cmd/internal/obj/loong64: add {,X}VSHUF.{B/H/W/V} instructions support

Go asm syntax:
         VSHUFB             V1, V2, V3, V4
        XVSHUFB             X1, X2, X3, X4
         VSHUF{H/W/V}       V1, V2, V3
        XVSHUF{H/W/V}       X1, X2, X3

Equivalent platform assembler syntax:
         vshuf.b            v4, v3, v2, v1
        xvshuf.b            x4, x3, x2, x1
         vshuf.{h/w/d}      v3, v2, v1
        xvshuf.{h/w/d}      x3, x2, x1

Change-Id: I8983467495f587cf46083fd81cb024400c7dc2a7
Reviewed-on: https://go-review.googlesource.com/c/go/+/716804
Reviewed-by: Meidan Li <limeidan@loongson.cn>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: sophie zhao <zhaoxiaolin@loongson.cn>

Author: abner-chenc

src/cmd/asm/internal/asm/testdata/loong64enc1.s

@@ -1023,6 +1023,16 @@ lable2:
 	XVSHUF4IV	$8, X1, X2	// 22209c77
 	XVSHUF4IV	$15, X1, X2	// 223c9c77
 
+	// [X]VSHUF.{B/H/W/V} instructions
+	VSHUFH		V1, V2, V3	// 43847a71
+	VSHUFW		V1, V2, V3	// 43047b71
+	VSHUFV		V1, V2, V3	// 43847b71
+	XVSHUFH		X1, X2, X3	// 43847a75
+	XVSHUFW		X1, X2, X3	// 43047b75
+	XVSHUFV		X1, X2, X3	// 43847b75
+	VSHUFB		V1, V2, V3, V4	// 6488500d
+	XVSHUFB		X1, X2, X3, X4	// 6488600d
+
 	// VPERMIW, XVPERMI{W,V,Q} instructions
 	VPERMIW		$0x1B, V1, V2 	// VPERMIW	$27, V1, V2 // 226ce473
 	XVPERMIW	$0x2B, X1, X2 	// XVPERMIW	$43, X1, X2 // 22ace477

src/cmd/internal/obj/loong64/a.out.go

@@ -1115,6 +1115,15 @@ const (
 	AXVSHUF4IW
 	AXVSHUF4IV
 
+	AVSHUFB
+	AVSHUFH
+	AVSHUFW
+	AVSHUFV
+	AXVSHUFB
+	AXVSHUFH
+	AXVSHUFW
+	AXVSHUFV
+
 	AVPERMIW
 	AXVPERMIW
 	AXVPERMIV

src/cmd/internal/obj/loong64/anames.go

@@ -155,6 +155,8 @@ var optab = []Optab{
 
 	{AFMADDF, C_FREG, C_FREG, C_NONE, C_FREG, C_NONE, 37, 4, 0, 0},
 	{AFMADDF, C_FREG, C_FREG, C_FREG, C_FREG, C_NONE, 37, 4, 0, 0},
+	{AVSHUFB, C_VREG, C_VREG, C_VREG, C_VREG, C_NONE, 37, 4, 0, 0},
+	{AXVSHUFB, C_XREG, C_XREG, C_XREG, C_XREG, C_NONE, 37, 4, 0, 0},
 
 	{AFSEL, C_FCCREG, C_FREG, C_FREG, C_FREG, C_NONE, 33, 4, 0, 0},
 	{AFSEL, C_FCCREG, C_FREG, C_NONE, C_FREG, C_NONE, 33, 4, 0, 0},
@@ -1561,6 +1563,8 @@ func buildop(ctxt *obj.Link) {
 			AMOVWU,
 			AVMOVQ,
 			AXVMOVQ,
+			AVSHUFB,
+			AXVSHUFB,
 			ANEGW,
 			ANEGV,
 			AWORD,
@@ -1698,6 +1702,9 @@ func buildop(ctxt *obj.Link) {
 			opset(AVMULD, r0)
 			opset(AVDIVF, r0)
 			opset(AVDIVD, r0)
+			opset(AVSHUFH, r0)
+			opset(AVSHUFW, r0)
+			opset(AVSHUFV, r0)
 
 		case AXVSEQB:
 			opset(AXVSEQH, r0)
@@ -1771,6 +1778,9 @@ func buildop(ctxt *obj.Link) {
 			opset(AXVMULD, r0)
 			opset(AXVDIVF, r0)
 			opset(AXVDIVD, r0)
+			opset(AXVSHUFH, r0)
+			opset(AXVSHUFW, r0)
+			opset(AXVSHUFV, r0)
 
 		case AVANDB:
 			opset(AVORB, r0)
@@ -3107,6 +3117,10 @@ func (c *ctxt0) oprrrr(a obj.As) uint32 {
 		return 0x8d << 20 // fnmsub.s
 	case AFNMSUBD:
 		return 0x8e << 20 // fnmsub.d
+	case AVSHUFB:
+		return 0x0D5 << 20 // vshuf.b
+	case AXVSHUFB:
+		return 0x0D6 << 20 // xvshuf.b
 	}
 
 	c.ctxt.Diag("bad rrrr opcode %v", a)
@@ -3775,6 +3789,18 @@ func (c *ctxt0) oprrr(a obj.As) uint32 {
 		return 0xea22 << 15 // xvbitrev.w
 	case AXVBITREVV:
 		return 0xea23 << 15 // xvbitrev.d
+	case AVSHUFH:
+		return 0x0E2F5 << 15 // vshuf.h
+	case AVSHUFW:
+		return 0x0E2F6 << 15 // vshuf.w
+	case AVSHUFV:
+		return 0x0E2F7 << 15 // vshuf.d
+	case AXVSHUFH:
+		return 0x0EAF5 << 15 // xvshuf.h
+	case AXVSHUFW:
+		return 0x0EAF6 << 15 // xvshuf.w
+	case AXVSHUFV:
+		return 0x0EAF7 << 15 // xvshuf.d
 	}
 
 	if a < 0 {