2018-02-08 07:40:33 -05:00
|
|
|
# frozen_string_literal: true
|
|
|
|
require 'test/unit'
|
2018-05-02 09:11:53 -04:00
|
|
|
require 'tmpdir'
|
2019-07-09 08:04:07 -04:00
|
|
|
require_relative '../lib/jit_support'
|
2018-02-11 22:33:00 -05:00
|
|
|
|
|
|
|
# Test for --jit option
|
|
|
|
class TestJIT < Test::Unit::TestCase
|
2018-02-22 10:11:12 -05:00
|
|
|
include JITSupport
|
2018-05-02 11:41:43 -04:00
|
|
|
|
mjit.c: introduce JIT compaction [experimental]
When all compilation finishes or the number of JIT-ed code reaches
--jit-max-cache, this compacts all generated code to a single .so file
and re-loads all methods from it.
In the future, it may trigger compaction more frequently and/or limit
the maximum times of compaction to prevent unlimited memory usage.
So the current behavior is experimental, but at least the performance
improvement in this commit won't be removed.
=== Benchmark ===
In this benchmark, I'll compare following four conditions:
* trunk: r64082
* trunk JIT: r64082 w/ --jit
* single-so JIT: This commit w/ --jit
* objfcn JIT: This branch https://github.com/k0kubun/ruby/tree/objfcn w/ --jit,
which is shinh's objfcn https://github.com/shinh/ruby/tree/objfcn rebased from this commit
```
$ uname -a
Linux bionic 4.15.0-29-generic #31-Ubuntu SMP Tue Jul 17 15:39:52 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
```
* Micro benchmark
Using this script https://gist.github.com/k0kubun/10e6d3387c9ab1b134622b2c9d76ef51,
calls some amount of different methods that just return `nil`. The following tables
are its average duration seconds of 3 measurements.
Smaller is better.
** 1 method (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.576067774333296 | 5.915551971666446 | 5.833641665666619 | 5.845915191666639 |
| Ratio | 1.00x | 1.06x | 1.05x | 1.05x |
** 50 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 3.1661167996666677| 6.125825928333342 | 4.135432743666665 | 3.750358728333348 |
| Ratio | 1.00x | 1.93x | 1.31x | 1.18x |
** 1500 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.971650823666664 | 19.579182102999994| 10.511108153999961| 10.854653588999932|
| Ratio | 1.00x | 3.28x | 1.76x | 1.82x |
* Discourse
Using the same benchmark strategy as https://bugs.ruby-lang.org/issues/14490 with
this branch https://github.com/k0kubun/discourse/commits/benchmark2 forked from discourse
v1.8.11 to support running trunk.
1. Run ruby script/bench.rb to warm up profiling database
2. Run RUBYOPT='--jit-verbose=1 --jit-max-cache=10000' RAILS_ENV=profile bin/puma -e production
3. WAIT 5-15 or so minutes for all jitting to stop so we have no cross talk
4. Run ab -n 100 http://localhost:9292/
5. Wait for all new jitting to finish
6. Run ab -n 100 http://localhost:9292/
** Response time (ms)
Here is the response time milliseconds for each percentile.
Skipping 99%ile because it's the same as 100%ile in 100 calls.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 38 | 45 | 41 | 43 |
| 66% | 39 | 50 | 44 | 44 |
| 75% | 47 | 51 | 46 | 45 |
| 80% | 49 | 52 | 47 | 47 |
| 90% | 50 | 63 | 50 | 52 |
| 95% | 60 | 79 | 52 | 55 |
| 98% | 91 | 114 | 91 | 91 |
|100% | 97 | 133 | 96 | 99 |
** Ratio (smaller is better)
Here is the response time increase ratio against no-JIT trunk's one.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 1.00x| 1.18x| 1.08x| 1.13x|
| 66% | 1.00x| 1.28x| 1.13x| 1.13x|
| 75% | 1.00x| 1.09x| 0.98x| 0.96x|
| 80% | 1.00x| 1.06x| 0.96x| 0.96x|
| 90% | 1.00x| 1.26x| 1.00x| 1.04x|
| 95% | 1.00x| 1.32x| 0.87x| 0.92x|
| 98% | 1.00x| 1.25x| 1.00x| 1.00x|
|100% | 1.00x| 1.37x| 0.99x| 1.02x|
While 50 and 60 %ile are still worse than no-JIT trunk, 75, 80, 90, 95,
98 and 100% are not slower than that.
So now it's a little harder to say "MJIT slows down Rails applications".
Probably I can close [Bug #14490] now. Let's start improving it.
Close https://github.com/ruby/ruby/pull/1921
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64094 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-07-28 12:14:56 -04:00
|
|
|
IGNORABLE_PATTERNS = [
|
2019-04-14 00:52:02 -04:00
|
|
|
/\AJIT recompile: .+\n\z/,
|
2019-07-02 21:00:50 -04:00
|
|
|
/\AJIT inline: .+\n\z/,
|
2021-08-13 01:54:40 -04:00
|
|
|
/\AJIT cancel: .+\n\z/,
|
mjit.c: introduce JIT compaction [experimental]
When all compilation finishes or the number of JIT-ed code reaches
--jit-max-cache, this compacts all generated code to a single .so file
and re-loads all methods from it.
In the future, it may trigger compaction more frequently and/or limit
the maximum times of compaction to prevent unlimited memory usage.
So the current behavior is experimental, but at least the performance
improvement in this commit won't be removed.
=== Benchmark ===
In this benchmark, I'll compare following four conditions:
* trunk: r64082
* trunk JIT: r64082 w/ --jit
* single-so JIT: This commit w/ --jit
* objfcn JIT: This branch https://github.com/k0kubun/ruby/tree/objfcn w/ --jit,
which is shinh's objfcn https://github.com/shinh/ruby/tree/objfcn rebased from this commit
```
$ uname -a
Linux bionic 4.15.0-29-generic #31-Ubuntu SMP Tue Jul 17 15:39:52 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
```
* Micro benchmark
Using this script https://gist.github.com/k0kubun/10e6d3387c9ab1b134622b2c9d76ef51,
calls some amount of different methods that just return `nil`. The following tables
are its average duration seconds of 3 measurements.
Smaller is better.
** 1 method (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.576067774333296 | 5.915551971666446 | 5.833641665666619 | 5.845915191666639 |
| Ratio | 1.00x | 1.06x | 1.05x | 1.05x |
** 50 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 3.1661167996666677| 6.125825928333342 | 4.135432743666665 | 3.750358728333348 |
| Ratio | 1.00x | 1.93x | 1.31x | 1.18x |
** 1500 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.971650823666664 | 19.579182102999994| 10.511108153999961| 10.854653588999932|
| Ratio | 1.00x | 3.28x | 1.76x | 1.82x |
* Discourse
Using the same benchmark strategy as https://bugs.ruby-lang.org/issues/14490 with
this branch https://github.com/k0kubun/discourse/commits/benchmark2 forked from discourse
v1.8.11 to support running trunk.
1. Run ruby script/bench.rb to warm up profiling database
2. Run RUBYOPT='--jit-verbose=1 --jit-max-cache=10000' RAILS_ENV=profile bin/puma -e production
3. WAIT 5-15 or so minutes for all jitting to stop so we have no cross talk
4. Run ab -n 100 http://localhost:9292/
5. Wait for all new jitting to finish
6. Run ab -n 100 http://localhost:9292/
** Response time (ms)
Here is the response time milliseconds for each percentile.
Skipping 99%ile because it's the same as 100%ile in 100 calls.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 38 | 45 | 41 | 43 |
| 66% | 39 | 50 | 44 | 44 |
| 75% | 47 | 51 | 46 | 45 |
| 80% | 49 | 52 | 47 | 47 |
| 90% | 50 | 63 | 50 | 52 |
| 95% | 60 | 79 | 52 | 55 |
| 98% | 91 | 114 | 91 | 91 |
|100% | 97 | 133 | 96 | 99 |
** Ratio (smaller is better)
Here is the response time increase ratio against no-JIT trunk's one.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 1.00x| 1.18x| 1.08x| 1.13x|
| 66% | 1.00x| 1.28x| 1.13x| 1.13x|
| 75% | 1.00x| 1.09x| 0.98x| 0.96x|
| 80% | 1.00x| 1.06x| 0.96x| 0.96x|
| 90% | 1.00x| 1.26x| 1.00x| 1.04x|
| 95% | 1.00x| 1.32x| 0.87x| 0.92x|
| 98% | 1.00x| 1.25x| 1.00x| 1.00x|
|100% | 1.00x| 1.37x| 0.99x| 1.02x|
While 50 and 60 %ile are still worse than no-JIT trunk, 75, 80, 90, 95,
98 and 100% are not slower than that.
So now it's a little harder to say "MJIT slows down Rails applications".
Probably I can close [Bug #14490] now. Let's start improving it.
Close https://github.com/ruby/ruby/pull/1921
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64094 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-07-28 12:14:56 -04:00
|
|
|
/\ASuccessful MJIT finish\n\z/,
|
|
|
|
]
|
2020-02-29 02:58:33 -05:00
|
|
|
MAX_CACHE_PATTERNS = [
|
|
|
|
/\AJIT compaction \([^)]+\): .+\n\z/,
|
2020-05-01 00:37:55 -04:00
|
|
|
/\AToo many JIT code, but skipped unloading units for JIT compaction\n\z/,
|
2020-02-29 02:58:33 -05:00
|
|
|
/\ANo units can be unloaded -- .+\n\z/,
|
|
|
|
]
|
mjit.c: introduce JIT compaction [experimental]
When all compilation finishes or the number of JIT-ed code reaches
--jit-max-cache, this compacts all generated code to a single .so file
and re-loads all methods from it.
In the future, it may trigger compaction more frequently and/or limit
the maximum times of compaction to prevent unlimited memory usage.
So the current behavior is experimental, but at least the performance
improvement in this commit won't be removed.
=== Benchmark ===
In this benchmark, I'll compare following four conditions:
* trunk: r64082
* trunk JIT: r64082 w/ --jit
* single-so JIT: This commit w/ --jit
* objfcn JIT: This branch https://github.com/k0kubun/ruby/tree/objfcn w/ --jit,
which is shinh's objfcn https://github.com/shinh/ruby/tree/objfcn rebased from this commit
```
$ uname -a
Linux bionic 4.15.0-29-generic #31-Ubuntu SMP Tue Jul 17 15:39:52 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
```
* Micro benchmark
Using this script https://gist.github.com/k0kubun/10e6d3387c9ab1b134622b2c9d76ef51,
calls some amount of different methods that just return `nil`. The following tables
are its average duration seconds of 3 measurements.
Smaller is better.
** 1 method (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.576067774333296 | 5.915551971666446 | 5.833641665666619 | 5.845915191666639 |
| Ratio | 1.00x | 1.06x | 1.05x | 1.05x |
** 50 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 3.1661167996666677| 6.125825928333342 | 4.135432743666665 | 3.750358728333348 |
| Ratio | 1.00x | 1.93x | 1.31x | 1.18x |
** 1500 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.971650823666664 | 19.579182102999994| 10.511108153999961| 10.854653588999932|
| Ratio | 1.00x | 3.28x | 1.76x | 1.82x |
* Discourse
Using the same benchmark strategy as https://bugs.ruby-lang.org/issues/14490 with
this branch https://github.com/k0kubun/discourse/commits/benchmark2 forked from discourse
v1.8.11 to support running trunk.
1. Run ruby script/bench.rb to warm up profiling database
2. Run RUBYOPT='--jit-verbose=1 --jit-max-cache=10000' RAILS_ENV=profile bin/puma -e production
3. WAIT 5-15 or so minutes for all jitting to stop so we have no cross talk
4. Run ab -n 100 http://localhost:9292/
5. Wait for all new jitting to finish
6. Run ab -n 100 http://localhost:9292/
** Response time (ms)
Here is the response time milliseconds for each percentile.
Skipping 99%ile because it's the same as 100%ile in 100 calls.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 38 | 45 | 41 | 43 |
| 66% | 39 | 50 | 44 | 44 |
| 75% | 47 | 51 | 46 | 45 |
| 80% | 49 | 52 | 47 | 47 |
| 90% | 50 | 63 | 50 | 52 |
| 95% | 60 | 79 | 52 | 55 |
| 98% | 91 | 114 | 91 | 91 |
|100% | 97 | 133 | 96 | 99 |
** Ratio (smaller is better)
Here is the response time increase ratio against no-JIT trunk's one.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 1.00x| 1.18x| 1.08x| 1.13x|
| 66% | 1.00x| 1.28x| 1.13x| 1.13x|
| 75% | 1.00x| 1.09x| 0.98x| 0.96x|
| 80% | 1.00x| 1.06x| 0.96x| 0.96x|
| 90% | 1.00x| 1.26x| 1.00x| 1.04x|
| 95% | 1.00x| 1.32x| 0.87x| 0.92x|
| 98% | 1.00x| 1.25x| 1.00x| 1.00x|
|100% | 1.00x| 1.37x| 0.99x| 1.02x|
While 50 and 60 %ile are still worse than no-JIT trunk, 75, 80, 90, 95,
98 and 100% are not slower than that.
So now it's a little harder to say "MJIT slows down Rails applications".
Probably I can close [Bug #14490] now. Let's start improving it.
