| Age | Commit message (Collapse) | Author |
|---|
|
|
|
Some tunings.
* add `inline` for vm_sendish()
* pass enum instead of func ptr to vm_sendish()
* reorder initial order of `calling` struct.
* add ALWAYS_INLINE for vm_search_method_fastpath()
* call vm_search_method_fastpath() from vm_sendish()
Notes:
Merged: https://github.com/ruby/ruby/pull/3922
|
|
```
$ benchmark-driver -v --rbenv 'before --jit;after --jit' --repeat-count=12 --alternate --output=all benchmark.yml
before --jit: ruby 3.0.0dev (2020-12-17T06:17:46Z master 3b4d698e0b) +JIT [x86_64-linux]
after --jit: ruby 3.0.0dev (2020-12-17T07:01:48Z master 843abb96f0) +JIT [x86_64-linux]
last_commit=Lazily move PC with RUBY_VM_CHECK_INTS
Calculating -------------------------------------
before --jit after --jit
Optcarrot Lan_Master.nes 80.29343646660429 83.15779723251525 fps
82.26755637885149 85.50197941326810
83.50682959728820 88.14657804306270
85.01236533133049 88.78201988978667
87.81799334561326 88.94841008936447
87.88228562393064 89.37925215601926
88.06695585889995 89.86143277214475
88.84730834922165 90.00773346420887
90.46317871213088 90.82603371104014
90.96308347148916 91.29797694822179
90.97945938504556 91.31086331868738
91.57127890154500 91.49949184318844
```
|
|
* Inline getconstant on JIT
* Support USE_MJIT=0
Notes:
Merged-By: k0kubun <takashikkbn@gmail.com>
|
|
`cd` is passed to method call functions to method invocation
functions, but `cd` can be manipulated by other ractors simultaneously
so it contains thread-safety issue.
To solve this issue, this patch stores `ci` and found `cc` to `calling`
and stops to pass `cd`.
Notes:
Merged: https://github.com/ruby/ruby/pull/3903
|
|
https://github.com/ruby/ruby/runs/1533401436
|
|
It can no longer issue a warning.
Notes:
Merged: https://github.com/ruby/ruby/pull/3879
|
|
Passing current ec can improve performance of newobj. This patch
tries it for Array and String literals ([] and '').
Notes:
Merged: https://github.com/ruby/ruby/pull/3842
|
|
iv_index_tbl manages instance variable indexes (ID -> index).
This data structure should be synchronized with other ractors
so introduce some VM locks.
This patch also introduced atomic ivar cache used by
set/getinlinecache instructions. To make updating ivar cache (IVC),
we changed iv_index_tbl data structure to manage (ID -> entry)
and an entry points serial and index. IVC points to this entry so
that cache update becomes atomically.
Notes:
Merged: https://github.com/ruby/ruby/pull/3662
|
|
* Remove freezestring instruction since this was the only usage for it.
* [Feature #17104]
Notes:
Merged: https://github.com/ruby/ruby/pull/3488
|
|
Noticed that struct rb_builtin_function is a purely compile-time
constant. MJIT can eliminate some runtime calculations by statically
generate dedicated C code generator for each builtin functions.
Notes:
Merged: https://github.com/ruby/ruby/pull/3305
|
|
Use ID instead of GENTRY for gvars.
Global variables are compiled into GENTRY (a pointer to struct
rb_global_entry). This patch replace this GENTRY to ID and
make the code simple.
We need to search GENTRY from ID every time (st_lookup), so
additional overhead will be introduced.
However, the performance of accessing global variables is not
important now a day and this simplicity helps Ractor development.
Notes:
Merged-By: ko1 <ko1@atdot.net>
|
|
using opt_invokebuiltin_delegate_leave insn.
Since Ruby 2.7, :return of methods using builtin have not been traced properly.
|
|
* Remove obsoleted opt_call_c_function insn
* Keep opt_call_c_function with DEFINE_INSN_IF
Notes:
Merged-By: k0kubun <takashikkbn@gmail.com>
|
|
These two function were almost identical, except in case of
T_STRING/T_FLOAT. Why not merge them into one, and let the difference be
handled in normal method calls (slowpath). This does not improve
runtime performance for me, but at least reduces for instance rb_eql_opt
from 653 bytes to 86 bytes on my machine, according to nm(1).
