Redo of 34a2acdac788602c14bf05fb616215187badd504 and
931138b00696419945dc03e10f033b1f53cd50f3 which were reverted.
GitHub PR #4340.
This change implements a cache for class variables. Previously there was
no cache for cvars. Cvar access is slow due to needing to travel all the
way up th ancestor tree before returning the cvar value. The deeper the
ancestor tree the slower cvar access will be.
The benefits of the cache are more visible with a higher number of
included modules due to the way Ruby looks up class variables. The
benchmark here includes 26 modules and shows with the cache, this branch
is 6.5x faster when accessing class variables.
```
compare-ruby: ruby 3.1.0dev (2021-03-15T06:22:34Z master 9e5105c) [x86_64-darwin19]
built-ruby: ruby 3.1.0dev (2021-03-15T12:12:44Z add-cache-for-clas.. c6be009) [x86_64-darwin19]
| |compare-ruby|built-ruby|
|:--------|-----------:|---------:|
|vm_cvar | 5.681M| 36.980M|
| | -| 6.51x|
```
Benchmark.ips calling `ActiveRecord::Base.logger` from within a Rails
application. ActiveRecord::Base.logger has 71 ancestors. The more
ancestors a tree has, the more clear the speed increase. IE if Base had
only one ancestor we'd see no improvement. This benchmark is run on a
vanilla Rails application.
Benchmark code:
```ruby
require "benchmark/ips"
require_relative "config/environment"
Benchmark.ips do |x|
x.report "logger" do
ActiveRecord::Base.logger
end
end
```
Ruby 3.0 master / Rails 6.1:
```
Warming up --------------------------------------
logger 155.251k i/100ms
Calculating -------------------------------------
```
Ruby 3.0 with cvar cache / Rails 6.1:
```
Warming up --------------------------------------
logger 1.546M i/100ms
Calculating -------------------------------------
logger 14.857M (± 4.8%) i/s - 74.198M in 5.006202s
```
Lastly we ran a benchmark to demonstate the difference between master
and our cache when the number of modules increases. This benchmark
measures 1 ancestor, 30 ancestors, and 100 ancestors.
Ruby 3.0 master:
```
Warming up --------------------------------------
1 module 1.231M i/100ms
30 modules 432.020k i/100ms
100 modules 145.399k i/100ms
Calculating -------------------------------------
1 module 12.210M (± 2.1%) i/s - 61.553M in 5.043400s
30 modules 4.354M (± 2.7%) i/s - 22.033M in 5.063839s
100 modules 1.434M (± 2.9%) i/s - 7.270M in 5.072531s
Comparison:
1 module: 12209958.3 i/s
30 modules: 4354217.8 i/s - 2.80x (± 0.00) slower
100 modules: 1434447.3 i/s - 8.51x (± 0.00) slower
```
Ruby 3.0 with cvar cache:
```
Warming up --------------------------------------
1 module 1.641M i/100ms
30 modules 1.655M i/100ms
100 modules 1.620M i/100ms
Calculating -------------------------------------
1 module 16.279M (± 3.8%) i/s - 82.038M in 5.046923s
30 modules 15.891M (± 3.9%) i/s - 79.459M in 5.007958s
100 modules 16.087M (± 3.6%) i/s - 81.005M in 5.041931s
Comparison:
1 module: 16279458.0 i/s
100 modules: 16087484.6 i/s - same-ish: difference falls within error
30 modules: 15891406.2 i/s - same-ish: difference falls within error
```
Co-authored-by: Aaron Patterson <tenderlove@ruby-lang.org>
Instead of on read. Once it's in the inline cache we never have to make
one again. We want to eventually put the value into the cache, and the
best opportunity to do that is when you write the value.
This change implements a cache for class variables. Previously there was
no cache for cvars. Cvar access is slow due to needing to travel all the
way up th ancestor tree before returning the cvar value. The deeper the
ancestor tree the slower cvar access will be.
The benefits of the cache are more visible with a higher number of
included modules due to the way Ruby looks up class variables. The
benchmark here includes 26 modules and shows with the cache, this branch
is 6.5x faster when accessing class variables.
