amdgpu: Add syncobj reset & signal wrappers.

Signed-off-by: Dave Airlie <airlied@redhat.com>

drm: Add drmSyncobjReset & drmSyncobjSignal wrappers.

anv already uses the ioctls but does not use libdrm, so these were
not wrapped yet.

Signed-off-by: Dave Airlie <airlied@redhat.com>

etnaviv: fix BO cache to properly work with different flags

Currently if the oldest BO in a bucket has different flags than what we
look for we'll miss the cache.Fix this by iterating over the cached BOs
until we find the oldest one with matching flags. This improves the hit
ratio for some of the buckets.

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>

tests/amdgpu: Add return CUE_SUCCESS to suite_vcn_tests_clean.

    fixes: 806d0803600000faecb4025d8e9c7490cb097c25 (amdgpu: Use new suite/test disabling functionality.)
    bug: https://bugs.freedesktop.org/show_bug.cgi?id=104280

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

etnaviv: support performance monitor requests

Add etna_cmd_stream_perf(..) to submit perform requests.
Userspace can submit pmrs via submit ioctl to sample perfmon
signals.

v3:
 - mark perfmon bos as RW

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>

etnaviv: add permon support

Query all domains and their signals and provide it this information
via struct etna_perfmon and the corresponding api functions.

v2:
 - code style changes
 - etna_perfmon_create(..): add missing clean up in error case

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>

etnaviv: sync uapi header

Import the etnaviv header changes from kernel commit 05916bed1 (drm-next)

The drm_etnaviv_gem_submit structure was extended to include performance
monitor requests. Also two new ioctls got added to be able to readout
performance monitor domains and their signals.

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Acked-by: Lucas Stach <l.stach@pengutronix.de>

amdgpu: Remove dummy CU_ASSERT_EQUAL.

Fixes test failure on rhel.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Qiang Yu <Qiang.Yu@amd.com>

amdgpu.ids: Refresh from AMD 17.40 release

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>

amdgpu: Only remember the device's marketing name

There's no point in keeping around the full table of marketing names,
when amdgpu_get_marketing_name only ever returns the device's marketing
name.

Acked-by: Slava Abramov <slava.abramov@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

amdgpu: Simplify error handling in parse_one_line

* Move empty/commented line check before the strdup and return -EAGAIN
  directly
* Initialize r = -EAGAIN and remove redundant assignments
* Set r = -ENOMEM if last strdup fails, and remove redundant goto

Acked-by: Slava Abramov <slava.abramov@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

amdgpu: Clean up amdgpu_parse_asic_ids error handling

* Move error message printing into amdgpu_parse_asic_ids and make it
  return void
* Print only "Invalid format" error message if parse_one_line returns
  -EINVAL
* Use strerror instead of printing the (negative) error code in hex

Acked-by: Slava Abramov <slava.abramov@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

amdgpu: Add explicit dependency test.

The test is as following:

1) Create context A & B
2) Send a command submission using context A which fires up a compute shader.
3) The shader wait a bit and then write a value to a memory location.
4) Send a command submission using context B which writes another value to the same memory location, but having an explicit dependency on the first command submission.
5) Wait with the CPU for both submissions to finish and inspect the written value.

Test passes if the value seen in the memory location after both submissions is from command B.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>

amdgpu: Add amdgpu_cs_create_syncobj2 to amdgpu-symbol-check

Fixes make check. Trivial.

amdgpu: Adding amdgpu_cs_create_syncobj2 to create syncobj as signaled initially

Signed-off-by: David Mao <david.mao@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

amdgpu: Dynamicly disable BO suite "Metadata" test.

Disable the test instead of doing  #ifdef 0

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

amdgpu: Switch amdgpu CS tests enabling to the new way.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

Android: disable warnings causing errors

AOSP master has changed the build default to -Werror making all the
warnings errors. Override that with -Wno-error.

Signed-off-by: Rob Herring <robh@kernel.org>

amdgpu: Disable deadlock test suite for Vega 10

The test stalls the CP, until RCA is done the test is
disabled to not disrupt regression testing.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>

amdgpu: Add memory over allocation test.

Allocates 1 TB of memory. Test is disabled by default
since it's triggers OOM killer.

v2:
FIx the test to only alloc the BO and assert if return value
not equal to -ENOMEM and remove test disable on start.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

amdgpu: Move memory alloc tests in bo suite.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

amdgpu: Use new suite/test disabling functionality.

