[Intel-gfx] [ANNOUNCE] intel-gpu-tools 1.8

Rodrigo Vivi rodrigo.vivi at gmail.com
Tue Sep 9 05:00:14 CEST 2014


Hi all,

Here is another release of Intel-gpu-tools 1.8.
This one came without any pre warning again and release was "time based" to
be on time for the Intel Linux Graphics stack validation that we are
starting today and that might be released by the end of this Quarter as
usual.

This is my last release since Thomas Wood is taking the role to maintain
Intel-gpu-tools. I'm confident that Thomas will do a great job and make the
releases more smooth and without my traditional rush.

1.8 NEWS:

- Added lib/igt.cocci semantic patch to catch often-seen patterns and
convert
  them to igt macros/infrastructure.

- Improvements to the documentation build systems (Thomas).

- Small fixes and improvements to the igt infrastructure and helpers all
over.

- As usual piles of new tests.

- Improved plane/pipe handling in the igt_kms library (Damien).

- Unified option parsing between simple tests and tests with subtests
(Thomas).
  This will allow us to merge the different Makefile targets once test
runners
  are converted.

- New commit functions for igt_kms to support the new universal planes
  interfaces (Matt Roper).

- Polish the debug output when test requirements aren't met a bit and
inject the
  program name/subtest in dmesg for easier backtrace/oom debugging (Chris).

- A bit of polish for the framebuffer helper functions (Damien).

- Robuster option parsing helpers, they now check for conflicts when merging
  different option lists (Thomas).

- MIPI DSI vbt support in intel_bios_read (Gaurav K Singh).

- Clarify the split between low-level helpers and the high-level library in
  igt_kms a bit by renaming some functions and improving and extending the
api
  documentation.

- Helper to restore the vt mode, useful to test lastclose/fbdev emulation
  behaviour (Thomas).

- Refactor the support for 64bit relocs. By specifying the number of
relocations
  explicit a lot of the gen8 checks can be removed from simple testcases
which
  only use the blitter (Chris).

The full list:

Arun Siluvery (1):
      igt/gem_workarounds: igt to test workaround registers

Ben Widawsky (1):
      lib: Add ndebug to the headers

Chris Wilson (60):
      API docs for intel_check_memory()
      tests: add gem_fence_upload
      igt/gem_mmap_gtt: Exercise concurrent pagefaulting
      igt/gem_fence_upload: Throw some threads into the mix
      igt/gem_fence_upload: Exercise contention on the fault handler
      igt/gem_exec_parse: Convert to igt_assert_eq
      igt/gem_exec_big: Update to new igt_assert_eq
      gem_fence_upload: Improve the contended messages
      gem_mmap_gtt: Test mmaping less than the full object
      Wait for any pid in order to reap failure quicker
      igt/gem_userptr_blits: Shared memory allocations
      igt/gem_userptr_blits: Verify that userptr bo work on unshared memory
      configure: Only build overlay if dri2proto is present
      igt/gem_tiled_swapping: Speed up slightly
      igt/gem_close_race: Trim object size
      overlay: Baytrail needs a custom GPU frequency parser
      igt/gem_close_race: Increase GPU load
      igt/gem_userptr_blits: Add missing RAM check before coherency tests
      igt/gem_userptr_blits: Add missing RAM check before coherency tests
      igt/kms_flip: Skip if headless
      Use SIGTERM in replace of SIGQUIT
      Quieten valgrind
      igt/gem_userptr_blits: New tests, old issues
      linux: Flush page/buffer cache before querying available memory
      linux: placate drop_caches and pass it a '\n'
      core: Only use signalsafe functions inside signal handlers
      igt_core: Refactor igt_stop_helper() to use igt_wait_helper()
      igt_core: Read manpages more carefully  for WNOHANG
      core: Inject subtest message into dmesg
      core: Improve phrasing for test requirements
      core: Inject program name into dmesg
      drmtest: First check that driver is i915
      igt/pm_rps: Fix assertion in load_helper_stop
      core: Add all kernel log levels
      igt/gem_cpu_reloc: Update the memory requirements
      igt: Prettify igt_assert_eq() failure messages
      igt/gem_concurrent_blit: Add GPU-vs-GPU checks
      igt/gem_concurrent_blit: Add ring race checks
      lib/batchbuffer: Store the gen in a local variable
      lib/batchbuffer: Guard intel_blt_copy with even more asserts
      igt: Tweak the igt_assert_cmpint macro
      igt/gem_concurrent_blit: Apply some fence pressure as well
      Prepare for 64bit relocation addresses
      igt/gem_seqno_wrap: Use debugfs singleton
      batch: Specify number of relocations to accommodate
      batch: Fix typo in BLIT_BATCH_BEGIN
      igt/gem_concurrent_blit: Use the preallocated dummy bo for temporary
copies
      rendercopy/gen8: Fix vertex buffer relocation
      igt/gem_gtt_hog: Convert opencoded fork()/wait() over to igt_fork()
      igt/gem_concurrent_blit: Move igt_require() out of children
      core: Display subtest walltime
      lib/igt_fb: Add debug output for creating fb
      lib/igt_fb: Ignore RGB888
      core: Prettify igt_require_f output
      aux: Squelch warning when emitting progress indicator
      igt/drv_missed_irq_hang: Increase busy load
      igt/kms_universal_plane: Only test existing planes
      igt/gem_mmap_gtt: Check coherency between GTT and CPU mmappings with
LLC
      igt/kms_flip_tiling: Use the associated pipe for the output
      igt/gem_cpu_reloc: Terminate the bad batch

