Clearing the watchdog while it is in progress can cause bus stalls on
the SAMD21. It appears that clearing the watchdog on the SAMD51 can
cause lockups of the watchdog device.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The sram matrix configuration is, bizarrely, configured with a slow
default. It will typically add an extra wait state to every memory
access. Set the matrix sram controller to improve the performance.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Avoid using the "low-level" library timing utilities. This is in
preparation for using SysTick as part of the timer implementation.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The ARM CortexM3 (and later) chips almost always have a 32bit timer in
the "Debug and Watch Trace" (DWT) unit which resides within the cpu
core itself. This timer is generally faster to access than timers
located on the chip's peripheral bus. When combined with the standard
ARM SysTick timer it provides a (generally) portable mechanism to
implement Klipper's software timer mechanism. It's generally also
faster than using the vendor specific timers.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Move the definition of SERIAL_BAUD to the main src/Kconfig file. Only
show this as an option if the user selects the ability to configure
low-level options.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The lpc176x ADC is extremely noisy. Implement a 5 order median filter
on each ADC read to weed out obviously incorrect readings.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Order the adc config code so that a shutdown from gpio_adc_setup()
doesn't result in analog_in_shutdown() running on a 'struct analog_in'
that hasn't yet been configured.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Make sure to explicitly cast to 32bit integers when doing math on
16bit integers that require the extra precision as the AVR uses a
16bit int.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The sam3 i2c code and the linux code use a 7-bit i2c address, while
the avr, lpc176x, and samd21 i2c code uses an 8-bit address with the
least significant bit always zero. A similar issue occurred in the
host code (sx1509.py and replicape.py use 7-bit addresses while
uc1701.py and mcp4451.py use 8-bit addresses).
Consistently use 7-bit addresses in all the code. This breaks
compatibility between host and mcu software, so make a change to the
config_i2c command to force users to synchronize software updates.
This also breaks common Smoothieboard configs, so update the mcp4451
code to validate the i2c_address.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Most of the peripherals on the sam4e8e are similar to the ones on the
sam3x8e mcu. Merge the code together and use just one code directory.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Remove the 'pin' and 'afec' variables from 'struct gpio_adc'. This is
in preparation to merge the sam3x8e and sam4e8e code.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Use internal.h in a similar way to the internal.h in sam3x8e. Also,
move the twi pin definitions from internal.h to i2c.c (as they are only
used there).
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
These two files from atmel have different licenses than the other
files. Neither files provide any real value, so just remove them to
avoid any confusion.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Pass the gpio id instead of the bank/pin to gpio_peripheral(). This
is in keeping with other ARM ports.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Scale the bus speed within the peripheral instead of on the system bus
as scaling on the system bus slows down general register access to the
peripheral.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Simultaneously define the hardware pin and the timer registers. This
makes the table a little easier to understand.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Increase the amount of time to "busy wait" in the timer dispatch loop
to 2 micro-seconds. This, in practice, causes stepper step and unstep
events to occur within a single hardware interrupt. Doing that helps
stabilize the single active stepper performance benchmarks.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The atmega328p is basically the same as the atmega328 - add explicit
support for it so that avrdude doesn't complain while flashing.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Sometime bossac reports an error during chip restart. It appears this
error is spurious - just suppress it.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Scaling the timer on the system bus has a notable impact on
performance. So, scale the timer within the timer hardware itself.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
It appears the sam3x8e has a primitive cache prefetch mechanism (it
prefetches 32 bytes at a time aligned to a 16 byte boundary).
Aligning the main loop in timer_dispatch_many() to a 16 byte boundary
significantly improves performance.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
If sched_try_shutdown() is not inlined into timer_dispatch_many() it
can have a significant impact on benchmarks. (Even though the
function isn't called in practice, the call causes gcc to organize the
function differently.)
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Some older versions of gcc need the __always_inline directive in order
to inline timer_read_time. Inlining that function is important for
performance on the stm32f1.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
As long as timer_read_time() is inlined, I get better performance with
gcc -O2 optimization. The binary is also dramatically smaller and O2
better matches the other platforms.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
It's possible for the code to stay in the timer irq for up to 100ms,
so serial irqs should have a higher irq priority to prevent them from
being starved. (The timer code disables irqs during event dispatch,
so serial irqs would only be permitted between events anyway.)
