From: Scott Wood <scottwood@freescale.com>
To: Yoder Stuart-B08248 <B08248@freescale.com>
Cc: Wood Scott-B07421 <B07421@freescale.com>,
Kim Phillips <kim.phillips@linaro.org>,
Christoffer Dall <christoffer.dall@linaro.org>,
Alex Williamson <alex.williamson@redhat.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"a.motakis@virtualopensystems.com"
<a.motakis@virtualopensystems.com>,
"agraf@suse.de" <agraf@suse.de>,
Sethi Varun-B16395 <B16395@freescale.com>,
Bhushan Bharat-R65777 <R65777@freescale.com>,
"peter.maydell@linaro.org" <peter.maydell@linaro.org>,
"santosh.shukla@linaro.org" <santosh.shukla@linaro.org>,
"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
"gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>
Subject: Re: RFC: (re-)binding the VFIO platform driver to a platform device
Date: Wed, 9 Oct 2013 14:21:47 -0500 [thread overview]
Message-ID: <1381346507.7979.334.camel@snotra.buserror.net> (raw)
In-Reply-To: <9F6FE96B71CF29479FF1CDC8046E15036DDA62@039-SN1MPN1-002.039d.mgd.msft.net>
On Wed, 2013-10-09 at 14:02 -0500, Yoder Stuart-B08248 wrote:
> Have been thinking about this issue some more. As Scott mentioned,
> 'wildcard' matching for a driver can be fairly done in the platform
> bus driver. We could add a new flag to the platform driver struct:
>
> diff --git a/drivers/base/platform.c b/drivers/base/platform.c
> index 4f8bef3..4d6cf14 100644
> --- a/drivers/base/platform.c
> +++ b/drivers/base/platform.c
> @@ -727,6 +727,10 @@ static int platform_match(struct device *dev, struct device_driver *drv)
> struct platform_device *pdev = to_platform_device(dev);
> struct platform_driver *pdrv = to_platform_driver(drv);
>
> + /* the driver matches any device */
> + if (pdrv->match_any)
> + return 1;
> +
> /* Attempt an OF style match first */
> if (of_driver_match_device(dev, drv))
> return 1;
>
> However, the more problematic issue is that a bus driver has no way to
> differentiate from an explicit bind request via sysfs and a bind that
> happened through bus probing.
Again, I think the wildcard match should be orthogonal to "don't bind by
default" as far as the mechanism goes.
There's already a "bool suppress_bind_attrs" to prevent sysfs
bind/unbind. I suggested a similar flag to mean the oppsosite -- bind
*only* through sysfs. Greg KH was skeptical and wanted to see a patch
before any further discussion.
> diff --git a/drivers/base/base.h b/drivers/base/base.h
> index 2cbc677..7a15ef3 100644
> --- a/drivers/base/base.h
> +++ b/drivers/base/base.h
> @@ -114,9 +114,9 @@ extern void driver_detach(struct device_driver *drv);
> extern int driver_probe_device(struct device_driver *drv, struct device *dev);
> extern void driver_deferred_probe_del(struct device *dev);
> static inline int driver_match_device(struct device_driver *drv,
> - struct device *dev)
> + struct device *dev, int explicit_bind)
> {
> - return drv->bus->match ? drv->bus->match(dev, drv) : 1;
> + return drv->bus->match ? drv->bus->match(dev, drv, explicit_bind) : 1;
> }
>
> Of, course the above change would need to be propagated to the different
> bus drivers that implement the 'match' function.
...which would not be a problem with my approach, because you could
handle it in the callers of driver_match_device().
