fix(asm): encode the CMOV condition the renderer prints

The renderer spells a conditional move CMOV plus the condition alone
(CMOVLE, CMOVG), the width carried by the operand registers, so
CMOVLE parsed as the size L and the condition E and encoded CMOVE.
A suffix that is itself a condition name now reads as that condition
with the width from the destination register, and the Plan 9
size-prefixed spellings keep their parse.

Assisted-by: GLM 5.3 Flash
This commit is contained in:
petrbalvin committed 2026-10-07 02:12:21 +02:00
1 parent 11cac26508
commit abc2d32b83
3 files changed
+63 -29

No files matched your search

+7 -9
View File
@@ -62,18 +62,16 @@ func Encodable(mnemonic string) bool {
return true return true
} }
// CMOV carries size then condition (CMOVLGT); SET carries the condition // CMOV carries size then condition (CMOVLGT), or the renderer's
// alone (SETNE). The size letter is checked exactly as encodeCmov does, // condition alone (CMOVLE) with the width from the operand; SET carries
// so a spelling like CMOVBGT is not reported encodable when Encode // the condition alone (SETNE). cmovCondition checks the suffix exactly
// would reject it. // as encodeCmov does, so a spelling like CMOVBGT is not reported
if rest, ok := strings.CutPrefix(upper, "CMOV"); ok && len(rest) >= 2 { // encodable when Encode would reject it.
switch rest[0] { if rest, ok := strings.CutPrefix(upper, "CMOV"); ok {
case 'W', 'L', 'Q': if _, _, ok := cmovCondition(rest); ok {
if _, ok := jccMap[rest[1:]]; ok {
return true return true
} }
} }
}
if rest, ok := strings.CutPrefix(upper, "SET"); ok { if rest, ok := strings.CutPrefix(upper, "SET"); ok {
if _, ok := jccMap[rest]; ok { if _, ok := jccMap[rest]; ok {
return true return true
+14 -3
View File
@@ -411,6 +411,16 @@ func TestScalarGroundTruth(t *testing.T) {
{"CMOVLEQ CX,AX", "CMOVLEQ", []Operand{CX, AX}, "0f44c1", "CMOVE"}, {"CMOVLEQ CX,AX", "CMOVLEQ", []Operand{CX, AX}, "0f44c1", "CMOVE"},
{"CMOVQGT R9,R8", "CMOVQGT", []Operand{r9, r8}, "4d0f4fc1", "CMOVG"}, {"CMOVQGT R9,R8", "CMOVQGT", []Operand{r9, r8}, "4d0f4fc1", "CMOVG"},
{"CMOVWLS R9W,R8W", "CMOVWLS", []Operand{r9w, r8w}, "66450f46c1", "CMOVBE"}, {"CMOVWLS R9W,R8W", "CMOVWLS", []Operand{r9w, r8w}, "66450f46c1", "CMOVBE"},
// The renderer's condition spellings carry no size letter; the width
// rides the operand registers and the bytes match the toolchain's own
// size-prefixed spellings (CMOVQLE/CMOVLLE/CMOVWLE pinned from go tool
// asm). CMOVLE with the 16-bit registers reproduces the disasm
// fixture's 660f4e13 row byte for byte.
{"CMOVLE (BX),DX", "CMOVLE", []Operand{Ptr(BX, 0, 2), DX}, "660f4e13", "CMOVLE"},
{"CMOVQLE AX,BX", "CMOVQLE", []Operand{Reg{idx: 0, size: 8}, Reg{idx: 3, size: 8}}, "480f4ed8", "CMOVLE"},
{"CMOVLLE AX,BX", "CMOVLLE", []Operand{Reg{idx: 0, size: 4}, Reg{idx: 3, size: 4}}, "0f4ed8", "CMOVLE"},
{"CMOVWLE AX,BX", "CMOVWLE", []Operand{AX, BX}, "660f4ed8", "CMOVLE"},
{"CMOVB AL,CL", "CMOVB", []Operand{AL, CL}, "0f42c8", "CMOVB"},
{"SETNE AL", "SETNE", []Operand{AL}, "0f95c0", "SETNE"}, {"SETNE AL", "SETNE", []Operand{AL}, "0f95c0", "SETNE"},
{"SETNE (AX)", "SETNE", []Operand{Ptr(AX, 0, 1)}, "0f9500", "SETNE"}, {"SETNE (AX)", "SETNE", []Operand{Ptr(AX, 0, 1)}, "0f9500", "SETNE"},
{"MOVBLZX AL,CX", "MOVBLZX", []Operand{AL, CX}, "0fb6c8", "MOVZX"}, {"MOVBLZX AL,CX", "MOVBLZX", []Operand{AL, CX}, "0fb6c8", "MOVZX"},
@@ -595,14 +605,15 @@ func TestImmediateTruncation(t *testing.T) {
// TestEncodableCmovSize pins the linter contract for CMOVcc: Encodable must // TestEncodableCmovSize pins the linter contract for CMOVcc: Encodable must
// reject the spellings Encode rejects, so a mnemonic like CMOVBGT (no size // reject the spellings Encode rejects, so a mnemonic like CMOVBGT (no size
// letter) is not reported as encodable. // letter) is not reported as encodable. The condition-name spellings the
