because the name "MJIT" is an internal code name, it's inconsistent with
--jit while they are related to each other, and I want to discourage future
JIT implementation-specific (e.g. MJIT-specific) APIs by this rename.
[Feature #17490]
`cd` is passed to method call functions to method invocation
functions, but `cd` can be manipulated by other ractors simultaneously
so it contains thread-safety issue.
To solve this issue, this patch stores `ci` and found `cc` to `calling`
and stops to pass `cd`.
Performance is probably improved?
$ benchmark-driver -v --rbenv 'before --jit;after --jit' --repeat-count=12 --alternate --output=all benchmark.yml
before --jit: ruby 3.0.0dev (2020-11-27T04:37:47Z master 69e77e81dc) +JIT [x86_64-linux]
after --jit: ruby 3.0.0dev (2020-11-27T05:28:19Z master df6b05c6dd) +JIT [x86_64-linux]
last_commit=Set VM_FRAME_FLAG_FINISH at once
Calculating -------------------------------------
before --jit after --jit
Optcarrot Lan_Master.nes 80.89292998533379 82.19497327502751 fps
80.93130641142331 85.13943315260148
81.06214830270119 87.43757879797808
82.29172808453910 87.89942441487113
84.61206450455929 87.91309779491075
85.44545883567997 87.98026086648694
86.02923132404449 88.03081060383973
86.07411817365879 88.14650206137341
86.34348799602836 88.32791633649961
87.90257338977324 88.57599644892220
88.58006509876580 88.67426384743277
89.26611118140011 88.81669430874207
This should have no bad impact on VM because this function is ALWAYS_INLINE.
Predefined macros are practices not very well recommended, but can be
better than having no documents at all. Without those predefined
macros, Doxygen confused for instace PUREFUNC(int foo()) to be a
declaration of PUREFUNC, not foo.
Saves comitters' daily life by avoid #include-ing everything from
internal.h to make each file do so instead. This would significantly
speed up incremental builds.
We take the following inclusion order in this changeset:
1. "ruby/config.h", where _GNU_SOURCE is defined (must be the very
first thing among everything).
2. RUBY_EXTCONF_H if any.
3. Standard C headers, sorted alphabetically.
4. Other system headers, maybe guarded by #ifdef
5. Everything else, sorted alphabetically.
Exceptions are those win32-related headers, which tend not be self-
containing (headers have inclusion order dependencies).
prelude.c is an automatically generated file by template/prelude.c.tmpl.
However it does not contain any required functions. So remove it from
dependency.
Also miniprelude.c is included by mini_builtin.c and does not need
to make miniprelude.o.
The script in prelude.rb was embed in MRI to load it (eval this
script at everyboot).
This commit change the loading process of prelude.rb. MRI doesn't
eval a script, but load from compiled binary with builtin feature.
So that Init_prelude() does not load `prelude.rb` now.
FreeBSD make works differently with `-j` option.
> -j max_jobs
> Specify the maximum number of jobs that `make` may have running
> at any one time. The value is saved in `.MAKE.JOBS.` Turns
> compatibility mode off, unless the `B` flag is also specified.
> When compatibility mode is off, all commands associated with a
> target are executed in a single shell invocation as opposed to
> the traditional one shell invocation per line. This can break
> traditional scripts which change directories on each command
> invocation and then expect to start with a fresh environment on
> the next line. It is more efficient to correct the scripts
> rather than turn backwards compatibility on.
Stop using exit, cd, exec in middle of commands.
Instead of reading from the files by the full-path at runtime. As
rbinc files need to be included in distributed tarballs, the
full-paths at the packaging are unavailable at compilation times.
Compilation of extension libraries written in C++ are reportedly
broken due to https://github.com/ruby/ruby/pull/2404
The root cause of this issue was that the definition of ANYARGS
differ between C and C++, and that of C++ is incompatible with the
updated ones.
We are using the incompatibility against itself. In C++ two distinct
function prototypes can be overloaded. We provide the old, ANYARGSed
prototypes in addition to the current granular ones; and let the
older ones warn about types.
* template/unicode_norm_gen.tmpl: Change formatting of output to produce only a
single item (or range) for each line to make future diffs shorter and easier
to understand and check.