Damien Lespiau (83):
      list-workarounds/chv: Add Cherryview to the list of valid platorms
      kms_universal_plane: Don't assert outside of fixtures/subtests
      kms_pipe_crc_basic: Split the main test function a bit more
      kms_pipe_crc_basic: Cycle between 2 differently colored buffer
      kms_pipe_crc_basic: Make the number of CRCs a parameter
      kms_pipe_crc_basic: Add a bit a debugging output
      lib: Reset 'position_changed' after a drmModeSetCrtc()
      pipe_crc: Warn if the CRC values is 0xffffffff
      pipe_crc: Make collect_crc() ensure the CRC looks somewhat valid
      kms_plane: Make the gathering of reference CRCs test agnostic
      kms_plane: Add a debug message when grabbing the CRC of a colored fb
      kms_plane: Specify the pipe when grabbing reference CRCs
      core: Put the requirement failure messages together
      core: Only display the errno message if errno is set
      core: Apply the same treatment to the in errno message in
__igt_fail_assert()
      drv_module_reload: Don't declare success when failing
      igt_kms: Factor out a generic get_property() out of
get_drm_plane_type()
      igt_kms: Provide a get_plane_property() shorthand
      igt_kms: Add a way to query of the plane supports rotation
      igt_kms: Add support for setting plane rotation
      igt_kms: Introduce a for_each_pipe() macro
      kms_rotation_crc: Make check_plane_type() static
      kms_rotation_crc: Make more functions static
      kms_rotation_crc: Align a few wrapped lines to the opening brace
      kms_rotation_crc: Update the copyright to have this year as well
      kms_rotation_crc: Test the validity of the output first
      kms_rotation_crc: Require universal planes for the testing primary
rotation
      kms_rotation_crc: Fix style issue: '{' at the end of lines
      kms_rotation_crc: Fix style issue: single statement conditionals
      kms_rotation_crc: Factor out the square drawing function
      kms_rotation_crc: Factor out common primary/sprite code in
prepare_crtc()
      kms_rotation_crc: Remove useless comments
      kms_rotation_crc: Use drm_plane from igt_plane_t
      kms_rotation_crc: Style issue: binary operators need spaces before
and after
      kms_rotation_crc: Skip the tests if rotation is not supported
      kms_rotation_crc: Just store the igt_plane_t in data
      kms_rotation_crc: Unify the two tests
      kms_rotation_crc: Always disable the plane in cleanup
      kms_rotation_crc: Don't store rotate in the test state
      kms_rotation_crc: Don't store 'pipe' in the state
      kms_rotation_crc: Use igt_plane_set_rotation()
      kms_rotation_crc: Remove now unnecessary defines
      kms_rotation_crc: Use for_each_pipe()
      kms_rotation_crc: Remove the test on output->valid
      kms_rotation_crc: Remove 'output' from the state
      kms_rotation_crc: Remove the sleep(2)
      kms_rotation_crc: Remove plane from the state
      kms_rotation_crc: No need to test for NULL before freeing the pipe
CRC object
      kms_rotation_crc: Allow the sprite test to run even without universal
planes
      kms_rotation_crc: Don't commit with no fb set up
      kms_rotation_crc: Properly paint the whole frame buffer
      kms_rotation_crc: Add the test to .gitignore
      kms_rotation_crc: Don't compile the test on Android with no cairo
support
      kms_rotation_crc: Document the two steps in prepare_crtc()
      kms_rotation_crc: Always use the primary plane to compute the
reference CRC
      kms_rotation_crc: Remove unnecessary includes
      kms_rotation_crc: Use the igt_kms enum to encode the plane rotation
      lib: Extract gem_get_tiling() from a couple of tests.
      tests: Move gem_render_copy_redux to the list of multi-subtests tests
      lib: Remove unused field from struct igt_fb
      lib: Split the GTT mapping out of get_cairo_surface()
      lib: NULLify ->cairo_surface once unmapped
      lib: Don't take a reference to the surface in get_cairo_surface()
      testdisplay: Destroy the cairo context once the fb is painted
      lib: Change the fb creation functions to take fully qualified tiling
formats
      kms_cursor_crc: Use I915_TILING_NONE to create fbs
      kms_fbc_crc: Use I915_TILING_X to create fbs
      kms_fence_pin_leak: Use I915_TILING_X to create fbs
      kms_flip: Adjust to the new igt_create_fb*() API
      kms_flip_tiling: Adjust to the new igt_create_.*fb() API
      kms_mmio_vs_cs_flip: Adjust to the new igt_create_.*fb() API
      kms_pipe_crc_basic: Adjust to the new igt_create.*fb() API
      kms_plane: Adjust to the new igt_create.*fb() API
      kms_psr_sink_crc: Adjust to the new igt_create.*fb() API
      kms_render: Adjust to the new igt_create.*fb() API
      kms_setmode: Adjust to the igt_create.*fb() API
      kms_sink_crc_basic: Adjust to the new igt_create.*fb() API
      pm_lpsp: Adjust to the new igt_create.*fb() API
      pm_rpm: Adjust to the new igt_create*fb() API
      testdisplay: Make the desired tiling mode an unsigned int
      lib/fb: Replace straight igt_fail() by asserts with debug messages
      lib/fb: Assert, instead of silently failing, when creating fbs
      gem_workarounds: intel_wa_registers is now prefixed with i915

