Hi all,
Here goes a new i-g-t quarterly release with the following changes:
- Changes to support Broadwell in the test suite. (Ben, Damien, et al.)
- Updated (now working again!) Android support from Oscar Mateo.
- Test coverage through i-g-t is now officially an integral part of any drm/i915
feature work and also for bugfixes. For more details see:
http://blog.ffwll.ch/2013/11/testing-requirements-for-drmi915.html
- Fix the gen4asm build depency handling, now that the assembler is included in
i-g-t (Ben).
- Improve the cairo object lifetime management of the kmstest helpers (Chris).
- Allow register access to succeed if i915 is loaded but debugfs isn't found
(i.e. nomodeset is used)
- New kernel modesetting helper library for simpler testcases (Damien).
- New structure logging support for tests. Message at the WARN level be
reflected in the piglit result, also included igt_warn_on macros to simplify
test asserts which should just result in warnings, but not in a test abort.
- Broadwell support for intel_audio_dump (Mengdong Lin).
- API documentation support (Thomas)
Complete list:
Ben Widawsky (52):
chipset: IS_I9XX macro
pciid/bdw: Add Broadwell PCI ids
bdw: Update obvious missing blit support
tests/bdw: gem_exec_faulting_reloc
tests/bdw: gem_cpu_reloc
tests/bdw: gem_evict_*
tests/bdw: gem_exec_blt
tests/bdw: gem_pin
tests/bdw: gem_linear_blits
tests/bdw: pwrite_pread
tests/bdw: Port storedw_loop_vebox to gen8
assembler/bdw: Small cleanup
quick_dump/bdw: Just basic stuff for now
NEWS: Broadwell
NEWS: Broadwell will be in 1.6, not 1.5
NEWS: Restore the 1.4 release info
gem_storedw_batches_loop: Update for gen8
gem_evict_everything: Fix broken blit generalization
gem_evict_alignment: Skip on GEN8 for now
gem_cpu_reloc: Fixed botched BLIT conversion
gem_linear_blits: Fix blit command length
gem_storedw_batches_loop: Fix for real this time
gem_evict_*: Fix batch lengths, broken on gen8 integration
gem_cs_prefetch: more gen8 integration fallout
gem_pipe_control_store_loop: BDW update
drm_vma_limiter_cached: Missed gen8 BLT updates
gem_write_read_ring_switch: Missed gen8 BLT updates
gem_persistent_relocs: Update for gen8 BLT cmds
gem_reloc_vs_gpu: Update for gen8 BLT cmds
Kill XY_COLOR_BLT_CMD
gen8 rendercpy: temporarily disable
gem_exec_faulting_reloc: Update gen8 BLT cmds
intel-gpu-tools: Version information
shader-debugger: Use in tree gen4asm
gem_exec_blit: gen8 updates for blit
configure.ac: Remove GEN4ASM define
configure.ac: Fixup conditional shader debugger
configure.ac: Fix shader debugger for real
gem_storedw_batches_loop: Fix for BDW
.gitignore: Add pycache
quick_dump: Fix the indentation
quick_dump: Allow comments in the register files
quick_dump/hsw: Watermark debug
quick_dump/broadwell: Use haswell_other
quick_dump/watermarks: lp3, misc and debug
gem_storedw_batches_loop: Fix for pre-BDW
quick_dump: Refactor to allow future code
quick_dump: Add a file argument
quick_dump/gen7: Add other GFX_MODE registers
quick_dump/haswell: Add gen7 stuff + VECS_MODE
quick_dump/broadwell: Add gen7 ring mode registers
quick_dump: Add missed broadwell autodetect
Chon Ming Lee (2):
tools/quick_dump: Update Baytrail DPIO names to reflect to the
driver change.
Update dpio read/write to take in extra PHY parameter.
