Commit Graph

28 Commits

Author SHA1 Message Date
Scott Shawcroft 379e73af2e
Finer grained, per port tick locking
Fixes #3504 hopefully.
2020-10-12 18:43:21 -07:00
Dan Halbert 6dbd369272 merge from upstream 2020-08-30 14:39:03 -04:00
Jeff Epler 1033e89561 supervisor: use mp_handle_pending to check for exceptions 2020-08-23 09:12:01 -05:00
Dan Halbert 0a60aee3e4 wip: compiles 2020-08-02 11:36:38 -04:00
Jeff Epler 9fd10322fe supervisor: rename some locals for clarity
It's perfectly OK for these variables with static linkage to have the
same name, but it's inconvenient for humans like me.
2020-07-20 08:45:31 -05:00
Jeff Epler 1df48176ce supervisor: factor supervisor_background_tasks from sundry ports 2020-07-15 11:49:44 -05:00
Jeff Epler 8c4a9f6444 supervisor: tick: only run background tasks once per tick 2020-07-15 09:26:47 -05:00
Jeff Epler 1474fccd2f supervisor: Add a linked list of background callbacks
In time, we should transition interrupt driven background tasks out of the
overall run_background_tasks into distinct background callbacks,
so that the number of checks that occur with each tick is reduced.
2020-07-15 09:26:47 -05:00
Jeff Epler 51b9a1aeca tick.c: adjust whitespace 2020-07-15 09:26:47 -05:00
DavePutz b80abf1a90
Update handing of a CTRL-C exception 2020-06-23 12:15:01 -05:00
DavePutz 65512cef1c
Update tick.c 2020-06-22 15:59:15 -05:00
DavePutz 182a8733a4
Rework handling of ctrl-c interrupt 2020-06-21 13:28:26 -05:00
DavePutz 0b52359190
Generate stacktrace and reset exception for ctrl-c interrupt
Added code in mp_hal_delay_ms() to generate stacktrace and reset exception for ctrl-c interrupt
2020-06-19 19:02:43 -05:00
Diego Elio Pettenò dd5d7c86d2 Fix up end of file and trailing whitespace.
This can be enforced by pre-commit, but correct it separately to make it easier to review.
2020-06-03 10:56:35 +01:00
Sean Cross c7efc94a33 watchdog: move timeout exception to shared-bindings
Make this exception globally available to all platforms that have
enabled the watchdog timer.

Signed-off-by: Sean Cross <sean@xobs.io>
2020-05-27 11:28:49 +08:00
Sean Cross 15530a69c7 supervisor: tick: check for watchdog exception if enabled
Check to see if the current exception is a Watchdog exception, if it's
enabled. This ensures we break out of the current sleep() if a watchdog
timeout hits.

Signed-off-by: Sean Cross <sean@xobs.io>
2020-05-27 11:28:49 +08:00
Sean Cross ec99dae953 tick: break on watchdog timeout exception
Signed-off-by: Sean Cross <sean@xobs.io>
2020-05-27 11:28:49 +08:00
Scott Shawcroft e99cf6e441
Rework sleep timing
It didn't account for background task time and could end up
sleeping for way longer than it should because the RTC compare time
had already passed.
2020-04-03 18:07:56 -07:00
Scott Shawcroft c248730bd1
Clean up 2020-04-02 17:36:09 -07:00
Scott Shawcroft 2623022c84
Initial work for STM32. Need to fix us delay and PulseIn still. 2020-03-20 12:58:34 -07:00
Scott Shawcroft 7100d5e485
Fix autoreload and ticks in general 2020-03-13 16:13:24 -07:00
Scott Shawcroft 418333979a
Fix autoreload, neopixel, monotonic_ns and sleep w/o SD 2020-03-13 11:12:31 -07:00
Scott Shawcroft 6f60afe8c5
First try at lowering the power consumption 2020-03-13 11:12:30 -07:00
Scott Shawcroft 7d8dac9211
Refine iMX RT memory layout and add three boards
Introduces a way to place CircuitPython code and data into
tightly coupled memory (TCM) which is accessible by the CPU in a
single cycle. It also frees up room in the corresponding cache for
intermittent data. Loading from external flash is slow!

The data cache is also now enabled.

Adds support for the iMX RT 1021 chip. Adds three new boards:
* iMX RT 1020 EVK
* iMX RT 1060 EVK
* Teensy 4.0

Related to #2492, #2472 and #2477. Fixes #2475.
2020-01-17 17:36:08 -08:00
Jeff Epler 46c5775ba4 supervisor: tick: add supervisor_fake_tick 2019-11-20 14:37:13 -06:00
Jeff Epler 70719597ab supervisor: tick: Rewrite without atomics 2019-11-20 14:37:13 -06:00
Jeff Epler 568636d562 run_background_tasks: Do nothing unless there has been a tick
This improves performance of running python code by 34%, based
on the "pystone" benchmark on metro m4 express at 5000 passes
(1127.65 -> 1521.6 passes/second).

In addition, by instrumenting the tick function and monitoring on an
oscilloscope, the time actually spent in run_background_tasks() on
the metro m4 decreases from average 43% to 0.5%. (however, there's
some additional overhead that is moved around and not accounted for
in that "0.5%" figure, each time supervisor_run_background_tasks_if_tick
is called but no tick has occurred)

On the CPB, it increases pystone from 633 to 769, a smaller percentage
increase of 21%.  I did not measure the time actually spent in
run_background_tasks() on CPB.

Testing performed: on metro m4 and cpb, run pystone adapted from python3.4
(change time.time to time.monotonic for sub-second resolution)

Besides running a 5000 pass test, I also ran a 50-pass test while
scoping how long an output pin was set.  Average: 34.59ms or 1445/s on m4,
67.61ms or 739/s on cbp, both matching the other pystone result reasonably
well.

import pystone
import board
import digitalio
import time

d = digitalio.DigitalInOut(board.D13)
d.direction = digitalio.Direction.OUTPUT

while True:
    d.value = 0
    time.sleep(.01)
    d.value = 1
    pystone.main(50)
2019-11-18 11:26:48 -06:00
Jeff Epler 7f744a2369 Supervisor: move most of systick to the supervisor
This code is shared by most parts, except where not all the #ifdefs
inside the tick function were present in all ports.  This mostly would
have broken gamepad tick support on non-samd ports.

The "ms32" and "ms64" variants of the tick functions are introduced
because there is no 64-bit atomic read.  Disabling interrupts avoids
a low probability bug where milliseconds could be off by ~49.5 days
once every ~49.5 days (2^32 ms).

Avoiding disabling interrupts when only the low 32 bits are needed is a minor
optimization.

Testing performed: on metro m4 express, USB still works and
time.monotonic_ns() still counts up
2019-11-18 11:01:23 -06:00