ruby/ruby.c

2608 строки
68 KiB
C
Исходник Обычный вид История

/**********************************************************************
ruby.c -
$Author$
created at: Tue Aug 10 12:47:31 JST 1993
* encoding.c: provide basic features for M17N. * parse.y: encoding aware parsing. * parse.y (pragma_encoding): encoding specification pragma. * parse.y (rb_intern3): encoding specified symbols. * string.c (rb_str_length): length based on characters. for older behavior, bytesize method added. * string.c (rb_str_index_m): index based on characters. rindex as well. * string.c (succ_char): encoding aware succeeding string. * string.c (rb_str_reverse): reverse based on characters. * string.c (rb_str_inspect): encoding aware string description. * string.c (rb_str_upcase_bang): encoding aware case conversion. downcase, capitalize, swapcase as well. * string.c (rb_str_tr_bang): tr based on characters. delete, squeeze, tr_s, count as well. * string.c (rb_str_split_m): split based on characters. * string.c (rb_str_each_line): encoding aware each_line. * string.c (rb_str_each_char): added. iteration based on characters. * string.c (rb_str_strip_bang): encoding aware whitespace stripping. lstrip, rstrip as well. * string.c (rb_str_justify): encoding aware justifying (ljust, rjust, center). * string.c (str_encoding): get encoding attribute from a string. * re.c (rb_reg_initialize): encoding aware regular expression * sprintf.c (rb_str_format): formatting (i.e. length count) based on characters. * io.c (rb_io_getc): getc to return one-character string. for older behavior, getbyte method added. * ext/stringio/stringio.c (strio_getc): ditto. * io.c (rb_io_ungetc): allow pushing arbitrary string at the current reading point. * ext/stringio/stringio.c (strio_ungetc): ditto. * ext/strscan/strscan.c: encoding support. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@13261 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2007-08-25 07:29:39 +04:00
Copyright (C) 1993-2007 Yukihiro Matsumoto
Copyright (C) 2000 Network Applied Communication Laboratory, Inc.
Copyright (C) 2000 Information-technology Promotion Agency, Japan
**********************************************************************/
#include "ruby/internal/config.h"
#include <ctype.h>
#include <stdio.h>
#include <sys/types.h>
#ifdef __CYGWIN__
# include <windows.h>
# include <sys/cygwin.h>
#endif
#ifdef __hpux
# include <sys/pstat.h>
#endif
#if defined(LOAD_RELATIVE) && defined(HAVE_DLADDR)
# include <dlfcn.h>
#endif
#ifdef HAVE_UNISTD_H
# include <unistd.h>
#endif
#if defined(HAVE_FCNTL_H)
# include <fcntl.h>
#elif defined(HAVE_SYS_FCNTL_H)
# include <sys/fcntl.h>
#endif
#ifdef HAVE_SYS_PARAM_H
# include <sys/param.h>
#endif
#include "dln.h"
#include "eval_intern.h"
#include "internal.h"
#include "internal/error.h"
#include "internal/file.h"
#include "internal/inits.h"
#include "internal/io.h"
#include "internal/load.h"
#include "internal/loadpath.h"
#include "internal/missing.h"
#include "internal/object.h"
#include "internal/parse.h"
#include "internal/process.h"
#include "internal/variable.h"
#include "mjit.h"
#include "ruby/encoding.h"
#include "ruby/thread.h"
#include "ruby/util.h"
#include "ruby/version.h"
#include "ruby/internal/error.h"
#ifndef MAXPATHLEN
# define MAXPATHLEN 1024
#endif
#ifndef O_ACCMODE
# define O_ACCMODE (O_RDONLY | O_WRONLY | O_RDWR)
#endif
void Init_ruby_description(void);
#ifndef HAVE_STDLIB_H
char *getenv();
#endif
#ifndef DISABLE_RUBYGEMS
# define DISABLE_RUBYGEMS 0
#endif
#if DISABLE_RUBYGEMS
#define DEFAULT_RUBYGEMS_ENABLED "disabled"
#else
#define DEFAULT_RUBYGEMS_ENABLED "enabled"
#endif
void rb_warning_category_update(unsigned int mask, unsigned int bits);
#define COMMA ,
#define FEATURE_BIT(bit) (1U << feature_##bit)
#define EACH_FEATURES(X, SEP) \
X(gems) \
SEP \
X(did_you_mean) \
SEP \
X(rubyopt) \
SEP \
X(frozen_string_literal) \
SEP \
X(jit) \
/* END OF FEATURES */
#define EACH_DEBUG_FEATURES(X, SEP) \
X(frozen_string_literal) \
/* END OF DEBUG FEATURES */
#define AMBIGUOUS_FEATURE_NAMES 0 /* no ambiguous feature names now */
#define DEFINE_FEATURE(bit) feature_##bit
#define DEFINE_DEBUG_FEATURE(bit) feature_debug_##bit
enum feature_flag_bits {
EACH_FEATURES(DEFINE_FEATURE, COMMA),
feature_debug_flag_first,
feature_debug_flag_begin = feature_debug_flag_first - 1,
EACH_DEBUG_FEATURES(DEFINE_DEBUG_FEATURE, COMMA),
feature_flag_count
};
#define DEBUG_BIT(bit) (1U << feature_debug_##bit)
#define DUMP_BIT(bit) (1U << dump_##bit)
#define DEFINE_DUMP(bit) dump_##bit
#define EACH_DUMPS(X, SEP) \
X(version) \
SEP \
X(copyright) \
SEP \
X(usage) \
SEP \
X(help) \
SEP \
X(yydebug) \
SEP \
X(syntax) \
SEP \
X(parsetree) \
SEP \
X(parsetree_with_comment) \
SEP \
X(insns) \
/* END OF DUMPS */
enum dump_flag_bits {
dump_version_v,
EACH_DUMPS(DEFINE_DUMP, COMMA),
dump_exit_bits = (DUMP_BIT(yydebug) | DUMP_BIT(syntax) |
DUMP_BIT(parsetree) | DUMP_BIT(parsetree_with_comment) |
DUMP_BIT(insns))
};
typedef struct ruby_cmdline_options ruby_cmdline_options_t;
typedef struct {
unsigned int mask;
unsigned int set;
} ruby_features_t;
static inline void
rb_feature_set_to(ruby_features_t *feat, unsigned int bit_mask, unsigned int bit_set)
{
feat->mask |= bit_mask;
feat->set = (feat->set & ~bit_mask) | bit_set;
}
#define FEATURE_SET_TO(feat, bit_mask, bit_set) \
rb_feature_set_to(&(feat), bit_mask, bit_set)
#define FEATURE_SET(feat, bits) FEATURE_SET_TO(feat, bits, bits)
2019-12-20 16:50:46 +03:00
#define FEATURE_SET_RESTORE(feat, save) FEATURE_SET_TO(feat, (save).mask, (save).set & (save).mask)
#define FEATURE_SET_P(feat, bits) ((feat).set & (bits))
struct ruby_cmdline_options {
const char *script;
VALUE script_name;
VALUE e_script;
struct {
struct {
VALUE name;
int index;
} enc;
} src, ext, intern;
VALUE req_list;
ruby_features_t features;
ruby_features_t warn;
unsigned int dump;
#if USE_MJIT
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
struct mjit_options mjit;
#endif
int sflag, xflag;
unsigned int warning: 1;
unsigned int verbose: 1;
unsigned int do_loop: 1;
unsigned int do_print: 1;
unsigned int do_line: 1;
unsigned int do_split: 1;
unsigned int do_search: 1;
unsigned int setids: 2;
};
static void init_ids(ruby_cmdline_options_t *);
#define src_encoding_index GET_VM()->src_encoding_index
enum {
COMPILATION_FEATURES = (
0
| FEATURE_BIT(frozen_string_literal)
| FEATURE_BIT(debug_frozen_string_literal)
),
DEFAULT_FEATURES = (
(FEATURE_BIT(debug_flag_first)-1)
#if DISABLE_RUBYGEMS
& ~FEATURE_BIT(gems)
#endif
& ~FEATURE_BIT(frozen_string_literal)
& ~FEATURE_BIT(jit)
)
};
static ruby_cmdline_options_t *
cmdline_options_init(ruby_cmdline_options_t *opt)
{
MEMZERO(opt, *opt, 1);
init_ids(opt);
opt->src.enc.index = src_encoding_index;
opt->ext.enc.index = -1;
opt->intern.enc.index = -1;
opt->features.set = DEFAULT_FEATURES;
#ifdef MJIT_FORCE_ENABLE /* to use with: ./configure cppflags="-DMJIT_FORCE_ENABLE" */
opt->features.set |= FEATURE_BIT(jit);
#endif
return opt;
}
static rb_ast_t *load_file(VALUE parser, VALUE fname, VALUE f, int script,
ruby_cmdline_options_t *opt);
static VALUE open_load_file(VALUE fname_v, int *xflag);
static void forbid_setid(const char *, const ruby_cmdline_options_t *);
#define forbid_setid(s) forbid_setid((s), opt)
static struct {
int argc;
char **argv;
} origarg;
2020-03-27 17:18:42 +03:00
static const char esc_standout[] = "\n\033[1;7m";
static const char esc_bold[] = "\033[1m";
static const char esc_reset[] = "\033[0m";
static const char esc_none[] = "";
static void
2020-04-12 07:00:27 +03:00
show_usage_line(const char *str, unsigned int namelen, unsigned int secondlen, int help, int highlight, unsigned int w)
{
2020-03-27 17:18:42 +03:00
const char *sb = highlight ? esc_bold : esc_none;
const char *se = highlight ? esc_reset : esc_none;
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
const int wrap = help && namelen + secondlen - 1 > w;
2020-03-27 17:18:42 +03:00
printf(" %s%.*s%-*.*s%s%-*s%s\n", sb, namelen-1, str,
(wrap ? 0 : w - namelen + 1),
2020-03-27 17:18:42 +03:00
(help ? secondlen-1 : 0), str + namelen, se,
(wrap ? w + 3 : 0), (wrap ? "\n" : ""),
str + namelen + secondlen);
}
static void
2020-04-12 07:00:27 +03:00
usage(const char *name, int help, int highlight, int columns)
{
/* This message really ought to be max 23 lines.
* Removed -h because the user already knows that option. Others? */
struct message {
const char *str;
unsigned short namelen, secondlen;
};
#define M(shortopt, longopt, desc) { \
shortopt " " longopt " " desc, \
(unsigned short)sizeof(shortopt), \
(unsigned short)sizeof(longopt), \
}
static const struct message usage_msg[] = {
M("-0[octal]", "", "specify record separator (\\0, if no argument)"),
M("-a", "", "autosplit mode with -n or -p (splits $_ into $F)"),
M("-c", "", "check syntax only"),
M("-Cdirectory", "", "cd to directory before executing your script"),
M("-d", ", --debug", "set debugging flags (set $DEBUG to true)"),
M("-e 'command'", "", "one line of script. Several -e's allowed. Omit [programfile]"),
M("-Eex[:in]", ", --encoding=ex[:in]", "specify the default external and internal character encodings"),
M("-Fpattern", "", "split() pattern for autosplit (-a)"),
M("-i[extension]", "", "edit ARGV files in place (make backup if extension supplied)"),
M("-Idirectory", "", "specify $LOAD_PATH directory (may be used more than once)"),
M("-l", "", "enable line ending processing"),
M("-n", "", "assume 'while gets(); ... end' loop around your script"),
M("-p", "", "assume loop like -n but print line also like sed"),
M("-rlibrary", "", "require the library before executing your script"),
M("-s", "", "enable some switch parsing for switches after script name"),
M("-S", "", "look for the script using PATH environment variable"),
M("-v", "", "print the version number, then turn on verbose mode"),
M("-w", "", "turn warnings on for your script"),
M("-W[level=2|:category]", "", "set warning level; 0=silence, 1=medium, 2=verbose"),
M("-x[directory]", "", "strip off text before #!ruby line and perhaps cd to directory"),
M("--jit", "", "enable JIT with default options (experimental)"),
M("--jit-[option]","", "enable JIT with an option (experimental)"),
M("-h", "", "show this message, --help for more info"),
};
static const struct message help_msg[] = {
M("--copyright", "", "print the copyright"),
M("--dump={insns|parsetree|...}[,...]", "",
"dump debug information. see below for available dump list"),
M("--enable={gems|rubyopt|...}[,...]", ", --disable={gems|rubyopt|...}[,...]",
"enable or disable features. see below for available features"),
M("--external-encoding=encoding", ", --internal-encoding=encoding",
"specify the default external or internal character encoding"),
M("--verbose", "", "turn on verbose mode and disable script from stdin"),
M("--version", "", "print the version number, then exit"),
M("--help", "", "show this message, -h for short message"),
M("--backtrace-limit=num", "", "limit the maximum length of backtrace"),
};
static const struct message dumps[] = {
M("insns", "", "instruction sequences"),
M("yydebug", "", "yydebug of yacc parser generator"),
M("parsetree", "", "AST"),
M("parsetree_with_comment", "", "AST with comments"),
};
static const struct message features[] = {
M("gems", "", "rubygems (default: "DEFAULT_RUBYGEMS_ENABLED")"),
M("did_you_mean", "", "did_you_mean (default: "DEFAULT_RUBYGEMS_ENABLED")"),
M("rubyopt", "", "RUBYOPT environment variable (default: enabled)"),
M("frozen-string-literal", "", "freeze all string literals (default: disabled)"),
M("jit", "", "JIT compiler (default: disabled)"),
};
static const struct message warn_categories[] = {
M("deprecated", "", "deprecated features"),
M("experimental", "", "experimental features"),
};
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
static const struct message mjit_options[] = {
M("--jit-warnings", "", "Enable printing JIT warnings"),
M("--jit-debug", "", "Enable JIT debugging (very slow), or add cflags if specified"),
M("--jit-wait", "", "Wait until JIT compilation finishes every time (for testing)"),
M("--jit-save-temps", "", "Save JIT temporary files in $TMP or /tmp (for testing)"),
M("--jit-verbose=num", "", "Print JIT logs of level num or less to stderr (default: 0)"),
Change defaults of --jit options * --jit-min-calls: 5 -> 10000 --jit-min-calls=5 obviously can compile non hotspot. This was not a problem for MJIT-benchmarks and Optcarrot because the former has very few hot optimiziable methods and the latter is likely to trigger compilation of hotspot by its intensive calls to optimizable hotspot methods and has a very short window to allow limited compilations. In real-world applications, it has more time to compile more methods and it pressures computer's limited resources like icache. We should avoid compiling too many methods. Also compiling many methods exhausts time budget for compilation in one ruby process lifetime and delays the "JIT compaction" of Ruby 2.6. JVM is known to use 1,500 for C1 (client) compiler and 10,000 for C2 (server) compiler for -XX:CompileThreshold by default. https://docs.oracle.com/javase/8/embedded/develop-apps-platforms/codecache.htm When things are called X,000 times, requiring 10,000 can eliminate compilation of methods which are called only once in these X,000 iterations and obviously not hotspot. And in fact things like unicorn-worker-killer restarts unicorn process every 4096 requests. So I felt 10,000 is good for such an environment. * --jit-max-cache: 1000 -> 100 By the same reason stated above, we should not allow compiling many methods especially on MJIT which has a larger overhead due to poor code locality by dlopen and whose code is also likely to be bigger by just inlining many VM instructions with -O3 rather than directly generating low-level code. In JVM -XX:ReservedCodeCacheSize is 32M for reserved and 48M for maximum. --jit-max-cache=1,000 could be closer to it, but in this case MJIT's compilation is slow due to data synchronization between threads (to be improved in Ruby 2.7 though) and we do not want to delay the "JIT compaction" for a long time. So I chose a really conservative number for this, but by having method inlining in the future, wider range could be optimized even with this value. * Optcarrot --disable-gems, --benchmark Lan_Master.nes 12 attempts. No significant impact. | r67276 | r67276 --jit | after --jit | |:-------------------|:------------------|:------------------| | 50.44369263063978 | 72.87390680773056 | 73.47873485047297 | | 50.58788746124193 | 78.06820808947026 | 78.29723420171945 | | 50.77509250801378 | 80.29010348842613 | 78.94689404460769 | | 50.935361702064405 | 80.42796829926374 | 80.39539527351525 | | 51.27352672981195 | 81.98758158033202 | 81.6754198664817 | | 51.720715743242124 | 82.00118535811626 | 82.22960569251283 | | 51.89643169822524 | 82.2290091613556 | 82.5013636146388 | | 51.95895898113868 | 82.37318990939565 | 82.74002377794454 | | 52.10124886807968 | 82.48796686037502 | 83.23354941183932 | | 52.292280637519376 | 83.0265226541066 | 84.01552618012572 | | 52.51856237784144 | 83.8797360318052 | 84.8588319093393 | | 52.65076845986818 | 84.80037351256634 | 85.13577756273656 | * Railsbench `WARMUP=20000 BENCHMARK=1000 bin/bench` of https://github.com/k0kubun/railsbench. It gets closer to --disable=jit. | | r67276 | r67276 | after | | | | --jit | --jit | |:----------|:-------|:-------|:-------| | req/s | 891.3 | 742.2 | 841.5 | |:----------|:-------|:-------|:-------| | 50%ile ms | 1.00 | 1.21 | 1.08 | | 66%ile ms | 1.02 | 1.24 | 1.09 | | 75%ile ms | 1.03 | 1.28 | 1.10 | | 80%ile ms | 1.03 | 1.30 | 1.11 | | 90%ile ms | 1.09 | 1.42 | 1.15 | | 95%ile ms | 1.32 | 1.65 | 1.27 | | 98%ile ms | 4.79 | 2.23 | 1.81 | | 99%ile ms | 5.68 | 7.52 | 6.64 | |100%ile ms | 6.52 | 9.69 | 8.59 | git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67277 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2019-03-17 08:13:38 +03:00
M("--jit-max-cache=num", "", "Max number of methods to be JIT-ed in a cache (default: 100)"),
M("--jit-min-calls=num", "", "Number of calls to trigger JIT (for testing, default: 10000)"),
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
};
int i;
2020-03-27 17:18:42 +03:00
const char *sb = highlight ? esc_standout+1 : esc_none;
const char *se = highlight ? esc_reset : esc_none;
const int num = numberof(usage_msg) - (help ? 1 : 0);
2020-04-12 07:00:27 +03:00
unsigned int w = (columns > 80 ? (columns - 79) / 2 : 0) + 16;
#define SHOW(m) show_usage_line((m).str, (m).namelen, (m).secondlen, help, highlight, w)
2020-03-27 17:18:42 +03:00
printf("%sUsage:%s %s [switches] [--] [programfile] [arguments]\n", sb, se, name);
for (i = 0; i < num; ++i)
SHOW(usage_msg[i]);
if (!help) return;
2020-03-27 17:18:42 +03:00
if (highlight) sb = esc_standout;
for (i = 0; i < numberof(help_msg); ++i)
SHOW(help_msg[i]);
2020-03-27 17:18:42 +03:00
printf("%s""Dump List:%s\n", sb, se);
for (i = 0; i < numberof(dumps); ++i)
SHOW(dumps[i]);
2020-03-27 17:18:42 +03:00
printf("%s""Features:%s\n", sb, se);
for (i = 0; i < numberof(features); ++i)
SHOW(features[i]);
2020-03-27 17:18:42 +03:00
printf("%s""Warning categories:%s\n", sb, se);
for (i = 0; i < numberof(warn_categories); ++i)
SHOW(warn_categories[i]);
2020-03-27 17:18:42 +03:00
printf("%s""JIT options (experimental):%s\n", sb, se);
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
for (i = 0; i < numberof(mjit_options); ++i)
SHOW(mjit_options[i]);
}
#define rubylib_path_new rb_str_new
static void
push_include(const char *path, VALUE (*filter)(VALUE))
{
const char sep = PATH_SEP_CHAR;
const char *p, *s;
VALUE load_path = GET_VM()->load_path;
p = path;
while (*p) {
while (*p == sep)
p++;
if (!*p) break;
for (s = p; *s && *s != sep; s = CharNext(s));
rb_ary_push(load_path, (*filter)(rubylib_path_new(p, s - p)));
p = s;
}
}
#ifdef __CYGWIN__
static void
push_include_cygwin(const char *path, VALUE (*filter)(VALUE))
{
const char *p, *s;
char rubylib[FILENAME_MAX];
VALUE buf = 0;
p = path;
while (*p) {
unsigned int len;
while (*p == ';')
p++;
if (!*p) break;
for (s = p; *s && *s != ';'; s = CharNext(s));
len = s - p;
if (*s) {
if (!buf) {
buf = rb_str_new(p, len);
p = RSTRING_PTR(buf);
}
else {
rb_str_resize(buf, len);
p = strncpy(RSTRING_PTR(buf), p, len);
}
}
#ifdef HAVE_CYGWIN_CONV_PATH
#define CONV_TO_POSIX_PATH(p, lib) \
cygwin_conv_path(CCP_WIN_A_TO_POSIX|CCP_RELATIVE, (p), (lib), sizeof(lib))
#else
# error no cygwin_conv_path
#endif
if (CONV_TO_POSIX_PATH(p, rubylib) == 0)
p = rubylib;
push_include(p, filter);
if (!*s) break;
p = s + 1;
}
}
#define push_include push_include_cygwin
#endif
void
ruby_push_include(const char *path, VALUE (*filter)(VALUE))
{
if (path == 0)
return;
push_include(path, filter);
}
static VALUE
identical_path(VALUE path)
{
return path;
}
static VALUE
locale_path(VALUE path)
{
rb_enc_associate(path, rb_locale_encoding());
return path;
}
void
ruby_incpush(const char *path)
{
ruby_push_include(path, locale_path);
}
static VALUE
expand_include_path(VALUE path)
{
char *p = RSTRING_PTR(path);
if (!p)
return path;
if (*p == '.' && p[1] == '/')
return path;
return rb_file_expand_path(path, Qnil);
}
void
ruby_incpush_expand(const char *path)
{
ruby_push_include(path, expand_include_path);
}
#undef UTF8_PATH
#if defined _WIN32 || defined __CYGWIN__
static HMODULE libruby;
BOOL WINAPI
DllMain(HINSTANCE dll, DWORD reason, LPVOID reserved)
{
if (reason == DLL_PROCESS_ATTACH)
libruby = dll;
return TRUE;
}
HANDLE
rb_libruby_handle(void)
{
return libruby;
}
static inline void
translit_char_bin(char *p, int from, int to)
{
while (*p) {
if ((unsigned char)*p == from)
*p = to;
p++;
}
}
#endif
#ifdef _WIN32
# define UTF8_PATH 1
#endif
#ifndef UTF8_PATH
# define UTF8_PATH 0
#endif
#if UTF8_PATH
# define IF_UTF8_PATH(t, f) t
#else
# define IF_UTF8_PATH(t, f) f
#endif
#if UTF8_PATH
static VALUE
str_conv_enc(VALUE str, rb_encoding *from, rb_encoding *to)
{
return rb_str_conv_enc_opts(str, from, to,
ECONV_UNDEF_REPLACE|ECONV_INVALID_REPLACE,
Qnil);
}
#else
# define str_conv_enc(str, from, to) (str)
#endif
void ruby_init_loadpath(void);
#if defined(LOAD_RELATIVE)
static VALUE
runtime_libruby_path(void)
{
#if defined _WIN32 || defined __CYGWIN__
DWORD len = RSTRING_EMBED_LEN_MAX, ret;
VALUE path;
VALUE wsopath = rb_str_new(0, len*sizeof(WCHAR));
WCHAR *wlibpath;
char *libpath;
while (wlibpath = (WCHAR *)RSTRING_PTR(wsopath),
ret = GetModuleFileNameW(libruby, wlibpath, len),
(ret == len))
{
rb_str_modify_expand(wsopath, len*sizeof(WCHAR));
rb_str_set_len(wsopath, (len += len)*sizeof(WCHAR));
}
if (!ret || ret > len) rb_fatal("failed to get module file name");
#if defined __CYGWIN__
{
const int win_to_posix = CCP_WIN_W_TO_POSIX | CCP_RELATIVE;
size_t newsize = cygwin_conv_path(win_to_posix, wlibpath, 0, 0);
if (!newsize) rb_fatal("failed to convert module path to cygwin");
path = rb_str_new(0, newsize);
libpath = RSTRING_PTR(path);
if (cygwin_conv_path(win_to_posix, wlibpath, libpath, newsize)) {
rb_str_resize(path, 0);
}
}
#else
{
DWORD i;
for (len = ret, i = 0; i < len; ++i) {
if (wlibpath[i] == L'\\') {
wlibpath[i] = L'/';
ret = i+1; /* chop after the last separator */
}
}
}
len = WideCharToMultiByte(CP_UTF8, 0, wlibpath, ret, NULL, 0, NULL, NULL);
path = rb_utf8_str_new(0, len);
libpath = RSTRING_PTR(path);
WideCharToMultiByte(CP_UTF8, 0, wlibpath, ret, libpath, len, NULL, NULL);
#endif
rb_str_resize(wsopath, 0);
return path;
#elif defined(HAVE_DLADDR)
Dl_info dli;
VALUE fname, path;
const void* addr = (void *)(VALUE)expand_include_path;
if (!dladdr((void *)addr, &dli)) {
return rb_str_new(0, 0);
}
#ifdef __linux__
else if (origarg.argc > 0 && origarg.argv && dli.dli_fname == origarg.argv[0]) {
fname = rb_str_new_cstr("/proc/self/exe");
path = rb_readlink(fname, NULL);
}
#endif
else {
fname = rb_str_new_cstr(dli.dli_fname);
path = rb_realpath_internal(Qnil, fname, 1);
}
rb_str_resize(fname, 0);
return path;
#else
# error relative load path is not supported on this platform.
#endif
}
#endif
#define INITIAL_LOAD_PATH_MARK rb_intern_const("@gem_prelude_index")
VALUE ruby_archlibdir_path, ruby_prefix_path;
void
ruby_init_loadpath(void)
{
VALUE load_path, archlibdir = 0;
ID id_initial_load_path_mark;
const char *paths = ruby_initial_load_paths;
#if defined LOAD_RELATIVE
#if !defined ENABLE_MULTIARCH
# define RUBY_ARCH_PATH ""
#elif defined RUBY_ARCH
# define RUBY_ARCH_PATH "/"RUBY_ARCH
#else
# define RUBY_ARCH_PATH "/"RUBY_PLATFORM
#endif
char *libpath;
VALUE sopath;
size_t baselen;
const char *p;
sopath = runtime_libruby_path();
libpath = RSTRING_PTR(sopath);
p = strrchr(libpath, '/');
if (p) {
static const char libdir[] = "/"
#ifdef LIBDIR_BASENAME
LIBDIR_BASENAME
#else
"lib"
#endif
RUBY_ARCH_PATH;
const ptrdiff_t libdir_len = (ptrdiff_t)sizeof(libdir)
- rb_strlen_lit(RUBY_ARCH_PATH) - 1;
static const char bindir[] = "/bin";
const ptrdiff_t bindir_len = (ptrdiff_t)sizeof(bindir) - 1;
const char *p2 = NULL;
#ifdef ENABLE_MULTIARCH
multiarch:
#endif
if (p - libpath >= bindir_len && !STRNCASECMP(p - bindir_len, bindir, bindir_len)) {
p -= bindir_len;
archlibdir = rb_str_subseq(sopath, 0, p - libpath);
rb_str_cat_cstr(archlibdir, libdir);
OBJ_FREEZE_RAW(archlibdir);
}
else if (p - libpath >= libdir_len && !strncmp(p - libdir_len, libdir, libdir_len)) {
archlibdir = rb_str_subseq(sopath, 0, (p2 ? p2 : p) - libpath);
OBJ_FREEZE_RAW(archlibdir);
p -= libdir_len;
}
#ifdef ENABLE_MULTIARCH
else if (p2) {
p = p2;
}
else {
p2 = p;
p = rb_enc_path_last_separator(libpath, p, rb_ascii8bit_encoding());
if (p) goto multiarch;
p = p2;
}
#endif
}
baselen = p - libpath;
rb_str_resize(sopath, baselen);
libpath = RSTRING_PTR(sopath);
#define PREFIX_PATH() sopath
#define BASEPATH() rb_str_buf_cat(rb_str_buf_new(baselen+len), libpath, baselen)
#define RUBY_RELATIVE(path, len) rb_str_buf_cat(BASEPATH(), (path), (len))
#else
const size_t exec_prefix_len = strlen(ruby_exec_prefix);
#define RUBY_RELATIVE(path, len) rubylib_path_new((path), (len))
#define PREFIX_PATH() RUBY_RELATIVE(ruby_exec_prefix, exec_prefix_len)
#endif
rb_gc_register_address(&ruby_prefix_path);
ruby_prefix_path = PREFIX_PATH();
OBJ_FREEZE_RAW(ruby_prefix_path);
if (!archlibdir) archlibdir = ruby_prefix_path;
rb_gc_register_address(&ruby_archlibdir_path);
ruby_archlibdir_path = archlibdir;
load_path = GET_VM()->load_path;
ruby_push_include(getenv("RUBYLIB"), identical_path);
id_initial_load_path_mark = INITIAL_LOAD_PATH_MARK;
while (*paths) {
size_t len = strlen(paths);
VALUE path = RUBY_RELATIVE(paths, len);
rb_ivar_set(path, id_initial_load_path_mark, path);
rb_ary_push(load_path, path);
paths += len + 1;
}
rb_const_set(rb_cObject, rb_intern_const("TMP_RUBY_PREFIX"), ruby_prefix_path);
}
static void
add_modules(VALUE *req_list, const char *mod)
{
VALUE list = *req_list;
VALUE feature;
if (!list) {
*req_list = list = rb_ary_tmp_new(0);
}
feature = rb_str_cat_cstr(rb_str_tmp_new(0), mod);
rb_ary_push(list, feature);
}
static void
require_libraries(VALUE *req_list)
{
VALUE list = *req_list;
VALUE self = rb_vm_top_self();
ID require;
rb_encoding *extenc = rb_default_external_encoding();
CONST_ID(require, "require");
while (list && RARRAY_LEN(list) > 0) {
VALUE feature = rb_ary_shift(list);
rb_enc_associate(feature, extenc);
* include/ruby/ruby.h: constify RBasic::klass and add RBASIC_CLASS(obj) macro which returns a class of `obj'. This change is a part of RGENGC branch [ruby-trunk - Feature #8339]. * object.c: add new function rb_obj_reveal(). This function reveal interal (hidden) object by rb_obj_hide(). Note that do not change class before and after hiding. Only permitted example is: klass = RBASIC_CLASS(obj); rb_obj_hide(obj); .... rb_obj_reveal(obj, klass); TODO: API design. rb_obj_reveal() should be replaced with others. TODO: modify constified variables using cast may be harmful for compiler's analysis and optimizaton. Any idea to prohibt inserting RBasic::klass directly? If rename RBasic::klass and force to use RBASIC_CLASS(obj), then all codes such as `RBASIC(obj)->klass' will be compilation error. Is it acceptable? (We have similar experience at Ruby 1.9, for example "RARRAY(ary)->ptr" to "RARRAY_PTR(ary)". * internal.h: add some macros. * RBASIC_CLEAR_CLASS(obj) clear RBasic::klass to make it internal object. * RBASIC_SET_CLASS(obj, cls) set RBasic::klass. * RBASIC_SET_CLASS_RAW(obj, cls) same as RBASIC_SET_CLASS without write barrier (planned). * RCLASS_SET_SUPER(a, b) set super class of a. * array.c, class.c, compile.c, encoding.c, enum.c, error.c, eval.c, file.c, gc.c, hash.c, io.c, iseq.c, marshal.c, object.c, parse.y, proc.c, process.c, random.c, ruby.c, sprintf.c, string.c, thread.c, transcode.c, vm.c, vm_eval.c, win32/file.c: Use above macros and functions to access RBasic::klass. * ext/coverage/coverage.c, ext/readline/readline.c, ext/socket/ancdata.c, ext/socket/init.c, * ext/zlib/zlib.c: ditto. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@40691 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2013-05-13 14:49:11 +04:00
RBASIC_SET_CLASS_RAW(feature, rb_cString);
OBJ_FREEZE(feature);
rb_funcallv(self, require, 1, &feature);
}
*req_list = 0;
}
static const struct rb_block*
toplevel_context(rb_binding_t *bind)
{
return &bind->block;
}
static void
process_sflag(int *sflag)
{
if (*sflag > 0) {
long n;
const VALUE *args;
VALUE argv = rb_argv;
n = RARRAY_LEN(argv);
args = RARRAY_CONST_PTR(argv);
while (n > 0) {
* eval.c (block_pass): should not downgrade safe level. * ext/dbm/extconf.rb: allow specifying dbm-type explicitly. * ext/dbm/extconf.rb: avoid gdbm if possible, because it leaks memory, whereas gdbm.so doesn't. potential incompatibility. * string.c (rb_str_insert): new method. * parse.y (yylex): lex_state after RESCUE_MOD should be EXPR_BEG. * array.c (rb_ary_insert): new method. * array.c (rb_ary_update): new utility function. * io.c (set_outfile): should check if closed before assignment. * eval.c (rb_eval): should preserve value of ruby_errinfo. * eval.c (rb_thread_schedule): infinite sleep should not cause dead lock. * array.c (rb_ary_flatten_bang): proper recursive detection. * eval.c (yield_under): need not to prohibit at safe level 4. * pack.c (pack_pack): p/P packs nil into NULL. * pack.c (pack_unpack): p/P unpacks NULL into nil. * pack.c (pack_pack): size check for P template. * ruby.c (set_arg0): wrong predicate when new $0 value is bigger than original space. * gc.c (id2ref): should use NUM2ULONG() * object.c (rb_mod_const_get): check whether name is a class variable name. * object.c (rb_mod_const_set): ditto. * object.c (rb_mod_const_defined): ditto. * marshal.c (w_float): precision changed to "%.16g" * eval.c (rb_call0): wrong retry behavior. * numeric.c (fix_aref): a bug on long>int architecture. * eval.c (rb_eval_string_wrap): should restore ruby_wrapper. * regex.c (re_compile_pattern): char class at either edge of range should be invalid. * eval.c (handle_rescue): use === to compare exception match. * error.c (syserr_eqq): comparison between SytemCallErrors should based on their error numbers. * eval.c (safe_getter): should use INT2NUM(). * bignum.c (rb_big2long): 2**31 cannot fit in 31 bit long. * regex.c (calculate_must_string): wrong length calculation. * eval.c (rb_thread_start_0): fixed memory leak. * parse.y (none): should clear cmdarg_stack too. * io.c (rb_fopen): use setvbuf() to avoid recursive malloc() on some platforms. * file.c (rb_stat_dev): device functions should honor stat field types (except long long such as dev_t). * eval.c (rb_mod_nesting): should not push nil for nesting array. * eval.c (rb_mod_s_constants): should not search array by rb_mod_const_at() for nil (happens for singleton class). * class.c (rb_singleton_class_attached): should modify iv_tbl by itself, no longer use rb_iv_set() to avoid freeze check error. * variable.c (rb_const_get): error message "uninitialized constant Foo at Bar::Baz" instead of "uninitialized constantBar::Baz::Foo". * eval.c (rb_mod_included): new hook called from rb_mod_include(). * io.c (opt_i_set): should strdup() inplace_edit string. * eval.c (exec_under): need to push cref too. * eval.c (rb_f_missing): raise NameError for "undefined local variable or method". * error.c (Init_Exception): new exception NoMethodError. NameError moved under ScriptError again. * eval.c (rb_f_missing): use NoMethodError instead of NameError. * file.c (Init_File): should redifine "new" class method. * eval.c (PUSH_CREF): sharing cref node was problematic. maintain runtime cref list instead. * eval.c (rb_eval): copy defn node before registering. * eval.c (rb_load): clear ruby_cref before loading. * variable.c (rb_const_get): no recursion to show full class path for modules. * eval.c (rb_set_safe_level): should set safe level in curr_thread as well. * eval.c (safe_setter): ditto. * object.c (rb_obj_is_instance_of): nil belongs to false, not true. * time.c (make_time_t): proper (I hope) daylight saving time handling for both US and Europe. I HATE DST! * eval.c (rb_thread_wait_for): non blocked signal interrupt should stop the interval. * eval.c (proc_eq): class check aded. * eval.c (proc_eq): typo fixed ("return" was ommitted). * error.c (Init_Exception): move NameError under StandardError. * class.c (rb_mod_clone): should copy method bodies too. * bignum.c (bigdivrem): should trim trailing zero bdigits of remainder, even if dd == 0. * file.c (check3rdbyte): safe string check moved here. * time.c (make_time_t): remove HAVE_TM_ZONE code since it sometimes reports wrong time. * time.c (make_time_t): remove unnecessary range check for platforms where negative time_t is available. * process.c (proc_waitall): should push Process::Status instead of Finuxm status. * process.c (waitall_each): should add all entries in pid_tbl. these changes are inspired by Koji Arai. Thanks. * process.c (proc_wait): should not iterate if pid_tbl is 0. * process.c (proc_waitall): ditto. * numeric.c (flodivmod): a bug in no fmod case. * process.c (pst_wifsignaled): should apply WIFSIGNALED for status (int), not st (VALUE). * io.c (Init_IO): value of $/ and $\ are no longer restricted to strings. type checks are done on demand. * class.c (rb_include_module): module inclusion should be check taints. * ruby.h (STR2CSTR): replace to StringType() and StringTypePtr(). * ruby.h (rb_str2cstr): ditto. * eval.c (rb_load): should not copy topleve local variables. It cause variable/method ambiguity. Thanks to L. Peter Deutsch. * class.c (rb_include_module): freeze check at first. * eval.c (rb_attr): sprintf() and rb_intern() moved into conditional body. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@1356 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2001-05-02 08:22:21 +04:00
VALUE v = *args++;
char *s = StringValuePtr(v);
char *p;
int hyphen = FALSE;
if (s[0] != '-')
break;
n--;
if (s[1] == '-' && s[2] == '\0')
break;
v = Qtrue;
/* check if valid name before replacing - with _ */
for (p = s + 1; *p; p++) {
if (*p == '=') {
*p++ = '\0';
v = rb_str_new2(p);
break;
}
if (*p == '-') {
hyphen = TRUE;
}
else if (*p != '_' && !ISALNUM(*p)) {
VALUE name_error[2];
name_error[0] =
rb_str_new2("invalid name for global variable - ");
if (!(p = strchr(p, '='))) {
rb_str_cat2(name_error[0], s);
}
else {
rb_str_cat(name_error[0], s, p - s);
}
name_error[1] = args[-1];
rb_exc_raise(rb_class_new_instance(2, name_error, rb_eNameError));
}
}
s[0] = '$';
if (hyphen) {
for (p = s + 1; *p; ++p) {
if (*p == '-')
*p = '_';
}
}
rb_gv_set(s, v);
}
n = RARRAY_LEN(argv) - n;
while (n--) {
rb_ary_shift(argv);
}
*sflag = -1;
}
}
static long proc_options(long argc, char **argv, ruby_cmdline_options_t *opt, int envopt);
static void
moreswitches(const char *s, ruby_cmdline_options_t *opt, int envopt)
{
long argc, i, len;
char **argv, *p;
const char *ap = 0;
VALUE argstr, argary;
void *ptr;
while (ISSPACE(*s)) s++;
if (!*s) return;
argstr = rb_str_tmp_new((len = strlen(s)) + (envopt!=0));
argary = rb_str_tmp_new(0);
p = RSTRING_PTR(argstr);
if (envopt) *p++ = ' ';
memcpy(p, s, len + 1);
ap = 0;
rb_str_cat(argary, (char *)&ap, sizeof(ap));
while (*p) {
ap = p;
rb_str_cat(argary, (char *)&ap, sizeof(ap));
while (*p && !ISSPACE(*p)) ++p;
if (!*p) break;
*p++ = '\0';
while (ISSPACE(*p)) ++p;
}
argc = RSTRING_LEN(argary) / sizeof(ap);
ap = 0;
rb_str_cat(argary, (char *)&ap, sizeof(ap));
argv = ptr = ALLOC_N(char *, argc);
MEMMOVE(argv, RSTRING_PTR(argary), char *, argc);
while ((i = proc_options(argc, argv, opt, envopt)) > 1 && envopt && (argc -= i) > 0) {
argv += i;
if (**argv != '-') {
*--*argv = '-';
}
if ((*argv)[1]) {
++argc;
--argv;
}
}
ruby_xfree(ptr);
/* get rid of GC */
rb_str_resize(argary, 0);
rb_str_resize(argstr, 0);
}
static int
name_match_p(const char *name, const char *str, size_t len)
{
if (len == 0) return 0;
while (1) {
while (TOLOWER(*str) == *name) {
if (!--len || !*++str) return 1;
++name;
}
if (*str != '-' && *str != '_') return 0;
while (ISALNUM(*name)) name++;
if (*name != '-' && *name != '_') return 0;
++name;
++str;
}
}
#define NAME_MATCH_P(name, str, len) \
((len) < (int)sizeof(name) && name_match_p((name), (str), (len)))
#define UNSET_WHEN(name, bit, str, len) \
if (NAME_MATCH_P((name), (str), (len))) { \
*(unsigned int *)arg &= ~(bit); \
return; \
}
#define SET_WHEN(name, bit, str, len) \
if (NAME_MATCH_P((name), (str), (len))) { \
*(unsigned int *)arg |= (bit); \
return; \
}
#define LITERAL_NAME_ELEMENT(name) #name
static void
feature_option(const char *str, int len, void *arg, const unsigned int enable)
{
static const char list[] = EACH_FEATURES(LITERAL_NAME_ELEMENT, ", ");
ruby_features_t *argp = arg;
unsigned int mask = ~0U;
unsigned int set = 0U;
#if AMBIGUOUS_FEATURE_NAMES
int matched = 0;
# define FEATURE_FOUND ++matched
#else
# define FEATURE_FOUND goto found
#endif
#define SET_FEATURE(bit) \
if (NAME_MATCH_P(#bit, str, len)) {set |= mask = FEATURE_BIT(bit); FEATURE_FOUND;}
EACH_FEATURES(SET_FEATURE, ;);
if (NAME_MATCH_P("all", str, len)) {
goto found;
}
#if AMBIGUOUS_FEATURE_NAMES
if (matched == 1) goto found;
if (matched > 1) {
VALUE mesg = rb_sprintf("ambiguous feature: `%.*s' (", len, str);
#define ADD_FEATURE_NAME(bit) \
if (FEATURE_BIT(bit) & set) { \
rb_str_cat_cstr(mesg, #bit); \
if (--matched) rb_str_cat_cstr(mesg, ", "); \
}
EACH_FEATURES(ADD_FEATURE_NAME, ;);
rb_str_cat_cstr(mesg, ")");
rb_exc_raise(rb_exc_new_str(rb_eRuntimeError, mesg));
#undef ADD_FEATURE_NAME
}
#endif
rb_warn("unknown argument for --%s: `%.*s'",
enable ? "enable" : "disable", len, str);
rb_warn("features are [%.*s].", (int)strlen(list), list);
return;
found:
FEATURE_SET_TO(*argp, mask, (mask & enable));
return;
}
static void
enable_option(const char *str, int len, void *arg)
{
feature_option(str, len, arg, ~0U);
}
static void
disable_option(const char *str, int len, void *arg)
{
feature_option(str, len, arg, 0U);
}
RUBY_EXTERN const int ruby_patchlevel;
int ruby_env_debug_option(const char *str, int len, void *arg);
static void
debug_option(const char *str, int len, void *arg)
{
static const char list[] = EACH_DEBUG_FEATURES(LITERAL_NAME_ELEMENT, ", ");
ruby_features_t *argp = arg;
#define SET_WHEN_DEBUG(bit) \
if (NAME_MATCH_P(#bit, str, len)) { \
FEATURE_SET(*argp, DEBUG_BIT(bit)); \
return; \
}
EACH_DEBUG_FEATURES(SET_WHEN_DEBUG, ;);
#ifdef RUBY_DEVEL
if (ruby_patchlevel < 0 && ruby_env_debug_option(str, len, 0)) return;
#endif
rb_warn("unknown argument for --debug: `%.*s'", len, str);
rb_warn("debug features are [%.*s].", (int)strlen(list), list);
}
static void
dump_option(const char *str, int len, void *arg)
{
static const char list[] = EACH_DUMPS(LITERAL_NAME_ELEMENT, ", ");
#define SET_WHEN_DUMP(bit) SET_WHEN(#bit, DUMP_BIT(bit), str, len)
EACH_DUMPS(SET_WHEN_DUMP, ;);
rb_warn("don't know how to dump `%.*s',", len, str);
rb_warn("but only [%.*s].", (int)strlen(list), list);
}
static void
set_option_encoding_once(const char *type, VALUE *name, const char *e, long elen)
{
VALUE ename;
if (!elen) elen = strlen(e);
ename = rb_str_new(e, elen);
if (*name &&
rb_funcall(ename, rb_intern("casecmp"), 1, *name) != INT2FIX(0)) {
rb_raise(rb_eRuntimeError,
"%s already set to %"PRIsVALUE, type, *name);
}
*name = ename;
}
#define set_internal_encoding_once(opt, e, elen) \
set_option_encoding_once("default_internal", &(opt)->intern.enc.name, (e), (elen))
#define set_external_encoding_once(opt, e, elen) \
set_option_encoding_once("default_external", &(opt)->ext.enc.name, (e), (elen))
#define set_source_encoding_once(opt, e, elen) \
set_option_encoding_once("source", &(opt)->src.enc.name, (e), (elen))
#if USE_MJIT
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
static void
setup_mjit_options(const char *s, struct mjit_options *mjit_opt)
{
#define opt_match(s, l, name) \
((((l) > rb_strlen_lit(name)) ? (s)[rb_strlen_lit(name)] == '=' : \
(l) == rb_strlen_lit(name)) && \
memcmp((s), name, rb_strlen_lit(name)) == 0 && \
(((s) += rb_strlen_lit(name)), 1))
#define opt_match_noarg(s, l, name) \
opt_match(s, l, name) && (*(s) ? (rb_warn("argument to --jit-" name " is ignored"), 1) : 1)
#define opt_match_arg(s, l, name) \
opt_match(s, l, name) && (*(s) ? 1 : (rb_raise(rb_eRuntimeError, "--jit-" name " needs an argument"), 0))
if (*s != '-') return;
const size_t l = strlen(++s);
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
if (*s == 0) return;
else if (opt_match_noarg(s, l, "warnings")) {
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
mjit_opt->warnings = 1;
}
else if (opt_match(s, l, "debug")) {
if (*s)
mjit_opt->debug_flags = strdup(s + 1);
else
mjit_opt->debug = 1;
}
else if (opt_match_noarg(s, l, "wait")) {
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
mjit_opt->wait = 1;
}
else if (opt_match_noarg(s, l, "save-temps")) {
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
mjit_opt->save_temps = 1;
}
else if (opt_match(s, l, "verbose")) {
mjit_opt->verbose = *s ? atoi(s + 1) : 1;
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
}
else if (opt_match_arg(s, l, "max-cache")) {
mjit_opt->max_cache_size = atoi(s + 1);
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
}
else if (opt_match_arg(s, l, "min-calls")) {
mjit_opt->min_calls = atoi(s + 1);
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
}
else {
rb_raise(rb_eRuntimeError,
"invalid MJIT option `%s' (--help will show valid MJIT options)", s);
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
}
}
#endif
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
static long
proc_options(long argc, char **argv, ruby_cmdline_options_t *opt, int envopt)
{
long n, argc0 = argc;
const char *s;
int warning = opt->warning;
if (argc <= 0 || !argv)
return 0;
for (argc--, argv++; argc > 0; argc--, argv++) {
const char *const arg = argv[0];
if (!arg || arg[0] != '-' || !arg[1])
break;
s = arg + 1;
reswitch:
switch (*s) {
case 'a':
if (envopt) goto noenvopt;
opt->do_split = TRUE;
s++;
goto reswitch;
case 'p':
if (envopt) goto noenvopt;
opt->do_print = TRUE;
/* through */
case 'n':
if (envopt) goto noenvopt;
opt->do_loop = TRUE;
s++;
goto reswitch;
case 'd':
ruby_debug = Qtrue;
ruby_verbose = Qtrue;
s++;
goto reswitch;
case 'y':
if (envopt) goto noenvopt;
opt->dump |= DUMP_BIT(yydebug);
s++;
goto reswitch;
case 'v':
if (opt->verbose) {
s++;
goto reswitch;
}
opt->dump |= DUMP_BIT(version_v);
opt->verbose = 1;
case 'w':
if (!opt->warning) {
warning = 1;
ruby_verbose = Qtrue;
}
FEATURE_SET(opt->warn, RB_WARN_CATEGORY_ALL_BITS);
s++;
goto reswitch;
case 'W':
if (s[1] == ':') {
unsigned int bits = 0;
static const char no_prefix[] = "no-";
int enable = strncmp(s += 2, no_prefix, sizeof(no_prefix)-1) != 0;
if (!enable) s += sizeof(no_prefix)-1;
size_t len = strlen(s);
if (NAME_MATCH_P("deprecated", s, len)) {
bits = 1U << RB_WARN_CATEGORY_DEPRECATED;
}
else if (NAME_MATCH_P("experimental", s, len)) {
bits = 1U << RB_WARN_CATEGORY_EXPERIMENTAL;
}
else {
rb_warn("unknown warning category: `%s'", s);
}
if (bits) FEATURE_SET_TO(opt->warn, bits, enable ? bits : 0);
break;
}
{
size_t numlen;
int v = 2; /* -W as -W2 */
if (*++s) {
v = scan_oct(s, 1, &numlen);
if (numlen == 0)
v = 1;
s += numlen;
}
if (!opt->warning) {
switch (v) {
case 0:
ruby_verbose = Qnil;
break;
case 1:
ruby_verbose = Qfalse;
break;
default:
ruby_verbose = Qtrue;
break;
}
}
warning = 1;
switch (v) {
case 0:
FEATURE_SET_TO(opt->warn, RB_WARN_CATEGORY_ALL_BITS, 0);
break;
case 1:
FEATURE_SET_TO(opt->warn, 1U << RB_WARN_CATEGORY_DEPRECATED, 0);
break;
default:
FEATURE_SET(opt->warn, RB_WARN_CATEGORY_ALL_BITS);
break;
}
}
goto reswitch;
case 'c':
if (envopt) goto noenvopt;
opt->dump |= DUMP_BIT(syntax);
s++;
goto reswitch;
case 's':
if (envopt) goto noenvopt;
forbid_setid("-s");
if (!opt->sflag) opt->sflag = 1;
s++;
goto reswitch;
case 'h':
if (envopt) goto noenvopt;
opt->dump |= DUMP_BIT(usage);
goto switch_end;
case 'l':
if (envopt) goto noenvopt;
opt->do_line = TRUE;
rb_output_rs = rb_rs;
s++;
goto reswitch;
case 'S':
if (envopt) goto noenvopt;
forbid_setid("-S");
opt->do_search = TRUE;
s++;
goto reswitch;
case 'e':
if (envopt) goto noenvopt;
forbid_setid("-e");
if (!*++s) {
if (!--argc)
rb_raise(rb_eRuntimeError, "no code specified for -e");
s = *++argv;
}
if (!opt->e_script) {
opt->e_script = rb_str_new(0, 0);
if (opt->script == 0)
opt->script = "-e";
}
rb_str_cat2(opt->e_script, s);
rb_str_cat2(opt->e_script, "\n");
break;
case 'r':
forbid_setid("-r");
if (*++s) {
add_modules(&opt->req_list, s);
}
else if (argc > 1) {
add_modules(&opt->req_list, argv[1]);
argc--, argv++;
}
break;
case 'i':
if (envopt) goto noenvopt;
forbid_setid("-i");
ruby_set_inplace_mode(s + 1);
break;
case 'x':
if (envopt) goto noenvopt;
forbid_setid("-x");
opt->xflag = TRUE;
s++;
if (*s && chdir(s) < 0) {
rb_fatal("Can't chdir to %s", s);
}
break;
case 'C':
case 'X':
if (envopt) goto noenvopt;
if (!*++s && (!--argc || !(s = *++argv) || !*s)) {
rb_fatal("Can't chdir");
}
if (chdir(s) < 0) {
rb_fatal("Can't chdir to %s", s);
}
break;
case 'F':
if (envopt) goto noenvopt;
if (*++s) {
rb_fs = rb_reg_new(s, strlen(s), 0);
}
break;
case 'E':
if (!*++s && (!--argc || !(s = *++argv))) {
rb_raise(rb_eRuntimeError, "missing argument for -E");
}
goto encoding;
case 'U':
set_internal_encoding_once(opt, "UTF-8", 0);
++s;
goto reswitch;
case 'K':
if (*++s) {
const char *enc_name = 0;
switch (*s) {
case 'E': case 'e':
enc_name = "EUC-JP";
break;
case 'S': case 's':
enc_name = "Windows-31J";
break;
case 'U': case 'u':
enc_name = "UTF-8";
break;
case 'N': case 'n': case 'A': case 'a':
enc_name = "ASCII-8BIT";
break;
}
if (enc_name) {
opt->src.enc.name = rb_str_new2(enc_name);
if (!opt->ext.enc.name)
opt->ext.enc.name = opt->src.enc.name;
}
s++;
}
goto reswitch;
case 'I':
forbid_setid("-I");
if (*++s)
ruby_incpush_expand(s);
else if (argc > 1) {
ruby_incpush_expand(argv[1]);
argc--, argv++;
}
break;
case '0':
if (envopt) goto noenvopt;
{
size_t numlen;
int v;
char c;
v = scan_oct(s, 4, &numlen);
s += numlen;
if (v > 0377)
rb_rs = Qnil;
else if (v == 0 && numlen >= 2) {
rb_rs = rb_str_new2("");
}
else {
c = v & 0xff;
rb_rs = rb_str_new(&c, 1);
}
}
goto reswitch;
case '-':
if (!s[1] || (s[1] == '\r' && !s[2])) {
argc--, argv++;
goto switch_end;
}
s++;
# define is_option_end(c, allow_hyphen) \
(!(c) || ((allow_hyphen) && (c) == '-') || (c) == '=')
# define check_envopt(name, allow_envopt) \
(((allow_envopt) || !envopt) ? (void)0 : \
rb_raise(rb_eRuntimeError, "invalid switch in RUBYOPT: --" name))
# define need_argument(name, s, needs_arg, next_arg) \
((*(s) ? !*++(s) : (next_arg) && (!argc || !((s) = argv[1]) || (--argc, ++argv, 0))) && (needs_arg) ? \
rb_raise(rb_eRuntimeError, "missing argument for --" name) \
: (void)0)
# define is_option_with_arg(name, allow_hyphen, allow_envopt) \
is_option_with_optarg(name, allow_hyphen, allow_envopt, Qtrue, Qtrue)
# define is_option_with_optarg(name, allow_hyphen, allow_envopt, needs_arg, next_arg) \
(strncmp((name), s, n = sizeof(name) - 1) == 0 && is_option_end(s[n], (allow_hyphen)) ? \
(check_envopt(name, (allow_envopt)), s += n, \
need_argument(name, s, needs_arg, next_arg), 1) : 0)
if (strcmp("copyright", s) == 0) {
if (envopt) goto noenvopt_long;
opt->dump |= DUMP_BIT(copyright);
}
else if (is_option_with_optarg("debug", Qtrue, Qtrue, Qfalse, Qfalse)) {
if (s && *s) {
ruby_each_words(s, debug_option, &opt->features);
}
else {
ruby_debug = Qtrue;
ruby_verbose = Qtrue;
}
}
else if (is_option_with_arg("enable", Qtrue, Qtrue)) {
ruby_each_words(s, enable_option, &opt->features);
}
else if (is_option_with_arg("disable", Qtrue, Qtrue)) {
ruby_each_words(s, disable_option, &opt->features);
}
else if (is_option_with_arg("encoding", Qfalse, Qtrue)) {
char *p;
encoding:
do {
# define set_encoding_part(type) \
if (!(p = strchr(s, ':'))) { \
set_##type##_encoding_once(opt, s, 0); \
break; \
} \
else if (p > s) { \
set_##type##_encoding_once(opt, s, p-s); \
}
set_encoding_part(external);
if (!*(s = ++p)) break;
set_encoding_part(internal);
if (!*(s = ++p)) break;
#if defined ALLOW_DEFAULT_SOURCE_ENCODING && ALLOW_DEFAULT_SOURCE_ENCODING
set_encoding_part(source);
if (!*(s = ++p)) break;
#endif
rb_raise(rb_eRuntimeError, "extra argument for %s: %s",
(arg[1] == '-' ? "--encoding" : "-E"), s);
# undef set_encoding_part
} while (0);
}
else if (is_option_with_arg("internal-encoding", Qfalse, Qtrue)) {
set_internal_encoding_once(opt, s, 0);
}
else if (is_option_with_arg("external-encoding", Qfalse, Qtrue)) {
set_external_encoding_once(opt, s, 0);
}
#if defined ALLOW_DEFAULT_SOURCE_ENCODING && ALLOW_DEFAULT_SOURCE_ENCODING
else if (is_option_with_arg("source-encoding", Qfalse, Qtrue)) {
set_source_encoding_once(opt, s, 0);
}
#endif
else if (strcmp("version", s) == 0) {
if (envopt) goto noenvopt_long;
opt->dump |= DUMP_BIT(version);
}
else if (strcmp("verbose", s) == 0) {
opt->verbose = 1;
ruby_verbose = Qtrue;
}
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
else if (strncmp("jit", s, 3) == 0) {
#if USE_MJIT
FEATURE_SET(opt->features, FEATURE_BIT(jit));
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
setup_mjit_options(s + 3, &opt->mjit);
#else
rb_warn("MJIT support is disabled.");
#endif
mjit.c: merge MJIT infrastructure that allows to JIT-compile Ruby methods by generating C code and using C compiler. See the first comment of mjit.c to know what this file does. mjit.c is authored by Vladimir Makarov <vmakarov@redhat.com>. After he invented great method JIT infrastructure for MRI as MJIT, Lars Kanis <lars@greiz-reinsdorf.de> sent the patch to support MinGW in MJIT. In addition to merging it, I ported pthread to Windows native threads. Now this MJIT infrastructure can be compiled on Visual Studio. This commit simplifies mjit.c to decrease code at initial merge. For example, this commit does not provide multiple JIT threads support. We can resurrect them later if we really want them, but I wanted to minimize diff to make it easier to review this patch. `/tmp/_mjitXXX` file is renamed to `/tmp/_ruby_mjitXXX` because non-Ruby developers may not know the name "mjit" and the file name should make sure it's from Ruby and not from some harmful programs. TODO: it may be better to store this to some temporary directory which Ruby is already using by Tempfile, if it's not bad for performance. mjit.h: New. It has `mjit_exec` interface similar to `vm_exec`, which is for triggering MJIT. This drops interface for AOT compared to the original MJIT. Makefile.in: define macros to let MJIT know the path of MJIT header. Probably we can refactor this to reduce the number of macros (TODO). win32/Makefile.sub: ditto. common.mk: compile mjit.o and mjit_compile.o. Unlike original MJIT, this commit separates MJIT infrastructure and JIT compiler code as independent object files. As initial patch is NOT going to have ultra-fast JIT compiler, it's likely to replace JIT compiler, e.g. original MJIT's compiler or some future JIT impelementations which are not public now. inits.c: define MJIT module. This is added because `MJIT.enabled?` was necessary for testing. test/lib/zombie_hunter.rb: skip if `MJIT.enabled?`. Obviously this wouldn't work with current code when JIT is enabled. test/ruby/test_io.rb: skip this too. This would make no sense with MJIT. ruby.c: define MJIT CLI options. As major difference from original MJIT, "-j:l"/"--jit:llvm" are renamed to "--jit-cc" because I want to support not only gcc/clang but also cl.exe (Visual Studio) in the future. But it takes only "--jit-cc=gcc", "--jit-cc=clang" for now. And only long "--jit" options are allowed since some Ruby committers preferred it at Ruby developers Meeting on January, and some of options are renamed. This file also triggers to initialize MJIT thread and variables. eval.c: finalize MJIT worker thread and variables. test/ruby/test_rubyoptions.rb: fix number of CLI options for --jit. thread_pthread.c: change for pthread abstraction in MJIT. Prefix rb_ for functions which are used by other files. thread_win32.c: ditto, for Windows. Those pthread porting is one of major works that YARV-MJIT created, which is my fork of MJIT, in Feature 14235. thread.c: follow rb_ prefix changes vm.c: trigger MJIT call on VM invocation. Also trigger `mjit_mark` to avoid SEGV by race between JIT and GC of ISeq. The improvement was provided by wanabe <s.wanabe@gmail.com>. In JIT compiler I created and am going to add in my next commit, I found that having `mjit_exec` after `vm_loop_start:` is harmful because the JIT-ed function doesn't proceed other ISeqs on RESTORE_REGS of leave insn. Executing non-FINISH frame is unexpected for my JIT compiler and `exception_handler` triggers executions of such ISeqs. So `mjit_exec` here should be executed only when it directly comes from `vm_exec` call. `RubyVM::MJIT` module and `.enabled?` method is added so that we can skip some tests which don't expect JIT threads or compiler file descriptors. vm_insnhelper.h: trigger MJIT on method calls during VM execution. vm_core.h: add fields required for mjit.c. `bp` must be `cfp[6]` because rb_control_frame_struct is likely to be casted to another struct. The last position is the safest place to add the new field. vm_insnhelper.c: save initial value of cfp->ep as cfp->bp. This is an optimization which are done in both MJIT and YARV-MJIT. So this change is added in this commit. Calculating bp from ep is a little heavy work, so bp is kind of cache for it. iseq.c: notify ISeq GC to MJIT. We should know which iseq in MJIT queue is GCed to avoid SEGV. TODO: unload some GCed units in some safe way. gc.c: add hooks so that MJIT can wait GC, and vice versa. Simultaneous JIT and GC executions may cause SEGV and so we should synchronize them. cont.c: save continuation information in MJIT worker. As MJIT shouldn't unload JIT-ed code which is being used, MJIT wants to know full list of saved execution contexts for continuation and detect ISeqs in use. mjit_compile.c: added empty JIT compiler so that you can reuse this commit to build your own JIT compiler. This commit tries to compile ISeqs but all of them are considered as not supported in this commit. So you can't use JIT compiler in this commit yet while we added --jit option now. Patch author: Vladimir Makarov <vmakarov@redhat.com>. Contributors: Takashi Kokubun <takashikkbn@gmail.com>. wanabe <s.wanabe@gmail.com>. Lars Kanis <lars@greiz-reinsdorf.de>. Part of Feature 12589 and 14235. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62189 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-02-04 09:58:09 +03:00
}
else if (strcmp("yydebug", s) == 0) {
if (envopt) goto noenvopt_long;
opt->dump |= DUMP_BIT(yydebug);
}
else if (is_option_with_arg("dump", Qfalse, Qfalse)) {
ruby_each_words(s, dump_option, &opt->dump);
}
else if (strcmp("help", s) == 0) {
if (envopt) goto noenvopt_long;
opt->dump |= DUMP_BIT(help);
goto switch_end;
}
else if (is_option_with_arg("backtrace-limit", Qfalse, Qfalse)) {
char *e;
long n = strtol(s, &e, 10);
if (errno == ERANGE || n < 0 || *e) rb_raise(rb_eRuntimeError, "wrong limit for backtrace length");
rb_backtrace_length_limit = n;
}
else {
rb_raise(rb_eRuntimeError,
"invalid option --%s (-h will show valid options)", s);
}
break;
case '\r':
if (!s[1])
break;
default:
{
rb_raise(rb_eRuntimeError,
"invalid option -%c (-h will show valid options)",
(int)(unsigned char)*s);
}
goto switch_end;
noenvopt:
/* "EIdvwWrKU" only */
rb_raise(rb_eRuntimeError, "invalid switch in RUBYOPT: -%c", *s);
break;
noenvopt_long:
rb_raise(rb_eRuntimeError, "invalid switch in RUBYOPT: --%s", s);
break;
case 0:
break;
# undef is_option_end
# undef check_envopt
# undef need_argument
# undef is_option_with_arg
# undef is_option_with_optarg
}
}
switch_end:
if (warning) opt->warning = warning;
return argc0 - argc;
}
void Init_builtin_features(void);
static void
ruby_init_prelude(void)
{
Init_builtin_features();
rb_const_remove(rb_cObject, rb_intern_const("TMP_RUBY_PREFIX"));
}
void rb_call_builtin_inits(void);
static void
ruby_opt_init(ruby_cmdline_options_t *opt)
{
if (opt->dump & dump_exit_bits) return;
if (opt->features.set & FEATURE_BIT(gems)) {
rb_define_module("Gem");
if (opt->features.set & FEATURE_BIT(did_you_mean)) {
rb_define_module("DidYouMean");
}
}
Init_ext(); /* load statically linked extensions before rubygems */
rb_call_builtin_inits();
ruby_init_prelude();
ruby_set_script_name(opt->script_name);
require_libraries(&opt->req_list);
}
static int
opt_enc_index(VALUE enc_name)
{
const char *s = RSTRING_PTR(enc_name);
int i = rb_enc_find_index(s);
if (i < 0) {
rb_raise(rb_eRuntimeError, "unknown encoding name - %s", s);
}
else if (rb_enc_dummy_p(rb_enc_from_index(i))) {
rb_raise(rb_eRuntimeError, "dummy encoding is not acceptable - %s ", s);
}
return i;
}
#define rb_progname (GET_VM()->progname)
#define rb_orig_progname (GET_VM()->orig_progname)
VALUE rb_argv0;
VALUE rb_e_script;
static VALUE
false_value(ID _x, VALUE *_y)
{
return Qfalse;
}
static VALUE
true_value(ID _x, VALUE *_y)
{
return Qtrue;
}
#define rb_define_readonly_boolean(name, val) \
rb_define_virtual_variable((name), (val) ? true_value : false_value, 0)
static VALUE
uscore_get(void)
{
VALUE line;
line = rb_lastline_get();
if (!RB_TYPE_P(line, T_STRING)) {
rb_raise(rb_eTypeError, "$_ value need to be String (%s given)",
NIL_P(line) ? "nil" : rb_obj_classname(line));
}
return line;
}
/*
* call-seq:
* sub(pattern, replacement) -> $_
* sub(pattern) {|...| block } -> $_
*
* Equivalent to <code>$_.sub(<i>args</i>)</code>, except that
* <code>$_</code> will be updated if substitution occurs.
* Available only when -p/-n command line option specified.
*/
static VALUE
rb_f_sub(int argc, VALUE *argv, VALUE _)
{
VALUE str = rb_funcall_passing_block(uscore_get(), rb_intern("sub"), argc, argv);
rb_lastline_set(str);
return str;
}
/*
* call-seq:
* gsub(pattern, replacement) -> $_
* gsub(pattern) {|...| block } -> $_
*
* Equivalent to <code>$_.gsub...</code>, except that <code>$_</code>
* will be updated if substitution occurs.
* Available only when -p/-n command line option specified.
*
*/
static VALUE
rb_f_gsub(int argc, VALUE *argv, VALUE _)
{
VALUE str = rb_funcall_passing_block(uscore_get(), rb_intern("gsub"), argc, argv);
rb_lastline_set(str);
return str;
}
/*
* call-seq:
* chop -> $_
*
* Equivalent to <code>($_.dup).chop!</code>, except <code>nil</code>
* is never returned. See String#chop!.
* Available only when -p/-n command line option specified.
*
*/
static VALUE
rb_f_chop(VALUE _)
{
VALUE str = rb_funcall_passing_block(uscore_get(), rb_intern("chop"), 0, 0);
rb_lastline_set(str);
return str;
}
/*
* call-seq:
* chomp -> $_
* chomp(string) -> $_
*
* Equivalent to <code>$_ = $_.chomp(<em>string</em>)</code>. See
* String#chomp.
* Available only when -p/-n command line option specified.
*
*/
static VALUE
rb_f_chomp(int argc, VALUE *argv, VALUE _)
{
VALUE str = rb_funcall_passing_block(uscore_get(), rb_intern("chomp"), argc, argv);
rb_lastline_set(str);
return str;
}
static void
setup_pager_env(void)
{
if (!getenv("LESS")) ruby_setenv("LESS", "-R"); // Output "raw" control characters.
