platform/upstream/intel-gpu-tools.git
12 years agobuild: Make autoreconf honour ACLOCAL_FLAGS
Damien Lespiau [Wed, 25 Jul 2012 15:04:32 +0000 (16:04 +0100)]
build: Make autoreconf honour ACLOCAL_FLAGS

When running autoreconf, it's possible to give flags to the underlying
aclocal by declaring a ACLOCAL_AMFLAGS variable in the top level
Makefile.am.

Putting ${ACLOCAL_FLAGS} there allows the user to set an environment
variable up before running autogen.sh and pull in the right directories
to look for m4 macros, say an up-to-date version of the xorg-util macros.

Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
12 years agotest/flip_test: don't fail if we can't find a crtc
Daniel Vetter [Tue, 22 May 2012 17:55:21 +0000 (19:55 +0200)]
test/flip_test: don't fail if we can't find a crtc

some connectors are forced to certain crcs ...

12 years agotests: fixup prime test paranoia
Daniel Vetter [Mon, 23 Jul 2012 09:25:47 +0000 (11:25 +0200)]
tests: fixup prime test paranoia

12 years agotests/prime: more paranoia
Daniel Vetter [Mon, 23 Jul 2012 08:06:25 +0000 (10:06 +0200)]
tests/prime: more paranoia

- check the old contents first in check_bo
- drop all GEM handles for the re-import test

12 years agotests: add missing stuff to .gitignore
Daniel Vetter [Mon, 23 Jul 2012 07:51:12 +0000 (09:51 +0200)]
tests: add missing stuff to .gitignore

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: add basic re-importing prime test
Daniel Vetter [Mon, 23 Jul 2012 07:49:06 +0000 (09:49 +0200)]
tests: add basic re-importing prime test

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: s/drm_reg_read/gem_reg_read
Daniel Vetter [Mon, 23 Jul 2012 04:41:10 +0000 (06:41 +0200)]
tests: s/drm_reg_read/gem_reg_read

gem_ is the prefix for low-level kernel tests, drm_ tests
libdrm_intel specific stuff.

12 years agoshut up the compiler a bit
Daniel Vetter [Sat, 21 Jul 2012 15:36:36 +0000 (17:36 +0200)]
shut up the compiler a bit

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agoreg_read: use good offset, and errno
Ben Widawsky [Sun, 22 Jul 2012 23:30:04 +0000 (16:30 -0700)]
reg_read: use good offset, and errno

The test was bonghits.

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agoreg_read: skip older kernels
Ben Widawsky [Wed, 18 Jul 2012 18:47:42 +0000 (11:47 -0700)]
reg_read: skip older kernels

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agoreg_read: basic register read ioctl test
Ben Widawsky [Thu, 12 Jul 2012 00:06:01 +0000 (17:06 -0700)]
reg_read: basic register read ioctl test

This will need to get modified when the ioctl expands, and so is only
here for reference/to make Daniel happy.

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agotests: improve concurrent blit tests
Daniel Vetter [Fri, 13 Jul 2012 08:35:11 +0000 (10:35 +0200)]
tests: improve concurrent blit tests

By adding another testcase that follows up with a gpu read. This
checks whether the kernel properly tracks the pending write and
doesn't lose it (or sync up to the wrong seqno).

For some odd reason only the cpu mmap variant blows up, the gtt one
works here. I need to look into that some more.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agogem_ctx_basic: fixes
Ben Widawsky [Tue, 3 Jul 2012 18:04:05 +0000 (11:04 -0700)]
gem_ctx_basic: fixes

I must have checked in the wrong version the first time.

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agoctx_basic: we stopped using rendercopy
Ben Widawsky [Tue, 3 Jul 2012 17:40:32 +0000 (10:40 -0700)]
ctx_basic: we stopped using rendercopy

so get rid of the remants. perhaps we'll use it again someday.

