cmd/internal/obj/loong64: add VPERMI.W, XVPERMI.{W,V,Q} instruction support

Go asm syntax:
	 VPERMIW        $0x1b, vj, vd
	XVPERMI{W,V,Q}  $0x1b, xj, xd

Equivalent platform assembler syntax:
	 vpermi.w       vd, vj, $0x1b
	xvpermi.{w,d,q} xd, xj, $0x1b

Change-Id: Ie23b2fdd09b4c93801dc804913206f1c5a496268
Reviewed-on: https://go-review.googlesource.com/c/go/+/716800
Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Meidan Li <limeidan@loongson.cn>
Reviewed-by: sophie zhao <zhaoxiaolin@loongson.cn>
Reviewed-by: Michael Knyszek <mknyszek@google.com>

5 files changed, 44 insertions, 0 deletions

diff --git a/src/cmd/asm/internal/asm/testdata/loong64enc1.s b/src/cmd/asm/internal/asm/testdata/loong64enc1.s
index 6e2a86969d..ca86ef8e6c 100644
--- a/src/cmd/asm/internal/asm/testdata/loong64enc1.s
+++ b/src/cmd/asm/internal/asm/testdata/loong64enc1.s
@@ -1017,6 +1017,12 @@ lable2:
 	XVSHUF4IV	$8, X1, X2	// 22209c77
 	XVSHUF4IV	$15, X1, X2	// 223c9c77
 
+	// VPERMIW, XVPERMI{W,V,Q} instructions
+	VPERMIW		$0x1B, V1, V2 	// VPERMIW	$27, V1, V2 // 226ce473
+	XVPERMIW	$0x2B, X1, X2 	// XVPERMIW	$43, X1, X2 // 22ace477
+	XVPERMIV	$0x3B, X1, X2 	// XVPERMIV	$59, X1, X2 // 22ece877
+	XVPERMIQ	$0x4B, X1, X2 	// XVPERMIQ	$75, X1, X2 // 222ced77
+
 	// [X]VSETEQZ.V, [X]VSETNEZ.V
 	VSETEQV		V1, FCC0	// 20989c72
 	VSETNEV		V1, FCC0	// 209c9c72
diff --git a/src/cmd/internal/obj/loong64/a.out.go b/src/cmd/internal/obj/loong64/a.out.go
index 3a676db922..762dc338e3 100644
--- a/src/cmd/internal/obj/loong64/a.out.go
+++ b/src/cmd/internal/obj/loong64/a.out.go
@@ -1115,6 +1115,11 @@ const (
 	AXVSHUF4IW
 	AXVSHUF4IV
 
+	AVPERMIW
+	AXVPERMIW
+	AXVPERMIV
+	AXVPERMIQ
+
 	AVSETEQV
 	AVSETNEV
 	AVSETANYEQB
diff --git a/src/cmd/internal/obj/loong64/anames.go b/src/cmd/internal/obj/loong64/anames.go
index 422ccbd9b0..607e606311 100644
--- a/src/cmd/internal/obj/loong64/anames.go
+++ b/src/cmd/internal/obj/loong64/anames.go
@@ -586,6 +586,10 @@ var Anames = []string{
 	"XVSHUF4IH",
 	"XVSHUF4IW",
 	"XVSHUF4IV",
+	"VPERMIW",
+	"XVPERMIW",
+	"XVPERMIV",
+	"XVPERMIQ",
 	"VSETEQV",
 	"VSETNEV",
 	"VSETANYEQB",
diff --git a/src/cmd/internal/obj/loong64/asm.go b/src/cmd/internal/obj/loong64/asm.go
index 7eb5668d82..8e2393bc1c 100644
--- a/src/cmd/internal/obj/loong64/asm.go
+++ b/src/cmd/internal/obj/loong64/asm.go
@@ -1778,6 +1778,7 @@ func buildop(ctxt *obj.Link) {
 			opset(AVSHUF4IH, r0)
 			opset(AVSHUF4IW, r0)
 			opset(AVSHUF4IV, r0)
+			opset(AVPERMIW, r0)
 
