linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sagi Grimberg <sagigrim@gmail.com>
To: Chuck Lever <chuck.lever@oracle.com>
Cc: leon@kernel.org, linux-rdma@vger.kernel.org,
	Linux NFS Mailing List <linux-nfs@vger.kernel.org>
Subject: Re: [PATCH v2 01/24] mlx4-ib: Use coherent memory for priv pages
Date: Sun, 19 Jun 2016 12:48:51 +0300	[thread overview]
Message-ID: <57666A83.8050502@gmail.com> (raw)
In-Reply-To: <652EBA09-2978-414C-8606-38A96C63365A@oracle.com>


>> First of all, IIRC the patch author was Christoph wasn't he.
>>
>> Plus, you do realize that this patch makes the pages allocation
>> in granularity of pages. In systems with a large page size this
>> is completely redundant, it might even be harmful as the storage
>> ULPs need lots of MRs.
>
> I agree that the official fix should take a conservative
> approach to allocating this resource; there will be lots
> of MRs in an active system. This fix doesn't seem too
> careful.
>
>
>> Also, I don't see how that solves the issue, I'm not sure I even
>> understand the issue. Do you? Were you able to reproduce it?
>
> The issue is that dma_map_single() does not seem to DMA map
> portions of a memory region that are past the end of the first
> page of that region. Maybe that's a bug?

That seems weird to me, from looking at the code I didn't see
any indication that such a mapping would fail. Maybe we are seeing
a mlx4 specific issue? If this is some kind of generic dma-mapping
bug mlx5 would suffer from the same problem right? does it?

> This patch works around that behavior by guaranteeing that
>
> a) the memory region starts at the beginning of a page, and
> b) the memory region is never larger than a page
>
> This patch is not sufficient to repair mlx5, because b)
> cannot be satisfied in that case; the array of __be64's can
> be larger than 512 entries.

If a single page boundary is indeed the root-cause then I agree
this would not solve the problem for mlx5.

>> IFF the pages buffer end not being aligned to a cacheline is problematic
>> then why not extent it to end in a cacheline? Why in the next full page?
>
> I think the patch description justifies the choice of
> solution, but does not describe the original issue at
> all. The original issue had nothing to do with cacheline
> alignment.
>
> Lastly, this patch should remove the definition of
> MLX4_MR_PAGES_ALIGN.

The mlx4 PRM explicitly states that the translation (pages) vector
should align to 64 bytes and this is where this define comes from,
hence I don't think it should be removed from the code.

  parent reply	other threads:[~2016-06-19  9:48 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
2016-06-21 14:35                                 ` Laurence Oberman
2016-06-19  9:58                 ` Sagi Grimberg
2016-06-19  9:48             ` Sagi Grimberg [this message]
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=57666A83.8050502@gmail.com \
    --to=sagigrim@gmail.com \
    --cc=chuck.lever@oracle.com \
    --cc=leon@kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.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;
as well as URLs for NNTP newsgroup(s).