netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Steve Wise <swise@opengridcomputing.com>
To: Caitlin Bestler <caitlinb@broadcom.com>
Cc: Tom Tucker <tom@opengridcomputing.com>,
	Sean Hefty <mshefty@ichips.intel.com>,
	netdev@vger.kernel.org, rdreier@cisco.com,
	linux-kernel@vger.kernel.org, openib-general@openib.org
Subject: RE: [openib-general] Re: [PATCH 1/2] iWARP Connection Manager.
Date: Fri, 02 Jun 2006 08:57:44 -0500	[thread overview]
Message-ID: <1149256664.791.3.camel@stevo-desktop> (raw)
In-Reply-To: <54AD0F12E08D1541B826BE97C98F99F150D3E6@NT-SJCA-0751.brcm.ad.broadcom.com>

> > 
> > The problem is that we can't synchronously cancel an
> > outstanding connect request. Once we've asked the adapter to
> > connect, we can't tell him to stop, we have to wait for it to
> > fail. During the time period between when we ask to connect
> > and the adapter says yeah-or-nay, the user hits ctrl-C. This
> > is the case where disconnect and/or destroy gets called and
> > we have to block it waiting for the outstanding connect
> > request to complete.
> > 
> > One alternative to this approach is to do the kfree of the
> > cm_id in the deref logic. This was the original design and
> > leaves the object around to handle the completion of the
> > connect and still allows the app to clean up and go away
> > without all this waitin' around. When the adapter finally
> > finishes and releases it's reference, the object is kfree'd.
> > 
> > Hope this helps.
> > 
> Why couldn't you synchronously put the cm_id in a state of
> "pending delete" and do the actual delete when the RNIC
> provides a response to the request? 

This is Tom's "alternative" mentioned above.  The provider already keeps
an explicit reference on the cm_id while it might possibly deliver an
event on that cm_id.  So if you change deref to kfree the cm_id on its
last deref (when the refcnt reaches 0), then you can avoid blocking
during destroy...  

> There could even be
> an optional method to see if the device is capable of
> cancelling the request. I know it can't yank a SYN back
> from the wire, but it could refrain from retransmitting.

I would suggest we don't add this optional method until we see an RNIC
that supports canceling a connect request or accept synchronously...

Steve.


  reply	other threads:[~2006-06-02 13:57 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2006-06-01 22:28 [openib-general] Re: [PATCH 1/2] iWARP Connection Manager Caitlin Bestler
2006-06-02 13:57 ` Steve Wise [this message]
  -- strict thread matches above, loose matches on Subject: below --
2006-05-31 18:26 [PATCH 0/2][RFC] iWARP Core Support Steve Wise
2006-05-31 18:26 ` [PATCH 1/2] iWARP Connection Manager Steve Wise
2006-05-31 22:22   ` Sean Hefty
2006-06-01 17:00     ` Steve Wise
2006-06-01 21:09       ` Sean Hefty
2006-06-01 22:21         ` [openib-general] " Tom Tucker
2006-06-01 17:11     ` Tom Tucker

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=1149256664.791.3.camel@stevo-desktop \
    --to=swise@opengridcomputing.com \
    --cc=caitlinb@broadcom.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mshefty@ichips.intel.com \
    --cc=netdev@vger.kernel.org \
    --cc=openib-general@openib.org \
    --cc=rdreier@cisco.com \
    --cc=tom@opengridcomputing.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).