It seems opening the canbus socket can sometimes throw an os.error (in
addition to can.CanError). Catch that error to avoid Klipper
reporting an internal error.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Temporarily postpone loading the CAN module so that moonraker (and
similar) can improve upgrade scripts.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The tmc2130 (and tmc5160) will respond back with the value written
during the next SPI command. Use this feature to verify that the
value written matches the value sent.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Don't try to detect a "real serial port" in serialhdl.py. Instead,
have the callers invoke either connect_uart(), connect_file(), or
connect_pipe().
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
When `serialhdl` opens a terminal device it must prevent the device from
becoming the controlling terminal of `klippy`, as such a terminal will
send additional messages to the session leader, e.g. a SIGHUP, which
would kill `klippy`. See e.g. #3981.
pySerial already does this, but for e.g. `klipper_mcu` we were not doing
this ourselves.
On Linux a process must set `O_NOCTTY` when opening any file that could
potentially be a terminal device, to avoid this.
Earlier process daemonization tools prevent this from being an issue by
double forking, but under `systemd` a process must take steps to avoid
opening a controlling terminal.
Signed-off-by: Lasse Dalegaard <dalegaard@gmail.com>
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The existence of a __del__() method prevents deallocation on python2
if there are circular references. Just remove the method as
disconnect() should already be called directly.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Fysetc Cheetah v1.2 boards require a special sequence to reset reliably.
This sequence works for me in all cases. Simpler sequences without
double reset did not work correctly. This is likely because of a weird
stateful circuitry for toggling the bootloader state.
Cheetah boards use RTS to configure bootloader triggering. By default,
pySerial sets RTS on connect, which unfortunately configures the board
to start the bootloader on reset.
Add a toggle for the RTS state to allow users to workaround. The RTS state
is set before the serial connection is opened, so there are no glitches.
Addresses #2026.
Signed-off-by: Grigori Goronzy <greg@chown.ath.cx>
Use new mcu.lookup_query_command() for all commands that query
information from the micro-controller.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Convert standard queries to use the acknowledgments of the sent
command to determine if the response was received or not. This also
controls command retransmissions (should the response have been lost).
The tmc_uart.py will continue to use the old mechanism as the tmcuart
responses are not sent synchronously with the query command.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add ability for the host code to get a notification when the ack for a
command sent to the micro-controller is received. This is in
preparation for improved detection of message loss between mcu and
host.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Move the serial command wrapper class from serialhdl.py to mcu.py.
This will allow that class to better support higher level
functionality.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Introduce a main send_with_response() helper function and use it
during identify and the setup of clocksync.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Reduce the delay in send_with_response() by introducing a reactor
mutex to wake the main thread when the response is obtained.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
If the mcu is using a traditional serial port, then only send a new
message block if there is space available in the mcu receive buffer.
This should make it significantly less likely that high load on the
mcu will result in retransmits.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Automatically clear the information printed at the start of each log
file rollover on a klippy internal restart.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add a lookup_command() method to the SerialReader class that provides
a wrapper that stores the serial and commandqueue references. This
makes it easier to run the send() method.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Store the gcc and binutils versions used in the compilation of the
firmware in the firmware data dictionary. Forward that information to
the log so it is available during debugging.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The logging module can build strings directly from printf syntax - no
need to build the string first.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Build a single (very large) logging message with the debug state.
This prevents the output from being fragmented.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The existing clock synchronization code has two flaws: intermittent
transmission latency during get_status requests can cause the
estimated clock to be too low, and the estimated clock calculation did
not take into account possible clock drift between samples. The
former could potentially lead to "Timer too close" errors and the
latter could potentially lead to "Move queue empty" errors.
Rework the code to avoid the above problems. It's not necessary to
estimate the micro-controller clock as an excellent estimate is
reported by the micro-controller (via the CLOCK_FREQ constant).
Account for a small drift from the reported value, and check on each
sample if the drift exceeds the expected limits. With a good starting
estimated clock, only the offset needs to be calculated. Use previous
offsets (and the estimated clock) in calculation of new offsets to
avoid intermittent latency from badly skewing the results. Finally,
add an additional time offset of one millisecond to account for any
minor inaccuracies.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Thijs Triemstra