From: Sagi Grimberg <sagigrim@gmail.com>
To: Yishai Hadas <yishaih@dev.mellanox.co.il>,
Chuck Lever <chuck.lever@oracle.com>
Cc: leon@kernel.org, Or Gerlitz <gerlitz.or@gmail.com>,
Yishai Hadas <yishaih@mellanox.com>,
linux-rdma <linux-rdma@vger.kernel.org>,
Linux NFS Mailing List <linux-nfs@vger.kernel.org>,
Majd Dibbiny <majd@mellanox.com>
Subject: Re: [PATCH v2 01/24] mlx4-ib: Use coherent memory for priv pages
Date: Tue, 21 Jun 2016 16:56:44 +0300 [thread overview]
Message-ID: <5769479C.8070605@gmail.com> (raw)
In-Reply-To: <12ee28bb-b838-ed4c-5f84-0cb8f1760d63@dev.mellanox.co.il>
> Just found the root cause of the problem, it was found to be a hardware
> limitation that is described as part of the PRM. The driver code had to
> be written accordingly, confirmed that internally with the relevant people.
>
> From PRM:
> "The PBL should be physically contiguous, must reside in a
> 64-byte-aligned address, and must not include the last 8 bytes of a page."
>
> The last sentence pointed that only one page can be used as the last 8
> bytes should not be included. That's why there is a hard limit in the
> code for 511 entries.
>
> Re the candidate fix that you sent, from initial review it makes sense,
> we'll formally confirm it soon after finalizing the regression testing
> in our side.
>
> Thanks Chuck and Sagi for evaluating and working on a solution.
Thanks Yishai,
That clears up the root-cause.
Does the same holds for mlx5? or we can leave it alone?
next prev parent reply other threads:[~2016-06-21 13:56 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-15 3:15 [PATCH v2 00/24] NFS/RDMA client patches proposed for v4.8 Chuck Lever
2016-06-15 3:15 ` [PATCH v2 01/24] mlx4-ib: Use coherent memory for priv pages Chuck Lever
2016-06-15 4:28 ` Leon Romanovsky
2016-06-15 16:40 ` Chuck Lever
2016-06-16 14:35 ` Leon Romanovsky
2016-06-16 21:10 ` Sagi Grimberg
2016-06-16 21:58 ` Chuck Lever
2016-06-17 9:20 ` Leon Romanovsky
2016-06-17 19:55 ` Chuck Lever
2016-06-18 10:56 ` Leon Romanovsky
2016-06-18 20:08 ` Chuck Lever
2016-06-19 10:04 ` Sagi Grimberg
2016-06-19 19:38 ` Or Gerlitz
2016-06-19 19:43 ` Or Gerlitz
2016-06-19 20:02 ` Chuck Lever
2016-06-20 5:44 ` Leon Romanovsky
2016-06-20 6:34 ` Sagi Grimberg
2016-06-20 7:01 ` Leon Romanovsky
2016-06-20 8:35 ` Sagi Grimberg
2016-06-20 13:41 ` Yishai Hadas
2016-06-21 13:56 ` Sagi Grimberg [this message]
2016-06-21 14:35 ` Laurence Oberman
2016-06-19 9:58 ` Sagi Grimberg
2016-06-19 9:48 ` Sagi Grimberg
2016-06-17 9:05 ` Leon Romanovsky
2016-06-19 7:05 ` Sagi Grimberg
2016-06-15 3:15 ` [PATCH v2 02/24] xprtrdma: Remove FMRs from the unmap list after unmapping Chuck Lever
2016-06-15 3:15 ` [PATCH v2 03/24] xprtrdma: Create common scatterlist fields in rpcrdma_mw Chuck Lever
2016-06-15 3:15 ` [PATCH v2 04/24] xprtrdma: Move init and release helpers Chuck Lever
2016-06-15 3:15 ` [PATCH v2 05/24] xprtrdma: Rename fields in rpcrdma_fmr Chuck Lever
2016-06-15 3:16 ` [PATCH v2 06/24] xprtrdma: Use scatterlist for DMA mapping and unmapping under FMR Chuck Lever
2016-06-15 3:16 ` [PATCH v2 07/24] xprtrdma: Refactor MR recovery work queues Chuck Lever
2016-06-15 3:16 ` [PATCH v2 08/24] xprtrdma: Do not leak an MW during a DMA map failure Chuck Lever
2016-06-15 3:16 ` [PATCH v2 09/24] xprtrdma: Remove ALLPHYSICAL memory registration mode Chuck Lever
2016-06-15 3:16 ` [PATCH v2 10/24] xprtrdma: Remove rpcrdma_map_one() and friends Chuck Lever
2016-06-15 3:16 ` [PATCH v2 11/24] xprtrdma: Reply buffer exhaustion can be catastrophic Chuck Lever
2016-06-15 3:16 ` [PATCH v2 12/24] xprtrdma: Honor ->send_request API contract Chuck Lever
2016-06-15 3:17 ` [PATCH v2 13/24] xprtrdma: Chunk list encoders must not return zero Chuck Lever
2016-06-15 3:17 ` [PATCH v2 14/24] xprtrdma: Allocate MRs on demand Chuck Lever
2016-06-15 3:17 ` [PATCH v2 15/24] xprtrdma: Release orphaned MRs immediately Chuck Lever
2016-06-15 3:17 ` [PATCH v2 16/24] xprtrdma: Place registered MWs on a per-req list Chuck Lever
2016-06-15 3:17 ` [PATCH v2 17/24] xprtrdma: Chunk list encoders no longer share one rl_segments array Chuck Lever
2016-06-15 3:17 ` [PATCH v2 18/24] xprtrdma: rpcrdma_inline_fixup() overruns the receive page list Chuck Lever
2016-06-15 3:17 ` [PATCH v2 19/24] xprtrdma: Do not update {head, tail}.iov_len in rpcrdma_inline_fixup() Chuck Lever
2016-06-15 3:18 ` [PATCH v2 20/24] xprtrdma: Update only specific fields in private receive buffer Chuck Lever
2016-06-15 3:18 ` [PATCH v2 21/24] xprtrdma: Clean up fixup_copy_count accounting Chuck Lever
2016-06-15 3:18 ` [PATCH v2 22/24] xprtrdma: No direct data placement with krb5i and krb5p Chuck Lever
2016-06-15 3:18 ` [PATCH v2 23/24] svc: Avoid garbage replies when pc_func() returns rpc_drop_reply Chuck Lever
2016-06-15 3:18 ` [PATCH v2 24/24] NFS: Don't drop CB requests with invalid principals Chuck Lever
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=5769479C.8070605@gmail.com \
--to=sagigrim@gmail.com \
--cc=chuck.lever@oracle.com \
--cc=gerlitz.or@gmail.com \
--cc=leon@kernel.org \
--cc=linux-nfs@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=majd@mellanox.com \
--cc=yishaih@dev.mellanox.co.il \
--cc=yishaih@mellanox.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).