Close https://github.com/ruby/ruby/pull/1921
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64094 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-07-28 12:14:56 -04:00
|
|
|
|
2018-05-02 11:41:43 -04:00
|
|
|
# trace_* insns are not compiled for now...
|
|
|
|
TEST_PENDING_INSNS = RubyVM::INSTRUCTION_NAMES.select { |n| n.start_with?('trace_') }.map(&:to_sym) + [
|
|
|
|
# not supported yet
|
|
|
|
:defineclass,
|
2019-11-10 00:56:38 -05:00
|
|
|
|
2019-11-18 12:31:36 -05:00
|
|
|
# to be tested
|
|
|
|
:invokebuiltin,
|
|
|
|
|
2019-11-10 00:56:38 -05:00
|
|
|
# never used
|
|
|
|
:opt_invokebuiltin_delegate,
|
2019-09-03 08:38:32 -04:00
|
|
|
].each do |insn|
|
|
|
|
if !RubyVM::INSTRUCTION_NAMES.include?(insn.to_s)
|
|
|
|
warn "instruction #{insn.inspect} is not defined but included in TestJIT::TEST_PENDING_INSNS"
|
|
|
|
end
|
|
|
|
end
|
2018-05-02 11:41:43 -04:00
|
|
|
|
|
|
|
def self.untested_insns
|
|
|
|
@untested_insns ||= (RubyVM::INSTRUCTION_NAMES.map(&:to_sym) - TEST_PENDING_INSNS)
|
|
|
|
end
|
2018-02-08 08:56:32 -05:00
|
|
|
|
2020-03-28 18:19:06 -04:00
|
|
|
def self.setup
|
|
|
|
return if defined?(@setup_hooked)
|
|
|
|
@setup_hooked = true
|
|
|
|
|
|
|
|
# ci.rvm.jp caches its build environment. Clean up temporary files left by SEGV.
|
|
|
|
if ENV['RUBY_DEBUG']&.include?('ci')
|
|
|
|
Dir.glob("#{ENV.fetch('TMPDIR', '/tmp')}/_ruby_mjit_p*u*.*").each do |file|
|
2020-11-26 22:23:34 -05:00
|
|
|
puts "test/ruby/test_jit.rb: removing #{file}"
|
|
|
|
File.unlink(file)
|
2020-03-28 18:19:06 -04:00
|
|
|
end
|
2018-10-19 10:26:29 -04:00
|
|
|
end
|
|
|
|
|
2018-05-02 11:41:43 -04:00
|
|
|
# ruby -w -Itest/lib test/ruby/test_jit.rb
|
2020-03-28 18:19:06 -04:00
|
|
|
if $VERBOSE
|
2021-10-25 08:28:44 -04:00
|
|
|
pid = $$
|
2018-05-02 11:41:43 -04:00
|
|
|
at_exit do
|
2021-10-25 08:28:44 -04:00
|
|
|
if pid == $$ && !TestJIT.untested_insns.empty?
|
2019-08-01 06:42:48 -04:00
|
|
|
warn "you may want to add tests for following insns, when you have a chance: #{TestJIT.untested_insns.join(' ')}"
|
2018-05-02 11:41:43 -04:00
|
|
|
end
|
|
|
|
end
|
|
|
|
end
|
2018-02-11 23:00:45 -05:00
|
|
|
end
|
|
|
|
|
2020-03-28 18:19:06 -04:00
|
|
|
def setup
|
|
|
|
unless JITSupport.supported?
|
|
|
|
skip 'JIT seems not supported on this platform'
|
|
|
|
end
|
|
|
|
self.class.setup
|
|
|
|
end
|
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_nop
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('nil rescue true', result_inspect: 'nil', insns: %i[nop])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_local
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '1', insns: %i[setlocal_WC_0 getlocal_WC_0])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
foo = 1
|
|
|
|
foo
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
2018-05-02 12:27:33 -04:00
|
|
|
|
|
|
|
insns = %i[setlocal getlocal setlocal_WC_0 getlocal_WC_0 setlocal_WC_1 getlocal_WC_1]
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", success_count: 3, stdout: '168', insns: insns)
|
|
|
|
begin;
|
|
|
|
def foo
|
|
|
|
a = 0
|
|
|
|
[1, 2].each do |i|
|
|
|
|
a += i
|
|
|
|
[3, 4].each do |j|
|
|
|
|
a *= j
|
|
|
|
end
|
|
|
|
end
|
|
|
|
a
|
|
|
|
end
|
|
|
|
|
|
|
|
print foo
|
|
|
|
end;
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_blockparam
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '3', success_count: 2, insns: %i[getblockparam setblockparam])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
def foo(&b)
|
|
|
|
a = b
|
|
|
|
b = 2
|
|
|
|
a.call + 2
|
|
|
|
end
|
|
|
|
|
|
|
|
print foo { 1 }
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_getblockparamproxy
|
2019-07-14 05:04:14 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '4', success_count: 3, insns: %i[getblockparamproxy])
|
|
|
|
begin;
|
|
|
|
def bar(&b)
|
|
|
|
b.call
|
|
|
|
end
|
|
|
|
|
|
|
|
def foo(&b)
|
|
|
|
bar(&b) * bar(&b)
|
|
|
|
end
|
|
|
|
|
|
|
|
print foo { 2 }
|
|
|
|
end;
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_getspecial
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('$1', result_inspect: 'nil', insns: %i[getspecial])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_setspecial
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: 'true', insns: %i[setspecial])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
true if nil.nil?..nil.nil?
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_instancevariable
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '1', insns: %i[getinstancevariable setinstancevariable])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
@foo = 1
|
|
|
|
@foo
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
2018-10-14 05:24:43 -04:00
|
|
|
|
|
|
|
# optimized getinstancevariable call
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '33', success_count: 1, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
class A
|
|
|
|
def initialize
|
|
|
|
@a = 1
|
|
|
|
@b = 2
|
|
|
|
end
|
|
|
|
|
|
|
|
def three
|
|
|
|
@a + @b
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
a = A.new
|
|
|
|
print(a.three) # set ic
|
|
|
|
print(a.three) # inlined ic
|
|
|
|
end;
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_classvariable
|
2018-05-22 11:48:58 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '1', success_count: 1, insns: %i[getclassvariable setclassvariable])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-05-22 11:48:58 -04:00
|
|
|
class Foo
|
|
|
|
def self.foo
|
|
|
|
@@foo = 1
|
|
|
|
@@foo
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
print Foo.foo
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_constant
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '1', insns: %i[getconstant setconstant])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
FOO = 1
|
|
|
|
FOO
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_global
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '1', insns: %i[getglobal setglobal])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
$foo = 1
|
|
|
|
$foo
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_putnil
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('nil', result_inspect: 'nil', insns: %i[putnil])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_putself
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: 'hello', success_count: 1, insns: %i[putself])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
proc { print "hello" }.call
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_putobject
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('0', result_inspect: '0', insns: %i[putobject_INT2FIX_0_])
|
|
|
|
assert_compile_once('1', result_inspect: '1', insns: %i[putobject_INT2FIX_1_])
|
|
|
|
assert_compile_once('2', result_inspect: '2', insns: %i[putobject])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2019-04-05 22:53:32 -04:00
|
|
|
def test_compile_insn_definemethod_definesmethod
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: 'helloworld', success_count: 3, insns: %i[definemethod definesmethod])
|
2019-04-05 22:35:37 -04:00
|
|
|
begin;
|
2019-04-05 22:42:30 -04:00
|
|
|
print 1.times.map {
|
2019-04-05 22:35:37 -04:00
|
|
|
def method_definition
|
|
|
|
'hello'
|
|
|
|
end
|
2019-04-05 22:53:32 -04:00
|
|
|
|
|
|
|
def self.smethod_definition
|
|
|
|
'world'
|
|
|
|
end
|
|
|
|
|
|
|
|
method_definition + smethod_definition
|
|
|
|
}.join
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
|
|
|
def test_compile_insn_putspecialobject
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: 'a', success_count: 2, insns: %i[putspecialobject])
|
|
|
|
begin;
|
|
|
|
print 1.times.map {
|
|
|
|
def a
|
|
|
|
'a'
|
|
|
|
end
|
|
|
|
|
|
|
|
alias :b :a
|
|
|
|
|
|
|
|
b
|
2019-04-05 22:35:37 -04:00
|
|
|
}.join
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
Optimize dynamic string interpolation for symbol/true/false/nil/0-9
This provides a significant speedup for symbol, true, false,
nil, and 0-9, class/module, and a small speedup in most other cases.
Speedups (using included benchmarks):
:symbol :: 60%
0-9 :: 50%
Class/Module :: 50%
nil/true/false :: 20%
integer :: 10%
[] :: 10%
"" :: 3%
One reason this approach is faster is it reduces the number of
VM instructions for each interpolated value.
Initial idea, approach, and benchmarks from Eric Wong. I applied
the same approach against the master branch, updating it to handle
the significant internal changes since this was first proposed 4
years ago (such as CALL_INFO/CALL_CACHE -> CALL_DATA). I also
expanded it to optimize true/false/nil/0-9/class/module, and added
handling of missing methods, refined methods, and RUBY_DEBUG.
This renames the tostring insn to anytostring, and adds an
objtostring insn that implements the optimization. This requires
making a few functions non-static, and adding some non-static
functions.
This disables 4 YJIT tests. Those tests should be reenabled after
YJIT optimizes the new objtostring insn.
Implements [Feature #13715]
Co-authored-by: Eric Wong <e@80x24.org>
Co-authored-by: Alan Wu <XrXr@users.noreply.github.com>
Co-authored-by: Yusuke Endoh <mame@ruby-lang.org>
Co-authored-by: Koichi Sasada <ko1@atdot.net>
2021-11-18 18:10:20 -05:00
|
|
|
def test_compile_insn_putstring_concatstrings_objtostring
|
|
|
|
assert_compile_once('"a#{}b" + "c"', result_inspect: '"abc"', insns: %i[putstring concatstrings objtostring])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_toregexp
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('/#{true}/ =~ "true"', result_inspect: '0', insns: %i[toregexp])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
mjit_compile.c: use local variables for stack
if catch_except_p is FALSE. If catch_except_p is TRUE, stack values
should be on VM's stack when exception is thrown and the JIT-ed frame
is re-executed by VM's exception handler. If it's FALSE, the JIT-ed
frame won't be re-executed and don't need to keep values on VM's stack.
Using local variables allows us to reduce cfp->sp motion. Moving cfp->sp
is needed only for insns whose handles_frame? is false. So it improves
performance.
_mjit_compile_insn.erb: Prepare `stack_size` variable for GET_SP,
STACK_ADDR_FROM_TOP, TOPN macros. Share pc and sp motion partial view.
Use cancel handler created in mjit_compile.c.
_mjit_compile_send.erb: ditto. Also, when iseq->body->catch_except_p is
TRUE, this stops to call mjit_exec directly. I described the reason in
vm_insnhelper.h's comment for EXEC_EC_CFP.
_mjit_compile_pc_and_sp.erb: Shared logic for moving sp and pc. As you
can see from thsi file, when status->local_stack_p is TRUE and
insn.handles_frame? is false, moving sp is skipped. But if
insn.handles_frame? is true, values should be rolled back to VM's stack.
common.mk: add dependency for the file
_mjit_compile_insn_body.erb: Set sp value before canceling JIT on
DISPATCH_ORIGINAL_INSN. Replace GET_SP, STACK_ADDR_FROM_TOP, TOPN macros
for the case ocal_stack_p is TRUE and insn.handles_frame? is false.
In that case, values are not available on VM's stack and those macros
should be replaced.
mjit_compile.inc.erb: updated comments of macros which are supported by
JIT compiler. All references to `cfp->sp` should be replaced and thus
INC_SP, SET_SV, PUSH are no longer supported for now, because they are
not used now.
vm_exec.h: moved EXEC_EC_CFP definition to vm_insnhelper.h because it's
tighly coupled to CALL_METHOD.
vm_insnhelper.h: Have revised EXEC_EC_CFP definition moved from vm_exec.h.
Now it triggers mjit_exec for VM, and has the guard for catch_except_p
on JIT-ed code. See comments for details. CALL_METHOD delegates
triggering mjit_exec to EXEC_EC_CFP.
insns.def: Stopped using EXEC_EC_CFP for the case we don't want to
trigger mjit_exec. Those insns (defineclass, opt_call_c_function) are
not supported by JIT and it's safe to use RESTORE_REGS(), NEXT_INSN().
expandarray is changed to pass GET_SP() to replace the macro in
_mjit_compile_insn_body.erb.
vm_insnhelper.c: change to take sp for the above reason.
[close https://github.com/ruby/ruby/pull/1828]
This patch resurrects the performance which was attached in
[Feature #14235].
* Benchmark
Optcarrot (with configuration for benchmark_driver.gem)
https://github.com/benchmark-driver/optcarrot
$ benchmark-driver benchmark.yml --verbose 1 --rbenv 'before;before+JIT::before,--jit;after;after+JIT::after,--jit' --repeat-count 10
before: ruby 2.6.0dev (2018-03-04 trunk 62652) [x86_64-linux]
before+JIT: ruby 2.6.0dev (2018-03-04 trunk 62652) +JIT [x86_64-linux]
after: ruby 2.6.0dev (2018-03-04 local-variable.. 62652) [x86_64-linux]
last_commit=mjit_compile.c: use local variables for stack
after+JIT: ruby 2.6.0dev (2018-03-04 local-variable.. 62652) +JIT [x86_64-linux]
last_commit=mjit_compile.c: use local variables for stack
Calculating -------------------------------------
before before+JIT after after+JIT
optcarrot 53.552 59.680 53.697 63.358 fps
Comparison:
optcarrot
after+JIT: 63.4 fps
before+JIT: 59.7 fps - 1.06x slower
after: 53.7 fps - 1.18x slower
before: 53.6 fps - 1.18x slower
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62655 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-04 02:04:40 -05:00
|
|
|
def test_compile_insn_newarray
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '[1, 2, 3]', insns: %i[newarray])
|
mjit_compile.c: use local variables for stack
if catch_except_p is FALSE. If catch_except_p is TRUE, stack values
should be on VM's stack when exception is thrown and the JIT-ed frame
is re-executed by VM's exception handler. If it's FALSE, the JIT-ed
frame won't be re-executed and don't need to keep values on VM's stack.
