Re: [PATCH v2] sched,livepatch: Untangle cond_resched() and live-patching

[Petr Mladek <pmladek@suse.com>]

On Tue 2025-05-13 16:05:51, Miroslav Benes wrote:
> On Tue, 13 May 2025, Peter Zijlstra wrote:
> 
> > On Tue, May 13, 2025 at 03:34:50PM +0200, Miroslav Benes wrote:
> > > Hi,
> > > 
> > > thanks for the updated version.
> > > 
> > > On Fri, 9 May 2025, Sebastian Andrzej Siewior wrote:
> > > 
> > > > From: Peter Zijlstra <peterz@infradead.org>
> > > > 
> > > > With the goal of deprecating / removing VOLUNTARY preempt, live-patch
> > > > needs to stop relying on cond_resched() to make forward progress.
> > > > 
> > > > Instead, rely on schedule() with TASK_FREEZABLE set. Just like
> > > > live-patching, the freezer needs to be able to stop tasks in a safe /
> > > > known state.
> > > > 
> > > > Compile tested only.
> > > 
> > > livepatch selftests pass and I also ran some more.
> > >  
> > > > [bigeasy: use likely() in __klp_sched_try_switch() and update comments]
> > > > 
> > > > Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> > > > Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
> > > 
> > > Acked-by: Miroslav Benes <mbenes@suse.cz>
> > > 
> > > A nit below if there is an another version, otherwise Petr might fix it 
> > > when merging.
> > 
> > Petr or Peter?
> > 
> > That is, who are we expecting to merge this :-)
> 
> Petr Mladek if it goes through the live patching tree, you if tip. Feel 
> free to pick it up :).

IMHO, it might be easier when it goes via tip. Peter, feel free to
take it.

The patch does not create any conflict with the klp tree.
But I guess that there might be some dependent patches in tip...

That said, I could take it via the livepatch tree if Peter preferred
it from some reasons.

Anyway, the patch looks good and passes the tests. Feel free to use:

Reviewed-by: Petr Mladek <pmladek@suse.com>
Tested-by: Petr Mladek <pmladek@suse.com>

Best Regards,
Petr