cmd/compile: use MOVBE instruction for GOAMD64>=v3

encoding/binary benchmark on my laptop:
name                      old time/op    new time/op     delta
ReadSlice1000Int32s-8       4.42µs ± 5%     4.20µs ± 1%   -4.94%  (p=0.046 n=9+8)
ReadStruct-8                 359ns ± 8%      368ns ± 5%   +2.35%  (p=0.041 n=9+10)
WriteStruct-8                349ns ± 1%      357ns ± 1%   +2.15%  (p=0.000 n=8+10)
ReadInts-8                   235ns ± 1%      233ns ± 1%   -1.01%  (p=0.005 n=10+10)
WriteInts-8                  265ns ± 1%      274ns ± 1%   +3.45%  (p=0.000 n=10+10)
WriteSlice1000Int32s-8      4.61µs ± 5%     4.59µs ± 5%     ~     (p=0.986 n=10+10)
PutUint16-8                 0.56ns ± 4%     0.57ns ± 4%     ~     (p=0.101 n=10+10)
PutUint32-8                 0.83ns ± 2%     0.56ns ± 6%  -32.91%  (p=0.000 n=10+10)
PutUint64-8                 0.81ns ± 3%     0.62ns ± 4%  -23.82%  (p=0.000 n=10+10)
LittleEndianPutUint16-8     0.55ns ± 4%     0.55ns ± 3%     ~     (p=0.926 n=10+10)
LittleEndianPutUint32-8     0.41ns ± 4%     0.42ns ± 3%     ~     (p=0.148 n=10+9)
LittleEndianPutUint64-8     0.55ns ± 2%     0.56ns ± 6%     ~     (p=0.897 n=10+10)
ReadFloats-8                60.4ns ± 4%     59.0ns ± 1%   -2.25%  (p=0.007 n=10+10)
WriteFloats-8               72.3ns ± 2%     71.5ns ± 7%     ~     (p=0.089 n=10+10)
ReadSlice1000Float32s-8     4.21µs ± 3%     4.18µs ± 2%     ~     (p=0.197 n=10+10)
WriteSlice1000Float32s-8    4.61µs ± 2%     4.68µs ± 7%     ~     (p=1.000 n=8+10)
ReadSlice1000Uint8s-8        250ns ± 4%      247ns ± 4%     ~     (p=0.324 n=10+10)
WriteSlice1000Uint8s-8       227ns ± 5%      229ns ± 2%     ~     (p=0.193 n=10+7)
PutUvarint32-8              15.3ns ± 2%     15.4ns ± 4%     ~     (p=0.782 n=10+10)
PutUvarint64-8              38.5ns ± 1%     38.6ns ± 5%     ~     (p=0.396 n=8+10)

name                      old speed      new speed       delta
ReadSlice1000Int32s-8      890MB/s ±17%    953MB/s ± 1%   +7.00%  (p=0.027 n=10+8)
ReadStruct-8               209MB/s ± 8%    204MB/s ± 5%   -2.42%  (p=0.043 n=9+10)
WriteStruct-8              214MB/s ± 3%    210MB/s ± 1%   -1.75%  (p=0.003 n=9+10)
ReadInts-8                 127MB/s ± 1%    129MB/s ± 1%   +1.01%  (p=0.006 n=10+10)
WriteInts-8                113MB/s ± 1%    109MB/s ± 1%   -3.34%  (p=0.000 n=10+10)
WriteSlice1000Int32s-8     868MB/s ± 5%    872MB/s ± 5%     ~     (p=1.000 n=10+10)
PutUint16-8               3.55GB/s ± 4%   3.50GB/s ± 4%     ~     (p=0.093 n=10+10)
PutUint32-8               4.83GB/s ± 2%   7.21GB/s ± 6%  +49.16%  (p=0.000 n=10+10)
PutUint64-8               9.89GB/s ± 3%  12.99GB/s ± 4%  +31.26%  (p=0.000 n=10+10)
LittleEndianPutUint16-8   3.65GB/s ± 4%   3.65GB/s ± 4%     ~     (p=0.912 n=10+10)
LittleEndianPutUint32-8   9.74GB/s ± 3%   9.63GB/s ± 3%     ~     (p=0.222 n=9+9)
LittleEndianPutUint64-8   14.4GB/s ± 2%   14.3GB/s ± 5%     ~     (p=0.912 n=10+10)
ReadFloats-8               199MB/s ± 4%    203MB/s ± 1%   +2.27%  (p=0.007 n=10+10)
WriteFloats-8              166MB/s ± 2%    168MB/s ± 7%     ~     (p=0.089 n=10+10)
ReadSlice1000Float32s-8    949MB/s ± 3%    958MB/s ± 2%     ~     (p=0.218 n=10+10)
WriteSlice1000Float32s-8   867MB/s ± 2%    857MB/s ± 6%     ~     (p=1.000 n=8+10)
ReadSlice1000Uint8s-8     4.00GB/s ± 4%   4.06GB/s ± 4%     ~     (p=0.353 n=10+10)
WriteSlice1000Uint8s-8    4.40GB/s ± 4%   4.36GB/s ± 2%     ~     (p=0.193 n=10+7)
PutUvarint32-8             262MB/s ± 2%    260MB/s ± 4%     ~     (p=0.739 n=10+10)
PutUvarint64-8             208MB/s ± 1%    207MB/s ± 5%     ~     (p=0.408 n=8+10)

