[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170413124014.63177422@gandalf.local.home>
Date: Thu, 13 Apr 2017 12:40:14 -0400
From: Steven Rostedt <rostedt@...dmis.org>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Alex Shi <alex.shi@...aro.org>, mingo@...hat.com, corbet@....net,
"open list:LOCKING PRIMITIVES" <linux-kernel@...r.kernel.org>,
Sebastian Siewior <bigeasy@...utronix.de>,
Thomas Gleixner <tglx@...utronix.de>
Subject: Re: [PATCH 2/3] rtmutex: deboost priority conditionally when
rt-mutex unlock
On Thu, 13 Apr 2017 18:21:13 +0200
Peter Zijlstra <peterz@...radead.org> wrote:
> On Thu, Apr 13, 2017 at 12:09:25PM -0400, Steven Rostedt wrote:
> > On Thu, 13 Apr 2017 16:39:52 +0200
> > Peter Zijlstra <peterz@...radead.org> wrote:
> >
> > > On Thu, Apr 13, 2017 at 10:02:53PM +0800, Alex Shi wrote:
> > > > /*
> > > > + * 'current' release this lock, so 'current' should be a higher prio
> > > > + * task than the next top waiter, unless the current prio was gotten
> > > > + * from this top waiter, iff so, we need to deboost 'current' after
> > > > + * the lock release.
> > > > + */
> > > > + if (current->prio == waiter->prio)
> > > > + deboost = true;
> > >
> > > This is wrong.
> >
> > The comment is, especially that "iff". What if current and waiter
> > happen to have the same priority? Then it too doesn't need to be
> > deboosted.
>
> The wrongness is in comparing prio and thinking it means anything.
Because of deadline scheduling?
-- Steve
Powered by blists - more mailing lists