Damien Lespiau [Tue, 4 Jun 2013 10:36:23 +0000 (11:36 +0100)]
lib: Use INSTDONE_I965 and INSTDONE_1 for gen6
The GEN6_INSTDONE_1 and GEN6_INSTDONE_2 registers are just the old
INSTDONE_I965 and INSTDONE_1 registers but renamed. Let's use the old
names (this is what the kernel does).
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 4 Jun 2013 08:06:50 +0000 (09:06 +0100)]
lib: Move the INSTDONE bit definitions to instdone.c
This is the only place where they are used and we've even started using
1 << n constants with gen 7.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 4 Jun 2013 05:48:49 +0000 (06:48 +0100)]
instdone: Add an assert to make sure we never overflow instdone_bits
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 4 Jun 2013 04:41:36 +0000 (05:41 +0100)]
intel_reg: Renamed INST_DONE to INSTDONE
That's how the registers are named in the kernel defines.
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Paulo Zanoni [Mon, 30 Dec 2013 13:40:03 +0000 (11:40 -0200)]
tests/kms_flip: free the test_output struct when counting modes
Looks like we have to do a lot of work just to count the number of
modes...
Caught by Valgrind.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Mon, 30 Dec 2013 13:33:16 +0000 (11:33 -0200)]
tests/kms_flip: don't leak the connector_config struct
... in case we can't find a compatible mode. We already have
config[0], we can't return without freeing it first.
Caught by Valgrind.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Mon, 30 Dec 2013 13:30:03 +0000 (11:30 -0200)]
tests/kms_flip: don't leak the CRTC
The kms_flip program calls kmstest_get_connector_config, which returns
a struct containing some allocated variables, including a pointer to
the CRTC. The problem is that we copy the values returned by this
structure to the test_output struct, but we ignore the CRTC pointer.
So free the CRTC pointer instead of leaking it.
Caught by Valgrind.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Mon, 30 Dec 2013 13:27:57 +0000 (11:27 -0200)]
tests/kms_flip: don't leak the connector when setting DPMS
Caught by Valgrind.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Mon, 30 Dec 2013 13:20:39 +0000 (11:20 -0200)]
drmtest: don't leak memory when parsing the arguments
Found this while investigating memory leaks on kms_flip. Detected by
Valgrind.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Daniel Vetter [Tue, 7 Jan 2014 07:14:33 +0000 (08:14 +0100)]
NEWS updates
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Damien Lespiau [Mon, 16 Dec 2013 11:28:39 +0000 (11:28 +0000)]
intel_error_decode: Factor out common decoding code
4 pieces of code were looking very similar. Let's factor out a common
function in the not so unlikely case we need to tweak that code.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Ben Widawsky [Fri, 3 Jan 2014 05:33:57 +0000 (19:33 -1000)]
configure.ac: Fix shader debugger for real
I tried to get fancy before, but ended up pushing something which I
didn't test. Keep the thing simple and stupid, and just make it work.
Reported-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Chris Wilson [Mon, 30 Dec 2013 17:48:43 +0000 (17:48 +0000)]
kmstest: Fix up lifetimes of cairo objects
cairo_t is the short lived drawing context, whereas cairo_surface_t is
the heavyweight object that persists and is also tied to underlying GEM
objects. So make the kmstest API reflect the different weights and fix
the lifetime and underlying object reference leaks.
