dev-util/bsdiff/files/4.3_bsdiff-divsufsort.patch

Make bsdiff use libdivsufsort + tiny / cosmetic fixes.
--- bsdiff.1
+++ bsdiff.1
@@ -33,20 +33,21 @@
 .Nd generate a patch between two binary files
 .Sh SYNOPSIS
 .Nm
-.Ao Ar oldfile Ac Ao Ar newfile Ac Ao Ar patchfile Ac
+.Ar oldfile newfile patchfile
 .Sh DESCRIPTION
 .Nm
 compares
-.Ao Ar oldfile Ac
+.Ar oldfile
 to
-.Ao Ar newfile Ac
+.Ar newfile
 and writes to
-.Ao Ar patchfile Ac
-a binary patch suitable for use by bspatch(1).
+.Ar patchfile
+a binary patch suitable for use by
+.Xr bspatch 1 .
 When
-.Ao Ar oldfile Ac
+.Ar oldfile
 and
-.Ao Ar newfile Ac
+.Ar newfile
 are two versions of an executable program, the
 patches produced are on average a factor of five smaller
 than those produced by any other binary patch tool known
@@ -54,7 +55,7 @@ to the author.
 .Pp
 .Nm
 uses memory equal to 17 times the size of 
-.Ao Ar oldfile Ac ,
+.Ar oldfile ,
 and requires
 an absolute minimum working set size of 8 times the size of oldfile.
 .Sh SEE ALSO
--- bsdiff.c
+++ bsdiff.c
@@ -38,106 +38,15 @@ __FBSDID("$FreeBSD: src/usr.bin/bsdiff/bsdiff/bsdiff.c,v 1.1 2005/08/06 01:59:05
 #include <string.h>
 #include <unistd.h>
 
-#define MIN(x,y) (((x)<(y)) ? (x) : (y))
-
-static void split(off_t *I,off_t *V,off_t start,off_t len,off_t h)
-{
-	off_t i,j,k,x,tmp,jj,kk;
-
-	if(len<16) {
-		for(k=start;k<start+len;k+=j) {
-			j=1;x=V[I[k]+h];
-			for(i=1;k+i<start+len;i++) {
-				if(V[I[k+i]+h]<x) {
-					x=V[I[k+i]+h];
-					j=0;
-				};
-				if(V[I[k+i]+h]==x) {
-					tmp=I[k+j];I[k+j]=I[k+i];I[k+i]=tmp;
-					j++;
-				};
-			};
-			for(i=0;i<j;i++) V[I[k+i]]=k+j-1;
-			if(j==1) I[k]=-1;
-		};
-		return;
-	};
-
-	x=V[I[start+len/2]+h];
-	jj=0;kk=0;
-	for(i=start;i<start+len;i++) {
-		if(V[I[i]+h]<x) jj++;
-		if(V[I[i]+h]==x) kk++;
-	};
-	jj+=start;kk+=jj;
-
-	i=start;j=0;k=0;
-	while(i<jj) {
-		if(V[I[i]+h]<x) {
-			i++;
-		} else if(V[I[i]+h]==x) {
-			tmp=I[i];I[i]=I[jj+j];I[jj+j]=tmp;
-			j++;
-		} else {
-			tmp=I[i];I[i]=I[kk+k];I[kk+k]=tmp;
-			k++;
-		};
-	};
-
-	while(jj+j<kk) {
-		if(V[I[jj+j]+h]==x) {
-			j++;
-		} else {
-			tmp=I[jj+j];I[jj+j]=I[kk+k];I[kk+k]=tmp;
-			k++;
-		};
-	};
-
-	if(jj>start) split(I,V,start,jj-start,h);
-
-	for(i=0;i<kk-jj;i++) V[I[jj+i]]=kk-1;
-	if(jj==kk-1) I[jj]=-1;
-
-	if(start+len>kk) split(I,V,kk,start+len-kk,h);
-}
-
-static void qsufsort(off_t *I,off_t *V,u_char *old,off_t oldsize)
-{
-	off_t buckets[256];
-	off_t i,h,len;
-
-	for(i=0;i<256;i++) buckets[i]=0;
-	for(i=0;i<oldsize;i++) buckets[old[i]]++;
-	for(i=1;i<256;i++) buckets[i]+=buckets[i-1];
-	for(i=255;i>0;i--) buckets[i]=buckets[i-1];
-	buckets[0]=0;
-
-	for(i=0;i<oldsize;i++) I[++buckets[old[i]]]=i;
-	I[0]=oldsize;
-	for(i=0;i<oldsize;i++) V[i]=buckets[old[i]];
-	V[oldsize]=0;
-	for(i=1;i<256;i++) if(buckets[i]==buckets[i-1]+1) I[buckets[i]]=-1;
-	I[0]=-1;
-
-	for(h=1;I[0]!=-(oldsize+1);h+=h) {
-		len=0;
-		for(i=0;i<oldsize+1;) {
-			if(I[i]<0) {
-				len-=I[i];
-				i-=I[i];
-			} else {
-				if(len) I[i-len]=-len;
-				len=V[I[i]]+1-i;
-				split(I,V,i,len,h);
-				i+=len;
-				len=0;
-			};
-		};
-		if(len) I[i-len]=-len;
-	};
+#if _FILE_OFFSET_BITS == 64
+#include "divsufsort64.h"
+#define saidx_t saidx64_t
+#define divsufsort divsufsort64
+#else
+#include "divsufsort.h"
+#endif
 