Using local variables allows us to reduce cfp->sp motion. Moving cfp->sp
is needed only for insns whose handles_frame? is false. So it improves
performance.
_mjit_compile_insn.erb: Prepare `stack_size` variable for GET_SP,
STACK_ADDR_FROM_TOP, TOPN macros. Share pc and sp motion partial view.
Use cancel handler created in mjit_compile.c.
_mjit_compile_send.erb: ditto. Also, when iseq->body->catch_except_p is
TRUE, this stops to call mjit_exec directly. I described the reason in
vm_insnhelper.h's comment for EXEC_EC_CFP.
_mjit_compile_pc_and_sp.erb: Shared logic for moving sp and pc. As you
can see from thsi file, when status->local_stack_p is TRUE and
insn.handles_frame? is false, moving sp is skipped. But if
insn.handles_frame? is true, values should be rolled back to VM's stack.
common.mk: add dependency for the file
_mjit_compile_insn_body.erb: Set sp value before canceling JIT on
DISPATCH_ORIGINAL_INSN. Replace GET_SP, STACK_ADDR_FROM_TOP, TOPN macros
for the case ocal_stack_p is TRUE and insn.handles_frame? is false.
In that case, values are not available on VM's stack and those macros
should be replaced.
mjit_compile.inc.erb: updated comments of macros which are supported by
JIT compiler. All references to `cfp->sp` should be replaced and thus
INC_SP, SET_SV, PUSH are no longer supported for now, because they are
not used now.
vm_exec.h: moved EXEC_EC_CFP definition to vm_insnhelper.h because it's
tighly coupled to CALL_METHOD.
vm_insnhelper.h: Have revised EXEC_EC_CFP definition moved from vm_exec.h.
Now it triggers mjit_exec for VM, and has the guard for catch_except_p
on JIT-ed code. See comments for details. CALL_METHOD delegates
triggering mjit_exec to EXEC_EC_CFP.
insns.def: Stopped using EXEC_EC_CFP for the case we don't want to
trigger mjit_exec. Those insns (defineclass, opt_call_c_function) are
not supported by JIT and it's safe to use RESTORE_REGS(), NEXT_INSN().
expandarray is changed to pass GET_SP() to replace the macro in
_mjit_compile_insn_body.erb.
vm_insnhelper.c: change to take sp for the above reason.
[close https://github.com/ruby/ruby/pull/1828]
This patch resurrects the performance which was attached in
[Feature #14235].
* Benchmark
Optcarrot (with configuration for benchmark_driver.gem)
https://github.com/benchmark-driver/optcarrot
$ benchmark-driver benchmark.yml --verbose 1 --rbenv 'before;before+JIT::before,--jit;after;after+JIT::after,--jit' --repeat-count 10
before: ruby 2.6.0dev (2018-03-04 trunk 62652) [x86_64-linux]
before+JIT: ruby 2.6.0dev (2018-03-04 trunk 62652) +JIT [x86_64-linux]
after: ruby 2.6.0dev (2018-03-04 local-variable.. 62652) [x86_64-linux]
last_commit=mjit_compile.c: use local variables for stack
after+JIT: ruby 2.6.0dev (2018-03-04 local-variable.. 62652) +JIT [x86_64-linux]
last_commit=mjit_compile.c: use local variables for stack
Calculating -------------------------------------
before before+JIT after after+JIT
optcarrot 53.552 59.680 53.697 63.358 fps
Comparison:
optcarrot
after+JIT: 63.4 fps
before+JIT: 59.7 fps - 1.06x slower
after: 53.7 fps - 1.18x slower
before: 53.6 fps - 1.18x slower
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62655 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-04 02:04:40 -05:00
|
|
|
begin;
|
|
|
|
a, b, c = 1, 2, 3
|
|
|
|
[a, b, c]
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2019-09-02 12:39:00 -04:00
|
|
|
def test_compile_insn_newarraykwsplat
|
|
|
|
assert_compile_once('[**{ x: 1 }]', result_inspect: '[{:x=>1}]', insns: %i[newarraykwsplat])
|
|
|
|
end
|
|
|
|
|
mjit_compile.c: use local variables for stack
if catch_except_p is FALSE. If catch_except_p is TRUE, stack values
should be on VM's stack when exception is thrown and the JIT-ed frame
is re-executed by VM's exception handler. If it's FALSE, the JIT-ed
frame won't be re-executed and don't need to keep values on VM's stack.
Using local variables allows us to reduce cfp->sp motion. Moving cfp->sp
is needed only for insns whose handles_frame? is false. So it improves
performance.
_mjit_compile_insn.erb: Prepare `stack_size` variable for GET_SP,
STACK_ADDR_FROM_TOP, TOPN macros. Share pc and sp motion partial view.
Use cancel handler created in mjit_compile.c.
_mjit_compile_send.erb: ditto. Also, when iseq->body->catch_except_p is
TRUE, this stops to call mjit_exec directly. I described the reason in
vm_insnhelper.h's comment for EXEC_EC_CFP.
_mjit_compile_pc_and_sp.erb: Shared logic for moving sp and pc. As you
can see from thsi file, when status->local_stack_p is TRUE and
insn.handles_frame? is false, moving sp is skipped. But if
insn.handles_frame? is true, values should be rolled back to VM's stack.
common.mk: add dependency for the file
_mjit_compile_insn_body.erb: Set sp value before canceling JIT on
DISPATCH_ORIGINAL_INSN. Replace GET_SP, STACK_ADDR_FROM_TOP, TOPN macros
for the case ocal_stack_p is TRUE and insn.handles_frame? is false.
In that case, values are not available on VM's stack and those macros
should be replaced.
mjit_compile.inc.erb: updated comments of macros which are supported by
JIT compiler. All references to `cfp->sp` should be replaced and thus
INC_SP, SET_SV, PUSH are no longer supported for now, because they are
not used now.
vm_exec.h: moved EXEC_EC_CFP definition to vm_insnhelper.h because it's
tighly coupled to CALL_METHOD.
vm_insnhelper.h: Have revised EXEC_EC_CFP definition moved from vm_exec.h.
Now it triggers mjit_exec for VM, and has the guard for catch_except_p
on JIT-ed code. See comments for details. CALL_METHOD delegates
triggering mjit_exec to EXEC_EC_CFP.
insns.def: Stopped using EXEC_EC_CFP for the case we don't want to
trigger mjit_exec. Those insns (defineclass, opt_call_c_function) are
not supported by JIT and it's safe to use RESTORE_REGS(), NEXT_INSN().
expandarray is changed to pass GET_SP() to replace the macro in
_mjit_compile_insn_body.erb.
vm_insnhelper.c: change to take sp for the above reason.
[close https://github.com/ruby/ruby/pull/1828]
This patch resurrects the performance which was attached in
[Feature #14235].
* Benchmark
Optcarrot (with configuration for benchmark_driver.gem)
https://github.com/benchmark-driver/optcarrot
$ benchmark-driver benchmark.yml --verbose 1 --rbenv 'before;before+JIT::before,--jit;after;after+JIT::after,--jit' --repeat-count 10
before: ruby 2.6.0dev (2018-03-04 trunk 62652) [x86_64-linux]
before+JIT: ruby 2.6.0dev (2018-03-04 trunk 62652) +JIT [x86_64-linux]
after: ruby 2.6.0dev (2018-03-04 local-variable.. 62652) [x86_64-linux]
last_commit=mjit_compile.c: use local variables for stack
after+JIT: ruby 2.6.0dev (2018-03-04 local-variable.. 62652) +JIT [x86_64-linux]
last_commit=mjit_compile.c: use local variables for stack
Calculating -------------------------------------
before before+JIT after after+JIT
optcarrot 53.552 59.680 53.697 63.358 fps
Comparison:
optcarrot
after+JIT: 63.4 fps
before+JIT: 59.7 fps - 1.06x slower
after: 53.7 fps - 1.18x slower
before: 53.6 fps - 1.18x slower
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62655 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-04 02:04:40 -05:00
|
|
|
def test_compile_insn_intern_duparray
|
2018-05-02 11:41:43 -04:00
|
|
|
assert_compile_once('[:"#{0}"] + [1,2,3]', result_inspect: '[:"0", 1, 2, 3]', insns: %i[intern duparray])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_expandarray
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('y = [ true, false, nil ]; x, = y; x', result_inspect: 'true', insns: %i[expandarray])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_concatarray
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('["t", "r", *x = "u", "e"].join', result_inspect: '"true"', insns: %i[concatarray])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_splatarray
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('[*(1..2)]', result_inspect: '[1, 2]', insns: %i[splatarray])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_newhash
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('a = 1; { a: a }', result_inspect: '{:a=>1}', insns: %i[newhash])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
Speed up hash literals by duping
This commit replaces the `newhashfromarray` instruction with a `duphash`
instruction. Instead of allocating a new hash from an array stored in
the Instruction Sequences, store a hash directly in the instruction
sequences and dup it on execution.
== Instruction sequence changes ==
```ruby
code = <<-eorby
{ "foo" => "bar", "baz" => "lol" }
eorby
insns = RubyVM::InstructionSequence.compile(code, __FILE__, nil, 0, frozen_string_literal: true)
puts insns.disasm
```
On Ruby 2.5:
```
== disasm: #<ISeq:<compiled>@test.rb:0 (0,0)-(0,36)>====================
0000 putobject "foo"
0002 putobject "bar"
0004 putobject "baz"
0006 putobject "lol"
0008 newhash 4
0010 leave
```
Ruby 2.6@r66174 3b6321083a2e3525da3b34d08a0b68bac094bd7f:
```
$ ./ruby test.rb
== disasm: #<ISeq:<compiled>@test.rb:0 (0,0)-(0,36)> (catch: FALSE)
0000 newhashfromarray 2, ["foo", "bar", "baz", "lol"]
0003 leave
```
Ruby 2.6 + This commit:
```
$ ./ruby test.rb
== disasm: #<ISeq:<compiled>@test.rb:0 (0,0)-(0,36)> (catch: FALSE)
0000 duphash {"foo"=>"bar", "baz"=>"lol"}
0002 leave
```
== Benchmark Results ==
Compared to 2.5.3:
```
$ make benchmark ITEM=hash_literal_small COMPARE_RUBY=/Users/aaron/.rbenv/versions/2.5.3/bin/ruby
generating known_errors.inc
known_errors.inc unchanged
./revision.h unchanged
/Users/aaron/.rbenv/shims/ruby --disable=gems -rrubygems -I./benchmark/lib ./benchmark/benchmark-driver/exe/benchmark-driver \
--executables="compare-ruby::/Users/aaron/.rbenv/versions/2.5.3/bin/ruby -I.ext/common --disable-gem" \
--executables="built-ruby::./miniruby -I./lib -I. -I.ext/common -r./prelude --disable-gem" \
$(find ./benchmark -maxdepth 1 -name '*hash_literal_small*.yml' -o -name '*hash_literal_small*.rb' | sort)
Calculating -------------------------------------
compare-ruby built-ruby
hash_literal_small2 1.498 1.877 i/s - 1.000 times in 0.667581s 0.532656s
hash_literal_small4 1.197 1.642 i/s - 1.000 times in 0.835375s 0.609160s
hash_literal_small8 0.620 1.215 i/s - 1.000 times in 1.611638s 0.823090s
Comparison:
hash_literal_small2
built-ruby: 1.9 i/s
compare-ruby: 1.5 i/s - 1.25x slower
hash_literal_small4
built-ruby: 1.6 i/s
compare-ruby: 1.2 i/s - 1.37x slower
hash_literal_small8
built-ruby: 1.2 i/s
compare-ruby: 0.6 i/s - 1.96x slower
```
Compared to r66255
```
$ make benchmark ITEM=hash_literal_small COMPARE_RUBY=/Users/aaron/.rbenv/versions/ruby-trunk/bin/ruby
generating known_errors.inc
known_errors.inc unchanged
./revision.h unchanged
/Users/aaron/.rbenv/shims/ruby --disable=gems -rrubygems -I./benchmark/lib ./benchmark/benchmark-driver/exe/benchmark-driver \
--executables="compare-ruby::/Users/aaron/.rbenv/versions/ruby-trunk/bin/ruby -I.ext/common --disable-gem" \
--executables="built-ruby::./miniruby -I./lib -I. -I.ext/common -r./prelude --disable-gem" \
$(find ./benchmark -maxdepth 1 -name '*hash_literal_small*.yml' -o -name '*hash_literal_small*.rb' | sort)
Calculating -------------------------------------
compare-ruby built-ruby
hash_literal_small2 1.567 1.831 i/s - 1.000 times in 0.638056s 0.546039s
hash_literal_small4 1.298 1.652 i/s - 1.000 times in 0.770214s 0.605182s
hash_literal_small8 0.873 1.216 i/s - 1.000 times in 1.145304s 0.822047s
Comparison:
hash_literal_small2
built-ruby: 1.8 i/s
compare-ruby: 1.6 i/s - 1.17x slower
hash_literal_small4
built-ruby: 1.7 i/s
compare-ruby: 1.3 i/s - 1.27x slower
hash_literal_small8
built-ruby: 1.2 i/s
compare-ruby: 0.9 i/s - 1.39x slower
```
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66258 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-12-06 13:28:21 -05:00
|
|
|
def test_compile_insn_duphash
|
|
|
|
assert_compile_once('{ a: 1 }', result_inspect: '{:a=>1}', insns: %i[duphash])
|
2018-10-24 10:24:34 -04:00
|
|
|
end
|
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_newrange
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('a = 1; 0..a', result_inspect: '0..1', insns: %i[newrange])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_pop
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '1', insns: %i[pop])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
a = false
|
|
|
|
b = 1
|
|
|
|
a || b
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_dup
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '3', insns: %i[dup])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
a = 1
|
|
|
|
a&.+(2)
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_dupn
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: 'true', insns: %i[dupn])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
klass = Class.new
|
|
|
|
klass::X ||= true
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_swap_topn
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('{}["true"] = true', result_inspect: 'true', insns: %i[swap topn])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_reput
|
|
|
|
skip "write test"
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_setn
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('[nil][0] = 1', result_inspect: '1', insns: %i[setn])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_adjuststack
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: 'true', insns: %i[adjuststack])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
x = [true]
|
|
|
|
x[0] ||= nil
|
|
|
|
x[0]
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_defined
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('defined?(a)', result_inspect: 'nil', insns: %i[defined])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_checkkeyword
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: 'true', success_count: 1, insns: %i[checkkeyword])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
def test(x: rand)
|
|
|
|
x
|
|
|
|
end
|
|
|
|
print test(x: true)
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_tracecoverage
|
|
|
|
skip "write test"
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_defineclass
|
|
|
|
skip "support this in mjit_compile (low priority)"
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_send
|
2020-07-03 21:02:43 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '1', success_count: 3, insns: %i[send])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
print proc { yield_self { 1 } }.call
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
|
|
|
def test_compile_insn_opt_str_freeze
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '"foo"', insns: %i[opt_str_freeze])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
|
|
|
'foo'.freeze
|
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2019-08-02 10:29:58 -04:00
|
|
|
def test_compile_insn_opt_nil_p
|
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: 'false', insns: %i[opt_nil_p])
|
|
|
|
begin;
|
|
|
|
nil.nil?.nil?
