Patchwork [v2,1/2] include: linux: sysfs: Add __ATTR_NAMED macro

login
register
mail settings
Submitter Himanshi Jain
Date Sept. 13, 2017, 8:44 a.m.
Message ID <2bc7a124ab925dd2c50aa8c778745d179777d50d.1505291907.git.himshijain.hj@gmail.com>
Download mbox | patch
Permalink /patch/9950677/
State New
Headers show

Comments

Himanshi Jain - Sept. 13, 2017, 8:44 a.m.
Add __ATTR_NAMED macro similar to __ATTR but taking name as a
string instead of implicit conversion of argument to string using
the macro _stringify(_name).

Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
---
 include/linux/sysfs.h | 7 +++++++
 1 file changed, 7 insertions(+)
Jonathan Cameron - Sept. 13, 2017, 5:03 p.m.
On Wed, 13 Sep 2017 14:14:07 +0530
Himanshi Jain <himshijain.hj@gmail.com> wrote:

> Add __ATTR_NAMED macro similar to __ATTR but taking name as a
> string instead of implicit conversion of argument to string using
> the macro _stringify(_name).
> 
> Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
> ---
>  include/linux/sysfs.h | 7 +++++++
>  1 file changed, 7 insertions(+)
> 
> diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
> index aa02c32..20321cf 100644
> --- a/include/linux/sysfs.h
> +++ b/include/linux/sysfs.h
> @@ -104,6 +104,13 @@ struct attribute_group {
>  	.store	= _store,						\
>  }
>  
> +#define __ATTR_NAMED(_name, _mode, _show, _store) {			\

I'm not sure about the naming here.  The normal __ATTR macro is also
'named'.  Maybe something as awful as

__ATTR_STRING_NAME ? 

Greg what do you think?

This is all to allow us to have names with operators in them without
checkpatch complaining about them... A worthwhile aim just to stop
more people wasting time trying to 'fix' those cases by adding spaces.

Jonathan

> +	.attr = {.name = _name,						\
> +		 .mode = VERIFY_OCTAL_PERMISSIONS(_mode) },		\
> +	.show = _show,							\
> +	.store = _store,						\
> +}
> +
>  #define __ATTR_PREALLOC(_name, _mode, _show, _store) {			\
>  	.attr = {.name = __stringify(_name),				\
>  		 .mode = SYSFS_PREALLOC | VERIFY_OCTAL_PERMISSIONS(_mode) },\
gregkh@linuxfoundation.org - Sept. 13, 2017, 6:58 p.m.
On Wed, Sep 13, 2017 at 06:03:10PM +0100, Jonathan Cameron wrote:
> On Wed, 13 Sep 2017 14:14:07 +0530
> Himanshi Jain <himshijain.hj@gmail.com> wrote:
> 
> > Add __ATTR_NAMED macro similar to __ATTR but taking name as a
> > string instead of implicit conversion of argument to string using
> > the macro _stringify(_name).
> > 
> > Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
> > ---
> >  include/linux/sysfs.h | 7 +++++++
> >  1 file changed, 7 insertions(+)
> > 
> > diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
> > index aa02c32..20321cf 100644
> > --- a/include/linux/sysfs.h
> > +++ b/include/linux/sysfs.h
> > @@ -104,6 +104,13 @@ struct attribute_group {
> >  	.store	= _store,						\
> >  }
> >  
> > +#define __ATTR_NAMED(_name, _mode, _show, _store) {			\
> 
> I'm not sure about the naming here.  The normal __ATTR macro is also
> 'named'.  Maybe something as awful as
> 
> __ATTR_STRING_NAME ? 
> 
> Greg what do you think?

ick ick ick.

> This is all to allow us to have names with operators in them without
> checkpatch complaining about them... A worthwhile aim just to stop
> more people wasting time trying to 'fix' those cases by adding spaces.

Yeah, but this really seems "heavy" for just a crazy sysfs name in a
macro.  Adding a whole new "core" define for that is a hard sell...

I also want to get rid of the "generic" __ATTR type macros, and force
people to use the proper _RW and friends instead.  I don't want to add
another new one that people will start to use that I later have to
change...