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The usb_cdc code will arrange to call usb_read_ep0_setup() after a
shutdown, so no need to have an explicit shutdown handler.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Call usb_read_bulk_out() before calling command_find_and_dispatch() as
this optimizes the common case where each usb packet contains a single
new message block.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Recent code additions have increased the stack usage and it is no
longer clear that everything will fit in 128 bytes. Increase all
targets to 256 bytes (the atmega2560 was already at 256 bytes).
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Fix travis-ci build failure by forcing the data field of the struct
usb_string_descriptor to be the same type as a 16bit unicode string.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Use new usb_read_ep0_setup() function when reading a setup packet -
this allows the low-level usb hardware code to better handle errors.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
A zero length packet should only be sent at the end of a transmission
if the host is expecting more data.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Maintain the state tracking code entirely within a usb_do_xfer()
method. This simplifies the callers.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Use "ep0" when referring to endpoint0 - don't use "setup" as that can
be confused with the low-level usb setup token.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The atmega168 doesn't have enough flash space to fit all features -
disable the "bit banging" interfaces to make room.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The bank_id was calculated erroneously. Also unify gpio_in_setup and gpio_out_setup a bit. Also disable make flash for the sam4e8e port as it currently references a version of bossac that is not present on all systems.
Signed-off-by: Florian Heilmann <Florian.Heilmann@gmx.net>
This can be flashed to e.g. the duet wifi using bossac. It requires a
later version as is currently included in the klipper repo (1.8
vs. 1.2). Comms are currently via UART0 only, USB serial is still TBD
Signed-off-by: Florian Heilmann <Florian.Heilmann@gmx.net>
It's common for boards to have restrictions on the type of each
endpoint. So, make it possible for the board to select the endpoint
ids for each endpoint.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
No need to include separate cmsis files from the mbed project - the
standard CMSIS_5 files work fine.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Update the sam3x8e port use the most recent SAM3X CMSIS files from
Atmel. The Atmel files appear to have a more consistent license and
it is unclear if there were Arduino specific changes in the CMSIS
files obtained from the Arduino repo.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Send the temperature before checking for an out of range error (that
makes it easier to debug).
The query_thermocouple clock parameter is unpredictable and a value of
zero can't be used to disable the query - use rest_ticks instead.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The MAX6675 chip has a different read sequence than the MAX31855 chip.
Signed-off-by: Ricardo Amézquita <ramezquitao@cihologramas.com>
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Change the simulator to use the generic timer_irq.c and serial_irq.c
code for (dummy) timer and io handling. This is just to make the code
a better example for other developers (most micro-controllers will use
the timer_irq.c and serial_irq.c code).
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The SPI interface needs to be enabled and configured to the correct
settings of a given oid before CS is asserted. The new function
spi_prepare() allows ports to do that.
This port only introduces the new function in all ports with no
implementation and adds the call to the Klipper generic firmware code.
That means everything still works as before. Ports need to be changed
to fix the underlying issue.
Discussion about the motivation here:
https://github.com/KevinOConnor/klipper/pull/453#issuecomment-403131149
Signed-off-by: Grigori Goronzy <greg@chown.ath.cx>
Add basic SPI support and associated documentation.
v2: remove baud rate check, fix baud rate calculations
v3: finish transaction with BSY check, disable SPI when not in use
Signed-off-by: Grigori Goronzy <greg@chown.ath.cx>
Extend the ADC out of range check so that it is possible to sample
multiple times before going into a shutdown state. This reduces the
chance that measurement noise will cause an error. In an actual over
temperature (or under temperature event) it is expected that the
sensor will consistently report the problem, so extra checks for an
additional second or two should not substantially increase risk.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add mcu support for periodically polling for a button press. Add host
code support for registering buttons and invoking callbacks for them.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
For some reason, the upper two pin definitions were missing. Found while
writing the port documentation.
