Commit Graph

35 Commits

Author SHA1 Message Date
Damien George
e52758da22 Merge tag 'v1.8.1' into parse-bytecode
Many ESP8266 improvements, enhanced WebREPL, and support for STM32L4 MCUs

This release brings general improvements and bug fixes, and some new
features.  There is now a uerror module for consistent errno handling
across ports, as well as textual names of OS errors that are printed when
an OSError is raised.  There is support for frozen packages, via both
frozen scripts and frozen bytecode.  WebREPL on the ESP8266 is greatly
improved with many bug fixes and now supports an unlimited (or very large)
number of reconnects.  The os module on the ESP8266 now has rename, chdir,
getcwd and stat.  The unix port now includes the ussl module by default.
The stmhal port has support for STM32L4 MCUs including the STM32L476
Discovery board and the LimiFrog board.

README:
- add explicit note that subdirs contain more READMEs
- add "make deplibs" to quick build section
- "quick build": Use "make axtls" after all

CODECONVENTIONS.md:
- describe git commit messages conventions

py core:
- obj: add warning note about get_array return value and GC blocks
- objstr: binary type of str/bytes for buffer protocol is 'B'
- runtime: properly handle passing user mappings to ** keyword args
- repl: if there're no better alternatives, try to complete "import"
- mpz: fix bug with overflowing C-shift in division routine
- mpz: do Python style division/modulo within bignum divmod routine
- mpz: fix mpn_div so that it doesn't modify memory of denominator
- vstr: vstr_null_terminated_str(): Extend string by at most one byte
- vstr: change allocation policy, +16 to requested size, instead of *2
- add mperrno.h file with uPy defined errno constants
- add uerrno module, with errno constants and dict
- parse: add uerrno to list of modules to look for constants in
- mperrno: add EAFNOSUPPORT definition
- repl: fix handling of backslash in quotes when checking continuation
- gc: gc_dump_alloc_table(): Show byte/str and (byte)array objects
- gc: make (byte)array type dumping conditional on these types being enabled
- gc: use '=' char for tail blocks when dumping heap
- mperrno: add some more MP_Exxx constants, related to networking
- moduerrno: add more constants to the errno module
- add mp_errno_to_str() and use it to provide nicer OSError msgs
- objfloat, py/modmath: ensure M_PI and M_E defined
- emitglue: fix build on AArch64 (ARMv8, etc.) related to loading .mpy files
- objexcept: don't convert errno to str in constructor, do it in print
- moduerrno: add EACCES, pretty common error on Unix
- gc: gc_dump_alloc_table(): dump heap offset instead of actual address
- objstr: make dedicated splitlines function, supporting diff newlines
- objstringio: add TODO comment about avoiding copying on .getvalue()
- modstruct: raise ValueError on unsupported format char
- stream: support both "exact size" and "one underlying call" operations
- declare constant data as properly constant
- stream: add mp_stream_close() helper function
- mphal.h: provide default prototypes for mp_hal_delay_us/mp_hal_ticks_us
- rework frozen modules support to support packages
- objstr: implement str.center()
- allow to stat and import frozen mpy files using new frozen "VFS"
- makeqstrdata.py: allow to have double-quote characters in qstrs
- objnamedtuple: allow passing field names as a tuple
- moduerrno: add EEXIST, EISDIR, ECONNREFUSED
- modstruct: allow to have "0s" in struct format

extmod:
- modlwip: convert errno's to use MP_Exxx symbols
- modlwip: rework how Python accept callback is called
- when including extmod headers, prefix path with extmod/
- modwebsocket: add close() method
- modwebrepl: add close() method
- moduos_dupterm: dumpterm subsystem is responsible for closing stream
- modussl: make more compatible with non-default obj representations
- machine_i2c: redo mp_hal_pin macros to use open_drain and od_low
- virtpin: initial implementation of open-ended C-level Pin interface
- vfs_fat: replace text error messages by POSIX error numbers
- vfs_fat: add chdir() method
- vfs_fat: add getcwd() method
- vfs_fat: add vfs.stat() method
- add machine time_pulse_us function (at C and Python level)
- machine: add MICROPY_PY_MACHINE_PULSE config for time_pulse_us

lib:
- timeutils/timeutils: timeutils_mktime may accept negative time values

drivers:
- cc3000: rename timeval to cc3000_timeval, to avoid clash
- add C-level function to read DHT11 and DHT22 devices
- display/ssd1306: update SSD1306_SPI to work with new API

tools:
- make-frozen.py: properly escape hex chars when making C strings
- make-frozen.py: quick fix to support package-modules
- mpy-tool.py: add checks for critical configuration vars
- make-frozen.py: update for latest changes in frozen modules support
- mpy-tool.py: include .py extension in frozen filename
- mpy-tool.py: don't strip directories from the frozen source name
- upgrade upip to 0.7: SSL cert warning, use uerrno, better usage message

tests:
- run-tests: factor out list of supported external boards
- disable memoryview tests that overflow int conversion
- basics/string_splitlines: reinstate feature test for splitlines
- struct1: add testcase for an unknown type char
- add testcase for str.center()
- extmod/vfs_fat_ramdisk: add testcases for chdir(), getcwd()
- misc/recursive_iternext.py: increase depth N from 1000 to 2000
- misc/recursive_iternext.py: provide more fine-grained selection of N
- pyb/rtc: make RTC test on pyboard more reliable by calling init()

mpy-cross:
- add -s option to specify the embedded source filename

unix port:
- mphalport: add mp_hal_delay_us() for consistency with other ports
- enable uerrno module
- add ability to include frozen bytecode in the build
- mpconfigport_coverage.h: add dedicated config file for coverage build
- unix_mphal: implement mp_hal_ticks_us()
- support frozen packages
- Makefile: nanbox build is not compatible with modussl, disable
- enable "ussl" module
- mpconfigport.mk: document MICROPY_STANDALONE make-level option
- Makefile: "make axtls": automatically fetch submodules if missing

windows port:
- enable multi-processor compilation for msvc

stmhal port:
- l4: adapt DMA to be able to support STM32L4 MCU series
- l4: adapt startup code, clock configuration and interrupts
- l4: make CCM/DTCM RAM start-up conditional on MCU type
- l4: add support for machine.sleep on STM32L4 MCUs
- dma: make DAC DMA descriptors conditional on having a DAC
- add board files for LIMIFROG board
- for LIMIFROG board, add early-init function to get to DFU mode
- dma: fix builds for boards with an F4 or F7 but no DAC
- sdcard: fix initialisation of DMA TX so that writes work
- can: allow to get existing CAN obj if constructed without args
- fix clock configuration for STM32L476-discovery; also add I2C2
- convert to use internal errno symbols; enable uerrno module
- for network drivers, convert to use MP_Exxx errno symbols
- led: allow LEDs to be in PWM mode with TIM1 and channels 1-4
- i2c: expose I2CHandle3 for use by custom C code
- sdcard: allow to do unaligned read-from/write-to SD card
- support frozen packages using .mpy files
- moduos: getcwd(): use mp_obj_new_exception_arg1()
- dac: add DAC deinit() method
- uart: fix wrong baudrate calculation for stm32l4 series