Chris Wilson (26):
tests/pm_pc8: Fix invalid pointer-to-integer casting
drmtest: Fix weird signage on errno
kmstest: Fix up lifetimes of cairo objects
tests: Add gem_gtt_hog
gem_gtt_hog: Print elapsed time in ms
tests/gem_pipe_control_store_loop: Delete bogus assertion
gem_gtt_hog: use a normal exit()
kms_flip: Try to consume all fences after a flip
intel_error_decode: Protect against missing ring registers
igt/gem_concurrent_blit: Scale resource usage to RAM correctly
igt/gem_close_race: Fix blit commands for BDW
lib: Capture errno on entry
tests: Add gem_madvise
tests: Add gen7_forcewake_mt
Add i915_pciids.h
kms_flip: Try to make hang_gpu() robust against hanging the GPU
tests: Compile all threaded tests with -pthread when appropriate
kms_flip: Assert that hang_gpu() should only fail is the GPU is
already hung
kms_flip: Skip tests that try to inject GPU hangs if already
terminally wedged
gem_render_linear_blits: Trim usage to fit within RAM
gem_render_tiled_blits: Trim usage to fit within RAM
lib: Test against available swap
evictions: Limit the number of minor eviction surfaces to fit in RAM
igt/gem_pwrite_pread: Fix XY_SRC_COPY command length for gen8+
igt/gem_pwrite_pread: Fix relocation offsets for gen8+
igt/gem_pwrite_pread: Fix relocation offsets for gen8+
Damien Lespiau (68):
bdw: Add gen8 to the IS_9XX() macro
bdw: Add gen8 to intel_gen()
bdw: Add gen8 specific instdone bits
tests/bdw: Port storedw_loop_render to bdw
assembler/bdw: Add gen8_instruction from mesa
assembler/bdw: Support some basic gen8 intructions
assembler/bdw: Make the validation functions take a
brw_program_instruction
assembler/bdw: Add the start of a gen8 disassembler
assembler/bdw: Preliminary gen8 send & msgtarget support
assembler/bdw: Use gen8_set_exec_size() to set the execution size
rendercopy/bdw: Add gen8_render.h to the file list
rendercopy/bdw: Fix the original implementation
rendercopy/bdw: Add AUB annotations for states
list-workarounds/bdw: Add Broadwell to the list of valid platorms
gitignore: Ignore the newly introduced pm_psr
gitignore: Ignore gem_reset_stats
intel_dump_decode: Actually parse the -d option
intel_dump_decode: Support the INTEL_DEVID_OVERRIDE env variable
rendercopy/bdw: Fix the STATE_SIP instruction length
rendercopy/bdw: Fix the various 3DSTATE_CONSTANT_* instruction length
rendercopy/bdw: Fix the 3DSTATE_HIER_DEPTH_BUFFER instruction length
rendercopy/bdw: Fix the 3DSTATE_STENCIL_BUFFER instruction length
kms_fbc_crc: Remove unused variable
gitignore: Ignore kms_fbc_crc
gitignore: Ignore gem_media_fill
lib: Include the necessary stdint.h in gen6_render.h
intel_error_decode: Factor out common decoding code
intel_reg: Renamed INST_DONE to INSTDONE
instdone: Add an assert to make sure we never overflow instdone_bits
lib: Move the INSTDONE bit definitions to instdone.c
lib: Use INSTDONE_I965 and INSTDONE_1 for gen6
tools: Removed unused tools/intel_iosf_read.c
tools: Remove intel_disable_clock_gating
gem_ring_sync_copy: Add a ring to ring synchronization test
lib: Introduce a modeset API
kms_pipe_crc_basic: Port to the new modeset API
lib: Introduce symbolic names for display planes
lib: Introduce a for_each_connected_output() macro
kms_pipe_crc_basic: Use for_each_connected_output()
lib/display: Add an accessor to retrieve the number of pipes
kms_pipe_crc_basic: Use igt_display_get_n_pipes() instead of hardcoding 3
