[MC] Allowing the use of $-prefixed integer as asm identifiers

Summary: Dollar signed prefixed integers were not allowed by the AsmParser to be used as Identifiers, differing from the GNU assembler behavior. This patch updates the parsing of Identifiers to consider such cases as valid, where the identifier string includes the $ prefix itself. As the Lexer currently splits these occurrences into separate tokens, those need to be combined by the AsmParser itself. Reviewers: efriedma, chill Reviewed By: efriedma Subscribers: sdardis, hiraditya, jrtc27, atanasyan, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D75111
llvm · Mar 6, 2020 · 0ba553d · 0ba553d
1 parent af1c2e5
commit 0ba553d
Show file tree

Hide file tree

Showing 6 changed files with 14 additions and 12 deletions.
diff --git a/llvm/lib/MC/MCParser/AsmParser.cpp b/llvm/lib/MC/MCParser/AsmParser.cpp
@@ -2854,18 +2854,18 @@ bool AsmParser::parseIdentifier(StringRef &Res) {
     AsmToken Buf[1];
     Lexer.peekTokens(Buf, false);
 
-    if (Buf[0].isNot(AsmToken::Identifier))
+    if (Buf[0].isNot(AsmToken::Identifier) && Buf[0].isNot(AsmToken::Integer))
       return true;
 
-    // We have a '$' or '@' followed by an identifier, make sure they are adjacent.
+    // We have a '$' or '@' followed by an identifier or integer token, make
+    // sure they are adjacent.
     if (PrefixLoc.getPointer() + 1 != Buf[0].getLoc().getPointer())
       return true;
 
     // eat $ or @
     Lexer.Lex(); // Lexer's Lex guarantees consecutive token.
     // Construct the joined identifier and consume the token.
-    Res =
-        StringRef(PrefixLoc.getPointer(), getTok().getIdentifier().size() + 1);
+    Res = StringRef(PrefixLoc.getPointer(), getTok().getString().size() + 1);
     Lex(); // Parser Lex to maintain invariants.
     return false;
   }

diff --git a/llvm/test/MC/ARM/arm-branches.s b/llvm/test/MC/ARM/arm-branches.s
@@ -19,17 +19,22 @@
 @------------------------------------------------------------------------------
 
         .global $foo
+        .global $4
         b $foo
         bl $foo
         beq $foo
         blx $foo
         b $foo + 4
+        bl $4
+        beq $4 + 4
 
 @ CHECK: b      ($foo)                      @ encoding: [A,A,A,0xea]
 @ CHECK: bl     ($foo)                      @ encoding: [A,A,A,0xeb]
 @ CHECK: beq    ($foo)                      @ encoding: [A,A,A,0x0a]
 @ CHECK: blx    ($foo)                      @ encoding: [A,A,A,0xfa]
 @ CHECK: b      #($foo)+4                   @ encoding: [A,A,A,0xea]
+@ CHECK: bl     ($4)                        @ encoding: [A,A,A,0xeb]
+@ CHECK: beq    #($4)+4                     @ encoding: [A,A,A,0x0a]
 
 @------------------------------------------------------------------------------
 @ Leading '$' should be allowed to introduce an expression

diff --git a/llvm/test/MC/MachO/bad-dollar.s b/llvm/test/MC/MachO/bad-dollar.s
diff --git a/llvm/test/MC/MachO/dollar-identifier.s b/llvm/test/MC/MachO/dollar-identifier.s
@@ -0,0 +1,4 @@
+// RUN: llvm-mc -triple x86_64-apple-darwin10 %s | FileCheck %s
+
+.long $1
+// CHECK: .long ($1)
diff --git a/llvm/test/MC/Mips/cpsetup-bad.s b/llvm/test/MC/Mips/cpsetup-bad.s
@@ -12,8 +12,6 @@ t1:
 # ASM: :[[@LINE-1]]:23: error: expected save register or stack offset
         .cpsetup $31, $32, __cerror
 # ASM: :[[@LINE-1]]:23: error: invalid register
-        .cpsetup $25, $2, $3
-# ASM: :[[@LINE-1]]:27: error: expected expression
         .cpsetup $25, $2, 4
 # ASM: :[[@LINE-1]]:28: error: expected symbol
         .cpsetup $25, $2, 4+65

diff --git a/llvm/test/MC/Mips/invalid-instructions-spellcheck.s b/llvm/test/MC/Mips/invalid-instructions-spellcheck.s
@@ -13,7 +13,7 @@
 
 $2, $1, $25
 
-# ALL:      error: unexpected token at start of statement
+# ALL:      error: unknown instruction
 # ALL-NEXT: $2, $1, $25
 # ALL-NEXT:  ^