So no, I don't like this, how about just changing your macros instead?
No one else has this problem :)

thanks,

greg k-h
Lars-Peter Clausen - Sept. 13, 2017, 7:23 p.m.
On 09/13/2017 08:58 PM, Greg KH wrote:
> On Wed, Sep 13, 2017 at 06:03:10PM +0100, Jonathan Cameron wrote:
>> On Wed, 13 Sep 2017 14:14:07 +0530
>> Himanshi Jain <himshijain.hj@gmail.com> wrote:
>>
>>> Add __ATTR_NAMED macro similar to __ATTR but taking name as a
>>> string instead of implicit conversion of argument to string using
>>> the macro _stringify(_name).
>>>
>>> Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
>>> ---
>>>  include/linux/sysfs.h | 7 +++++++
>>>  1 file changed, 7 insertions(+)
>>>
>>> diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
>>> index aa02c32..20321cf 100644
>>> --- a/include/linux/sysfs.h
>>> +++ b/include/linux/sysfs.h
>>> @@ -104,6 +104,13 @@ struct attribute_group {
>>>  	.store	= _store,						\
>>>  }
>>>  
>>> +#define __ATTR_NAMED(_name, _mode, _show, _store) {			\
>>
>> I'm not sure about the naming here.  The normal __ATTR macro is also
>> 'named'.  Maybe something as awful as
>>
>> __ATTR_STRING_NAME ? 
>>
>> Greg what do you think?
> 
> ick ick ick.
> 
>> This is all to allow us to have names with operators in them without
>> checkpatch complaining about them... A worthwhile aim just to stop
>> more people wasting time trying to 'fix' those cases by adding spaces.
> 
> Yeah, but this really seems "heavy" for just a crazy sysfs name in a
> macro.  Adding a whole new "core" define for that is a hard sell...
> 
> I also want to get rid of the "generic" __ATTR type macros, and force
> people to use the proper _RW and friends instead.  I don't want to add
> another new one that people will start to use that I later have to
> change...
> 
> So no, I don't like this, how about just changing your macros instead?
> No one else has this problem :)

Nobody else realized they have this problem yet. E.g. there are a few users
of __ATTR in block/genhd.c that have the same issue and are likely to
generate the same false positives from static checkers.
Jonathan Cameron - Sept. 13, 2017, 8:50 p.m.
On 13 September 2017 12:23:31 GMT-07:00, Lars-Peter Clausen <lars@metafoo.de> wrote:
>On 09/13/2017 08:58 PM, Greg KH wrote:
>> On Wed, Sep 13, 2017 at 06:03:10PM +0100, Jonathan Cameron wrote:
>>> On Wed, 13 Sep 2017 14:14:07 +0530
>>> Himanshi Jain <himshijain.hj@gmail.com> wrote:
>>>
>>>> Add __ATTR_NAMED macro similar to __ATTR but taking name as a
>>>> string instead of implicit conversion of argument to string using
>>>> the macro _stringify(_name).
>>>>
>>>> Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
>>>> ---
>>>>  include/linux/sysfs.h | 7 +++++++
>>>>  1 file changed, 7 insertions(+)
>>>>
>>>> diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
>>>> index aa02c32..20321cf 100644
>>>> --- a/include/linux/sysfs.h
>>>> +++ b/include/linux/sysfs.h
>>>> @@ -104,6 +104,13 @@ struct attribute_group {
>>>>  	.store	= _store,						\
>>>>  }
>>>>  
>>>> +#define __ATTR_NAMED(_name, _mode, _show, _store) {			\
>>>
>>> I'm not sure about the naming here.  The normal __ATTR macro is also
>>> 'named'.  Maybe something as awful as
>>>
>>> __ATTR_STRING_NAME ? 
>>>
>>> Greg what do you think?
>> 
>> ick ick ick.
>> 
>>> This is all to allow us to have names with operators in them without
>>> checkpatch complaining about them... A worthwhile aim just to stop
>>> more people wasting time trying to 'fix' those cases by adding
>spaces.
>> 
>> Yeah, but this really seems "heavy" for just a crazy sysfs name in a
>> macro.  Adding a whole new "core" define for that is a hard sell...
>> 
>> I also want to get rid of the "generic" __ATTR type macros, and force
>> people to use the proper _RW and friends instead.  I don't want to
>add
>> another new one that people will start to use that I later have to
>> change...
>> 
>> So no, I don't like this, how about just changing your macros
>instead?
>> No one else has this problem :)
>
>Nobody else realized they have this problem yet. E.g. there are a few
>users
>of __ATTR in block/genhd.c that have the same issue and are likely to
>generate the same false positives from static checkers.

For IIO there is the option of moving these over to the core generated available callbacks, but
that won't work in every case and is a more major change.  I need to shift a few more drivers
over to the available callbacks and see how well it works out.  Might find time to do one in a
 gap between interesting talks this afternoon...

If I am feeling really keen I might write this missing docs I promised a while back on that stuff. Jet lag dependant...  