Signed-off-by: Grigori Goronzy <greg@chown.ath.cx>
Add a command that can immediately set the value of a digital output
pin created with config_digital_out.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The AVR requires that the SS pin be an output pin for correct SPI
operation. Some boards use the SS pin to control devices separate
from SPI, however. Don't change the output level if the pin is
already an output, and prefer setting it low if it is not already an
output pin.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The PRU code size has grown too close to the maximum - remove the ADC
code to reduce its overall size. (On the beaglebone, the Linux
process can, and typically does, do the ADC reading.)
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
It appears the st7920 requires a longer delay when switching from
command to data mode (and vice-versa). Slower MCUs don't show a
problem because the klipper command processing time results in a
sufficient delay. However, some of the faster MCUs can process
klipper commands fast enough that the next st7920 transfer is sent too
fast. Add an additional delay to account for this.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Change the get_status command to get_clock. Don't report the shutdown
status in the new get_clock command.
The primary purpose of this change is to force the host code to report
a firmware version mismatch with older firmwares as recent changes
(namely the ordering of message block acks) have subtle
incompatibilities if different host/mcu code is used.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Normally, the MCU is restarted on a config change. However, that
should not be done automatically if the MCU is in a shutdown state.
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>
Send the ack after processing commands - this gives the host code more
information on serial buffer utilization.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Prefer using 'uint8_t' buffers as it is too easy to run into C sign
extension problems with 'char' buffers. Prefer using 'uint_fast8_t'
for buffer lengths as gcc does a better job compiling them on 32bit
mcus.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The gpio_out_toggle() function in the sam3x8e and stm32f1 code was
only valid if it was called with irqs disabled.
Commits 018c5daa and 9c52ad43 enabled the lcd code which called
gpio_out_toggle() with irqs enabled. This could cause corruption of
the gpio state.
Introduce a gpio_out_toggle_noirq() function that will only be invoked
with irqs disabled, and fix gpio_out_toggle() on sam3x8e and stm32f1
so that it safe to call even if irqs are enabled.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Use the generic spi send/receive code on Linux. Update the replicape
code to use the updated command format.
Also, update the replicape code to turn off the stepper motors on a
shutdown event.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Improve the SPI message transmit system. Add support for bus speed
and bus mode. Add support for sending SPI messages on shutdown.
Signed-off-by: Petri Honkala <cruwaller@gmail.com>
Signed-off-by: Douglas Hammond <wizhippo@gmail.com>
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The AVR chips (bizarrely) start an ADC conversion when entering Idle
mode. This behavior can cause the ADC to be busy when a sample is
required. Worse, if a series of events cause the cpu to enter and
leave Idle mode with a timing similar to the ADC checking rate then it
can cause the ADC to show as busy for extended periods. This could
cause high MCU load and possibly lead to a "Rescheduled timer in the
past" shutdown.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Don't logically or the ADSC bit on the start of a conversion -
explicitly set the full contents of the register. Also, clear the
ADIF flag on each write.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Define unique register aliases for all of the hardware serial port
definitions. This makes it easier to deal with the AVR chips that use
different register names.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Extract out common code from avr/serial.c, sam3x8e/serial.c, and
stm32f1/serial.c into a new generic/serial_irq.c file.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
With the optimized timer_read_time() it is no longer necessary to
implement custom timer dispatch code - use the generic mechanism in
timer_irq.c. This simplifies the code and provides a small
performance increase.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Implement 32bit timer conversion without disabling interrupts. This
uses the 16th bit of timer_high as a rollover detection flag.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add a fully functional STM32F1 port, currently mostly targeting STM32F103
microcontrollers. This requires an 8 MHz XTAL. The maximum possible step rate
is around 282K steps per second.
This uses stm32flash to burn the firmware. The bootloader needs to be started
by setting BOOT0 to 1 and resetting the MCU. There is no automatic bootloader,
unlike on Arduino.
Signed-off-by: Grigori Goronzy <greg@kinoho.net>
When initializing the ADC, explicitly clear the registers (instead of
logically or'ing them with their previous values).
