2f80854644
sensord: rewrite in Python ( #35353 )
...
* py sensord
* fix up mmc
* temp
* port over accel
* lil more
* kinda works
* rm that
* gpiochip
* mostly there
* lil more
* lil more
* irq timestamps
* fix ts
* fix double deg2rad
* test passes
* fix up mypy
* rm one more
* exception
* lint:
* read in all events
* bump that
* get under budget:
* accel self test
* gyro self-test
* keep these readable
* give it more cores
* debug
* valid
* rewrite that
---------
Co-authored-by: Comma Device <device@comma.ai >
2025-05-30 13:31:07 -07:00
67486ff92d
bump tinygrad ( #35208 )
...
* bump tinygrad
* fix
* why is mac different?
* fix sim
* relax that
2025-05-13 16:59:35 -07:00
f237649a7a
Reapply "Online lateral lag learning" ( #34975 )
...
* Online lateral lag learning (#34974 )
This reverts commit b4cc9e68d1
2025-04-09 22:31:07 +00:00
049193b178
test_onroad: test liveParameters service timing ( #34977 )
...
Add live params to test onroad
2025-04-04 03:31:55 +00:00
2382f1cab0
Run DM model twice, prep for bigger models ( #34894 )
...
* twice
* swap
* adjust
2025-03-24 15:54:42 -07:00
ffd54e41ee
bump manager starting time after 519ea9d
2025-03-17 15:19:34 -07:00
3f6fd3da78
test_onroad threshold tuning
2025-03-11 16:03:00 -07:00
65bf2ec0ea
test_onroad: add sanity check on ecode outputs ( #34784 )
...
* test_onroad: add sanity check on ecode outputs
* exact
2025-03-05 15:58:03 -08:00
821148150b
Revert "make test_onroad faster ( #34704 ) ( #34783 )
...
* Revert "make test_onroad faster (#34704 )"
This reverts commit 470ec46830
2025-03-05 15:42:22 -08:00
470ec46830
make test_onroad faster ( #34704 )
...
* revert that
* again
* port over
* clean
* round
2025-02-25 10:32:35 -08:00
952def7abf
start logging services first ( #34646 )
...
* start logging services first
* fix test
---------
Co-authored-by: Comma Device <device@comma.ai >
2025-02-20 16:53:07 -08:00
d9a8fab3e6
test_onroad: run same checks on both EncodeIdx and CameraState ( #34638 )
...
* try this
* cleanup
* lil more
---------
Co-authored-by: Comma Device <device@comma.ai >
2025-02-20 15:46:19 -08:00
52b66a1c0c
Stricter cam sync tests ( #34609 )
...
* strict
* fix that
---------
Co-authored-by: Comma Device <device@comma.ai >
2025-02-20 14:15:38 -08:00
a30d3f7d72
test_onroad: cleanup with time series LogReader ( #34613 )
2025-02-17 21:26:21 -08:00
73266a016c
test_onroad: always show model exec timings ( #34606 )
...
Co-authored-by: Comma Device <device@comma.ai >
2025-02-17 09:18:41 -08:00
958c8d1ce3
Beautiful camera timings ( #34558 )
...
* the start of something beautiful
* mv that
* cleanup
* cleanup output
* the good stuff
---------
Co-authored-by: Comma Device <device@comma.ai >
2025-02-14 21:33:50 -08:00
fbf918b311
test_onroad fixups ( #34572 )
...
* test_onroad fixpus
* fix
2025-02-11 20:15:30 -08:00
9ec54f59c6
loggerd: switch logging from raw file format to zstd compressed files ( #34549 )
...
* switch logging from raw file format to zstd compressed files
* more zst suffix
* compress bootlog
* remove class RawFile
* Optimize ZstdFileWriter by adding input caching
* use ZSTD_compressStream2
* cleanup
* LOG_COMPRESSION_LEVEL=10
* space
* add zst suffix to LOGS_SIZE_RATE
2025-02-08 12:55:31 -08:00
9bc35c0919
Revert "athena upload: reduce memory usage and improve efficiency with streaming ( #34528 )"
...
This reverts commit 4c65f51a55
2025-02-06 11:50:04 -08:00
4c65f51a55
athena upload: reduce memory usage and improve efficiency with streaming ( #34528 )
...
* reduce memory usage and improve upload efficiency with streaming
* space
2025-02-04 13:10:13 -08:00
89204b5464
test_onroad: adjust rlog size ( #34336 )
...
s
2025-01-07 19:45:53 -08:00
17ca6389e1
Tinygrad runner ( #34261 )
...