Jonathan
>
>--
>To unsubscribe from this list: send the line "unsubscribe linux-iio" in
>the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html
gregkh@linuxfoundation.org - Sept. 13, 2017, 9:29 p.m.
On Wed, Sep 13, 2017 at 09:23:31PM +0200, Lars-Peter Clausen wrote:
> On 09/13/2017 08:58 PM, Greg KH wrote:
> > On Wed, Sep 13, 2017 at 06:03:10PM +0100, Jonathan Cameron wrote:
> >> On Wed, 13 Sep 2017 14:14:07 +0530
> >> Himanshi Jain <himshijain.hj@gmail.com> wrote:
> >>
> >>> Add __ATTR_NAMED macro similar to __ATTR but taking name as a
> >>> string instead of implicit conversion of argument to string using
> >>> the macro _stringify(_name).
> >>>
> >>> Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
> >>> ---
> >>>  include/linux/sysfs.h | 7 +++++++
> >>>  1 file changed, 7 insertions(+)
> >>>
> >>> diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
> >>> index aa02c32..20321cf 100644
> >>> --- a/include/linux/sysfs.h
> >>> +++ b/include/linux/sysfs.h
> >>> @@ -104,6 +104,13 @@ struct attribute_group {
> >>>  	.store	= _store,						\
> >>>  }
> >>>  
> >>> +#define __ATTR_NAMED(_name, _mode, _show, _store) {			\
> >>
> >> I'm not sure about the naming here.  The normal __ATTR macro is also
> >> 'named'.  Maybe something as awful as
> >>
> >> __ATTR_STRING_NAME ? 
> >>
> >> Greg what do you think?
> > 
> > ick ick ick.
> > 
> >> This is all to allow us to have names with operators in them without
> >> checkpatch complaining about them... A worthwhile aim just to stop
> >> more people wasting time trying to 'fix' those cases by adding spaces.
> > 
> > Yeah, but this really seems "heavy" for just a crazy sysfs name in a
> > macro.  Adding a whole new "core" define for that is a hard sell...
> > 
> > I also want to get rid of the "generic" __ATTR type macros, and force
> > people to use the proper _RW and friends instead.  I don't want to add
> > another new one that people will start to use that I later have to
> > change...
> > 
> > So no, I don't like this, how about just changing your macros instead?
> > No one else has this problem :)
> 
> Nobody else realized they have this problem yet. E.g. there are a few users
> of __ATTR in block/genhd.c that have the same issue and are likely to
> generate the same false positives from static checkers.

Then fix the broken static checkers :)
Dan Carpenter - Sept. 13, 2017, 9:43 p.m.
On Wed, Sep 13, 2017 at 02:29:52PM -0700, Greg KH wrote:
> On Wed, Sep 13, 2017 at 09:23:31PM +0200, Lars-Peter Clausen wrote:
> > On 09/13/2017 08:58 PM, Greg KH wrote:
> > > On Wed, Sep 13, 2017 at 06:03:10PM +0100, Jonathan Cameron wrote:
> > >> On Wed, 13 Sep 2017 14:14:07 +0530
> > >> Himanshi Jain <himshijain.hj@gmail.com> wrote:
> > >>
> > >>> Add __ATTR_NAMED macro similar to __ATTR but taking name as a
> > >>> string instead of implicit conversion of argument to string using
> > >>> the macro _stringify(_name).
> > >>>
> > >>> Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
> > >>> ---
> > >>>  include/linux/sysfs.h | 7 +++++++
> > >>>  1 file changed, 7 insertions(+)
> > >>>
> > >>> diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
> > >>> index aa02c32..20321cf 100644
> > >>> --- a/include/linux/sysfs.h
> > >>> +++ b/include/linux/sysfs.h
> > >>> @@ -104,6 +104,13 @@ struct attribute_group {
> > >>>  	.store	= _store,						\
> > >>>  }
> > >>>  
> > >>> +#define __ATTR_NAMED(_name, _mode, _show, _store) {			\
> > >>
> > >> I'm not sure about the naming here.  The normal __ATTR macro is also
> > >> 'named'.  Maybe something as awful as
> > >>
> > >> __ATTR_STRING_NAME ? 
> > >>
> > >> Greg what do you think?
> > > 
> > > ick ick ick.
> > > 
> > >> This is all to allow us to have names with operators in them without
> > >> checkpatch complaining about them... A worthwhile aim just to stop
> > >> more people wasting time trying to 'fix' those cases by adding spaces.
> > > 
> > > Yeah, but this really seems "heavy" for just a crazy sysfs name in a
> > > macro.  Adding a whole new "core" define for that is a hard sell...
> > > 
> > > I also want to get rid of the "generic" __ATTR type macros, and force
> > > people to use the proper _RW and friends instead.  I don't want to add
> > > another new one that people will start to use that I later have to
> > > change...
> > > 
> > > So no, I don't like this, how about just changing your macros instead?
> > > No one else has this problem :)
> > 
> > Nobody else realized they have this problem yet. E.g. there are a few users
> > of __ATTR in block/genhd.c that have the same issue and are likely to
> > generate the same false positives from static checkers.
> 
> Then fix the broken static checkers :)