Notes:
Merged: https://github.com/ruby/ruby/pull/3169
|
|
This changes the following warnings:
* warning: class variable access from toplevel
* warning: class variable @foo of D is overtaken by C
into RuntimeErrors. Handle defined?(@@foo) at toplevel
by returning nil instead of raising an exception (the previous
behavior warned before returning nil when defined? was used).
Refactor the specs to avoid the warnings even in older versions.
The specs were checking for the warnings, but the purpose of
the related specs as evidenced from their description is to
test for behavior, not for warnings.
Fixes [Bug #14541]
Notes:
Merged: https://github.com/ruby/ruby/pull/2987
|
|
This patch contains several ideas:
(1) Disposable inline method cache (IMC) for race-free inline method cache
* Making call-cache (CC) as a RVALUE (GC target object) and allocate new
CC on cache miss.
* This technique allows race-free access from parallel processing
elements like RCU.
(2) Introduce per-Class method cache (pCMC)
* Instead of fixed-size global method cache (GMC), pCMC allows flexible
cache size.
* Caching CCs reduces CC allocation and allow sharing CC's fast-path
between same call-info (CI) call-sites.
(3) Invalidate an inline method cache by invalidating corresponding method
entries (MEs)
* Instead of using class serials, we set "invalidated" flag for method
entry itself to represent cache invalidation.
* Compare with using class serials, the impact of method modification
(add/overwrite/delete) is small.
* Updating class serials invalidate all method caches of the class and
sub-classes.
* Proposed approach only invalidate the method cache of only one ME.
See [Feature #16614] for more details.
Notes:
Merged: https://github.com/ruby/ruby/pull/2888
|
|
Now, rb_call_info contains how to call the method with tuple of
(mid, orig_argc, flags, kwarg). Most of cases, kwarg == NULL and
mid+argc+flags only requires 64bits. So this patch packed
rb_call_info to VALUE (1 word) on such cases. If we can not
represent it in VALUE, then use imemo_callinfo which contains
conventional callinfo (rb_callinfo, renamed from rb_call_info).
iseq->body->ci_kw_size is removed because all of callinfo is VALUE
size (packed ci or a pointer to imemo_callinfo).
To access ci information, we need to use these functions:
vm_ci_mid(ci), _flag(ci), _argc(ci), _kwarg(ci).
struct rb_call_info_kw_arg is renamed to rb_callinfo_kwarg.
rb_funcallv_with_cc() and rb_method_basic_definition_p_with_cc()
is temporary removed because cd->ci should be marked.
Notes:
Merged: https://github.com/ruby/ruby/pull/2888
|
|
|
|
This commit introduces an "inline ivar cache" struct. The reason we
need this is so compaction can differentiate from an ivar cache and a
regular inline cache. Regular inline caches contain references to
`VALUE` and ivar caches just contain references to the ivar index. With
this new struct we can easily update references for inline caches (but
not inline var caches as they just contain an int)
|
|
Asynchronous events such as signal trap, finalization timing,
thread switching and so on are managed by "interrupt_flag".
Ruby's threads check this flag periodically and if a thread
does not check this flag, above events doesn't happen.
This checking is CHECK_INTS() (related) macro and it is placed
at some places (laeve instruction and so on). However, at the end
of C methods, C blocks (IMEMO_IFUNC) etc there are no checking
and it can introduce uninterruptible thread.
To modify this situation, we decide to place CHECK_INTS() at
vm_pop_frame(). It increases interrupt checking points.
[Bug #16366]
This patch can introduce unexpected events...
|
|
This reverts commit 2e6f1cf8b264f4c8499c4e5f18bf662fdade04ff.
|
|
|
|
add casts to avoid compile error.
http://ci.rvm.jp/results/trunk_clang_39@silicon-docker/2402215
|
|
opt_invokebuiltin_delegate and opt_invokebuiltin_delegate_leave
invokes builtin functions with same parameters of the method.
