From: "Sean Hefty" <sean.hefty@intel.com>
To: "'Roland Dreier'" <rdreier@cisco.com>
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
general@lists.openfabrics.org
Subject: RE: [ofa-general] InfiniBand/RDMA merge plans for 2.6.24
Date: Tue, 2 Oct 2007 11:50:04 -0700 [thread overview]
Message-ID: <000601c80525$0b661f30$ff0da8c0@amr.corp.intel.com> (raw)
In-Reply-To: <adaprzxgyoi.fsf@cisco.com>
>Umm... this is a difficult situation for me to merge the changes then.
>We're changing the CM retry behavior blind here. How do we know that
>the MRA changes don't make the scalability issue worse?
What's currently upstream doesn't work for Intel MPI on our larger clusters.
The connection requests time out on the active side before the passive side can
respond.
The OFED release works because it provides a kernel patch to make the timeout a
module parameter. I'm trying to avoid adding a module parameter, and the MRA is
designed for this situation.
I tested this by simulating a slow passive side responder, and it worked as
expected for those tests. Using an MRA does add another MAD to the CM exchange,
which is why it is sent only after seeing a duplicate request. Alternatively,
we can take the OFED module parameter patch.
- Sean
next prev parent reply other threads:[~2007-10-02 18:50 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-09-13 17:57 InfiniBand/RDMA merge plans for 2.6.24 Roland Dreier
2007-09-13 18:04 ` [ofa-general] " Steve Wise
2007-09-13 18:56 ` Jeff Garzik
2007-09-13 18:59 ` Steve Wise
2007-09-13 19:55 ` Jeff Garzik
2007-09-14 12:55 ` Evgeniy Polyakov
2007-09-13 21:12 ` Roland Dreier
2007-09-13 21:11 ` Roland Dreier
2007-09-13 22:59 ` Michael Chan
2007-09-14 16:18 ` Roland Dreier
2007-09-14 21:09 ` Michael Chan
2007-09-15 14:03 ` Steve Wise
2007-09-13 18:20 ` Sean Hefty
2007-09-13 21:02 ` Roland Dreier
2007-09-14 17:45 ` Sean Hefty
2007-10-02 18:26 ` Roland Dreier
2007-10-02 18:50 ` Sean Hefty [this message]
2007-10-05 23:10 ` Roland Dreier
2007-09-17 13:00 ` Hal Rosenstock
2007-09-18 9:48 ` Tziporet Koren
2007-09-13 18:22 ` Shirley Ma
2007-09-13 21:00 ` Roland Dreier
2007-09-13 22:16 ` [ofa-general] " Shirley Ma
2007-09-14 16:09 ` Roland Dreier
2007-09-14 18:36 ` Shirley Ma
2007-09-17 21:47 ` [ofa-general] " Roland Dreier
2007-09-17 22:24 ` Shirley Ma
2007-10-03 18:43 ` Shirley Ma
2007-09-16 8:50 ` [ofa-general] " Or Gerlitz
2007-09-17 22:11 ` Roland Dreier
2007-09-18 7:09 ` Jack Morgenstein
2007-09-19 6:19 ` [ofa-general] " Michael S. Tsirkin
2007-09-18 16:34 ` Michael S. Tsirkin
2007-09-18 17:18 ` Roland Dreier
2007-09-18 17:22 ` Michael S. Tsirkin
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='000601c80525$0b661f30$ff0da8c0@amr.corp.intel.com' \
--to=sean.hefty@intel.com \
--cc=general@lists.openfabrics.org \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=rdreier@cisco.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).