He was exagerating a bit to call it a "static checker" warning...
It's just checkpatch.pl complaining about adding spaces around the -
operator.  The sysfs file has a hyphen in the middle.

regards,
dan carpenter
Joe Perches - Sept. 13, 2017, 9:58 p.m.
On Thu, 2017-09-14 at 00:43 +0300, Dan Carpenter wrote:
> He was exagerating a bit to call it a "static checker" warning...

Not really.

False positives and false negatives exist in just about
every static
checker.

> It's just checkpatch.pl complaining about adding spaces around the -
> operator.

checkpatch is a brain-damaged by design static checker.
regexes can only be sensitive to patterns, not compiled code.

> The sysfs file has a hyphen in the middle.

Another option would be to use an underscore instead.
Jonathan Cameron - Sept. 13, 2017, 11:16 p.m.
On 13 September 2017 14:58:23 GMT-07:00, Joe Perches <joe@perches.com> wrote:
>On Thu, 2017-09-14 at 00:43 +0300, Dan Carpenter wrote:
>> He was exagerating a bit to call it a "static checker" warning...
>
>Not really.
>
>False positives and false negatives exist in just about
>every static
>checker.
>
>> It's just checkpatch.pl complaining about adding spaces around the -
>> operator.
>
>checkpatch is a brain-damaged by design static checker.
>regexes can only be sensitive to patterns, not compiled code.
>
>> The sysfs file has a hyphen in the middle.
>
>Another option would be to use an underscore instead.

Userspace ABI plus it really does mean subtract.  Used in description of differential channels.
Most such IIO ABI is generated by the IIO core, this just comes up with corner cases of the ABI.

Jonathan
Julia Lawall - Sept. 14, 2017, 5:10 a.m.
On Wed, 13 Sep 2017, Joe Perches wrote:

> On Thu, 2017-09-14 at 00:43 +0300, Dan Carpenter wrote:
> > He was exagerating a bit to call it a "static checker" warning...
>
> Not really.
>
> False positives and false negatives exist in just about
> every static
> checker.
>
> > It's just checkpatch.pl complaining about adding spaces around the -
> > operator.
>
> checkpatch is a brain-damaged by design static checker.
> regexes can only be sensitive to patterns, not compiled code.
>
> > The sysfs file has a hyphen in the middle.
>
> Another option would be to use an underscore instead.

There are already underscores on either side.

