platform/upstream/intel-gpu-tools.git
14 years agodecode: Split decoder and reuse outside of intel_gpu_dump
Chris Wilson [Mon, 30 Nov 2009 15:20:29 +0000 (15:20 +0000)]
decode: Split decoder and reuse outside of intel_gpu_dump

Introduce a second decoder that simpler operates on a file so that it
can be run on dumped batchbuffers independent of the gpu.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
15 years agoAdd: tools/intel_audio_dump
Wu Fengguang [Fri, 6 Nov 2009 03:06:22 +0000 (11:06 +0800)]
Add: tools/intel_audio_dump

Signed-off-by: Zhenyu Wang <zhenyu.z.wang@intel.com>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
15 years agoAdd: tools/intel_gpu_time
Chris Wilson [Thu, 8 Oct 2009 05:48:02 +0000 (06:48 +0100)]
Add: tools/intel_gpu_time

A simple variant on time crossed with intel_gpu_top to estimate the
amount of GPU activity during a program's execution time.

The resource usage (sys+user, i.e. CPU %) is for the child, but the
GPU measure is global. Hmm.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
15 years agoAdd intel_gpu_dump from the 2D driver.
Eric Anholt [Wed, 7 Oct 2009 00:25:10 +0000 (17:25 -0700)]
Add intel_gpu_dump from the 2D driver.

15 years agoBump to 1.0.2 for release. 1.0.2
Eric Anholt [Mon, 21 Sep 2009 22:00:56 +0000 (15:00 -0700)]
Bump to 1.0.2 for release.

15 years agoPrint unmatched opcode value for all 3D UNKNOWN messages
Carl Worth [Thu, 17 Sep 2009 22:05:24 +0000 (15:05 -0700)]
Print unmatched opcode value for all 3D UNKNOWN messages

This just simplifies one step for fixing up an incompletely-parsed
dump.

15 years agoAdd support for new chips
Xiang, Haihao [Thu, 3 Sep 2009 01:50:00 +0000 (09:50 +0800)]
Add support for new chips

Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
15 years agoCS FENCE in URB_FENCE is 11-bits wide
Xiang, Haihao [Thu, 3 Sep 2009 01:35:50 +0000 (09:35 +0800)]
CS FENCE in URB_FENCE is 11-bits wide

Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
15 years agoFix the length of CONSTANT_BUFFER
Xiang, Haihao [Thu, 3 Sep 2009 01:33:36 +0000 (09:33 +0800)]
Fix the length of CONSTANT_BUFFER

Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
15 years agointel_gpu_dump: Don't pretend we understand vertex formats on 8xx.
Eric Anholt [Mon, 20 Jul 2009 18:46:11 +0000 (11:46 -0700)]
intel_gpu_dump: Don't pretend we understand vertex formats on 8xx.

15 years agointel_gpu_dump: Parse LOAD_STATE_IMMEDIATE_2 for 830.
Eric Anholt [Mon, 20 Jul 2009 18:44:56 +0000 (11:44 -0700)]
intel_gpu_dump: Parse LOAD_STATE_IMMEDIATE_2 for 830.

15 years agointel_gpu_dump: Add IPEIR and PGTBL_ER to output.
Eric Anholt [Mon, 20 Jul 2009 16:14:43 +0000 (09:14 -0700)]
intel_gpu_dump: Add IPEIR and PGTBL_ER to output.

15 years agoAdd INSTDONE bits for 830-865.
Eric Anholt [Mon, 20 Jul 2009 16:07:16 +0000 (09:07 -0700)]
Add INSTDONE bits for 830-865.

15 years agointel_gpu_dump: Fix decode for LOAD_STATE_IMMEDIATE_1 with S8 set.
Eric Anholt [Thu, 16 Jul 2009 21:15:47 +0000 (14:15 -0700)]
intel_gpu_dump: Fix decode for LOAD_STATE_IMMEDIATE_1 with S8 set.

