X-Git-Url: https://mattmccutchen.net/rsync/rsync.git/blobdiff_plain/ba582f753a4c3eb2dd793904fb6a378ed8ffd90a..bb6721dce6bac8ff2374609bf5f99ea7e6fe2c70:/match.c

diff --git a/match.c b/match.c
index 1eecf3c3..f3858e5b 100644
--- a/match.c
+++ b/match.c
@@ -23,6 +23,7 @@ extern int verbose;
 extern int am_server;
 extern int do_progress;
 extern int checksum_seed;
+extern int inplace;
 
 typedef unsigned short tag;
 
@@ -142,14 +143,14 @@ static void hash_search(int f,struct sum_struct *s,
 {
 	OFF_T offset, end;
 	unsigned int k;
-	size_t last_i;
+	size_t want_i;
 	char sum2[SUM_LENGTH];
 	uint32 s1, s2, sum;
 	schar *map;
 
-	/* last_i is used to encourage adjacent matches, allowing the RLL coding of the
-	   output to work more efficiently */
-	last_i = (size_t)-1;
+	/* want_i is used to encourage adjacent matches, allowing the RLL
+	 * coding of the output to work more efficiently. */
+	want_i = 0;
 
 	if (verbose > 2) {
 		rprintf(FINFO,"hash search b=%u len=%.0f\n",
@@ -188,7 +189,7 @@ static void hash_search(int f,struct sum_struct *s,
 
 		sum = (s1 & 0xffff) | (s2 << 16);
 		tag_hits++;
-		for (; j < s->count && targets[j].t == t; j++) {
+		do {
 			unsigned int l;
 			size_t i = targets[j].i;
 
@@ -200,6 +201,12 @@ static void hash_search(int f,struct sum_struct *s,
 			if (l != s->sums[i].len)
 				continue;
 
+			/* inplace: ensure chunk's offset is either >= our
+			 * offset or that the data didn't move. */
+			if (inplace && s->sums[i].offset < offset
+			    && !(s->sums[i].flags & SUMFLG_SAME_OFFSET))
+				continue;
+
 			if (verbose > 3)
 				rprintf(FINFO,"potential match at %.0f target=%.0f %.0f sum=%08x\n",
 					(double)offset,(double)j,(double)i,sum);
@@ -215,23 +222,42 @@ static void hash_search(int f,struct sum_struct *s,
 				continue;
 			}
 
-			/* we've found a match, but now check to see
-			 * if last_i can hint at a better match */
-			for (j++; j < s->count && targets[j].t == t; j++) {
-				size_t i2 = targets[j].i;
-				if (i2 == last_i + 1) {
-					if (sum != s->sums[i2].sum1)
-						break;
-					if (memcmp(sum2,s->sums[i2].sum2,s->s2length) != 0)
-						break;
-					/* we've found an adjacent match - the RLL coder
-					 * will be happy */
-					i = i2;
-					break;
-				}
+			/* If inplace is enabled, the best possible match is
+			 * one with an identical offset, so we prefer that over
+			 * the following want_i optimization. */
+			if (inplace) {
+				do {
+					size_t i2 = targets[j].i;
+					if (s->sums[i2].offset != offset)
+						continue;
+					if (i2 != i) {
+						if (sum != s->sums[i2].sum1)
+							break;
+						if (memcmp(sum2, s->sums[i2].sum2,
+							   s->s2length) != 0)
+							break;
+						i = i2;
+					}
+					/* This chunk was at the same offset on
+					 * both the sender and the receiver. */
+					s->sums[i].flags |= SUMFLG_SAME_OFFSET;
+					goto set_want_i;
+				} while (++j < s->count && targets[j].t == t);
 			}
 
-			last_i = i;
+			/* we've found a match, but now check to see
+			 * if want_i can hint at a better match. */
+			if (i != want_i && want_i < s->count
+			    && (!inplace || s->sums[want_i].offset >= offset
+			     || s->sums[want_i].flags & SUMFLG_SAME_OFFSET)
+			    && sum == s->sums[want_i].sum1
+			    && memcmp(sum2, s->sums[want_i].sum2, s->s2length) == 0) {
+				/* we've found an adjacent match - the RLL coder
+				 * will be happy */
+				i = want_i;
+			}
+		    set_want_i:
+			want_i = i + 1;
 
 			matched(f,s,buf,offset,i);
 			offset += s->sums[i].len - 1;
@@ -242,7 +268,7 @@ static void hash_search(int f,struct sum_struct *s,
 			s2 = sum >> 16;
 			matches++;
 			break;
-		}
+		} while (++j < s->count && targets[j].t == t);
 
 	null_tag:
 		/* Trim off the first byte from the checksum */
@@ -292,7 +318,6 @@ static void hash_search(int f,struct sum_struct *s,
 void match_sums(int f, struct sum_struct *s, struct map_struct *buf, OFF_T len)
 {
 	char file_sum[MD4_SUM_LENGTH];
-	extern int write_batch;
 
 	last_match = 0;
 	false_alarms = 0;
@@ -330,8 +355,6 @@ void match_sums(int f, struct sum_struct *s, struct map_struct *buf, OFF_T len)
 	if (verbose > 2)
 		rprintf(FINFO,"sending file_sum\n");
 	write_buf(f,file_sum,MD4_SUM_LENGTH);
-	if (write_batch)
-		write_batch_delta_file(file_sum, MD4_SUM_LENGTH);
 
 	if (targets) {
 		free(targets);