julia

>
> --
> You received this message because you are subscribed to the Google Groups "outreachy-kernel" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to outreachy-kernel+unsubscribe@googlegroups.com.
> To post to this group, send email to outreachy-kernel@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/outreachy-kernel/1505339903.8969.20.camel%40perches.com.
> For more options, visit https://groups.google.com/d/optout.
>
Himanshi Jain - Sept. 18, 2017, 10:49 a.m.
On Thu, Sep 14, 2017 at 2:20 AM, Jonathan Cameron
<jic23@jic23.retrosnub.co.uk> wrote:
>
>
> On 13 September 2017 12:23:31 GMT-07:00, Lars-Peter Clausen <lars@metafoo.de> wrote:
>>On 09/13/2017 08:58 PM, Greg KH wrote:
>>> On Wed, Sep 13, 2017 at 06:03:10PM +0100, Jonathan Cameron wrote:
>>>> On Wed, 13 Sep 2017 14:14:07 +0530
>>>> Himanshi Jain <himshijain.hj@gmail.com> wrote:
>>>>
>>>>> Add __ATTR_NAMED macro similar to __ATTR but taking name as a
>>>>> string instead of implicit conversion of argument to string using
>>>>> the macro _stringify(_name).
>>>>>
>>>>> Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
>>>>> ---
>>>>>  include/linux/sysfs.h | 7 +++++++
>>>>>  1 file changed, 7 insertions(+)
>>>>>
>>>>> diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
>>>>> index aa02c32..20321cf 100644
>>>>> --- a/include/linux/sysfs.h
>>>>> +++ b/include/linux/sysfs.h
>>>>> @@ -104,6 +104,13 @@ struct attribute_group {
>>>>>    .store  = _store,                                               \
>>>>>  }
>>>>>
>>>>> +#define __ATTR_NAMED(_name, _mode, _show, _store) {                       \
>>>>
>>>> I'm not sure about the naming here.  The normal __ATTR macro is also
>>>> 'named'.  Maybe something as awful as
>>>>
>>>> __ATTR_STRING_NAME ?
>>>>
>>>> Greg what do you think?
>>>
>>> ick ick ick.
>>>
>>>> This is all to allow us to have names with operators in them without
>>>> checkpatch complaining about them... A worthwhile aim just to stop
>>>> more people wasting time trying to 'fix' those cases by adding
>>spaces.
>>>
>>> Yeah, but this really seems "heavy" for just a crazy sysfs name in a
>>> macro.  Adding a whole new "core" define for that is a hard sell...
>>>
>>> I also want to get rid of the "generic" __ATTR type macros, and force
>>> people to use the proper _RW and friends instead.  I don't want to
>>add
>>> another new one that people will start to use that I later have to
>>> change...
>>>
>>> So no, I don't like this, how about just changing your macros
>>instead?
>>> No one else has this problem :)
>>
>>Nobody else realized they have this problem yet. E.g. there are a few
>>users
>>of __ATTR in block/genhd.c that have the same issue and are likely to
>>generate the same false positives from static checkers.
>
> For IIO there is the option of moving these over to the core generated available callbacks, but
> that won't work in every case and is a more major change.  I need to shift a few more drivers
> over to the available callbacks and see how well it works out.  Might find time to do one in a
>  gap between interesting talks this afternoon...

Can I help you in this? It is about exploring options as far as I can
make out, although can't really understand what options are those for
now.

Or do you want me to put comments to not to fix this checkpatch
warning as you suggested earlier?

>
> If I am feeling really keen I might write this missing docs I promised a while back on that stuff. Jet lag dependant...
>
> Jonathan
>>
>>--
>>To unsubscribe from this list: send the line "unsubscribe linux-iio" in
>>the body of a message to majordomo@vger.kernel.org
>>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
> --
> Sent from my Android device with K-9 Mail. Please excuse my brevity.
Lars-Peter Clausen - Sept. 19, 2017, 12:13 p.m.
On 09/13/2017 11:29 PM, Greg KH wrote:
> On Wed, Sep 13, 2017 at 09:23:31PM +0200, Lars-Peter Clausen wrote:
>> On 09/13/2017 08:58 PM, Greg KH wrote:
>>> On Wed, Sep 13, 2017 at 06:03:10PM +0100, Jonathan Cameron wrote:
>>>> On Wed, 13 Sep 2017 14:14:07 +0530
>>>> Himanshi Jain <himshijain.hj@gmail.com> wrote:
>>>>
>>>>> Add __ATTR_NAMED macro similar to __ATTR but taking name as a
>>>>> string instead of implicit conversion of argument to string using
>>>>> the macro _stringify(_name).
>>>>>
>>>>> Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
>>>>> ---
>>>>>  include/linux/sysfs.h | 7 +++++++
>>>>>  1 file changed, 7 insertions(+)
>>>>>
>>>>> diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
>>>>> index aa02c32..20321cf 100644
>>>>> --- a/include/linux/sysfs.h
>>>>> +++ b/include/linux/sysfs.h
>>>>> @@ -104,6 +104,13 @@ struct attribute_group {
>>>>>  	.store	= _store,						\
>>>>>  }
>>>>>  
>>>>> +#define __ATTR_NAMED(_name, _mode, _show, _store) {			\
>>>>
>>>> I'm not sure about the naming here.  The normal __ATTR macro is also
>>>> 'named'.  Maybe something as awful as
>>>>
>>>> __ATTR_STRING_NAME ? 
>>>>
>>>> Greg what do you think?
>>>
>>> ick ick ick.
>>>
>>>> This is all to allow us to have names with operators in them without
>>>> checkpatch complaining about them... A worthwhile aim just to stop
>>>> more people wasting time trying to 'fix' those cases by adding spaces.
>>>
>>> Yeah, but this really seems "heavy" for just a crazy sysfs name in a
>>> macro.  Adding a whole new "core" define for that is a hard sell...
>>>
>>> I also want to get rid of the "generic" __ATTR type macros, and force
>>> people to use the proper _RW and friends instead.  I don't want to add
>>> another new one that people will start to use that I later have to
>>> change...
>>>
>>> So no, I don't like this, how about just changing your macros instead?
>>> No one else has this problem :)
>>
>> Nobody else realized they have this problem yet. E.g. there are a few users
>> of __ATTR in block/genhd.c that have the same issue and are likely to
>> generate the same false positives from static checkers.
> 
> Then fix the broken static checkers :)