This technique eliminate stack push operations. However, delegation
parameters should be completely same as given parameters.
(e.g. `def foo(a, b, c) __builtin_foo(a, b, c)` is okay, but
__builtin_foo(b, c) is not allowed)
This patch relaxes this restriction. ISeq has a local variables
table which includes parameters. For example, the method defined
as `def foo(a, b, c) x=y=nil`, then local variables table contains
[a, b, c, x, y]. If calling builtin-function with arguments which
are sub-array of the lvar table, use opt_invokebuiltin_delegate
instruction with start index. For example, `__builtin_foo(b, c)`,
`__builtin_bar(c, x, y)` is okay, and so on.
|
|
This reverts commit 67c574736912003c377218153f9d3b9c0c96a17b.
[Feature #16275]
|
|
vm_invoke_builtin() accesses VM stack via cfp->sp. However, MJIT
can use their own stack. To access them appropriately, we need to
use STACK_ADDR_FROM_TOP().
|
|
Support loading builtin features written in Ruby, which implement
with C builtin functions.
[Feature #16254]
Several features:
(1) Load .rb file at boottime with native binary.
Now, prelude.rb is loaded at boottime. However, this file is contained
into the interpreter as a text format and we need to compile it.
This patch contains a feature to load from binary format.
(2) __builtin_func() in Ruby call func() written in C.
In Ruby file, we can write `__builtin_func()` like method call.
However this is not a method call, but special syntax to call
a function `func()` written in C. C functions should be defined
in a file (same compile unit) which load this .rb file.
Functions (`func` in above example) should be defined with
(a) 1st parameter: rb_execution_context_t *ec
(b) rest parameters (0 to 15).
(c) VALUE return type.
This is very similar requirements for functions used by
rb_define_method(), however `rb_execution_context_t *ec`
is new requirement.
(3) automatic C code generation from .rb files.
tool/mk_builtin_loader.rb creates a C code to load .rb files
needed by miniruby and ruby command. This script is run by
BASERUBY, so *.rb should be written in BASERUBY compatbile
syntax. This script load a .rb file and find all of __builtin_
prefix method calls, and generate a part of C code to export
functions.
tool/mk_builtin_binary.rb creates a C code which contains
binary compiled Ruby files needed by ruby command.
Notes:
Merged: https://github.com/ruby/ruby/pull/2655
|
|
To perform a regular method call, the VM needs two structs,
`rb_call_info` and `rb_call_cache`. At the moment, we allocate these two
structures in separate buffers. In the worst case, the CPU needs to read
4 cache lines to complete a method call. Putting the two structures
together reduces the maximum number of cache line reads to 2.
Combining the structures also saves 8 bytes per call site as the current
layout uses separate two pointers for the call info and the call cache.
This saves about 2 MiB on Discourse.
This change improves the Optcarrot benchmark at least 3%. For more
details, see attached bugs.ruby-lang.org ticket.
Complications:
- A new instruction attribute `comptime_sp_inc` is introduced to
calculate SP increase at compile time without using call caches. At
compile time, a `TS_CALLDATA` operand points to a call info struct, but
at runtime, the same operand points to a call data struct. Instruction
that explicitly define `sp_inc` also need to define `comptime_sp_inc`.
- MJIT code for copying call cache becomes slightly more complicated.
- This changes the bytecode format, which might break existing tools.
[Misc #16258]
Notes:
Merged: https://github.com/ruby/ruby/pull/2564
|
|
This reverts commits: 10d6a3aca7 8ba48c1b85 fba8627dc1 dd883de5ba
6c6a25feca 167e6b48f1 7cb96d41a5 3207979278 595b3c4fdd 1521f7cf89
c11c5e69ac cf33608203 3632a812c0 f56506be0d 86427a3219 .
The reason for the revert is that we observe ABA problem around
inline method cache. When a cache misshits, we search for a
method entry. And if the entry is identical to what was cached
before, we reuse the cache. But the commits we are reverting here
introduced situations where a method entry is freed, then the
identical memory region is used for another method entry. An
inline method cache cannot detect that ABA.
