circuitpython/py/nlr.h

191 lines
6.4 KiB
C
Raw Normal View History

/*
* This file is part of the MicroPython project, http://micropython.org/
*
* The MIT License (MIT)
*
2020-06-03 18:40:05 -04:00
* SPDX-FileCopyrightText: Copyright (c) 2013, 2014 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.
*/
#ifndef MICROPY_INCLUDED_PY_NLR_H
#define MICROPY_INCLUDED_PY_NLR_H
// non-local return
// exception handling, basically a stack of setjmp/longjmp buffers
#include <limits.h>
#include <assert.h>
#include "py/mpconfig.h"
#define MICROPY_NLR_NUM_REGS_X86 (6)
#define MICROPY_NLR_NUM_REGS_X64 (8)
#define MICROPY_NLR_NUM_REGS_X64_WIN (10)
#define MICROPY_NLR_NUM_REGS_ARM_THUMB (10)
#define MICROPY_NLR_NUM_REGS_ARM_THUMB_FP (10 + 6)
2021-02-17 18:28:42 -05:00
#define MICROPY_NLR_NUM_REGS_AARCH64 (13)
#define MICROPY_NLR_NUM_REGS_XTENSA (10)
#define MICROPY_NLR_NUM_REGS_XTENSAWIN (17)
// *FORMAT-OFF*
// If MICROPY_NLR_SETJMP is not enabled then auto-detect the machine arch
2018-07-12 14:13:51 -04:00
#if !defined(MICROPY_NLR_SETJMP) || !MICROPY_NLR_SETJMP
// A lot of nlr-related things need different treatment on Windows
#if defined(_WIN32) || defined(__CYGWIN__)
#define MICROPY_NLR_OS_WINDOWS 1
#else
#define MICROPY_NLR_OS_WINDOWS 0
#endif
#if defined(__i386__)
#define MICROPY_NLR_SETJMP (0)
#define MICROPY_NLR_X86 (1)
#define MICROPY_NLR_NUM_REGS (MICROPY_NLR_NUM_REGS_X86)
#elif defined(__x86_64__)
#define MICROPY_NLR_SETJMP (0)
#define MICROPY_NLR_X64 (1)
#if MICROPY_NLR_OS_WINDOWS
#define MICROPY_NLR_NUM_REGS (MICROPY_NLR_NUM_REGS_X64_WIN)
#else
#define MICROPY_NLR_NUM_REGS (MICROPY_NLR_NUM_REGS_X64)
#endif
#elif defined(__thumb2__) || defined(__thumb__) || defined(__arm__)
#define MICROPY_NLR_SETJMP (0)
#define MICROPY_NLR_THUMB (1)
#if defined(__SOFTFP__)
#define MICROPY_NLR_NUM_REGS (MICROPY_NLR_NUM_REGS_ARM_THUMB)
#else
// With hardware FP registers s16-s31 are callee save so in principle
// should be saved and restored by the NLR code. gcc only uses s16-s21
// so only save/restore those as an optimisation.
#define MICROPY_NLR_NUM_REGS (MICROPY_NLR_NUM_REGS_ARM_THUMB_FP)
#endif
2021-02-17 18:28:42 -05:00
#elif defined(__aarch64__)
#define MICROPY_NLR_AARCH64 (1)
#define MICROPY_NLR_NUM_REGS (MICROPY_NLR_NUM_REGS_AARCH64)
#elif defined(__xtensa__)
#define MICROPY_NLR_SETJMP (0)
#define MICROPY_NLR_XTENSA (1)
#define MICROPY_NLR_NUM_REGS (MICROPY_NLR_NUM_REGS_XTENSA)
#elif defined(__powerpc__)
#define MICROPY_NLR_SETJMP (0)
#define MICROPY_NLR_POWERPC (1)
// this could be less but using 128 for safety
#define MICROPY_NLR_NUM_REGS (128)
#else
#define MICROPY_NLR_SETJMP (1)
// #warning "No native NLR support for this arch, using setjmp implementation"
#endif
#endif
// If MICROPY_NLR_SETJMP is not defined above - define/disable it here
#if !defined(MICROPY_NLR_SETJMP)
#define MICROPY_NLR_SETJMP (0)
#endif
// *FORMAT-ON*
#if MICROPY_NLR_SETJMP
#include <setjmp.h>
Revert "py/nlr: Factor out common NLR code to generic functions." This reverts commit 6a3a742a6c9caaa2be0fd0aac7a5df4ac816081c. The above commit has number of faults starting from the motivation down to the actual implementation. 1. Faulty implementation. The original code contained functions like: NORETURN void nlr_jump(void *val) { nlr_buf_t **top_ptr = &MP_STATE_THREAD(nlr_top); nlr_buf_t *top = *top_ptr; ... __asm volatile ( "mov %0, %%edx \n" // %edx points to nlr_buf "mov 28(%%edx), %%esi \n" // load saved %esi "mov 24(%%edx), %%edi \n" // load saved %edi "mov 20(%%edx), %%ebx \n" // load saved %ebx "mov 16(%%edx), %%esp \n" // load saved %esp "mov 12(%%edx), %%ebp \n" // load saved %ebp "mov 8(%%edx), %%eax \n" // load saved %eip "mov %%eax, (%%esp) \n" // store saved %eip to stack "xor %%eax, %%eax \n" // clear return register "inc %%al \n" // increase to make 1, non-local return "ret \n" // return : // output operands : "r"(top) // input operands : // clobbered registers ); } Which clearly stated that C-level variable should be a parameter of the assembly, whcih then moved it into correct register. Whereas now it's: NORETURN void nlr_jump_tail(nlr_buf_t *top) { (void)top; __asm volatile ( "mov 28(%edx), %esi \n" // load saved %esi "mov 24(%edx), %edi \n" // load saved %edi "mov 20(%edx), %ebx \n" // load saved %ebx "mov 16(%edx), %esp \n" // load saved %esp "mov 12(%edx), %ebp \n" // load saved %ebp "mov 8(%edx), %eax \n" // load saved %eip "mov %eax, (%esp) \n" // store saved %eip to stack "xor %eax, %eax \n" // clear return register "inc %al \n" // increase to make 1, non-local return "ret \n" // return ); for (;;); // needed to silence compiler warning } Which just tries to perform operations on a completely random register (edx in this case). The outcome is the expected: saving the pure random luck of the compiler putting the right value in the random register above, there's a crash. 2. Non-critical assessment. The original commit message says "There is a small overhead introduced (typically 1 machine instruction)". That machine instruction is a call if a compiler doesn't perform tail optimization (happens regularly), and it's 1 instruction only with the broken code shown above, fixing it requires adding more. With inefficiencies already presented in the NLR code, the overhead becomes "considerable" (several times more than 1%), not "small". The commit message also says "This eliminates duplicated code.". An obvious way to eliminate duplication would be to factor out common code to macros, not introduce overhead and breakage like above. 3. Faulty motivation. All this started with a report of warnings/errors happening for a niche compiler. It could have been solved in one the direct ways: a) fixing it just for affected compiler(s); b) rewriting it in proper assembly (like it was before BTW); c) by not doing anything at all, MICROPY_NLR_SETJMP exists exactly to address minor-impact cases like thar (where a) or b) are not applicable). Instead, a backwards "solution" was put forward, leading to all the issues above. The best action thus appears to be revert and rework, not trying to work around what went haywire in the first place.
2017-12-26 11:39:51 -05:00
#endif
typedef struct _nlr_buf_t nlr_buf_t;
struct _nlr_buf_t {
// the entries here must all be machine word size
nlr_buf_t *prev;
void *ret_val; // always a concrete object (an exception instance)
#if MICROPY_NLR_SETJMP
jmp_buf jmpbuf;
#else
void *regs[MICROPY_NLR_NUM_REGS];
#endif
py: Introduce a Python stack for scoped allocation. This patch introduces the MICROPY_ENABLE_PYSTACK option (disabled by default) which enables a "Python stack" that allows to allocate and free memory in a scoped, or Last-In-First-Out (LIFO) way, similar to alloca(). A new memory allocation API is introduced along with this Py-stack. It includes both "local" and "nonlocal" LIFO allocation. Local allocation is intended to be equivalent to using alloca(), whereby the same function must free the memory. Nonlocal allocation is where another function may free the memory, so long as it's still LIFO. Follow-up patches will convert all uses of alloca() and VLA to the new scoped allocation API. The old behaviour (using alloca()) will still be available, but when MICROPY_ENABLE_PYSTACK is enabled then alloca() is no longer required or used. The benefits of enabling this option are (or will be once subsequent patches are made to convert alloca()/VLA): - Toolchains without alloca() can use this feature to obtain correct and efficient scoped memory allocation (compared to using the heap instead of alloca(), which is slower). - Even if alloca() is available, enabling the Py-stack gives slightly more efficient use of stack space when calling nested Python functions, due to the way that compilers implement alloca(). - Enabling the Py-stack with the stackless mode allows for even more efficient stack usage, as well as retaining high performance (because the heap is no longer used to build and destroy stackless code states). - With Py-stack and stackless enabled, Python-calling-Python is no longer recursive in the C mp_execute_bytecode function. The micropython.pystack_use() function is included to measure usage of the Python stack.