Updates #45453

Change-Id: Ifda0d48d54665cef45d46d3aad974062633142c4
Reviewed-on: https://go-review.googlesource.com/c/go/+/354670
Run-TryBot: Alberto Donizetti <alb.donizetti@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Trust: Matthew Dempsky <mdempsky@google.com>

1 files changed, 24 insertions, 12 deletions

diff --git a/test/codegen/memcombine.go b/test/codegen/memcombine.go
index 2a0c534df0..97e1d4bdfb 100644
--- a/test/codegen/memcombine.go
+++ b/test/codegen/memcombine.go
@@ -70,7 +70,8 @@ func load_le16_idx(b []byte, idx int) {
 }
 
 func load_be64(b []byte) {
-	// amd64:`BSWAPQ`,-`MOV[BWL]\t[^$]`,-`OR`
+	// amd64/v1,amd64/v2:`BSWAPQ`,-`MOV[BWL]\t[^$]`,-`OR`
+	// amd64/v3:`MOVBEQ`
 	// s390x:`MOVD\s\(.*\),`
 	// arm64:`REV`,`MOVD\s\(R[0-9]+\),`,-`MOV[BHW]`,-`REVW`,-`REV16W`
 	// ppc64le:`MOVDBR`,-`MOV[BHW]Z`
@@ -78,7 +79,8 @@ func load_be64(b []byte) {
 }
 
 func load_be64_idx(b []byte, idx int) {
-	// amd64:`BSWAPQ`,-`MOV[BWL]\t[^$]`,-`OR`
+	// amd64/v1,amd64/v2:`BSWAPQ`,-`MOV[BWL]\t[^$]`,-`OR`
+	// amd64/v3: `MOVBEQ`
 	// s390x:`MOVD\s\(.*\)\(.*\*1\),`
 	// arm64:`REV`,`MOVD\s\(R[0-9]+\)\(R[0-9]+\),`,-`MOV[WHB]`,-`REVW`,-`REV16W`
 	// ppc64le:`MOVDBR`,-`MOV[BHW]Z`
@@ -86,7 +88,8 @@ func load_be64_idx(b []byte, idx int) {
 }
 
 func load_be32(b []byte) {
-	// amd64:`BSWAPL`,-`MOV[BW]`,-`OR`
+	// amd64/v1,amd64/v2:`BSWAPL`,-`MOV[BW]`,-`OR`
+	// amd64/v3: `MOVBEL`
 	// s390x:`MOVWZ\s\(.*\),`
 	// arm64:`REVW`,`MOVWU\s\(R[0-9]+\),`,-`MOV[BH]`,-`REV16W`
 	// ppc64le:`MOVWBR`,-`MOV[BH]Z`
@@ -94,7 +97,8 @@ func load_be32(b []byte) {
 }
 
 func load_be32_idx(b []byte, idx int) {
-	// amd64:`BSWAPL`,-`MOV[BW]`,-`OR`
+	// amd64/v1,amd64/v2:`BSWAPL`,-`MOV[BW]`,-`OR`
+	// amd64/v3: `MOVBEL`
 	// s390x:`MOVWZ\s\(.*\)\(.*\*1\),`
 	// arm64:`REVW`,`MOVWU\s\(R[0-9]+\)\(R[0-9]+\),`,-`MOV[HB]`,-`REV16W`
 	// ppc64le:`MOVWBR`,-`MOV[BH]Z`
@@ -179,7 +183,8 @@ func load_be_byte4_uint32(s []byte) uint32 {
 
 func load_be_byte4_uint32_inv(s []byte) uint32 {
 	// arm64:`MOVWU\t\(R[0-9]+\)`,`REVW`,-`ORR`,-`REV16W`,-`MOV[BH]`
-	// amd64:`MOVL\s\([A-Z]+\)`,`BSWAPL`,-`MOV[BW]`,-`OR`
+	// amd64/v1,amd64/v2:`MOVL\s\([A-Z]+\)`,`BSWAPL`,-`MOV[BW]`,-`OR`
+	// amd64/v3: `MOVBEL`
 	return uint32(s[3]) | uint32(s[2])<<8 | uint32(s[1])<<16 | uint32(s[0])<<24
 }
 
