@matheusbsbs / matheus-bagliano / commits / 663c5ad035

diff histogram: intern strings

Histogram is the only diff algorithm not to call xdl_classify_record(). xdl_classify_record() ensures that the hash values of two strings that are not equal differ which means that it is not necessary to use xdl_recmatch() when comparing lines, all that is necessary is to compare the hash values. This gives a 7% reduction in the runtime of "git log --patch" when using the histogram diff algorithm. Test HEAD^ HEAD ----------------------------------------------------------------------------- 4000.1: log -3000 (baseline) 0.18(0.14+0.04) 0.19(0.17+0.02) +5.6% 4000.2: log --raw -3000 (tree-only) 0.99(0.77+0.21) 0.98(0.78+0.20) -1.0% 4000.3: log -p -3000 (Myers) 4.84(4.31+0.51) 4.81(4.15+0.64) -0.6% 4000.4: log -p -3000 --histogram 6.34(5.86+0.46) 5.87(5.19+0.66) -7.4% 4000.5: log -p -3000 --patience 5.39(4.60+0.76) 5.35(4.60+0.73) -0.7% Signed-off-by: Phillip Wood <phillip.wood@dunelm.org.uk> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Phillip Wood committed Nov 17, 2021 at 11:20 UTC 663c5ad0356f5501e2db7a39a7abf76e8db26b7d
2 files changed +10 -19
xdiff/xhistogram.c
+2 -3
@@ -91,9 +91,8 @@ struct region {
91 static int cmp_recs(xpparam_t const *xpp,
92 xrecord_t *r1, xrecord_t *r2)
93 {
94 - return r1->ha == r2->ha &&
95 - xdl_recmatch(r1->ptr, r1->size, r2->ptr, r2->size,
96 - xpp->flags);
94 + return r1->ha == r2->ha;
95 +
96 }
97
98 #define CMP_ENV(xpp, env, s1, l1, s2, l2) \
xdiff/xprepare.c
+8 -16
@@ -181,15 +181,11 @@ static int xdl_prepare_ctx(unsigned int pass, mmfile_t *mf, long narec, xpparam_
181 if (!(recs = (xrecord_t **) xdl_malloc(narec * sizeof(xrecord_t *))))
182 goto abort;
183
184 - if (XDF_DIFF_ALG(xpp->flags) == XDF_HISTOGRAM_DIFF)
185 - hbits = hsize = 0;
186 - else {
187 - hbits = xdl_hashbits((unsigned int) narec);
188 - hsize = 1 << hbits;
189 - if (!(rhash = (xrecord_t **) xdl_malloc(hsize * sizeof(xrecord_t *))))
190 - goto abort;
191 - memset(rhash, 0, hsize * sizeof(xrecord_t *));
192 - }
184 + hbits = xdl_hashbits((unsigned int) narec);
185 + hsize = 1 << hbits;
186 + if (!(rhash = (xrecord_t **) xdl_malloc(hsize * sizeof(xrecord_t *))))
187 + goto abort;
188 + memset(rhash, 0, hsize * sizeof(xrecord_t *));
189
190 nrec = 0;
191 if ((cur = blk = xdl_mmfile_first(mf, &bsize)) != NULL) {
@@ -208,9 +204,7 @@ static int xdl_prepare_ctx(unsigned int pass, mmfile_t *mf, long narec, xpparam_
204 crec->size = (long) (cur - prev);
205 crec->ha = hav;
206 recs[nrec++] = crec;
211 -
212 - if ((XDF_DIFF_ALG(xpp->flags) != XDF_HISTOGRAM_DIFF) &&
213 - xdl_classify_record(pass, cf, rhash, hbits, crec) < 0)
207 + if (xdl_classify_record(pass, cf, rhash, hbits, crec) < 0)
208 goto abort;
209 }
210 }
@@ -279,8 +273,7 @@ int xdl_prepare_env(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
273 enl1 = xdl_guess_lines(mf1, sample) + 1;
274 enl2 = xdl_guess_lines(mf2, sample) + 1;
275
282 - if (XDF_DIFF_ALG(xpp->flags) != XDF_HISTOGRAM_DIFF &&
283 - xdl_init_classifier(&cf, enl1 + enl2 + 1, xpp->flags) < 0)
276 + if (xdl_init_classifier(&cf, enl1 + enl2 + 1, xpp->flags) < 0)
277 return -1;
278
279 if (xdl_prepare_ctx(1, mf1, enl1, xpp, &cf, &xe->xdf1) < 0) {
@@ -305,8 +298,7 @@ int xdl_prepare_env(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
298 return -1;
299 }
300
308 - if (XDF_DIFF_ALG(xpp->flags) != XDF_HISTOGRAM_DIFF)
309 - xdl_free_classifier(&cf);
301 + xdl_free_classifier(&cf);
302
303 return 0;
304 }