Daniel Vetter (57):
      NEWS: post-release bump
      Revert "Add rc6_residency_counter subtest"
      lib/intel_os: Spelling fix
      tests/gem_fence_upload: Fix test wrapper
      tests/kms_psr_sink_crc: Skip properly
      tests/prime_self_import: Use igt_assert_cmpint
      lib/igt_core: Add igt_assert_eq
      lib: add igt.cocci
      tests: Run igt.cocci over tests
      Revert "tests: Run igt.cocci over tests"
      tests/pm_psr_sink_crc: Fix longjmp fun
      tests: Don't use stderr for informational messages
      lib/igt.cocci: Conversion to igt logging
      lib/igt.cocci: Convert abort() to igt_fail
      lib/igt.cocci: Also add rule to use igt_warn_on_f
      lib: Introduce igt_fail_on/_f
      NEWS: Mention igt.cocci
      tests: run igt.cocci
      Revert "Revert "Add rc6_residency_counter subtest""
      tests/Makefile.sources: Remove empty last line
      tests: Move gem_userptr_blits to the right Makefile target
      Paritially revert "lib: add missing includes to headers"
      tests/gem_mmap: Add short mmap test
      tests/gem_exec_big: Re-add gem_sync
      tests/kms_psr_sink_crc: Fix delayed gtt/cpu write tests
      tests/gem_exec_parse: use gem_uses_aliasing_ppgtt
      tests/gem_render_copy_redux needs legacy drm nodes
      drv_module_reload: Unbind the right console driver
      tests: Run igt.cocci
      NEWS: Updates
      lib/igt_core: Don't log when listing subtests
      lib/igt.cocci: Add stanza for for_each_pipe
      lib: Add argc/argv to igt_simple_init
      lib/igt_core: Fixup gtkdoc for logging functions
      tests/gem_userptr_blits: Polish
      lib/igt.cocci: Add s/assert/igt_assert/
      tests: Move root check to lib/drm_lib.sh
      lib/drm_lib.sh: Bare-bones long option parsing
      tests/pm_rpm: Convert to new optparsing
      NEWS: Updates
      NEWS: Don't forget new universal plane commit for igt_kms
      tests/gem_ringfill: add interruptible mode
      tests/kms_cursor_crc: Align opening {
      lib/igt_kms: Unify pipe name helpers
      lib/igt_kms: Clean up the other _name functions/macros
      lib/igt_kms: doc for type2str functions
      lib/igt_kms: set_vt_graphics_mode is a low-level helper
      lib/igt_kms: Shuffle kmtests_ functions
      lib/igt_kms: Simplify return value of
kmstest_get_connector_default_mode
      lib/igt_kms: Simplify return value of kmstest_get_connector_config
      lib/igt_kms: doc for the remaining kmstest_ functions
      lib/rendercopy*: Use igt_assert
      lib/intel_* Use igt checks and macros
      lib/igt_* Use igt macros in igt libaries
      lib: Use igt macros more
      NEWS: Updates in anticipation of the next release
      lib/igt_aux: Improve wait_for_keypress helper a bit