The static checkers aren't broken, the macro is. It takes a string
parameter, but instead of a string it is passed as an expression and then
transformed to string using preprocessor magic in the macro. That's not very
good semantics. And hence this gets detected as a false positive, because
nobody expects such strange behavior.
Jonathan Cameron - Sept. 21, 2017, 1:54 p.m.
On Mon, 18 Sep 2017 16:19:07 +0530
Himanshi Jain <himshijain.hj@gmail.com> wrote:

> On Thu, Sep 14, 2017 at 2:20 AM, Jonathan Cameron
> <jic23@jic23.retrosnub.co.uk> wrote:
> >
> >
> > On 13 September 2017 12:23:31 GMT-07:00, Lars-Peter Clausen <lars@metafoo.de> wrote:  
> >>On 09/13/2017 08:58 PM, Greg KH wrote:  
> >>> On Wed, Sep 13, 2017 at 06:03:10PM +0100, Jonathan Cameron wrote:  
> >>>> On Wed, 13 Sep 2017 14:14:07 +0530
> >>>> Himanshi Jain <himshijain.hj@gmail.com> wrote:
> >>>>  
> >>>>> Add __ATTR_NAMED macro similar to __ATTR but taking name as a
> >>>>> string instead of implicit conversion of argument to string using
> >>>>> the macro _stringify(_name).
> >>>>>
> >>>>> Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
> >>>>> ---
> >>>>>  include/linux/sysfs.h | 7 +++++++
> >>>>>  1 file changed, 7 insertions(+)
> >>>>>
> >>>>> diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
> >>>>> index aa02c32..20321cf 100644
> >>>>> --- a/include/linux/sysfs.h
> >>>>> +++ b/include/linux/sysfs.h
> >>>>> @@ -104,6 +104,13 @@ struct attribute_group {
> >>>>>    .store  = _store,                                               \
> >>>>>  }
> >>>>>
> >>>>> +#define __ATTR_NAMED(_name, _mode, _show, _store) {                       \  
> >>>>
> >>>> I'm not sure about the naming here.  The normal __ATTR macro is also
> >>>> 'named'.  Maybe something as awful as
> >>>>
> >>>> __ATTR_STRING_NAME ?
> >>>>
> >>>> Greg what do you think?  
> >>>
> >>> ick ick ick.
> >>>  
> >>>> This is all to allow us to have names with operators in them without
> >>>> checkpatch complaining about them... A worthwhile aim just to stop
> >>>> more people wasting time trying to 'fix' those cases by adding  
> >>spaces.  
> >>>
> >>> Yeah, but this really seems "heavy" for just a crazy sysfs name in a
> >>> macro.  Adding a whole new "core" define for that is a hard sell...
> >>>
> >>> I also want to get rid of the "generic" __ATTR type macros, and force
> >>> people to use the proper _RW and friends instead.  I don't want to  
> >>add  
> >>> another new one that people will start to use that I later have to
> >>> change...
> >>>
> >>> So no, I don't like this, how about just changing your macros  
> >>instead?  
> >>> No one else has this problem :)  
> >>
> >>Nobody else realized they have this problem yet. E.g. there are a few
> >>users
> >>of __ATTR in block/genhd.c that have the same issue and are likely to
> >>generate the same false positives from static checkers.  
> >
> > For IIO there is the option of moving these over to the core generated available callbacks, but
> > that won't work in every case and is a more major change.  I need to shift a few more drivers
> > over to the available callbacks and see how well it works out.  Might find time to do one in a
> >  gap between interesting talks this afternoon...  
> 
> Can I help you in this? It is about exploring options as far as I can
> make out, although can't really understand what options are those for
> now.

You are welcome to have a go.  However the relevant code isn't all that
well tested so I'd want to actually get enough of the setup to run to see
if it works (which is trickier).  I doubt anyone has ready access
to this hardware.