-	for(i=0;i<oldsize+1;i++) I[V[i]]=i;
-}
+#define MIN(x,y) (((x)<(y)) ? (x) : (y))
 
 static off_t matchlen(u_char *old,off_t oldsize,u_char *new,off_t newsize)
 {
@@ -149,7 +58,7 @@ static off_t matchlen(u_char *old,off_t oldsize,u_char *new,off_t newsize)
 	return i;
 }
 
-static off_t search(off_t *I,u_char *old,off_t oldsize,
+static off_t search(saidx_t *I,u_char *old,off_t oldsize,
 		u_char *new,off_t newsize,off_t st,off_t en,off_t *pos)
 {
 	off_t x,y;
@@ -168,7 +77,7 @@ static off_t search(off_t *I,u_char *old,off_t oldsize,
 	};
 
 	x=st+(en-st)/2;
-	if(memcmp(old+I[x],new,MIN(oldsize-I[x],newsize))<0) {
+	if(memcmp(old+I[x],new,MIN(oldsize-I[x],newsize))<=0) {
 		return search(I,old,oldsize,new,newsize,x,en,pos);
 	} else {
 		return search(I,old,oldsize,new,newsize,st,x,pos);
@@ -198,8 +107,8 @@ int main(int argc,char *argv[])
 	int fd;
 	u_char *old,*new;
 	off_t oldsize,newsize;
-	off_t *I,*V;
-	off_t scan,pos,len;
+	saidx_t *I;
+	off_t scan,pos=0,len;
 	off_t lastscan,lastpos,lastoffset;
 	off_t oldscore,scsc;
 	off_t s,Sf,lenf,Sb,lenb;
@@ -224,12 +133,9 @@ int main(int argc,char *argv[])
 		(read(fd,old,oldsize)!=oldsize) ||
 		(close(fd)==-1)) err(1,"%s",argv[1]);
 
-	if(((I=malloc((oldsize+1)*sizeof(off_t)))==NULL) ||
-		((V=malloc((oldsize+1)*sizeof(off_t)))==NULL)) err(1,NULL);
-
-	qsufsort(I,V,old,oldsize);
+	if(((I=malloc((oldsize+1)*sizeof(saidx_t)))==NULL)) err(1,NULL);
 
-	free(V);
+	if(divsufsort(old, I, oldsize)) err(1, "divsufsort");
 
 	/* Allocate newsize+1 bytes instead of newsize bytes to ensure
 		that we never try to malloc(0) and get a NULL pointer */
@@ -274,7 +180,17 @@ int main(int argc,char *argv[])
 	while(scan<newsize) {
 		oldscore=0;
 
+		/* If we come across a large block of data that only differs
+		 * by less than 8 bytes, this loop will take a long time to
+		 * go past that block of data. We need to track the number of
+		 * times we're stuck in the block and break out of it. */
+		int num_less_than_eight = 0;
+		off_t prev_len, prev_oldscore, prev_pos;
 		for(scsc=scan+=len;scan<newsize;scan++) {
+			prev_len=len;
+			prev_oldscore=oldscore;
+			prev_pos=pos;
+
 			len=search(I,old,oldsize,new+scan,newsize-scan,
 					0,oldsize,&pos);
 
@@ -289,6 +205,17 @@ int main(int argc,char *argv[])
 			if((scan+lastoffset<oldsize) &&
 				(old[scan+lastoffset] == new[scan]))
 				oldscore--;
+
+			const off_t fuzz = 8;
+			if (prev_len-fuzz<=len && len<=prev_len &&
+			    prev_oldscore-fuzz<=oldscore &&
+			    oldscore<=prev_oldscore &&
+			    prev_pos<=pos && pos <=prev_pos+fuzz &&
+			    oldscore<=len && len<=oldscore+fuzz)
+				++num_less_than_eight;
+			else
+				num_less_than_eight=0;
+			if (num_less_than_eight > 100) break;
 		};
 
 		if((len!=oldscore) || (scan==newsize)) {