Mercurial > hg
view contrib/fuzz/bdiff.cc @ 40978:42f59d3f714d
delta: exclude base candidate much smaller than the target
If a revision's full text is that much bigger than a base candidate full text,
we no longer consider that candidate.
This solves a pathological case we encountered on a very specify repository.
It contains a long series of changesets with a very small manifest (one file)
co-existing with others changesets using a very large manifest.
Without this filtering, we ended up considering a large number of tiny full
snapshots as a potential base. It resulted in very large delta (the size of
the full text) and mercurial spending 99% of its time compressing these
deltas.
The timing of a commit moved from about 400s to about 10s (still slow, but not
ridiculously slow).
author | Boris Feld <boris.feld@octobus.net> |
---|---|
date | Mon, 17 Dec 2018 10:42:19 +0100 |
parents | fa0ddd5e8fff |
children | dbc39f028c9f |
line wrap: on
line source
/* * bdiff.cc - fuzzer harness for bdiff.c * * Copyright 2018, Google Inc. * * This software may be used and distributed according to the terms of * the GNU General Public License, incorporated herein by reference. */ #include <memory> #include <stdlib.h> #include "fuzzutil.h" extern "C" { #include "bdiff.h" int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size) { auto maybe_inputs = SplitInputs(Data, Size); if (!maybe_inputs) { return 0; } auto inputs = std::move(maybe_inputs.value()); struct bdiff_line *a, *b; int an = bdiff_splitlines(inputs.left.get(), inputs.left_size, &a); int bn = bdiff_splitlines(inputs.right.get(), inputs.right_size, &b); struct bdiff_hunk l; bdiff_diff(a, an, b, bn, &l); free(a); free(b); bdiff_freehunks(l.next); return 0; // Non-zero return values are reserved for future use. } #ifdef HG_FUZZER_INCLUDE_MAIN int main(int argc, char **argv) { const char data[] = "asdf"; return LLVMFuzzerTestOneInput((const uint8_t *)data, 4); } #endif } // extern "C"