* Makefile.am (check-DEJAGNU): New target.

	(site.exp): Likewise.
	* acinclude.m4 (NEWLIB_CONFIGURE): Replace AC_CANONICAL_HOST
	with AC_CANONICAL_SYSTEM.  Remove AC_CANONICAL_BUILD.
	* libc/locale/locale.c (_setlocale_r): Add UTF-8 support.
	* libc/stdlib/mbtowc_r.c (_mbtowc_r): Likewise.
	* libc/stdlib/wctomb_r.c (_wctomb_r): Likewise.
	* testsuite: New directory.
	* testsuite/config: Likewise.
	* testsuite/lib: Likewise.
	* testsuite/newlib.locale: Likewise.
	* testsuite/newlib.string: Likewise.
	* testsuite/config/default.exp: New file.
	* testsuite/lib/checkoutput.exp: New file.
	* testsuite/lib/newlib.exp: New file.
	* testsuite/lib/passfail.exp: New file.
	* testsuite/newlib.locale/UTF-8.c: New file.
	* testsuite/newlib.locale/UTF-8.exp: New file.
	* testsuite/newlib.locale/locale.exp: New file.
	* testsuite/newlib.string/string.exp: New file.
	* testsuite/newlib.string/tstring.c: New file.

3 files changed, 248 insertions, 0 deletions

diff --git a/newlib/libc/stdlib/Makefile.in b/newlib/libc/stdlib/Makefile.in
index 8ae3d154d..b29bfe6f5 100644
--- a/newlib/libc/stdlib/Makefile.in
+++ b/newlib/libc/stdlib/Makefile.in
@@ -58,8 +58,12 @@ POST_INSTALL = :
 NORMAL_UNINSTALL = :
 PRE_UNINSTALL = :
 POST_UNINSTALL = :
+build_alias = @build_alias@
+build_triplet = @build@
 host_alias = @host_alias@
 host_triplet = @host@
+target_alias = @target_alias@
+target_triplet = @target@
 AR = @AR@
 AS = @AS@
 CC = @CC@
diff --git a/newlib/libc/stdlib/mbtowc_r.c b/newlib/libc/stdlib/mbtowc_r.c
index 4bf302359..866789f16 100644
--- a/newlib/libc/stdlib/mbtowc_r.c
+++ b/newlib/libc/stdlib/mbtowc_r.c
@@ -66,6 +66,193 @@ _DEFUN (_mbtowc_r, (r, pwc, s, n, state),
   if (r->_current_locale == NULL ||
       (strlen (r->_current_locale) <= 1))
     { /* fall-through */ }
+  else if (!strcmp (r->_current_locale, "UTF-8"))
+    {
+      wchar_t char1 = 0;
+
+      if (s == NULL)
+        return 0; /* UTF-8 character encodings are not state-dependent */
+
+      /* we know n >= 1 if we get here */
+      *pwc = 0;
+      char1 = (wchar_t)*t;
+
+      if (char1 == '\0')
+        return 0; /* s points to the null character */
+
+      if (char1 >= 0x0 && char1 <= 0x7f)
+        {
+          /* single-byte sequence */
+          *pwc = char1;
+          return 1;
+        }
+      else if (char1 >= 0xc0 && char1 <= 0xdf)
+        {
+          /* two-byte sequence */
+          if (n >= 2)
+            {
+              wchar_t char2 = (wchar_t)*(t+1);
+
+              if (char2 < 0x80 || char2 > 0xbf)
+                return -1;
+
+              if (char1 < 0xc2)
+                /* overlong UTF-8 sequence */
+                return -1;
+
+              *pwc = ((char1 & 0x1f) << 6)
+                |     (char2 & 0x3f);
+              return 2;
+            }
+          else
+            return -1;
+        }
+      else if (char1 >= 0xe0 && char1 <= 0xef)
+        {
+          /* three-byte sequence */
+          if (n >= 3)
+            {
+              wchar_t char2 = (wchar_t)*(t+1);
+              wchar_t char3 = (wchar_t)*(t+2);
+
+              if (char2 < 0x80 || char2 > 0xbf)
+                return -1;
+              if (char3 < 0x80 || char3 > 0xbf)
+                return -1;
+
