From: "Steve Wise" <swise@opengridcomputing.com>
To: "'Tom Talpey'" <tom@talpey.com>,
"'Jason Gunthorpe'" <jgunthorpe@obsidianresearch.com>
Cc: "'Chuck Lever'" <chuck.lever@oracle.com>,
<linux-rdma@vger.kernel.org>, <linux-nfs@vger.kernel.org>,
"'Doug Ledford'" <dledford@redhat.com>
Subject: RE: [PATCH v3 05/15] xprtrdma: Remove last ib_reg_phys_mr() call site
Date: Tue, 21 Jul 2015 16:22:52 -0500 [thread overview]
Message-ID: <00c901d0c3fb$66e7cfa0$34b76ee0$@opengridcomputing.com> (raw)
In-Reply-To:
> > >>>> B) why bother to check? Are machines with <4GB interesting, and worth
> > >>>> supporting a special optimization?
> > >>>
> > >>> No, but cxgb3 is still interesting to user applications, and perhaps NFSRDMA using FRMRs.
> > >>
> > >> I'm obviously not making myself clear. I am suggesting that cxgb3 fail
> > >> the ib_get_dma_mr() verb, regardless of installed memory.
> > >>
> > >> I am not suggesting it fail to load, or fail other memreg requests. It
> > >> should work normally in all other respects.
> > >
> > > Even with its limitation, doesn't it have utility for someone using cxgb3 in an embedded 32b environment?
> >
> > Sure, do you mean making it conditional on #if sizeof(physaddr) <= 32?
> > That would make sense I guess.
>
> No, a runtime check. x64 platforms will work too if the mem size takes <= 32b to describe.
Jason/Doug, do you think it should allow dma mr allocaton iff totalmem_pages < 4GB? Or do a check on the sizeof totalmem_pages and
only allow dma mrs if it is <= 4?
WARNING: multiple messages have this Message-ID (diff)
From: "Steve Wise" <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
To: 'Tom Talpey' <tom-CLs1Zie5N5HQT0dZR+AlfA@public.gmane.org>,
'Jason Gunthorpe'
<jgunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org>
Cc: 'Chuck Lever'
<chuck.lever-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
'Doug Ledford' <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Subject: RE: [PATCH v3 05/15] xprtrdma: Remove last ib_reg_phys_mr() call site
Date: Tue, 21 Jul 2015 16:22:52 -0500 [thread overview]
Message-ID: <00c901d0c3fb$66e7cfa0$34b76ee0$@opengridcomputing.com> (raw)
In-Reply-To:
> > >>>> B) why bother to check? Are machines with <4GB interesting, and worth
> > >>>> supporting a special optimization?
> > >>>
> > >>> No, but cxgb3 is still interesting to user applications, and perhaps NFSRDMA using FRMRs.
> > >>
> > >> I'm obviously not making myself clear. I am suggesting that cxgb3 fail
> > >> the ib_get_dma_mr() verb, regardless of installed memory.
> > >>
> > >> I am not suggesting it fail to load, or fail other memreg requests. It
> > >> should work normally in all other respects.
> > >
> > > Even with its limitation, doesn't it have utility for someone using cxgb3 in an embedded 32b environment?
> >
> > Sure, do you mean making it conditional on #if sizeof(physaddr) <= 32?
> > That would make sense I guess.
>
> No, a runtime check. x64 platforms will work too if the mem size takes <= 32b to describe.
Jason/Doug, do you think it should allow dma mr allocaton iff totalmem_pages < 4GB? Or do a check on the sizeof totalmem_pages and
only allow dma mrs if it is <= 4?