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agorendercopy_gen7: add support for haswell
Ben Widawsky [Sun, 1 Jul 2012 02:26:59 +0000 (19:26 -0700)]
rendercopy_gen7: add support for haswell

Only compile tested.

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agocontext_basic: the most stressful context test to date.
Ben Widawsky [Thu, 14 Jun 2012 18:04:01 +0000 (11:04 -0700)]
context_basic: the most stressful context test to date.

this context test requiring libdrm updates

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agocontext: libdrm wrappers
Ben Widawsky [Sun, 15 Jan 2012 22:52:33 +0000 (14:52 -0800)]
context: libdrm wrappers

This wraps libdrm functionality to exec with contexts. This patch
shouldn't be applied until libdrm for contexts is updated.

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agointel: update libdrm requirement
Ben Widawsky [Fri, 29 Jun 2012 18:23:34 +0000 (11:23 -0700)]
intel: update libdrm requirement

We want context APIs in order to use our super context test.

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agogem_stress: use the gen7 rendercopy
Ben Widawsky [Fri, 29 Jun 2012 05:43:47 +0000 (22:43 -0700)]
gem_stress: use the gen7 rendercopy

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agolib: add a gen7 rendercopy
Ben Widawsky [Fri, 29 Jun 2012 05:42:58 +0000 (22:42 -0700)]
lib: add a gen7 rendercopy

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agoctx_bad_destroy: danvet changed to ENOENT
Ben Widawsky [Wed, 27 Jun 2012 18:13:23 +0000 (11:13 -0700)]
ctx_bad_destroy: danvet changed to ENOENT

In all fairness if the test worked, he would have found this earlier

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agoctx_bad_destroy - ioctl is 2e, not 23
Ben Widawsky [Wed, 27 Jun 2012 18:13:06 +0000 (11:13 -0700)]
ctx_bad_destroy - ioctl is 2e, not 23

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agoctx_bad_destroy: a little more description is nice
Ben Widawsky [Wed, 27 Jun 2012 18:07:09 +0000 (11:07 -0700)]
ctx_bad_destroy: a little more description is nice

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agotests/gem_wait_rendering_timeout: needs more fudge
Daniel Vetter [Tue, 26 Jun 2012 17:47:30 +0000 (19:47 +0200)]
tests/gem_wait_rendering_timeout: needs more fudge

Due to the way we calculate the workload by doubling it each time we
might end up with almost twice as much as we want. Hence increase our
fudge by 1.5 to account for that.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=50666

12 years agointel_reg_dumper: use intel_register_access_init/fini
Eugeni Dodonov [Thu, 21 Jun 2012 17:31:34 +0000 (14:31 -0300)]
intel_reg_dumper: use intel_register_access_init/fini

We need to hold forcewake lock in order to be able to read GT registers.
Otherwise, when the GPU is in RC6 mode, we'll read all zeros.

Signed-off-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
12 years agocontext tests not requiring libdrm
Ben Widawsky [Thu, 14 Jun 2012 17:00:57 +0000 (10:00 -0700)]
context tests not requiring libdrm

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agocontext: update for new execbuf2 element
Ben Widawsky [Mon, 19 Mar 2012 01:42:44 +0000 (18:42 -0700)]
context: update for new execbuf2 element

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agointel_reg_dumper: dump more PM registers
Eugeni Dodonov [Wed, 13 Jun 2012 18:34:40 +0000 (15:34 -0300)]
intel_reg_dumper: dump more PM registers

Signed-off-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
12 years agoadd VLV PCI ID
Jesse Barnes [Mon, 11 Jun 2012 19:08:32 +0000 (12:08 -0700)]
add VLV PCI ID

This allows the tests to run on the prototype boards.