15 years agoAdd a manpage for intel_reg_write
Eric Anholt [Mon, 13 Jul 2009 17:03:17 +0000 (10:03 -0700)]
Add a manpage for intel_reg_write

15 years agodump: Minor typo
Chris Wilson [Thu, 3 Sep 2009 09:58:59 +0000 (10:58 +0100)]
dump: Minor typo

15 years agoAdd gem_fence_thrash test
Chris Wilson [Tue, 1 Sep 2009 09:09:55 +0000 (10:09 +0100)]
Add gem_fence_thrash test

This is a test case that overcommits fence registers between threads, which
are copying from one fenced bo to another. In earlier versions of the driver
this would cause excessive spinning as the first inactive (i.e. not in use
by the GPU) would be used to service the next page. After all the fence
registers had been allocated, in effect only the very first fence would then
be used for all subsequent faults.

15 years agoAdd CS_USB_STATE decoding
Ben Gamari [Wed, 1 Jul 2009 20:52:20 +0000 (16:52 -0400)]
Add CS_USB_STATE decoding

15 years agoAdd CONSTANT_BUFFER decoding
Ben Gamari [Wed, 1 Jul 2009 20:52:19 +0000 (16:52 -0400)]
Add CONSTANT_BUFFER decoding

15 years agoFix gs_fence output in URB_FENCE output
Ben Gamari [Wed, 1 Jul 2009 20:52:18 +0000 (16:52 -0400)]
Fix gs_fence output in URB_FENCE output

15 years agoadd realloc bits to URB_FENCE decoding
Ben Gamari [Wed, 1 Jul 2009 20:52:17 +0000 (16:52 -0400)]
add realloc bits to URB_FENCE decoding

15 years agoFix error message for case of kernel prior to 2.6.30
Carl Worth [Fri, 8 May 2009 23:04:24 +0000 (16:04 -0700)]
Fix error message for case of kernel prior to 2.6.30

The error message was falsely claiming that a debugfs directory
didn't exist, when it was supposed to report that the ringbuffer
file did not exist within that directory.

15 years ago.gitignore: catch up to intel_reg_write addition
Eric Anholt [Wed, 1 Jul 2009 01:05:57 +0000 (18:05 -0700)]
.gitignore: catch up to intel_reg_write addition

15 years agointel_gpu_dump: Add instdone register values and decoding.
Eric Anholt [Wed, 1 Jul 2009 00:52:43 +0000 (17:52 -0700)]
intel_gpu_dump: Add instdone register values and decoding.

15 years agointel_gpu_top: Move instdone bit definitions to lib to share with _dump.
Eric Anholt [Wed, 1 Jul 2009 00:13:27 +0000 (17:13 -0700)]
intel_gpu_top: Move instdone bit definitions to lib to share with _dump.

15 years agointel_gpu_dump: Use the correct IPEHR on pre-965.
Eric Anholt [Wed, 24 Jun 2009 02:35:19 +0000 (19:35 -0700)]
intel_gpu_dump: Use the correct IPEHR on pre-965.

15 years agointel_gpu_dump: Add details on URB_FENCE.
Eric Anholt [Wed, 24 Jun 2009 01:43:59 +0000 (18:43 -0700)]
intel_gpu_dump: Add details on URB_FENCE.

15 years agoOops, fix up printed hw offset since the HEAD pointer fixup
Eric Anholt [Wed, 24 Jun 2009 01:16:42 +0000 (18:16 -0700)]
Oops, fix up printed hw offset since the HEAD pointer fixup

15 years agoAdd more 965 INSTDONE bits.
Eric Anholt [Wed, 24 Jun 2009 00:10:44 +0000 (17:10 -0700)]
Add more 965 INSTDONE bits.

This shows off the units that are stuck busy in the ut2004 hang.

15 years agoCorrectly flag the acthd when it points within a packet.
Eric Anholt [Tue, 23 Jun 2009 23:22:41 +0000 (16:22 -0700)]
Correctly flag the acthd when it points within a packet.

15 years agoAdd the acthd to the GPU dump.
Eric Anholt [Tue, 23 Jun 2009 23:21:34 +0000 (16:21 -0700)]
Add the acthd to the GPU dump.

