This significantly speeds up readline on files opened directly from an
mpremote mount.
Signed-off-by: Andrew Leech <andrew.leech@planetinnovation.com.au>
The mantissa parsing code uses a floating point variable to accumulate
digits. Using an `mp_float_uint_t` variable instead and casting to
`mp_float_t` at the very end reduces code size. In some cases, it also
improves the rounding behaviour as extra digits are taken into account
by the int-to-float conversion code.
An extra test case handles the special case where mantissa overflow occurs
while processing deferred trailing zeros.
Signed-off-by: Yoctopuce dev <dev@yoctopuce.com>
This commit changes the Xtensa inline assembly parser to use a slightly
simpler (and probably a tiny bit more efficient) way to look up register
names when decoding instruction parameters.
Signed-off-by: Alessandro Gatti <a.gatti@frob.it>
This commit introduces a few changes aimed at reducing the amount of
space taken by the inline assembler once compiled:
* The register string table uses 2 bytes for each qstr rather than the
usual 4
* The opcode table uses 2 bytes for each qstr rather than the usual 4
* Opcode masks are not embedded in each opcode entry but looked up via
an additional smaller table, reducing the number of bytes taken by
an opcode's masks from 12 to 2 (with a fixed overhead of 24 bytes for
the the masks themselves stored elsewhere)
* Some error messages had a trailing period, now removed
* Error messages have been parameterised when possible, and the overall
text length is smaller.
Signed-off-by: Alessandro Gatti <a.gatti@frob.it>
This commit fixes a compilation warning (turned error) about a
potentially uninitialised variable being used. The warning can be
ignored as the variable in question is always written to, but the code
has been changed to silence that warning.
Signed-off-by: Alessandro Gatti <a.gatti@frob.it>
Testing with ROMFS shows that it is relatively easy to end up with a
corrupt filesystem on the device -- eg due to the ROMFS deploy process
stopping half way through -- which could lead to hard crashes. Notably,
there can be boot loops trying to mount a corrupt filesystem, crashes when
importing modules like `os` that first scan the filesystem for `os.py`, and
crashing when deploying a new ROMFS in certain cases because the old one is
removed while still mounted.
The main problem is that `mp_decode_uint()` has an loop that keeps going as
long as it reads 0xff byte values, which can happen in the case of erased
and unwritten flash.
This commit adds full bounds checking in the new `mp_decode_uint_checked()`
function, and that makes all ROMFS filesystem accesses robust.
Signed-off-by: Damien George <damien@micropython.org>
60 seconds is long enough that the USB serial connection drops out before
it times out (at least on my computer).
Also refactor out the timeout argument from sdcard_wait_finished, to try
and save a little code size.
This work was funded through GitHub Sponsors.
Signed-off-by: Angus Gratton <angus@redyak.com.au>
Manifests as `readblocks(-1, buf)` failing. The ST HAL does a bounds
check, but it checks `(block_num + num_blocks)` is within bounds, so if
these values overflow then it allows the read which seems to hang some SD
Cards (but not all).
Fix by explicitly testing for overflow in our layer of the driver.
This work was funded through GitHub Sponsors.
Signed-off-by: Angus Gratton <angus@redyak.com.au>
This commit upgrades from codespell==2.2.6 to the current codespell==2.4.1,
adding emac to the ignore-words-list.
Signed-off-by: Christian Clauss <cclauss@me.com>
Changing runner OS can change Python version, and ESP-IDF installs are
keyed on ESP-IDF and Python version together.
Signed-off-by: Angus Gratton <angus@redyak.com.au>
URLs in `package.json` may now be specified relative to the base URL of the
`package.json` file.
Relative URLs wil work for `package.json` files installed from the web as
well as local file paths.
Docs: update `docs/reference/packages.rst` to add documentation for:
- Installing packages from local filesystems (PR #12476); and
- Using relative URLs in the `package.json` file (PR #12477);
- Update the packaging example to encourage relative URLs as the default
in `package.json`.
Add `tools/mpremote/tests/test_mip_local_install.sh` to test the
installation of a package from local files using relative URLs in the
`package.json`.
Signed-off-by: Glenn Moloney <glenn.moloney@gmail.com>
It was set to 133Mhz, but that is not stable. Reduce to 100MHz.
The UF2 bootloader runs at 100MHz, so no need for a change of the
bootloader.
Signed-off-by: robert-hh <robert@hammelrath.com>
This commit removes "memory.h" from the ESP32 port tree, as it is no
longer needed with recent ESP-IDF versions.
Signed-off-by: Alessandro Gatti <a.gatti@frob.it>
This works similarly to the existing support in "bare metal" make ports,
with the caveat that CMake will only set this value on a clean build and
will reuse the previous value otherwise.