* squash
* dmonitoringmodeld: use cl transform (#34235 )
* needs cleanup
* only if tici
* bump tinygrad
* check width
* base modelframe
* .
* need to be args
* more cleanup
* no _frame in base
* tici only
* its DrivingModelFrame
* .6 is fair
---------
Co-authored-by: Comma Device <device@comma.ai >
* Update tinygrad
* tg upstream
* bump tg
* bump tg
* debug
* attr
* misc cleanup
* whitespace
* remove
* Add TODOs to make python proc for modelrunners
* whitespace
---------
Co-authored-by: ZwX1616 <zwx1616@gmail.com >
Co-authored-by: Comma Device <device@comma.ai >
Co-authored-by: Maxime Desroches <desroches.maxime@gmail.com >
2024-12-18 11:58:59 -08:00
f6885dcbec
Revert Tinygrad ( #34243 )
...
* Revert "dmonitoringmodeld: use cl transform (#34235 )"
This reverts commit 684b0b9d4d31606a7d1544f58ff758#34171 )"
This reverts commit 7b5a4fbb03
2024-12-14 12:15:36 -08:00
7b5a4fbb03
Tinygrad runner ( #34171 )
...
* squash
* bump tg
* bump tg
* debump tinygrad
* bump tinygrad
* bump tg
* Skip init iteration
* fixes
* cleanups
* skip first test sample
* typos
* linter unhappy
* update cpu usage
* OPENCL just zeros for now
* imports
* Try printing
* Runs again, but slower
* unused import
* Allow more buffer with tg and all on gpu
* bump tinygrad
* seems ok
* stricter timings for driving looser for dm
* try llvm
* check nvidia
* More timeout for now
* make test pass
* Revert "try llvm"
This reverts commit ef136e478320101fea262bae3579e558da991902.
* small fixes
* whitespace
* revert test timeout
* No model runners
* Always CPU always fast
* No onnx runtime GPU
* more cores
* cleanup
* Is this faster
* Is this faster
* at least runs
* FP32 is faster than 16
* fix deps
* whitespace
* comment
---------
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com >
2024-12-11 23:15:20 -08:00
00c10f6851
ci: adjust pandad cpu usage ( #34077 )
...
more
2024-11-21 11:32:06 -08:00
b206879e4d
ci: more robust memory usage test in test_onroad ( #34067 )
...
metric
2024-11-19 21:07:22 -08:00
bf21e10d81
ci: move manager test_startup_time to test_onroad ( #34062 )
...
* get
* fix
* now
* try
* better sign
* better
* better
* clean
* space
* fix
* more
* msg
2024-11-19 14:27:15 -08:00
d9d57e5d6f
Revert ISP image processing + tinygrad bump ( #34020 )
...
* Revert "Replace ThneedModel with TinygradModel (#33532 )"
This reverts commit da952e9b64#33959 )"
This reverts commit f2a1cce42b
2024-11-13 19:27:11 -08:00
4948f910f0
ci: fix test_onroad ( #34016 )
...
fix
2024-11-13 12:02:17 -08:00
da952e9b64
Replace ThneedModel with TinygradModel ( #33532 )
...
* squash
* bump tg
* bump tg
* debump tinygrad
* bump tinygrad
* bump tg
* Skip init iteration
* fixes
* cleanups
* skip first test sample
* typos
* linter unhappy
* update cpu usage
* OPENCL just zeros for now
* imports
* Try printing
* Runs again, but slower
* unused import
* Allow more buffer with tg and all on gpu
* bump tinygrad
---------
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com >
Co-authored-by: Bruce Wayne <harald.the.engineer@gmail.com >
2024-11-11 13:32:21 -08:00
700a5651bd
ci: nicer output for test_onroad ( #33961 )
...
* viz
* viz
* fix
* show more
* fix
* failure
* cleanup
2024-11-07 21:45:51 -08:00
6a2ae346ab
ci: relax test_onroad cpu usage bounds ( #33951 )
...
bounds
2024-11-06 20:30:38 -08:00
32c5254415
ci: faster test_onroad ( #33906 )
...
* 30s
* ruff
* remove
* timeout
* global
* value
* get value
* better
* format
* cleaner
2024-11-02 21:19:05 -07:00
fcbdd39aaf
test_onroad: lower bound for soundd cpu usage ( #33883 )
...
lb
2024-10-28 17:25:48 -07:00
abdd3ad2e9
lower card CPU usage threshold
...
from removing capnp conversion!