Here is a code that reproduce such situation:
```ruby
require 'prime'
class << Integer
alias org_sqrt sqrt
def sqrt(n)
raise
end
GC.stress = true
Prime.each(7*37){} rescue nil # <- Here we populate CC
class << Object.new; end
# These adjacent remove-then-alias maneuver
# frees a method entry, then immediately
# reuses it for another.
remove_method :sqrt
alias sqrt org_sqrt
end
Prime.each(7*37).to_a # <- SEGV
```
|
|
At last, not only myself but also your compiler are fully confident
that the method entries pointed from call caches are immutable. We
don't have to worry about silent updates. Just delete the branch
that is now always false.
Calculating -------------------------------------
ours trunk
vm2_poly_same_method 2.142M 2.070M i/s - 6.000M times in 2.801148s 2.898994s
Comparison:
vm2_poly_same_method
ours: 2141979.2 i/s
trunk: 2069683.8 i/s - 1.03x slower
Notes:
Merged: https://github.com/ruby/ruby/pull/2486
|
|
I noticed that in case of cache misshit, re-calculated cc->me can
be the same method entry than the pevious one. That is an okay
situation but can't we partially reuse the cache, because cc->call
should still be valid then?
One thing that has to be special-cased is when the method entry
gets amended by some refinements. That happens behind-the-scene
of call cache mechanism. We have to check if cc->me->def points to
the previously saved one.
Calculating -------------------------------------
trunk ours
vm2_poly_same_method 1.534M 2.025M i/s - 6.000M times in 3.910203s 2.962752s
Comparison:
vm2_poly_same_method
ours: 2025143.9 i/s
trunk: 1534447.2 i/s - 1.32x slower
Notes:
Merged: https://github.com/ruby/ruby/pull/2468
|
|
Notes:
Merged: https://github.com/ruby/ruby/pull/2420
|
|
Ignore empty keyword splats in arrays
Notes:
Merged: https://github.com/ruby/ruby/pull/2418
Merged-By: jeremyevans <code@jeremyevans.net>
|
|
----
trunk: ruby 2.6.0dev (2018-09-18 trunk 64767) [x86_64-darwin15]
ours: ruby 2.6.0dev (2018-09-18 opt_regexpmatch 64775) [x86_64-darwin15]
last_commit=opt_regexpmatch1 is actually making things slower.
Calculating -------------------------------------
trunk ours
Optcarrot Lan_Master.nes 33.877 35.282 fps
Comparison:
Optcarrot Lan_Master.nes
ours: 35.3 fps
trunk: 33.9 fps - 1.04x slower
Notes:
Merged: https://github.com/ruby/ruby/pull/1959
|
|
[Feature #16103]
Close: https://github.com/ruby/ruby/pull/2267
|
|
[Feature #16103]
|
|
Some tooling depends on the current bytecode, and adding an operand
changes the bytecode. While tooling can be updated for new bytecode,
this support doesn't warrant such a change.
|
|
Fixes error when using -Werror,-Wshorten-64-to-32.
|
|
This was an intentional bug added in 1.9.
The approach taken here is to add a second operand to the
getconstant instruction for whether nil should be allowed and
treated as current scope.
Fixes [Bug #11718]
|
|
Inspired by 346aa557b31fe96760e505d30da26eb7a846bac9
Closes: https://github.com/ruby/ruby/pull/2321
|
|
leaf is true by default. Other insns are not specifying it explicitly.
Also the comment describing why it was not leaf is outdated.
|
|
# Benchmark zero?
```
require 'benchmark/ips'
Numeric.class_eval do
def ruby_zero?
self == 0
end
end
Benchmark.ips do |x|
x.report('0.zero?') { 0.ruby_zero? }
x.report('1.zero?') { 1.ruby_zero? }
x.compare!
end
```
## VM
No significant impact for VM.
