From: Leon Romanovsky <leon@kernel.org>
To: Cheng Xu <chengyou@linux.alibaba.com>
Cc: jgg@ziepe.ca, dledford@redhat.com, linux-rdma@vger.kernel.org,
KaiShen@linux.alibaba.com, tonylu@linux.alibaba.com
Subject: Re: [PATCH rdma-next 00/11] Elastic RDMA Adapter (ERDMA) driver
Date: Thu, 23 Dec 2021 15:44:29 +0200 [thread overview]
Message-ID: <YcR9PVDS2jFsrJ4N@unreal> (raw)
In-Reply-To: <9496ca18-760c-f90e-8735-f7fb2982e7a4@linux.alibaba.com>
On Thu, Dec 23, 2021 at 08:59:14PM +0800, Cheng Xu wrote:
>
>
> On 12/23/21 6:23 PM, Leon Romanovsky wrote:
> > On Wed, Dec 22, 2021 at 11:35:44AM +0800, Cheng Xu wrote:
> > >
> >
> > <...>
> >
> > > > >
> > > > > For the ECS instance with RDMA enabled, there are two kinds of devices
> > > > > allocated, one for ERDMA, and one for the original netdev (virtio-net).
> > > > > They are different PCI deivces. ERDMA driver can get the information about
> > > > > which netdev attached to in its PCIe barspace (by MAC address matching).
> > > >
> > > > This is very questionable. The netdev part should be kept in the
> > > > drivers/ethernet/... part of the kernel.
> > > >
> > > > Thanks
> > >
> > > The net device used in Alibaba ECS instance is virtio-net device, driven
> > > by virtio-pci/virtio-net drivers. ERDMA device does not need its own net
> > > device, and will be attached to an existed virtio-net device. The
> > > relationship between ibdev and netdev in erdma is similar to siw/rxe.
> >
> > siw/rxe binds through RDMA_NLDEV_CMD_NEWLINK netlink command and not
> > through MAC's matching.
> >
> > Thanks
>
> Both siw/rxe/erdma don't need to implement netdev part, this is what I
> wanted to express when I said 'similar'.
> What you mentioned (the bind mechanism) is one major difference between
> erdma and siw/rxe. For siw/rxe, user can attach ibdev to every netdev if
> he/she wants, but it is not true for erdma. When user buys the erdma
> service, he/she must specify which ENI (elastic network interface) to be
> binded, it means that the attached erdma device can only be binded to
> the specific netdev. Due to the uniqueness of MAC address in our ECS
> instance, we use the MAC address as the identification, then the driver
> knows which netdev should be binded to.
Nothing prohibits from you to implement this MAC check in RDMA_NLDEV_CMD_NEWLINK.
I personally don't like the idea that bind logic is performed "magically".
BTW,
1. No module parameters
2. No driver versions
Thanks
>
> Thanks,
> Cheng Xu
next prev parent reply other threads:[~2021-12-23 13:44 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-12-21 2:48 [PATCH rdma-next 00/11] Elastic RDMA Adapter (ERDMA) driver Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 01/11] RDMA: Add ERDMA to rdma_driver_id definition Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 02/11] RDMA/erdma: Add the hardware related definitions Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 03/11] RDMA/erdma: Add main include file Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 04/11] RDMA/erdma: Add cmdq implementation Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 05/11] RDMA/erdma: Add event queue implementation Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 06/11] RDMA/erdma: Add verbs header file Cheng Xu
2021-12-21 13:28 ` Leon Romanovsky
2021-12-22 2:36 ` Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 07/11] RDMA/erdma: Add verbs implementation Cheng Xu
2021-12-21 13:32 ` Leon Romanovsky
2021-12-21 15:20 ` Bernard Metzler
2021-12-22 3:11 ` Cheng Xu
2021-12-22 4:18 ` Cheng Xu
2021-12-22 12:46 ` Bernard Metzler
2021-12-23 8:38 ` Cheng Xu
2021-12-22 2:50 ` Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 08/11] RDMA/erdma: Add connection management (CM) support Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 09/11] RDMA/erdma: Add the erdma module Cheng Xu
2021-12-21 13:26 ` Leon Romanovsky
2021-12-22 2:33 ` Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 10/11] RDMA/erdma: Add the ABI definitions Cheng Xu
2021-12-21 11:57 ` kernel test robot
2021-12-22 16:14 ` kernel test robot
2021-12-23 15:46 ` Yanjun Zhu
2021-12-23 18:45 ` Leon Romanovsky
2021-12-23 22:55 ` Yanjun Zhu
2021-12-24 6:04 ` Leon Romanovsky
2021-12-24 7:54 ` Yanjun Zhu
2021-12-24 18:11 ` Leon Romanovsky
2021-12-24 7:12 ` Cheng Xu
2021-12-24 8:02 ` Yanjun Zhu
2021-12-24 18:19 ` Leon Romanovsky
2021-12-25 0:03 ` Yanjun Zhu
2021-12-25 3:36 ` Cheng Xu
2021-12-21 2:48 ` [PATCH rdma-next 11/11] RDMA/erdma: Add driver to kernel build environment Cheng Xu
2021-12-22 0:58 ` kernel test robot
2021-12-21 13:09 ` [PATCH rdma-next 00/11] Elastic RDMA Adapter (ERDMA) driver Leon Romanovsky
2021-12-22 3:35 ` Cheng Xu
2021-12-23 10:23 ` Leon Romanovsky
2021-12-23 12:59 ` Cheng Xu
2021-12-23 13:44 ` Leon Romanovsky [this message]
2021-12-24 7:07 ` Cheng Xu
2021-12-24 18:26 ` Leon Romanovsky
2021-12-25 2:54 ` Cheng Xu
2021-12-25 2:57 ` Cheng Xu
2021-12-25 3:03 ` [Please ignore the two former responses]Re: " Cheng Xu
2022-01-07 14:24 ` Jason Gunthorpe
2022-01-10 10:07 ` Cheng Xu
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=YcR9PVDS2jFsrJ4N@unreal \
--to=leon@kernel.org \
--cc=KaiShen@linux.alibaba.com \
--cc=chengyou@linux.alibaba.com \
--cc=dledford@redhat.com \
--cc=jgg@ziepe.ca \
--cc=linux-rdma@vger.kernel.org \
--cc=tonylu@linux.alibaba.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).