From: Alan Cook <acook-osxm6dDZNBBZx8iatJs59jGjJy/sRE9J@public.gmane.org>
To: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: Converting from RC to UC
Date: Fri, 11 Mar 2011 20:10:18 +0000 (UTC) [thread overview]
Message-ID: <loom.20110311T210251-557@post.gmane.org> (raw)
In-Reply-To: 20110311185831.GX22729@obsidianresearch.com
Jason Gunthorpe <jgunthorpe@...> writes:
> I think what you are hitting is some code is calling ibv_modify_qp
> with an invalid qp_attr_mask for one of the state transitions. UC and
> RC have different requirements. Assuming you are not calling
> ibv_modify_qp in your app it is probably a kernel bug?
>
No calls are made to ibv_modify_qp. The application is failing on the connect,
so there has not been any application specific operations performed over
InfiniBand yet (other than the connection setup). The application itself is
quite basic: eight application specific messages to be passed over InfiniBand,
along with RDMA write messages. It can be thought of as a glorified scp
operation.
I am using the vanilla 2.6.27.57 kernel. If the problem is a kernel bug, I
would expect the kernel to not support UC at all, given the simplicity of the
application.
The only change made to the application for switching from RC to UC was to
change the queue type, which does not appear to work. It seems that there are
other setup operations performed for the RC server which are not compatible with
a UC server, but no calls return errors until rdma_accept(). What other
settings/setup would be causing the issue that would not return an error?
I have found no examples for setting up a basic IB application using UC. All
examples are with RC. Does anyone know of any examples for UC that they could
point me towards?
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2011-03-11 20:10 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-03-11 18:29 Converting from RC to UC Alan Cook
[not found] ` <loom.20110311T192125-878-eS7Uydv5nfjZ+VzJOa5vwg@public.gmane.org>
2011-03-11 18:58 ` Jason Gunthorpe
2011-03-11 20:10 ` Alan Cook [this message]
[not found] ` <AANLkTikqv-WYcL3_=5HRU6mVORLydNgcqv-5_VYkJrbC@mail.gmail.com>
[not found] ` <AANLkTikqv-WYcL3_=5HRU6mVORLydNgcqv-5_VYkJrbC-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2011-03-11 21:54 ` Jason Gunthorpe
2011-03-11 21:57 ` Alan Cook
2011-03-11 21:42 ` Roland Dreier
2011-03-11 21:55 ` Alan Cook
[not found] ` <loom.20110311T224601-402-eS7Uydv5nfjZ+VzJOa5vwg@public.gmane.org>
2011-03-11 22:07 ` Jason Gunthorpe
2011-03-13 9:56 ` Or Gerlitz
2011-03-14 1:08 ` Alan Cook
[not found] ` <loom.20110314T020520-335-eS7Uydv5nfjZ+VzJOa5vwg@public.gmane.org>
2011-03-14 15:05 ` frank zago
[not found] ` <AANLkTimM0PDiWuHFJfn3s-JVEKWHVVfL4+2Kcc4+G5C7-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2011-03-14 15:26 ` Hefty, Sean
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=loom.20110311T210251-557@post.gmane.org \
--to=acook-osxm6ddznbbzx8iatjs59jgjjy/sre9j@public.gmane.org \
--cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
/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