From: Jason Gunthorpe <jgg@nvidia.com>
To: Jacob Keller <jacob.e.keller@intel.com>
Cc: "Saleem, Shiraz" <shiraz.saleem@intel.com>,
Leon Romanovsky <leon@kernel.org>,
"dledford@redhat.com" <dledford@redhat.com>,
"kuba@kernel.org" <kuba@kernel.org>,
"davem@davemloft.net" <davem@davemloft.net>,
"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
"gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"Ertman, David M" <david.m.ertman@intel.com>,
"Nguyen, Anthony L" <anthony.l.nguyen@intel.com>,
"Ismail, Mustafa" <mustafa.ismail@intel.com>,
"jiri@nvidia.com" <jiri@nvidia.com>,
"Samudrala, Sridhar" <sridhar.samudrala@intel.com>,
"Williams, Dan J" <dan.j.williams@intel.com>
Subject: Re: [PATCH 07/22] RDMA/irdma: Register an auxiliary driver and implement private channel OPs
Date: Mon, 25 Jan 2021 21:10:43 -0400 [thread overview]
Message-ID: <20210126011043.GG4147@nvidia.com> (raw)
In-Reply-To: <5b3f609d-034a-826f-1e50-0a5f8ad8406e@intel.com>
On Mon, Jan 25, 2021 at 05:01:40PM -0800, Jacob Keller wrote:
>
>
> On 1/25/2021 4:39 PM, Saleem, Shiraz wrote:
> >> Subject: Re: [PATCH 07/22] RDMA/irdma: Register an auxiliary driver and
> >> implement private channel OPs
> >>
> >> On Sun, Jan 24, 2021 at 03:45:51PM +0200, Leon Romanovsky wrote:
> >>> On Fri, Jan 22, 2021 at 05:48:12PM -0600, Shiraz Saleem wrote:
> >>>> From: Mustafa Ismail <mustafa.ismail@intel.com>
> >>>>
> >>>> Register irdma as an auxiliary driver which can attach to auxiliary
> >>>> RDMA devices from Intel PCI netdev drivers i40e and ice. Implement
> >>>> the private channel ops, add basic devlink support in the driver and
> >>>> register net notifiers.
> >>>
> >>> Devlink part in "the RDMA client" is interesting thing.
> >>>
> >>> The idea behind auxiliary bus was that PCI logic will stay at one
> >>> place and devlink considered as the tool to manage that.
> >>
> >> Yes, this doesn't seem right, I don't think these auxiliary bus objects should have
> >> devlink instances, or at least someone from devlink land should approve of the
> >> idea.
> >>
> >
> > In our model, we have one auxdev (for RDMA) per PCI device function owned by netdev driver
> > and one devlink instance per auxdev. Plus there is an Intel netdev driver for each HW generation.
> > Moving the devlink logic to the PCI netdev driver would mean duplicating the same set of RDMA
> > params in each Intel netdev driver. Additionally, plumbing RDMA specific params in the netdev
> > driver sort of seems misplaced to me.
> >
>
> I agree that plumbing these parameters at the PCI side in the devlink of
> the parent device is weird. They don't seem to be parameters that the
> parent driver cares about.
It does, the PCI driver is not supposed to spawn any aux devices for
RDMA at all if RDMA is disabled.
For an iWarp driver I would consider ENABLE_ROCE to really be a
general ENABLE_RDMA.
Are you sure you need to implement this?
In any event, you just can't put the generic ENABLE_ROCE flag anyplace
but the PCI device for devlink, it breaks the expected user API
established by mlx5
Jason
next prev parent reply other threads:[~2021-01-26 19:43 UTC|newest]
Thread overview: 80+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-22 23:48 [PATCH 00/22] Add Intel Ethernet Protocol Driver for RDMA (irdma) Shiraz Saleem
2021-01-22 23:48 ` [PATCH 01/22] iidc: Introduce iidc.h Shiraz Saleem
2021-01-22 23:48 ` [PATCH 02/22] ice: Initialize RDMA support Shiraz Saleem
2021-01-22 23:48 ` [PATCH 03/22] ice: Implement iidc operations Shiraz Saleem
2021-01-22 23:48 ` [PATCH 04/22] ice: Register auxiliary device to provide RDMA Shiraz Saleem
2021-01-25 19:09 ` Jason Gunthorpe
2021-02-05 15:23 ` Saleem, Shiraz
2021-02-05 15:27 ` Jason Gunthorpe
2021-01-22 23:48 ` [PATCH 05/22] i40e: Prep i40e header for aux bus conversion Shiraz Saleem
2021-01-22 23:48 ` [PATCH 06/22] i40e: Register auxiliary devices to provide RDMA Shiraz Saleem
