diff options
| author | Amaury de la Vieuville <amaury.dlv@gmail.com> | 2013-06-18 08:13:05 +0000 |
|---|---|---|
| committer | Amaury de la Vieuville <amaury.dlv@gmail.com> | 2013-06-18 08:13:05 +0000 |
| commit | beb920fce6ccc89b4735f280f94cb8c227f4ef5e (patch) | |
| tree | b05f8b29168dcdff8045ea3d5b0f859beab8b8a0 /lib/Target/ARM/MCTargetDesc/ARMMCCodeEmitter.cpp | |
| parent | f8b60d6f30a8f25c84a71d36ff3a86fe1f52f671 (diff) | |
| download | external_llvm-beb920fce6ccc89b4735f280f94cb8c227f4ef5e.zip external_llvm-beb920fce6ccc89b4735f280f94cb8c227f4ef5e.tar.gz external_llvm-beb920fce6ccc89b4735f280f94cb8c227f4ef5e.tar.bz2 | |
ARM: fix literal load with positive offset encoding
When using a positive offset, literal loads where encoded
as if it was negative, because:
- The sign bit was not assigned to an operand
- The addrmode_imm12 operand was not encoding the sign bit correctly
This patch also makes the assembler look at the .w/.n specifier for
loads.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@184182 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/Target/ARM/MCTargetDesc/ARMMCCodeEmitter.cpp')
| -rw-r--r-- | lib/Target/ARM/MCTargetDesc/ARMMCCodeEmitter.cpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/lib/Target/ARM/MCTargetDesc/ARMMCCodeEmitter.cpp b/lib/Target/ARM/MCTargetDesc/ARMMCCodeEmitter.cpp index f324bc2..8631d81 100644 --- a/lib/Target/ARM/MCTargetDesc/ARMMCCodeEmitter.cpp +++ b/lib/Target/ARM/MCTargetDesc/ARMMCCodeEmitter.cpp @@ -743,10 +743,10 @@ getAddrModeImm12OpValue(const MCInst &MI, unsigned OpIdx, if (!MO.isReg()) { Reg = CTX.getRegisterInfo()->getEncodingValue(ARM::PC); // Rn is PC. Imm12 = 0; - isAdd = false ; // 'U' bit is set as part of the fixup. if (MO.isExpr()) { const MCExpr *Expr = MO.getExpr(); + isAdd = false ; // 'U' bit is set as part of the fixup. MCFixupKind Kind; if (isThumb2()) |