On Wed, 31 Jul 2019, Jan Kara wrote:
> On Tue 30-07-19 13:24:54, Thomas Gleixner wrote:
> > Bit spinlocks are problematic if PREEMPT_RT is enabled, because they
> > disable preemption, which is undesired for latency reasons and breaks when
> > regular spinlocks are taken within the bit_spinlock locked region because
> > regular spinlocks are converted to 'sleeping spinlocks' on RT. So RT
> > replaces the bit spinlocks with regular spinlocks to avoid this problem.
> >
> > To avoid ifdeffery at the source level, wrap all BH_Uptodate_Lock bitlock
> > operations with inline functions, so the spinlock substitution can be done
> > at one place.
> >
> > Using regular spinlocks can also be enabled for lock debugging purposes so
> > the lock operations become visible to lockdep.
> >
> > Signed-off-by: Thomas Gleixner <[email protected]>
> > Cc: "Theodore Ts'o" <[email protected]>
> > Cc: Matthew Wilcox <[email protected]>
> > Cc: Alexander Viro <[email protected]>
> > Cc: [email protected]
>
> Looks good to me. You can add:
>
> Reviewed-by: Jan Kara <[email protected]>
>
> BTW, it should be possible to get rid of BH_Uptodate_Lock altogether using
> bio chaining (which was non-existent when this bh code was written) to make
> sure IO completion function gets called only once all bios used to fill in
> / write out the page are done. It would be also more efficient. But I guess
> that's an interesting cleanup project for some other time...
While 'possible cleanup' is something which triggers a certain nerve, that
particular project certainly goes beyond my basic understanding of that
whole fs/block conglomerate. I rather leave that to people who actually
have a clue. :)
Thanks,
tglx