Gaurav K Singh (2):
      intel_bios_reader: Add support to dump MIPI Configuration Block #52
      intel_bios_reader: Add support to dump MIPI Sequence block #53

Jesse Barnes (1):
      gem_exec_parse: require PPGTT as well

Matt Roper (10):
      lib/kms: Drop igt_pipe->need_set_{crtc, cursor}
      lib/kms: Track need for vblank wait locally
      lib/kms: Add igt_display_commit2()
      lib/kms: Add igt_display_try_commit2()
      lib/kms: Add universal plane support (v2)
      lib/kms: Drop primary plane windowing test from
igt_plane_set_position()
      tests/kms_universal_plane: Universal plane testing (v6)
      tests/kms_cursor_crc: Combine data_t and test_data_t
      tests/kms_cursor_crc: Call drmModeMoveCursor() directly
      lib/kms: Provide universal plane #define's

Mika Kuoppala (6):
      lib/gen6_render: removed duplicate defines
      tools/null_state_gen: terminate relocs with -1
      lib/rendercopy: fix a typo in define
      lib/gen6_render: fix GEN6_3DSTATE_VF_STATISTICS define
      lib: Add MI_LOAD_REGISTER_IMM
      tools/null_state_gen: build cmd and state space separately

Mike Mason (4):
      Add kms_flip_event_leak to .gitignore
      scripts: Allow multiple -t and -x regular expressions for run-tests.sh
      tests: Fix seg fault when gem_mmap is run without specifying a subtest
      scripts: Add capability to resume interrupted run-tests.sh session

Paulo Zanoni (14):
      tests/pm_rpm: add dpms-mode-unset{,-non}-lpsp subtests
      tests/pm_rpm: add subtests for planes and cursors
      igt_kms: don't get drmModeRes just to free it later
      igt_kms: pass drm_fd instead of igt_display_t on some functions
      igt_kms: optionally return the property from get_property
      igt_kms: document and export kmstest_get_property()
      tests/pm_rpm: use kmstest_get_property()
      igt_kms: add kmstest_unset_all_crtcs()
      tests/pm_rpm: add subtests for fences
      tests/pm_rpm: add wait_for_suspended() calls to test_one_plane
      tests/pm_rpm: add macros to disable/enable screens and wait
      tests/pm_rpm: add disable_or_dpms_all_screens()
      tests/pm_rpm: refactor the modesetting code
      igt_core: zero exit_handler_count before forking

Rodrigo Vivi (2):
      tests/kms_psr_sink_crc: Create test to test PSR by checking panel CRC.
      bump version to 1.8 and add the release date

Rémi Cardona (1):
      shader-debugger: Force file/stdout IO as UTF-8

Sean V Kelley (1):
      lib/chv: CHV media pipeline command sequence

Sonika Jindal (2):
      tests/kms_rotation_crc: IGT for 180 degree HW rotation
      kms_rotation_crc: Calling commit2 instead of legacy commit