Based on the fix by Paulo Zanoni.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Ben Widawsky [Mon, 30 Dec 2013 18:41:43 +0000 (10:41 -0800)]
configure.ac: Fixup conditional shader debugger
The original patch was incomplete. Testing no assembler build is a pain
for me. Sorry for the mess.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Mon, 30 Dec 2013 18:32:06 +0000 (10:32 -0800)]
configure.ac: Remove GEN4ASM define
This was a leftover hunk that got accidently merged.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Tue, 24 Dec 2013 21:12:31 +0000 (13:12 -0800)]
gem_exec_blit: gen8 updates for blit
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Sat, 21 Dec 2013 05:15:16 +0000 (21:15 -0800)]
shader-debugger: Use in tree gen4asm
Reported-by: Patrick McMunn <doctorwhoguy@gmail.com>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Daniel Vetter [Fri, 20 Dec 2013 15:56:14 +0000 (16:56 +0100)]
README: Fix spelling fail
Thanks to Jani for spotting this.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Fri, 20 Dec 2013 14:49:59 +0000 (15:49 +0100)]
README: Document how to avoid Waffle with piglit
Suggested by Paulo.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Paulo Zanoni [Mon, 16 Dec 2013 20:47:39 +0000 (18:47 -0200)]
tests/pm_pc8: check if we can't become DRM master
QA reported a failure that I believe happened because we couldn't
become DRM master, so add code that checks for this and prints a nice
error message.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Oscar Mateo [Fri, 13 Dec 2013 20:48:24 +0000 (20:48 +0000)]
build: Create version and config information files in Android
Also, skip the compilation of pm_lpsp in Android (uses Cairo)
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Oscar Mateo [Fri, 13 Dec 2013 16:24:37 +0000 (16:24 +0000)]
tests/gem_close_race: Adapt the test for Full PPGTT
With Full PPGTT, each new fd creates a new context and thus a new
PPGTT, so we have to reduce the number of simultaneous fds or face
OOM problems. For every new PPGTT, its PDEs are stored in the GGTT
which imposes a limit of 1024 new contexts. We want to leave at
least 1/4 of the GGTT available for "important" stuff like scanout
buffers, so never open more than 768 fds.
v2: Unconditionally use the lower limit, as suggested by Daniel
Vetter
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Damien Lespiau [Fri, 13 Dec 2013 12:53:31 +0000 (12:53 +0000)]
lib: Include the necessary stdint.h in gen6_render.h
Instead of assuming someone else will do it.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Daniel Vetter [Thu, 12 Dec 2013 21:46:18 +0000 (22:46 +0100)]
tests: Document the Makefile variables a bit better
Also, this is a test for the patchwork hook.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Mika Kuoppala [Tue, 10 Dec 2013 08:50:48 +0000 (10:50 +0200)]
tests/gem_reset_stats: add reverse order in close-pending-fork
Use own copy of gem_quiescent_gpu() so that test still works
if it gets changed. Further improve the test by posting a batch
to rings in reverse order.
Suggested-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Thu, 12 Dec 2013 08:26:58 +0000 (09:26 +0100)]
tests: Move gem_reset_stats to the right place
Binaries should only be added to the HANG target if they're completely
broken. If they just cause a recoverable gpu hang as part of the test
then they need to be somewhere else.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Paulo Zanoni [Tue, 3 Dec 2013 17:03:02 +0000 (15:03 -0200)]
tests/kms_setmode: avoid 2 connectors on the same encoder
Don't try to set modes on two connectors that share the same encoder.
That will just fail.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68463
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Imre Deak [Tue, 10 Dec 2013 13:33:07 +0000 (15:33 +0200)]
kms_flip: consolidate run_test_on_{crtc,crtc_pair}
No functional change.
Signed-off-by: Imre Deak <imre.deak@intel.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Imre Deak [Tue, 10 Dec 2013 12:00:19 +0000 (14:00 +0200)]
kms_flip: keep drm_fd open for the dpms restore exit handler
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=69161
Signed-off-by: Imre Deak <imre.deak@intel.com>
Imre Deak [Mon, 9 Dec 2013 15:28:50 +0000 (17:28 +0200)]
kms_flip: fix seq_step for 2xcrtc subtests
This also raises the absolute wait-for-vblank sequence step from 5 to
10, just to make it consistent with the corresponding 1x subtests.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=69161
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Kenneth Graunke [Tue, 10 Dec 2013 07:29:36 +0000 (23:29 -0800)]
rendercopy/bdw: Fix length of a bunch of zeroed out packets.
A bunch of packets were too short.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Cc: Ben Widawsky <ben@bwidawsk.net>
Cc: Damien Lespiau <damien.lespiau@intel.com>
Kenneth Graunke [Tue, 10 Dec 2013 07:29:35 +0000 (23:29 -0800)]
rendercopy/bdw: Emit 3DSTATE_WM_HZ_OP.
