cmd/asm: add mode to collect symbol ABIs

This adds a -symabis flag that runs the assembler in a special mode
that outputs symbol definition and reference ABIs rather than
assembling the code. This uses a fast and somewhat lax parser because
the go_asm.h definitions may not be available.

For #27539.

Change-Id: I248ba0ebab7cc75dcb2a90e82a82eb445da7e88e
Reviewed-on: https://go-review.googlesource.com/c/147098
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: David Chase <drchase@google.com>
Reviewed-by: Cherry Zhang <cherryyz@google.com>

3 files changed, 125 insertions, 0 deletions

diff --git a/src/cmd/asm/internal/asm/operand_test.go b/src/cmd/asm/internal/asm/operand_test.go
index 69393b6b20..2ba3fd73df 100644
--- a/src/cmd/asm/internal/asm/operand_test.go
+++ b/src/cmd/asm/internal/asm/operand_test.go
@@ -122,6 +122,49 @@ func TestS390XOperandParser(t *testing.T) {
 	testOperandParser(t, parser, s390xOperandTests)
 }
 
+func TestFuncAddress(t *testing.T) {
+	type subtest struct {
+		arch  string
+		tests []operandTest
+	}
+	for _, sub := range []subtest{
+		{"amd64", amd64OperandTests},
+		{"386", x86OperandTests},
+		{"arm", armOperandTests},
+		{"arm64", arm64OperandTests},
+		{"ppc64", ppc64OperandTests},
+		{"mips", mipsOperandTests},
+		{"mips64", mips64OperandTests},
+		{"s390x", s390xOperandTests},
+	} {
+		t.Run(sub.arch, func(t *testing.T) {
+			parser := newParser(sub.arch)
+			for _, test := range sub.tests {
+				parser.start(lex.Tokenize(test.input))
+				name, ok := parser.funcAddress()
+
+				isFuncSym := strings.HasSuffix(test.input, "(SB)") &&
+					// Ignore static symbols.
+					!strings.Contains(test.input, "<>") &&
+					// Ignore symbols with offsets.
+					!strings.Contains(test.input, "+")
+
+				wantName := ""
+				if isFuncSym {
+					// Strip $|* and (SB).
+					wantName = test.output[:len(test.output)-4]
+					if strings.HasPrefix(wantName, "$") || strings.HasPrefix(wantName, "*") {
+						wantName = wantName[1:]
+					}
+				}
+				if ok != isFuncSym || name != wantName {
+					t.Errorf("fail at %s as function address: got %s, %v; expected %s, %v", test.input, name, ok, wantName, isFuncSym)
+				}
+			}
+		})
+	}
+}
+
 type operandTest struct {
 	input, output string
 }
diff --git a/src/cmd/asm/internal/asm/parse.go b/src/cmd/asm/internal/asm/parse.go
index 3620e31320..346976ef48 100644
--- a/src/cmd/asm/internal/asm/parse.go
+++ b/src/cmd/asm/internal/asm/parse.go
@@ -116,6 +116,22 @@ func (p *Parser) Parse() (*obj.Prog, bool) {
 	return p.firstProg, true
 }
 
+// ParseSymABIs parses p's assembly code to find text symbol
+// definitions and references and writes a symabis file to w.
+func (p *Parser) ParseSymABIs(w io.Writer) bool {
+	operands := make([][]lex.Token, 0, 3)
+	for {
+		word, _, operands1, ok := p.line(operands)
+		if !ok {
+			break
+		}
+		operands = operands1
+
+		p.symDefRef(w, word, operands)
+	}
+	return p.errorCount == 0
+}
+
 // line consumes a single assembly line from p.lex of the form
 //
 //   {label:} WORD[.cond] [ arg {, arg} ] (';' | '\n')
@@ -258,6 +274,42 @@ func (p *Parser) pseudo(word string, operands [][]lex.Token) bool {
 	return true
 }
 
+// symDefRef scans a line for potential text symbol definitions and
+// references and writes symabis information to w.
+//
+// The symabis format is documented at
+// cmd/compile/internal/gc.readSymABIs.
+func (p *Parser) symDefRef(w io.Writer, word string, operands [][]lex.Token) {
+	switch word {
+	case "TEXT":
+		// Defines text symbol in operands[0].
+		if len(operands) > 0 {
+			p.start(operands[0])
+			if name, ok := p.funcAddress(); ok {
+				fmt.Fprintf(w, "def %s ABI0\n", name)
+			}
+		}
+		return
+	case "GLOBL", "PCDATA":
+		// No text definitions or symbol references.
+	case "DATA", "FUNCDATA":
+		// For DATA, operands[0] is defined symbol.
+		// For FUNCDATA, operands[0] is an immediate constant.
+		// Remaining operands may have references.
+		if len(operands) < 2 {
+			return
+		}
+		operands = operands[1:]
+	}
+	// Search for symbol references.
+	for _, op := range operands {
+		p.start(op)
+		if name, ok := p.funcAddress(); ok {
+			fmt.Fprintf(w, "ref %s ABI0\n", name)
+		}
+	}
+}
+
 func (p *Parser) start(operand []lex.Token) {
 	p.input = operand
 	p.inputPos = 0
@@ -746,6 +798,35 @@ func (p *Parser) setPseudoRegister(addr *obj.Addr, reg string, isStatic bool, pr
 	}
 }
 
+// funcAddress parses an external function address. This is a
+// constrained form of the operand syntax that's always SB-based,
+// non-static, and has no additional offsets:
+//
+//    [$|*]sym(SB)
+func (p *Parser) funcAddress() (string, bool) {
+	switch p.peek() {
+	case '$', '*':
+		// Skip prefix.
+		p.next()
+	}
+
+	tok := p.next()
+	name := tok.String()
+	if tok.ScanToken != scanner.Ident || p.atStartOfRegister(name) {
+		return "", false
+	}
+	if p.next().ScanToken != '(' {
+		return "", false
+	}
+	if reg := p.next(); reg.ScanToken != scanner.Ident || reg.String() != "SB" {
+		return "", false
+	}
+	if p.next().ScanToken != ')' || p.peek() != scanner.EOF {
+		return "", false
+	}
+	return name, true
+}
+
 // registerIndirect parses the general form of a register indirection.
 // It is can be (R1), (R2*scale), (R1)(R2*scale), (R1)(R2.SXTX<<3) or (R1)(R2<<3)
 // where R1 may be a simple register or register pair R:R or (R, R) or (R+R).
diff --git a/src/cmd/asm/internal/flags/flags.go b/src/cmd/asm/internal/flags/flags.go
index 6acde29432..752a1d4526 100644
--- a/src/cmd/asm/internal/flags/flags.go
+++ b/src/cmd/asm/internal/flags/flags.go
@@ -22,6 +22,7 @@ var (
 	Shared     = flag.Bool("shared", false, "generate code that can be linked into a shared library")
 	Dynlink    = flag.Bool("dynlink", false, "support references to Go symbols defined in other shared libraries")
 	AllErrors  = flag.Bool("e", false, "no limit on number of errors reported")
+	SymABIs    = flag.Bool("symabis", false, "write symbol ABI information to output file, don't assemble")
 )
 
 var (