From: Guennadi Liakhovetski <g.liakhovetski-Mmb7MZpHnFY@public.gmane.org>
To: Darius <augulis.darius-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Ben Dooks <ben-linux-elnMNo+KYs3YtjvyW6yDsg@public.gmane.org>
Subject: Re: [PATCH][resend] iMX/MXC support for I2C
Date: Thu, 4 Dec 2008 11:20:43 +0100 (CET) [thread overview]
Message-ID: <Pine.LNX.4.64.0812041108460.5079@axis700.grange> (raw)
In-Reply-To: <4937AA1C.8050004-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
On Thu, 4 Dec 2008, Darius wrote:
> > > > + if (!request_mem_region(res->start, res_size, res->name)) {
> > > > + dev_err(&pdev->dev, "can't allocate %d bytes at %d address\n",
> > > > + res_size, res->start);
> > > > + return -ENOMEM;
> > > > + }
> > > I was once told, one doesn't need request_mem_region() on regions from
> > > platform data resources - this is already done implicitly.
> >
> > I belive that request_mem_region() is the correct thing to do, IIRC,
> > the platform_device_register() calls insert_resource() on all the
> > resource regions of the device so that it is guarnateed to appear in
> > the /proc/ioport or /proc/iomem map. However I belive this does not
> > actually make a reservation of that resource.
> >
>
> confused... Guennadi, can you please explain more your mention?
Look at this:
http://marc.info/?l=linux-sh&m=121642007016402&w=2
> > > > + /* Get I2C clock */
> > > > + i2c_imx->clk = clk_get(NULL, "i2c_clk");
> > > There can be several i2c busses on the system, so you want:
> > >
> > > + i2c_imx->clk = clk_get(&pdev->dev, "i2c_clk");
> >
> > tbh, the bus clock should be defined as NULL, any extra clocks for
> > a device get names. However this may need fixing in the platform
> > as well. I belive RMK is already making these changes for AMBA and
> > I will probably follow suit with the s3c architectures as soon as
> > the rest of the development work is out of the way.
Ben, what do you mean here? I thought Russell was against
platform-specific clock names and for function names. But i2c_clk is a
function name. I don't understand the difference between "bus clock" and
"extra clocks" - on i.MX31 we have 3 i2c busses with separate clock
sources, so, we have to use the platform device in clk_get to request a
specific i2c clock matching the platform-device index. That's all what I
meant.
> > > > +struct imxi2c_platform_data {
> > > > + int (*init)(struct device *dev);
> > > > + int (*exit)(struct device *dev);
> > > What are you going to use .exit() for? Is it really needed? Even if it is,
> > > it can easily return void I guess?
> > >
>
> Where to put commnets? Right here in driver code or is better in
> /Documentation?
Something like this:
+/**
+ * @init: initialise the interface
+ * @exit: free interface resources
+ */
+struct imxi2c_platform_data {
+ int (*init)(struct device *dev);
+ int (*exit)(struct device *dev);
Thanks
Guennadi
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
next prev parent reply other threads:[~2008-12-04 10:20 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-12-02 14:00 [PATCH][resend] iMX/MXC support for I2C Darius
2008-12-02 14:04 ` Darius
2008-12-03 8:28 ` Jean Delvare
2008-12-03 11:17 ` Guennadi Liakhovetski
[not found] ` <Pine.LNX.4.64.0812030917440.4717-0199iw4Nj15frtckUFj5Ag@public.gmane.org>
2008-12-03 12:52 ` Darius
2008-12-03 15:19 ` Mark Brown
2008-12-03 21:19 ` Guennadi Liakhovetski
[not found] ` <Pine.LNX.4.64.0812032158150.7961-0199iw4Nj15frtckUFj5Ag@public.gmane.org>
2008-12-03 23:45 ` Ben Dooks
2008-12-04 7:47 ` Darius
[not found] ` <49378B1A.7030909-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2008-12-04 13:30 ` Mark Brown
2008-12-03 23:43 ` Ben Dooks
[not found] ` <20081203234317.GA4256-elnMNo+KYs3pIgCt6eIbzw@public.gmane.org>
2008-12-04 9:59 ` Darius
[not found] ` <4937AA1C.8050004-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2008-12-04 10:20 ` Guennadi Liakhovetski [this message]
2008-12-05 15:17 ` Guennadi Liakhovetski
[not found] ` <Pine.LNX.4.64.0812051612500.5840-0199iw4Nj15frtckUFj5Ag@public.gmane.org>
2008-12-08 9:18 ` Darius
2008-12-03 12:56 ` Guennadi Liakhovetski
[not found] ` <Pine.LNX.4.64.0812031355120.4717-0199iw4Nj15frtckUFj5Ag@public.gmane.org>
2008-12-03 13:07 ` Darius
2008-12-03 13:27 ` Wolfram Sang
2008-12-03 20:58 ` Guennadi Liakhovetski
[not found] ` <Pine.LNX.4.64.0812032155350.7961-0199iw4Nj15frtckUFj5Ag@public.gmane.org>
2008-12-03 23:48 ` Ben Dooks
[not found] ` <20081203234833.GC4256-elnMNo+KYs3pIgCt6eIbzw@public.gmane.org>
2008-12-04 0:21 ` Guennadi Liakhovetski
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=Pine.LNX.4.64.0812041108460.5079@axis700.grange \
--to=g.liakhovetski-mmb7mzphnfy@public.gmane.org \
--cc=augulis.darius-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=ben-linux-elnMNo+KYs3YtjvyW6yDsg@public.gmane.org \
--cc=linux-i2c-u79uwXL29TY76Z2rM5mHXA@public.gmane.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