We don't want depth/stencil fast clears or HiZ resolves; we want normal
drawing. Without this, the pixel pipeline doesn't work.
Signed-off-by: Kenneth Graunke <kenneth@whitecape.org>
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Cc: Ben Widawsky <ben@bwidawsk.net>
Cc: Damien Lespiau <damien.lespiau@intel.com>
Daniel Vetter [Tue, 10 Dec 2013 09:30:59 +0000 (10:30 +0100)]
ignore version.h
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 10 Dec 2013 09:18:32 +0000 (10:18 +0100)]
tests: roll out igt_simple_init/igt_simple_main
Also use igt_skip a bit more to simplify some of the tests.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 10 Dec 2013 08:58:28 +0000 (09:58 +0100)]
tests: gem_close_race has subtests!
Oops.
v2: Found more.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 10 Dec 2013 08:41:48 +0000 (09:41 +0100)]
lib: igt_simple_main/init for subtest-less tests
Atm only used to print the version information.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 10 Dec 2013 08:26:08 +0000 (09:26 +0100)]
lib: print uname information in the version string
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 10 Dec 2013 08:23:44 +0000 (09:23 +0100)]
lib: Add host platform to version string
This is import on systems where we compile 32bit tests that run on
64bit platforms.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Ben Widawsky [Sat, 7 Dec 2013 04:35:29 +0000 (20:35 -0800)]
intel-gpu-tools: Version information
Provide two arguments version, and verbose, which allow printing from an
arbitrary igt test. It will show system information (from build time,
not runtime), as well as the git SHA being used.
This will help reduce errors when people try to reproduce problems.
As an example if I want to verify someone is running the correct version
of a test, I could ask them to do:
bwidawsk@ironside ~/intel-gfx/intel-gpu-tools (master)$ sudo ./tests/gem_exec_nop --verbose
gem_exec_nop-git-3c5423b (Linux ironside 3.12.0-1-ARCH #1 SMP PREEMPT
Wed Nov 6 09:06:27 CET 2013 x86_64 GNU/Linux)
Time to exec x 1: 35.000µs (ring=render)
Time to exec x 2: 28.000µs (ring=render)
Time to exec x 4: 20.000µs (ring=render)
Time to exec x 8: 14.625µs (ring=render)
Time to exec x 16: 11.188µs (ring=render)
Time to exec x 32: 11.125µs (ring=render)
Time to exec x 64: 10.328µs (ring=render)
Time to exec x 128: 10.172µs (ring=render)
Time to exec x 256: 10.234µs (ring=render)
Time to exec x 512: 10.232µs (ring=render)
Time to exec x 1024: 10.121µs (ring=render)
Time to exec x 2048: 10.151µs (ring=render)
Time to exec x 4096: 11.474µs (ring=render)
Time to exec x 8192: 9.432µs (ring=render)
Time to exec x 16384: 6.003µs (ring=render)
Time to exec x 32768: 5.029µs (ring=render)
Time to exec x 65536: 4.206µs (ring=render)
Time to exec x 131072: 3.630µs (ring=render)
Subtest render: SUCCESS
--verbose is provided for completeness, but doesn't seem too useful at
the moement.
bwidawsk@ironside ~/intel-gfx/intel-gpu-tools (master)$ sudo ./tests/gem_exec_nop --version
gem_exec_nop-git-3c5423b (Linux ironside 3.12.0-1-ARCH #1 SMP PREEMPT
Wed Nov 6 09:06:27 CET 2013 x86_64 GNU/Linux)
I've put version.h in the root directory so that any subdir can access
it. I've added the tests usage since it's immediately useful, and done
easily via Daniels igt infrastructure work.
v2:
- Always print the version number.
- We want to print uname at runtime.
