[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1232616517.11429.129.camel@ymzhang>
Date: Thu, 22 Jan 2009 17:28:37 +0800
From: "Zhang, Yanmin" <yanmin_zhang@...ux.intel.com>
To: Pekka Enberg <penberg@...helsinki.fi>
Cc: Christoph Lameter <cl@...ux-foundation.org>,
Andi Kleen <andi@...stfloor.org>,
Matthew Wilcox <matthew@....cx>,
Nick Piggin <nickpiggin@...oo.com.au>,
Andrew Morton <akpm@...ux-foundation.org>,
netdev@...r.kernel.org, sfr@...b.auug.org.au,
matthew.r.wilcox@...el.com, chinang.ma@...el.com,
linux-kernel@...r.kernel.org, sharad.c.tripathi@...el.com,
arjan@...ux.intel.com, suresh.b.siddha@...el.com,
harita.chilukuri@...el.com, douglas.w.styner@...el.com,
peter.xihong.wang@...el.com, hubert.nueckel@...el.com,
chris.mason@...cle.com, srostedt@...hat.com,
linux-scsi@...r.kernel.org, andrew.vasquez@...gic.com,
anirban.chakraborty@...gic.com
Subject: Re: Mainline kernel OLTP performance update
On Thu, 2009-01-22 at 11:15 +0200, Pekka Enberg wrote:
> On Thu, 2009-01-22 at 16:36 +0800, Zhang, Yanmin wrote:
> > On Wed, 2009-01-21 at 18:58 -0500, Christoph Lameter wrote:
> > > On Tue, 20 Jan 2009, Zhang, Yanmin wrote:
> > >
> > > > kmem_cache skbuff_head_cache's object size is just 256, so it shares the kmem_cache
> > > > with :0000256. Their order is 1 which means every slab consists of 2 physical pages.
> > >
> > > That order can be changed. Try specifying slub_max_order=0 on the kernel
> > > command line to force an order 0 alloc.
> > I tried slub_max_order=0 and there is no improvement on this UDP-U-4k issue.
> > Both get_page_from_freelist and __free_pages_ok's cpu time are still very high.
> >
> > I checked my instrumentation in kernel and found it's caused by large object allocation/free
> > whose size is more than PAGE_SIZE. Here its order is 1.
> >
> > The right free callchain is __kfree_skb => skb_release_all => skb_release_data.
> >
> > So this case isn't the issue that batch of allocation/free might erase partial page
> > functionality.
>
> So is this the kfree(skb->head) in skb_release_data() or the put_page()
> calls in the same function in a loop?
It's kfree(skb->head).
>
> If it's the former, with big enough size passed to __alloc_skb(), the
> networking code might be taking a hit from the SLUB page allocator
> pass-through.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists