perf: update gumbo utf8 decode #2735
Closed
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
from https://bjoern.hoehrmann.de/utf-8/decoder/dfa/ which apparently saves a shift instruction for every byte. benchmarking doesn't find a discernible performance improvement, though.
Comment on lines
-76
to
+73
| static inline uint32_t decode(uint32_t* state, uint32_t* codep, uint32_t byte) { | ||
| uint32_t static inline | ||
| decode(uint32_t* state, uint32_t* codep, uint32_t byte) { |
There was a problem hiding this comment.
It seems odd to put static inline between the return type and the function name. I don't think any of the other functions do that.
Comment on lines
-79
to
81
| *codep = | ||
| (*state != UTF8_ACCEPT) | ||
| ? (byte & 0x3fu) | (*codep << 6) | ||
| : (0xff >> type) & (byte); | ||
| *codep = (*state != UTF8_ACCEPT) ? | ||
| (byte & 0x3fu) | (*codep << 6) : | ||
| (0xff >> type) & (byte); | ||
|
|
||
| *state = utf8d[256 + *state + type]; | ||
| *state = utf8d[256 + *state*16 + type]; | ||
| return *state; |
There was a problem hiding this comment.
I believe this is the older, slower code. The *16 is the shift that Rich Felker eliminated by pre-multiplying
On 24th June 2010 Rich Felker pointed out that the state values in the transition table can be pre-multiplied with 16 which would save a shift instruction for every byte. D'oh! We actually just need 12 and can throw away the filler values previously in the table making the table 36 bytes shorter and save the shift in the code.
There was a problem hiding this comment.
Ah, I see, that's confusing. OK, will close.
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Outdated suggestions cannot be applied.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.
What problem is this PR intended to solve?
Related to #2722
update gumbo utf8 decode() from https://bjoern.hoehrmann.de/utf-8/decoder/dfa/ which apparently saves a shift instruction for every byte. benchmarking doesn't find a discernible performance improvement, though.
Have you included adequate test coverage?
N/A
Does this change affect the behavior of either the C or the Java implementations?
N/A