Switch from disabling tests during run to using the new disable
API.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

amdgpu: Add functions to disable suites and tests.

Suits are diasbled based on hooks they provide (e.g incompatible
ASIC or missing blocks). Single tests are diasbled explicitly.
Suit or test can be forced to execute even if disabled by adding -f
flag after specifying suit [test] ids.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

headers: Drop outdated node about a delta in drm_mode.h.

Fixed in 9433b702fc16 ("headers: Sync drm{,_mode}.h with the kernel")

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>

headers: Sync up kernel changes to use kernel types instead of stdint.h.

This pulls in pieces of drm-next d65d31388a23 ("Merge tag
'drm-misc-next-fixes-2017-11-07' of
git://anongit.freedesktop.org/drm/drm-misc into drm-next")

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>

headers: Sync up mga_drm.h from drm-next.

The kernel started using the struct from drm.h instead of the typedef.

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>

headers: Sync up some comment spelling and whitespace fixes from drm-next.

This pulls in pieces of drm-next d65d31388a23
("Merge tag 'drm-misc-next-fixes-2017-11-07' of
git://anongit.freedesktop.org/drm/drm-misc into drm-next")

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>

headers: Sync up some header guard changes from drm-next.

This pulls in pieces of drm-next d65d31388a23
("Merge tag 'drm-misc-next-fixes-2017-11-07' of
git://anongit.freedesktop.org/drm/drm-misc into drm-next")

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>

headers: Sync vc4 header from drm-next.

This updates the header with the contents from drm-next d65d31388a23
("Merge tag 'drm-misc-next-fixes-2017-11-07' of
git://anongit.freedesktop.org/drm/drm-misc into drm-next")

Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Emil Velikov <emil.l.velikov@gmail.com>

Revert "amdgpu: fix 32bit VA manager max address"

This reverts commit 944f6665de36b6a6c36263f23b7b9d1730e544fa.

Accidentially pushed an imcomplete patch.

Signed-off-by: Christian König <christian.koenig@amd.com>

Revert "amdgpu: use the high VA range if possible"

This reverts commit 6c0ea4b0c5452bfc1e67b74ce723696ef3c80b25.

Accidentially pushed an incomplete patch.

Signed-off-by: Christian König <christian.koenig@amd.com>

amdgpu: use the high VA range if possible

This frees up the low range for HMM.

Signed-off-by: Christian König <christian.koenig@amd.com>

amdgpu: fix 32bit VA manager max address

The range is exclusive not inclusive.

Signed-off-by: Christian König <christian.koenig@amd.com>

amdgpu: sanitize 64bit VA manager

Adding the extra reservation of the 32bit space to the 64bit manager is
complete nonsense and just a waste of memory and CPU cycles.

Signed-off-by: Christian König <christian.koenig@amd.com>

amdpgu: fix coding style in amdgpu_vamgr.c

No functional change.

Signed-off-by: Christian König <christian.koenig@amd.com>

freedreno: add the API fd_pipe_new2 to the symbol test

As kindly spotted by `make check'

Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com>

exynos: change the license to X11/MIT

Change GPL license of Exynos related code to X11/MIT.

I'd like to keep a consistent license across all Exynos code
because License checker notices two more licenses exist
in libdrm.

For the license change I need to get your agree - all committers.
So please give me Acked-by if you agree with me.

Signed-off-by: Inki Dae <inki.dae@samsung.com>
Acked-by: Hyungwon Hwang <human.hwang@samsung.com>
Acked-by: SooChan Lim <sc1.lim@samsung.com>
Acked-by: Sangjin LEE <lsj119@samsung.com>
Acked-by: Boram Park <boram1288.park@samsung.com>
Acked-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Acked-by: Joonyoung Shim <jy0922.shim@samsung.com>
Acked-by: Emil Velikov <emil.l.velikov@gmail.com>
Acked-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de>
Acked-by: Jan Vesely <jan.vesely@rutgers.edu>

freedreno: submit-queue context priority

With a new-enough kernel to support prioritized submit-queues, we can
expose priority level support to mesa.  Open a submit queue associated
with the fd_pipe and pass it's id back to SUBMIT ioctl.

