Re: [tip:sched/core] BUILD SUCCESS 3387ce4d8a5f2956fab827edf499fe6780e83faa

[Chen, Rong A]

On 3/25/2022 1:58 AM, Peter Zijlstra wrote:
> On Tue, Mar 22, 2022 at 10:39:41AM +0800, Chen, Rong A wrote:
> > On 3/22/2022 7:17 AM, Nathan Chancellor wrote:
> > > Hello,
> > >
> > > On Tue, Mar 22, 2022 at 06:36:16AM +0800, kernel test robot wrote:
> > > > tree/branch: https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched/core
> > > > branch HEAD: 3387ce4d8a5f2956fab827edf499fe6780e83faa  headers/prep: Fix header to build standalone: <linux/psi.h>
> > > >
> > > > elapsed time: 730m
> > > >
> > > > configs tested: 114
> > > > configs skipped: 3
> > > >
> > > > The following configs have been built successfully.
> > > > More configs may be tested in the coming days.
> > > >
> > > > gcc tested configs:
> > > ...
> > > > arm                              allyesconfig
> > > > arm                              allmodconfig
>
> > > Are you sure these configurations built sucessfully?
>
> ...
>
> > Thanks for your reminding, the problem commit is 4ff8f2ca6ccd
> > ("sched/headers: Reorganize, clean up and optimize kernel/sched/sched.h
> > dependencies"),
> > the bot has sent two build reports related to it:
> >
> >    https://lore.kernel.org/all/202203152116.qphmikIZ-lkp@xxxxxxxxx/
> >    https://lore.kernel.org/all/202203152144.uFQqgVUf-lkp@xxxxxxxxx/
> >
> > and more reports were stopped to avoid too many noise, like below one:
>
> ...
>
> > We'll adjust the strategy avoid blocking some important reports.
>
> Oh *PLEASE* never report a branch as building if there's errors.
> It could be I (force) push a branch multiple times before I get any
> 0day reports back (0day has gotten *soooo* slow) so I only ever look at
> the latest report -- possibly days later.

Hi Peter,

Sorry for the bad experience, I'm not sure the meaning of "a branch as
building", is there a way to know the branch is not ready, or we can
postpone the test for a branch by a day?


> If you then falsly report the branch as being good, because you're
> suppressing errors, things *will* go bad.

yes, current the summary report only shows the errors that bisected
successfully in a cycle, but it may causes some errors missing
and old errors still in the report, we plan to redesign the report
to present all (may a lot) recent errors.

Best Regards,
Rong Chen