Mercurial > hg
view mercurial/mpatch.c @ 39627:ee7ee0c516ca
narrow: intersect provided matcher with narrowmatcher in `hg diff`
This provides significant speedups when running diff, and no change in behavior
that I'm aware of (or that the tests found). I tested with a repo that I started
using narrow in after it was created and attempted to run `hg diff -c .` and
similar commands in it on a commit that had files not in the narrowspec.
Timing numbers below, using a similar setup as my previous commits.
before=9db85644, m-u is mozilla-unified at eb39298e432d (flatmanifest) and
0553b7f29eaf (treemanifest). l-d-r is a repo simulating a situation I've
encountered where there's one directory with 30k+ subdirectories. N means
narrow, T means treemanifest. The narrowspec is pretty small when in use, and
importantly the narrowspec is applied *after* doing the initial checkout
(without narrowing), so all of these files exist in the filesystem, which is not
normally the case if someone has been using narrow for the entire life of the
clone.
Anything less than a 5% difference in performance is most likely noise.
diff --git:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 1.292 s +- 0.009 s | 1.295 s +- 0.010 s | 100.2%
m-u | | x | 1.296 s +- 0.042 s | 1.299 s +- 0.026 s | 100.2%
m-u | x | | 1.292 s +- 0.010 s | 1.297 s +- 0.021 s | 100.4%
m-u | x | x | 84.2 ms +- 1.2 ms | 83.6 ms +- 0.2 ms | 99.3%
l-d-r | | | 188.7 ms +- 2.7 ms | 188.8 ms +- 2.0 ms | 100.1%
l-d-r | | x | 189.9 ms +- 1.5 ms | 189.4 ms +- 1.2 ms | 99.7%
l-d-r | x | | 97.1 ms +- 1.0 ms | 87.1 ms +- 1.0 ms | 89.7% <--
l-d-r | x | x | 96.9 ms +- 0.8 ms | 87.2 ms +- 0.7 ms | 90.0% <--
diff -c . --git:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 231.6 ms +- 3.1 ms | 228.9 ms +- 1.6 ms | 98.8%
m-u | | x | 150.5 ms +- 1.7 ms | 150.7 ms +- 1.4 ms | 100.1%
m-u | x | | 233.7 ms +- 2.4 ms | 232.2 ms +- 1.9 ms | 99.4%
m-u | x | x | 126.1 ms +- 1.2 ms | 126.8 ms +- 1.2 ms | 100.6%
l-d-r | | | 82.1 ms +- 2.0 ms | 81.8 ms +- 1.4 ms | 99.6%
l-d-r | | x | 3.732 s +- 0.020 s | 3.746 s +- 0.027 s | 100.4%
l-d-r | x | | 83.1 ms +- 0.8 ms | 107.6 ms +- 2.4 ms | 129.5% <--
l-d-r | x | x | 758.2 ms +- 38.8 ms | 188.5 ms +- 1.8 ms | 24.9% <--
rebase -r . --keep -d .^^:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 5.532 s +- 0.087 s | 5.496 s +- 0.016 s | 99.3%
m-u | | x | 5.554 s +- 0.061 s | 5.532 s +- 0.013 s | 99.6%
m-u | x | | 5.602 s +- 0.134 s | 5.508 s +- 0.035 s | 98.3%
m-u | x | x | 582.2 ms +- 15.2 ms | 572.9 ms +- 12.0 ms | 98.4%
l-d-r | | | 629.5 ms +- 12.3 ms | 622.5 ms +- 7.3 ms | 98.9%
l-d-r | | x | 6.173 s +- 0.062 s | 6.185 s +- 0.076 s | 100.2%
l-d-r | x | | 274.5 ms +- 10.0 ms | 272.1 ms +- 6.2 ms | 99.1%
l-d-r | x | x | 4.835 s +- 0.056 s | 4.826 s +- 0.034 s | 99.8%
status --change . --copies:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 214.4 ms +- 1.4 ms | 212.2 ms +- 1.7 ms | 99.0%
m-u | | x | 130.9 ms +- 1.2 ms | 131.7 ms +- 1.1 ms | 100.6%
m-u | x | | 215.0 ms +- 2.1 ms | 214.9 ms +- 2.7 ms | 100.0%
m-u | x | x | 109.5 ms +- 2.3 ms | 107.8 ms +- 0.9 ms | 98.4%
l-d-r | | | 79.6 ms +- 0.9 ms | 79.8 ms +- 1.6 ms | 100.3%
l-d-r | | x | 3.799 s +- 0.037 s | 3.928 s +- 0.021 s | 103.4% <--?
