Re: [patch 1/4] x86: FIFO ticket spinlocks

[Nick Piggin <npiggin@suse.de>]

On Fri, Nov 02, 2007 at 10:05:37AM -0400, Rik van Riel wrote:
> On Fri, 2 Nov 2007 07:42:20 +0100
> Nick Piggin <npiggin@suse.de> wrote:
> 
> > On Thu, Nov 01, 2007 at 06:19:41PM -0700, Linus Torvalds wrote:
> > > 
> > > 
> > > On Thu, 1 Nov 2007, Rik van Riel wrote:
> > > > 
> > > > Larry Woodman managed to wedge the VM into a state where, on his
> > > > 4x dual core system, only 2 cores (on the same CPU) could get the
> > > > zone->lru_lock overnight.  The other 6 cores on the system were
> > > > just spinning, without being able to get the lock.
> > 
> > That's quite incredible, considering that the CPUs actually _taking_
> > the locks also drop the locks and do quite a bit of work before taking
> > them again (ie. they take them to pull pages off the LRU, but then
> > do a reasonable amount of work to remove each one from pagecache
> > before refilling from the LRU).
> > 
> > Possibly actually that is a *more* difficult case for the HW to
> > handle: once the CPU actually goes away and operates on other
> > cachelines, it may get a little more difficult to detect that it is
> > causing starvation issues.
> 
> In case of the zone->lru_lock, grabbing the spinlock does
> not mean that the process is letting go of the cacheline.
> 
> On the contrary, once the spinlock has been grabbed, the
> real cacheline prodding begins.

I didn't say that, though. Obviously the hardware can't do anything
about starvating until a lock is released.