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_str_uminus
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '"bar"', insns: %i[opt_str_uminus])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
|
|
|
-'bar'
|
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_newarray_max
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '2', insns: %i[opt_newarray_max])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
a = 1
|
|
|
|
b = 2
|
2018-02-11 22:33:00 -05:00
|
|
|
[a, b].max
|
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_newarray_min
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '1', insns: %i[opt_newarray_min])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
|
|
|
a = 1
|
|
|
|
b = 2
|
|
|
|
[a, b].min
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
|
|
|
def test_compile_insn_opt_send_without_block
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('print', result_inspect: 'nil', insns: %i[opt_send_without_block])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_invokesuper
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '3', success_count: 4, insns: %i[invokesuper])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
mod = Module.new {
|
|
|
|
def test
|
|
|
|
super + 2
|
|
|
|
end
|
|
|
|
}
|
|
|
|
klass = Class.new {
|
|
|
|
prepend mod
|
|
|
|
def test
|
|
|
|
1
|
|
|
|
end
|
|
|
|
}
|
|
|
|
print klass.new.test
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_invokeblock_leave
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '2', success_count: 2, insns: %i[invokeblock leave])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
def foo
|
|
|
|
yield
|
|
|
|
end
|
|
|
|
print foo { 2 }
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_throw
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '4', success_count: 2, insns: %i[throw])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
def test
|
|
|
|
proc do
|
|
|
|
if 1+1 == 1
|
|
|
|
return 3
|
|
|
|
else
|
|
|
|
return 4
|
|
|
|
end
|
|
|
|
5
|
|
|
|
end.call
|
|
|
|
end
|
|
|
|
print test
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_jump_branchif
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~'end;'}", result_inspect: 'nil', insns: %i[jump branchif])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
a = false
|
2018-05-02 11:12:55 -04:00
|
|
|
1 + 1 while a
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_branchunless
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~'end;'}", result_inspect: '1', insns: %i[branchunless])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
a = true
|
|
|
|
if a
|
|
|
|
1
|
|
|
|
else
|
|
|
|
2
|
|
|
|
end
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_branchnil
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~'end;'}", result_inspect: '3', insns: %i[branchnil])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
a = 2
|
|
|
|
a&.+(1)
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
Optimize dynamic string interpolation for symbol/true/false/nil/0-9
This provides a significant speedup for symbol, true, false,
nil, and 0-9, class/module, and a small speedup in most other cases.
Speedups (using included benchmarks):
:symbol :: 60%
0-9 :: 50%
Class/Module :: 50%
nil/true/false :: 20%
integer :: 10%
[] :: 10%
"" :: 3%
One reason this approach is faster is it reduces the number of
VM instructions for each interpolated value.
Initial idea, approach, and benchmarks from Eric Wong. I applied
the same approach against the master branch, updating it to handle
the significant internal changes since this was first proposed 4
years ago (such as CALL_INFO/CALL_CACHE -> CALL_DATA). I also
expanded it to optimize true/false/nil/0-9/class/module, and added
handling of missing methods, refined methods, and RUBY_DEBUG.
This renames the tostring insn to anytostring, and adds an
objtostring insn that implements the optimization. This requires
making a few functions non-static, and adding some non-static
functions.
This disables 4 YJIT tests. Those tests should be reenabled after
YJIT optimizes the new objtostring insn.
Implements [Feature #13715]
Co-authored-by: Eric Wong <e@80x24.org>
Co-authored-by: Alan Wu <XrXr@users.noreply.github.com>
Co-authored-by: Yusuke Endoh <mame@ruby-lang.org>
Co-authored-by: Koichi Sasada <ko1@atdot.net>
2021-11-18 18:10:20 -05:00
|
|
|
def test_compile_insn_objtostring
|
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~'end;'}", result_inspect: '"42"', insns: %i[objtostring])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
a = '2'
|
|
|
|
"4#{a}"
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_inlinecache
|
2018-11-07 03:13:20 -05:00
|
|
|
assert_compile_once('Struct', result_inspect: 'Struct', insns: %i[opt_getinlinecache opt_setinlinecache])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_once
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('/#{true}/o =~ "true" && $~.to_a', result_inspect: '["true"]', insns: %i[once])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_checkmatch_opt_case_dispatch
|
2021-05-26 11:50:37 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '"world"', insns: %i[opt_case_dispatch])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
case 'hello'
|
2018-05-02 11:12:55 -04:00
|
|
|
when 'hello'
|
2018-02-10 11:09:55 -05:00
|
|
|
'world'
|
|
|
|
end
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_calc
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('4 + 2 - ((2 * 3 / 2) % 2)', result_inspect: '5', insns: %i[opt_plus opt_minus opt_mult opt_div opt_mod])
|
2018-11-12 02:14:43 -05:00
|
|
|
assert_compile_once('4.0 + 2.0 - ((2.0 * 3.0 / 2.0) % 2.0)', result_inspect: '5.0', insns: %i[opt_plus opt_minus opt_mult opt_div opt_mod])
|
2018-02-11 22:33:00 -05:00
|
|
|
assert_compile_once('4 + 2', result_inspect: '6')
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_cmp
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('(1 == 1) && (1 != 2)', result_inspect: 'true', insns: %i[opt_eq opt_neq])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_rel
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('1 < 2 && 1 <= 1 && 2 > 1 && 1 >= 1', result_inspect: 'true', insns: %i[opt_lt opt_le opt_gt opt_ge])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_ltlt
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('[1] << 2', result_inspect: '[1, 2]', insns: %i[opt_ltlt])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-09-25 22:38:45 -04:00
|
|
|
def test_compile_insn_opt_and
|
|
|
|
assert_compile_once('1 & 3', result_inspect: '1', insns: %i[opt_and])
|
|
|
|
end
|
|
|
|
|
|
|
|
def test_compile_insn_opt_or
|
|
|
|
assert_compile_once('1 | 3', result_inspect: '3', insns: %i[opt_or])
|
|
|
|
end
|
|
|
|
|
2018-02-13 10:58:38 -05:00
|
|
|
def test_compile_insn_opt_aref
|
|
|
|
# optimized call (optimized JIT) -> send call
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '21', success_count: 2, min_calls: 1, insns: %i[opt_aref])
|
2018-02-13 10:58:38 -05:00
|
|
|
begin;
|
|
|
|
obj = Object.new
|
|
|
|
def obj.[](h)
|
|
|
|
h
|
|
|
|
end
|
|
|
|
|
|
|
|
block = proc { |h| h[1] }
|
|
|
|
print block.call({ 1 => 2 })
|
|
|
|
print block.call(obj)
|
|
|
|
end;
|
|
|
|
|
|
|
|
# send call -> optimized call (send JIT) -> optimized call
|
2019-04-14 00:52:02 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '122', success_count: 2, min_calls: 2)
|
2018-02-13 10:58:38 -05:00
|
|
|
begin;
|
|
|
|
obj = Object.new
|
|
|
|
def obj.[](h)
|
|
|
|
h
|
|
|
|
end
|
|
|
|
|
|
|
|
block = proc { |h| h[1] }
|
|
|
|
print block.call(obj)
|
|
|
|
print block.call({ 1 => 2 })
|
|
|
|
print block.call({ 1 => 2 })
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2018-05-02 12:04:26 -04:00
|
|
|
def test_compile_insn_opt_aref_with
|
|
|
|
assert_compile_once("{ '1' => 2 }['1']", result_inspect: '2', insns: %i[opt_aref_with])
|
|
|
|
end
|
|
|
|
|
2018-02-13 10:58:38 -05:00
|
|
|
def test_compile_insn_opt_aset
|
2018-05-02 11:41:43 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '5', insns: %i[opt_aset opt_aset_with])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
hash = { '1' => 2 }
|
2018-02-13 10:58:38 -05:00
|
|
|
(hash['2'] = 2) + (hash[1.to_s] = 3)
|
2018-02-11 22:33:00 -05:00
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_length_size
|
2018-05-02 11:41:43 -04:00
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '4', insns: %i[opt_length opt_size])
|
2018-02-11 22:33:00 -05:00
|
|
|
begin;
|
2018-02-10 11:09:55 -05:00
|
|
|
array = [1, 2]
|
2018-02-11 22:33:00 -05:00
|
|
|
array.length + array.size
|
|
|
|
end;
|
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_empty_p
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('[].empty?', result_inspect: 'true', insns: %i[opt_empty_p])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_succ
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('1.succ', result_inspect: '2', insns: %i[opt_succ])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_not
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once('!!true', result_inspect: 'true', insns: %i[opt_not])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2018-02-11 22:33:00 -05:00
|
|
|
def test_compile_insn_opt_regexpmatch2
|
2019-09-02 01:33:29 -04:00
|
|
|
assert_compile_once("/true/ =~ 'true'", result_inspect: '0', insns: %i[opt_regexpmatch2])
|
2018-05-02 11:12:55 -04:00
|
|
|
assert_compile_once("'true' =~ /true/", result_inspect: '0', insns: %i[opt_regexpmatch2])
|
2018-02-11 22:33:00 -05:00
|
|
|
end
|
2018-02-10 11:09:55 -05:00
|
|
|
|
2019-11-10 00:56:38 -05:00
|
|
|
def test_compile_insn_opt_invokebuiltin_delegate_leave
|
2020-06-26 01:21:12 -04:00
|
|
|
iseq = eval(EnvUtil.invoke_ruby(['-e', <<~'EOS'], '', true).first)
|
|
|
|
p RubyVM::InstructionSequence.of("\x00".method(:unpack)).to_a
|
|
|
|
EOS
|
|
|
|
insns = collect_insns(iseq)
|
2019-11-10 00:56:38 -05:00
|
|
|
mark_tested_insn(:opt_invokebuiltin_delegate_leave, used_insns: insns)
|
|
|
|
assert_eval_with_jit('print "\x00".unpack("c")', stdout: '[0]', success_count: 1)
|
|
|
|
end
|
|
|
|
|
2021-10-25 08:30:16 -04:00
|
|
|
def test_compile_insn_checkmatch
|
|
|
|
assert_compile_once("#{<<~"begin;"}\n#{<<~"end;"}", result_inspect: '"world"', insns: %i[checkmatch])
|
|
|
|
begin;
|
|
|
|
ary = %w(hello good-bye)
|
|
|
|
case 'hello'
|
|
|
|
when *ary
|
|
|
|
'world'
|
|
|
|
end
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2018-02-08 07:40:33 -05:00
|
|
|
def test_jit_output
|
|
|
|
out, err = eval_with_jit('5.times { puts "MJIT" }', verbose: 1, min_calls: 5)
|
|
|
|
assert_equal("MJIT\n" * 5, out)
|
|
|
|
assert_match(/^#{JIT_SUCCESS_PREFIX}: block in <main>@-e:1 -> .+_ruby_mjit_p\d+u\d+\.c$/, err)
|
|
|
|
assert_match(/^Successful MJIT finish$/, err)
|
|
|
|
end
|
|
|
|
|
2020-02-29 02:20:44 -05:00
|
|
|
def test_nothing_to_unload_with_jit_wait
|
2020-02-29 02:58:33 -05:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: 'hello', success_count: 11, max_cache: 10, ignorable_patterns: MAX_CACHE_PATTERNS)
|
2020-02-29 02:20:44 -05:00
|
|
|
begin;
|
|
|
|
def a1() a2() end
|
|
|
|
def a2() a3() end
|
|
|
|
def a3() a4() end
|
|
|
|
def a4() a5() end
|
|
|
|
def a5() a6() end
|
|
|
|
def a6() a7() end
|
|
|
|
def a7() a8() end
|
|
|
|
def a8() a9() end
|
|
|
|
def a9() a10() end
|
|
|
|
def a10() a11() end
|
|
|
|
def a11() print('hello') end
|
|
|
|
a1
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2020-02-29 02:58:33 -05:00
|
|
|
def test_unload_units_on_fiber
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: 'hello', success_count: 12, max_cache: 10, ignorable_patterns: MAX_CACHE_PATTERNS)
|
|
|
|
begin;
|
|
|
|
def a1() a2(false); a2(true) end
|
|
|
|
def a2(a) a3(a) end
|
|
|
|
def a3(a) a4(a) end
|
|
|
|
def a4(a) a5(a) end
|
|
|
|
def a5(a) a6(a) end
|
|
|
|
def a6(a) a7(a) end
|
|
|
|
def a7(a) a8(a) end
|
|
|
|
def a8(a) a9(a) end
|
|
|
|
def a9(a) a10(a) end
|
|
|
|
def a10(a)
|
|
|
|
if a
|
|
|
|
Fiber.new { a11 }.resume
|
|
|
|
end
|
|
|
|
end
|
|
|
|
def a11() print('hello') end
|
|
|
|
a1
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2018-11-18 03:25:48 -05:00
|
|
|
def test_unload_units_and_compaction
|
2018-07-28 22:36:35 -04:00
|
|
|
Dir.mktmpdir("jit_test_unload_units_") do |dir|
|
2018-07-27 05:27:58 -04:00
|
|
|
# MIN_CACHE_SIZE is 10
|
|
|
|
out, err = eval_with_jit({"TMPDIR"=>dir}, "#{<<~"begin;"}\n#{<<~'end;'}", verbose: 1, min_calls: 1, max_cache: 10)
|
|
|
|
begin;
|
2018-08-02 11:35:05 -04:00
|
|
|
i = 0
|
|
|
|
while i < 11
|
2018-07-27 05:27:58 -04:00
|
|
|
eval(<<-EOS)
|
|
|
|
def mjit#{i}
|
|
|
|
print #{i}
|
|
|
|
end
|
|
|
|
mjit#{i}
|
|
|
|
EOS
|
2018-08-02 11:35:05 -04:00
|
|
|
i += 1
|
2018-07-27 05:27:58 -04:00
|
|
|
end
|
2018-11-18 03:25:48 -05:00
|
|
|
|
|
|
|
if defined?(fork)
|
|
|
|
# test the child does not try to delete files which are deleted by parent,
|
|
|
|
# and test possible deadlock on fork during MJIT unload and JIT compaction on child
|
|
|
|
Process.waitpid(Process.fork {})
|
|
|
|
end
|
2018-07-27 05:27:58 -04:00
|
|
|
end;
|
2018-09-11 07:09:59 -04:00
|
|
|
|
2019-01-10 09:55:24 -05:00
|
|
|
debug_info = %Q[stdout:\n"""\n#{out}\n"""\n\nstderr:\n"""\n#{err}"""\n]
|
2018-09-11 07:09:59 -04:00
|
|
|
assert_equal('012345678910', out, debug_info)
|
2018-07-28 22:05:08 -04:00
|
|
|
compactions, errs = err.lines.partition do |l|
|
2020-05-01 02:58:50 -04:00
|
|
|
l.match?(/\AJIT compaction \(\d+\.\dms\): Compacted \d+ methods /)
|
mjit.c: introduce JIT compaction [experimental]
When all compilation finishes or the number of JIT-ed code reaches
--jit-max-cache, this compacts all generated code to a single .so file
and re-loads all methods from it.