Signed-off-by: Rob Clark <robclark@freedesktop.org>

freedreno: sync uapi header (driver version 1.3.0)

Signed-off-by: Rob Clark <robclark@freedesktop.org>

configure.ac: bump version for release

amdgpu: Fix wrappers for AMDGPU_VM IOCTL.

Rmove amdgpu_context_handle from the interface and use
amdgpu_device_handle instead. Uupdate VMID reservation test
accordingly.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

configure.ac: bump version for release

amdgpu: fix 32-bit build

configure.ac: bump version for release

amdgpu: Add VMID reservation per GPU context test.

The test will Reserve a VMID, submit a command and
unreserve the VMID.

v2:
Wrappers names were changed.

Signed-off-by: Andrey Grodzovsky <Andrey.Grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

amdgpu: Add wrappers for AMDGPU_VM IOCTL.

v2:
Rename wrappers to match the IOCTL naming, fix
identation and fix make check error.

Signed-off-by: Andrey Grodzovsky <Andrey.Grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

drm/syncobj: fix some whitespace issues

These had tabs, just remove them.

Signed-off-by: Dave Airlie <airlied@redhat.com>

drm: sync drm headers from drm-next.

This updates the headers with the contents in drm-next
at 62884cd386b876638720ef88374b31a84ca7ee5f

This adds leasing and aspect ratio.

configure.ac: bump version to 2.4.85

amdgpu: add padding to the fence to handle ioctl

copied from Dave's kernel patch.

Add const qualifier to arguments of drmModeAddFB2()

Both drmModeAddFB2() and drmModeAddFB2WithModifiers() have some
arguments that are just pointers to uint32_t in disguise. These
are not modified (just copied) in the function, so we can add a
const qualifier here.

Signed-off-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>

amdgpu: implement context priority for amdgpu_cs_ctx_create2 v3

Add a new context creation function that allows specifying the context
priority.

A high priority context has the potential of starving lower priority
contexts. The current kernel driver implementation allows only apps
that hold CAP_SYS_NICE or DRM_MASTER to acquire a priority above
AMDGPU_CTX_PRIORITY_NORMAL.

v2: corresponding changes for kernel patch v2
v3: Fixed 'make check' symbol error

Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

headers: Sync amdgpu_drm.h with drm-next

Generated using make headers_install from:
airlied/drm-next 282dc83 Merge tag 'drm-intel-next-2017-10-12' ...

Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Acked-by: Dave Airlie <airlied@redhat.com>
Acked-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>

modetest: Allow full testing of primary planes

Allow the user to override the default configuration set by setcrtc
for the primary plane. On some hardware primary planes can be freely
positioned/sized, and it'd be nice if we can actually test that feature.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

modetest: Decode IN_FORMATS plane blob property

This teaches modetest about the new IN_FORMATS blob and decodes the
blob to show supported formats and modifiers.

Signed-off-by: Kristian H. Kristensen <hoegsberg@chromium.org>

configure.ac: bump version to 2.4.84

amdgpu: add amdgpu_cs_fence_to_handle

v2: update amdgpu-symbol-check

Reviewed-by: Chunming Zhou <david1.zhou@amd.com>

amdgpu: add amdgpu_cs_syncobj_wait

v2: update amdgpu-symbol-check

Reviewed-by: Chunming Zhou <david1.zhou@amd.com>

drm: add drmSyncobjWait wrapper

Reviewed-by: Chunming Zhou <david1.zhou@amd.com>

amdgpu: add sync_file import and export functions

v2: update amdgpu-symbol-check

Reviewed-by: Chunming Zhou <david1.zhou@amd.com>

include: sync drm.h and amdgpu_drm.h with airlied/drm-next

headers: sync syncobj ioctl defines.

This syncs the drm.h header with my drm-next branch as of
bb7a9c8d712f37385a706a594d6edf6e6d2669d0

Signed-off-by: Dave Airlie <airlied@redhat.com>

headers: sync DRM_MODE_ROTATE/REFLECT defines from kernel v4.14-rc1

Add the new DRM_MODE_ROTATE_* and DRM_MODE_REFLECT_* defines from v4.14-rc1
kernel.

Reviewed-by: Thierry Reding <treding@nvidia.com>
Reviewed-by: Robert Foss <robert.foss@collabora.com>
Signed-off-by: Rob Herring <robh@kernel.org>

Android: move libraries to /vendor

As part of Treble project in Android O, all the device specific files have
to be located in a separate vendor partition. This is done by setting
LOCAL_PROPRIETARY_MODULE (the name is misleading). This change will not
break existing platforms without a vendor partition as it will just move
files to /system/vendor.

