Re: [RFC] Avoid mutex starvation when optimistic spinning is disabled

From	Imre Deak <imre.deak@intel.com>
Newsgroups	linux.kernel
Subject	Re: [RFC] Avoid mutex starvation when optimistic spinning is disabled
Date	2016-07-22 22:00 +0200
Message-ID	<rXOM2-1Io-19@gated-at.bofh.it> (permalink)
References	(6 earlier) <rXtex-4cZ-11@gated-at.bofh.it> <rXN3A-Qn-19@gated-at.bofh.it> <rXN3A-Qn-17@gated-at.bofh.it> <rXNwB-117-7@gated-at.bofh.it> <rXOiZ-1yn-3@gated-at.bofh.it>
Organization	Intel

Show all headers | View raw

On Fri, 2016-07-22 at 12:26 -0700, Davidlohr Bueso wrote:
> On Fri, 22 Jul 2016, Imre Deak wrote:
> 
> > On Fri, 2016-07-22 at 11:03 -0700, Davidlohr Bueso wrote:
> > > On Fri, 22 Jul 2016, Waiman Long wrote:
> > > 
> > > > I think making mutex_trylock() fail maybe a bit too far. Do we
> > > > really
> > > > have any real workload that cause starvation problem  because
> > > > of
> > > > that.
> > > > Code that does mutex_trylock() in a loop can certainly cause
> > > > lock
> > > > starvation, but it is not how mutex_trylock() is supposed to be
> > > > used.
> > > > We can't build in safeguard for all the possible abuses of the
> > > > mutex
> > > > APIs.
> > > 
> > > True, and that's actually why I think that 'fixing' the
> > > !SPIN_ON_OWNER case
> > > is a bit too far in the first place: most of the archs that will
> > > care
> > > about
> > > this already have ARCH_SUPPORTS_ATOMIC_RMW. The extra code for
> > > dealing with
> > > this is not worth it imo.
> > 
> > SPIN_ON_OWNER is also disabled in case of DEBUG_MUTEXES, which is
> > the
> > config where I wanted to avoid starvation in the first place.
> 
> Well yes, but know of course that that option is even less common
> than
> archs with non atomic Rmw.

The point is that it's broken atm.

--Imre

Thread

[RFC] locking/mutex: Fix starvation of sleeping waiters Imre Deak <imre.deak@intel.com> - 2016-07-18 18:20 +0200
  Re: [RFC] locking/mutex: Fix starvation of sleeping waiters Peter Zijlstra <peterz@infradead.org> - 2016-07-18 19:20 +0200
    Re: [RFC] locking/mutex: Fix starvation of sleeping waiters Jason Low <jason.low2@hpe.com> - 2016-07-18 19:50 +0200
      Re: [RFC] locking/mutex: Fix starvation of sleeping waiters Imre Deak <imre.deak@intel.com> - 2016-07-19 19:00 +0200
        Re: [RFC] locking/mutex: Fix starvation of sleeping waiters Jason Low <jason.low2@hpe.com> - 2016-07-20 01:10 +0200
        [RFC] Avoid mutex starvation when optimistic spinning is disabled Jason Low <jason.low2@hpe.com> - 2016-07-20 01:20 +0200
          Re: [RFC] Avoid mutex starvation when optimistic spinning is  disabled Jason Low <jason.low2@hpe.com> - 2016-07-20 06:50 +0200
            Re: [RFC] Avoid mutex starvation when optimistic spinning is  disabled Imre Deak <imre.deak@intel.com> - 2016-07-20 15:40 +0200
              Re: [RFC] Avoid mutex starvation when optimistic spinning is  disabled Jason Low <jason.low2@hpe.com> - 2016-07-21 23:00 +0200
                Re: [RFC] Avoid mutex starvation when optimistic spinning is disabled Davidlohr Bueso <dave@stgolabs.net> - 2016-07-22 20:10 +0200
                Re: [RFC] Avoid mutex starvation when optimistic spinning is  disabled Imre Deak <imre.deak@intel.com> - 2016-07-22 20:40 +0200
                Re: [RFC] Avoid mutex starvation when optimistic spinning is disabled Davidlohr Bueso <dave@stgolabs.net> - 2016-07-22 21:30 +0200
                Re: [RFC] Avoid mutex starvation when optimistic spinning is  disabled Imre Deak <imre.deak@intel.com> - 2016-07-22 22:00 +0200
            Re: [RFC] Avoid mutex starvation when optimistic spinning is  disabled Jason Low <jason.low2@hpe.com> - 2016-07-22 00:40 +0200
              Re: [RFC] Avoid mutex starvation when optimistic spinning is  disabled Imre Deak <imre.deak@intel.com> - 2016-07-22 11:40 +0200
                Re: [RFC] Avoid mutex starvation when optimistic spinning is  disabled Jason Low <jason.low2@hpe.com> - 2016-07-22 20:50 +0200

csiph-web