circuitpython/mpy-cross/main.c

356 lines
14 KiB
C
Raw Normal View History

/*
* This file is part of the MicroPython project, http://micropython.org/
*
* The MIT License (MIT)
*
* Copyright (c) 2013-2016 Damien P. George
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
* THE SOFTWARE.
*/
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <unistd.h>
#include "py/builtin.h"
#include "py/compile.h"
#include "py/persistentcode.h"
#include "py/runtime.h"
#include "py/gc.h"
#include "py/stackctrl.h"
#include "genhdr/mpversion.h"
#ifdef _WIN32
#include "fmode.h"
#endif
// Command line options, with their defaults
STATIC uint emit_opt = MP_EMIT_OPT_NONE;
mp_uint_t mp_verbose_flag = 0;
// Heap size of GC heap (if enabled)
// Make it larger on a 64 bit machine, because pointers are larger.
2021-03-15 09:57:36 -04:00
long heap_size = 1024 * 1024 * (sizeof(mp_uint_t) / 4);
STATIC void stderr_print_strn(void *env, const char *str, size_t len) {
(void)env;
ssize_t dummy = write(STDERR_FILENO, str, len);
(void)dummy;
}
STATIC const mp_print_t mp_stderr_print = {NULL, stderr_print_strn};
STATIC int compile_and_save(const char *file, const char *output_file, const char *source_file) {
nlr_buf_t nlr;
if (nlr_push(&nlr) == 0) {
mp_lexer_t *lex = mp_lexer_new_from_file(file);
qstr source_name;
if (source_file == NULL) {
source_name = lex->source_name;
} else {
source_name = qstr_from_str(source_file);
}
#if MICROPY_PY___FILE__
mp_store_global(MP_QSTR___file__, MP_OBJ_NEW_QSTR(source_name));
#endif
mp_parse_tree_t parse_tree = mp_parse(lex, MP_PARSE_FILE_INPUT);
mp_compiled_module_t cm;
cm.context = m_new_obj(mp_module_context_t);
mp_compile_to_raw_code(&parse_tree, source_name, false, &cm);
vstr_t vstr;
vstr_init(&vstr, 16);
if (output_file == NULL) {
vstr_add_str(&vstr, file);
vstr_cut_tail_bytes(&vstr, 2);
vstr_add_str(&vstr, "mpy");
} else {
vstr_add_str(&vstr, output_file);
}
py: Rework bytecode and .mpy file format to be mostly static data. Background: .mpy files are precompiled .py files, built using mpy-cross, that contain compiled bytecode functions (and can also contain machine code). The benefit of using an .mpy file over a .py file is that they are faster to import and take less memory when importing. They are also smaller on disk. But the real benefit of .mpy files comes when they are frozen into the firmware. This is done by loading the .mpy file during compilation of the firmware and turning it into a set of big C data structures (the job of mpy-tool.py), which are then compiled and downloaded into the ROM of a device. These C data structures can be executed in-place, ie directly from ROM. This makes importing even faster because there is very little to do, and also means such frozen modules take up much less RAM (because their bytecode stays in ROM). The downside of frozen code is that it requires recompiling and reflashing the entire firmware. This can be a big barrier to entry, slows down development time, and makes it harder to do OTA updates of frozen code (because the whole firmware must be updated). This commit attempts to solve this problem by providing a solution that sits between loading .mpy files into RAM and freezing them into the firmware. The .mpy file format has been reworked so that it consists of data and bytecode which is mostly static and ready to run in-place. If these new .mpy files are located in flash/ROM which is memory addressable, the .mpy file can be executed (mostly) in-place. With this approach there is still a small amount of unpacking and linking of the .mpy file that needs to be done when it's imported, but it's still much better than loading an .mpy from disk into RAM (although not as good as freezing .mpy files into the firmware). The main trick to make static .mpy files is to adjust the bytecode so any qstrs that it references now go through a lookup table to convert from local qstr number in the module to global qstr number in the firmware. That means the bytecode does not need linking/rewriting of qstrs when it's loaded. Instead only a small qstr table needs to be built (and put in RAM) at import time. This means the bytecode itself is static/constant and can be used directly if it's in addressable memory. Also the qstr string data in the .mpy file, and some constant object data, can be used directly. Note that the qstr table is global to the module (ie not per function). In more detail, in the VM what used to be (schematically): qst = DECODE_QSTR_VALUE; is now (schematically): idx = DECODE_QSTR_INDEX; qst = qstr_table[idx]; That allows the bytecode to be fixed at compile time and not need relinking/rewriting of the qstr values. Only qstr_table needs to be linked when the .mpy is loaded. Incidentally, this helps to reduce the size of bytecode because what used to be 2-byte qstr values in the bytecode are now (mostly) 1-byte indices. If the module uses the same qstr more than two times then the bytecode is smaller than before. The following changes are measured for this commit compared to the previous (the baseline): - average 7%-9% reduction in size of .mpy files - frozen code size is reduced by about 5%-7% - importing .py files uses about 5% less RAM in total - importing .mpy files uses about 4% less RAM in total - importing .py and .mpy files takes about the same time as before The qstr indirection in the bytecode has only a small impact on VM performance. For stm32 on PYBv1.0 the performance change of this commit is: diff of scores (higher is better) N=100 M=100 baseline -> this-commit diff diff% (error%) bm_chaos.py 371.07 -> 357.39 : -13.68 = -3.687% (+/-0.02%) bm_fannkuch.py 78.72 -> 77.49 : -1.23 = -1.563% (+/-0.01%) bm_fft.py 2591.73 -> 2539.28 : -52.45 = -2.024% (+/-0.00%) bm_float.py 6034.93 -> 5908.30 : -126.63 = -2.098% (+/-0.01%) bm_hexiom.py 48.96 -> 47.93 : -1.03 = -2.104% (+/-0.00%) bm_nqueens.py 4510.63 -> 4459.94 : -50.69 = -1.124% (+/-0.00%) bm_pidigits.py 650.28 -> 644.96 : -5.32 = -0.818% (+/-0.23%) core_import_mpy_multi.py 564.77 -> 581.49 : +16.72 = +2.960% (+/-0.01%) core_import_mpy_single.py 68.67 -> 67.16 : -1.51 = -2.199% (+/-0.01%) core_qstr.py 64.16 -> 64.12 : -0.04 = -0.062% (+/-0.00%) core_yield_from.py 362.58 -> 354.50 : -8.08 = -2.228% (+/-0.00%) misc_aes.py 429.69 -> 405.59 : -24.10 = -5.609% (+/-0.01%) misc_mandel.py 3485.13 -> 3416.51 : -68.62 = -1.969% (+/-0.00%) misc_pystone.py 2496.53 -> 2405.56 : -90.97 = -3.644% (+/-0.01%) misc_raytrace.py 381.47 -> 374.01 : -7.46 = -1.956% (+/-0.01%) viper_call0.py 576.73 -> 572.49 : -4.24 = -0.735% (+/-0.04%) viper_call1a.py 550.37 -> 546.21 : -4.16 = -0.756% (+/-0.09%) viper_call1b.py 438.23 -> 435.68 : -2.55 = -0.582% (+/-0.06%) viper_call1c.py 442.84 -> 440.04 : -2.80 = -0.632% (+/-0.08%) viper_call2a.py 536.31 -> 532.35 : -3.96 = -0.738% (+/-0.06%) viper_call2b.py 382.34 -> 377.07 : -5.27 = -1.378% (+/-0.03%) And for unix on x64: diff of scores (higher is better) N=2000 M=2000 baseline -> this-commit diff diff% (error%) bm_chaos.py 13594.20 -> 13073.84 : -520.36 = -3.828% (+/-5.44%) bm_fannkuch.py 60.63 -> 59.58 : -1.05 = -1.732% (+/-3.01%) bm_fft.py 112009.15 -> 111603.32 : -405.83 = -0.362% (+/-4.03%) bm_float.py 246202.55 -> 247923.81 : +1721.26 = +0.699% (+/-2.79%) bm_hexiom.py 615.65 -> 617.21 : +1.56 = +0.253% (+/-1.64%) bm_nqueens.py 215807.95 -> 215600.96 : -206.99 = -0.096% (+/-3.52%) bm_pidigits.py 8246.74 -> 8422.82 : +176.08 = +2.135% (+/-3.64%) misc_aes.py 16133.00 -> 16452.74 : +319.74 = +1.982% (+/-1.50%) misc_mandel.py 128146.69 -> 130796.43 : +2649.74 = +2.068% (+/-3.18%) misc_pystone.py 83811.49 -> 83124.85 : -686.64 = -0.819% (+/-1.03%) misc_raytrace.py 21688.02 -> 21385.10 : -302.92 = -1.397% (+/-3.20%) The code size change is (firmware with a lot of frozen code benefits the most): bare-arm: +396 +0.697% minimal x86: +1595 +0.979% [incl +32(data)] unix x64: +2408 +0.470% [incl +800(data)] unix nanbox: +1396 +0.309% [incl -96(data)] stm32: -1256 -0.318% PYBV10 cc3200: +288 +0.157% esp8266: -260 -0.037% GENERIC esp32: -216 -0.014% GENERIC[incl -1072(data)] nrf: +116 +0.067% pca10040 rp2: -664 -0.135% PICO samd: +844 +0.607% ADAFRUIT_ITSYBITSY_M4_EXPRESS As part of this change the .mpy file format version is bumped to version 6. And mpy-tool.py has been improved to provide a good visualisation of the contents of .mpy files. In summary: this commit changes the bytecode to use qstr indirection, and reworks the .mpy file format to be simpler and allow .mpy files to be executed in-place. Performance is not impacted too much. Eventually it will be possible to store such .mpy files in a linear, read-only, memory- mappable filesystem so they can be executed from flash/ROM. This will essentially be able to replace frozen code for most applications. Signed-off-by: Damien George <damien@micropython.org>
2021-10-22 07:22:47 -04:00
mp_raw_code_save_file(&cm, vstr_null_terminated_str(&vstr));
vstr_clear(&vstr);
nlr_pop();
return 0;
} else {
// uncaught exception
mp_obj_print_exception(&mp_stderr_print, (mp_obj_t)nlr.ret_val);
return 1;
}
}
STATIC int usage(char **argv) {
printf(
"usage: %s [<opts>] [-X <implopt>] <input filename>\n"
"Options:\n"
"--version : show version information\n"
"-o : output file for compiled bytecode (defaults to input with .mpy extension)\n"
"-s : source filename to embed in the compiled bytecode (defaults to input file)\n"
"-v : verbose (trace various operations); can be multiple\n"
"-O[N] : apply bytecode optimizations of level N\n"
"\n"
"Target specific options:\n"
"-msmall-int-bits=number : set the maximum bits used to encode a small-int\n"
"-march=<arch> : set architecture for native emitter; x86, x64, armv6, armv6m, armv7m, armv7em, armv7emsp, armv7emdp, xtensa, xtensawin\n"
"\n"
"Implementation specific options:\n", argv[0]
);
int impl_opts_cnt = 0;
printf(
#if MICROPY_EMIT_NATIVE
2021-03-15 09:57:36 -04:00
" emit={bytecode,native,viper} -- set the default code emitter\n"
#else
" emit=bytecode -- set the default code emitter\n"
#endif
2021-03-15 09:57:36 -04:00
);
impl_opts_cnt++;
printf(
2021-03-15 09:57:36 -04:00
" heapsize=<n> -- set the heap size for the GC (default %ld)\n"
, heap_size);
impl_opts_cnt++;
if (impl_opts_cnt == 0) {
printf(" (none)\n");
}
return 1;
}
// Process options which set interpreter init options
