From: Sagi Grimberg <sagi@grimberg.me>
To: Leon Romanovsky <leonro@mellanox.com>
Cc: Idan Burstein <idanb@mellanox.com>,
Saeed Mahameed <saeedm@mellanox.com>,
"David S. Miller" <davem@davemloft.net>,
Doug Ledford <dledford@redhat.com>,
Jason Gunthorpe <jgg@mellanox.com>,
Or Gerlitz <ogerlitz@mellanox.com>,
Tal Gilboa <talgi@mellanox.com>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
Yamin Friedman <yaminf@mellanox.com>,
Max Gurtovoy <maxg@mellanox.com>
Subject: Re: [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs
Date: Wed, 3 Jul 2019 11:56:04 -0700 [thread overview]
Message-ID: <8d525d64-6da1-48c3-952d-8c6b0d541859@grimberg.me> (raw)
In-Reply-To: <20190702064107.GS4727@mtr-leonro.mtl.com>
> Hi Sagi,
>
> I'm not sharing your worries about bad out-of-the-box experience for a
> number of reasons.
>
> First of all, this code is part of upstream kernel and will take time
> till users actually start to use it as is and not as part of some distro
> backports or MOFED packages.
True, but I am still saying that this feature is damaging sync IO which
represents the majority of the users. It might not be an extreme impact
but it is still a degradation (from a very limited testing I did this
morning I'm seeing a consistent 5%-10% latency increase for low QD
workloads which is consistent with what Yamin reported AFAIR).
But having said that, the call is for you guys to make as this is a
Mellanox device. I absolutely think that this is useful (as I said
before), I just don't think its necessarily a good idea to opt it by
default given that only a limited set of users would take full advantage
of it while the rest would see a negative impact (even if its 10%).
I don't have a hard objection here, just wanted to give you my
opinion on this because mlx5 is an important driver for rdma
users.
> Second, Yamin did extensive testing and worked very close with Or G.
> and I have very high confident in the results of their team work.
Has anyone tested other RDMA ulps? NFS/RDMA or SRP/iSER?
Would be interesting to understand how other subsystems with different
characteristics behave with this.
next prev parent reply other threads:[~2019-07-03 18:56 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-25 20:57 [pull request][for-next V2 0/7] Generic DIM lib for netdev and RDMA Saeed Mahameed
2019-06-25 20:57 ` [for-next V2 01/10] linux/dim: Move logic to dim.h Saeed Mahameed
2019-06-25 21:53 ` Sagi Grimberg
2019-06-25 20:57 ` [for-next V2 02/10] linux/dim: Remove "net" prefix from internal DIM members Saeed Mahameed
2019-06-25 21:53 ` Sagi Grimberg
2019-06-25 20:57 ` [for-next V2 03/10] linux/dim: Rename externally exposed macros Saeed Mahameed
2019-06-25 21:54 ` Sagi Grimberg
2019-06-25 20:57 ` [for-next V2 04/10] linux/dim: Rename net_dim_sample() to net_dim_update_sample() Saeed Mahameed
2019-06-25 21:55 ` Sagi Grimberg
2019-06-25 20:57 ` [for-next V2 05/10] linux/dim: Rename externally used net_dim members Saeed Mahameed
2019-06-25 21:57 ` Sagi Grimberg
2019-06-26 6:38 ` Tal Gilboa
2019-06-25 20:57 ` [for-next V2 06/10] linux/dim: Move implementation to .c files Saeed Mahameed
2019-07-02 16:15 ` Geert Uytterhoeven
2019-06-25 20:57 ` [for-next V2 07/10] linux/dim: Add completions count to dim_sample Saeed Mahameed
2019-06-25 20:57 ` [for-next V2 08/10] linux/dim: Implement rdma_dim Saeed Mahameed
2019-06-25 22:02 ` Sagi Grimberg
2019-06-26 11:57 ` Or Gerlitz
2019-06-27 5:25 ` Yamin Friedman
2019-06-25 20:57 ` [for-next V2 09/10] RDMA/nldev: Added configuration of RDMA dynamic interrupt moderation to netlink Saeed Mahameed
2019-06-25 21:15 ` Sagi Grimberg
2019-06-27 5:29 ` Yamin Friedman
2019-06-25 20:57 ` [for-next V2 10/10] RDMA/core: Provide RDMA DIM support for ULPs Saeed Mahameed
2019-06-25 21:14 ` Sagi Grimberg
2019-06-26 7:56 ` Idan Burstein
2019-07-02 5:36 ` Sagi Grimberg
2019-07-02 6:41 ` Leon Romanovsky
2019-07-03 18:56 ` Sagi Grimberg [this message]
2019-07-04 7:51 ` Leon Romanovsky
2019-07-04 12:30 ` Idan Burstein
2019-06-27 5:28 ` Yamin Friedman
2019-06-25 21:07 ` [pull request][for-next V2 0/7] Generic DIM lib for netdev and RDMA Saeed Mahameed
2019-06-27 19:43 ` David Miller
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=8d525d64-6da1-48c3-952d-8c6b0d541859@grimberg.me \
--to=sagi@grimberg.me \
--cc=davem@davemloft.net \
--cc=dledford@redhat.com \
--cc=idanb@mellanox.com \
--cc=jgg@mellanox.com \
--cc=leonro@mellanox.com \
--cc=linux-rdma@vger.kernel.org \
--cc=maxg@mellanox.com \
--cc=netdev@vger.kernel.org \
--cc=ogerlitz@mellanox.com \
--cc=saeedm@mellanox.com \
--cc=talgi@mellanox.com \
--cc=yaminf@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).