--
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:[~2015-07-21 21:22 UTC|newest]
Thread overview: 114+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-20 19:02 [PATCH v3 00/15] NFS/RDMA client side for Linux 4.3 Chuck Lever
2015-07-20 19:02 ` Chuck Lever
2015-07-20 19:02 ` [PATCH v3 01/15] xprtrdma: Make xprt_setup_rdma() agnostic to family of server address Chuck Lever
2015-07-20 19:02 ` Chuck Lever
2015-07-26 16:49 ` Christoph Hellwig
2015-07-26 16:49 ` Christoph Hellwig
2015-07-20 19:02 ` [PATCH v3 02/15] xprtrdma: Raise maximum payload size to one megabyte Chuck Lever
2015-07-20 19:02 ` Chuck Lever
2015-07-20 19:02 ` [PATCH v3 03/15] xprtrdma: Increase default credit limit Chuck Lever
2015-07-20 19:02 ` Chuck Lever
2015-07-20 19:03 ` [PATCH v3 04/15] xprtrdma: Don't fall back to PHYSICAL memory registration Chuck Lever
2015-07-20 19:03 ` Chuck Lever
2015-07-26 16:50 ` Christoph Hellwig
2015-07-26 16:50 ` Christoph Hellwig
2015-07-20 19:03 ` [PATCH v3 05/15] xprtrdma: Remove last ib_reg_phys_mr() call site Chuck Lever
2015-07-20 19:03 ` Chuck Lever
2015-07-20 20:34 ` Tom Talpey
2015-07-20 20:34 ` Tom Talpey
2015-07-20 20:55 ` Chuck Lever
2015-07-20 20:55 ` Chuck Lever
2015-07-20 21:55 ` Tom Talpey
2015-07-20 21:55 ` Tom Talpey
2015-07-20 22:21 ` Chuck Lever
2015-07-20 22:21 ` Chuck Lever
2015-07-20 22:30 ` Tom Talpey
2015-07-20 22:30 ` Tom Talpey
2015-07-20 21:05 ` Jason Gunthorpe
2015-07-20 21:05 ` Jason Gunthorpe
2015-07-20 21:16 ` Steve Wise
2015-07-20 21:16 ` Steve Wise
2015-07-20 22:04 ` Tom Talpey
2015-07-20 22:04 ` Tom Talpey
2015-07-20 22:17 ` Jason Gunthorpe
2015-07-20 22:17 ` Jason Gunthorpe
2015-07-20 22:26 ` Tom Talpey
2015-07-20 22:26 ` Tom Talpey
2015-07-20 22:41 ` Steve Wise
2015-07-20 22:41 ` Steve Wise
2015-07-20 22:42 ` Jason Gunthorpe
2015-07-20 22:42 ` Jason Gunthorpe
2015-07-21 22:41 ` Steve Wise
2015-07-21 22:41 ` Steve Wise
2015-07-21 22:54 ` Jason Gunthorpe
2015-07-21 22:54 ` Jason Gunthorpe
2015-07-22 13:58 ` Steve Wise
2015-07-22 13:58 ` Steve Wise
2015-07-21 0:15 ` Tom Talpey
2015-07-21 0:15 ` Tom Talpey
2015-07-21 14:33 ` Steve Wise
2015-07-21 14:33 ` Steve Wise
2015-07-21 20:47 ` Tom Talpey
2015-07-21 20:47 ` Tom Talpey
2015-07-21 20:55 ` Steve Wise
2015-07-21 20:55 ` Steve Wise
2015-07-21 21:22 ` Steve Wise [this message]
2015-07-21 21:22 ` Steve Wise
2015-07-20 21:34 ` Steve Wise
2015-07-20 21:34 ` Steve Wise
2015-07-20 21:37 ` Steve Wise
2015-07-20 21:37 ` Steve Wise
2015-07-20 22:13 ` Jason Gunthorpe
2015-07-20 22:13 ` Jason Gunthorpe
2015-07-20 22:43 ` Steve Wise
2015-07-20 22:43 ` Steve Wise
2015-07-20 22:54 ` Jason Gunthorpe
