[llvm-bugs] [Bug 39299] New: [AMDGPU][MC] Specifying GLC modifier with MUBUF atomics result in incorrect OFFSET operand

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=39299

            Bug ID: 39299
           Summary: [AMDGPU][MC] Specifying GLC modifier with MUBUF
                    atomics result in incorrect OFFSET operand
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: All
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: Backend: AMDGPU
          Assignee: unassignedbugs at nondot.org
          Reporter: dpreobrazhensky at luxoft.com
                CC: llvm-bugs at lists.llvm.org

This issue does not affect any meaningful code. It only mainifests itself when
offset operand is a floating-point inline constant and GLC is specified. 

There are no issues without GLC:

  buffer_atomic_add v255, v0, s[12:15], -4.0 idxen

This is encoded as expected: [0x00,0x20,0x08,0xe1,0x00,0xff,0x03,0xf7]

Adding GLC modifier results in incorrect encoding of soffset:

  buffer_atomic_add v255, v0, s[12:15], -4.0 idxen glc

encoding: [0x00,0x60,0x08,0xe1,0x00,0xff,0x03,0x80]

When disassembled, the code results in:

   buffer_atomic_add v255, v0, s[12:15], 0 idxen glc

This looks like a minor issue, but should be investigated.

-- 
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/20181015/2e7d458c/attachment.html>