 		case AXVANDB:
 			opset(AXVORB, r0)
@@ -1787,6 +1788,9 @@ func buildop(ctxt *obj.Link) {
 			opset(AXVSHUF4IH, r0)
 			opset(AXVSHUF4IW, r0)
 			opset(AXVSHUF4IV, r0)
+			opset(AXVPERMIW, r0)
+			opset(AXVPERMIV, r0)
+			opset(AXVPERMIQ, r0)
 
 		case AVANDV:
 			opset(AVORV, r0)
@@ -4362,6 +4366,14 @@ func (c *ctxt0) opirr(a obj.As) uint32 {
 		return 0x1de6 << 18 // xvshuf4i.w
 	case AXVSHUF4IV:
 		return 0x1de7 << 18 // xvshuf4i.d
+	case AVPERMIW:
+		return 0x1cf9 << 18 // vpermi.w
+	case AXVPERMIW:
+		return 0x1df9 << 18 // xvpermi.w
+	case AXVPERMIV:
+		return 0x1dfa << 18 // xvpermi.d
+	case AXVPERMIQ:
+		return 0x1dfb << 18 // xvpermi.q
 	case AVBITCLRB:
 		return 0x1CC4<<18 | 0x1<<13 // vbitclri.b
 	case AVBITCLRH:
diff --git a/src/cmd/internal/obj/loong64/doc.go b/src/cmd/internal/obj/loong64/doc.go
index f7e5a4fb42..45f75e6e70 100644
--- a/src/cmd/internal/obj/loong64/doc.go
+++ b/src/cmd/internal/obj/loong64/doc.go
@@ -229,6 +229,23 @@ Note: In the following sections 3.1 to 3.6, "ui4" (4-bit unsigned int immediate)
          VMOVQ  8(R4), V5.W4     |      vldrepl.w  v5, r4, $2
          VMOVQ  8(R4), V5.V2     |      vldrepl.d  v5, r4, $1
 
+3.8 Vector permutation instruction
+	Instruction format:
+	VPERMIW    ui8, Vj, Vd
+
+	Mapping between Go and platform assembly:
+	     Go assembly     |   platform assembly   |                                 semantics
+	VPERMIW  ui8, Vj, Vd |  vpermi.w vd, vj, ui8 | VR[vd].W[0] = VR[vj].W[ui8[1:0]], VR[vd].W[1] = VR[vj].W[ui8[3:2]],
+	                     |                       | VR[vd].W[2] = VR[vd].W[ui8[5:4]], VR[vd].W[3] = VR[vd].W[ui8[7:6]]
+	XVPERMIW ui8, Xj, Xd | xvpermi.w xd, xj, ui8 | XR[xd].W[0] = XR[xj].W[ui8[1:0]],   XR[xd].W[1] = XR[xj].W[ui8[3:2]],
+	                     |                       | XR[xd].W[3] = XR[xd].W[ui8[7:6]],   XR[xd].W[2] = XR[xd].W[ui8[5:4]],
+	                     |                       | XR[xd].W[4] = XR[xj].W[ui8[1:0]+4], XR[xd].W[5] = XR[xj].W[ui8[3:2]+4],
+	                     |                       | XR[xd].W[6] = XR[xd].W[ui8[5:4]+4], XR[xd].W[7] = XR[xd].W[ui8[7:6]+4]
+	XVPERMIV ui8, Xj, Xd | xvpermi.d xd, xj, ui8 | XR[xd].D[0] = XR[xj].D[ui8[1:0]], XR[xd].D[1] = XR[xj].D[ui8[3:2]],
+	                     |                       | XR[xd].D[2] = XR[xj].D[ui8[5:4]], XR[xd].D[3] = XR[xj].D[ui8[7:6]]
+	XVPERMIQ ui8, Xj, Xd | xvpermi.q xd, xj, ui8 | vec = {XR[xd], XR[xj]}, XR[xd].Q[0] = vec.Q[ui8[1:0]], XR[xd].Q[1] = vec.Q[ui8[5:4]]
+
+
 # Special instruction encoding definition and description on LoongArch
 
  1. DBAR hint encoding for LA664(Loongson 3A6000) and later micro-architectures, paraphrased