- Also prefix the i-g-t release version.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net> (v1)
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 10 Dec 2013 07:07:11 +0000 (08:07 +0100)]
lib: Always print errno in igt assert/require macros
Useful to figure out why things have failed.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Tvrtko Ursulin [Fri, 6 Dec 2013 12:38:49 +0000 (12:38 +0000)]
tests/gem_media_fill: Remove unnecessary include
Causes trouble for Android builds.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Paulo Zanoni [Tue, 3 Dec 2013 21:44:45 +0000 (19:44 -0200)]
tests: add pm_lpsp
QA has asked me "How can we make sure LPSP is working?". Now, instead
of writing big paragraphs, I can just answer "make sure pm_lpsp
works".
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Tue, 3 Dec 2013 18:44:08 +0000 (16:44 -0200)]
lib: rename some power well bit names
I did the same change in the Kernel a few months ago. This should help
not getting confused about which bit does what.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Xiang, Haihao [Fri, 6 Dec 2013 01:16:58 +0000 (09:16 +0800)]
assembler/bdw: Update write(...)
write(...) is used for Render Target Write and Media Block Write.
The two message types no longer share the same cache agent on GEN8,
So a parameter is needed for cache agent. The 4th parameter of write()
is used for write commit bit which has been removed since GEN7. Hence
we can re-use the 4th parameter as cache agent on GEN8
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Daniel Vetter [Fri, 6 Dec 2013 09:48:25 +0000 (10:48 +0100)]
lib: make igt_pipe_crc_start never fail
It's what callers expect - pipe_crc_new is the function where
we pass a potential failure back to callers.
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Fri, 6 Dec 2013 09:47:10 +0000 (10:47 +0100)]
lib: add igt_pipe_crc_check
No need to duplicate this all over the place.
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Fri, 6 Dec 2013 09:45:49 +0000 (10:45 +0100)]
tests: drm_open_any doesn't fail
Or more precisely: It already has an igt_require. So we cant ditch it
from tests.
Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Damien Lespiau [Fri, 6 Dec 2013 13:02:51 +0000 (13:02 +0000)]
gitignore: Ignore gem_media_fill
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Xiang, Haihao [Fri, 6 Dec 2013 08:54:45 +0000 (16:54 +0800)]
rendercopy/bdw: Set Instruction Buffer size Modify Enable to 1
Otherwise it may result in GPU hang
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Xiang, Haihao [Fri, 6 Dec 2013 08:54:44 +0000 (16:54 +0800)]
lib: Clean the batch buffer store after reset
Otherwise the stale data in the buffer
Reviewed-by: Kenneth Graunke <kenneth@whitecape.org>
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Daniel Vetter [Fri, 6 Dec 2013 07:39:31 +0000 (08:39 +0100)]
tests/gem_reset_stats: document the new subtest a bit better
I wanted to add this to Mika's original patch but forgot to git add.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Xiang, Haihao [Mon, 2 Dec 2013 04:52:08 +0000 (12:52 +0800)]
tests/gem_media_fill: the assembly code for the shader used in the case
The code is for reference only
v2: Fixed the source register used for the send with EOT
Fixed the posted destination operand for the send with EOT
v3: Fixed the cache agent used in media_block_write message on GEN8
Reviewed-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Mon, 2 Dec 2013 04:49:14 +0000 (12:49 +0800)]
tests/gem_media_fill: add support for gen7
v2: Fixed the source register used for the send with EOT
Fixed the posted destination operand for the send with EOT
Reviewed-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Mon, 2 Dec 2013 04:36:15 +0000 (12:36 +0800)]
tests/gem_media_fill: add support for gen8
v2: Fixed the source register used for the send with EOT
Fixed the posted destination operand for the send with EOT
v3: Workaround: Insert MEDIA_STATE_FLUSH after MEDIA_OBJECT.
Fixed the cache agent used in media_block_write message
Set Instruction Buffer size Modify Enable to 1, otherwise it may result in GPU hang
Reviewed-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Fri, 29 Nov 2013 05:47:31 +0000 (13:47 +0800)]
tests: add gem_media_fill
It is to check whether media pipeline on render ring works. Codes
are copied and modified from the rendercopy case which uses 3D pipeline.
