ABataev added a comment.

In D83261#2166766 <https://reviews.llvm.org/D83261#2166766>, @Meinersbur wrote:

> In D83261#2162561 <https://reviews.llvm.org/D83261#2162561>, @ABataev wrote:
>
> > 1. OMPChildren class uses standard TrailingObjects harness instead of 
> > manual calculation.
>
>
> Note that that having a separate object defeats the purpose of 
> `TrailingObjects` of having just a single allocation per AST node. If we do 
> separate objects, we could also have member pointers to arrays.


I know. Will check what I can do about it.

> 
> 
> In D83261#2164929 <https://reviews.llvm.org/D83261#2164929>, @ABataev wrote:
> 
>> Sure, we can make `OMPChildren` common for declarative and executable 
>> directives. Do you want me to do it?
> 
> 
> Yes, I think it would increase its usefulness and remove code duplication of 
> handling clauses.
> 
>>>> There should be an additional patch, which, I hope, should simplify things 
>>>> for loop-based directives.
>>> 
>>> OK. What does this patch do? Are you going to upload it as well?
>> 
>> At first, need to deal with this one, at least come to an agreement with the 
>> design.
> 
> The reviewer list is surprisingly small. Aren't there any others with stakes 
> in the class hierarchy?




Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D83261/new/

https://reviews.llvm.org/D83261



_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to