From: Dave Hansen <haveblue@us.ibm.com>
To: maneesh@in.ibm.com
Cc: William Lee Irwin III <wli@holomorphy.com>,
Andrew Morton <akpm@digeo.com>,
linux-kernel@vger.kernel.org, viro@math.psu.edu,
Hanna Linder <hannal@us.ibm.com>
Subject: Re: 2.5.36-mm1 dbench 512 profiles
Date: Fri, 20 Sep 2002 07:34:32 -0700 [thread overview]
Message-ID: <3D8B31F8.40900@us.ibm.com> (raw)
In-Reply-To: 200209200747.g8K7la9B174532@northrelay01.pok.ibm.com
Maneesh Soni wrote:
> On Fri, 20 Sep 2002 05:48:38 +0530, William Lee Irwin III wrote:
>
>>Hanna Linder wrote:
>>
>>>>Looks like fastwalk might not behave so well on this 32 cpu numa
>>>>system...
>>
>>On Thu, Sep 19, 2002 at 04:38:14PM -0700, Andrew Morton wrote:
>>
>>>I've rather lost the plot. Have any of the dcache speedup patches been
>>>merged into 2.5?
>>
>>As far as the dcache goes, I'll stick to observing and reporting. I'll
>>rerun with dcache patches applied, though.
>>
> For a 32-way system fastwalk will perform badly from dcache_lock point of
> view, basically due to increased lock hold time. dcache_rcu-12 should reduce
> dcache_lock contention and hold time.
Isn't increased hold time _good_ on NUMA-Q? I thought that the really
costy operation was bouncing the lock around the interconnect, not
holding it. Has fastwalk ever been tested on NUMA-Q?
Remember when John Stultz tried MCS (fair) locks on NUMA-Q? They
sucked because low hold times, which result from fairness, aren't
efficient. It is actually faster to somewhat starve remote CPUs.
In any case, we all know often acquired global locks are a bad idea on
a 32-way, and should be avoided like the plague. I just wish we had a
dcache solution that didn't even need locks as much... :)
--
Dave Hansen
haveblue@us.ibm.com
next prev parent reply other threads:[~2002-09-20 14:29 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-09-19 22:30 2.5.36-mm1 dbench 512 profiles William Lee Irwin III
2002-09-19 23:18 ` Hanna Linder
2002-09-19 23:38 ` Andrew Morton
2002-09-19 23:45 ` Hanna Linder
2002-09-20 0:08 ` William Lee Irwin III
2002-09-20 4:02 ` William Lee Irwin III
2002-09-20 7:59 ` Maneesh Soni
2002-09-20 8:06 ` William Lee Irwin III
2002-09-20 12:03 ` William Lee Irwin III
2002-09-20 18:51 ` Hanna Linder
2002-09-20 20:32 ` Hanna Linder
2002-09-20 20:54 ` Dipankar Sarma
2002-09-20 20:39 ` William Lee Irwin III
2002-09-20 21:30 ` Martin J. Bligh
2002-09-20 23:11 ` William Lee Irwin III
2002-09-20 23:22 ` Martin J. Bligh
2002-09-21 7:52 ` William Lee Irwin III
2002-09-20 14:34 ` Dave Hansen [this message]
2002-09-20 16:07 ` Martin J. Bligh
2002-09-20 17:48 ` Dipankar Sarma
2002-09-20 17:40 ` Dipankar Sarma
2002-09-20 20:28 ` Dipankar Sarma
2002-09-20 5:14 ` William Lee Irwin III
2002-09-20 6:59 ` William Lee Irwin III
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=3D8B31F8.40900@us.ibm.com \
--to=haveblue@us.ibm.com \
--cc=akpm@digeo.com \
--cc=hannal@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=maneesh@in.ibm.com \
--cc=viro@math.psu.edu \
--cc=wli@holomorphy.com \
/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).