All of lore.kernel.org
 help / color / mirror / Atom feed
From: Yuval Shaia <yuval.shaia@oracle.com>
To: Ira Weiny <ira.weiny@intel.com>
Cc: Shamir Rabinovitch <srabinov7@gmail.com>,
	Jason Gunthorpe <jgg@ziepe.ca>,
	Christoph Hellwig <hch@infradead.org>,
	dledford@redhat.com, leon@kernel.org, monis@mellanox.com,
	parav@mellanox.com, danielj@mellanox.com, kamalheib1@gmail.com,
	markz@mellanox.com, swise@opengridcomputing.com,
	shamir.rabinovitch@oracle.com, johannes.berg@intel.com,
	willy@infradead.org, michaelgur@mellanox.com, markb@mellanox.com,
	dan.carpenter@oracle.com, bvanassche@acm.org, maxg@mellanox.com,
	israelr@mellanox.com, galpress@amazon.com, denisd@mellanox.com,
	yuvalav@mellanox.com, dennis.dalessandro@intel.com,
	will@kernel.org, ereza@mellanox.com, linux-rdma@vger.kernel.org
Subject: Re: [PATCH 00/25] Shared PD and MR
Date: Thu, 1 Aug 2019 07:05:05 +0300	[thread overview]
Message-ID: <20190801040504.GD2929@lap1> (raw)
In-Reply-To: <20190717235526.GB4936@iweiny-DESK2.sc.intel.com>

On Wed, Jul 17, 2019 at 04:55:26PM -0700, Ira Weiny wrote:
> On Wed, Jul 17, 2019 at 04:35:30PM +0300, Shamir Rabinovitch wrote:
> > On Wed, Jul 17, 2019 at 2:55 PM Jason Gunthorpe <jgg@ziepe.ca> wrote:
> > >
> > > On Wed, Jul 17, 2019 at 02:09:50PM +0300, Shamir Rabinovitch wrote:
> > > > On Wed, Jul 17, 2019 at 8:09 AM Christoph Hellwig <hch@infradead.org> wrote:
> > > > >
> > > > > On Tue, Jul 16, 2019 at 09:11:35PM +0300, Shamir Rabinovitch wrote:
> > > > > > Following patch-set introduce the shared object feature.
> > > > > >
> > > > > > A shared object feature allows one process to create HW objects (currently
> > > > > > PD and MR) so that a second process can import.
> > > > >
> > > > > That sounds like a major complication, so you'd better also explain
> > > > > the use case very well.
> > > >
> > > > The main use case was that there is a server that has giant shared
> > > > memory that is shared across many processes (lots of mtts).
> > > > Each process needs the same memory registration (lots of mrs that
> > > > register same memory).
> > > > In such scenario, the HCA runs out of mtts.
> > > > To solve this problem, an single memory registration is shared across
> > > > all the process in that server saving hca mtts.
> > >
> > > Well, why not just share the entire uverbs FD then? Once the PD is
> > > shared all security is lost anyhow..
> > >
> > > This is not the model that was explained to me last year
> > >
> > > Jason
> > 
> > We do share the whole uvrbs FD (context) with the second process and
> > let that process to instantiate the PD & MR from the shared FD.
> 
> Then the first (both) process(es) should have access to the MR right?

Yes.
(Please note that since we maintain refcount then the ib_mr object will be
destroyed only when the two will destroy it).

> 
> > The instantiation include creating new uobject in the second process
> > context that points to the same ib_x HW objects.
> > The second process does not own the shared context.
> > It just use it to get access to the shared ib_x objects and then it
> > mark those & shared FD as shared.
> 
> I'm not following this?

Shamir, correct me if i'm wrong here:
So there is one ib_mr object and two uobjects that "points" to it.

> 
> > 
> > What was the expectation from "import_from_xxx" ?
> 
> ... and I don't understand this question.

Shamir?