12 years agotests/debgufs_emon_crash: make it quiet
Daniel Vetter [Fri, 8 Jun 2012 17:35:19 +0000 (19:35 +0200)]
tests/debgufs_emon_crash: make it quiet

12 years agointel_reg_dumper: dump pre-pch-split blc regs
Daniel Vetter [Thu, 7 Jun 2012 14:37:02 +0000 (16:37 +0200)]
intel_reg_dumper: dump pre-pch-split blc regs

12 years agotest/gem_gtt_speed: Add a baseline test for the performance of a CPU mmap
Chris Wilson [Mon, 4 Jun 2012 16:29:20 +0000 (17:29 +0100)]
test/gem_gtt_speed: Add a baseline test for the performance of a CPU mmap

When looking at the pwrite/pread/wc performance, it is useful to judge
that against the performance of an ordinary CPU mmap.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotest/sysfs_l3_parity: skip test if the sysfs interface doesn't exist
Daniel Vetter [Thu, 31 May 2012 13:45:17 +0000 (15:45 +0200)]
test/sysfs_l3_parity: skip test if the sysfs interface doesn't exist

12 years agotest/sysfs_l3_parity: fixup logic
Daniel Vetter [Thu, 31 May 2012 13:45:17 +0000 (15:45 +0200)]
test/sysfs_l3_parity: fixup logic

It didn't fail properly.

12 years agocleanups after l3_parity merge
Daniel Vetter [Thu, 31 May 2012 11:49:32 +0000 (13:49 +0200)]
cleanups after l3_parity merge

- call it sysfs_l3_parity for more consistency
- add .gitignore for intel_l3_parity

12 years agotests/dpf: simple dpf test
Ben Widawsky [Fri, 25 May 2012 23:56:27 +0000 (16:56 -0700)]
tests/dpf: simple dpf test

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agotools/dpf: Tool to read and write l3 remap registers.
Ben Widawsky [Fri, 25 May 2012 23:56:26 +0000 (16:56 -0700)]
tools/dpf: Tool to read and write l3 remap registers.

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agotests/gem_wait_rendering_timeout: check polling behaviour
Daniel Vetter [Wed, 30 May 2012 18:16:24 +0000 (20:16 +0200)]
tests/gem_wait_rendering_timeout: check polling behaviour

... with timeout=0. We have special code in the kernel to optimize
that one.

12 years agolib/gem_wait_rendering_timeout: sprinkle busy_ioctl checks
Daniel Vetter [Tue, 29 May 2012 20:23:14 +0000 (22:23 +0200)]
lib/gem_wait_rendering_timeout: sprinkle busy_ioctl checks

While we have that testcase, let's cross-check the busy ioctl.

12 years agolib: fix gen5 workaround emission
Daniel Vetter [Tue, 29 May 2012 20:14:06 +0000 (22:14 +0200)]
lib: fix gen5 workaround emission

Using BEGIN_BATCH can lead to a nice inf recursion through require_space
-> flush_batch -> BEGIN_BATCH.

Also fix things up to always require BATCH_RESERVED. We need 2 dwords
for the gen5 workaround and 2 dwords for MI_BB_END.

12 years agotests/gem_wait_rendering_timeout: test the timeout case
Daniel Vetter [Tue, 29 May 2012 14:52:23 +0000 (16:52 +0200)]
tests/gem_wait_rendering_timeout: test the timeout case

When I've first tried this it somehow failed, but I can't reproduce
this any more. I guess I've hunted a ghost.

12 years agotests/gem_wait_rendering_timeout: fix up the autotuner
Daniel Vetter [Tue, 29 May 2012 14:49:41 +0000 (16:49 +0200)]
tests/gem_wait_rendering_timeout: fix up the autotuner

duuh

12 years agotests/gem_wait_rendering_timout: adjust .gitignore
Daniel Vetter [Tue, 29 May 2012 14:35:35 +0000 (16:35 +0200)]
tests/gem_wait_rendering_timout: adjust .gitignore

12 years agotests/get_wait_rendering_timeout: fixup the auto-tuner
Daniel Vetter [Tue, 29 May 2012 14:35:11 +0000 (16:35 +0200)]
tests/get_wait_rendering_timeout: fixup the auto-tuner

- start at 1 iterations
- actually adjust the number of iterations like the comment claims

12 years agotests/gem_wait_rendering_timeout: some fixes for the blt function
Daniel Vetter [Tue, 29 May 2012 14:34:32 +0000 (16:34 +0200)]
tests/gem_wait_rendering_timeout: some fixes for the blt function

- widht/height are in pixel, and the cmd uses 32bit, so we need to
  adjust this.
- we write to the buffer, set the right reloc domain.

