which has been developed by Takashi Kokubun <takashikkbn@gmail> as
YARV-MJIT. Many of its bugs are fixed by wanabe <s.wanabe@gmail.com>.
This JIT compiler is designed to be a safe migration path to introduce
JIT compiler to MRI. So this commit does not include any bytecode
changes or dynamic instruction modifications, which are done in original
MJIT.
This commit even strips off some aggressive optimizations from
YARV-MJIT, and thus it's slower than YARV-MJIT too. But it's still
fairly faster than Ruby 2.5 in some benchmarks (attached below).
Note that this JIT compiler passes `make test`, `make test-all`, `make
test-spec` without JIT, and even with JIT. Not only it's perfectly safe
with JIT disabled because it does not replace VM instructions unlike
MJIT, but also with JIT enabled it stably runs Ruby applications
including Rails applications.
I'm expecting this version as just "initial" JIT compiler. I have many
optimization ideas which are skipped for initial merging, and you may
easily replace this JIT compiler with a faster one by just replacing
mjit_compile.c. `mjit_compile` interface is designed for the purpose.
common.mk: update dependencies for mjit_compile.c.
internal.h: declare `rb_vm_insn_addr2insn` for MJIT.
vm.c: exclude some definitions if `-DMJIT_HEADER` is provided to
compiler. This avoids to include some functions which take a long time
to compile, e.g. vm_exec_core. Some of the purpose is achieved in
transform_mjit_header.rb (see `IGNORED_FUNCTIONS`) but others are
manually resolved for now. Load mjit_helper.h for MJIT header.
mjit_helper.h: New. This is a file used only by JIT-ed code. I'll
refactor `mjit_call_cfunc` later.
vm_eval.c: add some #ifdef switches to skip compiling some functions
like Init_vm_eval.
win32/mkexports.rb: export thread/ec functions, which are used by MJIT.
include/ruby/defines.h: add MJIT_FUNC_EXPORTED macro alis to clarify
that a function is exported only for MJIT.
array.c: export a function used by MJIT.
bignum.c: ditto.
class.c: ditto.
compile.c: ditto.
error.c: ditto.
gc.c: ditto.
hash.c: ditto.
iseq.c: ditto.
numeric.c: ditto.
object.c: ditto.
proc.c: ditto.
re.c: ditto.
st.c: ditto.
string.c: ditto.
thread.c: ditto.
variable.c: ditto.
vm_backtrace.c: ditto.
vm_insnhelper.c: ditto.
vm_method.c: ditto.
I would like to improve maintainability of function exports, but I
believe this way is acceptable as initial merging if we clarify the
new exports are for MJIT (so that we can use them as TODO list to fix)
and add unit tests to detect unresolved symbols.
I'll add unit tests of JIT compilations in succeeding commits.
Author: Takashi Kokubun <takashikkbn@gmail.com>
Contributor: wanabe <s.wanabe@gmail.com>
Part of [Feature #14235]
---
* Known issues
* Code generated by gcc is faster than clang. The benchmark may be worse
in macOS. Following benchmark result is provided by gcc w/ Linux.
* Performance is decreased when Google Chrome is running
* JIT can work on MinGW, but it doesn't improve performance at least
in short running benchmark.
* Currently it doesn't perform well with Rails. We'll try to fix this
before release.
---
* Benchmark reslts
Benchmarked with:
Intel 4.0GHz i7-4790K with 16GB memory under x86-64 Ubuntu 8 Cores
- 2.0.0-p0: Ruby 2.0.0-p0
- r62186: Ruby trunk (early 2.6.0), before MJIT changes
- JIT off: On this commit, but without `--jit` option
- JIT on: On this commit, and with `--jit` option
** Optcarrot fps
Benchmark: https://github.com/mame/optcarrot
| |2.0.0-p0 |r62186 |JIT off |JIT on |
|:--------|:--------|:--------|:--------|:--------|
|fps |37.32 |51.46 |51.31 |58.88 |
|vs 2.0.0 |1.00x |1.38x |1.37x |1.58x |
** MJIT benchmarks
Benchmark: https://github.com/benchmark-driver/mjit-benchmarks
(Original: https://github.com/vnmakarov/ruby/tree/rtl_mjit_branch/MJIT-benchmarks)
| |2.0.0-p0 |r62186 |JIT off |JIT on |
|:----------|:--------|:--------|:--------|:--------|
|aread |1.00 |1.09 |1.07 |2.19 |
|aref |1.00 |1.13 |1.11 |2.22 |
|aset |1.00 |1.50 |1.45 |2.64 |
|awrite |1.00 |1.17 |1.13 |2.20 |
|call |1.00 |1.29 |1.26 |2.02 |
|const2 |1.00 |1.10 |1.10 |2.19 |
|const |1.00 |1.11 |1.10 |2.19 |
|fannk |1.00 |1.04 |1.02 |1.00 |
|fib |1.00 |1.32 |1.31 |1.84 |
|ivread |1.00 |1.13 |1.12 |2.43 |
|ivwrite |1.00 |1.23 |1.21 |2.40 |
|mandelbrot |1.00 |1.13 |1.16 |1.28 |
|meteor |1.00 |2.97 |2.92 |3.17 |
|nbody |1.00 |1.17 |1.15 |1.49 |
|nest-ntimes|1.00 |1.22 |1.20 |1.39 |
|nest-while |1.00 |1.10 |1.10 |1.37 |
|norm |1.00 |1.18 |1.16 |1.24 |
|nsvb |1.00 |1.16 |1.16 |1.17 |
|red-black |1.00 |1.02 |0.99 |1.12 |
|sieve |1.00 |1.30 |1.28 |1.62 |
|trees |1.00 |1.14 |1.13 |1.19 |
|while |1.00 |1.12 |1.11 |2.41 |
** Discourse's script/bench.rb
Benchmark: https://github.com/discourse/discourse/blob/v1.8.7/script/bench.rb
NOTE: Rails performance was somehow a little degraded with JIT for now.