l-d-r | x | | 82.7 ms +- 0.7 ms | 83.2 ms +- 1.0 ms | 100.6%
l-d-r | x | x | 746.8 ms +- 6.1 ms | 739.0 ms +- 4.2 ms | 99.0%
status --copies:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 1.884 s +- 0.012 s | 1.885 s +- 0.013 s | 100.1%
m-u | | x | 1.897 s +- 0.027 s | 1.909 s +- 0.077 s | 100.6%
m-u | x | | 1.886 s +- 0.021 s | 1.891 s +- 0.030 s | 100.3%
m-u | x | x | 92.0 ms +- 0.7 ms | 92.4 ms +- 0.4 ms | 100.4%
l-d-r | | | 570.3 ms +- 18.7 ms | 552.2 ms +- 4.5 ms | 96.8%
l-d-r | | x | 568.9 ms +- 16.1 ms | 567.2 ms +- 11.9 ms | 99.7%
l-d-r | x | | 171.1 ms +- 2.5 ms | 170.4 ms +- 1.2 ms | 99.6%
l-d-r | x | x | 171.6 ms +- 3.4 ms | 171.5 ms +- 1.7 ms | 99.9%
update $rev^; ~/src/hg/hg{hg}/hg update $rev:
repo | N | T | before (mean +- stdev) | after (mean +- stdev) | % of before
------+---+---+------------------------+-----------------------+------------
m-u | | | 3.107 s +- 0.017 s | 3.116 s +- 0.012 s | 100.3%
m-u | | x | 2.943 s +- 0.010 s | 2.945 s +- 0.019 s | 100.1%
m-u | x | | 3.116 s +- 0.033 s | 3.118 s +- 0.027 s | 100.1%
m-u | x | x | 318.5 ms +- 2.7 ms | 320.8 ms +- 4.8 ms | 100.7%
l-d-r | | | 428.9 ms +- 4.4 ms | 429.5 ms +- 4.0 ms | 100.1%
l-d-r | | x | 9.593 s +- 0.081 s | 9.869 s +- 0.043 s | 102.9%
l-d-r | x | | 253.2 ms +- 3.6 ms | 254.0 ms +- 2.8 ms | 100.3%
l-d-r | x | x | 1.613 s +- 0.009 s | 1.630 s +- 0.017 s | 101.1%
Differential Revision: https://phab.mercurial-scm.org/D4587
author | spectral <spectral@google.com> |
---|---|
date | Thu, 13 Sep 2018 22:48:27 -0700 |
parents | 9c5ced5276d6 |
children | 763b45bc4483 |
line wrap: on
line source
/* mpatch.c - efficient binary patching for Mercurial This implements a patch algorithm that's O(m + nlog n) where m is the size of the output and n is the number of patches. Given a list of binary patches, it unpacks each into a hunk list, then combines the hunk lists with a treewise recursion to form a single hunk list. This hunk list is then applied to the original text. The text (or binary) fragments are copied directly from their source Python objects into a preallocated output string to avoid the allocation of intermediate Python objects. Working memory is about 2x the total number of hunks. Copyright 2005, 2006 Matt Mackall <mpm@selenic.com> This software may be used and distributed according to the terms of the GNU General Public License, incorporated herein by reference. */ #include <limits.h> #include <stdlib.h> #include <string.h> #include "bitmanipulation.h" #include "compat.h" #include "mpatch.h" /* VC9 doesn't include bool and lacks stdbool.h based on cext/util.h */ #if defined(_MSC_VER) || __STDC_VERSION__ < 199901L #define true 1 #define false 0 typedef unsigned char bool; #else #include <stdbool.h> #endif static struct mpatch_flist *lalloc(ssize_t size) { struct mpatch_flist *a = NULL; if (size < 1) size = 1; a = (struct mpatch_flist *)malloc(sizeof(struct mpatch_flist)); if (a) { a->base = (struct mpatch_frag *)malloc( sizeof(struct mpatch_frag) * size); if (a->base) { a->head = a->tail = a->base; return a; } free(a); } return