2017-11-26 07:28:40 -05:00
#if MICROPY_ENABLE_PYSTACK
void *pystack;
#endif
};
Revert "py/nlr: Factor out common NLR code to generic functions." This reverts commit 6a3a742a6c9caaa2be0fd0aac7a5df4ac816081c. The above commit has number of faults starting from the motivation down to the actual implementation. 1. Faulty implementation. The original code contained functions like: NORETURN void nlr_jump(void *val) { nlr_buf_t **top_ptr = &MP_STATE_THREAD(nlr_top); nlr_buf_t *top = *top_ptr; ... __asm volatile ( "mov %0, %%edx \n" // %edx points to nlr_buf "mov 28(%%edx), %%esi \n" // load saved %esi "mov 24(%%edx), %%edi \n" // load saved %edi "mov 20(%%edx), %%ebx \n" // load saved %ebx "mov 16(%%edx), %%esp \n" // load saved %esp "mov 12(%%edx), %%ebp \n" // load saved %ebp "mov 8(%%edx), %%eax \n" // load saved %eip "mov %%eax, (%%esp) \n" // store saved %eip to stack "xor %%eax, %%eax \n" // clear return register "inc %%al \n" // increase to make 1, non-local return "ret \n" // return : // output operands : "r"(top) // input operands : // clobbered registers ); } Which clearly stated that C-level variable should be a parameter of the assembly, whcih then moved it into correct register. Whereas now it's: NORETURN void nlr_jump_tail(nlr_buf_t *top) { (void)top; __asm volatile ( "mov 28(%edx), %esi \n" // load saved %esi "mov 24(%edx), %edi \n" // load saved %edi "mov 20(%edx), %ebx \n" // load saved %ebx "mov 16(%edx), %esp \n" // load saved %esp "mov 12(%edx), %ebp \n" // load saved %ebp "mov 8(%edx), %eax \n" // load saved %eip "mov %eax, (%esp) \n" // store saved %eip to stack "xor %eax, %eax \n" // clear return register "inc %al \n" // increase to make 1, non-local return "ret \n" // return ); for (;;); // needed to silence compiler warning } Which just tries to perform operations on a completely random register (edx in this case). The outcome is the expected: saving the pure random luck of the compiler putting the right value in the random register above, there's a crash. 2. Non-critical assessment. The original commit message says "There is a small overhead introduced (typically 1 machine instruction)". That machine instruction is a call if a compiler doesn't perform tail optimization (happens regularly), and it's 1 instruction only with the broken code shown above, fixing it requires adding more. With inefficiencies already presented in the NLR code, the overhead becomes "considerable" (several times more than 1%), not "small". The commit message also says "This eliminates duplicated code.". An obvious way to eliminate duplication would be to factor out common code to macros, not introduce overhead and breakage like above. 3. Faulty motivation. All this started with a report of warnings/errors happening for a niche compiler. It could have been solved in one the direct ways: a) fixing it just for affected compiler(s); b) rewriting it in proper assembly (like it was before BTW); c) by not doing anything at all, MICROPY_NLR_SETJMP exists exactly to address minor-impact cases like thar (where a) or b) are not applicable). Instead, a backwards "solution" was put forward, leading to all the issues above. The best action thus appears to be revert and rework, not trying to work around what went haywire in the first place.
2017-12-26 11:39:51 -05:00
// Helper macros to save/restore the pystack state
#if MICROPY_ENABLE_PYSTACK
#define MP_NLR_SAVE_PYSTACK(nlr_buf) (nlr_buf)->pystack = MP_STATE_THREAD(pystack_cur)
#define MP_NLR_RESTORE_PYSTACK(nlr_buf) MP_STATE_THREAD(pystack_cur) = (nlr_buf)->pystack
#else
#define MP_NLR_SAVE_PYSTACK(nlr_buf) (void)nlr_buf
#define MP_NLR_RESTORE_PYSTACK(nlr_buf) (void)nlr_buf
#endif
// Helper macro to use at the start of a specific nlr_jump implementation
#define MP_NLR_JUMP_HEAD(val, top) \
nlr_buf_t **_top_ptr = &MP_STATE_THREAD(nlr_top); \
nlr_buf_t *top = *_top_ptr; \
if (top == NULL) { \
nlr_jump_fail(val); \
} \
top->ret_val = val; \
MP_NLR_RESTORE_PYSTACK(top); \
*_top_ptr = top->prev; \