12 years agotests/gem_wait_rendering_timeout: properly skip test
Daniel Vetter [Tue, 29 May 2012 14:23:41 +0000 (16:23 +0200)]
tests/gem_wait_rendering_timeout: properly skip test

also assign the timeout value unconditionally, we want to know it even
when we've timeout out.

12 years agotests/gem_wait_rendering_timeout: make it compile
Daniel Vetter [Tue, 29 May 2012 14:14:31 +0000 (16:14 +0200)]
tests/gem_wait_rendering_timeout: make it compile

... we need to do this libdrm depency dance. Or just copy&paste it.

12 years agotests/wait render timeout test
Ben Widawsky [Fri, 11 May 2012 20:54:12 +0000 (13:54 -0700)]
tests/wait render timeout test

Assures that signals interrupting the wait works properly. Because of
the scheduling around signals, interrupted waits will *seem* faster as
the GPU continues to work while all the CPU scheduling stuff happens.

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agotests: fixup type in module_reload
Daniel Vetter [Fri, 25 May 2012 09:29:28 +0000 (11:29 +0200)]
tests: fixup type in module_reload

Oops, I've frobbed this to test the check but committed the wrong
version.

12 years agotests: add some help to module_reload
root [Fri, 25 May 2012 09:26:53 +0000 (11:26 +0200)]
tests: add some help to module_reload

... so that it's easier to get the kernel configuration right.

12 years agotests: add module reload testcase
root [Fri, 25 May 2012 09:02:09 +0000 (11:02 +0200)]
tests: add module reload testcase

We've broken this way too often in the past.

12 years agotests/hangman: fix when run with absolute path
Daniel Vetter [Wed, 23 May 2012 09:37:04 +0000 (11:37 +0200)]
tests/hangman: fix when run with absolute path

QA's test rig does that.

12 years agotest/flip_test: fail harder
Daniel Vetter [Tue, 22 May 2012 17:54:22 +0000 (19:54 +0200)]
test/flip_test: fail harder

We want testcase to return non-zero if they fail ...

12 years agolib: extract kmstest_dump_mode
Daniel Vetter [Tue, 22 May 2012 14:15:15 +0000 (16:15 +0200)]
lib: extract kmstest_dump_mode

12 years agotest/flip_test: use kmstest_create_fb
Daniel Vetter [Tue, 22 May 2012 14:01:32 +0000 (16:01 +0200)]
test/flip_test: use kmstest_create_fb

And bring back some frame indication that actually works.

12 years agolib: extract kmstest_create_fb
Daniel Vetter [Tue, 22 May 2012 12:37:19 +0000 (14:37 +0200)]
lib: extract kmstest_create_fb

We should get more kms tests soon, and not needing to copy-paste a
nice test pattern should be useful.

That establishes a firm depency of i-g-t on cairo over everything, but
I don't care so much about that.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: use drm_open_any more
Daniel Vetter [Tue, 22 May 2012 10:01:36 +0000 (12:01 +0200)]
tests: use drm_open_any more

And fix it up to not leak open fds, which kills all the master only
stuff.

12 years agotests/flip_test: rip out usage
Daniel Vetter [Tue, 22 May 2012 08:57:17 +0000 (10:57 +0200)]
tests/flip_test: rip out usage

Totally useless if the only option is "-h"

12 years agotests: add flip_test to .gitignore
Daniel Vetter [Tue, 22 May 2012 08:53:38 +0000 (10:53 +0200)]
tests: add flip_test to .gitignore

12 years agotests/flip_test: rip out funny pictures
Daniel Vetter [Tue, 22 May 2012 08:51:29 +0000 (10:51 +0200)]
tests/flip_test: rip out funny pictures

The path isn't properly autodetected and it yells stupid stuff into
stderr. Which causes spurious failures.

