[PATCH] D35384: PGOInstrumentation: Move profile matching warnings to remarks

Thu Jul 13 22:53:14 PDT 2017

On Thu, Jul 13, 2017 at 10:41 PM, Chandler Carruth <chandlerc at google.com>
wrote:

> On Thu, Jul 13, 2017 at 7:29 PM Xinliang David Li via llvm-commits <
> llvm-commits at lists.llvm.org> wrote:
>
>> On Thu, Jul 13, 2017 at 4:17 PM, David Blaikie <dblaikie at gmail.com>
>> wrote:
>>
>>>
>>>
>>> On Thu, Jul 13, 2017 at 2:03 PM David Li via Phabricator <
>>> reviews at reviews.llvm.org> wrote:
>>>
>>>> davidxl added a comment.
>>>>
>>>> Blindly downgrade the warning into remarks can be bad -- profile
>>>> mismatch problems can go undetected. People may spend more time diagnosing
>>>> performance regressions due to missed warnings. Worse, we may regress in
>>>> compiler without noticing.
>>>>
>>>
>>> Any other ideas?
>>>
>>> The diagnostic handling callbacks through to Clang aren't granular
>>> enough to demote only this warning to -Wno-error
>>>
>>
>> Can this be fixed?
>>
>>
>>
>>> (also as a warning it undermines the confidence as warnings as
>>> actionable things that Clang has helped provide users).
>>>
>>
>>
>> Warnings like this are usually actionable by users or compiler
>> developers.
>>
>
> Warnings are really to indicate a bug in user code.
>
> For example, we use some things like this to warn users when inline
> assembly is likely to have a bug, because that is user code that can be
> fixed.
>
> A profile mismatch seems much less like "user code has a bug" and so
> doesn't seem to fit any well in the "warning" category.
>

It is not user code bug, but still be a user bug in how PGO is used.  The
warning happens also to be a useful for catch potential compiler bug.

David

>
> On the other hand, remarks were specifically designed to be non-bug
> important information from optimization passes that users may need to act
> on.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170713/adb49878/attachment.html>