Signed-off-by: Rob Herring <robh@kernel.org>

tests/amdgpu: fix uvd enc data corruption issue

In uvd encode parameter package, parameters input_pic_luma_pitch and
input_pic_chroma_pitch should be picture width align with hardware alignment.
The hardware alignment is 16 for amdgpu family earlier than AMDGPU_FAMILY_AI,
and 256 for later than and including AMDGPU_FAMILY_AI.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>

tests/amdgpu: add new uvd enc support check

Query hardware IP information to find out if there are uvd encode rings
ready for use in kernel driver.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>

amdgpu: Add deadlock detection test suit.

Adding initial tests for locks detection when SW
scheduler FIFO is full.

The test works by submitting a batch of identical commands which make the CP
stall waiting for condition to become true. The condition is later satisfied
form a helper thread. Other events that happen during this time
might create deadlock situations. One such example is GPU reset
triggered by this stall when  amdgpu_lockup_timeout != 0.

v2:
Increase the delay from 2 to 100 ms.
Comment out the compute test until it's working.
Typos fix.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>

amdgpu: make userptr unit test more interesting

Do a fork() to force some MMU callbacks.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>

amdgpu: print error messages when amdgpu_device_initialize is failing

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>

intel: Change a KBL pci id to GT2 from GT1.5

See Mesa commit 9c588ff

Cc: Matt Turner <mattst88@gmail.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Anuj Phogat <anuj.phogat@gmail.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

freedreno/kgsl: fix pointer-to-int cast

Fixes this warning:

freedreno/kgsl/kgsl_ringbuffer.c: In function ‘kgsl_ringbuffer_flush’:
freedreno/kgsl/kgsl_ringbuffer.c:149:19: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
   req.timestamp = (uint32_t)kgsl_ring->bo->hostptr;
                   ^

Signed-off-by: Eric Engestrom <eric.engestrom@imgtec.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>

etnaviv: prevent deadlock in error path

If drmPrimeFDToHandle fails in etna_bo_from_dmabuf, the function must
not return with the table_lock mutex held. Unlock the mutex in the error
path.

Based on ceb70a6b1015 ("freedreno: prevent deadlock in error path").

Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>

amdgpu: Do not write beyond allocated memory when parsing ids

Fixes crash when/usr/share/libdrm/amdgpu.ids contains ASIC_ID_TABLE_NUM_ENTRIES + 1 entries.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102432
Fixes: 7e6bf88cac315a9fa41818cf72a7b5d18a2cb1fc (amdgpu: move asic id table to a separate file)
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>

tests/amdgpu: add missing header to SOURCES

Fixes the tarball generation.

fixes: 9d133dd08720d80dfc8ce098bf0972 (tests/amdgpu: add uvd encode unit tests)
bug: https://bugs.freedesktop.org/show_bug.cgi?id=102391
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

configure.ac: Bump version to 2.4.83

etnaviv: fix etna_bo_from_name

Look up BOs from the name table using the name parameter instead of
req.handle (which at this point is always zero).

Signed-off-by: Philipp Zabel <p.zabel@pengutronix.de>
Reviewed-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Wladimir J. van der Laan <laanwj@gmail.com>
Reviewed-by: Daniel Stone <daniels@collabora.com>

android: amdgpu: fix build break

Define two macros to avoid building errors.

Fixes: 7e6bf88cac (amdgpu: move asic id table to a separate file)

Signed-off-by: Chih-Wei Huang <cwhuang@linux.org.tw>

android: add rules to build amdgpu.ids

Signed-off-by: Chih-Wei Huang <cwhuang@linux.org.tw>

drmsltest: Check expected neighbours

Fixes: 7d8c9464081634f053e16e5eac9655a12fae1dc4
Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>

tests/amdgpu: add uvd encode unit tests

Signed-off-by: Boyuan Zhang <boyuan.zhang@amd.com>
Acked-by: Alex Deucher <alexander.deucher at amd.com>
Acked-by: Christian König <christian.koenig@amd.com>

drm: Pull new modifier uapi into drm_fourcc and drm_mode

Reviewed-by: Daniel Stone <daniels@collabora.com>

amdgpu: merge and cleanup amdgpu_bo_free

since bo_reference and bo_internal_free are
all only used by bo_free, so we just merge them
together

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

amdgpu: fix race issue between two bo functions(v2)

there is race issue between two threads on amdgpu_bo_reference and
amdgpu_bo_import, this patch tends to fix it by moving the
pthread_mutex_lock out of bo_free_internal and move to bo_reference
to cover the update_reference part.