In the future, it may trigger compaction more frequently and/or limit
the maximum times of compaction to prevent unlimited memory usage.
So the current behavior is experimental, but at least the performance
improvement in this commit won't be removed.
=== Benchmark ===
In this benchmark, I'll compare following four conditions:
* trunk: r64082
* trunk JIT: r64082 w/ --jit
* single-so JIT: This commit w/ --jit
* objfcn JIT: This branch https://github.com/k0kubun/ruby/tree/objfcn w/ --jit,
which is shinh's objfcn https://github.com/shinh/ruby/tree/objfcn rebased from this commit
```
$ uname -a
Linux bionic 4.15.0-29-generic #31-Ubuntu SMP Tue Jul 17 15:39:52 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
```
* Micro benchmark
Using this script https://gist.github.com/k0kubun/10e6d3387c9ab1b134622b2c9d76ef51,
calls some amount of different methods that just return `nil`. The following tables
are its average duration seconds of 3 measurements.
Smaller is better.
** 1 method (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.576067774333296 | 5.915551971666446 | 5.833641665666619 | 5.845915191666639 |
| Ratio | 1.00x | 1.06x | 1.05x | 1.05x |
** 50 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 3.1661167996666677| 6.125825928333342 | 4.135432743666665 | 3.750358728333348 |
| Ratio | 1.00x | 1.93x | 1.31x | 1.18x |
** 1500 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.971650823666664 | 19.579182102999994| 10.511108153999961| 10.854653588999932|
| Ratio | 1.00x | 3.28x | 1.76x | 1.82x |
* Discourse
Using the same benchmark strategy as https://bugs.ruby-lang.org/issues/14490 with
this branch https://github.com/k0kubun/discourse/commits/benchmark2 forked from discourse
v1.8.11 to support running trunk.
1. Run ruby script/bench.rb to warm up profiling database
2. Run RUBYOPT='--jit-verbose=1 --jit-max-cache=10000' RAILS_ENV=profile bin/puma -e production
3. WAIT 5-15 or so minutes for all jitting to stop so we have no cross talk
4. Run ab -n 100 http://localhost:9292/
5. Wait for all new jitting to finish
6. Run ab -n 100 http://localhost:9292/
** Response time (ms)
Here is the response time milliseconds for each percentile.
Skipping 99%ile because it's the same as 100%ile in 100 calls.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 38 | 45 | 41 | 43 |
| 66% | 39 | 50 | 44 | 44 |
| 75% | 47 | 51 | 46 | 45 |
| 80% | 49 | 52 | 47 | 47 |
| 90% | 50 | 63 | 50 | 52 |
| 95% | 60 | 79 | 52 | 55 |
| 98% | 91 | 114 | 91 | 91 |
|100% | 97 | 133 | 96 | 99 |
** Ratio (smaller is better)
Here is the response time increase ratio against no-JIT trunk's one.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 1.00x| 1.18x| 1.08x| 1.13x|
| 66% | 1.00x| 1.28x| 1.13x| 1.13x|
| 75% | 1.00x| 1.09x| 0.98x| 0.96x|
| 80% | 1.00x| 1.06x| 0.96x| 0.96x|
| 90% | 1.00x| 1.26x| 1.00x| 1.04x|
| 95% | 1.00x| 1.32x| 0.87x| 0.92x|
| 98% | 1.00x| 1.25x| 1.00x| 1.00x|
|100% | 1.00x| 1.37x| 0.99x| 1.02x|
While 50 and 60 %ile are still worse than no-JIT trunk, 75, 80, 90, 95,
98 and 100% are not slower than that.
So now it's a little harder to say "MJIT slows down Rails applications".
Probably I can close [Bug #14490] now. Let's start improving it.
Close https://github.com/ruby/ruby/pull/1921
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64094 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-07-28 12:14:56 -04:00
|
|
|
end
|
2018-08-02 11:35:05 -04:00
|
|
|
10.times do |i|
|
2018-09-11 07:09:59 -04:00
|
|
|
assert_match(/\A#{JIT_SUCCESS_PREFIX}: mjit#{i}@\(eval\):/, errs[i], debug_info)
|
2018-07-27 01:52:01 -04:00
|
|
|
end
|
2018-07-27 05:27:58 -04:00
|
|
|
|
2020-11-28 02:13:17 -05:00
|
|
|
assert_equal("No units can be unloaded -- incremented max-cache-size to 11 for --jit-wait\n", errs[10], debug_info)
|
|
|
|
assert_match(/\A#{JIT_SUCCESS_PREFIX}: mjit10@\(eval\):/, errs[11], debug_info)
|
2018-07-28 22:05:08 -04:00
|
|
|
# On --jit-wait, when the number of JIT-ed code reaches --jit-max-cache,
|
|
|
|
# it should trigger compaction.
|
2020-11-28 02:13:17 -05:00
|
|
|
unless RUBY_PLATFORM.match?(/mswin|mingw/) # compaction is not supported on Windows yet
|
2020-11-28 02:46:10 -05:00
|
|
|
assert_equal(1, compactions.size, debug_info)
|
2018-08-02 11:35:05 -04:00
|
|
|
end
|
2018-07-28 22:05:08 -04:00
|
|
|
|
2018-12-29 05:50:31 -05:00
|
|
|
if RUBY_PLATFORM.match?(/mswin/)
|
|
|
|
# "Permission Denied" error is preventing to remove so file on AppVeyor/RubyCI.
|
|
|
|
skip 'Removing so file is randomly failing on AppVeyor/RubyCI mswin due to Permission Denied.'
|
2018-10-07 21:08:37 -04:00
|
|
|
else
|
2020-05-01 02:58:50 -04:00
|
|
|
# verify .c files are deleted on unload_units
|
2020-05-03 19:49:30 -04:00
|
|
|
assert_send([Dir, :empty?, dir], debug_info)
|
2018-10-07 21:08:37 -04:00
|
|
|
end
|
2018-07-27 01:52:01 -04:00
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2020-04-18 04:38:52 -04:00
|
|
|
def test_newarraykwsplat_on_stack
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "[nil, [{:type=>:development}]]\n", success_count: 1, insns: %i[newarraykwsplat])
|
|
|
|
begin;
|
|
|
|
def arr
|
|
|
|
[nil, [:type => :development]]
|
|
|
|
end
|
|
|
|
p arr
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
mjit_compile.c: use local variables for stack
if catch_except_p is FALSE. If catch_except_p is TRUE, stack values
should be on VM's stack when exception is thrown and the JIT-ed frame
is re-executed by VM's exception handler. If it's FALSE, the JIT-ed
frame won't be re-executed and don't need to keep values on VM's stack.
Using local variables allows us to reduce cfp->sp motion. Moving cfp->sp
is needed only for insns whose handles_frame? is false. So it improves
performance.
_mjit_compile_insn.erb: Prepare `stack_size` variable for GET_SP,
STACK_ADDR_FROM_TOP, TOPN macros. Share pc and sp motion partial view.
Use cancel handler created in mjit_compile.c.
_mjit_compile_send.erb: ditto. Also, when iseq->body->catch_except_p is
TRUE, this stops to call mjit_exec directly. I described the reason in
vm_insnhelper.h's comment for EXEC_EC_CFP.
_mjit_compile_pc_and_sp.erb: Shared logic for moving sp and pc. As you
can see from thsi file, when status->local_stack_p is TRUE and
insn.handles_frame? is false, moving sp is skipped. But if
insn.handles_frame? is true, values should be rolled back to VM's stack.
common.mk: add dependency for the file
_mjit_compile_insn_body.erb: Set sp value before canceling JIT on
DISPATCH_ORIGINAL_INSN. Replace GET_SP, STACK_ADDR_FROM_TOP, TOPN macros
for the case ocal_stack_p is TRUE and insn.handles_frame? is false.
In that case, values are not available on VM's stack and those macros
should be replaced.
mjit_compile.inc.erb: updated comments of macros which are supported by
JIT compiler. All references to `cfp->sp` should be replaced and thus
INC_SP, SET_SV, PUSH are no longer supported for now, because they are
not used now.
vm_exec.h: moved EXEC_EC_CFP definition to vm_insnhelper.h because it's
tighly coupled to CALL_METHOD.
vm_insnhelper.h: Have revised EXEC_EC_CFP definition moved from vm_exec.h.
Now it triggers mjit_exec for VM, and has the guard for catch_except_p
on JIT-ed code. See comments for details. CALL_METHOD delegates
triggering mjit_exec to EXEC_EC_CFP.
insns.def: Stopped using EXEC_EC_CFP for the case we don't want to
trigger mjit_exec. Those insns (defineclass, opt_call_c_function) are
not supported by JIT and it's safe to use RESTORE_REGS(), NEXT_INSN().
expandarray is changed to pass GET_SP() to replace the macro in
_mjit_compile_insn_body.erb.
vm_insnhelper.c: change to take sp for the above reason.
[close https://github.com/ruby/ruby/pull/1828]
This patch resurrects the performance which was attached in
[Feature #14235].