2021-01-22 23:48 ` [PATCH 07/22] RDMA/irdma: Register an auxiliary driver and implement private channel OPs Shiraz Saleem
2021-01-24 13:45 ` Leon Romanovsky
2021-01-25 13:28 ` Jason Gunthorpe
2021-01-25 20:52 ` Jakub Kicinski
2021-01-26 0:39 ` Saleem, Shiraz
2021-01-26 0:47 ` Jason Gunthorpe
2021-01-26 0:57 ` Keller, Jacob E
2021-01-26 1:01 ` Jacob Keller
2021-01-26 1:10 ` Jason Gunthorpe [this message]
2021-01-27 0:42 ` Saleem, Shiraz
2021-01-27 2:44 ` Jason Gunthorpe
2021-01-30 1:19 ` Saleem, Shiraz
2021-01-26 5:29 ` Leon Romanovsky
2021-01-26 22:07 ` Jacob Keller
2021-01-27 1:02 ` Saleem, Shiraz
2021-01-26 0:39 ` Saleem, Shiraz
2021-01-25 18:42 ` Jason Gunthorpe
2021-01-26 0:42 ` Saleem, Shiraz
2021-01-26 0:59 ` Jason Gunthorpe
2021-01-27 0:41 ` Saleem, Shiraz
2021-01-27 12:18 ` Leon Romanovsky
2021-01-27 22:17 ` Saleem, Shiraz
2021-01-27 23:16 ` Jason Gunthorpe
2021-01-28 5:41 ` Leon Romanovsky
2021-01-30 1:19 ` Saleem, Shiraz
2021-02-01 6:09 ` Leon Romanovsky
2021-02-01 19:18 ` Jason Gunthorpe
2021-02-02 0:40 ` Saleem, Shiraz
2021-02-02 1:06 ` Dan Williams
2021-02-02 17:14 ` Jason Gunthorpe
2021-02-02 19:42 ` Saleem, Shiraz
2021-02-02 23:17 ` Jason Gunthorpe
2021-02-01 19:21 ` Jason Gunthorpe
2021-01-26 5:37 ` Leon Romanovsky
2021-01-30 1:19 ` Saleem, Shiraz
2021-02-01 19:19 ` Jason Gunthorpe
2021-01-25 19:16 ` Jason Gunthorpe
2021-01-30 1:19 ` Saleem, Shiraz
2021-01-22 23:48 ` [PATCH 08/22] RDMA/irdma: Implement device initialization definitions Shiraz Saleem
2021-01-22 23:48 ` [PATCH 09/22] RDMA/irdma: Implement HW Admin Queue OPs Shiraz Saleem
2021-01-25 19:23 ` Jason Gunthorpe
2021-01-27 0:41 ` Saleem, Shiraz
2021-01-27 2:41 ` Jason Gunthorpe
2021-01-30 1:18 ` Saleem, Shiraz
2021-01-22 23:48 ` [PATCH 10/22] RDMA/irdma: Add HMC backing store setup functions Shiraz Saleem
2021-01-22 23:48 ` [PATCH 11/22] RDMA/irdma: Add privileged UDA queue implementation Shiraz Saleem
2021-01-22 23:48 ` [PATCH 12/22] RDMA/irdma: Add QoS definitions Shiraz Saleem
2021-01-22 23:48 ` [PATCH 13/22] RDMA/irdma: Add connection manager Shiraz Saleem
2021-01-22 23:48 ` [PATCH 14/22] RDMA/irdma: Add PBLE resource manager Shiraz Saleem
2021-01-22 23:48 ` [PATCH 15/22] RDMA/irdma: Implement device supported verb APIs Shiraz Saleem
2021-01-24 14:18 ` Leon Romanovsky
2021-01-27 1:04 ` Saleem, Shiraz
2021-01-22 23:48 ` [PATCH 16/22] RDMA/irdma: Add RoCEv2 UD OP support Shiraz Saleem
2021-01-22 23:48 ` [PATCH 17/22] RDMA/irdma: Add user/kernel shared libraries Shiraz Saleem
2021-01-22 23:48 ` [PATCH 18/22] RDMA/irdma: Add miscellaneous utility definitions Shiraz Saleem
2021-01-25 19:37 ` Jason Gunthorpe
2021-01-22 23:48 ` [PATCH 19/22] RDMA/irdma: Add dynamic tracing for CM Shiraz Saleem
2021-01-22 23:48 ` [PATCH 20/22] RDMA/irdma: Add ABI definitions Shiraz Saleem
2021-01-25 19:45 ` Jason Gunthorpe
2021-01-30 1:18 ` Saleem, Shiraz
2021-02-01 19:21 ` Jason Gunthorpe
2021-02-05 20:12 ` Saleem, Shiraz
2021-01-22 23:48 ` [PATCH 21/22] RDMA/irdma: Add irdma Kconfig/Makefile and remove i40iw Shiraz Saleem
2021-01-25 18:50 ` Jason Gunthorpe
2021-01-26 0:39 ` Saleem, Shiraz
2021-01-26 5:47 ` Leon Romanovsky
2021-01-22 23:48 ` [PATCH 22/22] RDMA/irdma: Update MAINTAINERS file Shiraz Saleem
2021-01-25 13:29 ` [PATCH 00/22] Add Intel Ethernet Protocol Driver for RDMA (irdma) Jason Gunthorpe
2021-01-25 18:44 ` Jason Gunthorpe
2021-01-26 0:39 ` Saleem, Shiraz
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=20210126011043.GG4147@nvidia.com \
--to=jgg@nvidia.com \
--cc=anthony.l.nguyen@intel.com \
--cc=dan.j.williams@intel.com \
--cc=davem@davemloft.net \
--cc=david.m.ertman@intel.com \
--cc=dledford@redhat.com \
--cc=gregkh@linuxfoundation.org \
--cc=jacob.e.keller@intel.com \
--cc=jiri@nvidia.com \
--cc=kuba@kernel.org \
--cc=leon@kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=mustafa.ismail@intel.com \
--cc=netdev@vger.kernel.org \
--cc=shiraz.saleem@intel.com \
--cc=sridhar.samudrala@intel.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).