Chris Wilson [Mon, 26 Aug 2013 13:45:26 +0000 (14:45 +0100)]
overlay: Trim the padding slightly
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Mon, 26 Aug 2013 13:11:27 +0000 (14:11 +0100)]
overlay: Count number of semaphores used by each process
This required me to contract the per-process information considerably,
hopefully readability is not sacrificed too much.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 25 Aug 2013 22:37:58 +0000 (23:37 +0100)]
overlay: Allow simple positioning and resizing
Using window.size=<width>x<height> or window.size=<scale>% in
the config file, or --size=<scale>% or --size=<width>x<height>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 25 Aug 2013 19:13:31 +0000 (20:13 +0100)]
overlay: Monitor interrupts
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 25 Aug 2013 14:17:39 +0000 (15:17 +0100)]
overlay: Detach from terminal
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 25 Aug 2013 12:15:55 +0000 (13:15 +0100)]
overlay: Add a raw kms output
For when you don't have any display server, use brute force.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 24 Aug 2013 11:59:31 +0000 (12:59 +0100)]
overlay: Draw the cpu busy line on top of the waits
By drawing it behind the waits, it is obscured by the translucent waits.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 24 Aug 2013 11:37:59 +0000 (12:37 +0100)]
overlay: simplify types for object counts for 32/64bit porting
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 24 Aug 2013 09:00:36 +0000 (10:00 +0100)]
overlay: Expose sampling period/frequency as a user parameter
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Daniel Vetter [Fri, 23 Aug 2013 20:32:29 +0000 (22:32 +0200)]
tests/gem_evict_*: Add interruptible modes
Also I've stumbled over igt_fixture again. Somehow gcc loves to wreak
havoc with stack variables set up in igt_fixtures ... Duct-tape in the
form of moving fd out to global scope applied.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Chris Wilson [Fri, 23 Aug 2013 15:02:42 +0000 (16:02 +0100)]
overlay: Fix power reading from debugfs
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Fri, 23 Aug 2013 14:59:06 +0000 (15:59 +0100)]
overlay: Distinguish the root debugfs path and our dri node
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Fri, 23 Aug 2013 14:51:21 +0000 (15:51 +0100)]
overlay: Search for right debugfs path
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Daniel Vetter [Fri, 23 Aug 2013 07:30:47 +0000 (09:30 +0200)]
tests/sysfs_rps: Fix igt_require conversion
Another fumble caught by QA.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68461
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Chris Wilson [Thu, 22 Aug 2013 21:50:05 +0000 (22:50 +0100)]
overlay: Replace debug ticker with hostname
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Thu, 22 Aug 2013 19:42:58 +0000 (20:42 +0100)]
overlay: Close DRI2 device after authentication failure
So that we don't leave a master fd open when operating remotely.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Thu, 22 Aug 2013 18:20:01 +0000 (19:20 +0100)]
overlay: Double buffer the x11 window
For pleasant remote transport.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Thu, 22 Aug 2013 18:19:03 +0000 (19:19 +0100)]
overlay: Make the config parser more lenient
Allow a dot-notation outside of the [ini] style to make commandline
parameters easier to add.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Paulo Zanoni [Thu, 22 Aug 2013 14:53:09 +0000 (11:53 -0300)]
tests/pc8: don't work around the eDP I2C bug
On the current Kernels, i2c operations on disabled eDP panels don't
work because we don't enable the panel before doing the i2c
transactions. Whenever we try these transactions we print an error
message on dmesg and fail. The current test suite was trying to work
around this problem: we count how many eDP outputs we have and add
them to the final result.
Daniel asked me to not work around the bug, so now the i2c test will
fail until we actually fix the Kernel bug (or stop using eDP panels).
