crypto/sha512: use math/bits.RotateLeft64 instead of ad-hoc implementation

This makes code more readable and idiomatic and slightly increase performance.

Updates #31456

Benchstat:
name          old time/op    new time/op    delta
Hash8Bytes-8     281ns ± 4%     280ns ± 3%    ~     (p=0.640 n=10+10)
Hash1K-8        2.01µs ± 6%    2.02µs ± 3%    ~     (p=0.481 n=10+10)
Hash8K-8        14.2µs ± 6%    13.5µs ± 1%  -4.90%  (p=0.001 n=10+10)

name          old speed      new speed      delta
Hash8Bytes-8  28.5MB/s ± 4%  28.5MB/s ± 3%    ~     (p=0.516 n=10+10)
Hash1K-8       510MB/s ± 6%   507MB/s ± 4%    ~     (p=0.481 n=10+10)
Hash8K-8       576MB/s ± 6%   605MB/s ± 1%  +5.02%  (p=0.001 n=10+10)

Tested on macbook pro 2018 Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz

Change-Id: I1f5b78096dd49d14ffcb9129142c4a4e05b81ff9
Reviewed-on: https://go-review.googlesource.com/c/go/+/171736
Reviewed-by: Filippo Valsorda <filippo@golang.org>

diff --git a/src/crypto/sha512/sha512block.go b/src/crypto/sha512/sha512block.go
index 42e8d19fe8c850c87e9dbb78facf7c292acab8c6..81569c5f84e8964e955b5b5aa24e9f6e696c923d 100644 (file)
--- a/src/crypto/sha512/sha512block.go
+++ b/src/crypto/sha512/sha512block.go
@@ -8,6 +8,8 @@
 
 package sha512
 
+import "math/bits"
+
 var _K = []uint64{
        0x428a2f98d728ae22,
        0x7137449123ef65cd,
@@ -102,9 +104,9 @@ func blockGeneric(dig *digest, p []byte) {
                }
                for i := 16; i < 80; i++ {
                        v1 := w[i-2]
-                       t1 := (v1>>19 | v1<<(64-19)) ^ (v1>>61 | v1<<(64-61)) ^ (v1 >> 6)
+                       t1 := bits.RotateLeft64(v1, -19) ^ bits.RotateLeft64(v1, -61) ^ (v1 >> 6)
                        v2 := w[i-15]
-                       t2 := (v2>>1 | v2<<(64-1)) ^ (v2>>8 | v2<<(64-8)) ^ (v2 >> 7)
+                       t2 := bits.RotateLeft64(v2, -1) ^ bits.RotateLeft64(v2, -8) ^ (v2 >> 7)
 
                        w[i] = t1 + w[i-7] + t2 + w[i-16]
                }
@@ -112,9 +114,9 @@ func blockGeneric(dig *digest, p []byte) {
                a, b, c, d, e, f, g, h := h0, h1, h2, h3, h4, h5, h6, h7
 
                for i := 0; i < 80; i++ {
-                       t1 := h + ((e>>14 | e<<(64-14)) ^ (e>>18 | e<<(64-18)) ^ (e>>41 | e<<(64-41))) + ((e & f) ^ (^e & g)) + _K[i] + w[i]
+                       t1 := h + (bits.RotateLeft64(e, -14) ^ bits.RotateLeft64(e, -18) ^ bits.RotateLeft64(e, -41)) + ((e & f) ^ (^e & g)) + _K[i] + w[i]
 
-                       t2 := ((a>>28 | a<<(64-28)) ^ (a>>34 | a<<(64-34)) ^ (a>>39 | a<<(64-39))) + ((a & b) ^ (a & c) ^ (b & c))
+                       t2 := (bits.RotateLeft64(a, -28) ^ bits.RotateLeft64(a, -34) ^ bits.RotateLeft64(a, -39)) + ((a & b) ^ (a & c) ^ (b & c))
 
                        h = g
                        g = f