strcoll: Remove incorrect STRDIFF-based optimization (Bug 18589).

The optimization introduced in commit
f13c2a8dff2329c6692a80176262ceaaf8a6f74e, causes regressions in
sorting for languages that have digraphs that change sort order, like
cs_CZ which sorts ch between h and i.

My analysis shows the fast-forwarding optimization in STRCOLL advances
through a digraph while possibly stopping in the middle which results
in a subsequent skipping of the digraph and incorrect sorting. The
optimization is incorrect as implemented and because of that I'm
removing it for 2.23, and I will also commit this fix for 2.22 where
it was originally introduced.

This patch reverts the optimization, introduces a new bug-strcoll2.c
regression test that tests both cs_CZ.UTF-8 and da_DK.ISO-8859-1 and
ensures they sort one digraph each correctly. The optimization can't be
applied without regressing this test.

Checked on x86_64, bug-strcoll2.c fails without this patch and passes
after. This will also get a fix on 2.22 which has the same bug.

1 files changed, 1 insertions, 37 deletions

diff --git a/string/strcoll_l.c b/string/strcoll_l.c
index 8f1225fc7a..35bc0e42a2 100644
--- a/string/strcoll_l.c
+++ b/string/strcoll_l.c
@@ -29,7 +29,6 @@
 # define STRING_TYPE char
 # define USTRING_TYPE unsigned char
 # define STRCOLL __strcoll_l
-# define STRDIFF __strdiff
 # define STRCMP strcmp
 # define WEIGHT_H "../locale/weight.h"
 # define SUFFIX	MB
@@ -42,20 +41,6 @@
 #include "../locale/localeinfo.h"
 #include WEIGHT_H
 
-#define MASK_UTF8_7BIT  (1 << 7)
-#define MASK_UTF8_START (3 << 6)
-
-size_t
-STRDIFF (const STRING_TYPE *s, const STRING_TYPE *t)
-{
-  size_t n;
-
-  for (n = 0; *s != '\0' && *s++ == *t++; ++n)
-    continue;
-
-  return n;
-}
-
 /* Track status while looking for sequences in a string.  */
 typedef struct
 {
@@ -269,29 +254,9 @@ STRCOLL (const STRING_TYPE *s1, const STRING_TYPE *s2, __locale_t l)
   const USTRING_TYPE *extra;
   const int32_t *indirect;
 
-  /* In case there is no locale specific sort order (C / POSIX).  */
   if (nrules == 0)
     return STRCMP (s1, s2);
 
-  /* Fast forward to the position of the first difference.  Needs to be
-     encoding aware as the byte-by-byte comparison can stop in the middle
-     of a char sequence for multibyte encodings like UTF-8.  */
-  uint_fast32_t encoding =
-    current->values[_NL_ITEM_INDEX (_NL_COLLATE_ENCODING_TYPE)].word;
-  if (encoding != __cet_other)
-    {
-      size_t diff = STRDIFF (s1, s2);
-      if (diff > 0)
-	{
-	  if (encoding == __cet_utf8 && (*(s1 + diff) & MASK_UTF8_7BIT) != 0)
-	    do
-	      diff--;
-	    while (diff > 0 && (*(s1 + diff) & MASK_UTF8_START) != MASK_UTF8_START);
-	  s1 += diff;
-	  s2 += diff;
-	}
-    }
-
   /* Catch empty strings.  */
   if (__glibc_unlikely (*s1 == '\0') || __glibc_unlikely (*s2 == '\0'))
     return (*s1 != '\0') - (*s2 != '\0');
@@ -358,8 +323,7 @@ STRCOLL (const STRING_TYPE *s1, const STRING_TYPE *s2, __locale_t l)
 		     byte-level comparison to ensure that we don't waste time
 		     going through multiple passes for totally equal strings
 		     before proceeding to subsequent passes.  */
-		  if (pass == 0 && encoding == __cet_other &&
-		      STRCMP (s1, s2) == 0)
+		  if (pass == 0 && STRCMP (s1, s2) == 0)
 		    return result;
 		  else
 		    break;