* Benchmark
Optcarrot (with configuration for benchmark_driver.gem)
https://github.com/benchmark-driver/optcarrot
$ benchmark-driver benchmark.yml --verbose 1 --rbenv 'before;before+JIT::before,--jit;after;after+JIT::after,--jit' --repeat-count 10
before: ruby 2.6.0dev (2018-03-04 trunk 62652) [x86_64-linux]
before+JIT: ruby 2.6.0dev (2018-03-04 trunk 62652) +JIT [x86_64-linux]
after: ruby 2.6.0dev (2018-03-04 local-variable.. 62652) [x86_64-linux]
last_commit=mjit_compile.c: use local variables for stack
after+JIT: ruby 2.6.0dev (2018-03-04 local-variable.. 62652) +JIT [x86_64-linux]
last_commit=mjit_compile.c: use local variables for stack
Calculating -------------------------------------
before before+JIT after after+JIT
optcarrot 53.552 59.680 53.697 63.358 fps
Comparison:
optcarrot
after+JIT: 63.4 fps
before+JIT: 59.7 fps - 1.06x slower
after: 53.7 fps - 1.18x slower
before: 53.6 fps - 1.18x slower
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62655 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-04 02:04:40 -05:00
|
|
|
def test_local_stack_on_exception
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '3', success_count: 2)
|
|
|
|
begin;
|
|
|
|
def b
|
|
|
|
raise
|
|
|
|
rescue
|
|
|
|
2
|
|
|
|
end
|
|
|
|
|
|
|
|
def a
|
|
|
|
# Calling #b should be vm_exec, not direct mjit_exec.
|
|
|
|
# Otherwise `1` on local variable would be purged.
|
|
|
|
1 + b
|
|
|
|
end
|
|
|
|
|
|
|
|
print a
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
|
|
|
def test_local_stack_with_sp_motion_by_blockargs
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '1', success_count: 2)
|
|
|
|
begin;
|
|
|
|
def b(base)
|
|
|
|
1
|
|
|
|
end
|
|
|
|
|
|
|
|
# This method is simple enough to have false in catch_except_p.
|
|
|
|
# So local_stack_p would be true in JIT compiler.
|
|
|
|
def a
|
|
|
|
m = method(:b)
|
|
|
|
|
|
|
|
# ci->flag has VM_CALL_ARGS_BLOCKARG and cfp->sp is moved in vm_caller_setup_arg_block.
|
|
|
|
# So, for this send insn, JIT-ed code should use cfp->sp instead of local variables for stack.
|
|
|
|
Module.module_eval(&m)
|
|
|
|
end
|
|
|
|
|
|
|
|
print a
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
compile.c: mark all ISeq ancestors as catch_except_p
This change assumes that continuously reading `parent_iseq` from block
ISeq would reach non-block ISeq finally.
test/ruby/test_jit.rb: add test that catches 2-depth exception
Combination of r62654 and r62678 caused following error in this test.
-e:12:in `wrapper': Stack consistency error (sp: 14, bp: 13) (fatal)
== disasm: #<ISeq:wrapper@-e:10 (10,0)-(12,3)> (catch: FALSE)===========
local table (size: 2, argc: 2 [opts: 0, rest: -1, post: 0, block: -1, kw: -1@-1, kwrest: -1])
[ 2] paths<Arg> [ 1] prefixes<Arg>
0000 putself ( 11)[LiCa]
0001 getlocal_WC_0 paths
0003 getlocal_WC_0 prefixes
0005 opt_send_without_block <callinfo!mid:catch_true, argc:2, FCALL|ARGS_SIMPLE>, <callcache>
0008 leave ( 12)[Re]
As you can see, it says `catch: FALSE`, but obviously it catches
exception raised from `return path`.
As of r62655, it was kind of intentional because I only cared about
expiration of JIT-ed frame and I've thought calling `vm_exec` is only
needed once for it. So r62654 was NOT actually checking if it may catch
exception.
But for r62678, obviously we should set catch_except_p=TRUE for all
ISeqs which may catch exception. Otherwise catch table lookup would
fail.
With this bugfix, code generated by r62655 might be worse, but at least
while loop can be marked as `catch: FALSE` as expected.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@62717 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-03-10 09:52:12 -05:00
|
|
|
def test_catching_deep_exception
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '1', success_count: 4)
|
|
|
|
begin;
|
|
|
|
def catch_true(paths, prefixes) # catch_except_p: TRUE
|
|
|
|
prefixes.each do |prefix| # catch_except_p: TRUE
|
|
|
|
paths.each do |path| # catch_except_p: FALSE
|
|
|
|
return path
|
|
|
|
end
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
def wrapper(paths, prefixes)
|
|
|
|
catch_true(paths, prefixes)
|
|
|
|
end
|
|
|
|
|
|
|
|
print wrapper(['1'], ['2'])
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2021-01-18 17:52:07 -05:00
|
|
|
def test_inlined_builtin_methods
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '', success_count: 1, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
def test
|
|
|
|
float = 0.0
|
|
|
|
float.abs
|
2021-01-18 23:44:47 -05:00
|
|
|
float.-@
|
2021-01-18 17:52:07 -05:00
|
|
|
float.zero?
|
|
|
|
end
|
|
|
|
test
|
|
|
|
test
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2020-07-10 03:42:43 -04:00
|
|
|
def test_inlined_c_method
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "aaa", success_count: 2, recompile_count: 1, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
def test(obj, recursive: nil)
|
|
|
|
if recursive
|
|
|
|
test(recursive)
|
|
|
|
end
|
|
|
|
obj.to_s
|
|
|
|
end
|
|
|
|
|
|
|
|
print(test('a')) # set #to_s cc to String#to_s (expecting C method)
|
|
|
|
print(test('a')) # JIT with #to_s cc: String#to_s
|
2020-07-10 03:47:26 -04:00
|
|
|
# update #to_s cd->cc to Symbol#to_s, then go through the Symbol#to_s cd->cc
|
|
|
|
# after checking receiver class using inlined #to_s cc with String#to_s.
|
|
|
|
print(test('a', recursive: :foo))
|
2020-07-10 03:42:43 -04:00
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2020-03-31 01:27:01 -04:00
|
|
|
def test_inlined_exivar
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "aaa", success_count: 3, recompile_count: 1, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
class Foo < Hash
|
|
|
|
def initialize
|
|
|
|
@a = :a
|
|
|
|
end
|
|
|
|
|
|
|
|
def bar
|
|
|
|
@a
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
print(Foo.new.bar)
|
|
|
|
print(Foo.new.bar) # compile #initialize, #bar -> recompile #bar
|
|
|
|
print(Foo.new.bar) # compile #bar with exivar
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2018-10-24 11:41:14 -04:00
|
|
|
def test_inlined_undefined_ivar
|
2019-04-14 00:52:02 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "bbb", success_count: 3, min_calls: 3)
|
2018-10-24 11:41:14 -04:00
|
|
|
begin;
|
|
|
|
class Foo
|
|
|
|
def initialize
|
|
|
|
@a = :a
|
|
|
|
end
|
|
|
|
|
|
|
|
def bar
|
|
|
|
if @b.nil?
|
|
|
|
@b = :b
|
|
|
|
end
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
print(Foo.new.bar)
|
|
|
|
print(Foo.new.bar)
|
|
|
|
print(Foo.new.bar)
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2019-09-03 21:53:20 -04:00
|
|
|
def test_inlined_setivar_frozen
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "FrozenError\n", success_count: 2, min_calls: 3)
|
|
|
|
begin;
|
|
|
|
class A
|
|
|
|
def a
|
|
|
|
@a = 1
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
a = A.new
|
|
|
|
a.a
|
|
|
|
a.a
|
|
|
|
a.a
|
|
|
|
a.freeze
|
|
|
|
begin
|
|
|
|
a.a
|
|
|
|
rescue FrozenError => e
|
|
|
|
p e.class
|
|
|
|
end
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2021-01-04 16:09:01 -05:00
|
|
|
def test_inlined_getconstant
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '11', success_count: 1, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
FOO = 1
|
|
|
|
def const
|
|
|
|
FOO
|
|
|
|
end
|
|
|
|
print const
|
|
|
|
print const
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
_mjit_compile_send.erb: inline attr_reader call
_mjit_compile_send_guard.erb: carve out the shared logic to invalidate
inlined method call
common.mk: update dependency for this change
test_jit.rb: add test for attr_reader optimization
* Benchmark
```
require 'benchmark_driver'
Benchmark.driver do |x|
x.prelude %{
class C
attr_reader :a
def initialize
@a = 1
end
end
o = C.new
def l o
i = 0
while i < 1000000
o.a
i += 1
end
end
}
x.report 'aread', %{ l o }
x.loop_count 1000
x.rbenv 'before', 'before,--jit', 'after,--jit'
x.verbose
end
```
```
before: ruby 2.6.0dev (2018-04-20 trunk 63211) [x86_64-linux]
before,--jit: ruby 2.6.0dev (2018-04-20 trunk 63211) +JIT [x86_64-linux]
after,--jit: ruby 2.6.0dev (2018-04-20 trunk 63211) +JIT [x86_64-linux]
last_commit=_mjit_compile_send.erb: inline attr_reader call
Calculating -------------------------------------
before before,--jit after,--jit
aread 54.597 122.894 218.574 i/s - 1.000k times in 18.316102s 8.137089s 4.575106s
Comparison:
aread
after,--jit: 218.6 i/s
before,--jit: 122.9 i/s - 1.78x slower
before: 54.6 i/s - 4.00x slower
```
* Optcarrot
A little made faster?
fps: 71.35 -> 72.11
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63212 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-04-20 09:42:08 -04:00
|
|
|
def test_attr_reader
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "4nil\nnil\n6", success_count: 2, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
class A
|
|
|
|
attr_reader :a, :b
|
|
|
|
|
|
|
|
def initialize
|
|
|
|
@a = 2
|
|
|
|
end
|
|
|
|
|
|
|
|
def test
|
|
|
|
a
|
|
|
|
end
|
|
|
|
|
|
|
|
def undefined
|
|
|
|
b
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
a = A.new
|
|
|
|
print(a.test * a.test)
|
|
|
|
p(a.undefined)
|
|
|
|
p(a.undefined)
|
|
|
|
|
|
|
|
# redefinition
|
2019-04-05 22:42:30 -04:00
|
|
|
def a.test
|
|
|
|
3
|
_mjit_compile_send.erb: inline attr_reader call
_mjit_compile_send_guard.erb: carve out the shared logic to invalidate
inlined method call
common.mk: update dependency for this change
test_jit.rb: add test for attr_reader optimization
* Benchmark
```
require 'benchmark_driver'
Benchmark.driver do |x|
x.prelude %{
class C
attr_reader :a
def initialize
@a = 1
end
end
o = C.new
def l o
i = 0
while i < 1000000
o.a
i += 1
end
end
}
x.report 'aread', %{ l o }
x.loop_count 1000
x.rbenv 'before', 'before,--jit', 'after,--jit'
x.verbose
end
```
```
before: ruby 2.6.0dev (2018-04-20 trunk 63211) [x86_64-linux]
before,--jit: ruby 2.6.0dev (2018-04-20 trunk 63211) +JIT [x86_64-linux]
after,--jit: ruby 2.6.0dev (2018-04-20 trunk 63211) +JIT [x86_64-linux]
last_commit=_mjit_compile_send.erb: inline attr_reader call
Calculating -------------------------------------
before before,--jit after,--jit
aread 54.597 122.894 218.574 i/s - 1.000k times in 18.316102s 8.137089s 4.575106s
Comparison:
aread
after,--jit: 218.6 i/s
before,--jit: 122.9 i/s - 1.78x slower
before: 54.6 i/s - 4.00x slower
```
* Optcarrot
A little made faster?
fps: 71.35 -> 72.11
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63212 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-04-20 09:42:08 -04:00
|
|
|
end
|
|
|
|
|
|
|
|
print(2 * a.test)
|
|
|
|
end;
|
2018-05-03 04:47:32 -04:00
|
|
|
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "true", success_count: 1, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
class Hoge
|
|
|
|
attr_reader :foo
|
|
|
|
|
|
|
|
def initialize
|
|
|
|
@foo = []
|
|
|
|
@bar = nil
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
class Fuga < Hoge
|
|
|
|
def initialize
|
|
|
|
@bar = nil
|
|
|
|
@foo = []
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
def test(recv)
|
|
|
|
recv.foo.empty?
|
|
|
|
end
|
|
|
|
|
|
|
|
hoge = Hoge.new
|
|
|
|
fuga = Fuga.new
|
|
|
|
|
|
|
|
test(hoge) # VM: cc set index=1
|
|
|
|
test(hoge) # JIT: compile with index=1
|
|
|
|
test(fuga) # JIT -> VM: cc set index=2
|
|
|
|
print test(hoge) # JIT: should use index=1, not index=2 in cc
|
|
|
|
end;
|
_mjit_compile_send.erb: inline attr_reader call
_mjit_compile_send_guard.erb: carve out the shared logic to invalidate
inlined method call
common.mk: update dependency for this change
test_jit.rb: add test for attr_reader optimization
* Benchmark
```
require 'benchmark_driver'
Benchmark.driver do |x|
x.prelude %{
class C
attr_reader :a
def initialize
@a = 1
end
end
o = C.new
def l o
i = 0
while i < 1000000
o.a
i += 1
end
end
}
x.report 'aread', %{ l o }
x.loop_count 1000
x.rbenv 'before', 'before,--jit', 'after,--jit'
x.verbose
end
```
```
before: ruby 2.6.0dev (2018-04-20 trunk 63211) [x86_64-linux]
before,--jit: ruby 2.6.0dev (2018-04-20 trunk 63211) +JIT [x86_64-linux]
after,--jit: ruby 2.6.0dev (2018-04-20 trunk 63211) +JIT [x86_64-linux]
last_commit=_mjit_compile_send.erb: inline attr_reader call
Calculating -------------------------------------
before before,--jit after,--jit
aread 54.597 122.894 218.574 i/s - 1.000k times in 18.316102s 8.137089s 4.575106s
Comparison:
aread
after,--jit: 218.6 i/s
before,--jit: 122.9 i/s - 1.78x slower
before: 54.6 i/s - 4.00x slower
```
* Optcarrot
A little made faster?