STATIC void pre_process_options(int argc, char **argv) {
for (int a = 1; a < argc; a++) {
if (argv[a][0] == '-') {
if (strcmp(argv[a], "-X") == 0) {
if (a + 1 >= argc) {
exit(usage(argv));
}
if (strcmp(argv[a + 1], "emit=bytecode") == 0) {
emit_opt = MP_EMIT_OPT_BYTECODE;
#if MICROPY_EMIT_NATIVE
} else if (strcmp(argv[a + 1], "emit=native") == 0) {
emit_opt = MP_EMIT_OPT_NATIVE_PYTHON;
} else if (strcmp(argv[a + 1], "emit=viper") == 0) {
emit_opt = MP_EMIT_OPT_VIPER;
#endif
} else if (strncmp(argv[a + 1], "heapsize=", sizeof("heapsize=") - 1) == 0) {
char *end;
heap_size = strtol(argv[a + 1] + sizeof("heapsize=") - 1, &end, 0);
// Don't bring unneeded libc dependencies like tolower()
// If there's 'w' immediately after number, adjust it for
// target word size. Note that it should be *before* size
// suffix like K or M, to avoid confusion with kilowords,
// etc. the size is still in bytes, just can be adjusted
// for word size (taking 32bit as baseline).
bool word_adjust = false;
if ((*end | 0x20) == 'w') {
word_adjust = true;
end++;
}
if ((*end | 0x20) == 'k') {
heap_size *= 1024;
} else if ((*end | 0x20) == 'm') {
heap_size *= 1024 * 1024;
}
if (word_adjust) {
heap_size = heap_size * MP_BYTES_PER_OBJ_WORD / 4;
}
} else {
exit(usage(argv));
}
a++;
}
}
}
}
STATIC char *backslash_to_forwardslash(char *path) {
for (char *p = path; p != NULL && *p != '\0'; ++p) {
if (*p == '\\') {
*p = '/';
}
}
return path;
}
MP_NOINLINE int main_(int argc, char **argv) {
mp_stack_set_limit(40000 * (sizeof(void *) / 4));
pre_process_options(argc, argv);
char *heap = malloc(heap_size);
gc_init(heap, heap + heap_size);
mp_init();
2021-03-15 09:57:36 -04:00
#ifdef _WIN32
set_fmode_binary();
2021-03-15 09:57:36 -04:00
#endif
#if MICROPY_EMIT_NATIVE
// Set default emitter options
MP_STATE_VM(default_emit_opt) = emit_opt;
#else
(void)emit_opt;
#endif
// set default compiler configuration
mp_dynamic_compiler.small_int_bits = 31;
// don't support native emitter unless -march is specified
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_NONE;
mp_dynamic_compiler.nlr_buf_num_regs = 0;
const char *input_file = NULL;
const char *output_file = NULL;
const char *source_file = NULL;
// parse main options
for (int a = 1; a < argc; a++) {
if (argv[a][0] == '-') {
if (strcmp(argv[a], "-X") == 0) {
a += 1;
} else if (strcmp(argv[a], "--version") == 0) {
printf("CircuitPython " MICROPY_GIT_TAG " on " MICROPY_BUILD_DATE
"; mpy-cross emitting mpy v" MP_STRINGIFY(MPY_VERSION) "." MP_STRINGIFY(MPY_SUB_VERSION) "\n");
return 0;
} else if (strcmp(argv[a], "-v") == 0) {
mp_verbose_flag++;
} else if (strncmp(argv[a], "-O", 2) == 0) {
if (unichar_isdigit(argv[a][2])) {
MP_STATE_VM(mp_optimise_value) = argv[a][2] & 0xf;
} else {
MP_STATE_VM(mp_optimise_value) = 0;
2021-03-15 09:57:36 -04:00
for (char *p = argv[a] + 1; *p && *p == 'O'; p++, MP_STATE_VM(mp_optimise_value)++) {;
}
}
} else if (strcmp(argv[a], "-o") == 0) {
if (a + 1 >= argc) {
exit(usage(argv));
}
a += 1;
output_file = argv[a];
} else if (strcmp(argv[a], "-s") == 0) {
if (a + 1 >= argc) {
exit(usage(argv));
}
a += 1;
source_file = backslash_to_forwardslash(argv[a]);
} else if (strncmp(argv[a], "-msmall-int-bits=", sizeof("-msmall-int-bits=") - 1) == 0) {
char *end;
mp_dynamic_compiler.small_int_bits =
strtol(argv[a] + sizeof("-msmall-int-bits=") - 1, &end, 0);
if (*end) {
return usage(argv);
}
