* Revert "IsoTpParallelQuery: extend timeout for every frame type (#29654)"
This reverts commit a160f1a624.
* bump
* bump to master
* figure out a clean way to exclude segments
* revert
* bump
* better ordering and comments
* remove
* mark done if timed out
* response_timeouts should be only for extending timeouts, not marking done
* so the print still works
* more clear comment
* that can be removed
* revert isotpparallelquery returning rx addr for functional special case
* we don't really use the tx addr (and soon won't make sense with fun querying)
* stash
* process all functional addrs (stash)
* clean up
* rm
* simplify
* let user pass in tx for rx addrs
* revert panda
* simplify
* comment
order
* need to go by rx_addr now
* Revert "need to go by rx_addr now"
This reverts commit 1197ecfbc5.
* stash
* should also work
* this seems pretty clean
* not used
* properly use
* comment
* some fixes
* some fixes
* send consecutive frames on physical addrs
* bump panda
* looks better
* setup_only
* Revert VIN changes
* rev
* bump panda to master
* Update selfdrive/car/isotp_parallel_query.py
* Revert "VIN: query physical addresses (#25122)"
This reverts commit 0697ca2239.
* try sending tester present
* do CAN fingerprinting first
* looks like we can get rid of this!
* remove import
* no cache for testing
* revert
* revert
* move function to fw_versions
* Exception
* Revert fp order, sleep to let PubSocket connect
* comment
comment
* space
* Update selfdrive/car/car_helpers.py
* at 0.06 is where it becomes more consistent
* treat functional addrs like physical addrs (process all responses)
* fixes and debugging
* fix
* revert other changes
* Update selfdrive/car/isotp_parallel_query.py
* caps
* raise timeouts
* extend timeout for each message/frame (not multi-frame full response)
* bump panda
* 100 ms timeout between frames (max I've seen is 20, should be good)
* bump panda
* remove unused
* Add tester present to HKG queries (temp)
* send tester present to all ecus first
* vin and fw_versions.py sleep, should figure out why sendcan drops packets
* None is wildcard (some ecus respond with negative code, that's fine)
* typing
* try bus 0
* revert brand-specific changes
* Update selfdrive/car/isotp_parallel_query.py
* handle response pending
* match commit
* remove total timeout, just keep track of individual response timeouts
* fix
* add back total timeout
* this isn't reliable enough
* keep track of pending responses to print warning
* tx_addr is (addr, subaddr)
* debug
only query hyundai
import time
reponse pending
no cache
all cars
no timeout to test before
* Revert "debug"
This reverts commit abe9cfc1b6.
* always print pending
always debug
* Only debug
* Update selfdrive/car/isotp_parallel_query.py
* remove variable only for debugging
* log fw query response offset
* one line still
* log request bytes as well
* bump cereal
* prettier order
* bump cereal
* I wonder if the radar sends a response pending...
debug
dont cache
fix
debug
* Revert "I wonder if the radar sends a response pending..."
This reverts commit 12a9552c27.
* log addr too
* bump cereal
* improve tools.lib.kbhit and tools.sim.lib.keyboard_ctrl
* unpack more efficiently
* minor improvements
* agnos.py match spec better
* manual_ctrl test missing queue arg
* fix incorrect type annotation
* queues are generic
* varname reuse resulting in incorrect type inference
* bytes().hex() rather than bytes.hex(bytes())
* a bit of type hinting stuff
* FPv2 support for MQB
* Comment touch-ups
* Trim extra newlines
* Add Atlas and Tiguan
* Update submodule ref
* Revert "Add Atlas and Tiguan"
This reverts commit a47bc1bca7.
* Curb my enthusiasm
* Drop ABS/ESP from UDS fingerprinting for now
* Use multi-query to grab SW build ID and parameterization
* Resolve merge oops
* Resolve merge conflicts
* Make Ecu.fwdCamera optional for VW
* Where'd that come from?
* Propitiate process replay checks until ready for model dump
* FW values for Audi A3 Prestige
* Revert "Make Ecu.fwdCamera optional for VW"
This reverts commit 56852b99
* Don't try to FP fwdCamera for VW
* Remove FW comments
* Simplify RX offset handling
* Corrected scoping for MQB trans detection
* Add default response_offset
* Remove explicit list of MQB vehicles for now
* Deal with pylint warning
* Clarify and simplify comments
Co-authored-by: Comma Device <device@comma.ai>
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
Shane Smiskol
Adeeb Shihadeh
Dean Lee
Jason Young
Willem Melching
Josh Smith
Adeeb
George Hotz