[llvm-commits] [llvm] r166445 - in /llvm/trunk: include/llvm-c/Disassembler.h include/llvm/MC/MCInstPrinter.h lib/MC/MCDisassembler/Disassembler.cpp lib/Target/ARM/InstPrinter/ARMInstPrinter.cpp lib/Target/X86/InstPrinter/X86ATTInstPrinter.cpp te

[Meador Inge]

On 10/22/2012 05:41 PM, Chandler Carruth wrote:

> On Mon, Oct 22, 2012 at 3:31 PM, Kevin Enderby <enderby at apple.com> wrote:
>> @@ -265,8 +294,20 @@
>>    const MCOperand &MO1 = MI->getOperand(OpNum);
>>    if (MO1.isExpr())
>>      O << *MO1.getExpr();
>> -  else if (MO1.isImm())
>> -    O << "[pc, #" << MO1.getImm() << "]";
>> +  else if (MO1.isImm()) {
>> +    if (UseMarkup)
>> +      O << "<mem:";
>> +    O << "[pc, ";
>> +    if (UseMarkup)
>> +      O << "<imm:";
>> +    O << "#";
>> +    O << MO1.getImm();
>> +    if (UseMarkup)
>> +      O << ">";
>> +    O << "]";
>> +    if (UseMarkup)
>> +      O << ">";
>> +  }
> 
> I feel like a lot of these branch heavy changes could be made simpler
> with a tiny helper function that either pases a twine through or
> returns an empty twine:
> 
> O << markup("<mem:") << "[pc, " << markup("<imm:") << "#" <<
> MO1.getImm() << markup(">]>", "]");
> 
> Thoughts?

FWIW, I find your method much easier to read Chandler.

-- 
Meador Inge
CodeSourcery / Mentor Embedded
http://www.mentor.com/embedded-software