Anyhow, to see what I mean see : 

https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio.git/tree/drivers/iio/dac/dpot-dac.c

In particular dpot_dac_read_avail() and 

static const struct iio_chan_spec dpot_dac_iio_channel = {
	.type = IIO_VOLTAGE,
	.info_mask_separate = BIT(IIO_CHAN_INFO_RAW)
			    | BIT(IIO_CHAN_INFO_SCALE),
	.info_mask_separate_available = BIT(IIO_CHAN_INFO_RAW),
	.output = 1,
	.indexed = 1,
};

What this does is allow the core to create *_available attributes
(sometimes this is required by in kernel consumers of the channels as they
need to know the range of the channel for example).

The intent ultimately is to move drivers over to this interface and hopefully
get rid of the majority of remaining hand specified attrs.

It's always been well down the todo list though as it is easy to get wrong
in a given driver and we need to get more emulation in place to allow
testing of the various drivers. There are a couple of ways of doing such
emulation (i2c devices can use the i2c-stub framework, or we can do full blown
qemu emulation) - both are interesting diversions.

Also I promised to write the ABI docs (see the original patches introducing
it for some docs) but haven't actually done so yet which makes it tricky
to tell people to start using this stuff even in new drivers!

Anyhow, whilst you are welcome to look at this it might be a high risk choice
for where to get started during the outreachy applications period.

Still nothing wrong with being brave, but perhaps discuss with Alison, Daniel
or Julia.

Thanks,

Jonathan

> 
> Or do you want me to put comments to not to fix this checkpatch
> warning as you suggested earlier?
> 
> >
> > If I am feeling really keen I might write this missing docs I promised a while back on that stuff. Jet lag dependant...
> >
> > Jonathan  
> >>
> >>--
> >>To unsubscribe from this list: send the line "unsubscribe linux-iio" in
> >>the body of a message to majordomo@vger.kernel.org
> >>More majordomo info at  http://vger.kernel.org/majordomo-info.html  
> >
> > --
> > Sent from my Android device with K-9 Mail. Please excuse my brevity.  
> --
> To unsubscribe from this list: send the line "unsubscribe linux-iio" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
Julia Lawall - Sept. 21, 2017, 2:07 p.m.
On Thu, 21 Sep 2017, Jonathan Cameron wrote:

