lkml.org 
[lkml]   [2012]   [Sep]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH v4 1/7] mfd: add syscon driver based on regmap
On Fri, Aug 31, 2012 at 09:26:29AM +0800, Shawn Guo wrote:
> On Wed, Aug 29, 2012 at 06:56:52PM +0800, Dong Aisheng wrote:
> > +config MFD_SYSCON
> > + bool "System Controller Register R/W Based on Regmap"
>
> If the driver only compiles and works for device tree probe, we should
> have the following?
>
> depends on OF
>
Correct.
For currently it only supports dt.

> > + select REGMAP_MMIO
> > + help
> > + Select this option to enable accessing system control registers
> > + via regmap.
> > +
>
> <snip>
>
> > +struct regmap *syscon_node_to_regmap(struct device_node *np)
> > +{
> > + struct syscon *syscon;
> > + struct device *dev;
> > +
> > + dev = driver_find_device(&syscon_driver.driver, NULL, np,
> > + syscon_match);
> > + of_node_put(np);
>
> This looks a little unnatural to me. Function syscon_node_to_regmap
> becomes an API visible to clients, who might never know that np will
> be put inside the API. I'm saying the client may also call of_node_put
> to put the node they get.
We probably could add a comment here for the API to avoid this happen.

>
> I think of_node_put should be moved out from here and put into
> syscon_node_to_regmap and syscon_regmap_lookup_by_compatible.
>
I guess no, if you want to move of_node_put into syscon_regmap_lookup_by_phandle,
then syscon_regmap_lookup_by_phandle has the same issue.
Actually i had considered your concern when writing this API...
The original purpose of doing like that is saving some duplicated 'of_node_put'
and make the API easy to use.
I searched the kernel dt code and found it existed some similar cases.
e.g: of_irq_find_parent, of_get_next_parent
So it looks to me that it may be usually to do like that for the cases that
the conversion from a node to other thing since the client may only care
about the things converted. For our case, it's regmap.
So i can't think it make too much sense for all client driver have to write
duplicated and meaningless 'of_node_put' code.

Regards
Dong Aisheng

> > +
> > + if (!dev)
> > + return ERR_PTR(-EPROBE_DEFER);
> > +
> > + syscon = dev_get_drvdata(dev);
> > +
> > + return syscon->regmap;
> > +}
> > +EXPORT_SYMBOL_GPL(syscon_node_to_regmap);
> > +
> > +struct regmap *syscon_regmap_lookup_by_compatible(const char *s)
> > +{
> > + struct device_node *syscon_np;
> > +
> > + syscon_np = of_find_compatible_node(NULL, NULL, s);
> > + if (!syscon_np)
> > + return ERR_PTR(-ENODEV);
> > +
> > + return syscon_node_to_regmap(syscon_np);
> > +}
> > +EXPORT_SYMBOL_GPL(syscon_regmap_lookup_by_compatible);
> > +
> > +struct regmap *syscon_regmap_lookup_by_phandle(struct device_node *np,
> > + const char *property)
> > +{
> > + struct device_node *syscon_np;
> > +
> > + syscon_np = of_parse_phandle(np, property, 0);
> > + if (!syscon_np)
> > + return ERR_PTR(-ENODEV);
> > +
> > + return syscon_node_to_regmap(syscon_np);
> > +}
> > +EXPORT_SYMBOL_GPL(syscon_regmap_lookup_by_phandle);



\
 
 \ /
  Last update: 2012-09-03 05:21    [W:0.112 / U:0.160 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site