15 years agoAdd register writing tool
Ben Gamari [Thu, 11 Jun 2009 03:13:40 +0000 (23:13 -0400)]
Add register writing tool

Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agoFour new tests for error handling
Jesse Barnes [Fri, 19 Jun 2009 01:07:47 +0000 (18:07 -0700)]
Four new tests for error handling

Add four new tests for error the error handling cases:
  - gem_bad_address - store to a bad address, should generate a protection or
    page table error
  - gem_bad_batch - try to execute a bad batch, should generate a protection,
    invalid instruction or page table error
  - gem_bad_blit - blit to an invalid location, should generated a protection
    or page table error
  - gem_hang - hang the GPU on an event that will never happen, test hang
    detection & recovery code

15 years agoMake the precentage bars fit in 80 columns.
Eric Anholt [Tue, 9 Jun 2009 18:30:25 +0000 (11:30 -0700)]
Make the precentage bars fit in 80 columns.

Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agoA pedantic addition of a format string to a printf call
Robert Bragg [Fri, 17 Apr 2009 18:01:04 +0000 (19:01 +0100)]
A pedantic addition of a format string to a printf call

The printf used to clear the screen didn't have a format string, this
adds one to avoid a compiler warning.

Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agoAdds unicode bars for quickly visualizing the busy percentages
Robert Bragg [Fri, 17 Apr 2009 17:52:23 +0000 (18:52 +0100)]
Adds unicode bars for quickly visualizing the busy percentages

It can be a bit easier to digest the percentages with bar graphs than by
scanning continually changing numbers.

Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agoIncreases the sample frequency from 100/sec to 10,000/sec
Robert Bragg [Fri, 17 Apr 2009 15:25:52 +0000 (16:25 +0100)]
Increases the sample frequency from 100/sec to 10,000/sec

For a typical vsync locked application running at 60fps, sampling at just
under twice a frame doesn't seem to give very stable lists of relevent hardware
units because there are a number of units involved that may not be sampled one
second to the next.

This bumps the sample rate to 10,000 instead which is ~ 170 samples per
frame so we tend to hit all the units involved.

It also changes the report threshold to a sample count >= 1, so you don't
see as many units with a percentage of 0.

Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agoRevert "intel_gpu_top:print a visual clue on how big a certain use is"
Robert Bragg [Sun, 7 Jun 2009 22:26:18 +0000 (23:26 +0100)]
Revert "intel_gpu_top:print a visual clue on how big a certain use is"

This reverts commit 5e13b98da1e7e31cffba84fd257002357d5f2682.

We can print prettier graphs using unicode

Signed-off-by: Eric Anholt <eric@anholt.net>
15 years agointel_gpu_top: Print GPU clock information, which seems interesting.
Eric Anholt [Thu, 4 Jun 2009 14:38:17 +0000 (14:38 +0000)]
intel_gpu_top: Print GPU clock information, which seems interesting.

15 years agointel_gpu_top:print a visual clue on how big a certain use is
Arjan van de Ven [Tue, 28 Apr 2009 11:05:43 +0000 (12:05 +0100)]
intel_gpu_top:print a visual clue on how big a certain use is

humans are pretty bad at reading percentages quicky; this patch adds a
histogram capability to make it more visually clear as to which lines are
big ticket items

15 years agoBump version to 1.0.1 for release. 1.0.1
Eric Anholt [Tue, 19 May 2009 23:48:40 +0000 (16:48 -0700)]
Bump version to 1.0.1 for release.

15 years agointel_gpu_dump: Stop decoding batchbuffers after MI_BATCH_BUFFER_END.
Eric Anholt [Tue, 19 May 2009 23:32:39 +0000 (16:32 -0700)]
intel_gpu_dump: Stop decoding batchbuffers after MI_BATCH_BUFFER_END.

15 years agoAdd decode of gen4 PIPE_CONTROL
Eric Anholt [Tue, 19 May 2009 23:26:35 +0000 (16:26 -0700)]
Add decode of gen4 PIPE_CONTROL

15 years agoAdd decode of dest buffer variables (destination format)
Eric Anholt [Tue, 19 May 2009 23:12:43 +0000 (16:12 -0700)]
Add decode of dest buffer variables (destination format)

(cherry-pick from Mesa commit 0af7e9170fd7c0d906652378b9f78fe2ba9725ad)

15 years agoAdd manpages for all installed binaries.
Eric Anholt [Tue, 19 May 2009 23:09:08 +0000 (16:09 -0700)]
Add manpages for all installed binaries.

15 years agoAdd gem_largeobject to ignore.
Eric Anholt [Tue, 19 May 2009 22:53:55 +0000 (15:53 -0700)]
Add gem_largeobject to ignore.