tests/kms_cursor_crc: Move the array of cursor fb_ids to the stack
lib/display: Fix the SetCrtc disabling log message
lib/display: Add support for the cursor plane
lib/display: Add a way to specify we don't care about the pipe to use
kms_cursor_crc: Port the test to the new modeset API
lib/display: Allow to override the display verbosity with an env variale
lib/display: Check if we're trying to use the same pipe on 2 outputs
lib/crc: Factor out reading a single CRC value
lib/crc: Add a helper to read a single CRC value
tests/kms_cursor_crc: Use igt_pipe_crc_collect_crc()
lib/display: Add support for DRM planes
lib: Introduce a new helper kmstest_create_color_fb()
kms_cursor_crc: Use kmstest_create_color_fb()
lib: Include drm_fourcc.h from igt_kms.h
kms_pipe_crc_basic: Use kmstest_create_color_fb()
lib: Add a helper to wait for a keypress
lib/display: Add a way to wait at every commit for inspection
lib/display: Print the fb id, not its pointer in the set_fb() log message
lib/display: Wait for a vblank after SetPlane()
lib/display: Fix assertion in set_plane()
kms_plane: Start a basic display plane test
lib: Don't wait in igt_wait_for_keypress() if stdin is not a tty
lib: Fix igt_ouput_get_plane() typo (output)
kms_pipe_crc_basic: Remove igt_display_get_n_pipes() usage in main
lib/display: Fix typo in igt_plane_get_fd_id() (fb)
kms_plane: Remove spurious inclusion of glib.h
kms_pipe_crc_basic: Remove spurious inclusion of glib.h
Daniel Vetter (91):
NEWS: Remove spurious -
tests/gem_reloc_overflow: Polish after Rafael's patch
tests/gem_overflow_reloc: Rework subtest enumeration
Revert "tests/gem_ctx_bad_exec: Check the errno, too"
NEWS: Android.mk updates!
NEWS: Make testing requirements official
lib: adjust oom_score
tests/gem_cs_prefetch: Fix bdw damage
tests: More bdw fallout
tests: More bdw fallout
tests/gem_reset_stats: Use macros a bit more
tools/reg_dumper: Add FW_BLC regs
tests/gem_reset_stats: More macros!
tests: Document per-ring naming conventions
tests: Document exec naming convention for execbuf related tests
lib: Don't put functional code into asserts
test/gem_concurrent_blt
Revert "test/gem_concurrent_blt"
tests: rename get_client_auth to have a core_ prefix
tests/core_get_client_auth: new subtests for always authenticated
lib: add gem_available_aperture_size helper
tests/gem_ctx_exec: Add eviction test
tests/gem_reset_stats: document the new subtest a bit better
tests: drm_open_any doesn't fail
lib: add igt_pipe_crc_check
lib: make igt_pipe_crc_start never fail
lib: Always print errno in igt assert/require macros
lib: Add host platform to version string
lib: print uname information in the version string
lib: igt_simple_main/init for subtest-less tests
tests: gem_close_race has subtests!
tests: roll out igt_simple_init/igt_simple_main
ignore version.h
tests: Move gem_reset_stats to the right place
tests: Document the Makefile variables a bit better
README: Document how to avoid Waffle with piglit
README: Fix spelling fail
NEWS updates
tests/NAMING-CONVENTION: s/snooped/snoop/
tests/gem_ctx_exec: fix ram requirement fumble
tests: fixup igt_simple_main fallout
tests/gem_largeobject: Prefault instead of pin
tests/prime_nv_pcopy: disable swizzle tests
tests/primve_nv_pcopy: Comment out unused check_swizzle function
tests/gem_gtt_hog: Use igt_simple_main
tools: Install them all
tests: Move pm_rps to the right Makefile target
