Dynamic test of byteorder when htonl was removed failed for half of the

machines.... Always assumed little endian....

git-svn-id: svn://anonsvn.mit.edu/krb5/trunk@8279 dc483132-0cff-0310-8789-dd5450dbe970

diff --git a/src/lib/crypto/sha/ChangeLog b/src/lib/crypto/sha/ChangeLog
index 6573f253c28a2593343ede4ee8b26f130219f7c0..bdc56a96246f8b474fe28f29932174637f7f0d29 100644 (file)
--- a/src/lib/crypto/sha/ChangeLog
+++ b/src/lib/crypto/sha/ChangeLog
@@ -1,3 +1,8 @@
+Sat Jun  8 07:44:35 1996  Ezra Peisach  (epeisach@mit.edu)
+
+       * shs.c (longReverse): Test for big vs little endian failed for
+               big endian machines.
+
 Thu Jun  6 15:43:26 1996  Theodore Y. Ts'o  <tytso@mit.edu>
 
        * shs.c (longReverse): Don't use htonl(); it doesn't exist under

diff --git a/src/lib/crypto/sha/shs.c b/src/lib/crypto/sha/shs.c
index 1dc134c580c2e914250f5fe679f468ec3e5a0105..be4d8fd0d5f87f985bb925c6b8cfbc93114cc01a 100644 (file)
--- a/src/lib/crypto/sha/shs.c
+++ b/src/lib/crypto/sha/shs.c
@@ -219,6 +219,7 @@ void longReverse( LONG *buffer, int byteCount )
 
     switch (init) {
     case 0:
+       init=1;
        cp = (char *) &init;
        if (*cp == 1) {
            init=2;