[llvm-bugs] [Bug 25126] New: PPC disassembly should tag registers with r

via llvm-bugs llvm-bugs at lists.llvm.org
Fri Oct 9 11:36:24 PDT 2015


https://llvm.org/bugs/show_bug.cgi?id=25126

            Bug ID: 25126
           Summary: PPC disassembly should tag registers with r
           Product: tools
           Version: trunk
          Hardware: Other
                OS: Linux
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: llvm-dis
          Assignee: unassignedbugs at nondot.org
          Reporter: kyle+llvm at iteratee.net
                CC: llvm-bugs at lists.llvm.org
    Classification: Unclassified

For legibility, register N should be marked as "rN" instead of just N.
gcc does this:

0000000000000030 <_GLOBAL__sub_I_libA.cc>:
  30:   00 00 4c 3c     addis   r2,r12,0
                        30: R_PPC64_REL16_HA    .TOC.
  34:   00 00 42 38     addi    r2,r2,0
                        34: R_PPC64_REL16_LO    .TOC.+0x4
  38:   a6 02 08 7c     mflr    r0
  3c:   f8 ff e1 fb     std     r31,-8(r1)
  40:   10 00 01 f8     std     r0,16(r1)
  44:   91 ff 21 f8     stdu    r1,-112(r1)
  48:   78 0b 3f 7c     mr      r31,r1

compare for legibility with this:

_GLOBAL__sub_I_libA.cc:
      30:       00 00 4c 3c     addis 2, 12, 0
                0000000000000030:  R_PPC64_REL16_HA     Unknown
      34:       00 00 42 38     addi 2, 2, 0
                0000000000000034:  R_PPC64_REL16_LO     Unknown
      38:       a6 02 08 7c     mflr 0
      3c:       f8 ff e1 fb     std 31, -8(1)
      40:       10 00 01 f8     std 0, 16(1)
      44:       91 ff 21 f8     stdu 1, -112(1)
      48:       78 0b 3f 7c     mr 31, 1

tagging registers makes it easier to visually differentiate registers from
immediates without having to keep the full instruction set in memory.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20151009/2fdfb1b0/attachment.html>


More information about the llvm-bugs mailing list