And really, these are automated tests. If you need to look at the output,
something is wrong.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agointel_infoframes: options -p and -P require an argument
Paulo Zanoni [Fri, 27 Apr 2012 19:14:34 +0000 (16:14 -0300)]
intel_infoframes: options -p and -P require an argument

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
12 years agointel_gpu_top: allow to work on Gen7
Eugeni Dodonov [Wed, 16 May 2012 13:14:55 +0000 (10:14 -0300)]
intel_gpu_top: allow to work on Gen7

Somehow nobody noticed this before, but we were missing blt and bsd6
initialization on Gen7.

Reported-by: Philippe Lecluse <philippe.lecluse@intel.com>
Signed-off-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
12 years agotests/gtt_speed: try to outwit gcc
Daniel Vetter [Thu, 10 May 2012 18:25:33 +0000 (20:25 +0200)]
tests/gtt_speed: try to outwit gcc

It optimized away the read loops ;-)

Also fix up the size to not be 1/4th of the one we want for tiled
buffer tests.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests/gem_tiled_pread_pwrite fix progname for pretty print progress
Chris Wilson [Wed, 9 May 2012 13:06:28 +0000 (14:06 +0100)]
tests/gem_tiled_pread_pwrite fix progname for pretty print progress

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
12 years agointel_error_decode: Be more lax for whitespace around parsing PCI-ID
Chris Wilson [Wed, 9 May 2012 12:50:53 +0000 (13:50 +0100)]
intel_error_decode: Be more lax for whitespace around parsing PCI-ID

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
12 years agotools: add Haswell registers into intel_reg_dumper
Eugeni Dodonov [Tue, 8 May 2012 18:55:05 +0000 (15:55 -0300)]
tools: add Haswell registers into intel_reg_dumper

For now, only print their content for diffing, but also add the necessary
bits that can be used for more verbose output in the fugure.

Signed-off-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests/flip_test: test on each CRTC for each detected output
Jesse Barnes [Mon, 7 May 2012 19:57:05 +0000 (12:57 -0700)]
tests/flip_test: test on each CRTC for each detected output

This allows us to test on 3 pipe configs for example.

12 years agotests: add simple flip test
Jesse Barnes [Mon, 7 May 2012 19:36:28 +0000 (12:36 -0700)]
tests: add simple flip test

Add a simple flip test to test flipping on a single output with the
preferred mode.

TODO: test each crtc for 3s and print pass/fail results

12 years agointel_infoframe: fix assertion off by 1
Ben Widawsky [Sun, 29 Apr 2012 03:10:09 +0000 (20:10 -0700)]
intel_infoframe: fix assertion off by 1

this makes my compiler very unhappy

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agochipset: accidentally left the old IS_GEN7 macro
Ben Widawsky [Sun, 29 Apr 2012 03:07:40 +0000 (20:07 -0700)]
chipset: accidentally left the old IS_GEN7 macro

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agotest/ZZ_hangman: adjust to new error_state reset semantics
Daniel Vetter [Fri, 27 Apr 2012 12:08:34 +0000 (14:08 +0200)]
test/ZZ_hangman: adjust to new error_state reset semantics

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agointel_infoframes: add support for Gen 4
Paulo Zanoni [Tue, 24 Apr 2012 21:10:25 +0000 (18:10 -0300)]
intel_infoframes: add support for Gen 4

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agointel_infoframes: AVI InfoFrame version should be 2
Paulo Zanoni [Tue, 24 Apr 2012 21:10:24 +0000 (18:10 -0300)]
intel_infoframes: AVI InfoFrame version should be 2

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agochipset updates
Ben Widawsky [Wed, 25 Apr 2012 20:33:17 +0000 (13:33 -0700)]
chipset updates