```
compare-ruby: ruby 3.1.0dev (2021-03-15T06:22:34Z master 9e5105ca45) [x86_64-darwin19]
built-ruby: ruby 3.1.0dev (2021-03-15T12:12:44Z add-cache-for-clas.. c6be0093ae) [x86_64-darwin19]
| |compare-ruby|built-ruby|
|:--------|-----------:|---------:|
|vm_cvar | 5.681M| 36.980M|
| | -| 6.51x|
```
Benchmark.ips calling `ActiveRecord::Base.logger` from within a Rails
application. ActiveRecord::Base.logger has 71 ancestors. The more
ancestors a tree has, the more clear the speed increase. IE if Base had
only one ancestor we'd see no improvement. This benchmark is run on a
vanilla Rails application.
Benchmark code:
```ruby
require "benchmark/ips"
require_relative "config/environment"
Benchmark.ips do |x|
x.report "logger" do
ActiveRecord::Base.logger
end
end
```
Ruby 3.0 master / Rails 6.1:
```
Warming up --------------------------------------
logger 155.251k i/100ms
Calculating -------------------------------------
```
Ruby 3.0 with cvar cache / Rails 6.1:
```
Warming up --------------------------------------
logger 1.546M i/100ms
Calculating -------------------------------------
logger 14.857M (± 4.8%) i/s - 74.198M in 5.006202s
```
Lastly we ran a benchmark to demonstate the difference between master
and our cache when the number of modules increases. This benchmark
measures 1 ancestor, 30 ancestors, and 100 ancestors.
Ruby 3.0 master:
```
Warming up --------------------------------------
1 module 1.231M i/100ms
30 modules 432.020k i/100ms
100 modules 145.399k i/100ms
Calculating -------------------------------------
1 module 12.210M (± 2.1%) i/s - 61.553M in 5.043400s
30 modules 4.354M (± 2.7%) i/s - 22.033M in 5.063839s
100 modules 1.434M (± 2.9%) i/s - 7.270M in 5.072531s
Comparison:
1 module: 12209958.3 i/s
30 modules: 4354217.8 i/s - 2.80x (± 0.00) slower
100 modules: 1434447.3 i/s - 8.51x (± 0.00) slower
```
Ruby 3.0 with cvar cache:
```
Warming up --------------------------------------
1 module 1.641M i/100ms
30 modules 1.655M i/100ms
100 modules 1.620M i/100ms
Calculating -------------------------------------
1 module 16.279M (± 3.8%) i/s - 82.038M in 5.046923s
30 modules 15.891M (± 3.9%) i/s - 79.459M in 5.007958s
100 modules 16.087M (± 3.6%) i/s - 81.005M in 5.041931s
Comparison:
1 module: 16279458.0 i/s
100 modules: 16087484.6 i/s - same-ish: difference falls within error
30 modules: 15891406.2 i/s - same-ish: difference falls within error
```
Co-authored-by: Aaron Patterson <tenderlove@ruby-lang.org>
rb_funcall* (rb_funcall(), rb_funcallv(), ...) functions invokes
Ruby's method with given receiver. Ruby 2.7 introduced inline method
cache with static memory area. However, Ruby 3.0 reimplemented the
method cache data structures and the inline cache was removed.
Without inline cache, rb_funcall* searched methods everytime.
Most of cases per-Class Method Cache (pCMC) will be helped but
pCMC requires VM-wide locking and it hurts performance on
multi-Ractor execution, especially all Ractors calls methods
with rb_funcall*.
This patch introduced Global Call-Cache Cache Table (gccct) for
rb_funcall*. Call-Cache was introduced from Ruby 3.0 to manage
method cache entry atomically and gccct enables method-caching
without VM-wide locking. This table solves the performance issue
on multi-ractor execution.
[Bug #17497]
Ruby-level method invocation does not use gccct because it has
inline-method-cache and the table size is limited. Basically
rb_funcall* is not used frequently, so 1023 entries can be enough.
We will revisit the table size if it is not enough.
constant cache `IC` is accessed by non-atomic manner and there are
thread-safety issues, so Ruby 3.0 disables to use const cache on
non-main ractors.
This patch enables it by introducing `imemo_constcache` and allocates
it by every re-fill of const cache like `imemo_callcache`.
[Bug #17510]
Now `IC` only has one entry `IC::entry` and it points to
`iseq_inline_constant_cache_entry`, managed by T_IMEMO object.