However media pipeline is simpler than 3D pipeline and there is few changes
between gen6,gen7 and gen8
Reviewed-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Ben Widawsky [Fri, 6 Dec 2013 00:02:52 +0000 (16:02 -0800)]
gem_exec_faulting_reloc: Update gen8 BLT cmds
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Thu, 5 Dec 2013 23:35:01 +0000 (15:35 -0800)]
gen8 rendercpy: temporarily disable
We don't yet have this working, and the excessive hangs are annoying.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Thu, 5 Dec 2013 23:00:44 +0000 (15:00 -0800)]
Kill XY_COLOR_BLT_CMD
Since we now always want a length for this command, and we've created a
non-length variant, remove the #define to prevent further foot shooting.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Thu, 5 Dec 2013 22:44:23 +0000 (14:44 -0800)]
gem_reloc_vs_gpu: Update for gen8 BLT cmds
NOTE: Some subtests are still failing, but I haven't found the bug yet.
At least this should help QA meanwhile.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Thu, 5 Dec 2013 22:44:23 +0000 (14:44 -0800)]
gem_persistent_relocs: Update for gen8 BLT cmds
NOTE: Some subtests are still failing, but I haven't found the bug yet.
At least this should help QA meanwhile.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Thu, 5 Dec 2013 22:36:20 +0000 (14:36 -0800)]
gem_write_read_ring_switch: Missed gen8 BLT updates
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Thu, 5 Dec 2013 22:34:35 +0000 (14:34 -0800)]
drm_vma_limiter_cached: Missed gen8 BLT updates
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Ben Widawsky [Thu, 5 Dec 2013 22:14:35 +0000 (14:14 -0800)]
gem_pipe_control_store_loop: BDW update
I've opted to not use the PIPE_CONTROL w/a for now. I am unclear if it
is actually required (the test does pass).
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Daniel Vetter [Thu, 5 Dec 2013 14:49:58 +0000 (15:49 +0100)]
tests/gem_ctx_exec: Add eviction test
It didn't pan out at all since atm we don't evict context objects. But
maybe it's useful for when that happens.
v2: Actually test eviction. Took a while to figure that one out.
And indeed the testcase now fails because we don't evict the previous
context as we should.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Thu, 5 Dec 2013 14:50:18 +0000 (15:50 +0100)]
lib: add gem_available_aperture_size helper
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Thu, 5 Dec 2013 12:41:51 +0000 (13:41 +0100)]
tests/core_get_client_auth: new subtests for always authenticated
This exercise the bug fixed in
commit
1020dc6990168a5081ffad620c440e220f05b460
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Tue Oct 29 08:55:57 2013 +0000
drm: Do not drop root privileges for a fancier younger process
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Thu, 5 Dec 2013 12:35:50 +0000 (13:35 +0100)]
tests: rename get_client_auth to have a core_ prefix
Also sort .gitignore.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Ville Syrjälä [Thu, 5 Dec 2013 11:01:58 +0000 (13:01 +0200)]
quick_dump: Restore execture permission on quick_dump.py
quick_dump.py execute permission was accidentally lost in:
commit
d56697233699844db9262a9a056b62b24b69dd39
Author: Mengdong Lin <mengdong.lin@intel.com>
Date: Sun Aug 25 23:07:22 2013 -0400
quick_dump/bdw: support Broadwell in device auto-detection
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Chon Ming Lee [Wed, 4 Dec 2013 12:44:33 +0000 (20:44 +0800)]
Update dpio read/write to take in extra PHY parameter.
The extra parameter is for future platform.
Signed-off-by: Chon Ming Lee <chon.ming.lee@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Chon Ming Lee [Wed, 4 Dec 2013 12:44:32 +0000 (20:44 +0800)]
tools/quick_dump: Update Baytrail DPIO names to reflect to the driver change.
DPIO name still using old name. Change it according to the driver name.
