Optimize Gem.already_loaded? #3793
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Profiling a simple `ruby -e '1'` I see:
```
==================================
Mode: wall(10)
Samples: 3414 (55.10% miss rate)
GC: 856 (25.07%)
==================================
TOTAL (pct) SAMPLES (pct) FRAME
689 (20.2%) 669 (19.6%) Gem.already_loaded?
462 (13.5%) 462 (13.5%) (marking)
393 (11.5%) 393 (11.5%) (sweeping)
815 (23.9%) 365 (10.7%) Gem::Specification.load
1050 (30.8%) 156 (4.6%) Gem.register_default_spec
100 (2.9%) 84 (2.5%) Gem::Specification#files
64 (1.9%) 64 (1.9%) Gem::BasicSpecification#internal_init
136 (4.0%) 59 (1.7%) <top (required)>
2312 (67.7%) 58 (1.7%) <top (required)>
57 (1.7%) 57 (1.7%) RbConfig.expand
81 (2.4%) 55 (1.6%) Gem::Requirement.parse
191 (5.6%) 51 (1.5%) <top (required)>
128 (3.7%) 47 (1.4%) Gem::Requirement#initialize
41 (1.2%) 41 (1.2%) Gem.suffix_regexp
229 (6.7%) 35 (1.0%) <module:Gem>
260 (7.6%) 34 (1.0%) Kernel#require
```
So clearly `Gem.already_loaded?` is a major hotspot.
After this optimization, it's down to:
```
==================================
Mode: wall(10)
Samples: 2653 (58.21% miss rate)
GC: 718 (27.06%)
==================================
TOTAL (pct) SAMPLES (pct) FRAME
416 (15.7%) 416 (15.7%) (marking)
715 (27.0%) 312 (11.8%) Gem::Specification.load
299 (11.3%) 299 (11.3%) (sweeping)
279 (10.5%) 279 (10.5%) Gem.already_loaded?
564 (21.3%) 106 (4.0%) Gem.register_default_spec
95 (3.6%) 80 (3.0%) Gem::Specification#files
72 (2.7%) 72 (2.7%) Gem::BasicSpecification#internal_init
129 (4.9%) 58 (2.2%) <top (required)>
53 (2.0%) 53 (2.0%) RbConfig.expand
1697 (64.0%) 52 (2.0%) <top (required)>
68 (2.6%) 49 (1.8%) Gem::Requirement.parse
183 (6.9%) 48 (1.8%) <top (required)>
112 (4.2%) 44 (1.7%) Gem::Requirement#initialize
220 (8.3%) 33 (1.2%) <module:Gem>
250 (9.4%) 32 (1.2%) Kernel#require
```
The idea is that the vast majority of the time `already_loaded?` won't match
anything. So by first looking for candidate paths that `end_with?` the file we
look for, we save `default_gem_load_paths.size` iterations and string concatenations.
casperisfine
force-pushed
the
optimize-gem-already-loaded
branch
from
2f4f441
to
c60ce88
Compare
4 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Profiling a simple
ruby -e '1'I see:So clearly
Gem.already_loaded?is a major hotspot.After this optimization, it's down to:
The idea is that the vast majority of the time
already_loaded?won't matchanything. So by first looking for candidate paths that
end_with?the file welook for, we save
default_gem_load_paths.sizeiterations and string concatenations.