* lib/unicode_normalize/tables.rb: output of the above
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67439 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* To be able to run spec/ruby/command_line/dash_encoding_spec.rb
with the in-repo build.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64567 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
We see several occurrence of "diagnostic push/pop" so why not
make them macros. Tested on GCC8 / Clang 6.
Note that ruby.h is intentionally left untouched because we don't
want to introduce new public macros.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64118 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/fake.rb.in: switch RUBY_DESCRIPTION at runtime
depending on the MJIT status.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63523 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* parse.y (yycompile): in some cases (warning, error, dtrace,...),
ruby_sourcefile is expected to be NUL-terminated, so ensure it.
* template/prelude.c.tmpl (prelude_name): NUL-terminate to get rid
of copying static data.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62841 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* configure.ac (MAINLIBS): moved library options for main program
and static libruby, and append MAINLIBS to LIBRUBYARG_STATIC, as
these libraries are not needed for linking to shared libruby.
[ruby-core:85882] [Bug #14422]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62627 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
If MINIRUBY had arguments, which is the case of cross compiling
they wouldn't be parsed correctly and compiling would fail as a RUBY
without arguments would then be present in the Makefile's in ext/*
[ruby-core:85620] [Bug #14486] [Fix GH-1819]
Author: Carl Hörberg <carl.hoerberg@gmail.com>
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62479 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* configure.ac: MJIT_HEADER_INSTALL_DIR to rubyarchhdrdir to
support multiarch.
* Makefile.in (MJIT_HEADER_INSTALL_DIR): configured by multiarch.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62320 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (MJIT_HEADER, MJIT_MIN_HEADER): added hook to separate
intermediate headers per archs.
* defs/gmake.mk: build mjit headers per -arch options, and then
merge the headers with `#ifdef`s.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62245 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Previous commit changed insns.def format. Now is the time for its
generators. In doing so I chose to modernize the system, not just
patch. My attempt includes
- extensive use of Onigumo regular expressions
- split from one big file (instruction.rb) into separated MVC
- partial view
Also, let me take this opportunity to kill old unused features
such as
- stack caching
- minsns / yasmdata which are never seriously used
- yarvarch document generation (moved to doc/)
- vast majority of unused arguments to insns2vm.rb
This commit generates VM source codes that cleanly compile, and
the generated binary passes tests. At least for me.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61784 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Previous commit changed insns.def format. Now is the time for its
generators. In doing so I chose to modernize the system, not just
patch. My attempt includes
- extensive use of Onigumo regular expressions
- split from one big file (instruction.rb) into separated MVC
- partial view
Also, let me take this opportunity to kill old unused features
such as
- stack caching
- minsns / yasmdata which are never seriously used
- yarvarch document generation (moved to doc/)
- vast majority of unused arguments to insns2vm.rb
This commit generates VM source codes that cleanly compile, and
the generated binary passes tests. At least for me.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61733 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/insns_info.inc.tmpl (rb_vm_insn_name_info): split
instruction names pool not to exceed C90 limit.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61723 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/insns_info.inc.tmpl (insn_name_info): make position
independent for large strings.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61589 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/prelude.c.tmpl: ignore missing-field-initializers on
old gcc, e.g. 4.4, which does not support pushing/popping
diagnostics.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61524 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/prelude.c.tmpl: moved diagnostic pragmas outside
prelude_eval() for very old gcc. [ruby-core:84449] [Bug #14234]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61478 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/prelude.c.tmpl (PRELUDE_STR): make string objects with
static literals not to copy them.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61460 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/prelude.c.tmpl (prelude_eval): fix optimization
options. trace_instruction member has been removed. filled
coverage_enabled and debug_level members instead.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61459 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/prelude.c.tmpl (Init_prelude): fix line number of
preludes. line of prelude_eval is an int, not a VALUE.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61042 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk: download emoji-data.txt. As emoji data files are
located in a separate directory in Unicode.org site, reearranged
Unicode data files directories same as the site.