fps: 71.35 -> 72.11
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@63212 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-04-20 09:42:08 -04:00
|
|
|
end
|
|
|
|
|
2020-07-04 18:37:20 -04:00
|
|
|
def test_heap_promotion_of_ivar_in_the_middle_of_jit
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "true\ntrue\n", success_count: 2, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
class A
|
|
|
|
def initialize
|
|
|
|
@iv0 = nil
|
|
|
|
@iv1 = []
|
|
|
|
@iv2 = nil
|
|
|
|
end
|
|
|
|
|
|
|
|
def test(add)
|
|
|
|
@iv0.nil?
|
|
|
|
@iv2.nil?
|
|
|
|
add_ivar if add
|
|
|
|
@iv1.empty?
|
|
|
|
end
|
|
|
|
|
|
|
|
def add_ivar
|
|
|
|
@iv3 = nil
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
a = A.new
|
|
|
|
p a.test(false)
|
|
|
|
p a.test(true)
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2020-02-19 02:16:19 -05:00
|
|
|
def test_jump_to_precompiled_branch
|
|
|
|
assert_eval_with_jit("#{<<~'begin;'}\n#{<<~'end;'}", stdout: ".0", success_count: 1, min_calls: 1)
|
|
|
|
begin;
|
|
|
|
def test(foo)
|
|
|
|
".#{foo unless foo == 1}" if true
|
|
|
|
end
|
|
|
|
print test(0)
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2018-04-27 03:39:00 -04:00
|
|
|
def test_clean_so
|
2018-12-28 04:34:51 -05:00
|
|
|
if RUBY_PLATFORM.match?(/mswin/)
|
|
|
|
skip 'Removing so file is randomly failing on AppVeyor/RubyCI mswin due to Permission Denied.'
|
2018-10-07 21:08:37 -04:00
|
|
|
end
|
2018-04-27 03:39:00 -04:00
|
|
|
Dir.mktmpdir("jit_test_clean_so_") do |dir|
|
|
|
|
code = "x = 0; 10.times {|i|x+=i}"
|
|
|
|
eval_with_jit({"TMPDIR"=>dir}, code)
|
2020-05-03 19:49:30 -04:00
|
|
|
assert_send([Dir, :empty?, dir])
|
2018-04-27 03:39:00 -04:00
|
|
|
eval_with_jit({"TMPDIR"=>dir}, code, save_temps: true)
|
|
|
|
assert_not_send([Dir, :empty?, dir])
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
2018-11-26 08:39:53 -05:00
|
|
|
def test_clean_objects_on_exec
|
2018-11-26 11:13:56 -05:00
|
|
|
if /mswin|mingw/ =~ RUBY_PLATFORM
|
|
|
|
# TODO: check call stack and close handle of code which is not on stack, and remove objects on best-effort basis
|
|
|
|
skip 'Removing so file being used does not work on Windows'
|
|
|
|
end
|
2018-11-26 08:39:53 -05:00
|
|
|
Dir.mktmpdir("jit_test_clean_objects_on_exec_") do |dir|
|
|
|
|
eval_with_jit({"TMPDIR"=>dir}, "#{<<~"begin;"}\n#{<<~"end;"}", min_calls: 1)
|
|
|
|
begin;
|
|
|
|
def a; end; a
|
|
|
|
exec "true"
|
|
|
|
end;
|
|
|
|
error_message = "Undeleted files:\n #{Dir.glob("#{dir}/*").join("\n ")}\n"
|
2020-05-03 19:49:30 -04:00
|
|
|
assert_send([Dir, :empty?, dir], error_message)
|
2018-11-26 08:39:53 -05:00
|
|
|
end
|
2018-11-26 11:13:56 -05:00
|
|
|
end
|
2018-11-26 08:39:53 -05:00
|
|
|
|
2018-05-02 08:55:57 -04:00
|
|
|
def test_lambda_longjmp
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '5', success_count: 1)
|
|
|
|
begin;
|
|
|
|
fib = lambda do |x|
|
|
|
|
return x if x == 0 || x == 1
|
|
|
|
fib.call(x-1) + fib.call(x-2)
|
|
|
|
end
|
|
|
|
print fib.call(5)
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2018-07-05 11:56:48 -04:00
|
|
|
def test_stack_pointer_with_assignment
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "nil\nnil\n", success_count: 1)
|
|
|
|
begin;
|
|
|
|
2.times do
|
2018-07-24 07:54:15 -04:00
|
|
|
a, _ = nil
|
2018-07-05 11:56:48 -04:00
|
|
|
p a
|
|
|
|
end
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2019-08-12 07:40:13 -04:00
|
|
|
def test_frame_omitted_inlining
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "true\ntrue\ntrue\n", success_count: 1, min_calls: 2)
|
|
|
|
begin;
|
2020-06-20 17:55:09 -04:00
|
|
|
class Integer
|
2019-08-12 07:40:13 -04:00
|
|
|
remove_method :zero?
|
|
|
|
def zero?
|
|
|
|
self == 0
|
|
|
|
end
|
|
|
|
end
|
|
|
|
|
|
|
|
3.times do
|
|
|
|
p 0.zero?
|
|
|
|
end
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2019-07-02 21:00:50 -04:00
|
|
|
def test_block_handler_with_possible_frame_omitted_inlining
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "70.0\n70.0\n70.0\n", success_count: 2, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
def multiply(a, b)
|
|
|
|
a *= b
|
|
|
|
end
|
|
|
|
|
|
|
|
3.times do
|
|
|
|
p multiply(7.0, 10.0)
|
|
|
|
end
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2020-06-25 01:58:22 -04:00
|
|
|
def test_builtin_frame_omitted_inlining
|
|
|
|
assert_eval_with_jit('0.zero?; 0.zero?; 3.times { p 0.zero? }', stdout: "true\ntrue\ntrue\n", success_count: 1, min_calls: 2)
|
|
|
|
end
|
|
|
|
|
2018-11-18 03:22:30 -05:00
|
|
|
def test_program_counter_with_regexpmatch
|
2018-07-07 10:28:34 -04:00
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "aa", success_count: 1)
|
2018-07-05 11:56:48 -04:00
|
|
|
begin;
|
2018-07-07 10:28:34 -04:00
|
|
|
2.times do
|
|
|
|
break if /a/ =~ "ab" && !$~[0]
|
|
|
|
print $~[0]
|
|
|
|
end
|
2018-07-05 11:56:48 -04:00
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2018-07-07 10:41:03 -04:00
|
|
|
def test_pushed_values_with_opt_aset_with
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "{}{}", success_count: 1)
|
|
|
|
begin;
|
|
|
|
2.times do
|
|
|
|
print(Thread.current["a"] = {})
|
|
|
|
end
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
|
|
|
def test_pushed_values_with_opt_aref_with
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: "nil\nnil\n", success_count: 1)
|
|
|
|
begin;
|
|
|
|
2.times do
|
|
|
|
p(Thread.current["a"])
|
|
|
|
end
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2021-01-03 03:05:00 -05:00
|
|
|
def test_mjit_pause_wait
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", stdout: '', success_count: 0, min_calls: 1)
|
|
|
|
begin;
|
2021-01-14 01:36:45 -05:00
|
|
|
RubyVM::JIT.pause
|
2021-01-03 03:05:00 -05:00
|
|
|
proc {}.call
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2021-08-13 02:19:15 -04:00
|
|
|
def test_not_cancel_by_tracepoint_class
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", success_count: 1, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
TracePoint.new(:class) {}.enable
|
|
|
|
2.times {}
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2021-08-13 01:54:40 -04:00
|
|
|
def test_cancel_by_tracepoint
|
|
|
|
assert_eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", success_count: 0, min_calls: 2)
|
|
|
|
begin;
|
|
|
|
TracePoint.new(:line) {}.enable
|
|
|
|
2.times {}
|
|
|
|
end;
|
|
|
|
end
|
|
|
|
|
2018-08-11 01:33:15 -04:00
|
|
|
def test_caller_locations_without_catch_table
|
|
|
|
out, _ = eval_with_jit("#{<<~"begin;"}\n#{<<~"end;"}", min_calls: 1)
|
|
|
|
begin;
|
|
|
|
def b # 2
|
|
|
|
caller_locations.first # 3
|
|
|
|
end # 4
|
|
|
|
# 5
|
|
|
|
def a # 6
|
|
|
|
print # <-- don't leave PC here # 7
|
|
|
|
b # 8
|
|
|
|
end
|
|
|
|
puts a
|
|
|
|
puts a
|
|
|
|
end;
|
|
|
|
lines = out.lines
|
|
|
|
assert_equal("-e:8:in `a'\n", lines[0])
|
|
|
|
assert_equal("-e:8:in `a'\n", lines[1])
|
|
|
|
end
|
|
|
|
|
2018-11-18 03:25:48 -05:00
|
|
|
def test_fork_with_mjit_worker_thread
|
|
|
|
Dir.mktmpdir("jit_test_fork_with_mjit_worker_thread_") do |dir|
|
|
|
|
# min_calls: 2 to skip fork block
|
|
|
|
out, err = eval_with_jit({ "TMPDIR" => dir }, "#{<<~"begin;"}\n#{<<~"end;"}", min_calls: 2, verbose: 1)
|
|
|
|
begin;
|
|
|
|
def before_fork; end
|
|
|
|
def after_fork; end
|
|
|
|
|
|
|
|
before_fork; before_fork # the child should not delete this .o file
|
|
|
|
pid = Process.fork do # this child should not delete shared .pch file
|
2019-11-09 16:22:51 -05:00
|
|
|
sleep 2.0 # to prevent mixing outputs on Solaris
|
2018-11-18 03:25:48 -05:00
|
|
|
after_fork; after_fork # this child does not share JIT-ed after_fork with parent
|
|
|
|
end
|
|
|
|
after_fork; after_fork # this parent does not share JIT-ed after_fork with child
|
|
|
|
|
|
|
|
Process.waitpid(pid)
|
|
|
|
end;
|
|
|
|
success_count = err.scan(/^#{JIT_SUCCESS_PREFIX}:/).size
|
2018-11-18 03:44:16 -05:00
|
|
|
debug_info = "stdout:\n```\n#{out}\n```\n\nstderr:\n```\n#{err}```\n"
|
2018-12-21 09:01:49 -05:00
|
|
|
assert_equal(3, success_count, debug_info)
|
2018-11-18 03:44:16 -05:00
|
|
|
|
|
|
|
# assert no remove error
|
2018-11-18 04:22:48 -05:00
|
|
|
assert_equal("Successful MJIT finish\n" * 2, err.gsub(/^#{JIT_SUCCESS_PREFIX}:[^\n]+\n/, ''), debug_info)
|
2018-11-18 03:25:48 -05:00
|
|
|
|
|
|
|
# ensure objects are deleted
|
2020-05-03 19:49:30 -04:00
|
|
|
assert_send([Dir, :empty?, dir], debug_info)
|
2018-11-18 03:25:48 -05:00
|
|
|
end
|
|
|
|
end if defined?(fork)
|
|
|
|
|
2018-02-08 07:40:33 -05:00
|
|
|
private
|
|
|
|
|
2018-02-10 11:09:55 -05:00
|
|
|
# The shortest way to test one proc
|
2019-08-02 04:51:45 -04:00
|
|
|
def assert_compile_once(script, result_inspect:, insns: [], uplevel: 1)
|
2018-02-11 23:35:16 -05:00
|
|
|
if script.match?(/\A\n.+\n\z/m)
|
|
|
|
script = script.gsub(/^/, ' ')
|
|
|
|
else
|
|
|
|
script = " #{script} "
|
|
|
|
end
|
2019-08-02 04:51:45 -04:00
|
|
|
assert_eval_with_jit("p proc {#{script}}.call", stdout: "#{result_inspect}\n", success_count: 1, insns: insns, uplevel: uplevel + 1)
|
2018-02-10 11:09:55 -05:00
|
|
|
end
|
|
|
|
|
2018-02-08 07:40:33 -05:00
|
|
|
# Shorthand for normal test cases
|
2020-03-31 01:27:01 -04:00
|
|
|
def assert_eval_with_jit(script, stdout: nil, success_count:, recompile_count: nil, min_calls: 1, max_cache: 1000, insns: [], uplevel: 1, ignorable_patterns: [])
|
2020-02-29 02:20:44 -05:00
|
|
|
out, err = eval_with_jit(script, verbose: 1, min_calls: min_calls, max_cache: max_cache)