### before
ruby 2.7.0dev (2019-08-04T02:56:02Z master 2d8c037e97) [x86_64-linux]
0.zero?: 21855445.5 i/s
1.zero?: 21770817.3 i/s - same-ish: difference falls within error
### after
ruby 2.7.0dev (2019-08-04T11:17:10Z opt-eq-leaf 6404bebd6a) [x86_64-linux]
1.zero?: 21958912.3 i/s
0.zero?: 21881625.9 i/s - same-ish: difference falls within error
## JIT
The performance improves about 1.23x.
### before
ruby 2.7.0dev (2019-08-04T02:56:02Z master 2d8c037e97) +JIT [x86_64-linux]
0.zero?: 36343111.6 i/s
1.zero?: 36295153.3 i/s - same-ish: difference falls within error
### after
ruby 2.7.0dev (2019-08-04T11:17:10Z opt-eq-leaf 6404bebd6a) +JIT [x86_64-linux]
0.zero?: 44740467.2 i/s
1.zero?: 44363616.1 i/s - same-ish: difference falls within error
# Benchmark str == str / str != str
```
# frozen_string_literal: true
require 'benchmark/ips'
Benchmark.ips do |x|
x.report('a == a') { 'a' == 'a' }
x.report('a == b') { 'a' == 'b' }
x.report('a != a') { 'a' != 'a' }
x.report('a != b') { 'a' != 'b' }
x.compare!
end
```
## VM
No significant impact for VM.
### before
ruby 2.7.0dev (2019-08-04T02:56:02Z master 2d8c037e97) [x86_64-linux]
a == a: 27286219.0 i/s
a != a: 24892389.5 i/s - 1.10x slower
a == b: 23623635.8 i/s - 1.16x slower
a != b: 21800958.0 i/s - 1.25x slower
### after
ruby 2.7.0dev (2019-08-04T11:17:10Z opt-eq-leaf 6404bebd6a) [x86_64-linux]
a == a: 27224016.2 i/s
a != a: 24490109.5 i/s - 1.11x slower
a == b: 23391052.4 i/s - 1.16x slower
a != b: 21811321.7 i/s - 1.25x slower
## JIT
The performance improves on JIT a little.
### before
ruby 2.7.0dev (2019-08-04T02:56:02Z master 2d8c037e97) +JIT [x86_64-linux]
a == a: 42010674.7 i/s
a != a: 38920311.2 i/s - same-ish: difference falls within error
a == b: 32574262.2 i/s - 1.29x slower
a != b: 32099790.3 i/s - 1.31x slower
### after
ruby 2.7.0dev (2019-08-04T11:17:10Z opt-eq-leaf 6404bebd6a) +JIT [x86_64-linux]
a == a: 46902738.8 i/s
a != a: 43097258.6 i/s - 1.09x slower
a == b: 35822018.4 i/s - 1.31x slower
a != b: 33377257.8 i/s - 1.41x slower
This is needed towards Bug#15589.
Closes: https://github.com/ruby/ruby/pull/2318
|
|
This reverts commit a0980f2446c0db735b8ffeb37e241370c458a626.
Retry for macOS Mojave.
|
|
This reverts commit 9faef3113fb4331524b81ba73005ba13fa0ef6c6.
It seemed to cause a failure on macOS Mojave, though I'm unsure how.
https://rubyci.org/logs/rubyci.s3.amazonaws.com/osx1014/ruby-master/log/20190802T034503Z.fail.html.gz
This tentative revert is to check if the issue is actually caused by the
change or not.