`IC` is atomic data structure so `rb_mjit_before_vm_ic_update()` and
`rb_mjit_after_vm_ic_update()` is not needed.
add cc_found_in_ccs (renamed from cc_found_ccs), cc_not_found_in_ccs,
call0_public, call0_other debug counters to measure more details.
also it contains several modification.
This commit adds a debug counter for the case where the inline cache
*missed* but the ivar index table has an entry for that ivar. This is a
case where a polymorphic cache could help
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.
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.
for debug counters
```
../include/ruby/intern.h:1175:137: warning: passing argument 3 of 'rb_define_singleton_method0' from incompatible pointer type [-Wincompatible-pointer-types]
#define rb_define_singleton_method(klass, mid, func, arity) rb_define_singleton_method_choose_prototypem3((arity),(func))((klass),(mid),(func),(arity));
^
../vm.c:2958:5: note: in expansion of macro 'rb_define_singleton_method'
rb_define_singleton_method(rb_cRubyVM, "show_debug_counters", rb_debug_counter_show, 0);
^~~~~~~~~~~~~~~~~~~~~~~~~~
../include/ruby/intern.h:1139:99: note: expected 'VALUE (*)(VALUE) {aka long unsigned int (*)(long unsigned int)}' but argument is of type 'VALUE (*)(void) {aka long unsigned int (*)(void)}'
__attribute__((__unused__,__weakref__("rb_define_singleton_method"),__nonnull__(2,3)))static void rb_define_singleton_method0 (VALUE,const char*,VALUE(*)(VALUE),int);
```
I am trying to study debug counters inside a Rails application.
Accessing debug counters by killing the process is hard because child
processes don't get the same TRAP as the parent, and Rails seems to
intercept calls to `exit`. Adding this method lets me print the debug
counters when I want (at the end of requests for example)
On ar_table, Do not keep a full-length hash value (FLHV, 8 bytes)
but keep a 1 byte hint from a FLHV (lowest byte of FLHV).
An ar_table only contains at least 8 entries, so hints consumes
8 bytes at most. We can store hints in RHash::ar_hint.
On 32bit CPU, we use 4 entries ar_table.
The advantages:
* We don't need to keep FLHV so ar_table only consumes
16 bytes (VALUEs of key and value) * 8 entries = 128 bytes.
* We don't need to scan ar_table, but only need to check hints
in many cases. Especially we don't need to access ar_table
if there is no match entries (in many cases).
It will increase memory cache locality.
The disadvantages:
* This technique can increase `#eql?` time because hints can
conflicts (in theory, it conflicts once in 256 times).
It can introduce incompatibility if there is a object x where
x.eql? returns true even if hash values are different.
I believe we don't need to care such irregular case.
* We need to re-calculate FLHV if we need to switch from ar_table
to st_table (e.g. exceeds 8 entries).
It also can introduce incompatibility, on mutating key objects.
I believe we don't need to care such irregular case too.
Add new debug counters to measure the performance:
* artable_hint_hit - hint is matched and eql?#=>true
* artable_hint_miss - hint is not matched but eql?#=>false
* artable_hint_notfound - lookup counts
Shared arrays created by Array#dup and so on points
a shared_root object to manage lifetime of Array buffer.
However, sometimes shared_root is called only shared so
it is confusing. So I fixed these wording "shared" to "shared_root".
* RArray::heap::aux::shared -> RArray::heap::aux::shared_root
* ARY_SHARED() -> ARY_SHARED_ROOT()
* ARY_SHARED_NUM() -> ARY_SHARED_ROOT_REFCNT()
Also, add some debug_counters to count shared array objects.
* ary_shared_create: shared ary by Array#dup and so on.
* ary_shared: finished in shard.
* ary_shared_root_occupied: shared_root but has only 1 refcnt.
The number (ary_shared - ary_shared_root_occupied) is meaningful.
is defined. It's 0 by default and so it dissappears on actual build.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@67544 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
eileencodes
Aaron Patterson
Ryuta Kamizono
Koichi Sasada
Nobuyoshi Nakada
Takashi Kokubun
卜部昌平
Kazuhiro NISHIYAMA
Koichi Sasada
Lourens Naudé
k0kubun