From: Nick Piggin <nickpiggin@yahoo.com.au>
To: Arjan van de Ven <arjan@infradead.org>
Cc: Ingo Molnar <mingo@elte.hu>, Andi Kleen <ak@suse.de>,
Arjan van de Ven <arjan@intel.linux.com>,
linux-kernel@vger.kernel.org, akpm@osdl.org
Subject: Re: [Patch 3/3] prepopulate/cache cleared pages
Date: Fri, 24 Feb 2006 20:26:00 +1100 [thread overview]
Message-ID: <43FED128.1030500@yahoo.com.au> (raw)
In-Reply-To: <1140772543.2874.20.camel@laptopd505.fenrus.org>
Arjan van de Ven wrote:
>>Arjan, just to get an idea of your workload: obviously it is a mix of
>>read and write on the mmap_sem (read only will not really benefit from
>>reducing lock width because cacheline transfers will still be there).
>
>
> yeah it's threads that each allocate, use and then free memory with
> mmap()
>
>
>>Is it coming from brk() from the allocator? Someone told me a while ago
>>that glibc doesn't have a decent amount of hysteresis in its allocator
>>and tends to enter the kernel quite a lot... that might be something
>>to look into.
>
>
> we already are working on that angle; I just posted the kernel stuff as
> a side effect basically
>
OK.
[aside]
Actually I have a scalability improvement for rwsems, that moves the
actual task wakeups out from underneath the rwsem spinlock in the up()
paths. This was useful exactly on a mixed read+write workload on mmap_sem.
The difference was quite large for the "generic rwsem" algorithm because
it uses the spinlock in fastpaths a lot more than the xadd algorithm. I
think x86-64 uses the former, which is what I presume you're testing with?
Obviously this is a slightly different issue from the one you're trying to
address here, but I'll dig out patch when I get some time and you can see
if it helps.
--
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com
next prev parent reply other threads:[~2006-02-24 9:26 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-02-23 9:17 [Patch 0/3] threaded mmap tweaks Arjan van de Ven
2006-02-23 9:29 ` [Patch 3/3] prepopulate/cache cleared pages Arjan van de Ven
2006-02-23 9:41 ` Andi Kleen
2006-02-23 12:41 ` Ingo Molnar
2006-02-23 13:06 ` Andi Kleen
2006-02-23 13:15 ` Nick Piggin
2006-02-23 13:29 ` Ingo Molnar
2006-02-24 6:36 ` Nick Piggin
2006-02-24 6:49 ` Ingo Molnar
2006-02-24 7:01 ` Nick Piggin
2006-02-24 12:33 ` Andi Kleen
2006-02-24 12:55 ` Hugh Dickins
2006-02-24 9:15 ` Arjan van de Ven
2006-02-24 9:26 ` Nick Piggin [this message]
2006-02-24 12:27 ` Andi Kleen
2006-02-24 15:31 ` Andrea Arcangeli
2006-02-25 16:48 ` Nick Piggin
2006-02-25 17:22 ` Nick Piggin
2006-02-28 22:30 ` Pavel Machek
2006-02-23 18:25 ` Paul Jackson
2006-02-23 9:30 ` [Patch 2/3] fast VMA recycling Arjan van de Ven
2006-02-23 9:42 ` Andi Kleen
2006-02-23 9:48 ` Arjan van de Ven
2006-02-23 10:05 ` Andi Kleen
2006-02-23 10:15 ` Arjan van de Ven
2006-02-23 11:00 ` Andi Kleen
2006-02-23 11:22 ` Arjan van de Ven
2006-02-23 11:57 ` Andi Kleen
2006-02-24 18:52 ` Christoph Hellwig
2006-02-24 19:05 ` Andi Kleen
2006-02-24 19:09 ` Christoph Hellwig
2006-02-23 16:37 ` Benjamin LaHaise
2006-02-23 20:02 [Patch 3/3] prepopulate/cache cleared pages Chuck Ebbert
2006-02-23 21:10 Chuck Ebbert
2006-02-23 21:18 ` Arjan van de Ven
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=43FED128.1030500@yahoo.com.au \
--to=nickpiggin@yahoo.com.au \
--cc=ak@suse.de \
--cc=akpm@osdl.org \
--cc=arjan@infradead.org \
--cc=arjan@intel.linux.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).