}
static VALUE
process_options(int argc, char **argv, ruby_cmdline_options_t *opt)
{
rb_ast_t *ast = 0;
VALUE parser;
VALUE script_name;
2015-07-22 01:52:59 +03:00
const rb_iseq_t *iseq;
rb_encoding *enc, *lenc;
#if UTF8_PATH
rb_encoding *uenc, *ienc = 0;
#endif
const char *s;
char fbuf[MAXPATHLEN];
int i = (int)proc_options(argc, argv, opt, 0);
unsigned int dump = opt->dump & dump_exit_bits;
if (opt->dump & (DUMP_BIT(usage)|DUMP_BIT(help))) {
2020-03-27 17:18:42 +03:00
int tty = isatty(1);
const char *const progname =
(argc > 0 && argv && argv[0] ? argv[0] :
origarg.argc > 0 && origarg.argv && origarg.argv[0] ? origarg.argv[0] :
ruby_engine);
2020-04-12 07:00:27 +03:00
int columns = 0;
2020-03-27 17:18:42 +03:00
if ((opt->dump & DUMP_BIT(help)) && tty) {
const char *pager_env = getenv("RUBY_PAGER");
if (!pager_env) pager_env = getenv("PAGER");
2020-03-27 17:18:42 +03:00
if (pager_env && *pager_env && isatty(0)) {
2020-04-12 07:00:27 +03:00
const char *columns_env = getenv("COLUMNS");
if (columns_env) columns = atoi(columns_env);
VALUE pager = rb_str_new_cstr(pager_env);
#ifdef HAVE_WORKING_FORK
int fds[2];
if (rb_pipe(fds) == 0) {
rb_pid_t pid = rb_fork();
if (pid > 0) {
/* exec PAGER with reading from child */
dup2(fds[0], 0);
}
else if (pid == 0) {
/* send the help message to the parent PAGER */
dup2(fds[1], 1);
dup2(fds[1], 2);
}
close(fds[0]);
close(fds[1]);
if (pid > 0) {
setup_pager_env();
rb_f_exec(1, &pager);
kill(SIGTERM, pid);
rb_waitpid(pid, 0, 0);
}
}
#else
setup_pager_env();
VALUE port = rb_io_popen(pager, rb_str_new_lit("w"), Qnil, Qnil);
if (!NIL_P(port)) {
int oldout = dup(1);
int olderr = dup(2);
int fd = RFILE(port)->fptr->fd;
dup2(fd, 1);
dup2(fd, 2);
2020-03-27 17:18:42 +03:00
/* more.com doesn't support CSI sequence */
2020-04-12 07:00:27 +03:00
usage(progname, 1, 0, columns);
fflush(stdout);
dup2(oldout, 1);
dup2(olderr, 2);
rb_io_close(port);
return Qtrue;
}
#endif
}
}
2020-04-12 07:00:27 +03:00
usage(progname, (opt->dump & DUMP_BIT(help)), tty, columns);
return Qtrue;
}
argc -= i;
argv += i;
if ((opt->features.set & FEATURE_BIT(rubyopt)) && (s = getenv("RUBYOPT"))) {
VALUE src_enc_name = opt->src.enc.name;
VALUE ext_enc_name = opt->ext.enc.name;
VALUE int_enc_name = opt->intern.enc.name;
ruby_features_t feat = opt->features;
ruby_features_t warn = opt->warn;
opt->src.enc.name = opt->ext.enc.name = opt->intern.enc.name = 0;
moreswitches(s, opt, 1);
if (src_enc_name)
opt->src.enc.name = src_enc_name;
if (ext_enc_name)
opt->ext.enc.name = ext_enc_name;
if (int_enc_name)
opt->intern.enc.name = int_enc_name;
2019-12-20 16:50:46 +03:00
FEATURE_SET_RESTORE(opt->features, feat);
FEATURE_SET_RESTORE(opt->warn, warn);
}
if (opt->src.enc.name)
rb_warning("-K is specified; it is for 1.8 compatibility and may cause odd behavior");
#if USE_MJIT
if (opt->features.set & FEATURE_BIT(jit)) {
opt->mjit.on = TRUE; /* set mjit.on for ruby_show_version() API and check to call mjit_init() */
}
#endif
if (opt->dump & (DUMP_BIT(version) | DUMP_BIT(version_v))) {
#if USE_MJIT
mjit_opts.on = opt->mjit.on; /* used by ruby_show_version(). mjit_init() still can't be called here. */
#endif
ruby_show_version();
if (opt->dump & DUMP_BIT(version)) return Qtrue;
}
if (opt->dump & DUMP_BIT(copyright)) {
ruby_show_copyright();
return Qtrue;
}
if (!opt->e_script) {
if (argc <= 0) { /* no more args */
if (opt->verbose)
return Qtrue;
opt->script = "-";
}
else {
opt->script = argv[0];
if (!opt->script || opt->script[0] == '\0') {
opt->script = "-";
}
else if (opt->do_search) {
char *path = getenv("RUBYPATH");
opt->script = 0;
if (path) {
opt->script = dln_find_file_r(argv[0], path, fbuf, sizeof(fbuf));
}
if (!opt->script) {
opt->script = dln_find_file_r(argv[0], getenv(PATH_ENV), fbuf, sizeof(fbuf));
}
if (!opt->script)
opt->script = argv[0];
}
argc--;
argv++;
}
if (opt->script[0] == '-' && !opt->script[1]) {
forbid_setid("program input from stdin");
}
}
opt->script_name = rb_str_new_cstr(opt->script);
opt->script = RSTRING_PTR(opt->script_name);
#if _WIN32
translit_char_bin(RSTRING_PTR(opt->script_name), '\\', '/');
#elif defined DOSISH
translit_char(RSTRING_PTR(opt->script_name), '\\', '/');
#endif
ruby_gc_set_params();
ruby_init_loadpath();
#if USE_MJIT
if (opt->mjit.on)
/* Using TMP_RUBY_PREFIX created by ruby_init_loadpath(). */
mjit_init(&opt->mjit);
#endif
Init_ruby_description();
Init_enc();
lenc = rb_locale_encoding();
rb_enc_associate(rb_progname, lenc);
rb_obj_freeze(rb_progname);
parser = rb_parser_new();
if (opt->dump & DUMP_BIT(yydebug)) {
rb_parser_set_yydebug(parser, Qtrue);
}
if (opt->ext.enc.name != 0) {
opt->ext.enc.index = opt_enc_index(opt->ext.enc.name);
}
if (opt->intern.enc.name != 0) {
opt->intern.enc.index = opt_enc_index(opt->intern.enc.name);
}
if (opt->src.enc.name != 0) {
opt->src.enc.index = opt_enc_index(opt->src.enc.name);
src_encoding_index = opt->src.enc.index;
}
if (opt->ext.enc.index >= 0) {
enc = rb_enc_from_index(opt->ext.enc.index);
}
else {
enc = lenc;
}
rb_enc_set_default_external(rb_enc_from_encoding(enc));
if (opt->intern.enc.index >= 0) {
enc = rb_enc_from_index(opt->intern.enc.index);
rb_enc_set_default_internal(rb_enc_from_encoding(enc));
opt->intern.enc.index = -1;
#if UTF8_PATH
ienc = enc;
#endif
}
script_name = opt->script_name;
rb_enc_associate(opt->script_name,
IF_UTF8_PATH(uenc = rb_utf8_encoding(), lenc));
#if UTF8_PATH
if (uenc != lenc) {
opt->script_name = str_conv_enc(opt->script_name, uenc, lenc);
opt->script = RSTRING_PTR(opt->script_name);
}
#endif
rb_obj_freeze(opt->script_name);
if (IF_UTF8_PATH(uenc != lenc, 1)) {
long i;
rb_vm_t *vm = GET_VM();
VALUE load_path = vm->load_path;
const ID id_initial_load_path_mark = INITIAL_LOAD_PATH_MARK;
int modifiable = FALSE;
rb_get_expanded_load_path();
for (i = 0; i < RARRAY_LEN(load_path); ++i) {
VALUE path = RARRAY_AREF(load_path, i);
int mark = rb_attr_get(path, id_initial_load_path_mark) == path;
#if UTF8_PATH
VALUE newpath = rb_str_conv_enc(path, uenc, lenc);
if (newpath == path) continue;
path = newpath;
#else
path = rb_enc_associate(rb_str_dup(path), lenc);
#endif
if (mark) rb_ivar_set(path, id_initial_load_path_mark, path);
if (!modifiable) {
rb_ary_modify(load_path);
modifiable = TRUE;
}
RARRAY_ASET(load_path, i, path);
}
if (modifiable) {
rb_ary_replace(vm->load_path_snapshot, load_path);
}
}
if (opt->features.mask & COMPILATION_FEATURES) {
VALUE option = rb_hash_new();
#define SET_COMPILE_OPTION(h, o, name) \
rb_hash_aset((h), ID2SYM(rb_intern_const(#name)), \
(FEATURE_SET_P(o->features, FEATURE_BIT(name)) ? Qtrue : Qfalse));
SET_COMPILE_OPTION(option, opt, frozen_string_literal);
SET_COMPILE_OPTION(option, opt, debug_frozen_string_literal);
rb_funcallv(rb_cISeq, rb_intern_const("compile_option="), 1, &option);
#undef SET_COMPILE_OPTION
}
ruby_set_argv(argc, argv);
process_sflag(&opt->sflag);
rb_parser_set_context(parser, 0, TRUE);
if (opt->e_script) {
VALUE progname = rb_progname;
rb_encoding *eenc;
if (opt->src.enc.index >= 0) {
eenc = rb_enc_from_index(opt->src.enc.index);
}
else {
eenc = lenc;
#if UTF8_PATH
if (ienc) eenc = ienc;
#endif
}
#if UTF8_PATH
if (eenc != uenc) {
opt->e_script = str_conv_enc(opt->e_script, uenc, eenc);
}
#endif
rb_enc_associate(opt->e_script, eenc);
ruby_opt_init(opt);
ruby_set_script_name(progname);
rb_parser_set_options(parser, opt->do_print, opt->do_loop,
opt->do_line, opt->do_split);
rb_warning_category_update(opt->warn.mask, opt->warn.set);
ast = rb_parser_compile_string(parser, opt->script, opt->e_script, 1);
}
else {
VALUE f;
f = open_load_file(script_name, &opt->xflag);
ast = load_file(parser, opt->script_name, f, 1, opt);
}
ruby_set_script_name(opt->script_name);
if (dump & DUMP_BIT(yydebug)) {
dump &= ~DUMP_BIT(yydebug);
if (!dump) return Qtrue;
}
if (opt->ext.enc.index >= 0) {
enc = rb_enc_from_index(opt->ext.enc.index);
}
else {
enc = lenc;
}
rb_enc_set_default_external(rb_enc_from_encoding(enc));
if (opt->intern.enc.index >= 0) {
/* Set in the shebang line */
enc = rb_enc_from_index(opt->intern.enc.index);
rb_enc_set_default_internal(rb_enc_from_encoding(enc));
}
else if (!rb_default_internal_encoding())
/* Freeze default_internal */
rb_enc_set_default_internal(Qnil);
rb_stdio_set_default_encoding();
if (!ast->body.root) {
rb_ast_dispose(ast);
return Qfalse;
}
process_sflag(&opt->sflag);
opt->xflag = 0;
if (dump & DUMP_BIT(syntax)) {
printf("Syntax OK\n");
dump &= ~DUMP_BIT(syntax);
if (!dump) return Qtrue;
}
if (opt->do_loop) {
rb_define_global_function("sub", rb_f_sub, -1);
rb_define_global_function("gsub", rb_f_gsub, -1);
rb_define_global_function("chop", rb_f_chop, 0);
rb_define_global_function("chomp", rb_f_chomp, -1);
}
if (dump & (DUMP_BIT(parsetree)|DUMP_BIT(parsetree_with_comment))) {
rb_io_write(rb_stdout, rb_parser_dump_tree(ast->body.root, dump & DUMP_BIT(parsetree_with_comment)));
rb_io_flush(rb_stdout);
dump &= ~DUMP_BIT(parsetree)&~DUMP_BIT(parsetree_with_comment);
if (!dump) {
rb_ast_dispose(ast);
return Qtrue;
}
}
{
VALUE path = Qnil;
if (!opt->e_script && strcmp(opt->script, "-")) {
path = rb_realpath_internal(Qnil, script_name, 1);
#if UTF8_PATH
if (uenc != lenc) {
path = str_conv_enc(path, uenc, lenc);
}
#endif
if (!ENCODING_GET(path)) { /* ASCII-8BIT */
rb_enc_copy(path, opt->script_name);
}
}
rb_binding_t *toplevel_binding;
GetBindingPtr(rb_const_get(rb_cObject, rb_intern("TOPLEVEL_BINDING")),
toplevel_binding);
const struct rb_block *base_block = toplevel_context(toplevel_binding);
iseq = rb_iseq_new_main(&ast->body, opt->script_name, path, vm_block_iseq(base_block));
rb_ast_dispose(ast);
}
if (dump & DUMP_BIT(insns)) {
2015-07-22 01:52:59 +03:00
rb_io_write(rb_stdout, rb_iseq_disasm((const rb_iseq_t *)iseq));
rb_io_flush(rb_stdout);
dump &= ~DUMP_BIT(insns);
if (!dump) return Qtrue;
}
if (opt->dump & dump_exit_bits) return Qtrue;
rb_define_readonly_boolean("$-p", opt->do_print);
rb_define_readonly_boolean("$-l", opt->do_line);
rb_define_readonly_boolean("$-a", opt->do_split);
rb_gvar_ractor_local("$-p");
rb_gvar_ractor_local("$-l");
rb_gvar_ractor_local("$-a");
if ((rb_e_script = opt->e_script) != 0) {
rb_gc_register_mark_object(opt->e_script);
}
{
rb_execution_context_t *ec = GET_EC();
if (opt->e_script) {
/* -e */
rb_exec_event_hook_script_compiled(ec, iseq, opt->e_script);
}
else {
/* file */
rb_exec_event_hook_script_compiled(ec, iseq, Qnil);
}
}
2015-07-22 01:52:59 +03:00
return (VALUE)iseq;
}
#ifndef DOSISH
static void
warn_cr_in_shebang(const char *str, long len)
{
if (str[len-1] == '\n' && str[len-2] == '\r') {
rb_warn("shebang line ending with \\r may cause problems");
}
}
#else
#define warn_cr_in_shebang(str, len) (void)0
#endif
struct load_file_arg {
VALUE parser;
VALUE fname;
int script;
ruby_cmdline_options_t *opt;
VALUE f;
};
static VALUE
load_file_internal(VALUE argp_v)
{
struct load_file_arg *argp = (struct load_file_arg *)argp_v;
VALUE parser = argp->parser;
VALUE orig_fname = argp->fname;
int script = argp->script;
ruby_cmdline_options_t *opt = argp->opt;
VALUE f = argp->f;
int line_start = 1;
rb_ast_t *ast = 0;
rb_encoding *enc;
ID set_encoding;