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Paulo Zanoni [Thu, 22 Aug 2013 14:50:05 +0000 (11:50 -0300)]
tests/pc8: fix function return value
Looks like I forgot to push this chunk... The test was passing in
cases where you only have eDP monitors due to the fact that we
consider i2c won't work on eDP monitors.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Chris Wilson [Thu, 22 Aug 2013 14:07:48 +0000 (15:07 +0100)]
overlay: Rudiments of config files and option parsing
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Thu, 22 Aug 2013 10:59:58 +0000 (10:59 +0000)]
overlay: Fill the blank charts with error messages
Instead of showing nothing, show a "space left intentionally blank"
message.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Thu, 22 Aug 2013 00:30:02 +0000 (01:30 +0100)]
overlay: Remove busy estimation
Too inaccurate to be useful.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Rodrigo Vivi [Wed, 21 Aug 2013 18:13:13 +0000 (15:13 -0300)]
tests: ddx_intel_after_fbdev loads intel ddx after fbdev was loaded.
v2: Don't add to kernel tests.
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@gmail.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Chris Wilson [Wed, 21 Aug 2013 12:49:45 +0000 (13:49 +0100)]
overlay: Fix underflow for clamping initial values
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Wed, 21 Aug 2013 12:44:44 +0000 (13:44 +0100)]
overlay: i915 pmu doesn't require a specific frequency
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Wed, 21 Aug 2013 12:36:29 +0000 (13:36 +0100)]
overlay: Put a faint outline around each chart
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Wed, 21 Aug 2013 10:06:51 +0000 (11:06 +0100)]
overlay: Fade the graphs behind the text
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Wed, 21 Aug 2013 08:24:59 +0000 (09:24 +0100)]
tests: Add gem_evict_alignment
Clone gem_evict_everything and use it to exercise the misalignment
paths.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Daniel Vetter [Wed, 21 Aug 2013 08:22:13 +0000 (10:22 +0200)]
lib/drmtest: fix logic fumble in gem_context_create
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68363
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Chris Wilson [Wed, 21 Aug 2013 08:16:25 +0000 (09:16 +0100)]
overlay: Fix old mmio gpu busyness probing
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 22:19:45 +0000 (23:19 +0100)]
overlay: Fix sampling of i915 perf event
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 20:58:22 +0000 (21:58 +0100)]
overlay: Fix display of rc6 states
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 20:39:27 +0000 (21:39 +0100)]
overlay: Add copyright statements
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 18:11:44 +0000 (19:11 +0100)]
overlay: Use the new i915 PMU to query GPU busyness
And so avoid having to hold forcewake indefinitely.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 18:24:33 +0000 (19:24 +0100)]
overlay: Take a snapshot on SIGUSR1
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Daniel Vetter [Tue, 20 Aug 2013 20:03:30 +0000 (22:03 +0200)]
tests/pc8: Update to latest igt infrastructure
- Use igt_require for magic SKIP results while still printing results
for all subtests.
- Wrap setup/teardown code outside of subtests with igt_fixture. This
will ensure that the so wrapped code isn't run when an
igt_require/igt_assert failed (and we're essentially just printing
the SKIP/FAIL result for all subtests) or when just enumerating
subtests.
- Use igt_exit. Otherwise SKIP/FAIL results of subtests won't be propagated
correctly.
- igt_subtest checks should be run unconditionally to make sure we
always enumerate the same set of subtests. A bit on the line here
since this is a subtest hidden behind a runtime flag.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Paulo Zanoni [Mon, 19 Aug 2013 16:20:02 +0000 (13:20 -0300)]
tests: add pc8
This test chekcs our code that enables Package C8+. The environment
requirements for this test are quite complicated:
- The machine needs to be properly configured to reach PC8+ when
possible, which means all the power management policies and device
drivers must be properly configured.
- You need at least one output connected.
- You need the /dev/cpu/0/msr file available.
- You need the /dev/i2c-* files available.
v2: - Many many changes due to the latest review comments and rebase.
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Damien Lespiau [Tue, 20 Aug 2013 17:56:54 +0000 (18:56 +0100)]
gitignore: Add a couple of recent binaries to .gitignore files
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 20 Aug 2013 17:55:16 +0000 (18:55 +0100)]
quick_dump: Use AM_V_GEN when generating the bindings
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 20 Aug 2013 17:52:53 +0000 (18:52 +0100)]
overlay: Include gpu-perf.h for distribution
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 20 Aug 2013 17:25:52 +0000 (18:25 +0100)]
assembler: Disable the declare test
It's not hitting a valid assertion that it tries to write an instruction
without a defined execution size (because the "default" exec_size never
end up being set).
