Files
gasm-sdk/asm/evex_test.go
T

226 lines
10 KiB
Go
Raw Normal View History

// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause
package asm
import (
"os"
"strings"
"testing"
"golang.org/x/arch/x86/x86asm"
"sourcedock.dev/petrbalvin/gasm-devkit/parser"
)
// TestEvexGroundTruth checks the EVEX (AVX-512) encodings byte for byte
// against machine code extracted from the Go toolchain's assembly of the
// same instructions, covering every operand shape the go-flac AVX-512
// kernels use: NDS arithmetic, immediate and variable shifts, shuffles with
// an immediate, lane extracts, narrowing stores, broadcasts from a GPR or
// memory, mask destinations, mask moves, disp8×N compression and the 5-bit
// register fields (X/Y 16–31, Z 0–31).
func TestEvexGroundTruth(t *testing.T) {
cases := []struct {
name string
mnem string
ops []Operand
want string
}{
// NDS integer arithmetic / logic.
{"VPXORD Z12,Z12,Z12", "VPXORD", []Operand{vreg(t, "Z12"), vreg(t, "Z12"), vreg(t, "Z12")}, "62511d48efe4"},
{"VPXORQ Z8,Z9,Z10", "VPXORQ", []Operand{vreg(t, "Z8"), vreg(t, "Z9"), vreg(t, "Z10")}, "6251b548efd0"},
{"VPADDD Z1,Z0,Z0", "VPADDD", []Operand{vreg(t, "Z1"), vreg(t, "Z0"), vreg(t, "Z0")}, "62f17d48fec1"},
{"VPSUBQ Z8,Z11,Z11", "VPSUBQ", []Operand{vreg(t, "Z8"), vreg(t, "Z11"), vreg(t, "Z11")}, "6251a548fbd8"},
{"VPUNPCKLDQ Z5,Z3,Z6", "VPUNPCKLDQ", []Operand{vreg(t, "Z5"), vreg(t, "Z3"), vreg(t, "Z6")}, "62f1654862f5"},
{"VPUNPCKHDQ Z5,Z3,Z7", "VPUNPCKHDQ", []Operand{vreg(t, "Z5"), vreg(t, "Z3"), vreg(t, "Z7")}, "62f165486afd"},
{"VPMULLQ Z9,Z10,Z10", "VPMULLQ", []Operand{vreg(t, "Z9"), vreg(t, "Z10"), vreg(t, "Z10")}, "6252ad4840d1"},
{"VPMULLD Z13,Z11,Z2", "VPMULLD", []Operand{vreg(t, "Z13"), vreg(t, "Z11"), vreg(t, "Z2")}, "62d2254840d5"},
{"VPERMD Z0,Z15,Z8", "VPERMD", []Operand{vreg(t, "Z0"), vreg(t, "Z15"), vreg(t, "Z8")}, "6272054836c0"},
// Packed-double arithmetic (EVEX forms carry W=1).
{"VADDPD Z11,Z10,Z10", "VADDPD", []Operand{vreg(t, "Z11"), vreg(t, "Z10"), vreg(t, "Z10")}, "6251ad4858d3"},
{"VMULPD Z13,Z12,Z12", "VMULPD", []Operand{vreg(t, "Z13"), vreg(t, "Z12"), vreg(t, "Z12")}, "62519d4859e5"},
{"VFMADD231PD Z14,Z12,Z10", "VFMADD231PD", []Operand{vreg(t, "Z14"), vreg(t, "Z12"), vreg(t, "Z10")}, "62529d48b8d6"},
// Align (NDS + imm8).
{"VALIGND $12,Z12,Z0,Z1", "VALIGND", []Operand{Imm(12), vreg(t, "Z12"), vreg(t, "Z0"), vreg(t, "Z1")}, "62d37d4803cc0c"},
{"VALIGND $15,Z9,Z0,Z1", "VALIGND", []Operand{Imm(15), vreg(t, "Z9"), vreg(t, "Z0"), vreg(t, "Z1")}, "62d37d4803c90f"},
// Shifts: immediate (/digit) and variable (XMM count).