CONST_ID(set_encoding, "set_encoding");
if (script) {
VALUE c = 1; /* something not nil */
VALUE line;
char *p, *str;
long len;
int no_src_enc = !opt->src.enc.name;
int no_ext_enc = !opt->ext.enc.name;
int no_int_enc = !opt->intern.enc.name;
enc = rb_ascii8bit_encoding();
rb_funcall(f, set_encoding, 1, rb_enc_from_encoding(enc));
if (opt->xflag) {
line_start--;
search_shebang:
while (!NIL_P(line = rb_io_gets(f))) {
line_start++;
RSTRING_GETMEM(line, str, len);
if (len > 2 && str[0] == '#' && str[1] == '!') {
if (line_start == 1) warn_cr_in_shebang(str, len);
if ((p = strstr(str+2, ruby_engine)) != 0) {
goto start_read;
}
}
}
rb_loaderror("no Ruby script found in input");
}
* encoding.c: provide basic features for M17N. * parse.y: encoding aware parsing. * parse.y (pragma_encoding): encoding specification pragma. * parse.y (rb_intern3): encoding specified symbols. * string.c (rb_str_length): length based on characters. for older behavior, bytesize method added. * string.c (rb_str_index_m): index based on characters. rindex as well. * string.c (succ_char): encoding aware succeeding string. * string.c (rb_str_reverse): reverse based on characters. * string.c (rb_str_inspect): encoding aware string description. * string.c (rb_str_upcase_bang): encoding aware case conversion. downcase, capitalize, swapcase as well. * string.c (rb_str_tr_bang): tr based on characters. delete, squeeze, tr_s, count as well. * string.c (rb_str_split_m): split based on characters. * string.c (rb_str_each_line): encoding aware each_line. * string.c (rb_str_each_char): added. iteration based on characters. * string.c (rb_str_strip_bang): encoding aware whitespace stripping. lstrip, rstrip as well. * string.c (rb_str_justify): encoding aware justifying (ljust, rjust, center). * string.c (str_encoding): get encoding attribute from a string. * re.c (rb_reg_initialize): encoding aware regular expression * sprintf.c (rb_str_format): formatting (i.e. length count) based on characters. * io.c (rb_io_getc): getc to return one-character string. for older behavior, getbyte method added. * ext/stringio/stringio.c (strio_getc): ditto. * io.c (rb_io_ungetc): allow pushing arbitrary string at the current reading point. * ext/stringio/stringio.c (strio_ungetc): ditto. * ext/strscan/strscan.c: encoding support. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@13261 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2007-08-25 07:29:39 +04:00
c = rb_io_getbyte(f);
if (c == INT2FIX('#')) {
* encoding.c: provide basic features for M17N. * parse.y: encoding aware parsing. * parse.y (pragma_encoding): encoding specification pragma. * parse.y (rb_intern3): encoding specified symbols. * string.c (rb_str_length): length based on characters. for older behavior, bytesize method added. * string.c (rb_str_index_m): index based on characters. rindex as well. * string.c (succ_char): encoding aware succeeding string. * string.c (rb_str_reverse): reverse based on characters. * string.c (rb_str_inspect): encoding aware string description. * string.c (rb_str_upcase_bang): encoding aware case conversion. downcase, capitalize, swapcase as well. * string.c (rb_str_tr_bang): tr based on characters. delete, squeeze, tr_s, count as well. * string.c (rb_str_split_m): split based on characters. * string.c (rb_str_each_line): encoding aware each_line. * string.c (rb_str_each_char): added. iteration based on characters. * string.c (rb_str_strip_bang): encoding aware whitespace stripping. lstrip, rstrip as well. * string.c (rb_str_justify): encoding aware justifying (ljust, rjust, center). * string.c (str_encoding): get encoding attribute from a string. * re.c (rb_reg_initialize): encoding aware regular expression * sprintf.c (rb_str_format): formatting (i.e. length count) based on characters. * io.c (rb_io_getc): getc to return one-character string. for older behavior, getbyte method added. * ext/stringio/stringio.c (strio_getc): ditto. * io.c (rb_io_ungetc): allow pushing arbitrary string at the current reading point. * ext/stringio/stringio.c (strio_ungetc): ditto. * ext/strscan/strscan.c: encoding support. git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@13261 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2007-08-25 07:29:39 +04:00
c = rb_io_getbyte(f);
if (c == INT2FIX('!') && !NIL_P(line = rb_io_gets(f))) {
RSTRING_GETMEM(line, str, len);
warn_cr_in_shebang(str, len);
if ((p = strstr(str, ruby_engine)) == 0) {
/* not ruby script, assume -x flag */
goto search_shebang;
}
start_read:
str += len - 1;
if (*str == '\n') *str-- = '\0';
if (*str == '\r') *str-- = '\0';
/* ruby_engine should not contain a space */
if ((p = strstr(p, " -")) != 0) {
opt->warning = 0;
moreswitches(p + 1, opt, 0);
}
/* push back shebang for pragma may exist in next line */
rb_io_ungetbyte(f, rb_str_new2("!\n"));
}
else if (!NIL_P(c)) {
rb_io_ungetbyte(f, c);
}
rb_io_ungetbyte(f, INT2FIX('#'));
if (no_src_enc && opt->src.enc.name) {
opt->src.enc.index = opt_enc_index(opt->src.enc.name);
src_encoding_index = opt->src.enc.index;
}
if (no_ext_enc && opt->ext.enc.name) {
opt->ext.enc.index = opt_enc_index(opt->ext.enc.name);
}
if (no_int_enc && opt->intern.enc.name) {
opt->intern.enc.index = opt_enc_index(opt->intern.enc.name);
}
}
else if (!NIL_P(c)) {
rb_io_ungetbyte(f, c);
}
2019-06-04 03:36:32 +03:00
if (NIL_P(c)) {
argp->f = f = Qnil;
}
ruby_opt_init(opt);
}
if (opt->src.enc.index >= 0) {
enc = rb_enc_from_index(opt->src.enc.index);
}
else if (f == rb_stdin) {
enc = rb_locale_encoding();
}
else {
enc = rb_utf8_encoding();
}
rb_parser_set_options(parser, opt->do_print, opt->do_loop,
opt->do_line, opt->do_split);
rb_warning_category_update(opt->warn.mask, opt->warn.set);
if (NIL_P(f)) {
f = rb_str_new(0, 0);
rb_enc_associate(f, enc);
return (VALUE)rb_parser_compile_string_path(parser, orig_fname, f, line_start);
}
rb_funcall(f, set_encoding, 2, rb_enc_from_encoding(enc), rb_str_new_cstr("-"));
ast = rb_parser_compile_file_path(parser, orig_fname, f, line_start);
rb_funcall(f, set_encoding, 1, rb_parser_encoding(parser));
if (script && rb_parser_end_seen_p(parser)) {
/*
* DATA is a File that contains the data section of the executed file.
* To create a data section use <tt>__END__</tt>:
*
* $ cat t.rb
* puts DATA.gets
* __END__
* hello world!
*
* $ ruby t.rb
* hello world!
*/
rb_define_global_const("DATA", f);
argp->f = Qnil;
}
return (VALUE)ast;
}
static VALUE
open_load_file(VALUE fname_v, int *xflag)
{
const char *fname = (fname_v = rb_str_encode_ospath(fname_v),
StringValueCStr(fname_v));
long flen = RSTRING_LEN(fname_v);
VALUE f;
int e;
if (flen == 1 && fname[0] == '-') {
f = rb_stdin;
}
else {
int fd;
/* open(2) may block if fname is point to FIFO and it's empty. Let's
use O_NONBLOCK. */
#if defined O_NONBLOCK && HAVE_FCNTL && !(O_NONBLOCK & O_ACCMODE)
/* TODO: fix conflicting O_NONBLOCK in ruby/win32.h */
# define MODE_TO_LOAD (O_RDONLY | O_NONBLOCK)
#elif defined O_NDELAY && HAVE_FCNTL && !(O_NDELAY & O_ACCMODE)
# define MODE_TO_LOAD (O_RDONLY | O_NDELAY)
#else
# define MODE_TO_LOAD (O_RDONLY)
#endif
int mode = MODE_TO_LOAD;
#if defined DOSISH || defined __CYGWIN__
# define isdirsep(x) ((x) == '/' || (x) == '\\')
{
static const char exeext[] = ".exe";
enum {extlen = sizeof(exeext)-1};
if (flen > extlen && !isdirsep(fname[flen-extlen-1]) &&
STRNCASECMP(fname+flen-extlen, exeext, extlen) == 0) {
mode |= O_BINARY;
*xflag = 1;
}
}
#endif
if ((fd = rb_cloexec_open(fname, mode, 0)) < 0) {
e = errno;
if (!rb_gc_for_fd(e)) {
rb_load_fail(fname_v, strerror(e));
}
if ((fd = rb_cloexec_open(fname, mode, 0)) < 0) {
rb_load_fail(fname_v, strerror(errno));
}
}
rb_update_max_fd(fd);
#if defined HAVE_FCNTL && MODE_TO_LOAD != O_RDONLY
/* disabling O_NONBLOCK */
if (fcntl(fd, F_SETFL, 0) < 0) {
e = errno;
(void)close(fd);
rb_load_fail(fname_v, strerror(e));
}
#endif
e = ruby_is_fd_loadable(fd);
if (!e) {
e = errno;
(void)close(fd);
rb_load_fail(fname_v, strerror(e));
}
f = rb_io_fdopen(fd, mode, fname);
if (e < 0) {
/*
We need to wait if FIFO is empty. It's FIFO's semantics.
rb_thread_wait_fd() release GVL. So, it's safe.
*/
rb_thread_wait_fd(fd);
}
}
return f;
}
static VALUE
restore_load_file(VALUE arg)
{
struct load_file_arg *argp = (struct load_file_arg *)arg;
VALUE f = argp->f;
if (!NIL_P(f) && f != rb_stdin) {
rb_io_close(f);
}
return Qnil;
}
static rb_ast_t *
load_file(VALUE parser, VALUE fname, VALUE f, int script, ruby_cmdline_options_t *opt)
{
struct load_file_arg arg;
arg.parser = parser;
arg.fname = fname;
arg.script = script;
arg.opt = opt;
arg.f = f;
return (rb_ast_t *)rb_ensure(load_file_internal, (VALUE)&arg,
restore_load_file, (VALUE)&arg);
}
void *
rb_load_file(const char *fname)
{
VALUE fname_v = rb_str_new_cstr(fname);
return rb_load_file_str(fname_v);
}
void *
rb_load_file_str(VALUE fname_v)
{
return rb_parser_load_file(rb_parser_new(), fname_v);
}
void *
rb_parser_load_file(VALUE parser, VALUE fname_v)
{
ruby_cmdline_options_t opt;
VALUE f = open_load_file(fname_v, &cmdline_options_init(&opt)->xflag);
return load_file(parser, fname_v, f, 0, &opt);
}
/*
* call-seq:
* Process.argv0 -> frozen_string
*
* Returns the name of the script being executed. The value is not
* affected by assigning a new value to $0.
*
* This method first appeared in Ruby 2.1 to serve as a global
* variable free means to get the script name.
*/
static VALUE
proc_argv0(VALUE process)
{
return rb_orig_progname;
}
static VALUE ruby_setproctitle(VALUE title);
/*
* call-seq:
* Process.setproctitle(string) -> string
*
* Sets the process title that appears on the ps(1) command. Not
* necessarily effective on all platforms. No exception will be
* raised regardless of the result, nor will NotImplementedError be
* raised even if the platform does not support the feature.
*
* Calling this method does not affect the value of $0.
*
* Process.setproctitle('myapp: worker #%d' % worker_id)
*
* This method first appeared in Ruby 2.1 to serve as a global
* variable free means to change the process title.
*/
static VALUE
proc_setproctitle(VALUE process, VALUE title)
{
return ruby_setproctitle(title);
}
static VALUE
ruby_setproctitle(VALUE title)
{
const char *ptr = StringValueCStr(title);
setproctitle("%.*s", RSTRING_LENINT(title), ptr);
return title;
}
static void
set_arg0(VALUE val, ID id, VALUE *_)
{
if (origarg.argv == 0)
rb_raise(rb_eRuntimeError, "$0 not initialized");
rb_progname = rb_str_new_frozen(ruby_setproctitle(val));
}
static inline VALUE
external_str_new_cstr(const char *p)
{
#if UTF8_PATH
VALUE str = rb_utf8_str_new_cstr(p);
str = str_conv_enc(str, NULL, rb_default_external_encoding());
return str;
#else
return rb_external_str_new_cstr(p);
#endif
}
/*! Sets the current script name to this value.
*
* This is similar to <code>$0 = name</code> in Ruby level but also affects
* <code>Method#location</code> and others.
*/
void
ruby_script(const char *name)
{
if (name) {
rb_orig_progname = rb_progname = external_str_new_cstr(name);
rb_vm_set_progname(rb_progname);
}
}
/*! Sets the current script name to this value.
*
* Same as ruby_script() but accepts a VALUE.