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agotest: fix up hangman script and enable by default
Daniel Vetter [Wed, 25 Apr 2012 11:21:38 +0000 (13:21 +0200)]
test: fix up hangman script and enable by default

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: add gem_unfence_active_buffers
Daniel Vetter [Sun, 22 Apr 2012 10:35:23 +0000 (12:35 +0200)]
tests: add gem_unfence_active_buffers

Unfortunately this requires slab poisoning to catch anything :(

Also add a new helper to drmtest to get the available fence count.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agolib: Fix render copy on gen2
Chris Wilson [Thu, 19 Apr 2012 13:11:51 +0000 (14:11 +0100)]
lib: Fix render copy on gen2

12 years agotests/gem_ringfill: add progress indicator
Daniel Vetter [Thu, 19 Apr 2012 10:34:16 +0000 (12:34 +0200)]
tests/gem_ringfill: add progress indicator

This thing takes ages on older chips.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: re-enable commented-out testcase
Daniel Vetter [Tue, 17 Apr 2012 20:38:44 +0000 (22:38 +0200)]
tests: re-enable commented-out testcase

Meh, I suck.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: add gem_set_tiling_vs_pwrite
Daniel Vetter [Tue, 17 Apr 2012 20:02:33 +0000 (22:02 +0200)]
tests: add gem_set_tiling_vs_pwrite

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests/gem_set_tiling_vs_gtt: update copyright header
Daniel Vetter [Tue, 17 Apr 2012 16:01:48 +0000 (18:01 +0200)]
tests/gem_set_tiling_vs_gtt: update copyright header

Meh, I've forgotten these.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: add gem_set_tiling_vs_gtt
Daniel Vetter [Tue, 17 Apr 2012 15:25:24 +0000 (17:25 +0200)]
tests: add gem_set_tiling_vs_gtt

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests/gem_set_tiling_vs_blt: fix on gen4+
Daniel Vetter [Tue, 17 Apr 2012 11:25:04 +0000 (13:25 +0200)]
tests/gem_set_tiling_vs_blt: fix on gen4+

Oops, the new checks need correctly tiled blts to work.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: rename gem_tiled_after_untiled_blt into gem_set_tiling_vs_blt
Daniel Vetter [Sun, 15 Apr 2012 21:59:12 +0000 (23:59 +0200)]
tests: rename gem_tiled_after_untiled_blt into gem_set_tiling_vs_blt

... it checks much more now.

Also update the comments a bit in the testcase.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: pimp tiled_after_untiled_blt
Daniel Vetter [Sun, 15 Apr 2012 21:40:20 +0000 (23:40 +0200)]
tests: pimp tiled_after_untiled_blt

Now with more tiling transition tests. Changes:
- refactoring to reuse the tiling transition functions.
- add a tiled->untiled transition
- add a tiled->tiled transition with different stride
- add a check at the end to ensure that the tiling has indeed changed
  on the test_bo

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agoAdd tools/intel_infoframes
Paulo Zanoni [Fri, 13 Apr 2012 13:27:01 +0000 (10:27 -0300)]
Add tools/intel_infoframes

This is a command-line tool that allows us to display and modify the
InfoFrames we send.

v2: use argv instead of stdin

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: skip rc6_residency test if not supported
Daniel Vetter [Thu, 12 Apr 2012 21:12:24 +0000 (23:12 +0200)]
tests: skip rc6_residency test if not supported

Fixes spurious failures.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agoUse mmap64() for performing the GTT mapping
Chris Wilson [Thu, 12 Apr 2012 14:47:00 +0000 (15:47 +0100)]
Use mmap64() for performing the GTT mapping

Although the address space for GTT mappings may only be 32-bits, we need
to use the explicit 64-bit mmap interface so that on a 32-bit platform
the offset we pass is not truncated to 31-bits.

