[llvm-bugs] [Bug 35998] New: [AMDGPU][MC] 64-bit image atomics are not supported

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=35998

            Bug ID: 35998
           Summary: [AMDGPU][MC] 64-bit image atomics are not supported
           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

According with AMD documentation, 

"Image atomic operations are supported on 32- and 64-bit-per pixel surfaces"

"For atomic operations, DMASK is set to the number of VGPRs (Dwords) to send to
the texture unit. DMASK legal values for atomic image operations:

0x1 = 32-bit atomics except cmpswap.
0x3 = 32-bit atomic cmpswap.
0x3 = 64-bit atomics except cmpswap.
0xf = 64-bit atomic cmpswap."

Currently assembler supports 32-bit atomics only. An example of a failed test:

    image_atomic_add v[5:6], v1, s[8:15] dmask:0x3

-- 
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/20180118/d340e6b8/attachment.html>