15 years agotests: add large object test
Jesse Barnes [Thu, 14 May 2009 21:39:48 +0000 (14:39 -0700)]
tests: add large object test

The large object test simply tries to allocate a 128M object, pin it, then
pwrite the whole thing.  This should make obvious any leaks on close or
page pointer allocation failures.

15 years agoFix distcheck 1.0
Eric Anholt [Mon, 27 Apr 2009 16:45:22 +0000 (09:45 -0700)]
Fix distcheck

15 years ago.gitignore libtool junk
Eric Anholt [Mon, 27 Apr 2009 16:41:41 +0000 (09:41 -0700)]
.gitignore libtool junk

15 years agointel_gpu_top: Add ring information.
Eric Anholt [Mon, 27 Apr 2009 16:39:05 +0000 (09:39 -0700)]
intel_gpu_top: Add ring information.

15 years agoAdd a test to try to fill the ring.
Eric Anholt [Mon, 27 Apr 2009 16:35:40 +0000 (09:35 -0700)]
Add a test to try to fill the ring.

Did this while trying to debug the Ubuntu DRI1+compiz issue.  Didn't succeed
in filling the ring.

15 years agointel_gpu_dump: Use asprintf instead of a static-sized array
Carl Worth [Mon, 13 Apr 2009 21:35:11 +0000 (14:35 -0700)]
intel_gpu_dump: Use asprintf instead of a static-sized array

Not that we're really concerned about buffer-overrun attacks
against this utility, but still, good habits and clean code
are always preferred.

15 years agointel_gpu_dump: Give a nicer message for missing batchbuffer-dumping code
Carl Worth [Mon, 13 Apr 2009 21:33:09 +0000 (14:33 -0700)]
intel_gpu_dump: Give a nicer message for missing batchbuffer-dumping code

This is the case where debugfs is mounted, the dri/0 directory
exists but there's no i915_ringbuffer_info file. Point the user
toward upgrading the driver.

15 years agointel_gpu_dump: Mention filename in error message
Carl Worth [Mon, 13 Apr 2009 21:06:00 +0000 (14:06 -0700)]
intel_gpu_dump: Mention filename in error message

This helps for when the user makes a typo trying to pass a file
or directory name to intel_gpu_dump.

15 years agoAdd example command for mounting debugfs
Carl Worth [Mon, 13 Apr 2009 21:02:04 +0000 (14:02 -0700)]
Add example command for mounting debugfs

This is a nice opportunity to teach the user something, (and help our
users to actually succeed in creating useful bug reports for us).

15 years agointel_gpu_dump: Add no-arguments automatic mode, and clean up comments.
Eric Anholt [Mon, 13 Apr 2009 01:03:25 +0000 (18:03 -0700)]
intel_gpu_dump: Add no-arguments automatic mode, and clean up comments.

15 years agointel_gpu_dump: Dump ring, batch, and head/tail info when given the directory.
Eric Anholt [Mon, 13 Apr 2009 00:49:31 +0000 (17:49 -0700)]
intel_gpu_dump: Dump ring, batch, and head/tail info when given the directory.

15 years agointel_gpu_dump: Decode the GTT offset of batchbuffers.
Eric Anholt [Mon, 13 Apr 2009 00:04:31 +0000 (17:04 -0700)]
intel_gpu_dump: Decode the GTT offset of batchbuffers.

15 years agointel_gpu_dump: Get the devid so we can decode correctly on gen4.
Eric Anholt [Sun, 12 Apr 2009 22:57:07 +0000 (15:57 -0700)]
intel_gpu_dump: Get the devid so we can decode correctly on gen4.

15 years agoFix warning about assert in intel_batchbuffer.h
Eric Anholt [Wed, 8 Apr 2009 02:07:18 +0000 (19:07 -0700)]
Fix warning about assert in intel_batchbuffer.h

15 years agoAdd a test for reading tiled objects with pread.
Eric Anholt [Tue, 7 Apr 2009 23:21:16 +0000 (16:21 -0700)]
Add a test for reading tiled objects with pread.

15 years agoAdd a test for pread after blitting to an object.
Eric Anholt [Tue, 7 Apr 2009 23:18:11 +0000 (16:18 -0700)]
Add a test for pread after blitting to an object.

