Make revised search algorithm thread safe

In commit 85b1c53, I introduced a new "max_len" variable, used by the
search() function to store the maximum match length found in course of
the binary search.

However, I overlooked the fact that the search() function then lost
thread safety, as "max_len" would be shared by all threads of a process
using libbsdiff.

Fix the issue by not using a global variable, instead updating "max_len"
via another pointer, just like the "pos" variable is handled already.

Signed-off-by: Patrick McCarty <[email protected]>

Author: phmccarty

src/diff.c

@@ -88,42 +88,41 @@ static int64_t matchlen(u_char *old, int64_t oldsize, u_char *new,
 	return i;
 }
 
-int64_t max_len = 0;
-
 /**
  * Finds the longest matching array of bytes between the OLD and NEW file. The
  * old file is suffix-sorted; the suffix-sorted array is stored at I, and
- * indices to search between are indicated by ST (start) and EN (end). Returns
- * the length of the match, and POS is updated to the position of the match
- * within OLD.
+ * indices to search between are indicated by ST (start) and EN (end). The
+ * function does not return a value, but once a match is determined, POS is
+ * updated to the position of the match within OLD, and MAX_LEN is set to the
+ * match length.
  */
-static int64_t search(int64_t *I, u_char *old, int64_t oldsize,
-		      u_char *new, int64_t newsize, int64_t st, int64_t en,
-		      int64_t *pos)
+static void search(int64_t *I, u_char *old, int64_t oldsize,
+		   u_char *new, int64_t newsize, int64_t st, int64_t en,
+		   int64_t *pos, int64_t *max_len)
 {
 	int64_t x, y;
 
 	/* Initialize max_len for the binary search */
 	if (st == 0 && en == oldsize) {
-		max_len = matchlen(old, oldsize, new, newsize);
+		*max_len = matchlen(old, oldsize, new, newsize);
 		*pos = I[st];
 	}
 
 	/* The binary search terminates here when "en" and "st" are adjacent
 	 * indices in the suffix-sorted array. */
 	if (en - st < 2) {
 		x = matchlen(old + I[st], oldsize - I[st], new, newsize);
-		if (x > max_len) {
-			max_len = x;
+		if (x > *max_len) {
+			*max_len = x;
 			*pos = I[st];
 		}
 		y = matchlen(old + I[en], oldsize - I[en], new, newsize);
-		if (y > max_len) {
-			max_len = y;
+		if (y > *max_len) {
+			*max_len = y;
 			*pos = I[en];
 		}
 
-		return max_len;
+		return;
 	}
 
 	x = st + (en - st) / 2;
@@ -133,16 +132,16 @@ static int64_t search(int64_t *I, u_char *old, int64_t oldsize,
 
 	/* This match *could* be the longest one, so check for that here */
 	int64_t tmp = matchlen(oldoffset, length, new, length);
-	if (tmp > max_len) {
-		max_len = tmp;
+	if (tmp > *max_len) {
+		*max_len = tmp;
 		*pos = I[x];
 	}
 
 	/* Determine how to continue the binary search */
 	if (memcmp(oldoffset, new, length) < 0) {
-		return search(I, old, oldsize, new, newsize, x, en, pos);
+		return search(I, old, oldsize, new, newsize, x, en, pos, max_len);
 	} else {
-		return search(I, old, oldsize, new, newsize, st, x, pos);
+		return search(I, old, oldsize, new, newsize, st, x, pos, max_len);
 	}
 }
 
@@ -617,9 +616,8 @@ int make_bsdiff_delta(char *old_filename, char *new_filename, char *delta_filena
 		oldscore = 0;
 
 		for (scsc = scan += len; scan < newsize; scan++) {
-			len =
-			    search(I, old_data, oldsize, new_data + scan, newsize - scan,
-				   0, oldsize, &pos);
+			search(I, old_data, oldsize, new_data + scan, newsize - scan,
+			       0, oldsize, &pos, &len);
 
 			for (; scsc < scan + len; scsc++) {
 				if ((scsc + lastoffset < oldsize) &&