2024-10-14 14:07:48 -07:00
7556233cca
Reapply "move car.capnp to opendbc ( #33722 )" ( #33728 )
...
* Reapply "move car.capnp to opendbc" (#33725 )
This reverts commit 9d52a5b485
2024-10-03 22:47:03 -07:00
af774d894e
Deprecate car onroadEvents ( #33687 )
...
* new OnroadEvent struct
* and migrate python
* more forgotten
* re-index new OnroadEvent struct
* fix!
* more missing
* migrate onroadEvents
migrate onroadEvents
* migrate dm events
* hacks to get proc replay to say succeeded
* Revert "hacks to get proc replay to say succeeded"
This reverts commit 0bb8803e5755d606ae9f09da5395d9f50678c7c7.
* update refs
2024-10-03 15:20:58 -07:00
af609212f7
bump qcamera too
2024-10-02 10:10:11 -07:00
c87f953396
test_onroad: bump up hevc size
2024-10-01 23:21:33 -07:00
d826b08369
ci: lower soundd CPU budget ( #33684 )
...
tr
2024-09-30 18:59:46 -07:00
3b6f1a0b39
test_onroad: update CPU usages after recent changes
2024-09-29 10:55:39 -07:00
876f192112
exec DM model with gpu ( #33609 )
...
* half
old-commit-hash: 9f72eca003d4637ca7fe847ebaf925c694fc2e84
* optimed
old-commit-hash: 6e36e2a12e09275ec21d1590012a92b05ca52ff5
* thneed
old-commit-hash: 419a06c09c0c767d828bcd1e118bc575898c343a
* exec
old-commit-hash: 0059c27ec11b076a37f65d604ed135ea6541b1a6
* runner
old-commit-hash: 34232ada94450ce541eaef546197fa219810a891
* runs but
old-commit-hash: 3db37c00b6a64908293b4de8d8b56e80308cd8f2
* it is 01
old-commit-hash: a160d81eb1a7e77abbef959b44f602610f68f665
* np
old-commit-hash: c1caff6ba648cc2c0094c71b2ea074f01c3c2dc8
* module url
old-commit-hash: 6f4902c4d384263a53e2c1d14d93b5ff864b6a5f
* new
old-commit-hash: 779ae79b1bc3df6374fb6663ac8592e107a6e504
* ds fast
* is this work
* corcention
* real timing
* no reg
* interim gather
* 0e4a9c7b
* fa69be01, and halve
* list
* cleanup
* slighly faster
* setprotlt
* expected
* replay ref
* more powar
* reluctantly
* bump tg
* 8
* less
* less
* bump tg
* better than exp
* closer
* cc
* see diff
* commits
* was right
* to 32 cast
* remove dlc file
* support both
* dspExecutionTime -> gpuExecutionTime
* ignore
* time ref
* ref commit
* last
---------
Co-authored-by: Comma Device <device@comma.ai >
2024-09-26 16:40:44 -07:00
ffb6e11f96
Revert "exec DM model with gpu" ( #33604 )
...
Revert "exec DM model with gpu (#33397 )"
This reverts commit f79aca8e1e
2024-09-19 20:55:11 -07:00
f79aca8e1e
exec DM model with gpu ( #33397 )
...
* half
old-commit-hash: 9f72eca003d4637ca7fe847ebaf925c694fc2e84
* optimed
old-commit-hash: 6e36e2a12e09275ec21d1590012a92b05ca52ff5
* thneed
old-commit-hash: 419a06c09c0c767d828bcd1e118bc575898c343a
* exec
old-commit-hash: 0059c27ec11b076a37f65d604ed135ea6541b1a6
* runner
old-commit-hash: 34232ada94450ce541eaef546197fa219810a891
* runs but
old-commit-hash: 3db37c00b6a64908293b4de8d8b56e80308cd8f2
* it is 01
old-commit-hash: a160d81eb1a7e77abbef959b44f602610f68f665
* np
old-commit-hash: c1caff6ba648cc2c0094c71b2ea074f01c3c2dc8
* module url
old-commit-hash: 6f4902c4d384263a53e2c1d14d93b5ff864b6a5f
* new
old-commit-hash: 779ae79b1bc3df6374fb6663ac8592e107a6e504
* ds fast
* is this work
* corcention
* real timing
* no reg
* interim gather
* 0e4a9c7b
* fa69be01, and halve
* list
* cleanup
* slighly faster
* setprotlt
* expected
* replay ref
* more powar
* reluctantly
* bump tg
* 8
* less
* less
* bump tg
* better than exp
* closer
* cc
* see diff
* commits
* was right
* to 32 cast
* remove dlc file
* support both
---------
Co-authored-by: Comma Device <device@comma.ai >
2024-09-19 17:42:44 -07:00
f294162b2c
bump up cpu budget
2024-09-16 19:23:04 -07:00
df5b45a285
modeld has gone up a bit
2024-09-16 18:05:07 -07:00
e04455cbaa
selfdrived: controlsd only does controls ( #33485 )
...
