[llvm-bugs] [Bug 41070] New: Failure to compile CL code with address space conversion

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=41070

            Bug ID: 41070
           Summary: Failure to compile CL code with address space
                    conversion
           Product: clang
           Version: trunk
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: enhancement
          Priority: P
         Component: OpenCL
          Assignee: unassignedclangbugs at nondot.org
          Reporter: anastasia.stulova at arm.com
                CC: anastasia.stulova at arm.com, drohr at jwdt.org,
                    llvm-bugs at lists.llvm.org, Matthew.Arsenault at amd.com,
                    yaxun.liu at amd.com

I am creating this bug on behalf of David although he is experiencing slightly
different behavior that we are discussing on this bug:
https://bugs.llvm.org/show_bug.cgi?id=40778

while working on his code I have encountered the following failure if OpenCL
code like this is compiled:

generic int* foo(private int* i){
 return i;
}

Example command line to reproduce:
clang -cl-std=CL2.0 -x cl --target=amdgcn-amd-amdhsa a.cl


It fails with:
fatal error: error in backend: Attempting to emit FLAT_LOAD_DWORD_ci
instruction but the Feature_isCIOnly predicate(s) are not met


Not sure it is expected to work or not. So I am adding Matt and Sam that might
be able to shed some light on the problem. Feel free to reclassify the bug if
needed. 
Thanks!

-- 
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/20190314/259f9f4f/attachment.html>