[lkml]   [2008]   [May]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [patch 21/21] slab defrag: Obsolete SLAB
On Mon, 12 May 2008, Pekka Enberg wrote:

> Christoph fixed a tbench regression that was in the same ballpark as
> the TPC regression reported by Matthew which is why we've asked the
> Intel folks to re-test. But yeah, we're working on it.

I suspect that the TPC regression was due to the page allocator order 0
inefficiencies like the tbench regression but we have no data yet to
establish that.

Fundamentally there is no way to avoid complex queueing on free() unless
one directly frees the object. This is serialized in SLUB by taking a page
lock. If we can establish that the object is from the current cpu slab
then no lock is taken because the slab is reserved for the current
processor. So the bad case is a free of a object with a long life span or
an object freed on a remote processor.

Howver, the "slow" case in SLUB is still much less complex
than comparable processing in SLAB. It is quite fast.

SLAB freeing can avoid taking a lock if

1. We can establish that the object is node local (trivial if !NUMA
otherwise we need to get the node information from the page struct and
compare to the current node).

2. There is space in the per cpu queue

If the object is *not* node local then we have to take an alien lock for
the remote node in order to put the object in an alien queue. That is much
less efficient than the SLUB case. SLAB then needs to run the cache reaper
to expire these object into the remote nodes queues (later the cache
reaper may then actually free these objects). This management overhead
does not exist in SLUB. The cache reaper causes processors to not be
available for short time frames (the reaper scans through all slab
caches!) which in turn cause regression in applications that need to
respond in a short time frame (HPC appls, network applications that are
timing critical).

Note that the lock granularity in SLUB is finer than the locks in SLAB.
SLUB can concurrently free multiple objects to the same remote node etc
etc. If the objects belong to different slabs then there is no dirtying of
any shared cachelines.

The main issue for SLAB vs. SLUB on free is likely the !NUMA case in which
SLAB can avoid the overhead of the node check (which does not exist in
SLUB) and in which case we can always immediately batch the object (if
there is space). The additional overhead in SLUB is mainly one
atomic instruction over the SLAB fastpath.

So I think that the free need to stay as is. The disadvantages in terms
of the complexity of handling the objects and expiring them and the issue
of having to take per node locks in SLAB makes it hard to justify adding a
queue for free in SLUB. Maybe someone has an inspiration on how to do this
effective that is better than my attempts which always ultimately ended
implementing code that thad the same issues that we have in SLAB.

 \ /
  Last update: 2008-05-14 19:33    [W:0.114 / U:1.188 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site