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
It appears the most common type of avrdude protocol for devices other
than the atmega2560 is the "arduino" protocol. Update the build to
select a different protocol based on the avr processor type.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
On the AVR platform (and possibly others) the longjmp() call will
restore the interrupt state saved during the setjmp() call. So, the
setjmp() call must be invoked with interrupts disabled to ensure that
shutdown handling is run with interrupts disabled. This fixes
potential corruption of the shutdown processing on AVR.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Properly handle the (unlikely) case that a schedule_soft_pwm_out
command is received before a previous command is fully processed,
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Only open the watchdog device after the console has been opened. The
machine should not reboot in the unlikely event the console can't be
opened.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Allow the start value to be different from the default/shutdown value
for the pin. This will be useful for "heater fans" that should
startup in the off state, and transition to full on in a shutdown
state.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Continue to sample the ADC input pins even if the MCU goes into a
shutdown state. This enables the printer to continue reporting
temperatures even on an mcu error.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Pass the sample_ticks and sample_count parameters directly in the
end_stop_home command instead. This simplifies the code.
Also, simplify calculation of next wakeup time in
end_stop_oversample_event().
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Some printers can show occasional noise on the endstop pin. Support
sampling the endstop pin multiple times to attempt to filter out this
noise.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Update the scripts used to install and "flash" the pru
micro-controller code. Also, add a "flash" script for the linux
micro-controller code. This makes it easier to install Klipper on a
Beaglebone board that uses a Replicape.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add support for reading analog values via the standard Linux IIO
interface. This can be used on Replicape boards to sample analog
input pins.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add support for sending SPI messages to devices via the standard Linux
SPI interface. This can be used to configure the shift registers on
Replicape boards.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add support for controlling pca9685 PWM drivers using the standard
Linux I2C interface. The pca9685 device is found on Replicape boards.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add support for compiling the Klipper micro-controller code as a
real-time process capable of running on standard Linux systems.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
If pru0 waits in sleep mode then a full beaglebone reboot is needed to
reprogram the pru0 firmware. For now, avoid using the "slp"
instruction as a workaround.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Allow pru0 to gather multiple outgoing message blocks into a single
rpmsg. This can reduce communication overhead.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Each irq flag must only be cleared after it has been serviced. This
fixes a race condition that could cause incoming commands to be
delayed for extended period.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
It's possible for multiple blocks to be pending on the incoming
"rpmsg" stream. Don't sleep unless the input is confirmed to be
empty.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Compile the PRU binary with -Os optimization, but request that the
timer dispatch code be compiled with -O2 optimization. This improves
the performance of timers slightly.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The gcc -Os option significantly reduces the size of the PRU binary
and it has little impact on performance.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
The command_encodef() can read the max_size parameter directly from
the 'struct command_encoder' passed into it. Also, there is no need
to check that a message will fit in a buffer if the buffer is declared
to be MESSAGE_MAX in size.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Go back to scheduling the unstep time instead of busy waiting in the
timer dispatch. With the unstep time increased to 2us, it no longer
makes sense to spin while waiting for the unstep.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Allow timer_dispatch_many() to run for extended periods if there are
no tasks pending. This reduces the amount of lost cpu time spent
entering and exiting the irq handler.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Send a "startup" message after completing the mcu init functions.
This may help detect mcu reboots during debugging.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Allow the timer dispatch irq handler to run for extended periods if
there are no tasks pending. This reduces the amount of lost cpu time
spent entering and exiting the irq handler.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Track when tasks are pending and spin in irq_wait() when no tasks are
pending. This improves the mechanism for sleeping the processor -
it's simpler for the board specific code and it reduces the
possibility of the processor sleeping when tasks are busy.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Add function to indicate when tasks need to be run. This will allow
the scheduler code to know if there are any tasks that need to be
processed.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Don't rely on the generic scheduler code to always have a timer no
more than 1ms in the future. Instead, create an avr specific timer
that will be called every 0x8000 ticks. This simplifies the generic
code and it reduces the amount of code that needs to be run every
millisecond.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
If sched_add_timer() is called on a timer that would make it the new
head of the list, then add it and signal the board code that the timer
should be rescheduled.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
It's not necessary to keep a millisecond counter. Replace the two
users of sched_check_periodic() with explicit task wakeup flags.
Signed-off-by: Kevin O'Connor <kevin@koconnor.net>
Kevin O'Connor