[PATCH] D56411: [CUDA][HIP][Sema] Fix template kernel with function as template parameter

[John McCall via Phabricator via cfe-commits]

rjmccall added a comment.

But what we've just been talking about is not a validity rule, it's an overload-resolution rule.  It's not *invalid* to use a device function as a template argument to a host function template (or to a class template, which of course is neither host nor device).  All you need to do is to resolve otherwise-intractable overload ambiguities by matching with the host-ness of the current context, which there's probably already code to do for when an overload set is used as e.g. a function argument.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D56411/new/

https://reviews.llvm.org/D56411