linux-rdma.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Bart Van Assche <bart.vanassche@sandisk.com>
To: Christoph Hellwig <hch@lst.de>
Cc: "dledford@redhat.com" <dledford@redhat.com>,
	"sagig@mellanox.com" <sagig@mellanox.com>,
	"swise@opengridcomputing.com" <swise@opengridcomputing.com>,
	"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
	"target-devel@vger.kernel.org" <target-devel@vger.kernel.org>
Subject: Re: [PATCH v4 7/9] IB/core: generic RDMA READ/WRITE API
Date: Thu, 10 Mar 2016 07:30:36 -0800	[thread overview]
Message-ID: <56E1931C.1040206@sandisk.com> (raw)
In-Reply-To: <20160310075722.GA551@lst.de>

On 03/09/16 23:57, Christoph Hellwig wrote:
> On Wed, Mar 09, 2016 at 10:55:50AM -0800, Bart Van Assche wrote:
>> Regarding the iSERt patches that are present in your rdma-rw-api branch:
>
> I haven't submitted those yet, as the signature MR support hasn't been
> tested and is most likely broken.  (and even if it's not broken it's too
> ugly to live :))
>
>> what is the impact of these patches on memory registration for the iSERt
>> driver in combination with an IB HCA? Is memory registration still enabled
>> for this combination? If not: how about changing this patch series such
>> that memory registration is performed either if the transport protocol is
>> iWARP or the ULP driver requests memory registration explicitly, e.g. by
>> setting a flag in struct rdma_rw_ctx and/or struct ib_qp_init_attr?
>
> I don't want the ULP to ask for memory registrations, and it's not really
> the ULPs business.  It should be done either if needed (iWarp, signature
> MRs), or if the HCA driver prefers it for a larger enough transfers.
>
> Once the basic API is in I'd like to add an interface for the HCA driver,
> as Mellanox has indicated they'd like to register for larger transfers.
>
> Any chance to get a review of the remaining patches so we can get it
> in for this merge window?

Hello Christoph,

I will continue with reviewing these patches. But I disagree that ULPs 
should not have control over whether or not memory registration is 
enabled. My experience is that it is essential to always enable memory 
registration while debugging an RDMA driver. Without memory registration 
it can be extremely hard to determine whether memory corruption is 
caused by code that is running locally or by a misplaced RDMA write from 
another host. With memory registration enabled misplaced writes result 
in an access violation error. Additionally, certain users prefer to 
enable memory registration for all RDMA transfers because of the 
additional safety it provides.

Bart.

  reply	other threads:[~2016-03-10 15:30 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-03-08 18:16 generic RDMA READ/WRITE API V4 Christoph Hellwig
2016-03-08 18:16 ` [PATCH v4 1/9] IB/cma: pass the port number to ib_create_qp Christoph Hellwig
2016-03-08 19:25   ` Bart Van Assche
2016-03-08 19:27   ` Leon Romanovsky
     [not found]     ` <20160308192709.GJ13396-2ukJVAZIZ/Y@public.gmane.org>
2016-03-08 19:38       ` Steve Wise
2016-03-08 19:55         ` Leon Romanovsky
2016-03-09 13:58           ` 'Christoph Hellwig'
     [not found]             ` <20160309135804.GA32315-jcswGhMUV9g@public.gmane.org>
2016-03-09 14:43               ` Leon Romanovsky
2016-03-09 15:51               ` Steve Wise
2016-03-09 15:52                 ` 'Christoph Hellwig'
2016-03-08 18:16 ` [PATCH v4 2/9] IB/core: allow passing mapping an offset into the SG in ib_map_mr_sg Christoph Hellwig
2016-03-08 19:27   ` Bart Van Assche
2016-03-08 18:16 ` [PATCH v4 3/9] IB/core: add a helper to check for READ WITH INVALIDATE support Christoph Hellwig
2016-03-08 18:16 ` [PATCH v4 4/9] IB/core: refactor ib_create_qp Christoph Hellwig
2016-03-08 19:33   ` Bart Van Assche
     [not found]     ` <56DF2917.309-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
2016-03-09 12:55       ` Christoph Hellwig
2016-03-08 18:16 ` [PATCH v4 5/9] IB/core: add a simple MR pool Christoph Hellwig
     [not found]   ` <1457461000-24088-6-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
2016-03-08 19:37     ` Bart Van Assche
2016-03-09 12:56       ` Christoph Hellwig
     [not found] ` <1457461000-24088-1-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
2016-03-08 18:16   ` [PATCH v4 6/9] IB/core: add a need_inval flag to struct ib_mr Christoph Hellwig
2016-03-08 19:41     ` Bart Van Assche
2016-03-08 18:16   ` [PATCH v4 7/9] IB/core: generic RDMA READ/WRITE API Christoph Hellwig
2016-03-08 21:32     ` Bart Van Assche
2016-03-09 13:07       ` Christoph Hellwig
2016-03-09 15:54         ` Christoph Hellwig
2016-03-09 18:55           ` Bart Van Assche
2016-03-10  7:57             ` Christoph Hellwig
2016-03-10 15:30               ` Bart Van Assche [this message]
     [not found]                 ` <56E1931C.1040206-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
2016-03-10 16:21                   ` Christoph Hellwig
     [not found]                     ` <20160310162150.GA9544-jcswGhMUV9g@public.gmane.org>
2016-03-10 17:00                       ` Bart Van Assche
     [not found]                         ` <56E1A810.1080001-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
2016-03-10 17:01                           ` Christoph Hellwig
2016-03-11 22:39                       ` Bart Van Assche
     [not found]                         ` <56E34914.2020108-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
2016-03-12  6:12                           ` Christoph Hellwig
     [not found]                             ` <20160312061232.GA15236-jcswGhMUV9g@public.gmane.org>
2016-03-14  1:04                               ` Bart Van Assche
     [not found]                                 ` <56E60E0F.7060900-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
2016-03-15  8:45                                   ` Christoph Hellwig
     [not found]           ` <20160309155457.GA1898-jcswGhMUV9g@public.gmane.org>
2016-03-11  1:26             ` Bart Van Assche
2016-03-11  9:22               ` Christoph Hellwig
2016-03-08 18:16   ` [PATCH v4 8/9] target: enhance and export target_alloc_sgl/target_free_sgl Christoph Hellwig
2016-03-08 21:37     ` Bart Van Assche
2016-03-08 18:16 ` [PATCH v4 9/9] IB/srpt: convert to the generic RDMA READ/WRITE API Christoph Hellwig
2016-03-08 20:13   ` Leon Romanovsky
2016-03-09 12:59     ` Christoph Hellwig
2016-03-10 22:08       ` Bart Van Assche
2016-03-11  9:09         ` Christoph Hellwig
     [not found]   ` <1457461000-24088-10-git-send-email-hch-jcswGhMUV9g@public.gmane.org>
2016-03-10 22:40     ` Bart Van Assche
     [not found]       ` <56E1F7C8.9000101-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
2016-03-11  9:14         ` Christoph Hellwig

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=56E1931C.1040206@sandisk.com \
    --to=bart.vanassche@sandisk.com \
    --cc=dledford@redhat.com \
    --cc=hch@lst.de \
    --cc=linux-rdma@vger.kernel.org \
    --cc=sagig@mellanox.com \
    --cc=swise@opengridcomputing.com \
    --cc=target-devel@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).