Re: [PATCH 1/1] spi: Remove unused definitions

Re: [PATCH 1/1] spi: Remove unused definitions

[Valdis.Kletnieks]

[-- Attachment #1: Type: text/plain, Size: 1932 bytes --]

(Adding Al Viro and linux-fsdevel, dropping Mark Brown and the SPI list, because this is
heading off in a different direction now)

On Wed, 20 Aug 2014 22:26:02 +0200, Pavel Machek said:
> On Wed 2014-08-06 14:27:20, Valdis.Kletnieks@vt.edu wrote:
> > On Wed, 06 Aug 2014 13:53:17 -0400, Nick Krause said:
> > > Remove unused definition which cause the following warnings
> > >
> > > drivers/spi/spi-omap-100k.c:73:0: warning: "WRITE" redefined [enabled by default]
> > > include/linux/fs.h:193:0: note: this is the location of the previous definition
> > > drivers/spi/spi-omap-100k.c:74:0: warning: "READ" redefined [enabled by default]
> > > include/linux/fs.h:192:0: note: this is the location of the previous definition
> >
> > > -#define WRITE 0
> > > -#define READ  1
> >
> > NAK.  Full stop.  These are potentially used in an inner macro someplace, and by
> > removing these, the conflicting values from fs.h will be used instead.
> >
> > #define READ                    0
> > #define WRITE                   RW_MASK
> >
> > So if there *is* a use in an inner macro, you just screwed the pooch
> > and introduced a bug in this "clean up" - somebody will be expecting to see
> > a 0 for a READ, and will receive a 1 instead.  This can't end well.
>
> Actually.. having macros called READ and WRITE in fs.h is already something I'd say
> can't end well. Can we rename those?

I had the same thought, but other than a test rename to XYZZY_READ and PLUGH_WRITE
and doing a 'make allmodconfig' and seeing what throws an error, I'm not sure how
to track down all the users.  On my fairly stripped-down .config, I have:

[/usr/src/linux-next] find * -name '.*.cmd' | wc -l
4671
[/usr/src/linux-next] find * -name '.*.cmd' | xargs grep include/linux/fs.h | wc -l
2339

Which is telling me that pretty much half the world ends up including fs.h indirectly.

Now for the mandatory bikeshedding:

What do we want to rename them to? :)

[-- Attachment #2: Type: application/pgp-signature, Size: 848 bytes --]

Re: [PATCH 1/1] spi: Remove unused definitions

[Pavel Machek]

On Wed 2014-08-20 17:12:42, Valdis.Kletnieks@vt.edu wrote:
> (Adding Al Viro and linux-fsdevel, dropping Mark Brown and the SPI list, because this is
> heading off in a different direction now)
> 
> On Wed, 20 Aug 2014 22:26:02 +0200, Pavel Machek said:
> > On Wed 2014-08-06 14:27:20, Valdis.Kletnieks@vt.edu wrote:
> > > On Wed, 06 Aug 2014 13:53:17 -0400, Nick Krause said:
> > > > Remove unused definition which cause the following warnings
> > > >
> > > > drivers/spi/spi-omap-100k.c:73:0: warning: "WRITE" redefined [enabled by default]
> > > > include/linux/fs.h:193:0: note: this is the location of the previous definition
> > > > drivers/spi/spi-omap-100k.c:74:0: warning: "READ" redefined [enabled by default]
> > > > include/linux/fs.h:192:0: note: this is the location of the previous definition
> > >
> > > > -#define WRITE 0
> > > > -#define READ  1
> > >
> > > NAK.  Full stop.  These are potentially used in an inner macro someplace, and by
> > > removing these, the conflicting values from fs.h will be used instead.
> > >
> > > #define READ                    0
> > > #define WRITE                   RW_MASK
> > >
> > > So if there *is* a use in an inner macro, you just screwed the pooch
> > > and introduced a bug in this "clean up" - somebody will be expecting to see
> > > a 0 for a READ, and will receive a 1 instead.  This can't end well.
> >
> > Actually.. having macros called READ and WRITE in fs.h is already something I'd say
> > can't end well. Can we rename those?
> 
> I had the same thought, but other than a test rename to XYZZY_READ and PLUGH_WRITE
> and doing a 'make allmodconfig' and seeing what throws an error, I'm not sure how
> to track down all the users.  On my fairly stripped-down .config, I have:
> 
> [/usr/src/linux-next] find * -name '.*.cmd' | wc -l
> 4671
> [/usr/src/linux-next] find * -name '.*.cmd' | xargs grep include/linux/fs.h | wc -l
> 2339
> 
> Which is telling me that pretty much half the world ends up including fs.h indirectly.

Yep.

I hope sh math emulator does not include fs.h.

arch/sh/math-emu/math.c:#define READ(d,a)       ({if(get_user(d, (typeof (d)*)a)

> Now for the mandatory bikeshedding:
> 
> What do we want to rename them to? :)

REQ_ prefix would fit there well, I'd say.
									Pavel


-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html