|
2020-03-31 01:27:01 -04:00
|
|
|
success_actual = err.scan(/^#{JIT_SUCCESS_PREFIX}:/).size
|
|
|
|
recompile_actual = err.scan(/^#{JIT_RECOMPILE_PREFIX}:/).size
|
2021-12-13 19:08:01 -05:00
|
|
|
# Add --mjit-verbose=2 logs for cl.exe because compiler's error message is suppressed
|
|
|
|
# for cl.exe with --mjit-verbose=1. See `start_process` in mjit_worker.c.
|
2020-03-31 01:27:01 -04:00
|
|
|
if RUBY_PLATFORM.match?(/mswin/) && success_count != success_actual
|
2020-02-29 02:20:44 -05:00
|
|
|
out2, err2 = eval_with_jit(script, verbose: 2, min_calls: min_calls, max_cache: max_cache)
|
2019-03-21 00:47:59 -04:00
|
|
|
end
|
2018-03-11 05:32:48 -04:00
|
|
|
|
2018-05-02 11:12:55 -04:00
|
|
|
# Make sure that the script has insns expected to be tested
|
|
|
|
used_insns = method_insns(script)
|
|
|
|
insns.each do |insn|
|
2019-11-10 17:12:13 -05:00
|
|
|
mark_tested_insn(insn, used_insns: used_insns, uplevel: uplevel + 3)
|
2018-05-02 11:12:55 -04:00
|
|
|
end
|
|
|
|
|
2020-03-31 01:27:01 -04:00
|
|
|
suffix = "script:\n#{code_block(script)}\nstderr:\n#{code_block(err)}#{(
|
|
|
|
"\nstdout(verbose=2 retry):\n#{code_block(out2)}\nstderr(verbose=2 retry):\n#{code_block(err2)}" if out2 || err2
|
|
|
|
)}"
|
2018-02-11 22:33:00 -05:00
|
|
|
assert_equal(
|
2020-03-31 01:27:01 -04:00
|
|
|
success_count, success_actual,
|
|
|
|
"Expected #{success_count} times of JIT success, but succeeded #{success_actual} times.\n\n#{suffix}",
|
2018-02-11 22:33:00 -05:00
|
|
|
)
|
2020-03-31 01:27:01 -04:00
|
|
|
if recompile_count
|
|
|
|
assert_equal(
|
|
|
|
recompile_count, recompile_actual,
|
|
|
|
"Expected #{success_count} times of JIT recompile, but recompiled #{success_actual} times.\n\n#{suffix}",
|
|
|
|
)
|
|
|
|
end
|
2018-02-08 07:40:33 -05:00
|
|
|
if stdout
|
2018-02-10 11:09:55 -05:00
|
|
|
assert_equal(stdout, out, "Expected stdout #{out.inspect} to match #{stdout.inspect} with script:\n#{code_block(script)}")
|
2018-02-08 07:40:33 -05:00
|
|
|
end
|
mjit.c: introduce JIT compaction [experimental]
When all compilation finishes or the number of JIT-ed code reaches
--jit-max-cache, this compacts all generated code to a single .so file
and re-loads all methods from it.
In the future, it may trigger compaction more frequently and/or limit
the maximum times of compaction to prevent unlimited memory usage.
So the current behavior is experimental, but at least the performance
improvement in this commit won't be removed.
=== Benchmark ===
In this benchmark, I'll compare following four conditions:
* trunk: r64082
* trunk JIT: r64082 w/ --jit
* single-so JIT: This commit w/ --jit
* objfcn JIT: This branch https://github.com/k0kubun/ruby/tree/objfcn w/ --jit,
which is shinh's objfcn https://github.com/shinh/ruby/tree/objfcn rebased from this commit
```
$ uname -a
Linux bionic 4.15.0-29-generic #31-Ubuntu SMP Tue Jul 17 15:39:52 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
```
* Micro benchmark
Using this script https://gist.github.com/k0kubun/10e6d3387c9ab1b134622b2c9d76ef51,
calls some amount of different methods that just return `nil`. The following tables
are its average duration seconds of 3 measurements.
Smaller is better.
** 1 method (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.576067774333296 | 5.915551971666446 | 5.833641665666619 | 5.845915191666639 |
| Ratio | 1.00x | 1.06x | 1.05x | 1.05x |
** 50 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 3.1661167996666677| 6.125825928333342 | 4.135432743666665 | 3.750358728333348 |
| Ratio | 1.00x | 1.93x | 1.31x | 1.18x |
** 1500 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.971650823666664 | 19.579182102999994| 10.511108153999961| 10.854653588999932|
| Ratio | 1.00x | 3.28x | 1.76x | 1.82x |
* Discourse
Using the same benchmark strategy as https://bugs.ruby-lang.org/issues/14490 with
this branch https://github.com/k0kubun/discourse/commits/benchmark2 forked from discourse
v1.8.11 to support running trunk.
1. Run ruby script/bench.rb to warm up profiling database
2. Run RUBYOPT='--jit-verbose=1 --jit-max-cache=10000' RAILS_ENV=profile bin/puma -e production
3. WAIT 5-15 or so minutes for all jitting to stop so we have no cross talk
4. Run ab -n 100 http://localhost:9292/
5. Wait for all new jitting to finish
6. Run ab -n 100 http://localhost:9292/
** Response time (ms)
Here is the response time milliseconds for each percentile.
Skipping 99%ile because it's the same as 100%ile in 100 calls.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 38 | 45 | 41 | 43 |
| 66% | 39 | 50 | 44 | 44 |
| 75% | 47 | 51 | 46 | 45 |
| 80% | 49 | 52 | 47 | 47 |
| 90% | 50 | 63 | 50 | 52 |
| 95% | 60 | 79 | 52 | 55 |
| 98% | 91 | 114 | 91 | 91 |
|100% | 97 | 133 | 96 | 99 |
** Ratio (smaller is better)
Here is the response time increase ratio against no-JIT trunk's one.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 1.00x| 1.18x| 1.08x| 1.13x|
| 66% | 1.00x| 1.28x| 1.13x| 1.13x|
| 75% | 1.00x| 1.09x| 0.98x| 0.96x|
| 80% | 1.00x| 1.06x| 0.96x| 0.96x|
| 90% | 1.00x| 1.26x| 1.00x| 1.04x|
| 95% | 1.00x| 1.32x| 0.87x| 0.92x|
| 98% | 1.00x| 1.25x| 1.00x| 1.00x|
|100% | 1.00x| 1.37x| 0.99x| 1.02x|
While 50 and 60 %ile are still worse than no-JIT trunk, 75, 80, 90, 95,
98 and 100% are not slower than that.
So now it's a little harder to say "MJIT slows down Rails applications".
Probably I can close [Bug #14490] now. Let's start improving it.
Close https://github.com/ruby/ruby/pull/1921
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64094 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-07-28 12:14:56 -04:00
|
|
|
err_lines = err.lines.reject! do |l|
|
2020-02-29 02:20:44 -05:00
|
|
|
l.chomp.empty? || l.match?(/\A#{JIT_SUCCESS_PREFIX}/) || (IGNORABLE_PATTERNS + ignorable_patterns).any? { |pat| pat.match?(l) }
|
mjit.c: introduce JIT compaction [experimental]
When all compilation finishes or the number of JIT-ed code reaches
--jit-max-cache, this compacts all generated code to a single .so file
and re-loads all methods from it.
In the future, it may trigger compaction more frequently and/or limit
the maximum times of compaction to prevent unlimited memory usage.
So the current behavior is experimental, but at least the performance
improvement in this commit won't be removed.
=== Benchmark ===
In this benchmark, I'll compare following four conditions:
* trunk: r64082
* trunk JIT: r64082 w/ --jit
* single-so JIT: This commit w/ --jit
* objfcn JIT: This branch https://github.com/k0kubun/ruby/tree/objfcn w/ --jit,
which is shinh's objfcn https://github.com/shinh/ruby/tree/objfcn rebased from this commit
```
$ uname -a
Linux bionic 4.15.0-29-generic #31-Ubuntu SMP Tue Jul 17 15:39:52 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
```
* Micro benchmark
Using this script https://gist.github.com/k0kubun/10e6d3387c9ab1b134622b2c9d76ef51,
calls some amount of different methods that just return `nil`. The following tables
are its average duration seconds of 3 measurements.
Smaller is better.
** 1 method (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.576067774333296 | 5.915551971666446 | 5.833641665666619 | 5.845915191666639 |
| Ratio | 1.00x | 1.06x | 1.05x | 1.05x |
** 50 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 3.1661167996666677| 6.125825928333342 | 4.135432743666665 | 3.750358728333348 |
| Ratio | 1.00x | 1.93x | 1.31x | 1.18x |
** 1500 methods (seconds)
| | trunk | trunk JIT | single-so JIT | objfcn JIT |
|:------|:------------------|:------------------|:------------------|:------------------|
| Time | 5.971650823666664 | 19.579182102999994| 10.511108153999961| 10.854653588999932|
| Ratio | 1.00x | 3.28x | 1.76x | 1.82x |
* Discourse
Using the same benchmark strategy as https://bugs.ruby-lang.org/issues/14490 with
this branch https://github.com/k0kubun/discourse/commits/benchmark2 forked from discourse
v1.8.11 to support running trunk.
1. Run ruby script/bench.rb to warm up profiling database
2. Run RUBYOPT='--jit-verbose=1 --jit-max-cache=10000' RAILS_ENV=profile bin/puma -e production
3. WAIT 5-15 or so minutes for all jitting to stop so we have no cross talk
4. Run ab -n 100 http://localhost:9292/
5. Wait for all new jitting to finish
6. Run ab -n 100 http://localhost:9292/
** Response time (ms)
Here is the response time milliseconds for each percentile.
Skipping 99%ile because it's the same as 100%ile in 100 calls.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 38 | 45 | 41 | 43 |
| 66% | 39 | 50 | 44 | 44 |
| 75% | 47 | 51 | 46 | 45 |
| 80% | 49 | 52 | 47 | 47 |
| 90% | 50 | 63 | 50 | 52 |
| 95% | 60 | 79 | 52 | 55 |
| 98% | 91 | 114 | 91 | 91 |
|100% | 97 | 133 | 96 | 99 |
** Ratio (smaller is better)
Here is the response time increase ratio against no-JIT trunk's one.
| | trunk| trunk|single|objfcn|
| | | JIT|so JIT| JIT|
|:----|:-----|:-----|:-----|:-----|
| 50% | 1.00x| 1.18x| 1.08x| 1.13x|
| 66% | 1.00x| 1.28x| 1.13x| 1.13x|
| 75% | 1.00x| 1.09x| 0.98x| 0.96x|
| 80% | 1.00x| 1.06x| 0.96x| 0.96x|
| 90% | 1.00x| 1.26x| 1.00x| 1.04x|
| 95% | 1.00x| 1.32x| 0.87x| 0.92x|
| 98% | 1.00x| 1.25x| 1.00x| 1.00x|
|100% | 1.00x| 1.37x| 0.99x| 1.02x|
While 50 and 60 %ile are still worse than no-JIT trunk, 75, 80, 90, 95,
98 and 100% are not slower than that.
So now it's a little harder to say "MJIT slows down Rails applications".
Probably I can close [Bug #14490] now. Let's start improving it.
Close https://github.com/ruby/ruby/pull/1921
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@64094 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
2018-07-28 12:14:56 -04:00
|
|
|
end
|
2018-05-22 11:48:58 -04:00
|
|
|
unless err_lines.empty?
|
|
|
|
warn err_lines.join(''), uplevel: uplevel
|
|
|
|
end
|
2018-02-08 07:40:33 -05:00
|
|
|
end
|
|
|
|
|
2019-11-10 17:12:13 -05:00
|
|
|
def mark_tested_insn(insn, used_insns:, uplevel: 1)
|
2021-11-12 02:40:49 -05:00
|
|
|
# Currently, this check emits a false-positive warning against opt_regexpmatch2,
|
|
|
|
# so the insn is excluded explicitly. See https://bugs.ruby-lang.org/issues/18269
|
|
|
|
if !used_insns.include?(insn) && insn != :opt_regexpmatch2
|
2019-11-10 00:16:10 -05:00
|
|
|
$stderr.puts
|
2019-11-10 17:12:13 -05:00
|
|
|
warn "'#{insn}' insn is not included in the script. Actual insns are: #{used_insns.join(' ')}\n", uplevel: uplevel
|
2019-11-10 00:16:10 -05:00
|
|
|
end
|
|
|
|
TestJIT.untested_insns.delete(insn)
|
|
|
|
end
|
|
|
|
|
2018-05-02 11:12:55 -04:00
|
|
|
# Collect block's insns or defined method's insns, which are expected to be JIT-ed.
|
2018-05-02 11:41:43 -04:00
|
|
|
# Note that this intentionally excludes insns in script's toplevel because they are not JIT-ed.
|
2018-05-02 11:12:55 -04:00
|
|
|
def method_insns(script)
|
|
|
|
insns = []
|
|
|
|
RubyVM::InstructionSequence.compile(script).to_a.last.each do |(insn, *args)|
|
|
|
|
case insn
|
2019-04-05 04:15:11 -04:00
|
|
|
when :send
|
2018-05-02 11:12:55 -04:00
|
|
|
insns += collect_insns(args.last)
|
2019-04-05 04:15:11 -04:00
|
|
|
when :definemethod, :definesmethod
|
|
|
|
insns += collect_insns(args[1])
|
2018-05-22 11:48:58 -04:00
|
|
|
when :defineclass
|
|
|
|
insns += collect_insns(args[1])
|
2018-05-02 11:12:55 -04:00
|
|
|
end
|
|
|
|
end
|
|
|
|
insns.uniq
|
|
|
|
end
|
|
|
|
|
|
|
|
# Recursively collect insns in iseq_array
|
|
|
|
def collect_insns(iseq_array)
|
|
|
|
return [] if iseq_array.nil?
|
|
|
|
|
|
|
|
insns = iseq_array.last.select { |x| x.is_a?(Array) }.map(&:first)
|
|
|
|
iseq_array.last.each do |(insn, *args)|
|
|
|
|
case insn
|
2019-04-05 22:53:32 -04:00
|
|
|
when :definemethod, :definesmethod, :send
|
2018-05-02 11:12:55 -04:00
|
|
|
insns += collect_insns(args.last)
|
|
|
|
end
|
|
|
|
end
|
|
|
|
insns
|
|
|
|
end
|
2018-02-08 07:40:33 -05:00
|
|
|
end
|