// TODO check that small_int_bits is within range of host's capabilities
} else if (strncmp(argv[a], "-march=", sizeof("-march=") - 1) == 0) {
const char *arch = argv[a] + sizeof("-march=") - 1;
if (strcmp(arch, "x86") == 0) {
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_X86;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_X86;
} else if (strcmp(arch, "x64") == 0) {
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_X64;
mp_dynamic_compiler.nlr_buf_num_regs = MAX(MICROPY_NLR_NUM_REGS_X64, MICROPY_NLR_NUM_REGS_X64_WIN);
} else if (strcmp(arch, "armv6") == 0) {
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_ARMV6;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_ARM_THUMB_FP;
} else if (strcmp(arch, "armv6m") == 0) {
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_ARMV6M;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_ARM_THUMB_FP; // need to be conservative so this code can run on armv7emdp
} else if (strcmp(arch, "armv7m") == 0) {
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_ARMV7M;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_ARM_THUMB_FP;
} else if (strcmp(arch, "armv7em") == 0) {
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_ARMV7EM;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_ARM_THUMB_FP;
} else if (strcmp(arch, "armv7emsp") == 0) {
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_ARMV7EMSP;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_ARM_THUMB_FP;
} else if (strcmp(arch, "armv7emdp") == 0) {
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_ARMV7EMDP;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_ARM_THUMB_FP;
} else if (strcmp(arch, "xtensa") == 0) {
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_XTENSA;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_XTENSA;
} else if (strcmp(arch, "xtensawin") == 0) {
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_XTENSAWIN;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_XTENSAWIN;
} else if (strcmp(arch, "host") == 0) {
#if defined(__i386__) || defined(_M_IX86)
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_X86;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_X86;
#elif defined(__x86_64__) || defined(_M_X64)
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_X64;
mp_dynamic_compiler.nlr_buf_num_regs = MAX(MICROPY_NLR_NUM_REGS_X64, MICROPY_NLR_NUM_REGS_X64_WIN);
#elif defined(__arm__) && !defined(__thumb2__)
mp_dynamic_compiler.native_arch = MP_NATIVE_ARCH_ARMV6;
mp_dynamic_compiler.nlr_buf_num_regs = MICROPY_NLR_NUM_REGS_ARM_THUMB_FP;
#else
mp_printf(&mp_stderr_print, "unable to determine host architecture for -march=host\n");
exit(1);
#endif
} else {
return usage(argv);
}
} else {
return usage(argv);
}
} else {
if (input_file != NULL) {
mp_printf(&mp_stderr_print, "multiple input files\n");
exit(1);
}
input_file = backslash_to_forwardslash(argv[a]);
}
}
if (input_file == NULL) {
mp_printf(&mp_stderr_print, "no input file\n");
exit(1);
}
int ret = compile_and_save(input_file, output_file, source_file);
#if MICROPY_PY_MICROPYTHON_MEM_INFO
if (mp_verbose_flag) {
mp_micropython_mem_info(0, NULL);
}
#endif
mp_deinit();
return ret & 0xff;
}
int main(int argc, char **argv) {
mp_stack_ctrl_init();
return main_(argc, argv);
}
mp_import_stat_t mp_import_stat(const char *path) {
(void)path;
return MP_IMPORT_STAT_NO_EXIST;
}
void nlr_jump_fail(void *val) {
fprintf(stderr, "FATAL: uncaught NLR %p\n", val);
exit(1);
}