Fixes gem_mmap_offset_exhaustion on 32-bit platforms.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
12 years agotests: add gem_gtt_concurrent_blit
Daniel Vetter [Wed, 11 Apr 2012 13:28:26 +0000 (15:28 +0200)]
tests: add gem_gtt_concurrent_blit

Same test as Chris Wilson's gem_cpu_concurrent_blit, but for
gtt mmaps.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: add sysfs_rc6_residency to .gitignore
Daniel Vetter [Wed, 11 Apr 2012 13:02:32 +0000 (15:02 +0200)]
tests: add sysfs_rc6_residency to .gitignore

12 years agotests/sysfs_rc6_residency: add some sleep before testing
Daniel Vetter [Wed, 11 Apr 2012 12:55:30 +0000 (14:55 +0200)]
tests/sysfs_rc6_residency: add some sleep before testing

That way I can run i-g-t on my ivb without spurious failures.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: rc6 residency test
Ben Widawsky [Sun, 25 Mar 2012 02:04:10 +0000 (19:04 -0700)]
tests: rc6 residency test

This is meant to test the sysfs entry for showing rc6 residency in
milliseconds. Remember, sysfs is a permanent interface.

v2: use new get_card interface to try "all" devices
check rc6p and rc6pp in addition to rc6

v3: rename rc6_residency.c to sysfs_rc6_residency.c
print better error messages
skip test if rc6 isn't enabled

v4: update to new sysfs names

Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Ben Widawsky <benjamin.widawsky@intel.com>
12 years agodrm/i915: extract card getting
Ben Widawsky [Sun, 25 Mar 2012 23:04:07 +0000 (16:04 -0700)]
drm/i915: extract card getting

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agobuild: make sure we have asprintf
Ben Widawsky [Sun, 25 Mar 2012 02:03:38 +0000 (19:03 -0700)]
build: make sure we have asprintf

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
12 years agotest/gem_cpu_concurrent_blit: Limit memory usage to half aperture
Chris Wilson [Tue, 10 Apr 2012 18:41:43 +0000 (19:41 +0100)]
test/gem_cpu_concurrent_blit: Limit memory usage to half aperture

So that we don't cause needless thrashing on older devices and spoil the
test.

12 years agotest: Exercise concurrent GPU read/write with CPU domain access
Chris Wilson [Tue, 10 Apr 2012 17:21:33 +0000 (18:21 +0100)]
test: Exercise concurrent GPU read/write with CPU domain access

Designed to exercise this patch to i915.ko:

diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index fbf1118..57ae1f2 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -3181,9 +3181,11 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_obj
        if (ret)
                return ret;

-       ret = i915_gem_object_wait_rendering(obj);
-       if (ret)
-               return ret;
+       if (write || obj->pending_gpu_write) {
+               ret = i915_gem_object_wait_rendering(obj);
+               if (ret)
+                       return ret;
+       }

        i915_gem_object_flush_gtt_write_domain(obj);

By exercising the conditions whereby should either of the checks be missed
an error is detected.

12 years agotests: explicit use bash
Daniel Vetter [Tue, 10 Apr 2012 14:23:18 +0000 (16:23 +0200)]
tests: explicit use bash

We use bashisms, which debian's default sh dash doesn't like.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: add wedging tests
Ben Widawsky [Wed, 4 Apr 2012 06:02:59 +0000 (23:02 -0700)]
tests: add wedging tests

Used to check that i915_error_state works. Unfortunately this
kills the gpu because wedging is permanent.

Base on a patch by Ben Widawsky.

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: fixup drm_lib.sh sourcing
Daniel Vetter [Wed, 4 Apr 2012 09:07:44 +0000 (11:07 +0200)]
tests: fixup drm_lib.sh sourcing

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
12 years agotests: use / as path separator instead of _
Daniel Vetter [Wed, 4 Apr 2012 08:56:50 +0000 (10:56 +0200)]
tests: use / as path separator instead of _

Oops. Originally by me, noticed while reviewing Ben Widawsky's patch.

Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>