[cfe-dev] OpenCL SPIR/NVPTX code generation

[Jordan Rose]

On Jan 23, 2013, at 4:04 , "Benyei, Guy" <guy.benyei at intel.com> wrote:

> Hi Ankur,
> Since you use –Xclang, the clang executable passes multiple triples to “clang -cc1”. You can see that if you add the -v option. I’m sure there is someone here who can explain it better than I…
> Anyhow, I think you better use clang -cc1. Make sure -cc1 is the first command line option you use.
>  
> $ clang -cc1 -fno-builtin -emit-llvm-bc -triple spir-unknown-unknown Simple_Kernel.cl -o Simple_Kernel.bc
>  
> should work for you.

clang -cc1 shouldn't ever be necessary for end users of clang, particularly because cc1 options are not guaranteed to be stable. You also lose all the default options that the driver normally passes down. It's possible this doesn't matter much for SPIR, at least not right now, but it's not something we should be recommending or supporting.

-target is the driver version of -triple. This "worked" for me, where by "worked" I mean "generated some output but I don't know enough about SPIR to validate it".

% clang -x cl -fno-builtin -target spir -c -emit-llvm

Jordan

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20130123/10237ef5/attachment.html>