Hi,
I have a project to rewrite llvm output as OpenCL,
https://github.com/hughperkins/cuda-on-cl It currently contains lots of
conditional branches, whch become /labels / gotos in the output OpenCL.
What options are available to 'reloop' this to contain strictly
ifs/fors/switches/breaks/whiles ?
(Oh i see, the NamedMDNode contains MDNodes )
On Tue, Sep 27, 2016 at 3:22 PM, Hugh Perkins wrote:
> In the IR text, I see lines like:
>
> !0 = !{void (float*)* @_Z11somekernel1Pf, !"kernel", i32 1}
> !1 = !{void (float*)* @_Z3fooPf, !"kernel", i32 1}
>
> What are these? How to retrieve them, g
In the IR text, I see lines like:
!0 = !{void (float*)* @_Z11somekernel1Pf, !"kernel", i32 1}
!1 = !{void (float*)* @_Z3fooPf, !"kernel", i32 1}
What are these? How to retrieve them, given eg a Module *. I've tried
iterating over the global variables, named metadata, and aliases, but
nothing.
Question: is there a converter from IR to OpenCL?
I know there are SPIR-V converters around. But SPIR-V is not very
supported yet. IR written as OpenCL should probably be really ugly, but
maybe useable-ish?
___
cfe-users mailing list
cfe-users@lists.ll
( found it: --cuda-device-only cool :-) )
On Tue, Sep 27, 2016 at 12:17 AM, Hugh Perkins
wrote:
> I'm trying:
>
> clang++-3.8 -I/usr/local/cuda-7.5/include llvm-sample.cu -emit-llvm -S -o
> llvm-sample.ll
>
> This seems to emit host-side IR only?
>
_
I'm trying:
clang++-3.8 -I/usr/local/cuda-7.5/include llvm-sample.cu -emit-llvm -S -o
llvm-sample.ll
This seems to emit host-side IR only?
___
cfe-users mailing list
cfe-users@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-users