* tool/enc-unicode.rb (get_file): search emoji data files in the
second argument path.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60977 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/prelude.c.tmpl: split prelude code into blocks so that
each elements do not exceed the string literal size limit in
C89.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60953 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/prelude.c.tmpl (translate): revert r60873 because when some errors or
warnings are shown their line numbers are shifted.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60950 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/prelude.c.tmpl (translate): empty (only LF) lines are not necessary.
so skip them, but for safety only when they are made from comment line.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60873 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* iseq.c (rb_insn_operand_intern): show local variable operand
name in unified instructions.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60732 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
This re-introduces r60485.
This reverts commit 5a176b75b1.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60488 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
NODEs in AST are no longer objects managed by GC. This change will
remove the restriction imposed by the GC. For example, a NODE can use
more than five words (this is my primary purpose; we want to store the
position data for each NODE, for coverage library), or even a NODE can
have variable length (some kinds of NODEs have unused fields).
To do this, however, we need more work, since Ripper still uses T_NODE
objects managed by the GC.
The life time of NODEs is more obvious than other kinds of objects; they
are created at parsing, and they become disused immediately after
compilation. This change releases all NODEs by a few `xfree`s after
compilation, so performance will be improved a bit. In extreme example,
`eval("x=1;" * 10000000)` runs much faster (40 sec. -> 7.8 sec. on my
machine).
The most important part of this change is `ast_t` struct, which has
three contents: (1) NODE buffer (malloc'ed memory), (2) a reference to
the root NODE, and (3) an array that contains objects that must be
marked during parsing (such as literal objects). Some functions that
had received `NODE*` arguments, must now receive `ast_t*`.
* node.c, node.h: defines `ast_t` struct and related operations.
* gc.c, internal.h: defines `imemo_ast`.
* parse.y: makes `parser_params` struct have a reference to `ast_t`.
Instead of `rb_node_newnode`, use `rb_ast_newnode` to create a NODE.
* iseq.c, load.c, ruby.c, template/prelude.c.tmpl: modifies some
functions to handle `ast_t*` instead of `NODE*`.
* test/ruby/test_gc.rb: ad-hoc fix for a failed test. The test assumes
GC eden is increased at startup by NODE object creation. However,
this change now create no NODE object, so GC eden is not necessarily
increased.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60485 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* configure.ac (LIBRUBY_SO): get rid of referrence to
LIBRUBY_SONAME which is not present in ruby.pc.
* template/ruby.pc.in (RUBY_API_VERSION, SOEXT): add new
variables. [ruby-core:83208] [Bug #14002]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60187 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/verconf.h.tmpl: disable site and vendor directories
when removed. based on the path by arnoldwald (arnold w) at
[ruby-core:81563]. [Bug #13631]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60141 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/sizes.c.tmpl (Init_limits): turned a K&R style
declaration into a prototype declaration.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60065 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ruby-runner.c (insert_env_path): extracted the function which
insert path list to an environment variable.
* ruby-runner.c (main): append library paths to RUBYLIB.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@60006 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/insns.inc.tmpl (ASSERT_VM_INSTRUCTION_SIZE): static
assertion for VM instruction info tables.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59635 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/exts.mk.tmpl: get rid of making multiple rubies
simultaneously. it can cause making libruby in parallel.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58711 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/extmk.rb: split notes into header and footer, which are
common, from bodies which are unique for each extensions.
* template/exts.mk.tmpl: now each notes are not one line, should
not unique.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58425 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (build-ext): separate note from building extensions,
not to be intermingled. [ruby-core:80759] [Feature #13302]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58414 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/extmk.rb: fail if a mandatory extension library failed to
configure. [ruby-core:80759] [Feature #13302]
* template/exts.mk.tmpl: move `exit` at the end.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58412 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/limits.c.tmpl (Init_limits): rename RbConfig::Limits as
RbConfig::LIMITS, constants other than class or module are all
uppercase with underscores by convention.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58260 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/exts.mk.tmpl: place note footer message at the end.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@58226 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/configure-ext.mk.tmpl (EXTMK_ARGS): add MINIRUBY for
cross-compling, which is used in extmk.rb.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57764 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* parse.y (TOKEN2ID): add macro which maps static tokens to IDs.
* template/id.h.tmpl (TOKEN2*ID, DEFINE_*ID_FROM_TOKEN): separate
into macros, token to ID mapping and enum definitions.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57598 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (mflags): pass make flags to sub-makes, for nmake
which cannot pass them by the environment variable.
