Merge in some clean-up and optimization changes from glibc's md5.c.

1 files changed, 39 insertions, 10 deletions

diff --git a/lib/sha.c b/lib/sha.c
index 18d7f190f..150bf0e42 100644
--- a/lib/sha.c
+++ b/lib/sha.c
@@ -217,13 +217,14 @@ sha_process_bytes (const void *buffer, size_t len, struct sha_ctx *ctx)
       memcpy (&ctx->buffer[left_over], buffer, add);
       ctx->buflen += add;
 
-      if (left_over + add > 64)
+      if (ctx->buflen > 64)
 	{
-	  sha_process_block (ctx->buffer, (left_over + add) & ~63, ctx);
+	  sha_process_block (ctx->buffer, ctx->buflen & ~63, ctx);
+
+	  ctx->buflen &= 63;
 	  /* The regions in the following copy operation cannot overlap.  */
 	  memcpy (ctx->buffer, &ctx->buffer[(left_over + add) & ~63],
-		  (left_over + add) & 63);
-	  ctx->buflen = (left_over + add) & 63;
+		  ctx->buflen);
 	}
 
       buffer = (const char *) buffer + add;
@@ -231,18 +232,46 @@ sha_process_bytes (const void *buffer, size_t len, struct sha_ctx *ctx)
     }
 
   /* Process available complete blocks.  */
-  if (len > 64)
+  if (len >= 64)
     {
-      sha_process_block (buffer, len & ~63, ctx);
-      buffer = (const char *) buffer + (len & ~63);
-      len &= 63;
+#if !_STRING_ARCH_unaligned
+/* To check alignment gcc has an appropriate operator.  Other
+   compilers don't.  */
+# if __GNUC__ >= 2
+#  define UNALIGNED_P(p) (((md5_uintptr) p) % __alignof__ (md5_uint32) != 0)
+# else
+#  define UNALIGNED_P(p) (((md5_uintptr) p) % sizeof (md5_uint32) != 0)
+# endif
+      if (UNALIGNED_P (buffer))
+	while (len > 64)
+	  {
+	    sha_process_block (memcpy (ctx->buffer, buffer, 64), 64, ctx);
+	    buffer = (const char *) buffer + 64;
+	    len -= 64;
+	  }
+      else
+#endif
+	{
+	  sha_process_block (buffer, len & ~63, ctx);
+	  buffer = (const char *) buffer + (len & ~63);
+	  len &= 63;
+	}
     }
 
   /* Move remaining bytes in internal buffer.  */
   if (len > 0)
     {
-      memcpy (ctx->buffer, buffer, len);
-      ctx->buflen = len;
+      size_t left_over = ctx->buflen;
+
+      memcpy (&ctx->buffer[left_over], buffer, len);
+      left_over += len;
+      if (left_over >= 64)
+	{
+	  sha_process_block (ctx->buffer, 64, ctx);
+	  left_over -= 64;
+	  memcpy (ctx->buffer, &ctx->buffer[64], left_over);
+	}
+      ctx->buflen = left_over;
     }
 }