Signed-off-by: Chon Ming Lee <chon.ming.lee@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Mika Kuoppala [Wed, 4 Dec 2013 14:39:09 +0000 (16:39 +0200)]
tests/gem_reset_stats: add close-pending-fork
This triggers use after free oops on request->batch_obj when
going through the rings and setting reset status on requests,
after a gpu hang.
v2: Streamlined the test and added comments (Daniel)
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Mika Kuoppala [Tue, 26 Nov 2013 07:26:17 +0000 (09:26 +0200)]
lib/drmtest: Fix filedesc leak on gem_quiescent_gpu
to not run out of filedescs as some tests call gem_quiescent_gpu
directly.
Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Tvrtko Ursulin [Tue, 3 Dec 2013 16:44:54 +0000 (16:44 +0000)]
Revert "lib/drmtest: ducttape over fork race"
This reverts commit
a031a1bf93b828585e7147f06145fc5030814547.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Conflicts:
lib/drmtest.c
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 3 Dec 2013 16:57:06 +0000 (17:57 +0100)]
Revert "test/gem_concurrent_blt"
This reverts commit
aee0dcb1ec2075991d310dd6f3fb5e50160847d1.
Oops, accidentally commit a bit of wip changes.
Tvrtko Ursulin [Tue, 3 Dec 2013 16:44:55 +0000 (16:44 +0000)]
drmtest: Avoid wrong PID/TID after clone races
Various C library implementations have various races with regards
to caching getpid() or TID inside pthread_kill() implementations.
For example see clone(2) glibc man page and pthread_kill
Bionic C library source.
Work around that by making sure correct PID/TGID and TID values
are retrieved from the kernel when re-raising the signal. It
can be delivered immediately after the clone system call while C
library cached copies have not yet been updated.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Tvrtko Ursulin [Tue, 3 Dec 2013 15:35:41 +0000 (15:35 +0000)]
build: Skip kms_fbc_crc on Android
We do not have GLib there so it does not build.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Tvrtko Ursulin [Tue, 3 Dec 2013 15:09:57 +0000 (15:09 +0000)]
build: Make sure asserts are enabled for tests
Tests depend on assertions being enabled since they can, and do,
contain actual test steps. They are also mandatory for ensuring
sane test case behaviour.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 3 Dec 2013 15:32:52 +0000 (16:32 +0100)]
test/gem_concurrent_blt
Limit working set size also with available ram.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=72255
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Tue, 3 Dec 2013 12:07:28 +0000 (13:07 +0100)]
lib: Don't put functional code into asserts
Android apparently compiles with NDEBUG. Which is strongly advised
against, since the big pile of asserts mostly check correct usage of
the helper code in testcases.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Chris Wilson [Tue, 3 Dec 2013 10:12:37 +0000 (10:12 +0000)]
drmtest: Fix weird signage on errno
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Damien Lespiau [Thu, 28 Nov 2013 14:35:45 +0000 (14:35 +0000)]
gitignore: Ignore kms_fbc_crc
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Thu, 28 Nov 2013 14:34:51 +0000 (14:34 +0000)]
kms_fbc_crc: Remove unused variable
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Fri, 22 Nov 2013 18:14:26 +0000 (18:14 +0000)]
rendercopy/bdw: Fix the 3DSTATE_STENCIL_BUFFER instruction length
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Fri, 22 Nov 2013 18:12:35 +0000 (18:12 +0000)]
rendercopy/bdw: Fix the 3DSTATE_HIER_DEPTH_BUFFER instruction length
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Fri, 22 Nov 2013 18:11:03 +0000 (18:11 +0000)]
rendercopy/bdw: Fix the various 3DSTATE_CONSTANT_* instruction length
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Fri, 22 Nov 2013 17:36:54 +0000 (17:36 +0000)]
rendercopy/bdw: Fix the STATE_SIP instruction length
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Fri, 22 Nov 2013 17:29:02 +0000 (17:29 +0000)]
intel_dump_decode: Support the INTEL_DEVID_OVERRIDE env variable
This is the one that already works in libdrm, so don't disappoint people
coming with expectations.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Fri, 22 Nov 2013 17:24:52 +0000 (17:24 +0000)]
intel_dump_decode: Actually parse the -d option
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Ville Syrjälä [Sat, 9 Nov 2013 18:58:27 +0000 (20:58 +0200)]
kms_fbc_crc: Add a CRC based FBC test
kms_fbc_crc will perform various write operations to the scanout buffer
whilc FBC is enabled. CRC checks will be used to make sure the
modifcations to scanout buffer are detected.
