chore: add memory allocation benchmark with bump pointer allocator #20302
+631
−0
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
Adds a thread-local bump pointer allocator behind -DBUMP_ALLOCATOR that preallocates 1GB per thread via mmap and never frees. This allows measuring the performance impact of malloc/free in bb.
Instruments operator new/delete with backward-cpp stack traces, sizes, lifetimes, and thread affinity. Dumps a report at exit sorted by peak concurrent bytes to identify candidates for region-based allocation. Build with: cmake --preset alloc-profiler
- Add -fno-omit-frame-pointer for reliable backtrace() stack walking - Use RelWithDebInfo build type for addr2line symbolization - Replace backward-cpp symbolization with raw PC dump (avoids SIGSEGV from re-entrant allocation in TraceResolver during atexit) - Fix static destruction order by registering dump_report via atexit inside state() initialization - Use net CRS factory in chonk_bench when ALLOC_PROFILER defined
Use dladdr to get binary base address and subtract it from all PC addresses in the report. This makes addresses directly usable with addr2line without needing to know the ASLR base.
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Outdated suggestions cannot be applied.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.
Create an experimental version of bb with a bump pointer allocator that preallocates 1GB of RAM per thread and never deallocates memory. This allows us to measure the performance impact and total time loss attributable to memory allocation operations in the current implementation. This is a temporary branch for profiling purposes only.