On Fri, Jun 08, 2018 at 07:23:06PM +0200, Thomas Huth wrote: > On 08.06.2018 18:24, Michael S. Tsirkin wrote: > > On Fri, Jun 08, 2018 at 05:16:30PM +0100, Peter Maydell wrote: > >> On 8 June 2018 at 17:03, Michael S. Tsirkin <m...@redhat.com> wrote: > >>> Pull requests are somewhat different, they are usually tested for lack > >>> of warnings. This change didn't arrive as a result of a pull request > >>> maybe that's why it slipped through the cracks. Peter? > >>> > >>> Maybe we need a "pedantic" flag to fail on any warnings, or just catch > >>> output to stderr. > >> > >> If there's a situation that shouldn't exist in the tree (ie > >> a bug), then make check should catch it, and result in a > >> failure, not just printing random stuff to stderr. Otherwise > >> I'm not going to notice it, whether I'm applying a pull request > >> or an individual patch. > >> > >> thanks > >> -- PMM > > > > It's ok if it happens, but it just makes debugging and reviewing > > ACPI patches a little bit harder until it's fixed. > > It's maybe ok for *you*, but this certainly confuses everybody else. If > I want to check my patches and suddenly some strange warnings are > popping up, I first assume that there is something wrong in my patches > (since I assume that the git repository is clean by default). So I've > got to waste my time debugging issues that are not my own. Thanks for > that :-/ > > Thomas
Right so normally these do not pop out at all as I fix expected with a patch on top. -- MST