* selfdrived
* process replay
* lil more
* set the valids
* rename that
2024-09-06 17:16:32 -07:00
51ebf44f48
card: remove redundant CAN data conversion ( #33488 )
...
* remove redundant CAN data conversion
* 1 less %
* rename
---------
Co-authored-by: Shane Smiskol <shane@smiskol.com >
2024-09-06 13:51:19 -07:00
922348f33d
card parses radar points ( #33443 )
...
* interfaces returns radarinterface
old-commit-hash: 9ad1f096bfca3712320f19b1b49aa9e6ac9b68e4
* bump
old-commit-hash: 20334a8b257c6037e11d02f2ba7b1f02c59f3240
* get RI from opendbc
old-commit-hash: b5f6d0c48c90927926e9dd557130075aeec5edee
* stash so far
old-commit-hash: 5aa2c842eb152316434c17a661df05bb8af61f47
* new liveTracks message (radard expects and needs RadarData)
* this should just work?
* whoops
* fix that
* rm liveTracks from radard pm
* fix proceess replay
* lol fcw diff, something's not right
* actually there's fcw in original route. it's pretty close
* no tracks!
* fix test_leads
* CPU moved across procs
* fix not engageable from onroadEvents
* bump
* fixes
* bump to master
* radard publishes w/ modelV2 now, so it will always be sent. check valid which radard sets using liveTracks avg freq
* fix that (it works!)
* combine
join
* bump
* bump
* deprecate
* why
* fix incorrect args
* remove cumLagMs from process_replay
* update refs
2024-09-05 22:40:17 -07:00
236dffe400
locationd no GPS ( #33029 )
...
* Pose kf draft
old-commit-hash: 17dd4d576e597792f0e18826498c00076739f92b
* Fix it
old-commit-hash: 13ac120affe58fd22e871586ea5f4d335b3e9d2b
* Add translation noise
old-commit-hash: 166529cb612858c4ce80649367ac35b2b6007e1d
* Add gravity to acc
old-commit-hash: 8fcfed544b8e090ccc86b189c13bc03c6c190613
* Use pyx
old-commit-hash: 8e69e0baa0a4c43b4d0c22535711f296f05420aa
* Indent
old-commit-hash: 25b19a73644cdcb571ccf1a1d8acb88a0d066c67
* Reset function
old-commit-hash: ca5d2736da15e4fd6539f7268133320735b7c9cc
* Add device_from_ned and ned_from_device transformations
old-commit-hash: a60d25da0edc311e583549dc015fa595749fd4ae
* Fix rotations
old-commit-hash: d6d582f7f6d19a2bc2308dbcb0c9f81363e325b6
* kms
old-commit-hash: 681bc4e50374795ccc61422c3ce4ffb51389fce2
* Centripetal acceleration
old-commit-hash: 6e574506d27e5b76a04b2097d94efa4ca91ead71
* Rewrite draft
old-commit-hash: 4a2aad0146267460e5d30036c8cdb2bef94d1d7c
* Remove old locationd stuff
old-commit-hash: c2be9f7dbf22fb5cd29e437cd7891a7d52266fba
* Python process now
old-commit-hash: 83fac85f28c0b546b6965aafe1dd8a089e67f5b3
* Process replay fix
old-commit-hash: c44f9de98583c49dad0b22497869b3bb0266fcd9
* Add checks for timing and validity
old-commit-hash: aed4fbe2d00ca620e01a0e0ee99a4871c939de36
* Fixes
old-commit-hash: 3f052c658c16984a34915f38afdfbfd0fb19a267
* Process replay config fixes
old-commit-hash: 1c56690ee7ceb3c23c9ec2b2713352191212716e
* static analysis fixes
old-commit-hash: 6145e2c140ea9aa97e75069c3ddd82172cadc866
* lp in latcontrol
old-commit-hash: 9abf7359d68e794c69052724f3aca14b04dd3cca
* Fix SensorEvent name for acceleration
old-commit-hash: 91a1ad6c604727c9c898ba4aefe9478022b167fd
* Ignore sensor readings from segments with multiple imus
