Re: [PATCH 7.0 000/307] 7.0.7-rc1 review

[Greg Kroah-Hartman]

On Wed, May 13, 2026 at 08:05:45AM +0200, Andrea Righi wrote:
> Hello,
> 
> On Wed, May 13, 2026 at 01:31:00AM +0200, Peter Schneider wrote:
> > Hi Greg,
> > 
> > Am 12.05.2026 um 19:36 schrieb Greg Kroah-Hartman:
> > > This is the start of the stable review cycle for the 7.0.7 release.
> > > There are 307 patches in this series, all will be posted as a response
> > > to this one.  If anyone has any issues with these being applied, please
> > > let me know.
> > 
> > Trying to build 7.0.7-rc1, I get this build error.
> > 
> > In file included from kernel/sched/build_policy.c:62:
> > kernel/sched/ext.c: In function ‘bypass_lb_cpu’:
> > kernel/sched/ext.c:4019:35: error: ‘donor_rq’ undeclared (first use in this function); did you mean ‘donee_rq’?
> >  4019 |                 if (task_rq(p) != donor_rq)
> >       |                                   ^~~~~~~~
> >       |                                   donee_rq
> > kernel/sched/ext.c:4019:35: note: each undeclared identifier is reported only once for each function it appears in
> > make[4]: *** [scripts/Makefile.build:289: kernel/sched/build_policy.o] Fehler 1
> > make[3]: *** [scripts/Makefile.build:548: kernel/sched] Fehler 2
> > make[2]: *** [scripts/Makefile.build:548: kernel] Fehler 2
> > make[1]: *** [/usr/src/linux-stable-rc/Makefile:2108: .] Fehler 2
> > make: *** [Makefile:248: __sub-make] Fehler 2
> > root@linus:/usr/src/linux-stable-rc#
> > 
> > The offending line seems to be part of eb5b997dadc517 (sched_ext: Skip tasks with stale task_rq in bypass_lb_cpu())
> > Adding Tejun and Andrea to CC.
> 
> The upstream commit (da2d81b4118a) was written on top of ff06f727a941
> ("sched_ext: Move bypass_dsq into scx_sched_pcpu"), which renamed @rq to
> @donor_rq (among other things). That refactor is not in 7.0.y and it's not
> stable material.
> 
> I think The minimal fix is to rename donor_rq -> rq in the backported hunk, in
> 7.0.y the function still takes struct rq *rq directly, and that is the donor rq.
> 
> Greg, can you fold the following into the queued patch? Do you prefer a
> separate/updated patch?
> 
> Thanks,
> -Andrea
> 
>  kernel/sched/ext.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
> index 3cb8025b433e0..2bb7c7a902679 100644
> --- a/kernel/sched/ext.c
> +++ b/kernel/sched/ext.c
> @@ -4013,10 +4013,10 @@ static u32 bypass_lb_cpu(struct scx_sched *sch, struct rq *rq,
>  		/*
>  		 * If an earlier pass placed @p on @donor_dsq from a different
>  		 * CPU and the donee hasn't consumed it yet, @p is still on the
> -		 * previous CPU and task_rq(@p) != @donor_rq. @p can't be moved
> +		 * previous CPU and task_rq(@p) != @rq. @p can't be moved
>  		 * without its rq locked. Skip.
>  		 */
> -		if (task_rq(p) != donor_rq)
> +		if (task_rq(p) != rq)
>  			continue;
>  
>  		donee = cpumask_any_and_distribute(donee_mask, p->cpus_ptr);
> 

A whole new patch please.

thanks,

greg k-h