[PATCH] [NVPTX] noop when kernel pointers are already global

[Jingyue Wu]

Yes. I'll have Tobias test case crash again :)

On Tue, Jun 30, 2015 at 9:37 AM, Justin Holewinski <jholewinski at nvidia.com>
wrote:

> Agreed.  Jingyue, did you want to make this a part of your patch?
>
> > On Jun 30, 2015, at 12:30 PM, Tobias Grosser <tobias at grosser.es> wrote:
> >
> > On 06/30/2015 06:26 PM, Justin Holewinski wrote:
> >> Hi Tobias,
> >>
> >> One thing to keep in mind is that the CUDA interface expects kernel
> parameters to be in the generic address space.  If you write kernels that
> pass arguments as global memory pointers, you are technically violating the
> contract between host and device.  Now, this just so happens to work since
> the global->generic and generic->global pointer conversions are no-ops, but
> as far as I know this is not actually guaranteed and may not be true in the
> future.
> >
> > Interesting. Then maybe the right solution is not to allow this, but
> rather to give a proper error message in the backend (instead of just
> crashing)?
> >
> > Tobias
>
>
> -----------------------------------------------------------------------------------
> This email message is for the sole use of the intended recipient(s) and
> may contain
> confidential information.  Any unauthorized review, use, disclosure or
> distribution
> is prohibited.  If you are not the intended recipient, please contact the
> sender by
> reply email and destroy all copies of the original message.
>
> -----------------------------------------------------------------------------------
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150630/c584fea3/attachment.html>