esp8266 port:
- scripts/: remove use of pin.PULL_NONE
- scripts/inisetup: don't start WebREPL on boot in master branch
- scripts/: add fill() to NeoPixel
- scripts/webrepl: add optional password argument to webrepl.start()
- scripts/webrepl: add start_foreground() method
- main: bump heap size to 28K
- mpconfigport: reduce various parser-related allocation params
- help: add "sta_if.active(True)" command
- convert to use new MP_Exxx errno symbols
- enable uerrno module, weak linked also as errno
- change to use internal errno's
- moduos.c: addition of the rename method to module uos
- scripts/port_diag: add network diagnostic output
- scripts/webrepl_setup: show password placeholder char
- scripts/webrepl_setup: add max password length check
- README: add a very first start section
- add APA102 serial individually controllable LEDs support
- enable collections.OrderedDict
- main: update _boot module loading for recent frozen modules refactors
- scripts/port_diag: dump network interface IP settings
- esp_mphal: fix NLR buffer leak in call_dupterm_read()
- esp_mphal: handle Ctrl+C from dupterm (e.g. WebREPL)
- esp_mphal: mp_uos_dupterm_deactivate() may raise exception
- add mp_hal_pin_input() and mp_hal_pin_output() functions
- modpybspi: configure pins when initialising an SPI object
- xtirq: add xtirq.h for controlling xtensa irqs
- ets_alt_task: don't run ets_loop_iter if irqs are disabled
- modmachine: add disable_irq and enable_irq functions
- enable DHT C-level driver
- add dht.py script for high-level control of DHT11/DHT22 sensor
- Makefile: document "disable" value for UART_OS
- modnetwork: scan() is only supported by STA when it's enabled
- modnetwork: protect scan() callback against memory errors
- modnetwork: allow to press ctrl-C while scan() is running
- uart: properly initialise UART0 RXD pin in uart_config
- moduos: add chdir() and getcwd() functions
- scripts/ntptime: allow to override NTP server
- modmachine: add machine.time_pulse_us function
- enable MICROPY_PY_IO_FILEIO to get compliant text/binary streams
- moduos.c: add stat() to the module uos of esp8266
- rtc: set RTC user memory length to 0 on first boot
- provide a dedicated variable to disable ets_loop_iter
- modpybrtc: handle RTC overflow

docs:
- machine.UART: filter out unimplemented UART methods from esp8266 docs
- esp8266/quickref: new way to get MAC address
- esp8266/quickstart: remove i2c examples with stop=False
- ustruct: describe supported type codes
- ussl: add basic description of axTLS-based modussl
- esp8266: Include ussl module in the docs
- machine: make disable_irq and enable_irq docs available for all
- library/machine: add documentation for machine.time_pulse_us
- math, cmath: add port availability information
- library/index: add intro paragraph regarding availability of modules
- README: add some hints for PDF docs generation
- wipy/tutorial: add note about screen key bindings on OS X
- esp8266/quickref: update WebREPL section for 1.8.1 release
- esp8266: fix ESP8266 Network tutorial
- esp8266/quickref: use local image of Adafruit Huzzah board
- esp8266/general: add note about RTC overflow

travis:
- install gcc-arm-none-eabi with --force-yes for now
2017-08-13 21:33:40 +10:00
Paul Sokolovsky
6f34e138f1 py/vstr: Change allocation policy, +16 to requested size, instead of *2.
Effect measured on esp8266 port:

Before:
>>> pystone_lowmem.main(10000)
Pystone(1.2) time for 10000 passes = 44214 ms
This machine benchmarks at 226 pystones/second
>>> pystone_lowmem.main(10000)
Pystone(1.2) time for 10000 passes = 44246 ms
This machine benchmarks at 226 pystones/second

After:
>>> pystone_lowmem.main(10000)
Pystone(1.2) time for 10000 passes = 44343ms
This machine benchmarks at 225 pystones/second
>>> pystone_lowmem.main(10000)
Pystone(1.2) time for 10000 passes = 44376ms
This machine benchmarks at 225 pystones/second
2016-05-10 00:56:51 +03:00
Paul Sokolovsky
a1f2245a81 py/vstr: vstr_null_terminated_str(): Extend string by at most one byte.
vstr_null_terminated_str is almost certainly a vstr finalization operation,
so it should add the requested NUL byte, and not try to pre-allocate more.
The previous implementation could actually allocate double of the buffer
size.
2016-05-09 22:39:57 +03:00
Damien George
2d9531a777 py/vstr: Change allocation policy from *2 to +64. 2016-04-18 15:09:34 +01:00
Damien George
5bf649f370 py: Expose vstr_in_blank_bytes as public. 2016-04-18 15:09:34 +01:00
Dave Hylands
9f76dcd682 py: Prevent many extra vstr allocations.
I checked the entire codebase, and every place that vstr_init_len
was called, there was a call to mp_obj_new_str_from_vstr after it.

mp_obj_new_str_from_vstr always tries to reallocate a new buffer
1 byte larger than the original to store the terminating null
character.