+              if (char1 == 0xe0)
+                {
+                  if (char2 < 0xa0)
+                    /* overlong UTF-8 sequence */
+                    return -1;
+                }
+
+              *pwc = ((char1 & 0x0f) << 12)
+                |    ((char2 & 0x3f) << 6)
+                |     (char3 & 0x3f);
+
+              if (*pwc >= 0xd800 && *pwc <= 0xdfff)
+                {
+                  return -1;
+                }
+              else
+                return 3;
+            }
+          else
+            return -1;
+        }
+      else if (char1 >= 0xf0 && char1 <= 0xf7)
+        {
+          /* four-byte sequence */
+          if (n >= 4)
+            {
+              wchar_t char2 = (wchar_t)*(t+1);
+              wchar_t char3 = (wchar_t)*(t+2);
+              wchar_t char4 = (wchar_t)*(t+3);
+
+              if (char2 < 0x80 || char2 > 0xbf)
+                return -1;
+              if (char3 < 0x80 || char3 > 0xbf)
+                return -1;
+              if (char4 < 0x80 || char4 > 0xbf)
+                return -1;
+
+              if (char1 == 0xf0)
+                {
+                  if (char2 < 0x90)
+                    /* overlong UTF-8 sequence */
+                    return -1;
+                }
+                    
+              *pwc = ((char1 & 0x07) << 18)
+                |    ((char2 & 0x3f) << 12)
+                |    ((char3 & 0x3f) << 6)
+                |     (char4 & 0x3f);
+
+              return 4;
+            }
+          else
+            return -1;
+        }
+      else if (char1 >= 0xf8 && char1 <= 0xfb)
+        {
+          /* five-byte sequence */
+          if (n >= 5)
+            {
+              wchar_t char2 = (wchar_t)*(t+1);
+              wchar_t char3 = (wchar_t)*(t+2);
+              wchar_t char4 = (wchar_t)*(t+3);
+              wchar_t char5 = (wchar_t)*(t+4);
+
+              if (char2 < 0x80 || char2 > 0xbf)
+                return -1;
+              if (char3 < 0x80 || char3 > 0xbf)
+                return -1;
+              if (char4 < 0x80 || char4 > 0xbf)
+                return -1;
+              if (char5 < 0x80 || char5 > 0xbf)
+                return -1;
+
+              if (char1 == 0xf8)
+                {
+                  if (char2 < 0x88)
+                    /* overlong UTF-8 sequence */
+                    return -1;
+                }
+                    
+              *pwc = ((char1 & 0x03) << 24)
+                |    ((char2 & 0x3f) << 18)
+                |    ((char3 & 0x3f) << 12)
+                |    ((char4 & 0x3f) << 6)
+                |     (char5 & 0x3f);
+              return 5;
+            }
+          else
+            return -1;
+        }
+      else if (char1 >= 0xfc && char1 <= 0xfd)
+        {
+          /* six-byte sequence */
+          if (n >= 6)
+            {
+              wchar_t char2 = (wchar_t)*(t+1);
+              wchar_t char3 = (wchar_t)*(t+2);
+              wchar_t char4 = (wchar_t)*(t+3);
+              wchar_t char5 = (wchar_t)*(t+4);
+              wchar_t char6 = (wchar_t)*(t+5);
+
+              if (char2 < 0x80 || char2 > 0xbf)
+                return -1;
+              if (char3 < 0x80 || char3 > 0xbf)
+                return -1;