tests: Add gem_fd_exhaustion
lib: extract igt_drop_root
debugger: avoid global variable shadowing
tests/pm_rps: Add missing static
lib: add __gem_create
tests/gem_fd_exhaustion: Make it work
tests/pm_rps: rename basic test to "basic-api"
fix make distcheck
tests/pm_rps: Fix compilation on Linux
tests/pm_rps: Round requested freq correctly
tests/pm_rps: remove setfreq
tests/gem_reset_stats: Correctly wrap global test code
lib/drmtest: Fix up skip_henceforth checks
lib: fix header include
tests/gem_close_race: Properly wrap init code in fixtures
tests/gem_bad_ctx_exec: Properly wrap in igt_fixtures
tests/gem_madvise: use correct main block
test/gem_gtt_hog: bail out earlier
lib: install exit handler only on success for prefault control
lib: Make igt_skip noreturn
NEWS: Mention Damien's new helper library
lib: (somewhat) structured logging support
test/pm_rps: use igt logging
tests/gem_seqno_wrap: use igt logging
tests/kms_flip: use igt logging
NEWS: Mention new logging support
testdisplay: implement dpms test
testdisplay: Allow to specify the dpms level
test/kms_flip: fix up igt logging conversion
configure: Bump libdrm requirements
NEWS: intel_audio_dump has now bdw support
tests/drv_missed_irq_hang: Remove debug output
tests/gem_exec_blt: More reasonable runtime
tests/drv_missed_irq_hand: Make it robuster
gtk-doc: Document the docs
lib: extract ioctl_wrappers.c
lib/ioctl_wrappers: api doc
lib: extract igt_core.c
lib/igt_core: api documentation
lib: move prefault helpers to igt_debugfs.c
lib/ioctl_wrappers: api doc fixup
lib: make igt_debugfs_open infallible
lib/igt_debugfs: drop drm_fd argument from igt_pipe_crc_new
lib/igt_debugfs: api docs
Devon Davies (1):
build: Added -UNDEBUG to tools/Android.mk
Imre Deak (5):
kms_flip: fix seq_step for 2xcrtc subtests
kms_flip: keep drm_fd open for the dpms restore exit handler
kms_flip: consolidate run_test_on_{crtc,crtc_pair}
lib/drmtest: fix hang in quiescent_gpu_at_exit due to
signal-unsafe asprintf
kms_setmode: Add clone-single-crtc subtest
Jeff McGee (18):
pm_rps: Use unbuffered I/O on sysfs files
pm_rps: Assert that valid sysfs writes return success
pm_rps: Fix test to target original min and max
pm_rps: Use igt exit handler for restore
pm_rps: Add read back checking on sysfs writes
pm_rps: Convert to subtest structure
pm_rps: Fix verbose option and streamline its use
pm_rps: Expand on min and max config testing
pm_rps: Remove repeat sysfs reads
pm_rps: Make frequency logging more compact
pm_rps: New subtest min-max-config-idle
lib/drmtest: Add igt_wait_helper
pm_rps: Add load helper for loading gpu
pm_rps: New subtest min-max-config-loaded
pm_rps: Add stop rings injection utility
pm_rps: Load helper should stall for last write
pm_rps: Add variable load support to load helper
pm_rps: New subtest for gpu reset
Jesse Barnes (8):
rendercopy/bdw: Initial gen8 rendercopy
mmio: allow register init even if debugfs isn't available and
i915 isn't loaded
quick_dump: more vlv dpio regs
NEWS: update with nomodeset register access change
mmio: use intel_iosf.c for DPIO reads and writes
mmio: drop unused i915_loaded fn
dpio: make dpio reg read/write output like normal reg read/write output
iosf: make iosf_rw take a larger addr
Joao Santos (1):
build: Skip kms_plane on Android
Kenneth Graunke (2):
rendercopy/bdw: Emit 3DSTATE_WM_HZ_OP.
rendercopy/bdw: Fix length of a bunch of zeroed out packets.