*/
void
ruby_set_script_name(VALUE name)
{
rb_orig_progname = rb_progname = rb_str_dup(name);
rb_vm_set_progname(rb_progname);
}
static void
init_ids(ruby_cmdline_options_t *opt)
{
rb_uid_t uid = getuid();
rb_uid_t euid = geteuid();
rb_gid_t gid = getgid();
rb_gid_t egid = getegid();
if (uid != euid) opt->setids |= 1;
if (egid != gid) opt->setids |= 2;
}
#undef forbid_setid
static void
forbid_setid(const char *s, const ruby_cmdline_options_t *opt)
{
if (opt->setids & 1)
* array.c: replace rb_protect_inspect() and rb_inspecting_p() by rb_exec_recursive() in eval.c. * eval.c (rb_exec_recursive): new function. * array.c (rb_ary_join): use rb_exec_recursive(). * array.c (rb_ary_inspect, rb_ary_hash): ditto. * file.c (rb_file_join): ditto. * hash.c (rb_hash_inspect, rb_hash_to_s, rb_hash_hash): ditto. * io.c (rb_io_puts): ditto. * object.c (rb_obj_inspect): ditto * struct.c (rb_struct_inspect): ditto. * lib/set.rb (SortedSet::setup): a hack to shut up warning. [ruby-talk:132866] * lib/time.rb (Time::strptime): add new function. inspired by [ruby-talk:132815]. * lib/parsedate.rb (ParseDate::strptime): ditto. * regparse.c: move st_*_strend() functions from st.c. fixed some potential memory leaks. * exception error messages updated. [ruby-core:04497] * ext/socket/socket.c (Init_socket): add bunch of Socket constants. Patch from Sam Roberts <sroberts@uniserve.com>. [ruby-core:04409] * array.c (rb_ary_s_create): no need for negative argc check. [ruby-core:04463] * array.c (rb_ary_unshift_m): ditto. * lib/xmlrpc/parser.rb (XMLRPC::FaultException): make it subclass of StandardError class, not Exception class. [ruby-core:04429] * parse.y (fcall_gen): lvar(arg) will be evaluated as lvar.call(arg) when lvar is a defined local variable. [new] * object.c (rb_class_initialize): call inherited method before calling initializing block. * eval.c (rb_thread_start_1): initialize newly pushed frame. * lib/open3.rb (Open3::popen3): $? should not be EXIT_FAILURE. fixed: [ruby-core:04444] * eval.c (is_defined): NODE_IASGN is an assignment. * ext/readline/readline.c (Readline.readline): use rl_outstream and rl_instream. [ruby-dev:25699] * ext/etc/etc.c (Init_etc): sGroup needs HAVE_ST_GR_PASSWD check [ruby-dev:25675] * misc/ruby-mode.el: [ruby-core:04415] * lib/rdoc/generators/html_generator.rb: [ruby-core:04412] * lib/rdoc/generators/ri_generator.rb: ditto. * struct.c (make_struct): fixed: [ruby-core:04402] * ext/curses/curses.c (window_color_set): [ruby-core:04393] * ext/socket/socket.c (Init_socket): SO_REUSEPORT added. [ruby-talk:130092] * object.c: [ruby-doc:818] * parse.y (open_args): fix too verbose warnings for the space before argument parentheses. [ruby-dev:25492] * parse.y (parser_yylex): ditto. * parse.y (parser_yylex): the first expression in the parentheses should not be a command. [ruby-dev:25492] * lib/irb/context.rb (IRB::Context::initialize): [ruby-core:04330] * object.c (Init_Object): remove Object#type. [ruby-core:04335] * st.c (st_foreach): report success/failure by return value. [ruby-Bugs-1396] * parse.y: forgot to initialize parser struct. [ruby-dev:25492] * parse.y (parser_yylex): no tLABEL on EXPR_BEG. [ruby-talk:127711] * document updates - [ruby-core:04296], [ruby-core:04301], [ruby-core:04302], [ruby-core:04307] * dir.c (rb_push_glob): should work for NUL delimited patterns. * dir.c (rb_glob2): should aware of offset in the pattern. * string.c (rb_str_new4): should propagate taintedness. * env.h: rename member names in struct FRAME; last_func -> callee, orig_func -> this_func, last_class -> this_class. * struct.c (rb_struct_set): use original method name, not callee name, to retrieve member slot. [ruby-core:04268] * time.c (time_strftime): protect from format modification from GC finalizers. * object.c (Init_Object): remove rb_obj_id_obsolete() * eval.c (rb_mod_define_method): incomplete subclass check. [ruby-dev:25464] * gc.c (rb_data_object_alloc): klass may be NULL. [ruby-list:40498] * bignum.c (rb_big_rand): should return positive random number. [ruby-dev:25401] * bignum.c (rb_big_rand): do not use rb_big_modulo to generate random bignums. [ruby-dev:25396] * variable.c (rb_autoload): [ruby-dev:25373] * eval.c (svalue_to_avalue): [ruby-dev:25366] * string.c (rb_str_justify): [ruby-dev:25367] * io.c (rb_f_select): [ruby-dev:25312] * ext/socket/socket.c (sock_s_getservbyport): [ruby-talk:124072] * struct.c (make_struct): [ruby-dev:25249] * dir.c (dir_open_dir): new function. [ruby-dev:25242] * io.c (rb_f_open): add type check for return value from to_open. * lib/pstore.rb (PStore#transaction): Use the empty content when a file is not found. [ruby-dev:24561] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@8068 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2005-03-04 09:47:45 +03:00
rb_raise(rb_eSecurityError, "no %s allowed while running setuid", s);
if (opt->setids & 2)
* array.c: replace rb_protect_inspect() and rb_inspecting_p() by rb_exec_recursive() in eval.c. * eval.c (rb_exec_recursive): new function. * array.c (rb_ary_join): use rb_exec_recursive(). * array.c (rb_ary_inspect, rb_ary_hash): ditto. * file.c (rb_file_join): ditto. * hash.c (rb_hash_inspect, rb_hash_to_s, rb_hash_hash): ditto. * io.c (rb_io_puts): ditto. * object.c (rb_obj_inspect): ditto * struct.c (rb_struct_inspect): ditto. * lib/set.rb (SortedSet::setup): a hack to shut up warning. [ruby-talk:132866] * lib/time.rb (Time::strptime): add new function. inspired by [ruby-talk:132815]. * lib/parsedate.rb (ParseDate::strptime): ditto. * regparse.c: move st_*_strend() functions from st.c. fixed some potential memory leaks. * exception error messages updated. [ruby-core:04497] * ext/socket/socket.c (Init_socket): add bunch of Socket constants. Patch from Sam Roberts <sroberts@uniserve.com>. [ruby-core:04409] * array.c (rb_ary_s_create): no need for negative argc check. [ruby-core:04463] * array.c (rb_ary_unshift_m): ditto. * lib/xmlrpc/parser.rb (XMLRPC::FaultException): make it subclass of StandardError class, not Exception class. [ruby-core:04429] * parse.y (fcall_gen): lvar(arg) will be evaluated as lvar.call(arg) when lvar is a defined local variable. [new] * object.c (rb_class_initialize): call inherited method before calling initializing block. * eval.c (rb_thread_start_1): initialize newly pushed frame. * lib/open3.rb (Open3::popen3): $? should not be EXIT_FAILURE. fixed: [ruby-core:04444] * eval.c (is_defined): NODE_IASGN is an assignment. * ext/readline/readline.c (Readline.readline): use rl_outstream and rl_instream. [ruby-dev:25699] * ext/etc/etc.c (Init_etc): sGroup needs HAVE_ST_GR_PASSWD check [ruby-dev:25675] * misc/ruby-mode.el: [ruby-core:04415] * lib/rdoc/generators/html_generator.rb: [ruby-core:04412] * lib/rdoc/generators/ri_generator.rb: ditto. * struct.c (make_struct): fixed: [ruby-core:04402] * ext/curses/curses.c (window_color_set): [ruby-core:04393] * ext/socket/socket.c (Init_socket): SO_REUSEPORT added. [ruby-talk:130092] * object.c: [ruby-doc:818] * parse.y (open_args): fix too verbose warnings for the space before argument parentheses. [ruby-dev:25492] * parse.y (parser_yylex): ditto. * parse.y (parser_yylex): the first expression in the parentheses should not be a command. [ruby-dev:25492] * lib/irb/context.rb (IRB::Context::initialize): [ruby-core:04330] * object.c (Init_Object): remove Object#type. [ruby-core:04335] * st.c (st_foreach): report success/failure by return value. [ruby-Bugs-1396] * parse.y: forgot to initialize parser struct. [ruby-dev:25492] * parse.y (parser_yylex): no tLABEL on EXPR_BEG. [ruby-talk:127711] * document updates - [ruby-core:04296], [ruby-core:04301], [ruby-core:04302], [ruby-core:04307] * dir.c (rb_push_glob): should work for NUL delimited patterns. * dir.c (rb_glob2): should aware of offset in the pattern. * string.c (rb_str_new4): should propagate taintedness. * env.h: rename member names in struct FRAME; last_func -> callee, orig_func -> this_func, last_class -> this_class. * struct.c (rb_struct_set): use original method name, not callee name, to retrieve member slot. [ruby-core:04268] * time.c (time_strftime): protect from format modification from GC finalizers. * object.c (Init_Object): remove rb_obj_id_obsolete() * eval.c (rb_mod_define_method): incomplete subclass check. [ruby-dev:25464] * gc.c (rb_data_object_alloc): klass may be NULL. [ruby-list:40498] * bignum.c (rb_big_rand): should return positive random number. [ruby-dev:25401] * bignum.c (rb_big_rand): do not use rb_big_modulo to generate random bignums. [ruby-dev:25396] * variable.c (rb_autoload): [ruby-dev:25373] * eval.c (svalue_to_avalue): [ruby-dev:25366] * string.c (rb_str_justify): [ruby-dev:25367] * io.c (rb_f_select): [ruby-dev:25312] * ext/socket/socket.c (sock_s_getservbyport): [ruby-talk:124072] * struct.c (make_struct): [ruby-dev:25249] * dir.c (dir_open_dir): new function. [ruby-dev:25242] * io.c (rb_f_open): add type check for return value from to_open. * lib/pstore.rb (PStore#transaction): Use the empty content when a file is not found. [ruby-dev:24561] git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@8068 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2005-03-04 09:47:45 +03:00
rb_raise(rb_eSecurityError, "no %s allowed while running setgid", s);
}
static VALUE
verbose_getter(ID id, VALUE *ptr)
{
return *rb_ruby_verbose_ptr();
}
static void
verbose_setter(VALUE val, ID id, VALUE *variable)
{
*rb_ruby_verbose_ptr() = RTEST(val) ? Qtrue : val;
}
static VALUE
opt_W_getter(ID id, VALUE *dmy)
{
VALUE v = *rb_ruby_verbose_ptr();
switch (v) {
case Qnil:
return INT2FIX(0);
case Qfalse:
return INT2FIX(1);
case Qtrue:
return INT2FIX(2);
default:
return Qnil;
}
}
static VALUE
debug_getter(ID id, VALUE *dmy)
{
return *rb_ruby_debug_ptr();
}
static void
debug_setter(VALUE val, ID id, VALUE *dmy)
{
*rb_ruby_debug_ptr() = val;
}
/*! Defines built-in variables */
void
ruby_prog_init(void)
{
rb_define_virtual_variable("$VERBOSE", verbose_getter, verbose_setter);
rb_define_virtual_variable("$-v", verbose_getter, verbose_setter);
rb_define_virtual_variable("$-w", verbose_getter, verbose_setter);
rb_define_virtual_variable("$-W", opt_W_getter, rb_gvar_readonly_setter);
rb_define_virtual_variable("$DEBUG", debug_getter, debug_setter);
rb_define_virtual_variable("$-d", debug_getter, debug_setter);
rb_gvar_ractor_local("$VERBOSE");
rb_gvar_ractor_local("$-v");
rb_gvar_ractor_local("$-w");
rb_gvar_ractor_local("$-W");
rb_gvar_ractor_local("$DEBUG");
rb_gvar_ractor_local("$-d");
rb_define_hooked_variable("$0", &rb_progname, 0, set_arg0);
rb_define_hooked_variable("$PROGRAM_NAME", &rb_progname, 0, set_arg0);
rb_define_module_function(rb_mProcess, "argv0", proc_argv0, 0);
rb_define_module_function(rb_mProcess, "setproctitle", proc_setproctitle, 1);
/*
* ARGV contains the command line arguments used to run ruby.
*
* A library like OptionParser can be used to process command-line
* arguments.
*/
rb_define_global_const("ARGV", rb_argv);
}
void
ruby_set_argv(int argc, char **argv)
{
int i;
VALUE av = rb_argv;
#if defined(USE_DLN_A_OUT)
if (origarg.argc > 0 && origarg.argv)
dln_argv0 = origarg.argv[0];
else if (argc > 0 && argv)
dln_argv0 = argv[0];
#endif
rb_ary_clear(av);
for (i = 0; i < argc; i++) {
VALUE arg = external_str_new_cstr(argv[i]);
OBJ_FREEZE(arg);
rb_ary_push(av, arg);
}
}
void *
ruby_process_options(int argc, char **argv)
{
ruby_cmdline_options_t opt;
VALUE iseq;
const char *script_name = (argc > 0 && argv[0]) ? argv[0] : ruby_engine;
if (!origarg.argv || origarg.argc <= 0) {
origarg.argc = argc;
origarg.argv = argv;
}
ruby_script(script_name); /* for the time being */
rb_argv0 = rb_str_new4(rb_progname);
rb_gc_register_mark_object(rb_argv0);
iseq = process_options(argc, argv, cmdline_options_init(&opt));
#ifndef HAVE_SETPROCTITLE
ruby_init_setproctitle(argc, argv);
#endif
return (void*)(struct RData*)iseq;
}
static void
fill_standard_fds(void)
{
int f0, f1, f2, fds[2];
struct stat buf;
f0 = fstat(0, &buf) == -1 && errno == EBADF;
f1 = fstat(1, &buf) == -1 && errno == EBADF;
f2 = fstat(2, &buf) == -1 && errno == EBADF;
if (f0) {
if (pipe(fds) == 0) {
close(fds[1]);
if (fds[0] != 0) {
dup2(fds[0], 0);
close(fds[0]);
}
}
}
if (f1 || f2) {
if (pipe(fds) == 0) {
close(fds[0]);
if (f1 && fds[1] != 1)
dup2(fds[1], 1);
if (f2 && fds[1] != 2)
dup2(fds[1], 2);
if (fds[1] != 1 && fds[1] != 2)
close(fds[1]);
}
}
}
/*! Initializes the process for libruby.
*
* This function assumes this process is ruby(1) and it has just started.
* Usually programs that embed CRuby interpreter may not call this function,
* and may do their own initialization.
* argc and argv cannot be NULL.
*/
void
ruby_sysinit(int *argc, char ***argv)
{
#if defined(_WIN32)
rb_w32_sysinit(argc, argv);
#endif
if (*argc >= 0 && *argv) {
origarg.argc = *argc;
origarg.argv = *argv;
#if defined(USE_DLN_A_OUT)
dln_argv0 = origarg.argv[0];
#endif
}
fill_standard_fds();
}