In many cases, if we allocated the initial buffer to be 1 byte
longer, we can prevent this extra allocation, and just reuse
the originally allocated buffer.

Asking to read 256 bytes and only getting 100 will still cause
the extra allocation, but if you ask to read 256 and get 256
then the extra allocation will be optimized away.

Yes - the reallocation is optimized in the heap to try and reuse
the buffer if it can, but it takes quite a few cycles to figure
this out.

Note by Damien: vstr_init_len should now be considered as a
string-init convenience function and used only when creating
null-terminated objects.
2015-07-06 17:29:27 +01:00
Damien George
1c9a499135 py/vstr.c: Allow vstr_printf to print correctly to a fixed buffer.
This patch allows vstr_printf to use all the available space of a fixed
vstr buffer.  vstr_printf is a good alternative to snprintf.
2015-04-18 14:23:13 +01:00
Damien George
7f9d1d6ab9 py: Overhaul and simplify printf/pfenv mechanism.
Previous to this patch the printing mechanism was a bit of a tangled
mess.  This patch attempts to consolidate printing into one interface.

All (non-debug) printing now uses the mp_print* family of functions,
mainly mp_printf.  All these functions take an mp_print_t structure as
their first argument, and this structure defines the printing backend
through the "print_strn" function of said structure.

Printing from the uPy core can reach the platform-defined print code via
two paths: either through mp_sys_stdout_obj (defined pert port) in
conjunction with mp_stream_write; or through the mp_plat_print structure
which uses the MP_PLAT_PRINT_STRN macro to define how string are printed
on the platform.  The former is only used when MICROPY_PY_IO is defined.

With this new scheme printing is generally more efficient (less layers
to go through, less arguments to pass), and, given an mp_print_t*
structure, one can call mp_print_str for efficiency instead of
mp_printf("%s", ...).  Code size is also reduced by around 200 bytes on
Thumb2 archs.
2015-04-16 14:30:16 +00:00
Damien George
827b0f747b py: Change vstr_null_terminate -> vstr_null_terminated_str, returns str. 2015-01-29 13:57:23 +00:00
Damien George
0d3cb6726d py: Change vstr so that it doesn't null terminate buffer by default.
This cleans up vstr so that it's a pure "variable buffer", and the user
can decide whether they need to add a terminating null byte.  In most
places where vstr is used, the vstr did not need to be null terminated
and so this patch saves code size, a tiny bit of RAM, and makes vstr
usage more efficient.  When null termination is needed it must be
done explicitly using vstr_null_terminate.
2015-01-28 23:43:01 +00:00
Damien George
77089bebd4 py: Add comments for vstr_init and mp_obj_new_str. 2015-01-21 23:18:02 +00:00
Damien George
05005f679e py: Remove mp_obj_str_builder and use vstr instead.
With this patch str/bytes construction is streamlined.  Always use a
vstr to build a str/bytes object.  If the size is known beforehand then
use vstr_init_len to allocate only required memory.  Otherwise use
vstr_init and the vstr will grow as needed.  Then use
mp_obj_new_str_from_vstr to create a str/bytes object using the vstr
memory.

Saves code ROM: 68 bytes on stmhal, 108 bytes on bare-arm, and 336 bytes
on unix x64.
2015-01-21 23:18:02 +00:00
Damien George
0b9ee86133 py: Add mp_obj_new_str_from_vstr, and use it where relevant.
This patch allows to reuse vstr memory when creating str/bytes object.
This improves memory usage.

