From: Max Gurtovoy <maxg@mellanox.com>
To: Sagi Grimberg <sagi@grimberg.me>, Tom Talpey <tom@talpey.com>,
santosh.shilimkar@oracle.com,
Aron Silverton <aron.silverton@oracle.com>
Cc: bvanassche@acm.org, Jason Gunthorpe <jgg@mellanox.com>,
linux-rdma@vger.kernel.org, dledford@redhat.com, leon@kernel.org,
israelr@mellanox.com, shlomin@mellanox.com
Subject: Re: [PATCH 0/8 v1] Remove FMR support from RDMA drivers
Date: Sun, 17 May 2020 13:51:44 +0300 [thread overview]
Message-ID: <3e4c0d12-d0f9-87f4-a256-ebe7d881de08@mellanox.com> (raw)
In-Reply-To: <53ac195c-dccd-3d30-9e11-f1389dc8332d@grimberg.me>
On 5/15/2020 9:59 PM, Sagi Grimberg wrote:
>
>> It's not only the fmr_pools. The FMR API operates on a page granularity,
>> so a sub-page registration, or a non-page-aligned one, ends up exposing
>> data outside the buffer. This is done silently, and is a significant
>> vulnerability for most upper layers.
>
> You're right Tom, I forgot about that...
> I guess I'd vote to remove it altogether then...
Santosh,
I don't really understand the mechanism you use from user/kernel
regarding the registration.
But I saw you use wait_event and wait for completing the
FRWR/INVALIDATION. For sure this is not optimal.
But can't really say what is the bottleneck in your case.
Can you share the numbers you get for FMR/FRWR benchmarks ?
next prev parent reply other threads:[~2020-05-17 10:51 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-14 12:02 [PATCH 0/8 v1] Remove FMR support from RDMA drivers Max Gurtovoy
2020-05-14 12:02 ` [PATCH 1/8] RDMA/mlx4: remove FMR support for memory registration Max Gurtovoy
2020-05-14 12:02 ` [PATCH 2/8] RDMA/rds: " Max Gurtovoy
2020-05-14 12:03 ` [PATCH 3/8] RDMA/mthca: " Max Gurtovoy
2020-05-14 12:03 ` [PATCH 4/8] RDMA/rdmavt: remove FMR " Max Gurtovoy
2020-05-14 12:03 ` [PATCH 5/8] RDMA/iser: Remove support for " Max Gurtovoy
2020-05-14 12:03 ` [PATCH 6/8] RDMA/srp: remove " Max Gurtovoy
2020-05-14 14:02 ` Bart Van Assche
2020-05-14 12:03 ` [PATCH 7/8] RDMA/core: remove FMR pool API Max Gurtovoy
2020-05-14 12:03 ` [PATCH 8/8] RDMA/core: remove FMR device ops Max Gurtovoy
2020-05-14 15:13 ` [PATCH 0/8 v1] Remove FMR support from RDMA drivers Aron Silverton
2020-05-14 18:18 ` santosh.shilimkar
2020-05-14 19:42 ` Max Gurtovoy
2020-05-14 22:23 ` Sagi Grimberg
2020-05-14 23:41 ` santosh.shilimkar
2020-05-15 16:52 ` Tom Talpey
2020-05-15 18:59 ` Sagi Grimberg
2020-05-17 10:51 ` Max Gurtovoy [this message]
2020-05-18 16:34 ` santosh.shilimkar
2020-05-15 0:37 ` Max Gurtovoy
2020-05-14 16:00 ` Gal Pressman
2020-05-17 10:37 ` Max Gurtovoy
2020-05-18 15:20 ` Dennis Dalessandro
2020-05-18 18:10 ` Jason Gunthorpe
2020-05-19 13:43 ` Dennis Dalessandro
2020-05-19 13:53 ` Jason Gunthorpe
2020-05-19 14:19 ` Leon Romanovsky
2020-05-19 14:26 ` Dennis Dalessandro
2020-05-19 14:30 ` Jason Gunthorpe
2020-05-19 14:37 ` Dennis Dalessandro
2020-05-23 22:08 ` Jason Gunthorpe
2020-05-24 1:27 ` Tom Talpey
2020-05-26 15:38 ` Dennis Dalessandro
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=3e4c0d12-d0f9-87f4-a256-ebe7d881de08@mellanox.com \
--to=maxg@mellanox.com \
--cc=aron.silverton@oracle.com \
--cc=bvanassche@acm.org \
--cc=dledford@redhat.com \
--cc=israelr@mellanox.com \
--cc=jgg@mellanox.com \
--cc=leon@kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=sagi@grimberg.me \
--cc=santosh.shilimkar@oracle.com \
--cc=shlomin@mellanox.com \
--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 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).