{"VPSRAD $31,Z3,Z5", "VPSRAD", []Operand{Imm(31), vreg(t, "Z3"), vreg(t, "Z5")}, "62f1554872e31f"},
{"VPSLLD $1,Z3,Z4", "VPSLLD", []Operand{Imm(1), vreg(t, "Z3"), vreg(t, "Z4")}, "62f15d4872f301"},
{"VPSRAQ X31,Z8,Z8", "VPSRAQ", []Operand{vreg(t, "X31"), vreg(t, "Z8"), vreg(t, "Z8")}, "6211bd48e2c7"},
// Mask destinations (the K register occupies the reg field).
{"VPCMPEQD Z0,Z3,K1", "VPCMPEQD", []Operand{vreg(t, "Z0"), vreg(t, "Z3"), vreg(t, "K1")}, "62f1654876c8"},
{"VPCMPEQD Y30,Y11,K1", "VPCMPEQD", []Operand{vreg(t, "Y30"), vreg(t, "Y11"), vreg(t, "K1")}, "6291252876ce"},
// Mask moves and test (VEX-encoded).
{"KMOVW K1,CX", "KMOVW", []Operand{vreg(t, "K1"), CX}, "c5f893c9"},
{"KMOVW K1,R12", "KMOVW", []Operand{vreg(t, "K1"), vreg(t, "R12")}, "c57893e1"},
{"KTESTW K1,K1", "KTESTW", []Operand{vreg(t, "K1"), vreg(t, "K1")}, "c5f899c9"},
// Moves, incl. disp8×N (64 for a 512-bit operand).
{"VMOVDQU32 (SI)(R15*4),Z3", "VMOVDQU32", []Operand{Idx(SI, vreg(t, "R15"), 4, 0, 64), vreg(t, "Z3")}, "62b17e486f1cbe"},
{"VMOVDQU32 4(SI)(AX*1),Z4", "VMOVDQU32", []Operand{Idx(SI, AX, 1, 4, 64), vreg(t, "Z4")}, "62f17e486fa40604000000"},
{"VMOVDQU32 16(SI)(R15*4),Z4", "VMOVDQU32", []Operand{Idx(SI, vreg(t, "R15"), 4, 16, 64), vreg(t, "Z4")}, "62b17e486fa4be10000000"},
{"VMOVDQU32 Z0,4(SI)(AX*1)", "VMOVDQU32", []Operand{vreg(t, "Z0"), Idx(SI, AX, 1, 4, 64)}, "62f17e487f840604000000"},
{"VMOVDQU32 Z3,(DI)(R15*4)", "VMOVDQU32", []Operand{vreg(t, "Z3"), Idx(DI, vreg(t, "R15"), 4, 0, 64)}, "62b17e487f1cbf"},
// VMOVDQU64 — the W1 qword variant.
{"VMOVDQU64 (SI)(R15*4),Z3", "VMOVDQU64", []Operand{Idx(SI, vreg(t, "R15"), 4, 0, 64), vreg(t, "Z3")}, "62b1fe486f1cbe"},
{"VMOVDQU64 Z0,4(SI)(AX*1)", "VMOVDQU64", []Operand{vreg(t, "Z0"), Idx(SI, AX, 1, 4, 64)}, "62f1fe487f840604000000"},
{"VMOVDQU64 Z1,Z2", "VMOVDQU64", []Operand{vreg(t, "Z1"), vreg(t, "Z2")}, "62f1fe487fca"},
{"VMOVUPD (DI),Z14", "VMOVUPD", []Operand{Ptr(DI, 0, 64), vreg(t, "Z14")}, "6271fd481037"},
{"VMOVUPD 64(DI),Z14", "VMOVUPD", []Operand{Ptr(DI, 64, 64), vreg(t, "Z14")}, "6271fd48107701"},
// Conversions and narrowing stores (reg = wide source).
