sha1-alpha.pl: commentary update.

diff --git a/crypto/sha/asm/sha1-alpha.pl b/crypto/sha/asm/sha1-alpha.pl
index e8989307112b91f98301776b53fb7b7ce5fb50ac..9d1e9435acd6a39fa275ba0aa28cf780a156a336 100644 (file)
--- a/crypto/sha/asm/sha1-alpha.pl
+++ b/crypto/sha/asm/sha1-alpha.pl
@@ -10,8 +10,9 @@
 # SHA1 block procedure for Alpha.
 
 # On 21264 performance is 33% better than code generated by vendor
 # SHA1 block procedure for Alpha.
 
 # On 21264 performance is 33% better than code generated by vendor

-# compiler, and 75% better than GCC [3.4]. Implementation features
-# vectorized byte swap, but not Xupdate.
+# compiler, and 75% better than GCC [3.4], and in absolute terms is
+# 8.7 cycles per processed byte. Implementation features vectorized
+# byte swap, but not Xupdate.

 
 @X=(   "\$0",  "\$1",  "\$2",  "\$3",  "\$4",  "\$5",  "\$6",  "\$7",
        "\$8",  "\$9",  "\$10", "\$11", "\$12", "\$13", "\$14", "\$15");
 
 @X=(   "\$0",  "\$1",  "\$2",  "\$3",  "\$4",  "\$5",  "\$6",  "\$7",
        "\$8",  "\$9",  "\$10", "\$11", "\$12", "\$13", "\$14", "\$15");