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 20 Aug 2013 17:22:45 +0000 (18:22 +0100)]
assembler: Disable tests that where already failing in the gen4asm repo
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 20 Aug 2013 17:12:49 +0000 (18:12 +0100)]
assembler: Ignore make check output
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 20 Aug 2013 17:02:37 +0000 (18:02 +0100)]
assembler: Fix the path of intel-gen4asm
With the move to intel-gpu-tools, we need to update that as well.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 20 Aug 2013 16:58:48 +0000 (17:58 +0100)]
assembler: Revert "Fix missing environment variables problem in test/run-test.sh"
Same as:
commit
497814f2f2828efdc5bdd787ebc490d5083f61b8
Author: Damien Lespiau <damien.lespiau@intel.com>
Date: Tue Aug 20 14:52:05 2013 +0100
assembler: Revert "Automatically run all test cases."
make check will define srcdir and buildir variables for us.
This reverts commit
1c009349bc894bd195b5522540536898b0bee574.
Damien Lespiau [Tue, 20 Aug 2013 13:52:05 +0000 (14:52 +0100)]
assembler: Revert "Automatically run all test cases."
The tests where supposed to be run through make check, not running the
"run-test.sh" standalone. So revert that patch to have make check work
as intended.
This reverts commit
6983eebf47f37def8f2315d5af1800b81644f240.
Damien Lespiau [Tue, 20 Aug 2013 13:25:52 +0000 (14:25 +0100)]
assembler: Tune the error message for invalid send on gen6+
And be a bit more descriptive.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 20 Aug 2013 13:20:06 +0000 (14:20 +0100)]
intel_infoframes: Be future-proof about showing 3D_Ext_Data
As Ville noted, future 3D_Struct must also send 3D_Ext_Data in the
vendor infoframe.
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Damien Lespiau [Tue, 20 Aug 2013 13:18:58 +0000 (14:18 +0100)]
build: Fix the overlay build summary at the end of configure
Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
Chris Wilson [Tue, 20 Aug 2013 11:45:55 +0000 (12:45 +0100)]
lib: Make igt_assert() abort under gdb
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 11:40:56 +0000 (12:40 +0100)]
gem_evict_everything: Fix the error code checking after drmIoctl
drmIoctl doesn't return the errno, so look it up after an error.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 10:53:22 +0000 (11:53 +0100)]
tests: Add gem_evict_everything
Exercise the eviction logic. This is just a naive test to ensure that we
can evict old buffers to accommodate new batches, and in the process
trigger the evict everything logic.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 10:09:11 +0000 (11:09 +0100)]
overlay: Sample power every second
Found the bug, but still only read the msr once every second - mainly as
an exercise in handling different sample rates.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 10:08:13 +0000 (11:08 +0100)]
overlay: use rc6_enable to drop useless information
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 09:58:02 +0000 (10:58 +0100)]
overlay: Fix counting fail
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 09:04:23 +0000 (10:04 +0100)]
overlay: Add graph for GPU power consumption
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Tue, 20 Aug 2013 08:27:34 +0000 (09:27 +0100)]
overlay: Add rc6 residency
Not much point at the moment since we use forcewake to continuously
probe the GPU busyness. But that will eventually change to a more
power-efficient in-kernel.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Daniel Vetter [Tue, 20 Aug 2013 04:39:43 +0000 (06:39 +0200)]
tests/gem_concurrent_blit: Fix segmentation fault
Apparently the simple changes introduced in
commit
78865847f9fae7e590960f9836f2ec8b611a190e
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
AuthorDate: Mon Aug 19 07:23:49 2013 +0200
Commit: Daniel Vetter <daniel.vetter@ffwll.ch>
CommitDate: Mon Aug 19 08:09:25 2013 +0200
lib/drmtest: skip fixtures after an igt_skip
pushed the stack space requirement for run_modes over some magic limit
(I guess where the strack grow logic starts to fail, but I didn't
check). Fix this by moving the data into .bss.
