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Outdated suggestions cannot be applied.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.
Why?
in my company, the jfr to pprof allocate too much times of objects, and those cost most of cpu.
this is the allocation memory profile of my parca-agent:
What?
this pr reduce 99% percent of memory allocation and reduce 80% of cpu cost in jfr to pprof.
How?
1, cache parseArgs and getFileName and getPid and dimensions.
2, reuse slice as more as we can.
3, use bytes.buffer to build function key, and use unsafe string in map search to avoid memory allocation.
4, use unsafe string sample key in map search, only create locations when new sample created.
5, use uint64 as location key to avoid int to string convert.