* defs/gmake.mk (mflags): filter out -j option for sub-makes.
* template/exts.mk.tmpl (MFLAGS): extract MFLAGS from sub extmk
files for nmake.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57594 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
It is unnecessary to pass MFLAGS to sub-makes, because it is
evaluated implicitly if set by make. GNU make 4.2 sets both of
-jN and --jobserver-auth (or --jobserver-fds in older versions)
options which conflict. And it is useless for command line macros
since Microsoft NMAKE does not set them.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57489 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/limits.c.tmpl (ULLONG_MAX): add a fallback definition.
On old CentOS lacks ULLONG_MAX.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57475 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Because our tests now have several places where FIXNUM_MAX is needed,
we decided to provide it along with several other constants.
* template/limits.c.tmpl: new file, defining RbConfig::Limits
* ext/rbconfig/sizeof/depend (limits.c): rule to generate limits.c
* test/-ext-/num2int/test_num2int.rb: use RbConfig::Limits
* bootstraptest/test_insns.rb: ditto.
* .gitignore: ignore new generated file.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57465 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/exts.mk.tmpl: use `cd` then `make` instead of `make -C`
when not GNU make.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57438 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/configure-ext.mk.tmpl: fixed inverted names of target
and directory, and chomp the last slash.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57433 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/configure-ext.mk.tmpl: embed MINIRUBY and SCRIPT_ARGS
to get rid of quoting problems of nmake.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57404 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* common.mk (EXT_MK): use double-quotes and remove SCRIPT_ARGS
which contains both types of quotes.
* template/configure-ext.mk.tmpl (all): use single-quotes for MAKE
which is set by nmake and contains spaces in the path. do not
use SCRIPT_ARGS.
* template/exts.mk.tmpl (all, static): separate dependency lines,
not to become a default target unintentionally.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57403 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/configure-ext.mk: configure each directories underneath ext
in parallel.
* template/exts.mk.tmpl: then collect the results.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57401 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/extinit.c.tmpl: drop rest from the first dot in the
base name of a feature is ignored since r30464.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@57398 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* {cygwin,template}/GNUmakefile.in (MAKEFLAGS): remove -jN option
from MAKEFLAGS as well as MFLAGS.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@55505 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* ext/rbconfig/sizeof/extconf.rb: just check the existence of each
types, to reduce configuration time, especially cross-compiling.
* template/sizes.c.tmpl: calculate sizes of checked types at
compilation time.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54759 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/ruby-runner.h.in: separate configured part from
ruby-runner.c.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54572 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
We should try with Ruby 2.4 or 3.0.
[Bug #11842][ruby-core:72374]
* lib/rdoc/erb_partial.rb: ditto.
* template/verconf.h.tmpl: ditto.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@53216 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* compile.c (iseq_peephole_optimize): peephole optimization for
branchnil jumps.
* compile.c (iseq_compile_each): generate save navigation operator
code.
* insns.def (branchnil): new opcode to pop the tos and branch if
it is nil.
* parse.y (NEW_QCALL, call_op, parser_yylex): parse token '.?'.
[Feature #11537]
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52214 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/ruby-runner.c.in: wrapper to set dynamic loading path
environment variable. /bin/sh on Mac OS X 10.11 (El Capitan)
clears DYLD_LIBRARY_PATH.
it must:
- do nothing even if current directory is not present
- do not set other environment variables, e.g. PWD, SHLVL, etc
- do not open other FDs, e.g. pipes for timer thread
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52008 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* template/unicode_norm_gen.tmpl: pragma needs the end marker too,
not only the beginning marker.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51972 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
get rid of many .freeze commands by using frozen_string_literal
pragma.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@51971 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Nobuyoshi Nakada
Yusuke Endoh
Benoit Daloze
Takashi Kokubun
Koichi Sasada
Stefan Stüben
卜部昌平
Jeremy Evans
Dylan Thacker-Smith
Kazuhiro NISHIYAMA
Samuel Williams
nobu
duerst
mrkn
eregon
shyouhei
mame
hsbt
usa
ko1
svn
yugui
naruse
stomar
kazu