Also saves code ROM: 128 bytes on stmhal, 92 bytes on bare-arm, and 88
bytes on unix x64.
2015-01-21 23:17:27 +00:00
Damien George
963a5a3e82 py, unix: Allow to compile with -Wsign-compare.
See issue #699.
2015-01-16 17:47:07 +00:00
Damien George
51dfcb4bb7 py: Move to guarded includes, everywhere in py/ core.
Addresses issue #1022.
2015-01-01 20:32:09 +00:00
Damien George
969a6b37bf py: Make functions static where appropriate. 2014-12-10 22:08:14 +00:00
Damien George
42f3de924b py: Convert [u]int to mp_[u]int_t where appropriate.
Addressing issue #50.
2014-10-03 17:44:14 +00:00
Damien George
b0261341d3 py: For malloc and vstr functions, use size_t exclusively for int type.
It seems most sensible to use size_t for measuring "number of bytes" in
malloc and vstr functions (since that's what size_t is for).  We don't
use mp_uint_t because malloc and vstr are not Micro Python specific.
2014-09-25 15:49:26 +01:00
Paul Sokolovsky
165eb69b86 vstr: Restore bytestr compatibility. 2014-06-27 00:04:18 +03:00
Chris Angelico
2ba2299d28 lexer, vstr: Add unicode support. 2014-06-27 00:04:18 +03:00
Paul Sokolovsky
59c675a64c py: Include mpconfig.h before all other includes.
It defines types used by all other headers.

Fixes #691.
2014-06-21 22:43:22 +03:00
Damien George
04b9147e15 Add license header to (almost) all files.
Blanket wide to all .c and .h files.  Some files originating from ST are
difficult to deal with (license wise) so it was left out of those.

Also merged modpyb.h, modos.h, modstm.h and modtime.h in stmhal/.
2014-05-03 23:27:38 +01:00
Damien George
8cd72bdf92 py: Fix vstr_init for case that alloc = 0. 2014-03-31 17:10:59 +01:00
xbe
efe3422394 py: Clean up includes.
Remove unnecessary includes. Add includes that improve portability.
2014-03-17 02:43:40 -07:00
Damien George
ecd58aec08 py: Fix bug in vstr_ins_blank_bytes. 2014-03-15 16:54:06 +00:00
Damien George
280e7208d8 Add vstr_ins and vstr_cut_out; improve stmhal readline. 2014-03-15 14:33:09 +00:00
Paul Sokolovsky
520e2f58a5 Replace global "static" -> "STATIC", to allow "analysis builds". Part 2. 2014-02-12 18:31:30 +02:00
Damien George
354d15a964 Implement fixed buffer vstrs; use them for import path. 2014-02-06 21:11:19 +00:00
Damien George
f64086f80f Fix 1 warning and 1 bug. 2014-01-22 23:18:50 +00:00
Paul Sokolovsky
5d2499c638 Add "buffer management" and "shrink" API calls to vstr.
vstr is initially intended to deal with arbitrary-length strings. By
providing a bit lower-level API calls, it will be also useful to deal
with arbitrary-length I/O buffers (the difference from strings is that
buffers are filled from "outside", via I/O).

Another issue, especially aggravated by I/O buffer use, is alloc size
vs actual size length. If allocated 1Mb for buffer, but actually
read 1 byte, we don't want to keep rest of 1Mb be locked by this I/O
result, but rather return it to heap ASAP ("shrink" buffer before passing
it to qstr_from_str_take()).
2014-01-13 23:31:06 +02:00
Damien George
66028ab6dc Basic implementation of import.
import works for simple cases.  Still work to do on finding the right
script, and setting globals/locals correctly when running an imported
function.
2014-01-03 14:03:48 +00:00
Damien
732407f1bf Change memory allocation API to require size for free and realloc. 2013-12-29 19:33:23 +00:00
Damien
2f06c57f8a Add simple var-arg functions; add simple string.format. 2013-11-03 18:20:56 +00:00
Damien
8b3a7c2237 Fix func decls with no arguments: () -> (void). 2013-10-23 20:20:17 +01:00
Damien
68f59a96e3 Add vstr and its functions. 2013-10-20 14:39:58 +01:00