2015-07-20 22:54 ` Jason Gunthorpe
2015-07-20 22:54 ` Steve Wise
2015-07-20 22:54 ` Steve Wise
2015-07-20 22:26 ` Jason Gunthorpe
2015-07-20 22:26 ` Jason Gunthorpe
2015-07-20 22:31 ` Chuck Lever
2015-07-20 22:31 ` Chuck Lever
2015-07-20 22:41 ` Jason Gunthorpe
2015-07-20 22:41 ` Jason Gunthorpe
2015-07-20 23:36 ` Chuck Lever
2015-07-20 23:36 ` Chuck Lever
2015-07-21 0:11 ` Tom Talpey
2015-07-21 0:11 ` Tom Talpey
2015-07-21 0:34 ` Chuck Lever
2015-07-21 0:34 ` Chuck Lever
2015-07-21 1:03 ` Tom Talpey
2015-07-21 1:03 ` Tom Talpey
2015-07-20 19:03 ` [PATCH v3 06/15] xprtrdma: Clean up rpcrdma_ia_open() Chuck Lever
2015-07-20 19:03 ` Chuck Lever
2015-07-26 16:53 ` Christoph Hellwig
2015-07-26 16:53 ` Christoph Hellwig
2015-07-26 18:21 ` Chuck Lever
2015-07-26 18:21 ` Chuck Lever
2015-07-26 18:51 ` Christoph Hellwig
2015-07-26 18:51 ` Christoph Hellwig
2015-07-20 19:03 ` [PATCH v3 07/15] xprtrdma: Remove logic that constructs RDMA_MSGP type calls Chuck Lever
2015-07-20 19:03 ` Chuck Lever
2015-07-20 19:03 ` [PATCH v3 08/15] xprtrdma: Account for RPC/RDMA header size when deciding to inline Chuck Lever
2015-07-20 19:03 ` Chuck Lever
2015-07-20 19:03 ` [PATCH v3 09/15] xprtrdma: Always provide a write list when sending NFS READ Chuck Lever
2015-07-20 19:03 ` Chuck Lever
2015-07-20 19:03 ` [PATCH v3 10/15] xprtrdma: Don't provide a reply chunk when expecting a short reply Chuck Lever
2015-07-20 19:03 ` Chuck Lever
2015-07-20 19:04 ` [PATCH v3 11/15] xprtrdma: Fix XDR tail buffer marshalling Chuck Lever
2015-07-20 19:04 ` Chuck Lever
2015-07-20 19:04 ` [PATCH v3 12/15] xprtrdma: Fix large NFS SYMLINK calls Chuck Lever
2015-07-20 19:04 ` Chuck Lever
2015-07-20 19:04 ` [PATCH v3 13/15] xprtrdma: Clean up xprt_rdma_print_stats() Chuck Lever
2015-07-20 19:04 ` Chuck Lever
2015-07-20 19:04 ` [PATCH v3 14/15] xprtrdma: Count RDMA_NOMSG type calls Chuck Lever
2015-07-20 19:04 ` Chuck Lever
2015-07-20 19:04 ` [PATCH v3 15/15] core: Remove the ib_reg_phys_mr() and ib_rereg_phys_mr() verbs Chuck Lever
2015-07-20 19:04 ` Chuck Lever
2015-07-21 20:08 ` Anna Schumaker
2015-07-21 20:08 ` Anna Schumaker
2015-07-21 20:16 ` Chuck Lever
2015-07-21 20:16 ` Chuck Lever
2015-07-21 20:18 ` Anna Schumaker
2015-07-21 20:18 ` Anna Schumaker
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='00c901d0c3fb$66e7cfa0$34b76ee0$@opengridcomputing.com' \
--to=swise@opengridcomputing.com \
--cc=chuck.lever@oracle.com \
--cc=dledford@redhat.com \
--cc=jgunthorpe@obsidianresearch.com \
--cc=linux-nfs@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=tom@talpey.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.