linux-kernel - Re: [PATCH 1/6] staging: media: intel-ipu3: replace bit shifts with BIT() macro

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <YHQnOJ2t0+q65La2@kroah.com>
Date:   Mon, 12 Apr 2021 12:55:52 +0200
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Sakari Ailus <sakari.ailus@...ux.intel.com>
Cc:     Mitali Borkar <mitaliborkar810@...il.com>, bingbu.cao@...el.com,
        tian.shu.qiu@...el.com, mchehab@...nel.org,
        linux-media@...r.kernel.org, linux-staging@...ts.linux.dev,
        linux-kernel@...r.kernel.org, outreachy-kernel@...glegroups.com,
        mitali_s@...iitr.ac.in, laurent.pinchart@...asonboard.com
Subject: Re: [PATCH 1/6] staging: media: intel-ipu3: replace bit shifts with
 BIT() macro

On Mon, Apr 12, 2021 at 01:44:35PM +0300, Sakari Ailus wrote:
> Hi Greg,
> 
> On Mon, Apr 12, 2021 at 11:49:43AM +0200, Greg KH wrote:
> > On Mon, Apr 12, 2021 at 12:42:30PM +0300, Sakari Ailus wrote:
> > > Hi Mitali,
> > > 
> > > On Mon, Apr 12, 2021 at 04:38:39AM +0530, Mitali Borkar wrote:
> > > > Added #include <linux/bitops.h> and replaced bit shifts by BIT() macro.
> > > > This BIT() macro from linux/bitops.h is used to define IPU3_UAPI_GRID_Y_START_EN
> > > > and IPU3_UAPI_AWB_RGBS_THR_B_* bitmask.
> > > > Use of macro is better and neater. It maintains consistency.
> > > > Reported by checkpatch.
> > > > 
> > > > Signed-off-by: Mitali Borkar <mitaliborkar810@...il.com>
> > > > ---
> > > >  drivers/staging/media/ipu3/include/intel-ipu3.h | 7 ++++---
> > > >  1 file changed, 4 insertions(+), 3 deletions(-)
> > > > 
> > > > diff --git a/drivers/staging/media/ipu3/include/intel-ipu3.h b/drivers/staging/media/ipu3/include/intel-ipu3.h
> > > > index edd8edda0647..589d5ccee3a7 100644
> > > > --- a/drivers/staging/media/ipu3/include/intel-ipu3.h
> > > > +++ b/drivers/staging/media/ipu3/include/intel-ipu3.h
> > > > @@ -5,6 +5,7 @@
> > > >  #define __IPU3_UAPI_H
> > > >  
> > > >  #include <linux/types.h>
> > > > +#include <linux/bitops.h>
> > > >  
> > > >  /* from /drivers/staging/media/ipu3/include/videodev2.h */
> > > >  
> > > > @@ -22,11 +23,11 @@
> > > >  #define IPU3_UAPI_MAX_BUBBLE_SIZE			10
> > > >  
> > > >  #define IPU3_UAPI_GRID_START_MASK			((1 << 12) - 1)
> > > > -#define IPU3_UAPI_GRID_Y_START_EN			(1 << 15)
> > > > +#define IPU3_UAPI_GRID_Y_START_EN			BIT(15)
> > > 
> > > This header is used in user space where you don't have the BIT() macro.
> > 
> > If that is true, why is it not in a "uapi" subdir within this driver?
> > 
> > Otherwise it is not obvious at all that this is the case :(
> 
> It defines an interface towards the user space and the argument has been a
> staging driver shouldn't be doing that (for the lack of ABI stability),
> hence leaving it where it is currently.

I think we are talking past each other here...

If you have a userspace api, then put that in a .h file that has a
"uapi" in the path.  Just because you have this in a staging driver does
not mean you should not have such a thing, otherwise you are going to
constantly fight against people sending you patches like this as there
is no obvious way to determine this.

So how about moving this file to:
	drivers/staging/media/ipu3/include/uapi/intel-ipu3.h

thanks,

greg k-h