Also add some asserts to check that fixtures aren't abused while at
it (i.e. the useful parts of my debug printf/assert craze).
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68299
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Chris Wilson [Mon, 19 Aug 2013 22:29:08 +0000 (23:29 +0100)]
overlay: Wrap the colour index around the rgba array
And stop assigning colours beyond the end of the array. Still needs a
better method for colour assignment, big list of good colours & ida?
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Mon, 19 Aug 2013 21:26:01 +0000 (22:26 +0100)]
overlay: Add missing include "debugfs.h"
Later versions of gcc are apparently stricter.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Mon, 19 Aug 2013 17:15:04 +0000 (18:15 +0100)]
overlay: Auatomatically mount debugfs
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Mengdong Lin [Mon, 19 Aug 2013 00:36:03 +0000 (20:36 -0400)]
quick_dump/bdw: use hex() to convert devid to hex string in error message
This patch is to avoid the error on device auto-detection failure
"TypeError: Can't convert 'int' object to str implicitly".
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Mengdong Lin <mengdong.lin@intel.com>
Imre Deak [Wed, 31 Jul 2013 08:59:38 +0000 (11:59 +0300)]
tests: add kms_setmode
Iterate through all valid/invalid crtc/connector combinations. At the
moment only clone configurations are tested as the single output cases
are tested already by testdisplay. Also from combinations where all
connectors are on the same crtc (clone-single-crtc) only those are
tested that are invalid, as I haven't found any machine that supports
these (have to be GT2 with dvo and vga output).
For configurations with one crtc per connector the FBs are per-crtc atm.
Signed-off-by: Imre Deak <imre.deak@intel.com>
Imre Deak [Fri, 2 Aug 2013 23:10:11 +0000 (02:10 +0300)]
lib: add subtest extra command line option handling
At the moment any command line option handling done by tests will
interfere with the option handling of the subtest interface. To fix this
add a new version of the subtest_init function accepting optional short
and long command line options. Merge these together with the subtest
interface's own long options and handle both together in the same
getopt_long call.
Signed-off-by: Imre Deak <imre.deak@intel.com>
Imre Deak [Fri, 2 Aug 2013 14:13:37 +0000 (17:13 +0300)]
lib: handle SIGSEGV similarly to other error signals
Signed-off-by: Imre Deak <imre.deak@intel.com>
Imre Deak [Fri, 2 Aug 2013 14:13:21 +0000 (17:13 +0300)]
lib: shorten DP/eDP connector names
Signed-off-by: Imre Deak <imre.deak@intel.com>
Imre Deak [Thu, 1 Aug 2013 12:08:19 +0000 (15:08 +0300)]
lib: export kmstest_get_connector_default_mode
Signed-off-by: Imre Deak <imre.deak@intel.com>
Daniel Vetter [Mon, 19 Aug 2013 09:09:25 +0000 (11:09 +0200)]
lib/drmtest: igt_skip when drm_get_card fails
Also add printf support to igt_skip to make sure there's always a
reason why we fail a testcase.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Mon, 19 Aug 2013 08:55:28 +0000 (10:55 +0200)]
lib/drmtest: extract gem_create_context
libdrm requirements are now up to it.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Mon, 19 Aug 2013 08:47:53 +0000 (10:47 +0200)]
tests: roll out igt_require some more
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Mon, 19 Aug 2013 08:34:34 +0000 (10:34 +0200)]
lib/drmtest: rip out drm_open_any_master
It's unused. Also most of our tests failed to ask for the right type
of drm fd anyway. So it's imo better to just let them fall over when
they don't get master but want it, like they already do today.
This also allows us to garbage-collect the master parameter to
drm_get_card and associated code.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Mon, 19 Aug 2013 05:57:03 +0000 (07:57 +0200)]
lib/drmtest: Properly skip if no intel gfx supported
drm/* kernel test nirvana!