The mutex_unlock in bo_import should also cover bo refcount
increasement.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

amdgpu: fix missing mutex unlock before return

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

freedreno: prevent deadlock in error path

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Rob Clark <robdclark@gmail.com>

freedreno/msm: remove dead error path

`ring` cannot be non-null, so the label reduces to a simple return.
Then, there is no point initialising `ring` just to overwrite it before
anyone reads it.

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Rob Clark <robdclark@gmail.com>

freedreno: remove dead error path

`pipe` cannot be non-null, so the label reduces to a simple return.
Then, there is no point initialising `pipe` just to overwrite it before
anyone reads it.

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Rob Clark <robdclark@gmail.com>

test/amdgpu: fix test failure for SI

Signed-off-by: Flora Cui <Flora.Cui@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

tests/amdgpu: bypass VCE tests on raven

raven doesn't support VCE

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Xiaojie Yuan <Xiaojie.Yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

tests/amdgpu: bypass UVD CS tests on raven

raven doesn't support UVD decode

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Xiaojie Yuan <Xiaojie.Yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

amdgpu: Add FX-9800P Bristol Ridge iGPU id

Signed-off-by: Jan Vesely <jan.vesely@rutgers.edu>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>

radeon: add fallthrough annotation

GCC 7 started warning when a switch case has neither a `break` nor
a "fallthrough" comment.
Let's be explicit that we meant to fall through here.

Signed-off-by: Eric Engestrom <eric@engestrom.ch>
Reviewed-by: Michel Dänzer <michel.daenzer@amd.com>

xf86drm: continue with next device if drmProcessUsbDevice fails

Analogous to previous commit (and the rest of the codebase), simply
discard the device if we cannot parse it.

Fixes: f8484ccbd12 ("xf86drm: Add USB support")
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

xf86drm: continue after drmProcessPlatformDevice failure

On ChromeOS devices, readdir() processes the directory in
the following order:

-NAME-              -TYPE-
.                    n/a
..                   n/a
vgem                 n/a
card1           DRM_BUS_PLATFORM
renderD129      DRM_BUS_PLATFORM
card0             DRM_BUS_PCI
renderD128        DRM_BUS_PCI
controlD64        DRM_BUS_PCI

In drmGetDevices2, after drmProcessPlatformDevice fails for
/dev/dri/card1, we don't process the remaining directory entries.
As such, Vulkan fails to initialize since Mesa uses drmGetDevices2.
To fix this, continue if drmProcessPlatformDevice fails.

Fixes: 7b1f37f474d ("xf86drm: Add platform and host1x bus support")
Reviewed-by: Emil Velikov <emil.velikov@collabora.com>
[Emil: correct the host1x platforms as well]
Signed-off-by: Emil Velikov <emil.velikov@collabora.com>

configure.ac: bump version for release

Signed-off-by: Lucas Stach <l.stach@pengutronix.de>

amdgpu: add new symbols to tests.

drm/amdgpu: add new low overhead command submission API. (v2)

This just sends chunks to the kernel API for a single command
stream.

This should provide a more future proof and extensible API
for command submission.

v2: use amdgpu_bo_list_handle, add two helper functions to
access bo and context internals.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

drm/amdgpu: add syncobj create/destroy/import/export apis

These are just wrappers using the amdgpu device handle.

Acked-by: Chunming Zhou <david1.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

Remove redundant memclear

drmMalloc will zero out the memory for us

Reviewed-by: Eric Engestrom <eric.engestrom@imgtec.com>

etnaviv: submit full struct drm_etnaviv_gem_submit

It is safe to submit the full struct even on older kernels as such
kernels do not process the full struct. Without this change it
becomes quite challenging to extned the submit struct.

Freedreno has no special treatment too. See git commits
- freedreno: sync uapi header
- freedreno: add fence fd support

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Tested-by: Philipp Zabel <p.zabel@pengutronix.de>