We should fix this.
(At least I know opt_aref is performing badly in JIT and I have an idea
to fix it. Please wait for the fix.)
*** JIT off
Your Results: (note for timings- percentile is first, duration is second in millisecs)
categories_admin:
50: 17
75: 18
90: 22
99: 29
home_admin:
50: 21
75: 21
90: 27
99: 40
topic_admin:
50: 17
75: 18
90: 22
99: 32
categories:
50: 35
75: 41
90: 43
99: 77
home:
50: 39
75: 46
90: 49
99: 95
topic:
50: 46
75: 52
90: 56
99: 101
*** JIT on
Your Results: (note for timings- percentile is first, duration is second in millisecs)
categories_admin:
50: 19
75: 21
90: 25
99: 33
home_admin:
50: 24
75: 26
90: 30
99: 35
topic_admin:
50: 19
75: 20
90: 25
99: 30
categories:
50: 40
75: 44
90: 48
99: 76
home:
50: 42
75: 48
90: 51
99: 89
topic:
50: 49
75: 55
90: 58
99: 99
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62197 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* tool/ruby_vm/models/bare_instructions.rb (predefine_attributes):
`sp_inc` attribute which may return negative values must be
signed `rb_snum_t`, to be signed-expanded at type promotion.
* vm_insnhelper.h (ADJ_SP): removed the workaround for platforms
where rb_num_t is wider than int.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62103 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Now that sp_inc attributes are officially provided as inline
functions. Why not use them directly from the vm core, not just
by the compiler. By doing so, it is now possible for us to
optimize stack manipulations. We can now know exactly how many
words of stack space an instruction consumes before it actually
does. This changeset deletes some lines from insns.def because
they are no longer needed. As a result it reduces the size of
vm_exec_core function from 32,400 bytes to 32,352 bytes on my
machine.
It seems it does not affect performance:
-----------------------------------------------------------
benchmark results:
minimum results in each 3 measurements.
Execution time (sec)
name before after
loop_for 1.093 1.061
loop_generator 1.156 1.152
loop_times 0.982 0.974
loop_whileloop 0.549 0.587
loop_whileloop2 0.115 0.121
Speedup ratio: compare with the result of `before' (greater is better)
name after
loop_for 1.030
loop_generator 1.003
loop_times 1.008
loop_whileloop 0.935
loop_whileloop2 0.949
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62087 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Instead of using magic numbers, let us define a series of attributes
and use them from the VM core. Proper function declarations makes
these attributes inlined in most modern compilers. On my machine
exact same binary is generated with or without this changeset.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62085 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
as it's helpful for debugging.
I'm not sure what's the good output for RubyVM::TraceInstructions, so I
left it as it is.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62068 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
which is only used by #router.
To make it private, we need to stop defining them on top-level.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62065 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
via current directory. It's kind of a global state and fragile to
implementation changes in other places, and how the `Dir.getwd` is set
is not obvious from just reading around `RubyVM::Dumper#initialize` if
it depends on the global state.
tool/ruby_vm/controllers/application_controller.rb: explicitly pass
destdir to RubyVM::Dumper.
tool/ruby_vm/scripts/insns2vm.rb: explicitly pass destdir parsed from
optparse.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62064 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
`signed` is required for Rasbian (x86_64).
* tool/ruby_vm/views/_insn_stack_increase.erb: specify `signed` explicitly
for systems which use `unsigned` for `char` type.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61792 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
This was not requested :) but actually easier than the previous
so I just did it anyway.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61787 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
- Gave up @j comments
- Room for sp_inc to be a proper grammer element
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61782 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
* tool/ruby_vm/views/_insn_stack_increase.erb: specify `signed` explicitly
for systems which use `unsigned` for `char` type.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61746 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@61735 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
- Gave up @j comments
- Room for sp_inc to be a proper grammer element
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@61729 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
k0kubun
nobu
shyouhei
hsbt
ko1
naruse
svn