{"VCVTQQ2PD Z12,Z12", "VCVTQQ2PD", []Operand{vreg(t, "Z12"), vreg(t, "Z12")}, "6251fe48e6e4"},
{"VCVTQQ2PD X13,X13", "VCVTQQ2PD", []Operand{vreg(t, "X13"), vreg(t, "X13")}, "6251fe08e6ed"},
{"VPMOVSXDQ 32(SI),Z12", "VPMOVSXDQ", []Operand{Ptr(SI, 32, 32), vreg(t, "Z12")}, "62727d48256601"},
{"VPMOVDW Z0,Y0", "VPMOVDW", []Operand{vreg(t, "Z0"), vreg(t, "Y0")}, "62f27e4833c0"},
{"VPMOVQD Z11,Y11", "VPMOVQD", []Operand{vreg(t, "Z11"), vreg(t, "Y11")}, "62527e4835db"},
// Lane extracts.
{"VEXTRACTI64X4 $1,Z8,Y9", "VEXTRACTI64X4", []Operand{Imm(1), vreg(t, "Z8"), vreg(t, "Y9")}, "6253fd483bc101"},
{"VEXTRACTF64X4 $1,Z10,Y11", "VEXTRACTF64X4", []Operand{Imm(1), vreg(t, "Z10"), vreg(t, "Y11")}, "6253fd481bd301"},
// Broadcasts: GPR source (0x7C) vs memory source (0x58/0x59, disp8×4/8).
{"VPBROADCASTD AX,Z15", "VPBROADCASTD", []Operand{AX, vreg(t, "Z15")}, "62727d487cf8"},
{"VPBROADCASTD (SI),Z8", "VPBROADCASTD", []Operand{Ptr(SI, 0, 4), vreg(t, "Z8")}, "62727d485806"},
{"VPBROADCASTD 4(SI),Z10", "VPBROADCASTD", []Operand{Ptr(SI, 4, 4), vreg(t, "Z10")}, "62727d48585601"},
{"VPBROADCASTQ R8,X31", "VPBROADCASTQ", []Operand{vreg(t, "R8"), vreg(t, "X31")}, "6242fd087cf8"},
{"VPBROADCASTQ AX,Z9", "VPBROADCASTQ", []Operand{AX, vreg(t, "Z9")}, "6272fd487cc8"},
// Register indices 16–31 exist only in EVEX encodings.
{"VPBROADCASTD AX,Y30", "VPBROADCASTD", []Operand{AX, vreg(t, "Y30")}, "62627d287cf0"},
}
for _, c := range cases {
want := strings.ReplaceAll(c.want, " ", "")
code, err := Encode(c.mnem, c.ops...)
if err != nil {
t.Errorf("%s: Encode: %v", c.name, err)
continue
}
if got := hexCompact(code); got != want {
t.Errorf("%s: bytes %s, want %s", c.name, got, want)
continue
}
inst, err := x86asm.Decode(code, 64)
if err != nil {
t.Errorf("%s: Decode(%x): %v", c.name, code, err)
continue
}
if inst.Len != len(code) {
t.Errorf("%s: Decode consumed %d of %d bytes", c.name, inst.Len, len(code))
}
if inst.Op.String() != c.mnem {
t.Errorf("%s: decoded as %s", c.name, inst.Op.String())
}
}
}
// TestEvexErrors checks the EVEX-specific error paths.
func TestEvexErrors(t *testing.T) {
cases := []struct {
name string
mnem string
ops []Operand
}{
{"NDS arity", "VPXORD", []Operand{vreg(t, "Z0"), vreg(t, "Z1")}},
{"KMOVW arity", "KMOVW", []Operand{vreg(t, "K1")}},
{"KMOVW no K", "KMOVW", []Operand{AX, CX}},
{"VMOVUPD Z gpr", "VMOVUPD", []Operand{AX, vreg(t, "Z1")}},
{"broadcast src", "VPBROADCASTD", []Operand{Imm(1), vreg(t, "Z1")}},
{"VPMOVDW src", "VPMOVDW", []Operand{AX, vreg(t, "Y0")}},
{"align arity", "VALIGND", []Operand{Imm(1), vreg(t, "Z0"), vreg(t, "Z1")}},
// VEX-only mnemonics reject registers only EVEX can encode.
