From: Jason Gunthorpe <jgg@ziepe.ca>
To: "Wan, Kaike" <kaike.wan@intel.com>
Cc: "Dalessandro, Dennis" <dennis.dalessandro@intel.com>,
"dledford@redhat.com" <dledford@redhat.com>,
"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
"Marciniszyn, Mike" <mike.marciniszyn@intel.com>
Subject: Re: [PATCH for-next 3/3] IB/hfi1: Use the ibdev in hfi1_devdata as the parent of cdev
Date: Fri, 20 Mar 2020 14:33:43 -0300 [thread overview]
Message-ID: <20200320173343.GU20941@ziepe.ca> (raw)
In-Reply-To: <MW3PR11MB46657936C2E8286A3CB2234CF4F50@MW3PR11MB4665.namprd11.prod.outlook.com>
On Fri, Mar 20, 2020 at 05:30:40PM +0000, Wan, Kaike wrote:
> > > If this is not desirable, we could keep the current approach to create
> > > the struct device dynamically through device_create(). In that case,
> > > all we need to do is to clean up the code. Which one do you prefer?
> >
> > The issue here was parentage. There should not be a virtual device involved.
> >
> > The hfi1 user_class device should be parented to the ib_device, look at how
> > things like umad work to do this properly.
> So all we need to do is:
> -- Change user_device from struct device * to struct device in hfi1_devdata;
> -- Set up dd->user_device properly including setting its parent to ib_device;
> -- call cdev_device_all().
Yes, but keep in mind that putting multiple krefs inside the same
structure is very tricky - be sure to do it right.
Jason
next prev parent reply other threads:[~2020-03-20 17:33 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-16 21:04 [PATCH for-next 0/3] Clean up and improvements for 5.7 Dennis Dalessandro
2020-03-16 21:04 ` [PATCH for-next 1/3] IB/rdmavt: Delete unused routine Dennis Dalessandro
2020-03-16 21:05 ` [PATCH for-next 2/3] IB/hfi1: Remove kobj from hfi1_devdata Dennis Dalessandro
2020-03-16 21:05 ` [PATCH for-next 3/3] IB/hfi1: Use the ibdev in hfi1_devdata as the parent of cdev Dennis Dalessandro
2020-03-18 13:31 ` Jason Gunthorpe
2020-03-18 16:02 ` Wan, Kaike
2020-03-18 16:34 ` Jason Gunthorpe
2020-03-18 17:13 ` Wan, Kaike
2020-03-18 23:18 ` Jason Gunthorpe
2020-03-20 12:19 ` Wan, Kaike
2020-03-20 16:09 ` Wan, Kaike
2020-03-20 16:32 ` Jason Gunthorpe
2020-03-20 17:30 ` Wan, Kaike
2020-03-20 17:33 ` Jason Gunthorpe [this message]
2020-03-18 23:28 ` [PATCH for-next 0/3] Clean up and improvements for 5.7 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=20200320173343.GU20941@ziepe.ca \
--to=jgg@ziepe.ca \
--cc=dennis.dalessandro@intel.com \
--cc=dledford@redhat.com \
--cc=kaike.wan@intel.com \
--cc=linux-rdma@vger.kernel.org \
--cc=mike.marciniszyn@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).