[PATCH] D18010: AMDGPU: R600 code splitting cleanup

[David Blaikie via llvm-commits]

On Wed, Mar 9, 2016 at 6:11 PM, Matt Arsenault <Matthew.Arsenault at amd.com>
wrote:

> On 03/09/2016 05:21 PM, David Blaikie wrote:
>
>> What's the reason for marking the classes final? We don't usually go out
>> of our way to do that, except for -Wnon-virtual-dtor silencing (in my
>> experience/practice) - so perhaps it's that case? (just not clear to me
>> from the commit/code (without going & looking at the hierarchies, etc)
>>
> Because they can be. There's no reason to further subclass these, and
> might as well allow the compiler to emit direct calls to them.


I don't think we've really made any habit of that across LLVM - I'd be
slightly hesitant to bother without some evidence that it helps, but it's
just a single keyword, so I'm not terribly fussed if you prefer it that way.

- Dave


>
>
> -Matt
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160309/975b6d0b/attachment.html>