// renderer prints (CMOVB, CMOVLE) encode with the width from the operands.
func TestEncodableCmovSize(t *testing.T) { func TestEncodableCmovSize(t *testing.T) {
for _, m := range []string{"CMOVBGT", "CMOVXEQ", "CMOVB", "CMOV", "CMOVWXX"} { for _, m := range []string{"CMOVBGT", "CMOVXEQ", "CMOV", "CMOVWXX"} {
if Encodable(m) { if Encodable(m) {
t.Errorf("Encodable(%q) = true, want false", m) t.Errorf("Encodable(%q) = true, want false", m)
} }
} }
for _, m := range []string{"CMOVLGT", "CMOVQGT", "CMOVWLS", "CMOVLEQ"} { for _, m := range []string{"CMOVLGT", "CMOVQGT", "CMOVWLS", "CMOVLEQ", "CMOVB", "CMOVLE"} {
if !Encodable(m) { if !Encodable(m) {
t.Errorf("Encodable(%q) = false, want true", m) t.Errorf("Encodable(%q) = false, want true", m)
} }
+42 -17
View File
@@ -1068,29 +1068,46 @@ func immediate(v int64, size int, full64 bool) ([]byte, error) {
// --- CMOVcc / SETcc --------------------------------------------------------- // --- CMOVcc / SETcc ---------------------------------------------------------
// cmovCondition splits a CMOVcc suffix into an optional size letter and the
// condition code. Two vocabularies meet here: the Plan 9 spellings prefix
// the condition with a size letter (CMOVLGT, CMOVQEQ), while the toolchain's
// renderer prints the condition alone (CMOVLE, CMOVG) and leaves the width
// to the operand registers. A suffix that is itself a condition name reads
// as that condition, so the renderer's text re-encodes; CMOVBGT, CMOVWXX and
// the bare CMOV still find no condition and stay rejected.
func cmovCondition(rest string) (size int, cc int, ok bool) {
if cc, ok := jccMap[rest]; ok {
return 0, cc, true
}
if len(rest) >= 2 {
switch rest[0] {
case 'W':
if cc, ok := jccMap[rest[1:]]; ok {
return 2, cc, true
}
case 'L':
if cc, ok := jccMap[rest[1:]]; ok {
return 4, cc, true
}
case 'Q':
if cc, ok := jccMap[rest[1:]]; ok {
return 8, cc, true
}
}
}
return 0, 0, false
}
// encodeCmov encodes a conditional move: CMOV + size (W/L/Q) + condition // encodeCmov encodes a conditional move: CMOV + size (W/L/Q) + condition
// (CMOVLGT, CMOVQEQ, …). The condition reads exactly like the Jcc spellings; // (CMOVLGT, CMOVQEQ, …), or the renderer's condition alone (CMOVLE) with the
// the instruction is 0F 40+cc with reg = dst, rm = src. // width taken from the destination register. The instruction is 0F 40+cc
// with reg = dst, rm = src.
func (e *enc) encodeCmov(upper string, ops []Operand) error { func (e *enc) encodeCmov(upper string, ops []Operand) error {
if len(ops) != 2 { if len(ops) != 2 {
return fmt.Errorf("CMOVcc expects 2 operands, got %d", len(ops)) return fmt.Errorf("CMOVcc expects 2 operands, got %d", len(ops))
} }
rest := upper[len("CMOV"):] rest := upper[len("CMOV"):]
if len(rest) < 2 { size, cc, ok := cmovCondition(rest)
return fmt.Errorf("unsupported instruction %q", upper)
}
var size int
switch rest[0] {
case 'W':
size = 2
case 'L':
size = 4
case 'Q':
size = 8
default:
return fmt.Errorf("unsupported instruction %q", upper)
}
cc, ok := jccMap[rest[1:]]
if !ok { if !ok {
return fmt.Errorf("unsupported instruction %q", upper) return fmt.Errorf("unsupported instruction %q", upper)
} }
@@ -1099,6 +1116,14 @@ func (e *enc) encodeCmov(upper string, ops []Operand) error {
if !ok { if !ok {
return fmt.Errorf("CMOVcc destination must be a register") return fmt.Errorf("CMOVcc destination must be a register")
} }
if size == 0 {
// The renderer's spelling carries no size letter: the width rides
// the destination register's own size class.
size = dstReg.size
if size != 1 && size != 2 && size != 4 && size != 8 {
return fmt.Errorf("CMOVcc destination must be a general register")
}
}
i := newInstr(size, []byte{0x0F, byte(0x40 + cc)}) i := newInstr(size, []byte{0x0F, byte(0x40 + cc)})
if err := setRM(i, dstReg, src, size); err != nil { if err := setRM(i, dstReg, src, size); err != nil {
return err return err