Commit r345200 (new ARC reclamation threads) looks suspicious to me - second potential problem
Slawa Olhovchenkov
slw at zxy.spb.ru
Mon May 20 16:54:49 UTC 2019
On Mon, May 20, 2019 at 12:42:10PM -0400, Mark Johnston wrote:
> On Mon, May 20, 2019 at 07:05:07PM +0300, Lev Serebryakov wrote:
> >
> > I'm looking at last commit to
> > 'sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c' (r345200) and
> > have another question.
> >
> > Here are such code:
> >
> > 4960 /*
> > 4961 * Kick off asynchronous kmem_reap()'s of all our caches.
> > 4962 */
> > 4963 arc_kmem_reap_soon();
> > 4964
> > 4965 /*
> > 4966 * Wait at least arc_kmem_cache_reap_retry_ms between
> > 4967 * arc_kmem_reap_soon() calls. Without this check it is
> > possible to
> > 4968 * end up in a situation where we spend lots of time reaping
> > 4969 * caches, while we're near arc_c_min. Waiting here also
> > gives the
> > 4970 * subsequent free memory check a chance of finding that the
> > 4971 * asynchronous reap has already freed enough memory, and
> > we don't
> > 4972 * need to call arc_reduce_target_size().
> > 4973 */
> > 4974 delay((hz * arc_kmem_cache_reap_retry_ms + 999) / 1000);
> > 4975
> >
> > But looks like `arc_kmem_reap_soon()` is synchronous on FreeBSD! So,
> > this `delay()` looks very wrong. Am I right?
> >
> > Looks like it should be `#ifdef illumos`.
>
> See also r338142, which I believe was reverted by the update.
Some throttling realy need for reducing ARC size (and for reap calls
too, and this is not equal) (memory freeing to
kernel-wide is too expensive and can realy take seconds and more), but
this implementation for FreeBSD is too bad.
More information about the freebsd-hackers
mailing list