This is slightly different to the CMake built-in support for CFLAGS,
as this variable is used when evaluating source files for qstr
generation, etc.
This work was funded through GitHub Sponsors.
Signed-off-by: Angus Gratton <angus@redyak.com.au>
Regression in 3af006ef meant that pendsv.c no longer compiled if threads
were disabled in the build config. Add an implementation based on the
earlier one (simple counter) for the non-threads case.
It seems like with the current usage patterns there's no need for the
counter to be incremented/decremented atomically on a single core config.
This work was funded through GitHub Sponsors.
Signed-off-by: Angus Gratton <angus@redyak.com.au>
This adds a multi-test for DTLS server and client behaviour. It works on
all ports that enable this feature (eg unix, esp32, rp2, stm32), but
bare-metal ports that use lwIP are not reliable as the DTLS server because
the lwIP bindings only support queuing one UDP packet at a time (that needs
to be fixed).
Also, to properly implement a DTLS server sockets need to support
`socket.recvfrom(n, MSG_PEEK)`. That can be implemented in the future.
Signed-off-by: Damien George <damien@micropython.org>
This commit enables support for DTLS, i.e. TLS over datagram transport
protocols like UDP. While support for DTLS is absent in CPython, it is
worth supporting it in MicroPython because it is the basis of the
ubiquitous CoAP protocol, used in many IoT projects.
To select DTLS, a new set of "protocols" are added to SSLContext:
- ssl.PROTOCOL_DTLS_CLIENT
- ssl.PROTOCOL_DTLS_SERVER
If one of these is set, the library assumes that the underlying socket is a
datagram-like socket (i.e. UDP or similar).
Our own timer callbacks are implemented because the out of the box
implementation relies on `gettimeofday()`.
This new DTLS feature is enabled on all ports that use mbedTLS.
This commit is an update to a previous PR #10062.
Addresses issue #5270 which requested DTLS support.
Signed-off-by: Keenan Johnson <keenan.johnson@gmail.com>
Despite the code comments claiming one is sufficient, the mDNS application
is capable of using up to twelve timers. Three per IP protocol are started
at once in `mdns_start_multicast_timeouts_ipvX`, then another two per
protocol can be started in `mdns_handle_question`. Further timers can be
started for two additional callbacks.
Having certain timers, such as `MDNS_MULTICAST_TIMEOUT`, fail to start due
to none being free will break mDNS forever as the app will never realize
it's safe to transmit a packet. Therefore, this commit goes somewhat
overkill and allocates the maximal amount of timers; it's uncertain if all
can run simultaneously, or how many callback timers are needed.
Each timer struct is 16 bytes on standard 32 bit builds. Plus, say, 8
bytes of allocater overhead, that's 288 more bytes of RAM used which
shouldn't be too horrible. Users who don't need mDNS can manually disable
it to recover the RAM if necessary.
This fixes mDNS on W5500_EVB_PICO (among other boards). Before, mDNS would
work for a bit after connection until the host's cache expired a minute or
two later. Then the board would never respond to further queries. With
this patch, all works well.
Signed-off-by: Thomas Watson <twatson52@icloud.com>
The linker scripts for most of these microcontrollers contain a
non-contiguous flash section for the ID code that results in big binary
files, which exceed the flash size. This commit removes the ID code
section from the main firmware binary, and outputs it to a separate binary,
which can be deployed manually if ID code is enabled.
Signed-off-by: iabdalkader <i.abdalkader@gmail.com>
Functions would return NULL instead of `mp_const_false` if failed to init.
This work was funded through GitHub Sponsors.
Signed-off-by: Angus Gratton <angus@redyak.com.au>
The current code evaluates `pyb.RTC().datetime()` resulting in a remote
side exception, as `pyb` is not defined on most ports (only stm32).
The code should evaluate `machine.RTC().datetime()` and hence return the
current time.
Signed-off-by: rufusclark <50201718+rufusclark@users.noreply.github.com>
Signed-off-by: Damien George <damien@micropython.org>
It's not necessary to support this, which allows an arbitrary memory
address to be specified and potentially allows invalid memory accesses.
Requiring an object with the buffer protocol is safer, and also means that
the length of the region is always specified.
Signed-off-by: Damien George <damien@micropython.org>
This commit implements a small subset of the CPython `marshal` module. It
implements `marshal.dumps()` and `marshal.loads()`, but only supports
(un)marshalling code objects at this stage. The semantics match CPython,
except that the actual marshalled bytes is not compatible with CPython's
marshalled bytes.
The module is enabled at the everything level (only on the unix coverage
build at this stage).