Thomas Wood (46):
      quick_dump: read the architecture files for items to add to EXTRA_DIST
      README: update piglit instructions
      lib: remove /** from comments that are not API documentation
      README: update the section on modifying and rebuilding documentation
      gitignore: add missing files and keep lists sorted
      lib: various documentation fixes
      docs: add private headers to IGNORE_HFILES
      docs: always rebuild the sections file
      docs: remove unused annotation glossary include
      lib: ensure igt_display_init clears the memory for the display struct
      lib: add missing includes to headers
      lib: add function to change connector states
      lib: add the ability to set an EDID data block on a connector
      lib: add igt_enable_connectors and igt_reset_connectors
      tests: enable extra connectors in kms_flip and kms_pipe_crc_basic
      lib: allow the edid to be reset
      scripts: add a script to help run tests with Piglit
      README: update the test run instructions
      core: increase the exit handler limit
      lib: warn when attempting to run an unknown subtest
      tests: remove unused getopt header includes
      lib: move option parsing into common_init
      lib: add igt_simple_init_parse_opts
      lib: don't ignore unknown options in multi-tests
      tests: convert simple tests to use igt_simple_init_parse_opts
      lib: always warn about unknown options
      lib: add a command line option to enable debug output in tests
      lib: add an exit code to use when invalid options are specified
      build: improve version.h generation
      testdisplay: only set terminal attributes when in foreground process
group
      quick_dump: fix distcheck
      testdisplay: set a non-zero exit code if getopt detected an error
      lib: avoid getopt value conflicts with tests
      lib: check test options for conflicts
      lib: don't abort if forcing the connector state fails
      Update .gitignore files
      testdisplay: ignore invalid subtest options
      lib: add igt_restore_vt_mode
      tests: check plane rotation is reset after the VT mode is restored
      lib: add kmstest_edid_add_3d
      lib: move create_stereo_fb from testdisplay to igt_fb
      tests: add kms_3d test
      lib/igt_fb: ensure igt_create_fb parameters are consistent
      lib: don't force HDMI or DP connectors on HSW and BDW
      tests/kms_3d: skip if connectors cannot be forced
      lib: keep a list of modified connectors to reset

Tim Gore (7):
      intel-gpu-tools: remove the --cmd option from gem_seqno_wrap
      intel-gpu-tools: add igt_core init func calls to some tests
      intel-gpu-tools: Dont build kms_univeral_plane if no cairo
      intel-gpu-tools: Dont build kms_force_connector if no cairo
      intel-gpu-tools: add sys/wait.h to pm_rps.c
      intel-gpu-tools: fix version.h creation in android
      intel-gpu-tools: skip kms_flip_event_leak unless we have cairo

Tvrtko Ursulin (4):
      tests/gem_userptr_blits: Race between object creation and
multi-threaded mm ops
      igt/gem_userptr_benchmark: Fix for upstream ioctl number
      igt/gem_userptr_blits: Fix forked access test
      igt/gem_userptr_blits: Fix multi-threaded mm stress tester

Ville Syrjälä (7):
      tests/kms_fbc_crc: Update blit code for BDW
      lib/igt_debugfs: Don't fail if debugfs is already mounted
      lib/igt_debufs: Add IGT_NO_FORCEWAKE environment variable
      tools: Add intel_iosf_sb_{read,write} tools
      tools/intel_display_poller: Add a new tool that will poll various
display registers
      tests/gem_ctx_exec: Add reset-pin-leak subtest
      tests: Add kms_flip_event_leak test

Wendy Wang (2):
      Fixed the review issues for pm_rc6_residency IGT case
      Merge two subtests for pm_rc6_residency IGT case

Yi Sun (1):
      kms_plane: Add panning test for primary plane

git tag: intel-gpu-tools-1.8

http://xorg.freedesktop.org/archive/individual/app/intel-gpu-tools-1.8.tar.bz2
MD5:  49d2c3c65204d889189c4d8c14c598b3  intel-gpu-tools-1.8.tar.bz2
SHA1: bc727f19263c167390bc8b018da52d05e4f17fd1  intel-gpu-tools-1.8.tar.bz2
SHA256: ffe2a11bca57f7fe36e93d55e4b3685127640e3e6cdae19973b193fe25ff3759
intel-gpu-tools-1.8.tar.bz2

http://xorg.freedesktop.org/archive/individual/app/intel-gpu-tools-1.8.tar.gz
MD5:  78eee01f2f5a53b03d8d88613e07c471  intel-gpu-tools-1.8.tar.gz
SHA1: 7f286412de6d8ff7927c4bcb66aee19ca32a4476  intel-gpu-tools-1.8.tar.gz
SHA256: 1330098ff84176768530ce60df561c681a51b56f0f2b76fdeb055e73814745da
intel-gpu-tools-1.8.tar.gz

Thanks,
Rodrigo.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/intel-gfx/attachments/20140908/cc76079a/attachment.html>


More information about the Intel-gfx mailing list