The operations include:
- page flip
- GTT mmap
- CPU mmap
- blit
- rendercopy
- context switch + rendercopy
- combination of a page flip and each operation listed above
v2: Use gem_sw_finish instead of drmModeDirtyFB after CPU access
v3: Drop pwrite tests, call gem_bo_busy() after rendering, drop
set_domain() calls after mmap access, wait for 2 vblanks
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Ville Syrjälä [Mon, 25 Nov 2013 15:31:13 +0000 (17:31 +0200)]
lib: Add gem_bo_busy
Move gem_bo_busy() from gem_wait_render_timeout.c to lib.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Ville Syrjälä [Fri, 22 Nov 2013 20:33:21 +0000 (22:33 +0200)]
lib: Add gem_sw_finish()
Wrap DRM_IOCTL_I915_GEM_SW_FINISH into gem_sw_finish()
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Ville Syrjälä [Thu, 21 Nov 2013 17:05:17 +0000 (19:05 +0200)]
rendercopy: Pass context to rendercopy functions
rendercopy does the batch buffer flush internally, so if we want
to use it with multiple contexts, we need to pass the context
in from caller.
v2: Modify rendercopy_gen8 as well
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Ville Syrjälä [Fri, 22 Nov 2013 21:46:32 +0000 (23:46 +0200)]
kms_cursor_crc: Fix crc readout after a subtest failure
If a subtest fails, it'll leave the pipe CRC file open, which will
prevent subsequent subtests from opening the file. Make sure the file
is cloesed before trying to open it again.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Ville Syrjälä [Thu, 21 Nov 2013 17:48:51 +0000 (19:48 +0200)]
lib: Fix pipe_crc->ctl_fd assert
Use the correct fd in the assert().
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Daniel Vetter [Wed, 27 Nov 2013 10:31:12 +0000 (11:31 +0100)]
tests: Document exec naming convention for execbuf related tests
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Wed, 27 Nov 2013 10:14:19 +0000 (11:14 +0100)]
tests: Document per-ring naming conventions
Inspired by the recent pile of tests for the CS, which doesn't split
up the tests into per-ring subtests.
Cc: bradley.d.volkin@intel.com
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Paulo Zanoni [Tue, 26 Nov 2013 13:32:33 +0000 (11:32 -0200)]
tests/pm_pc8: add "resuming" state
This one was missing. For some reason we never really detected it on
our test suite. I checked the Kernel source and now we should be fine.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Thu, 21 Nov 2013 20:19:07 +0000 (18:19 -0200)]
tests/pm_pc8: try to reuse FBs when possible
Instead of creating a new FB every time we need one, create a cache of
FBs and reuse whenever possible. That means we'll create at most two
FBs, and reuse them hundreds and hundreds of times.
The kmstest_paint_test_pattern function takes about 1-2 seconds to
run, and we'll avoid it whenever we reuse the FB.
This makes the time taken to run the modeset-lpsp-stress subtest go
from 2:29 to 1:29.
A full "time ./pm_pc8 --quick" goes from 8:14 to 6:27.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Thu, 21 Nov 2013 19:32:42 +0000 (17:32 -0200)]
tests/pm_pc8: try harder to reach PC8+ residencies
If we want to actually reach PC8+ states, we need to properly
configure all the devices on the system to allow this. This function
will try to setup the things we know we need, but won't scream in case
anything fails: we don't know which devices are present on your
machine, so we can't really expect anything, just try to help with the
more common problems.
Another reason for this commit is that I got tired of having to
readjust the runtime PM policies every time I reboot my machine.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Thu, 21 Nov 2013 17:09:58 +0000 (15:09 -0200)]
tests/pm_pc8: rename some subtests
Due to copy/paste error, we had 3 subtests called gem-execbuf-stress.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>