> 
> Ira
> 

  reply	other threads:[~2019-08-01  4:07 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-07-16 18:11 [PATCH 00/25] Shared PD and MR Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 01/25] RDMA/uverbs: uobj_get_obj_read should return the ib_uobject Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 02/25] RDMA/uverbs: Delete the macro uobj_put_obj_read Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 03/25] RDMA/nldev: ib_pd can be pointed by multiple ib_ucontext Shamir Rabinovitch
2019-07-18 16:05   ` Leon Romanovsky
2019-07-16 18:11 ` [PATCH 04/25] IB/{core,hw}: ib_pd should not have ib_uobject pointer Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 05/25] IB/core: ib_uobject need HW object reference count Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 06/25] IB/uverbs: Helper function to initialize ufile member of uverbs_attr_bundle Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 07/25] IB/uverbs: Add context import lock/unlock helper Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 08/25] IB/uverbs: ufile must be freed only when not used anymore Shamir Rabinovitch
2019-07-17 11:53   ` Jason Gunthorpe
2019-07-17 19:25     ` Shamir Rabinovitch
2019-07-17 19:33       ` Jason Gunthorpe
2019-07-17 20:31         ` Yuval Shaia
2019-07-17 20:45           ` Matthew Wilcox
2019-07-17 21:36             ` Yuval Shaia
2019-07-17 23:51               ` Ira Weiny
2019-07-18 12:17               ` Jason Gunthorpe
2019-07-18 20:45                 ` Yuval Shaia
2019-07-19 11:46                   ` Jason Gunthorpe
2019-07-16 18:11 ` [PATCH 09/25] IB/verbs: Prototype of HW object clone callback Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 10/25] IB/core: Install clone ib_pd in device ops Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 11/25] IB/mlx4: Add implementation of clone_pd callback Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 12/25] IB/mlx5: " Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 13/25] RDMA/rxe: " Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 14/25] IB/uverbs: Add clone reference counting to ib_pd Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 15/25] IB/uverbs: Add PD import verb Shamir Rabinovitch
2019-07-17 11:44   ` Jason Gunthorpe
2019-07-17 20:15     ` Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 16/25] IB/mlx4: Enable import from FD verb Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 17/25] IB/mlx5: " Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 18/25] RDMA/rxe: " Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 19/25] IB/core: ib_mr should not have ib_uobject pointer Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 20/25] IB/core: Install clone ib_mr in device ops Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 21/25] IB/mlx4: Add implementation of clone_pd callback Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 22/25] IB/mlx5: " Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 23/25] RDMA/rxe: " Shamir Rabinovitch
2019-07-16 18:11 ` [PATCH 24/25] IB/uverbs: Add clone reference counting to ib_mr Shamir Rabinovitch
2019-07-16 18:12 ` [PATCH 25/25] IB/uverbs: Add MR import verb Shamir Rabinovitch
2019-07-17  5:09 ` [PATCH 00/25] Shared PD and MR Christoph Hellwig
2019-07-17 11:09   ` Shamir Rabinovitch
2019-07-17 11:55     ` Jason Gunthorpe
2019-07-17 13:35       ` Shamir Rabinovitch
2019-07-17 23:55         ` Ira Weiny
2019-08-01  4:05           ` Yuval Shaia [this message]
2019-07-18 12:16         ` Jason Gunthorpe

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=20190801040504.GD2929@lap1 \
    --to=yuval.shaia@oracle.com \
    --cc=bvanassche@acm.org \
    --cc=dan.carpenter@oracle.com \
    --cc=danielj@mellanox.com \
    --cc=denisd@mellanox.com \
    --cc=dennis.dalessandro@intel.com \
    --cc=dledford@redhat.com \
    --cc=ereza@mellanox.com \
    --cc=galpress@amazon.com \
    --cc=hch@infradead.org \
    --cc=ira.weiny@intel.com \
    --cc=israelr@mellanox.com \
    --cc=jgg@ziepe.ca \
    --cc=johannes.berg@intel.com \
    --cc=kamalheib1@gmail.com \
    --cc=leon@kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=markb@mellanox.com \
    --cc=markz@mellanox.com \
    --cc=maxg@mellanox.com \
    --cc=michaelgur@mellanox.com \
    --cc=monis@mellanox.com \
    --cc=parav@mellanox.com \
    --cc=shamir.rabinovitch@oracle.com \
    --cc=srabinov7@gmail.com \
    --cc=swise@opengridcomputing.com \
    --cc=will@kernel.org \
    --cc=willy@infradead.org \
    --cc=yuvalav@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 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.