{"VPSHUFD X16", "VPSHUFD", []Operand{Imm(1), vreg(t, "X16"), vreg(t, "X17")}},
}
for _, c := range cases {
if _, err := Encode(c.mnem, c.ops...); err == nil {
t.Errorf("%s: expected an error, got none", c.name)
}
}
}
// TestAssembleGoFlacAVX512Kernel assembles the whole production AVX-512
// kernel — all functions plus the file-global idx16 constant — and checks
// that the static-symbol load resolves to the right bytes in the image.
// Skipped when the sibling repository is not checked out.
func TestAssembleGoFlacAVX512Kernel(t *testing.T) {
path := "../../go-libraries/go-flac/avx512_amd64.s"
if _, err := os.Stat(path); err != nil {
t.Skip("go-libraries repository not present next to gasm-devkit")
}
src, err := os.ReadFile(path)
if err != nil {
t.Fatal(err)
}
f, errs := parser.Parse(path, string(src))
if len(errs) > 0 {
t.Fatalf("parse: %v", errs)
}
img, err := AssembleFile(f)
if err != nil {
t.Fatalf("AssembleFile: %v", err)
}
if len(img.Funcs) != 10 {
t.Errorf("functions = %d, want 10", len(img.Funcs))
}
// idx16 as the DATA directives define it: dwords 1..16.
idx := make([]byte, 0, 64)
for i := 1; i <= 16; i++ {
idx = append(idx, byte(i), 0, 0, 0)
}
image := img.Bytes()
base := img.Symbols["idx16"]
if base == 0 {
t.Fatal("idx16 not laid out")
}
if got := image[base : base+64]; hexCompact(got) != hexCompact(idx) {
t.Errorf("idx16 contents %x, want %x", got, idx)
}
// The VMOVDQU32 idx16(SB), Z13 load (62 71 7e 48 6f 2d + rel32) must
// resolve to idx16 within the image.
loads := 0
for _, fn := range img.Funcs {
code := img.Code[fn.Offset : fn.Offset+fn.Size]
pat := []byte{0x62, 0x71, 0x7e, 0x48, 0x6f, 0x2d}
for pos := 0; ; {
i := indexOf(code[pos:], pat)
if i < 0 {
break
}
i += pos
rel := int32(uint32(code[i+6]) | uint32(code[i+7])<<8 | uint32(code[i+8])<<16 | uint32(code[i+9])<<24)
target := fn.Offset + i + 10 + int(rel)
if target != base {
t.Errorf("%s: idx16 load at +%d targets 0x%x, want 0x%x", fn.Name, i, target, base)
}
loads++
pos = i + 10
}
}
if loads != 1 {
t.Errorf("idx16 loads found = %d, want 1", loads)
}
}
// hexCompact renders bytes as a lowercase hex string without separators.
func hexCompact(b []byte) string {
const hexdig = "0123456789abcdef"
out := make([]byte, len(b)*2)
for i, c := range b {
out[i*2] = hexdig[c>>4]
out[i*2+1] = hexdig[c&0xf]
}
return string(out)
}
// indexOf returns the index of the first occurrence of pat in b, or -1.
func indexOf(b, pat []byte) int {
for i := 0; i+len(pat) <= len(b); i++ {
j := 0
for j < len(pat) && b[i+j] == pat[j] {
j++
}
if j == len(pat) {
return i
}
}
return -1
}