llvm dev - Aug 2012 - [LLVMdev] clang thread-local compilation error on windows

Hello, I am trying to compile some code to LLVM IR with a simple "__thread
int x" but hitting this error:
    test.cpp:1:1: error: thread-local storage is unsupported for the
current target

I'm using both the -S and -emit-llvm options on clang, and was expecting to
see "@x = thread_local global i32 0" come out of clang.

I am curious why clang even cares about this since its my understanding it
woud just attach thread_local to the declaration and then allow me to
handle it later on in llc to do the actual traget specific implementation.
 Can someone help clear where the thread-local implementation should go?

Thanks,
-Stephen
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20120808/2f811fe2/attachment.html>

On Wed, Aug 8, 2012 at 12:04 PM, Stephen Schiffli <sschiffli at gmail.com>
wrote:
> Hello, I am trying to compile some code to LLVM IR with a simple
"__thread
> int x" but hitting this error:
>     test.cpp:1:1: error: thread-local storage is unsupported for the
current
> target
>
> I'm using both the -S and -emit-llvm options on clang, and was
expecting to
> see "@x = thread_local global i32 0" come out of clang.
>
> I am curious why clang even cares about this since its my understanding it
> woud just attach thread_local to the declaration and then allow me to
handle
> it later on in llc to do the actual traget specific implementation.  Can
> someone help clear where the thread-local implementation should go?
>
> Thanks,
> -Stephen
It's saying this because it knows that LLVM doesn't support thread
local storage on Windows. Although I believe this has changed
recently. The reason clang is reporting it is because in general it is
nicer for a user to get a nice diagnostic from clang instead of an
assert from inside codegen.

- Michael Spencer