Signed-off-by: Damien George <damien@micropython.org>
This allows retrieving the code object of a function using
`function.__code__`, and then reconstructing a function from a code object
using `FunctionType(code_object)`.
This feature is controlled by `MICROPY_PY_FUNCTION_ATTRS_CODE` and is
enabled at the full-features level.
Signed-off-by: Damien George <damien@micropython.org>
The `mp_obj_code_t` and `mp_type_code` code object was defined internally
in both `py/builtinevex.c` and `py/profile.c`, with completely different
implementations (the former very minimal, the latter quite complete).
This commit factors these implementations into a new, separate source file,
and allows the code object to have four different modes, selected at
compile-time:
- MICROPY_PY_BUILTINS_CODE_NONE: code object not included in the build.
- MICROPY_PY_BUILTINS_CODE_MINIMUM: very simple code object that just holds
a reference to the function that it represents. This level is used when
MICROPY_PY_BUILTINS_COMPILE is enabled.
- MICROPY_PY_BUILTINS_CODE_BASIC: simple code object that holds a reference
to the proto-function and its constants.
- MICROPY_PY_BUILTINS_CODE_FULL: almost complete implementation of the code
object. This level is used when MICROPY_PY_SYS_SETTRACE is enabled.
Signed-off-by: Damien George <damien@micropython.org>
This commit adds a clarification for the ESPNow module's documentation
regarding its "config" method.
The original documentation for that method could be interpreted as having
all its configuration keys being able to be queried, but the "rate"
configuration key is actually write-only due to ESP-IDF's lack of a way to
retrieve that bit of information from the radio's configuration. The
documentation changes highlight the fact that said configuration key is
actually write-only.
Signed-off-by: Alessandro Gatti <a.gatti@frob.it>
Also add some additional context links, suggestions for alternative
classes, etc.
This work was funded through GitHub Sponsors.
Signed-off-by: Angus Gratton <angus@redyak.com.au>
Also add the default values of these macros to the respective
`mpconfigport.h` files, to improve discoverability.
Signed-off-by: Angus Gratton <angus@redyak.com.au>
Also cross-link with the other WIZNET5K driver, to avoid confusion.
This work was funded through GitHub Sponsors.
Signed-off-by: Angus Gratton <angus@redyak.com.au>
The `rp2` package use a global dict `_pio_funcs` to populate a namespace
for `@asm_pio` functions to be executed in. That dict is not cleaned up
after use, keeping references to bound methods of a `PIOASMEmit`. By not
setting/clearing all the functions, `asm_pio_encode` unintentionally allows
the use of the old directives (harmless) as well as `jmp` (in general,
produces the wrong output).
Fix that by making sure `_pio_funcs` is returned to its original state
after using it:
- For `@asm_pio` update the target dict from `_pio_funcs` and then set
additional functions as needed, leaving `_pio_funcs` unchanged.
- For `asm_pio_encode`, borrow `_pio_funcs` to use as globals (avoiding a
bunch of memory alloc/free) but delete the instruction entries after use.
Signed-off-by: Neil Ludban <neil.ludban@gmail.com>
Add support for `mpremote mip install package.json` where `package.json` is
a json file on the local filesystem.
Without this, package json files can only be loaded from http, https,
github or gitlab URLs.
This is useful for testing `package.json` files for pacages in development
and for constructing one's own `package.json` files for Python packages
which are not yet available for installation using mip.
Signed-off-by: Glenn Moloney <glenn.moloney@gmail.com>
Add MSC support using internal flash storage or SD card.
Note this is disabled by default, and can be enabled by boards if needed.
Signed-off-by: iabdalkader <i.abdalkader@gmail.com>
The flash clock frequency may have been set to a different value by a
bootloader. Set the frequency according to the configured value. Use a
table of pre-calculated dividers to get the closest value for the flash
frequency, achieving for MIMXRT10xx:
30 -> 30.85 MHz
50 -> 49.65 MHz
60 -> 60 MHz
75 -> 75.13 MHz
80 -> 80 MHz
100 -> 99.31 Mhz
133 -> 132.92 MHz
166 -> 166.15 MHz
for MIMXRT1176:
30 -> 31 MHz
50 -> 52.8 MJz
60 -> 58.7 MHz
75 -> 75.4 MHz
80 -> 75.4 MHz
100 -> 105.6 MHz
133 -> 132 MHz
166 -> 176 MHz
Signed-off-by: robert-hh <robert@hammelrath.com>
And use these to initialize the LUT table properly for the various flash
types. The different flash types differ by 3 parameters. Thus it is
easier to just keep one copy of the qspiflash_config structure with the LUT
table and update it during flash initialisation as needed.
Signed-off-by: robert-hh <robert@hammelrath.com>