Mengdong Lin (15):
intel_audio_dump/bdw: add support for Broadwell
intel_audio_dump/bdw: dump debug registers for audio immediate commands
intel_audio_dump/bdw: dump audio chicken bit register
intel_audio_dump/bdw: dump audio DP and DIP FIFO debug status
intel_audio_dump/bdw: dump audio M CTS readback register
quick_dump/bdw: dump audio configuration registers for Haswell
and Broadwell
quick_dump/bdw: dump audio debug registers for Haswell and Broadwell
quick_dump/bdw: support Broadwell in device auto-detection
intel_audio_dump: support using base address plus an offset to
dump registers
intel_audio_dump: move definitions of
transcoder/pipe/port/converter earlier
intel_audio_dump: define IS_HASWELL_PLUS to cover Haswell and
its successors
intel_audio_dump: define get_num_pipes() to get number of pipes
for a device ID
intel_audio_dump: share common audio dump code for Ironlake,
Haswell & Broadwell
intel_audio_dump: adjust code alignment
intel_audio_dump: add support for Valleyview
Mika Kuoppala (10):
tests: add gem_reset_stats
tests/gem_reset_stats: add support for BDW+
tests/gem_reset_stats: stop rings after injecting hang
tests/gem_reset_stats: check non root access to reset_stats
lib/drmtest: Fix filedesc leak on gem_quiescent_gpu
tests/gem_reset_stats: add close-pending-fork
tests/gem_reset_stats: add reverse order in close-pending-fork
tests/gem_reset_stats: split ctx tests
tests/gem_reset_stats: run tests on all rings
tests/gem_reset_stats: stop only one ring when submitting hang
Oscar Mateo (14):
tests/drv_suspend: Replace "&> /dev/null" with ">
/dev/null 2>&1"
build: list all test/tool/lib source files in their own Makefile.sources
build: Fix missing "\" in tests/Makefile.sources
build: Move logic to tests/Makefile.am and away from Makefile.sources
lib: Move kms stuff from drmtest.c over to igt_kms.c
rendercopy: Remove rendercopy.c
tests/drm_get_client_auth: In Android, use gettid() instead of
syscall(SYS_gettid)
tests/gem_vmap_blits: Finish extracting gem_read
build: Fix assorted compilation warnings
lib/drmtest: Get the correct basename() under Android
lib/igt_display: workaround a name conflict in Android
build: New Android makefiles
tests/gem_close_race: Adapt the test for Full PPGTT
build: Create version and config information files in Android
Paulo Zanoni (40):
tests/pm_pc8: move teardown_environment next to setup_environment
tests/pm_pc8: return void on test_i2c
tests/pm_pc8: extract i2c_subtest_check_environment
tests/pm_pc8: add modeset-lpsp-stress
tests/pm_pc8: add debugfs/sysfs tests
tests/pm_pc8: remove "batch" and "stress-test"
subtests
tests/pm_pc8: add some GEM tests
tests/pm_pc8: reorganize main function
tests/pm_pc8: add support for runtime PM
tests/pm_pc8: check for PC8 status, not residency in most cases
tests/pm_pc8: fix debugfs-forcewake-user for PC8-only cases
tests/pm_pc8: rename some variables
tests/pm_pc8: remove compare_registers subtest
tests/pm_pc8: add some more stress tests
tests/pm_pc8: add --quick option
tests/pm_pc8: rename some subtests
tests/pm_pc8: try harder to reach PC8+ residencies
tests/pm_pc8: try to reuse FBs when possible
tests/pm_pc8: add "resuming" state
lib: rename some power well bit names
tests: add pm_lpsp
tests/kms_setmode: avoid 2 connectors on the same encoder
tests/pm_pc8: check if we can't become DRM master
drmtest: don't leak memory when parsing the arguments
tests/kms_flip: don't leak the connector when setting DPMS
tests/kms_flip: don't leak the CRTC
tests/kms_flip: don't leak the connector_config struct
tests/kms_flip: free the test_output struct when counting modes
tests/gen7_forcewake_mt: use -lpthread
tests/pm_pc8: remove sleep() call when setting up the environment
tests/pm_pc8: add gem-idle subtest
tests/pm_pc8: try to modprobe i2c-dev
tests/pm_pc8: try to modprobe msr
tests/pm_pc8: add reg-read-ioctl subtest
tests/pm_pc8: add pci-d3-state subtest
tests/pm_pc8: add "stay" mode
