xdiff/xprepare: use a smaller sample size for histogram diff
For histogram diff, we can afford a smaller sample size and thus a poorer estimate of the number of lines, as the hash table (rhash) won't be filled up/grown. This is safe as the final count of lines (xdf.nrecs) will be updated correctly anyway by xdl_prepare_ctx(). This gives us a small boost in performance. Signed-off-by: Tay Ray Chuan <rctay89@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Tay Ray Chuan committed
Jul 12, 2011 at 14:10 UTC
86abba801575892a8a2b161aa29518c1ed05e1f1
3 files changed
+17
-10
xdiff/xprepare.c
+14
-3
@@ -26,6 +26,8 @@
26
#define XDL_KPDIS_RUN 4
27
#define XDL_MAX_EQLIMIT 1024
28
#define XDL_SIMSCAN_WINDOW 100
29
+#define XDL_GUESS_NLINES1 256
30
+#define XDL_GUESS_NLINES2 20
31
32
33
typedef struct s_xdlclass {
@@ -239,11 +241,20 @@ static void xdl_free_ctx(xdfile_t *xdf) {
241
242
int xdl_prepare_env(mmfile_t *mf1, mmfile_t *mf2, xpparam_t const *xpp,
243
xdfenv_t *xe) {
242
- long enl1, enl2;
244
+ long enl1, enl2, sample;
245
xdlclassifier_t cf;
246
245
- enl1 = xdl_guess_lines(mf1) + 1;
246
- enl2 = xdl_guess_lines(mf2) + 1;
247
+ /*
248
+ * For histogram diff, we can afford a smaller sample size and
249
+ * thus a poorer estimate of the number of lines, as the hash
250
+ * table (rhash) won't be filled up/grown. The number of lines
251
+ * (nrecs) will be updated correctly anyway by
252
+ * xdl_prepare_ctx().
253
+ */
254
+ sample = xpp->flags & XDF_HISTOGRAM_DIFF ? XDL_GUESS_NLINES2 : XDL_GUESS_NLINES1;
255
+
256
+ enl1 = xdl_guess_lines(mf1, sample) + 1;
257
+ enl2 = xdl_guess_lines(mf2, sample) + 1;
258
259
if (!(xpp->flags & XDF_HISTOGRAM_DIFF) &&
260
xdl_init_classifier(&cf, enl1 + enl2 + 1, xpp->flags) < 0) {
xdiff/xutils.c
+2
-6
@@ -24,10 +24,6 @@
24
25
26
27
-#define XDL_GUESS_NLINES 256
28
-
29
-
30
-
27
28
long xdl_bogosqrt(long n) {
29
long i;
@@ -159,12 +155,12 @@ void *xdl_cha_next(chastore_t *cha) {
155
}
156
157
162
-long xdl_guess_lines(mmfile_t *mf) {
158
+long xdl_guess_lines(mmfile_t *mf, long sample) {
159
long nl = 0, size, tsize = 0;
160
char const *data, *cur, *top;
161
162
if ((cur = data = xdl_mmfile_first(mf, &size)) != NULL) {
167
- for (top = data + size; nl < XDL_GUESS_NLINES;) {
163
+ for (top = data + size; nl < sample;) {
164
if (cur >= top) {
165
tsize += (long) (cur - data);
166
if (!(cur = data = xdl_mmfile_next(mf, &size)))
xdiff/xutils.h
+1
-1
@@ -33,7 +33,7 @@ void xdl_cha_free(chastore_t *cha);
33
void *xdl_cha_alloc(chastore_t *cha);
34
void *xdl_cha_first(chastore_t *cha);
35
void *xdl_cha_next(chastore_t *cha);
36
-long xdl_guess_lines(mmfile_t *mf);
36
+long xdl_guess_lines(mmfile_t *mf, long sample);
37
int xdl_recmatch(const char *l1, long s1, const char *l2, long s2, long flags);
38
unsigned long xdl_hash_record(char const **data, char const *top, long flags);
39
unsigned int xdl_hashbits(unsigned int size);