old-commit-hash: 1f05af63d6cc605ea98d7da0d727a5bd8d7819b0
* Update shebang
old-commit-hash: e3f2f5c10df3a4ba698421335bfeffc63d1a8797
* Replace llk with lp
old-commit-hash: 99b6c7ba08de6b703708fef0b8fd2d8cb24b92c0
* Refactor locationd scenario test
old-commit-hash: 7f5f788f071b7647e36f854df927cc5b6f819a84
* Add more debugging tools
old-commit-hash: 8d4e364867e143ea35f4bfd00d8212aaf170a1d1
* Param name update
old-commit-hash: 5151e8f5520f067e7808e3f0baa628fbf8fb7337
* Fix expected observations
old-commit-hash: d6a0d4c1a96c438fb6893e8b6ff43becf6061b75
* Handle invalid measurements
old-commit-hash: 549362571e74ad1e7ec9368f6026378c54a29adf
* Fix spelling
old-commit-hash: eefd7c4c92fb486452e9b83c7121d2599811852b
* Include observations in debug info too
old-commit-hash: 625806d1110b3bffe249cd1d03416f2a3f2c1868
* Store error instead of expected observation
old-commit-hash: 1cb7a799b67e56af4eddc6608d5b0e295f2d888c
* Renames
old-commit-hash: a7eec74640fc5cc7a5e86172d2087b66cb93d17d
* Zero the yaw
old-commit-hash: 96150779590fcb8ac50e8ffe8f8df03105983179
* New state_dot for orientation
old-commit-hash: c1456bf3a0c5af2f888aa6ff0b5ffc2e5516ddf7
* Fix the state transformations
old-commit-hash: 7cb9b91e2f99caa4ac3fb748c7f23bb8bf9f65db
* Update process in test_onroad
old-commit-hash: 854afab7c39ca7dec2b42974cecbb5310b82b617
* Test polling on cameraOdometry
old-commit-hash: a78e8b7d61618177f15c9892e2fa1e51620deca8
* Keep the copy of x and P returned from predict
old-commit-hash: 3c4159a6a7d7383265a99f3f78f8805d2fcfc8cd
* Remove polling again
old-commit-hash: f7228675c5fd2de5f879c4786859f1abcec27d68
* Remove locationd.cc
old-commit-hash: d7005599b2b178e688c3bd1959d6b69357d3a663
* Optim in _finite_check
old-commit-hash: 58ad6a06b9380960e9f69eb98663ddb97461e8d5
* Access .t once
old-commit-hash: d03252e75ed4cbdb49291a60c904568e6a3d3399
* Move the timing check to cam odo code path
old-commit-hash: 6a1c26f8c201e1feb601753f0cb7299cf981b47e
* Call all_checks only once
old-commit-hash: 373809cebf8d9db89d1ab00f4c8c933f33038e78
* Do not sort
old-commit-hash: 2984cd02c0ab76827b8c7e32f7e637b261425025
* Check sm.updated
old-commit-hash: 11c48de3f0802eb4783899f6a37737078dbf2da4
* Remove test_params_gps
old-commit-hash: 82db4fd1a876cc2402702edc74eba0a8ac5da858
* Increase tolerance
old-commit-hash: 927d6f05249d2c8ec40b32e2a0dcde8e1a469fb3
* Fix static
old-commit-hash: 2d86d62c74d5ac0ad56ec3855a126e00a26cd490
* Try separate sockets for sensors
old-commit-hash: 5dade63947ab237f0b4555f45d941a8851449ab1
* sensor_all_checks
old-commit-hash: e25f40dd6b37ee76cd9cc2b19be552baf1355ec3
* Fix static
old-commit-hash: 328cf1ad86079746b4f3fde55539e4acb92d285e
* Set the cpu limit to 25
old-commit-hash: 7ba696ff54c5d3bfa42e42624d124f2a1914a96d
* Make it prettier
old-commit-hash: cd6270dec80d8b9dac784ddd4767a1a46bcff4b7
* Prettier
old-commit-hash: 1b17931d23d37f299dad54139eaf283a89592bf5
* Increase the cpu budget to 260
old-commit-hash: 20173afb937a2609c8a9905aee0b2b093cb8bba4
* Change trans std mult. 2 works better
* Update ref commit
* Update ref commit
2024-09-04 11:54:57 +02:00
Adeeb Shihadeh