[patch] Check for isDiagnosticEnabled even if we have a dig handler

Diego Novillo dnovillo at google.com
Wed Mar 4 08:02:06 PST 2015


On Tue, Mar 3, 2015 at 6:28 PM, Rafael EspĂ­ndola <rafael.espindola at gmail.com
> wrote:

> > RespectDiagnosticFilters does look like more flexibility than
> > necessary, but that is independent. I will try writing a patch to see
> > what breaks.
>
> And yes, it breaks clang. The issue is that clang expects to be passed
> all the remarks and to do its own filtering for -Rpass.
>
> Diego, do you remember why it was implemented like that? The filtering
> looks duplicated with the one in LLVM. Couldn't clang just pass the
> pattern down to LLVM?
>

I don't recall exactly, sorry. I do remember being annoyed by the
duplication, but at the time I did not find a way around that. I'm fine
with whatever approach works.


Diego.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150304/2c42fb47/attachment.html>


More information about the llvm-commits mailing list