NULL; } void mpatch_lfree(struct mpatch_flist *a) { if (a) { free(a->base); free(a); } } static ssize_t lsize(struct mpatch_flist *a) { return a->tail - a->head; } /* add helper to add src and *dest iff it won't overflow */ static inline bool safeadd(int src, int *dest) { if ((src > 0) == (*dest > 0)) { if (*dest > 0) { if (src > (INT_MAX - *dest)) { return false; } } else { if (src < (INT_MIN - *dest)) { return false; } } } *dest += src; return true; } /* subtract src from dest and store result in dest */ static inline bool safesub(int src, int *dest) { if (((src > 0) && (*dest < INT_MIN + src)) || ((src < 0) && (*dest > INT_MAX + src))) { return false; } *dest -= src; return true; } /* move hunks in source that are less cut to dest, compensating for changes in offset. the last hunk may be split if necessary. */ static int gather(struct mpatch_flist *dest, struct mpatch_flist *src, int cut, int offset) { struct mpatch_frag *d = dest->tail, *s = src->head; int postend, c, l; while (s != src->tail) { int soffset = s->start; if (!safeadd(offset, &soffset)) break; /* add would overflow, oh well */ if (soffset >= cut) break; /* we've gone far enough */ postend = offset; if (!safeadd(s->start, &postend) || !safeadd(s->len, &postend)) { break; } if (postend <= cut) { /* save this hunk */ int tmp = s->start; if (!safesub(s->end, &tmp)) { break; } if (!safeadd(s->len, &tmp)) { break; } if (!safeadd(tmp, &offset)) { break; /* add would overflow, oh well */ } *d++ = *s++; } else { /* break up this hunk */ c = cut; if (!safesub(offset, &c)) { break; } if (s->end < c) c = s->end; l = cut - offset - s->start; if (s->len < l) l = s->len; offset += s->start + l - c; d->start = s->start; d->end = c; d->len = l; d->data = s->data; d++; s->start = c; s->len = s->len - l; s->data = s->data + l; break; } } dest->tail = d; src->head = s; return offset; } /* like gather, but with no output list */ static int discard(struct mpatch_flist *src, int cut, int offset) { struct mpatch_frag *s = src->head; int postend, c, l; while (s != src->tail) { int cmpcut = s->start; if (!safeadd(offset, &cmpcut)) { break; } if (cmpcut >= cut) break; postend = offset; if (!safeadd(s->start, &postend)) { break; } if (!safeadd(s->len, &postend)) { break; } if (postend <= cut) { /* do the subtraction first to avoid UB integer overflow */ int tmp = s->start; if (!safesub(s->end, &tmp)) { break; } if (!safeadd(s->len, &tmp)) { break; } if (!safeadd(tmp, &offset)) { break; } s++; } else { c = cut; if (!safesub(offset, &c)) { break; } if (s->end < c) c = s->end; l = cut - offset - s->start; if (s->len < l) l = s->len; offset += s->start + l - c; s->start = c; s->len = s->len - l; s->data = s->data + l; break; } } src->head = s; return offset; } /* combine hunk lists a and b, while adjusting b for offset changes in a/ this deletes a and b and returns the resultant list. */ static struct mpatch_flist *combine(struct mpatch_flist *a, struct mpatch_flist *b) { struct mpatch_flist *c = NULL; struct mpatch_frag *bh, *ct; int offset = 0, post; if (a && b) c = lalloc((lsize(a) + lsize(b)) * 2); if (c) { for (bh = b->head; bh != b->tail; bh++) { /* save old hunks */ offset = gather(c, a, bh->start, offset); /* discard replaced hunks */ post = discard(a, bh->end, offset); /* insert new hunk */ ct = c->tail; ct->start = bh->start; ct->end = bh->end; if (!safesub(offset, &(ct->start)) || !safesub(post, &(ct->end))) { /* It was already possible to exit * this function with a return value * of NULL before the safesub()s were * added, so this should be fine. */ mpatch_lfree(c); c = NULL; goto done; } ct->len = bh->len; ct->data = bh->data; c->tail++; offset = post; } /* hold on to tail from a */ memcpy(c->tail, a->head, sizeof(struct mpatch_frag) * lsize(a)); c->tail += lsize(a); } done: mpatch_lfree(a); mpatch_lfree(b); return c; } /* decode a binary patch into a hunk list */ int mpatch_decode(const char *bin, ssize_t len, struct mpatch_flist **res) { struct mpatch_flist *l; struct mpatch_frag *lt; int pos = 0; /* assume worst case size, we won't have many of these lists */ l = lalloc(len / 12 + 1); if (!l) return MPATCH_ERR_NO_MEM; lt = l->tail; /* We check against len-11 to ensure we have at least 12 bytes left in the patch so we can read our three be32s out of it. */ while (pos >= 0 && pos < (len - 11)) { lt->start = getbe32(bin + pos); lt->end = getbe32(bin + pos + 4); lt->len = getbe32(bin + pos + 8); if (lt->start < 0 || lt->start > lt->end || lt->len < 0) break; /* sanity check */ if (!safeadd(12, &pos)) { break; } lt->data = bin + pos; if (!safeadd(lt->len, &pos)) { break; } lt++; } if (pos != len) { mpatch_lfree(l); return MPATCH_ERR_CANNOT_BE_DECODED; } l->tail = lt; *res = l; return 0; } /* calculate the size of resultant text */ ssize_t mpatch_calcsize(ssize_t len, struct mpatch_flist *l) { ssize_t outlen = 0, last = 0; struct mpatch_frag *f = l->head; while (f != l->tail) { if (f->start < last || f->end > len) { return MPATCH_ERR_INVALID_PATCH; } outlen += f->start - last; last = f->end; outlen += f->len; f++; } outlen += len - last; return outlen; } int mpatch_apply(char *buf, const char *orig, ssize_t len, struct mpatch_flist *l) { struct mpatch_frag *f = l->head; int last = 0; char *p = buf; while (f != l->tail) { if (f->start < last || f->start > len || f->end > len || last < 0) { return MPATCH_ERR_INVALID_PATCH; } memcpy(p, orig + last, f->start - last); p += f->start - last; memcpy(p, f->data, f->len); last = f->end; p += f->len; f++; } if (last < 0) { return MPATCH_ERR_INVALID_PATCH; } memcpy(p, orig + last, len - last); return 0; } /* recursively generate a patch of all bins between start and end */ struct mpatch_flist * mpatch_fold(void *bins, struct mpatch_flist *(*get_next_item)(void *, ssize_t), ssize_t start, ssize_t end) { ssize_t len; if (start + 1 == end) { /* trivial case, output a decoded list */ return get_next_item(bins, start); } /* divide and conquer, memory management is elsewhere */ len = (end - start) / 2; return combine(mpatch_fold(bins, get_next_item, start, start + len), mpatch_fold(bins, get_next_item, start + len, end)); }