@@ -191,7 +196,8 @@ func load_be_byte8_uint64(s []byte) uint64 {
 
 func load_be_byte8_uint64_inv(s []byte) uint64 {
 	// arm64:`MOVD\t\(R[0-9]+\)`,`REV`,-`ORR`,-`REVW`,-`REV16W`,-`MOV[BHW]`
-	// amd64:`MOVQ\s\([A-Z]+\),\s[A-Z]+`,`BSWAPQ`,-`MOV[BWL]\t[^$]`,-`OR`
+	// amd64/v1,amd64/v2:`MOVQ\s\([A-Z]+\),\s[A-Z]+`,`BSWAPQ`,-`MOV[BWL]\t[^$]`,-`OR`
+	// amd64/v3: `MOVBEQ`
 	// ppc64le:`MOVDBR\t\(R[0-9]+\)`,-`MOV[BHW]Z`
 	return uint64(s[7]) | uint64(s[6])<<8 | uint64(s[5])<<16 | uint64(s[4])<<24 | uint64(s[3])<<32 | uint64(s[2])<<40 | uint64(s[1])<<48 | uint64(s[0])<<56
 }
@@ -409,7 +415,8 @@ func store_le16_idx(b []byte, idx int) {
 }
 
 func store_be64(b []byte) {
-	// amd64:`BSWAPQ`,-`SHR.`
+	// amd64/v1,amd64/v2:`BSWAPQ`,-`SHR.`
+	// amd64/v3: `MOVBEQ`
 	// arm64:`MOVD`,`REV`,-`MOV[WBH]`,-`REVW`,-`REV16W`
 	// ppc64le:`MOVDBR`
 	// s390x:`MOVD\s.*\(.*\)$`,-`SRW\s`,-`SRD\s`
@@ -417,7 +424,8 @@ func store_be64(b []byte) {
 }
 
 func store_be64_idx(b []byte, idx int) {
-	// amd64:`BSWAPQ`,-`SHR.`
+	// amd64/v1,amd64/v2:`BSWAPQ`,-`SHR.`
+	// amd64/v3:`MOVBEQ`
 	// arm64:`REV`,`MOVD\sR[0-9]+,\s\(R[0-9]+\)\(R[0-9]+\)`,-`MOV[BHW]`,-`REV16W`,-`REVW`
 	// ppc64le:`MOVDBR`
 	// s390x:`MOVD\s.*\(.*\)\(.*\*1\)$`,-`SRW\s`,-`SRD\s`
@@ -425,7 +433,8 @@ func store_be64_idx(b []byte, idx int) {
 }
 
 func store_be32(b []byte) {
-	// amd64:`BSWAPL`,-`SHR.`
+	// amd64/v1,amd64/v2:`BSWAPL`,-`SHR.`
+	// amd64/v3:`MOVBEL`
 	// arm64:`MOVW`,`REVW`,-`MOV[BH]`,-`REV16W`
 	// ppc64le:`MOVWBR`
 	// s390x:`MOVW\s.*\(.*\)$`,-`SRW\s`,-`SRD\s`
@@ -445,7 +454,8 @@ func store_be32_load(b, x *[8]byte) {
 }
 
 func store_be32_idx(b []byte, idx int) {
-	// amd64:`BSWAPL`,-`SHR.`
+	// amd64/v1,amd64/v2:`BSWAPL`,-`SHR.`
+	// amd64/v3:`MOVBEL`
 	// arm64:`REVW`,`MOVW\sR[0-9]+,\s\(R[0-9]+\)\(R[0-9]+\)`,-`MOV[BH]`,-`REV16W`
 	// ppc64le:`MOVWBR`
 	// s390x:`MOVW\s.*\(.*\)\(.*\*1\)$`,-`SRW\s`,-`SRD\s`
@@ -508,14 +518,16 @@ func store_be_byte_2(b []byte, val uint16) {
 func store_be_byte_4(b []byte, val uint32) {
 	_ = b[4]
 	// arm64:`REVW`,`MOVW\sR[0-9]+,\s1\(R[0-9]+\)`,-`MOVB`,-`MOVH`,-`REV16W`
-	// amd64:`MOVL\s[A-Z]+,\s1\([A-Z]+\)`,-`MOVB`,-`MOVW`
+	// amd64/v1,amd64/v2:`MOVL\s[A-Z]+,\s1\([A-Z]+\)`,-`MOVB`,-`MOVW`
+	// amd64/v3:`MOVBEL\s[A-Z]+,\s1\([A-Z]+\)`
 	b[1], b[2], b[3], b[4] = byte(val>>24), byte(val>>16), byte(val>>8), byte(val)
 }
 
 func store_be_byte_8(b []byte, val uint64) {
 	_ = b[8]
 	// arm64:`REV`,`MOVD\sR[0-9]+,\s1\(R[0-9]+\)`,-`MOVB`,-`MOVH`,-`MOVW`,-`REV16W`,-`REVW`
-	// amd64:`MOVQ\s[A-Z]+,\s1\([A-Z]+\)`,-`MOVB`,-`MOVW`,-`MOVL`
+	// amd64/v1,amd64/v2:`MOVQ\s[A-Z]+,\s1\([A-Z]+\)`,-`MOVB`,-`MOVW`,-`MOVL`
+	// amd64/v3:`MOVBEQ\s[A-Z]+,\s1\([A-Z]+\)`, -`MOVBEL`
 	b[1], b[2], b[3], b[4], b[5], b[6], b[7], b[8] = byte(val>>56), byte(val>>48), byte(val>>40), byte(val>>32), byte(val>>24), byte(val>>16), byte(val>>8), byte(val)
 }