-Scott
next prev parent reply other threads:[~2013-10-09 19:23 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-01 18:38 Kim Phillips
2013-10-01 19:15 ` Scott Wood
2013-10-01 19:17 ` Scott Wood
2013-10-01 22:01 ` Kim Phillips
2013-10-01 21:59 ` Kim Phillips
2013-10-01 22:44 ` Scott Wood
2013-10-01 20:00 ` Greg Kroah-Hartman
2013-10-01 22:02 ` Kim Phillips
2013-10-02 1:53 ` Christoffer Dall
2013-10-02 2:35 ` Alex Williamson
2013-10-02 15:14 ` Christoffer Dall
2013-10-02 15:29 ` Alex Williamson
2013-10-02 18:25 ` Yoder Stuart-B08248
2013-10-02 18:32 ` Scott Wood
2013-10-02 18:43 ` Christoffer Dall
2013-10-02 20:04 ` Kim Phillips
2013-10-02 20:13 ` Christoffer Dall
2013-10-02 20:19 ` Scott Wood
2013-10-02 20:14 ` Scott Wood
2013-10-02 20:27 ` Christoffer Dall
2013-10-02 20:39 ` gregkh
2013-10-02 20:44 ` Christoffer Dall
2013-10-02 20:37 ` gregkh
2013-10-02 20:42 ` Christoffer Dall
2013-10-02 21:08 ` Scott Wood
2013-10-02 21:16 ` gregkh
2013-10-02 21:35 ` Scott Wood
2013-10-02 23:40 ` gregkh
2013-10-03 18:33 ` Scott Wood
2013-10-03 18:54 ` gregkh
2013-10-03 19:11 ` Scott Wood
2013-10-03 20:32 ` gregkh
2013-10-09 19:02 ` Yoder Stuart-B08248
2013-10-09 19:16 ` gregkh
2013-10-09 19:49 ` Scott Wood
2013-10-09 19:21 ` Scott Wood [this message]
2013-10-09 19:44 ` Yoder Stuart-B08248
2013-10-09 20:03 ` Scott Wood
2013-10-10 3:05 ` Kim Phillips
2013-10-10 8:01 ` Bhushan Bharat-R65777
2013-10-10 15:27 ` Scott Wood
2013-10-11 6:27 ` [PATCH 1/4] driver core: Add new device_driver flag to allow binding via sysfs only Kim Phillips
2013-10-11 6:27 ` [PATCH 2/4] driver core: platform: allow platform drivers to bind to any device Kim Phillips
2013-10-11 6:27 ` [PATCH 3/4] VFIO: pci: amend vfio-pci for explicit binding via sysfs only Kim Phillips
2013-10-11 20:43 ` Scott Wood
2013-10-11 23:17 ` Kim Phillips
2013-10-14 13:01 ` Yoder Stuart-B08248
2013-10-14 17:13 ` Scott Wood
2013-10-24 11:32 ` Bhushan Bharat-R65777
2013-10-28 17:47 ` Alex Williamson
2013-10-28 18:00 ` Scott Wood
2013-10-28 18:09 ` Scott Wood
2013-10-29 3:38 ` Bhushan Bharat-R65777
2013-10-29 3:40 ` Scott Wood
2013-10-29 3:52 ` Bhushan Bharat-R65777
2013-10-29 4:29 ` Scott Wood
2013-10-29 4:31 ` Bhushan Bharat-R65777
2013-10-29 4:35 ` Scott Wood
2013-10-29 4:45 ` Bhushan Bharat-R65777
2013-10-29 4:54 ` Scott Wood
2013-10-29 6:39 ` Bhushan Bharat-R65777
2013-10-11 6:27 ` [PATCH] VFIO: platform: allow the driver to bind to any device explicitly via sysfs Kim Phillips
2013-10-10 7:45 ` RFC: (re-)binding the VFIO platform driver to a platform device Bhushan Bharat-R65777
2013-10-10 13:43 ` Yoder Stuart-B08248
2013-10-10 15:23 ` Scott Wood
2013-10-10 15:25 ` Bhushan Bharat-R65777
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=1381346507.7979.334.camel@snotra.buserror.net \
--to=scottwood@freescale.com \
--cc=B07421@freescale.com \
--cc=B08248@freescale.com \
--cc=B16395@freescale.com \
--cc=R65777@freescale.com \
--cc=a.motakis@virtualopensystems.com \
--cc=agraf@suse.de \
--cc=alex.williamson@redhat.com \
--cc=christoffer.dall@linaro.org \
--cc=gregkh@linuxfoundation.org \
--cc=kim.phillips@linaro.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=peter.maydell@linaro.org \
--cc=santosh.shukla@linaro.org \
/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
Powered by JetHome