+              if (char4 < 0x80 || char4 > 0xbf)
+                return -1;
+              if (char5 < 0x80 || char5 > 0xbf)
+                return -1;
+              if (char6 < 0x80 || char6 > 0xbf)
+                return -1;
+
+              if (char1 == 0xfc)
+                {
+                  if (char2 < 0x84)
+                    /* overlong UTF-8 sequence */
+                    return -1;
+                }
+
+              *pwc = ((char1 & 0x01) << 30)
+                |    ((char2 & 0x3f) << 24)
+                |    ((char3 & 0x3f) << 18)
+                |    ((char4 & 0x3f) << 12)
+                |    ((char5 & 0x3f) << 6)
+                |     (char6 & 0x3f);
+              return 6;
+            }
+          else
+            return -1;
+        }
+      else
+        return -1;
+    }      
   else if (!strcmp (r->_current_locale, "C-SJIS"))
     {
       int char1;
diff --git a/newlib/libc/stdlib/wctomb_r.c b/newlib/libc/stdlib/wctomb_r.c
index b382c995f..991e0a610 100644
--- a/newlib/libc/stdlib/wctomb_r.c
+++ b/newlib/libc/stdlib/wctomb_r.c
@@ -12,6 +12,63 @@ _DEFUN (_wctomb_r, (r, s, wchar, state),
 {
   if (strlen (r->_current_locale) <= 1)
     { /* fall-through */ }
+  else if (!strcmp (r->_current_locale, "UTF-8"))
+    {
+      if (s == NULL)
+        return 0; /* UTF-8 encoding is not state-dependent */
+
+      if (wchar <= 0x7f)
+        {
+          *s = wchar;
+          return 1;
+        }
+      else if (wchar >= 0x80 && wchar <= 0x7ff)
+        {
+          *s++ = 0xc0 | ((wchar & 0x7c0) >> 6);
+          *s   = 0x80 |  (wchar &  0x3f);
+          return 2;
+        }
+      else if (wchar >= 0x800 && wchar <= 0xffff)
+        {
+          /* UTF-16 surrogates -- must not occur in normal UCS-4 data */
+          if (wchar >= 0xd800 && wchar <= 0xdfff)
+            return -1;
+
+          *s++ = 0xe0 | ((wchar & 0xf000) >> 12);
+          *s++ = 0x80 | ((wchar &  0xfc0) >> 6);
+          *s   = 0x80 |  (wchar &   0x3f);
+          return 3;
+        }
+      else if (wchar >= 0x10000 && wchar <= 0x1fffff)
+        {
+          *s++ = 0xf0 | ((wchar & 0x1c0000) >> 18);
+          *s++ = 0x80 | ((wchar &  0x3f000) >> 12);
+          *s++ = 0x80 | ((wchar &    0xfc0) >> 6);
+          *s   = 0x80 |  (wchar &     0x3f);
+          return 4;
+        }
+      else if (wchar >= 0x200000 && wchar <= 0x3ffffff)
+        {
+          *s++ = 0xf8 | ((wchar & 0x3000000) >> 24);
+          *s++ = 0x80 | ((wchar &  0xfc0000) >> 18);
+          *s++ = 0x80 | ((wchar &   0x3f000) >> 12);
+          *s++ = 0x80 | ((wchar &     0xfc0) >> 6);
+          *s   = 0x80 |  (wchar &      0x3f);
+          return 5;
+        }
+      else if (wchar >= 0x4000000 && wchar <= 0x7fffffff)
+        {
+          *s++ = 0xfc | ((wchar & 0x40000000) >> 30);
+          *s++ = 0x80 | ((wchar & 0x3f000000) >> 24);
+          *s++ = 0x80 | ((wchar &   0xfc0000) >> 18);
+          *s++ = 0x80 | ((wchar &    0x3f000) >> 12);
+          *s++ = 0x80 | ((wchar &      0xfc0) >> 6);
+          *s   = 0x80 |  (wchar &       0x3f);
+          return 6;
+        }
+      else
+        return -1;
+    }
   else if (!strcmp (r->_current_locale, "C-SJIS"))
     {
       unsigned char char2 = (unsigned char)wchar;