|
|
This commit adds a specialized instruction for called to `.nil?`. It is
about 27% faster than master in the case where the object is nil or not
nil. In the case where an object implements `nil?`, I think it may be
slightly slower. Here is a benchmark:
```ruby
require "benchmark/ips"
class Niller
def nil?; true; end
end
not_nil = Object.new
xnil = nil
niller = Niller.new
Benchmark.ips do |x|
x.report("nil?") { xnil.nil? }
x.report("not nil") { not_nil.nil? }
x.report("niller") { niller.nil? }
end
```
On Ruby master:
```
[aaron@TC ~/g/ruby (master)]$ ./ruby compil.rb
Warming up --------------------------------------
nil? 429.195k i/100ms
not nil 437.889k i/100ms
niller 437.935k i/100ms
Calculating -------------------------------------
nil? 20.166M (± 8.1%) i/s - 100.002M in 5.002794s
not nil 20.046M (± 7.6%) i/s - 99.839M in 5.020086s
niller 22.467M (± 6.1%) i/s - 112.111M in 5.013817s
[aaron@TC ~/g/ruby (master)]$ ./ruby compil.rb
Warming up --------------------------------------
nil? 449.660k i/100ms
not nil 433.836k i/100ms
niller 443.073k i/100ms
Calculating -------------------------------------
nil? 19.997M (± 8.8%) i/s - 99.375M in 5.020458s
not nil 20.529M (± 7.0%) i/s - 102.385M in 5.020689s
niller 21.796M (± 8.0%) i/s - 108.110M in 5.002300s
[aaron@TC ~/g/ruby (master)]$ ./ruby compil.rb
Warming up --------------------------------------
nil? 402.119k i/100ms
not nil 438.968k i/100ms
niller 398.226k i/100ms
Calculating -------------------------------------
nil? 20.050M (±12.2%) i/s - 98.519M in 5.008817s
not nil 20.614M (± 8.0%) i/s - 102.280M in 5.004531s
niller 22.223M (± 8.8%) i/s - 110.309M in 5.013106s
```
On this branch:
```
[aaron@TC ~/g/ruby (specialized-nilp)]$ ./ruby compil.rb
Warming up --------------------------------------
nil? 468.371k i/100ms
not nil 456.517k i/100ms
niller 454.981k i/100ms
Calculating -------------------------------------
nil? 27.849M (± 7.8%) i/s - 138.169M in 5.001730s
not nil 26.417M (± 8.7%) i/s - 131.020M in 5.011674s
niller 21.561M (± 7.5%) i/s - 107.376M in 5.018113s
[aaron@TC ~/g/ruby (specialized-nilp)]$ ./ruby compil.rb
Warming up --------------------------------------
nil? 477.259k i/100ms
not nil 428.712k i/100ms
niller 446.109k i/100ms
Calculating -------------------------------------
nil? 28.071M (± 7.3%) i/s - 139.837M in 5.016590s
not nil 25.789M (±12.9%) i/s - 126.470M in 5.011144s
niller 20.002M (±12.2%) i/s - 98.144M in 5.001737s
[aaron@TC ~/g/ruby (specialized-nilp)]$ ./ruby compil.rb
Warming up --------------------------------------
nil? 467.676k i/100ms
not nil 445.791k i/100ms
niller 415.024k i/100ms
Calculating -------------------------------------
nil? 26.907M (± 8.0%) i/s - 133.755M in 5.013915s
not nil 25.319M (± 7.9%) i/s - 125.713M in 5.007758s
niller 19.569M (±11.8%) i/s - 96.286M in 5.008533s
```
Co-Authored-By: Ashe Connor <kivikakk@github.com>
|
|
* insns.def: add definemethod and definesmethod (singleton method)
instructions. Old YARV contains these instructions, but it is moved
to methods of FrozenCore class because remove number of instructions
can improve performance for some techniques (static stack caching
and so on). However, we don't employ these technique and it is hard
to optimize/analysis definition sequence. So I decide to introduce
them (and remove definition methods). `putiseq` insn is also removed.
* vm_method.c (rb_scope_visibility_get): renamed to
`vm_scope_visibility_get()` and make it accept `ec`.
Same for `vm_scope_module_func_check()`.
These fixes are result of refactoring `vm_define_method`.
* vm_insnhelper.c (rb_vm_get_cref): renamed to `vm_get_cref`
because of consistency with other functions.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67442 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
|
|
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67372 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
|
|
GET_CFP() macro contains performance counter logic.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67371 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
|
|
related: r66982
Sadly opt_regexpmatch2 was not a leaf insn either.
http://ci.rvm.jp/results/trunk-vm-asserts@silicon-docker/1751213
CHECK_INTERRUPT_IN_MATCH_AT is just like RUBY_VM_CHECK_INTS, and it may
call arbitrary Ruby method, for example a GC finalizer from postponed
job in this case.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67091 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
|