Revert "py/nlr: Factor out common NLR code to generic functions." This reverts commit 6a3a742a6c9caaa2be0fd0aac7a5df4ac816081c. The above commit has number of faults starting from the motivation down to the actual implementation. 1. Faulty implementation. The original code contained functions like: NORETURN void nlr_jump(void *val) { nlr_buf_t **top_ptr = &MP_STATE_THREAD(nlr_top); nlr_buf_t *top = *top_ptr; ... __asm volatile ( "mov %0, %%edx \n" // %edx points to nlr_buf "mov 28(%%edx), %%esi \n" // load saved %esi "mov 24(%%edx), %%edi \n" // load saved %edi "mov 20(%%edx), %%ebx \n" // load saved %ebx "mov 16(%%edx), %%esp \n" // load saved %esp "mov 12(%%edx), %%ebp \n" // load saved %ebp "mov 8(%%edx), %%eax \n" // load saved %eip "mov %%eax, (%%esp) \n" // store saved %eip to stack "xor %%eax, %%eax \n" // clear return register "inc %%al \n" // increase to make 1, non-local return "ret \n" // return : // output operands : "r"(top) // input operands : // clobbered registers ); } Which clearly stated that C-level variable should be a parameter of the assembly, whcih then moved it into correct register. Whereas now it's: NORETURN void nlr_jump_tail(nlr_buf_t *top) { (void)top; __asm volatile ( "mov 28(%edx), %esi \n" // load saved %esi "mov 24(%edx), %edi \n" // load saved %edi "mov 20(%edx), %ebx \n" // load saved %ebx "mov 16(%edx), %esp \n" // load saved %esp "mov 12(%edx), %ebp \n" // load saved %ebp "mov 8(%edx), %eax \n" // load saved %eip "mov %eax, (%esp) \n" // store saved %eip to stack "xor %eax, %eax \n" // clear return register "inc %al \n" // increase to make 1, non-local return "ret \n" // return ); for (;;); // needed to silence compiler warning } Which just tries to perform operations on a completely random register (edx in this case). The outcome is the expected: saving the pure random luck of the compiler putting the right value in the random register above, there's a crash. 2. Non-critical assessment. The original commit message says "There is a small overhead introduced (typically 1 machine instruction)". That machine instruction is a call if a compiler doesn't perform tail optimization (happens regularly), and it's 1 instruction only with the broken code shown above, fixing it requires adding more. With inefficiencies already presented in the NLR code, the overhead becomes "considerable" (several times more than 1%), not "small". The commit message also says "This eliminates duplicated code.". An obvious way to eliminate duplication would be to factor out common code to macros, not introduce overhead and breakage like above. 3. Faulty motivation. All this started with a report of warnings/errors happening for a niche compiler. It could have been solved in one the direct ways: a) fixing it just for affected compiler(s); b) rewriting it in proper assembly (like it was before BTW); c) by not doing anything at all, MICROPY_NLR_SETJMP exists exactly to address minor-impact cases like thar (where a) or b) are not applicable). Instead, a backwards "solution" was put forward, leading to all the issues above. The best action thus appears to be revert and rework, not trying to work around what went haywire in the first place.
2017-12-26 11:39:51 -05:00
#if MICROPY_NLR_SETJMP
// nlr_push() must be defined as a macro, because "The stack context will be
// invalidated if the function which called setjmp() returns."
// For this case it is safe to call nlr_push_tail() first.
#define nlr_push(buf) (nlr_push_tail(buf), setjmp((buf)->jmpbuf))
#else
unsigned int nlr_push(nlr_buf_t *);
#endif
unsigned int nlr_push_tail(nlr_buf_t *top);
void nlr_pop(void);
NORETURN void nlr_jump(void *val);
// This must be implemented by a port. It's called by nlr_jump
// if no nlr buf has been pushed. It must not return, but rather
// should bail out with a fatal error.
NORETURN void nlr_jump_fail(void *val);
// use nlr_raise instead of nlr_jump so that debugging is easier
#ifndef MICROPY_DEBUG_NLR
#define nlr_raise(val) nlr_jump(MP_OBJ_TO_PTR(val))
#else
#include "mpstate.h"
#define nlr_raise(val) \
do { \
/*printf("nlr_raise: nlr_top=%p\n", MP_STATE_THREAD(nlr_top)); \
fflush(stdout);*/ \
void *_val = MP_OBJ_TO_PTR(val); \
assert(_val != NULL); \
assert(mp_obj_is_exception_instance(val)); \
nlr_jump(_val); \
} while (0)
#if !MICROPY_NLR_SETJMP
#define nlr_push(val) \
assert(MP_STATE_THREAD(nlr_top) != val),nlr_push(val)
/*
#define nlr_push(val) \
printf("nlr_push: before: nlr_top=%p, val=%p\n", MP_STATE_THREAD(nlr_top), val),assert(MP_STATE_THREAD(nlr_top) != val),nlr_push(val)
*/
#endif
#endif
#endif // MICROPY_INCLUDED_PY_NLR_H