPU usage. But page_address works
only for normal pages. clear_highpage can be used to clear both type of
pages.

Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reported-and-tested-by: Daniel Charles <daniel.charles@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Move NULL check outside spin lock

pending_flip->event is owned by single kernel thread at a time so it can
be checked outside of the spin lock. Spin lock is only protecting the
linked list where the event is added.

Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: pvr: Dereference a pointer after null check

Dereferencing a pointer may only happen after NULL check to avoid crash
if the pointer is NULL.

[pvr/services4/srvkm/common/buffer_manager.c:1152]: (error) Possible
null pointer dereference: psBMHeap - otherwise it is redundant to check
if psBMHeap is null at line 1156

Reported-by: Olev Kartau <olev.kartau@intel.com>
Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: imgv: Dereference a pointer after null check

Dereferencing a pointer may only happen after NULL check to avoid crash
if the pointer is NULL.

[imgv/psb_fence.c:33]: (error) Possible null pointer dereference:
dev_priv - otherwise it is redundant to check if dev_priv is null at
line 36

Reported-by: Olev Kartau <olev.kartau@intel.com>
Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>

gfx: drv: Remove unused variable

Remove unused stack variable which would crash -O0 optimised kernel.

[drv/psb_page_flip.c:61]: (error) Uninitialized variable: e

Reported-by: Olev Kartau <olev.kartau@intel.com>
Signed-off-by: Pauli Nieminen <pauli.nieminen@linux.intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>