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Mon, 19 Aug 2013 05:28:42 +0000 (07:28 +0200)]
lib/drmtest: make igt_fail work outside of subtests
With the abortable fixtures we can now correctly fail all subsequent
sutbests if we hit upon an igt_assert outside of a subtest.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Daniel Vetter [Mon, 19 Aug 2013 05:23:49 +0000 (07:23 +0200)]
lib/drmtest: skip fixtures after an igt_skip
This way we can just enclose all igt_skip/igt_require calls into
fixtures even when we have subtests, and still output correct SKIP
message for all of them. Wohoo, magic!
The only thing which doesn't work yet is enumerating failed subtests,
but I think that should work out on top of this.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Chris Wilson [Mon, 19 Aug 2013 07:20:15 +0000 (08:20 +0100)]
overlay: Tweak label locations to include baseline offset.
Still approximately, too lazy to query the metrics.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 20:10:26 +0000 (21:10 +0100)]
overlay: Add a X11 window backend
Useful for remote hosts.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 18:20:24 +0000 (19:20 +0100)]
overlay: Tweak a couple of the fill colours to be slightly less horrid
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 18:16:29 +0000 (19:16 +0100)]
overlay: Don't smooth gpu freq
This is supposed to be discrete jumps, so use straight lines to emphasis
this nature.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Daniel Vetter [Sun, 18 Aug 2013 17:58:28 +0000 (19:58 +0200)]
test/gem_dummy_reloc_loop: fix igt_require conversion
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68170
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Chris Wilson [Sun, 18 Aug 2013 17:54:04 +0000 (18:54 +0100)]
overlay: Remove the double ':' from object comm names
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 17:45:14 +0000 (18:45 +0100)]
overlay: Use a more compact 2x2 layout for the graphs
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 17:40:03 +0000 (18:40 +0100)]
overlay: Rearrange GUI state
Planning a new layout...
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 17:27:47 +0000 (18:27 +0100)]
overlay: Print min/max GPU frequencies.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 17:24:43 +0000 (18:24 +0100)]
overlay: Convert gem objects to megabytes
Less confusing, I hope.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 17:17:05 +0000 (18:17 +0100)]
overlay: Parse gem objects
Condense the information and begin graphing it. Remaining todo for
memory is to measure bind/evict flux, and perhaps clflush.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 15:42:25 +0000 (16:42 +0100)]
overlay: Add GPU frequency
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 14:56:22 +0000 (15:56 +0100)]
overlay: Include CPU usage in the overview chart
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sun, 18 Aug 2013 10:15:08 +0000 (11:15 +0100)]
overlay: Count flips per plane
Stop the misleading double-accounting of flips when we have multiple
displays active.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 22:50:21 +0000 (23:50 +0100)]
overlay: Improve the horizontal fit of the charts
Notably fix the fill boundary to not extend past the right hand side of
the chart.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 22:07:01 +0000 (23:07 +0100)]
overlay: Correct layout of ring/seqno in raw sample
For seqno completion, the events are too coarse i.e. one event may
signal the completion of a few seqno. We will need to sort the events to
properly compute the busy times.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 21:33:35 +0000 (22:33 +0100)]
overlay: Accumulate busy times
Still a little too course as we add multiple overlapping seqnos and
waits.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 21:22:21 +0000 (22:22 +0100)]
overlay: Couple wait begin/end events together to fix accounting
Since the events may be processed out of order (due to per-cpu
ringbuffers) we need to be careful to associated wait pairs in order to
compute the correct elapsed time.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 21:07:49 +0000 (22:07 +0100)]
overlay: Perf output redirection must be done after mmap()
Effectively you choose to redirect the output of an event instead of
mmapping it directly - but the target must already be mmapped.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 20:14:10 +0000 (21:14 +0100)]
overlay: Show per-process wait times
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 19:32:58 +0000 (20:32 +0100)]
overlay: Graph per-process requests over time
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 19:04:11 +0000 (20:04 +0100)]
overlay: Track requests per-process
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 17:48:28 +0000 (18:48 +0100)]
overlay: Drop unused mmap/comm events
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 17:24:39 +0000 (18:24 +0100)]
overlay: Constify perf ringbuffer
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Chris Wilson [Sat, 17 Aug 2013 17:09:40 +0000 (18:09 +0100)]
overlay: Fix wraparound handling of perf ringbuffer
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>