public inbox for linux-arm-kernel@lists.infradead.org 
 help / color / mirror / Atom feed
From: arnd@arndb•de (Arnd Bergmann)
To: linux-arm-kernel@lists•infradead.org
Subject: [RFC PATCH dtc] C-based DT schema checker integrated into dtc
Date: Tue, 5 Nov 2013 09:22:01 +0100	[thread overview]
Message-ID: <201311050922.01970.arnd@arndb.de> (raw)
In-Reply-To: <527816AE.1080508@wwwdotorg.org>

On Monday 04 November 2013, Stephen Warren wrote:
> On 11/04/2013 01:43 PM, Arnd Bergmann wrote:
> > int devm_probe_gpio(struct device *dev, const struct devm_probe *probe)
> ...
> > #define DEVM_GPIO(_struct, _member, _index) {				 \
> ...
> > #define DEVM_GPIO_NAMED(_struct, _member, _name) { 			 \
> 
> Should those save the GPIO flags too? Or, do we assume that gpiod-based
> GPIOs already handle the flags inside the gpiod API?

The latter. I would certainly hope to get away with assigning just one
structure field per callback function.
 
> > /*
> >  * this lists all properties we access from the driver. The list
> >  * is interpreted by devm_probe() and can be programmatically
> >  * verified to match the binding.
> >  */
> > static const struct devm_probe foo_probe_list[] = {
> > 	DEVM_ALLOC(foo_priv),
> > 	DEVM_IOMAP(foo_priv, regs, 0, 0),
> > 	DEVM_PROP_BOOL(foo_priv, oldstyle_dma, "foo,oldstyle-dma"),
> 
> I wonder if it makes sense to handle pure data here, or whether this
> only makes sense for resources/services that are provided by some other
> device?
> 
> I guess it's nice for all resources and configuration to come through to
> probe() in the same way, but I rather wonder how we'd handle bindings
> that have large custom (driver-specific) tables of data. Would be simply
> defer such things to custom code in probe()? If so, it feels slightly
> inconsistent to handle some data in the probe_list[] and some in code in
> probe(). Still, I guess there's something to be said for keeping the
> simple cases simple.

Right, and I didn't intend to solve all cases in a completely generic way.
Thinking about it some more, a driver could actually add a custom callback
for some properties, but I don't know if that adds any value over just
interpreting them from the regular probe() function.

> > 	DEVM_DMA_SLAVE(foo_priv, rxdma, "rx");
> > 	DEVM_DMA_SLAVE(foo_priv, txdma, "tx");
> > 	DEVM_GPIO(foo_priv, gpio, 0);
> > 	DEVM_IRQ_NAMED(foo_priv, irq, foo_irq_handler, "fifo", IRQF_SHARED),
> 
> What about the case where some resources are optional, or only required
> based on the values of certain other properties? I suppose that probe()
> could call devm_probe() multiple times, with different probe_list[]s,
> based on whatever algorithm they need.

That's one open question that I haven't found the best solution for yet.
My first take on that was to add another field in struct devm_probe to
mark a property as optional, but that would increase the overall
complexity. I'd first want to do a survey of what kinds of properties
are typically optional. If it's only a few of them, we could have
something like

DEVM_DMA_SLAVE_OPTIONAL()

for the cases that need it, with a variant of the callback function.

	Arnd

  reply	other threads:[~2013-11-05  8:22 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-24 21:51 [RFC PATCH dtc] C-based DT schema checker integrated into dtc Stephen Warren
2013-10-24 23:43 ` Grant Likely
2013-10-25  4:00   ` Kumar Gala
2013-10-25 14:44   ` Stephen Warren
2013-10-25 15:21     ` Jon Loeliger
2013-10-25 17:38       ` Rob Herring
2013-10-25 23:11       ` David Gibson
2013-11-03 23:15         ` Tomasz Figa
2013-11-03 23:26           ` Tomasz Figa
2013-11-04  9:28             ` Arnd Bergmann
2013-11-04 12:31               ` Tomasz Figa
2013-11-04 16:37               ` Stephen Warren
2013-11-04 18:57                 ` Olof Johansson
2013-11-04 20:43                 ` Arnd Bergmann
2013-11-04 21:29                   ` Jason Gunthorpe
2013-11-04 21:43                     ` Stephen Warren
2013-11-04 22:21                       ` Jason Gunthorpe
2013-11-05 12:14                         ` Arnd Bergmann
2013-11-05  8:39                     ` Arnd Bergmann
2013-11-05 18:03                       ` Jason Gunthorpe
2013-11-05 18:48                         ` Arnd Bergmann
2013-11-05 19:12                           ` Jason Gunthorpe
2013-11-05 19:34                             ` Arnd Bergmann
2013-11-05 19:58                               ` Jason Gunthorpe
2013-11-05 20:17                                 ` Arnd Bergmann
2013-11-05 20:36                                   ` Jason Gunthorpe
2013-11-04 21:50                   ` Stephen Warren
2013-11-05  8:22                     ` Arnd Bergmann [this message]
2013-11-06 12:17                   ` Thierry Reding
2013-11-04 14:28           ` David Gibson
2013-11-04 16:42           ` Stephen Warren
2013-10-28 10:17     ` David Gibson
2013-10-31 21:13       ` Stephen Warren
2013-11-01 13:24         ` David Gibson
2013-10-25 23:29 ` David Gibson
2013-10-31 21:11   ` Stephen Warren
2013-11-10 11:00     ` David Gibson
2013-11-12 22:06       ` Stephen Warren
2013-11-13  0:33         ` David Gibson

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=201311050922.01970.arnd@arndb.de \
    --to=arnd@arndb$(echo .)de \
    --cc=linux-arm-kernel@lists$(echo .)infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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