> On Mon, 18 Sep 2017 16:19:07 +0530
> Himanshi Jain <himshijain.hj@gmail.com> wrote:
>
> > On Thu, Sep 14, 2017 at 2:20 AM, Jonathan Cameron
> > <jic23@jic23.retrosnub.co.uk> wrote:
> > >
> > >
> > > On 13 September 2017 12:23:31 GMT-07:00, Lars-Peter Clausen <lars@metafoo.de> wrote:
> > >>On 09/13/2017 08:58 PM, Greg KH wrote:
> > >>> On Wed, Sep 13, 2017 at 06:03:10PM +0100, Jonathan Cameron wrote:
> > >>>> On Wed, 13 Sep 2017 14:14:07 +0530
> > >>>> Himanshi Jain <himshijain.hj@gmail.com> wrote:
> > >>>>
> > >>>>> Add __ATTR_NAMED macro similar to __ATTR but taking name as a
> > >>>>> string instead of implicit conversion of argument to string using
> > >>>>> the macro _stringify(_name).
> > >>>>>
> > >>>>> Signed-off-by: Himanshi Jain <himshijain.hj@gmail.com>
> > >>>>> ---
> > >>>>>  include/linux/sysfs.h | 7 +++++++
> > >>>>>  1 file changed, 7 insertions(+)
> > >>>>>
> > >>>>> diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
> > >>>>> index aa02c32..20321cf 100644
> > >>>>> --- a/include/linux/sysfs.h
> > >>>>> +++ b/include/linux/sysfs.h
> > >>>>> @@ -104,6 +104,13 @@ struct attribute_group {
> > >>>>>    .store  = _store,                                               \
> > >>>>>  }
> > >>>>>
> > >>>>> +#define __ATTR_NAMED(_name, _mode, _show, _store) {                       \
> > >>>>
> > >>>> I'm not sure about the naming here.  The normal __ATTR macro is also
> > >>>> 'named'.  Maybe something as awful as
> > >>>>
> > >>>> __ATTR_STRING_NAME ?
> > >>>>
> > >>>> Greg what do you think?
> > >>>
> > >>> ick ick ick.
> > >>>
> > >>>> This is all to allow us to have names with operators in them without
> > >>>> checkpatch complaining about them... A worthwhile aim just to stop
> > >>>> more people wasting time trying to 'fix' those cases by adding
> > >>spaces.
> > >>>
> > >>> Yeah, but this really seems "heavy" for just a crazy sysfs name in a
> > >>> macro.  Adding a whole new "core" define for that is a hard sell...
> > >>>
> > >>> I also want to get rid of the "generic" __ATTR type macros, and force
> > >>> people to use the proper _RW and friends instead.  I don't want to
> > >>add
> > >>> another new one that people will start to use that I later have to
> > >>> change...
> > >>>
> > >>> So no, I don't like this, how about just changing your macros
> > >>instead?
> > >>> No one else has this problem :)
> > >>
> > >>Nobody else realized they have this problem yet. E.g. there are a few
> > >>users
> > >>of __ATTR in block/genhd.c that have the same issue and are likely to
> > >>generate the same false positives from static checkers.
> > >
> > > For IIO there is the option of moving these over to the core generated available callbacks, but
> > > that won't work in every case and is a more major change.  I need to shift a few more drivers
> > > over to the available callbacks and see how well it works out.  Might find time to do one in a
> > >  gap between interesting talks this afternoon...
> >
> > Can I help you in this? It is about exploring options as far as I can
> > make out, although can't really understand what options are those for
> > now.
>
> You are welcome to have a go.  However the relevant code isn't all that
> well tested so I'd want to actually get enough of the setup to run to see
> if it works (which is trickier).  I doubt anyone has ready access
> to this hardware.
>
> Anyhow, to see what I mean see :
>
> https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio.git/tree/drivers/iio/dac/dpot-dac.c
>
> In particular dpot_dac_read_avail() and
>
> static const struct iio_chan_spec dpot_dac_iio_channel = {
> 	.type = IIO_VOLTAGE,
> 	.info_mask_separate = BIT(IIO_CHAN_INFO_RAW)
> 			    | BIT(IIO_CHAN_INFO_SCALE),
> 	.info_mask_separate_available = BIT(IIO_CHAN_INFO_RAW),
> 	.output = 1,
> 	.indexed = 1,
> };
>
> What this does is allow the core to create *_available attributes
> (sometimes this is required by in kernel consumers of the channels as they
> need to know the range of the channel for example).
>
> The intent ultimately is to move drivers over to this interface and hopefully
> get rid of the majority of remaining hand specified attrs.
>
> It's always been well down the todo list though as it is easy to get wrong
> in a given driver and we need to get more emulation in place to allow
> testing of the various drivers. There are a couple of ways of doing such
> emulation (i2c devices can use the i2c-stub framework, or we can do full blown
> qemu emulation) - both are interesting diversions.
>
> Also I promised to write the ABI docs (see the original patches introducing
> it for some docs) but haven't actually done so yet which makes it tricky
> to tell people to start using this stuff even in new drivers!
>
> Anyhow, whilst you are welcome to look at this it might be a high risk choice
> for where to get started during the outreachy applications period.
>
> Still nothing wrong with being brave, but perhaps discuss with Alison, Daniel
> or Julia.

Himanshi,

From a practical point of view, being brave would definitely be taken into
account.  But be sure that you have some accepted patches during the
application period as well.  If you spend all your time on this and it
doesn't work out, that would be too bad.

julia

>
> Thanks,
>
> Jonathan
>
> >
> > Or do you want me to put comments to not to fix this checkpatch
> > warning as you suggested earlier?
> >
> > >
> > > If I am feeling really keen I might write this missing docs I promised a while back on that stuff. Jet lag dependant...
> > >
> > > Jonathan
> > >>
> > >>--
> > >>To unsubscribe from this list: send the line "unsubscribe linux-iio" in
> > >>the body of a message to majordomo@vger.kernel.org
> > >>More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > >
> > > --
> > > Sent from my Android device with K-9 Mail. Please excuse my brevity.
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-iio" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
> --
> You received this message because you are subscribed to the Google Groups "outreachy-kernel" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to outreachy-kernel+unsubscribe@googlegroups.com.
> To post to this group, send email to outreachy-kernel@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/outreachy-kernel/20170921145442.0000160b%40huawei.com.
> For more options, visit https://groups.google.com/d/optout.
>

Patch

diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index aa02c32..20321cf 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -104,6 +104,13 @@  struct attribute_group {
 	.store	= _store,						\
 }
 
+#define __ATTR_NAMED(_name, _mode, _show, _store) {			\
+	.attr = {.name = _name,						\
+		 .mode = VERIFY_OCTAL_PERMISSIONS(_mode) },		\
+	.show = _show,							\
+	.store = _store,						\
+}
+
 #define __ATTR_PREALLOC(_name, _mode, _show, _store) {			\
 	.attr = {.name = __stringify(_name),				\
 		 .mode = SYSFS_PREALLOC | VERIFY_OCTAL_PERMISSIONS(_mode) },\