This caught a bug with get_user_pages usage in the kernel, which would
result in zeroes being read out of the object when faulting in a new page.

15 years agoAdd very preliminary intel_gpu_dump program
Carl Worth [Tue, 7 Apr 2009 23:22:46 +0000 (16:22 -0700)]
Add very preliminary intel_gpu_dump program

Right now it requires a filename on the command line, (such as
/debug/dri/0/i915_batchbuffers). We plan to make it smarter in
the future by going out and collecting all the data it needs.

15 years agoAdd Eric's batchbuffer-decode code
Carl Worth [Thu, 2 Apr 2009 18:53:38 +0000 (11:53 -0700)]
Add Eric's batchbuffer-decode code

This comes from the 3D driver within mesa, with identically named files.
In its new life here it will become a standalone program.

15 years agoAdd a regression test for tiled object blitting.
Eric Anholt [Fri, 3 Apr 2009 21:23:06 +0000 (14:23 -0700)]
Add a regression test for tiled object blitting.

15 years agoAdd a couple of other intel_upload_blit_large variants for comparison.
Eric Anholt [Tue, 31 Mar 2009 02:17:12 +0000 (19:17 -0700)]
Add a couple of other intel_upload_blit_large variants for comparison.

15 years agointel_upload_blit_*: Fix reporting and generate junk instead of using random.
Eric Anholt [Tue, 31 Mar 2009 02:09:00 +0000 (19:09 -0700)]
intel_upload_blit_*: Fix reporting and generate junk instead of using random.

Turns out that random completely overwhelmed the cost of the driver stuff
in _large.c.  I definitely want to generate data, but still be able to see
the driver's responsibility.

15 years agoAdd intel_chipset.h to fix dist.
Eric Anholt [Tue, 31 Mar 2009 02:08:43 +0000 (19:08 -0700)]
Add intel_chipset.h to fix dist.

15 years agoMove i810_reg.h to lib/intel_reg.h and get it disted.
Eric Anholt [Fri, 27 Mar 2009 23:11:50 +0000 (16:11 -0700)]
Move i810_reg.h to lib/intel_reg.h and get it disted.

15 years agotests/: warning fixes
Eric Anholt [Fri, 27 Mar 2009 23:04:25 +0000 (16:04 -0700)]
tests/: warning fixes

15 years agorename project
Eric Anholt [Fri, 27 Mar 2009 23:03:12 +0000 (16:03 -0700)]
rename project

15 years agoreadme fixups
Eric Anholt [Fri, 27 Mar 2009 23:03:02 +0000 (16:03 -0700)]
readme fixups

15 years agointel_gpu_top: Add more 965 bits.
Eric Anholt [Fri, 27 Mar 2009 20:13:47 +0000 (13:13 -0700)]
intel_gpu_top: Add more 965 bits.

15 years agoPort intel_idle from 2D driver as intel_gpu_top with a better interface.
Eric Anholt [Fri, 27 Mar 2009 19:25:09 +0000 (12:25 -0700)]
Port intel_idle from 2D driver as intel_gpu_top with a better interface.

15 years agointel_stepping: Add G45 and GM45.
Eric Anholt [Fri, 27 Mar 2009 18:07:33 +0000 (11:07 -0700)]
intel_stepping: Add G45 and GM45.

15 years agoAdd intel_stepping from the 2D driver.
Eric Anholt [Fri, 27 Mar 2009 17:59:09 +0000 (10:59 -0700)]
Add intel_stepping from the 2D driver.

15 years agoAdd the tests we want from libdrm.
Eric Anholt [Fri, 27 Mar 2009 01:47:42 +0000 (18:47 -0700)]
Add the tests we want from libdrm.

15 years agoFix project name/version for distfiles.
Eric Anholt [Fri, 27 Mar 2009 01:45:01 +0000 (18:45 -0700)]
Fix project name/version for distfiles.

15 years agoAdd notes about running things as root.
Eric Anholt [Fri, 27 Mar 2009 01:44:13 +0000 (18:44 -0700)]
Add notes about running things as root.

15 years agoInitial import of intel-graphics-tools with some microbenchmarks.
Eric Anholt [Fri, 27 Mar 2009 00:15:11 +0000 (17:15 -0700)]
Initial import of intel-graphics-tools with some microbenchmarks.