tests/pm_pc8: use XY_COLOR_BLT instead of just COLOR_BLT
tests/pm_pc8: BDW should also support PC8 residencies
tests/pm_pc8: use intel_gen instead of a big IS_GEN check
tests/pm_pc8: be more verbose on test_i2c failures
Rafael Barbalho (6):
tests/gem_reloc_overflow: Add gen8+ specifc tests
android: Handle the case when android doesn't have mmap64
android: Clean-up common makefile directives
android: Add makefile for the lib directory
android: Add the generation of vesion.h & config.h to lib
android: Change tests & tools directory to use the lib directory
Robert Beckett (1):
igt: tests/Android.mk: fix LOCAL_PATH usage
Rodrigo Vivi (1):
bump version to 1.6 and add the release date
Thomas Wood (10):
assembler: define YY_NO_INPUT to prevent unused symbol warnings
lib: fix signed/unsigned comparison issues
debugger: fix the -p option
assembler: fix condition for printing a warning
tests/gem_seqno_wrap: fix over allocation of arrays
tests: fix sub-test enumeration
Add API documentation support
drmtest: reformat the API documentation
lib: Add missing header to Makefile.sources
tests: add missing assertion
Tvrtko Ursulin (8):
build: Make sure asserts are enabled for tests
build: Skip kms_fbc_crc on Android
drmtest: Avoid wrong PID/TID after clone races
Revert "lib/drmtest: ducttape over fork race"
tests/gem_media_fill: Remove unnecessary include
tools: Allow building on Android after noinst_PROGRAMS is not
defined any more
tests/gem_evict_everything: Factor out eviction logic
tests/eviction_common: Avoid submitting duplicate objects
Ville Syrj?l? (11):
lib: Fix pipe_crc->ctl_fd assert
kms_cursor_crc: Fix crc readout after a subtest failure
rendercopy: Pass context to rendercopy functions
lib: Add gem_sw_finish()
lib: Add gem_bo_busy
kms_fbc_crc: Add a CRC based FBC test
quick_dump: Restore execture permission on quick_dump.py
tests: Drop glib.h include from kms_cursor_crc and kms_fbc_crc
kms_flip: Restore rings to running state in unhang_gpu()
kms_flip: Fail the subtest if page flip hang recovery wasn't
actually tested
kms_flip: Add dpms-vs-vblank-race and modeset-vs-vblank-race subtests
Xiang, Haihao (16):
assembler/bdw: Set math function
assembler/bdw: Disable mask control for advanced mode
assembler/bdw: Set jip/uip offsets used by flow control instructions
assembler/bdw: Set thread switch for multiple branch instructions
assembler/bdw: Data port message
assembler/bdw: Thread Spawn message
assembler/bdw: Video Motion Estimation(VME) message
assembler/bdw: Check & Refinement Engine message
assembler/bdw: SEND instruction
tests: add gem_media_fill
tests/gem_media_fill: add support for gen8
tests/gem_media_fill: add support for gen7
tests/gem_media_fill: the assembly code for the shader used in the case
lib: Clean the batch buffer store after reset
rendercopy/bdw: Set Instruction Buffer size Modify Enable to 1
assembler/bdw: Update write(...)
Zhao Yakui (5):
tests/bdw: Port storedw_loop_bsd to gen8
tests/bdw: Port storedw_loop_blt to gen8
assembler/bdw: Add the support of align1 register-indirect
addressing mode on Gen8
assembler/bdw: Add the DATA_PORT_CACHE1 shared function for Gen8+
Assembler/bdw: Remove the unsupported cache agent for WRITE(...)
git tag: intel-gpu-tools-1.6
http://xorg.freedesktop.org/archive/individual/app/intel-gpu-tools-1.6.tar.bz2
MD5: e823be446e06d31195c66e354752fb0b intel-gpu-tools-1.6.tar.bz2
SHA1: 87a066862f6cd4090b7bbb6a4ebc59e7aee330cd intel-gpu-tools-1.6.tar.bz2
SHA256: 6b29e04fd08c3b66639969c6e128db979d729f891017932e30034037df106cc6
intel-gpu-tools-1.6.tar.bz2
http://xorg.freedesktop.org/archive/individual/app/intel-gpu-tools-1.6.tar.gz
MD5: ae257d0231eea0992afe55be2db40d88 intel-gpu-tools-1.6.tar.gz
SHA1: 54b49a28144072a43d94df48f407e3d08e74bba6 intel-gpu-tools-1.6.tar.gz
SHA256: 9aff9c59ba315c500e0329d1421bb2892d5749797444c2661d6ef290678b63ac
intel-gpu-tools-1.6.tar.gz
--
Rodrigo Vivi
Blog: http://blog.vivi.eng.br