Reduce memory usage of Cartridge::Mbc1 #23
Closed
Conversation
This file contains hidden or 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
Precomputing one Proc per possible address amount of 8MiB of memory and almost 100k objects the GC as to scan etc. This can be reduced to just 24 objects with only an extra bitshit on access.
byroot
added a commit
to byroot/rubyboy
that referenced
this pull request
Feb 9, 2025
Closes: sacckey#23 Having to call a proc for every memory access isn't ideal, that quite a bit of overhead. If instead we use a `case` statement with only immediate values and a few bitwise operations we can let YJIT generate very efficient code. two runs on main with ruby 3.4.1 / YJIT: ``` Ruby: 3.4.1 YJIT: true 1: 4.051815 sec 2: 3.989457 sec 3: 3.963555 sec FPS: 374.849216902501 Ruby: 3.4.1 YJIT: true 1: 3.872855 sec 2: 3.943231 sec 3: 4.024268 sec FPS: 380.05620440064547 ``` On this branch: ``` Ruby: 3.4.1 YJIT: true 1: 3.625805 sec 2: 3.615032 sec 3: 3.572852 sec FPS: 416.1392102177157 Ruby: 3.4.1 YJIT: true 1: 3.543953 sec 2: 3.520738 sec 3: 3.506004 sec FPS: 425.70521616601366 ```
Merged
Contributor
Author
|
Actually there's much simpler: #24 |
byroot
added a commit
to byroot/rubyboy
that referenced
this pull request
Feb 10, 2025
Closes: sacckey#23 Having to call a proc for every memory access isn't ideal, that quite a bit of overhead. If instead we use a `case` statement with only immediate values and a few bitwise operations we can let YJIT generate very efficient code. two runs on main with ruby 3.4.1 / YJIT: ``` Ruby: 3.4.1 YJIT: true 1: 4.051815 sec 2: 3.989457 sec 3: 3.963555 sec FPS: 374.849216902501 Ruby: 3.4.1 YJIT: true 1: 3.872855 sec 2: 3.943231 sec 3: 4.024268 sec FPS: 380.05620440064547 ``` On this branch: ``` Ruby: 3.4.1 YJIT: true 1: 3.625805 sec 2: 3.615032 sec 3: 3.572852 sec FPS: 416.1392102177157 Ruby: 3.4.1 YJIT: true 1: 3.543953 sec 2: 3.520738 sec 3: 3.506004 sec FPS: 425.70521616601366 ```
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
Precomputing one Proc per possible address amount of 8MiB of memory and almost 100k objects the GC as to scan etc.
This can be reduced to just 24 objects with only an extra bitshift on access.