[vim/vim] cache NFA regex list buffers to reduce malloc/free overhead (PR #19956)

[mattn]

nfa_regmatch() allocates two large thread list buffers (~40KB each) on every call. For :%s on a large file, this means millions of malloc/free calls with the same compiled regexp. This patch caches those buffers in nfa_regprog_T and reuses them on subsequent calls. Recursive calls still allocate their own buffers.

Benchmark: 10K lines, :%s x50 iterations

Pattern	Before	After	Improvement
\<\(\w\+\%(ing|tion|ed|ly\)|\w\{3,}\)\> (many matches)	4.384s	4.299s	-2%
\(foo|bar|baz\)\{3,}\(qux|quux|corge\)\{2,}... (no match, high nstate)	16.927s	3.015s	-82%

The effect is largest when nstate is high and per-line matching work is small, making malloc/free the dominant cost.

---

You can view, comment on, or merge this pull request online at:

  https://github.com/vim/vim/pull/19956

Commit Summary

• 4cad87b cache NFA regex list buffers to reduce malloc/free overhead

File Changes

(2 files)

• M src/regexp.h (2)
• M src/regexp_nfa.c (35)

Patch Links:

• https://github.com/vim/vim/pull/19956.patch
• https://github.com/vim/vim/pull/19956.diff

—
Reply to this email directly, view it on GitHub.
You are receiving this because you are subscribed to this thread.Message ID: <vim/vim/pull/19956@github.com>