Date:   Mon, 29 Nov 2021 09:44:06 +0100
From:   Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
To:     Nathan Chancellor <nathan@kernel.org>
Cc:     linuxarm@huawei.com, mauro.chehab@huawei.com,
        Andrey Utkin <andrey_utkin@fastmail.com>,
        Anton Sviridenko <anton@corp.bluecherry.net>,
        Bluecherry Maintainers <maintainers@bluecherrydvr.com>,
        Ismael Luceno <ismael@iodev.co.uk>,
        Mauro Carvalho Chehab <mchehab@kernel.org>,
        Nick Desaulniers <ndesaulniers@google.com>,
        linux-kernel@vger.kernel.org, linux-media@vger.kernel.org,
        llvm@lists.linux.dev
Subject: Re: [PATCH 08/20] media: solo6x10: add _maybe_unused to currently
 unused functions
Message-ID: <20211129094406.39904314@coco.lan>
In-Reply-To: <YaE8aRz1OIJ+x5P5@archlinux-ax161>
References: <cover.1637781097.git.mchehab+huawei@kernel.org>
        <18daa194878d8815beef39d6fed2a838e1c2cb68.1637781097.git.mchehab+huawei@kernel.org>
        <YaE8aRz1OIJ+x5P5@archlinux-ax161>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Precedence: bulk

Em Fri, 26 Nov 2021 12:58:33 -0700
Nathan Chancellor <nathan@kernel.org> escreveu:

> On Wed, Nov 24, 2021 at 08:13:11PM +0100, Mauro Carvalho Chehab wrote:
> > There are several unused helper macros there, meant to parse some
> > fields.
> > 
> > While there's not wrong with that, it generates clang warnings
> > with W=1, causing build to break with CONFIG_WERROR.
> > 
> > So, add __maybe_unused to fix such warnings.
> > 
> > Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>  
> 
> I'll comment on this one patch but my opinion applies for all the
> patches adding '__maybe_unused' to truly unused functions.
> 
> I agree with Laurent's comment [1]: unless this code is going to be used
> soon, it should be deleted and resurrected when it is actually needed.
> We have git for a reason and by adding this attribute, you are making it
> harder to catch and eliminate unused functions, as no compiler will
> catch them with an unused attribute (it is possible other static
> analysis tools will but I doubt those are run as frequently as compilers
> with W=1).

In this specific case (and on a few other patches on this series), those
macros are used to document a field. That's why I opted to keep it.

> 
> However, you are the maintainer so if you really want to keep these
> around, I would recommend adding '__always_unused' instead of
> '__maybe_unused' for documentation and auditing purposes, even though
> they evaluate to the same thing:
> 
> $ rg __always_unused | wc -l
> 337
> 
> $ rg __maybe_unused | wc -l
> 4335

Good point. I'll use __always_unused on such patches.

> 
> [1]: https://lore.kernel.org/r/YZtpnjPcGxVwhe61@pendragon.ideasonboard.com/
> 
> Cheers,
> Nathan
> 
> > ---
> > 
> > To avoid mailbombing on a large number of people, only mailing lists were C/C on the cover.
> > See [PATCH 00/20] at: https://lore.kernel.org/all/cover.1637781097.git.mchehab+huawei@kernel.org/
> > 
> >  drivers/media/pci/solo6x10/solo6x10-v4l2-enc.c | 12 ++++++------
> >  1 file changed, 6 insertions(+), 6 deletions(-)
> > 
> > diff --git a/drivers/media/pci/solo6x10/solo6x10-v4l2-enc.c b/drivers/media/pci/solo6x10/solo6x10-v4l2-enc.c
> > index 0abcad4e84fa..85eaf5d00e9b 100644
> > --- a/drivers/media/pci/solo6x10/solo6x10-v4l2-enc.c
> > +++ b/drivers/media/pci/solo6x10/solo6x10-v4l2-enc.c
> > @@ -391,12 +391,12 @@ static int solo_send_desc(struct solo_enc_dev *solo_enc, int skip,
> >  }
> >  
> >  /* Extract values from VOP header - VE_STATUSxx */
> > -static inline int vop_interlaced(const vop_header *vh)
> > +static inline __maybe_unused int vop_interlaced(const vop_header *vh)
> >  {
> >  	return (__le32_to_cpu((*vh)[0]) >> 30) & 1;
> >  }
> >  
> > -static inline u8 vop_channel(const vop_header *vh)
> > +static inline __maybe_unused u8 vop_channel(const vop_header *vh)
> >  {
> >  	return (__le32_to_cpu((*vh)[0]) >> 24) & 0x1F;
> >  }
> > @@ -411,12 +411,12 @@ static inline u32 vop_mpeg_size(const vop_header *vh)
> >  	return __le32_to_cpu((*vh)[0]) & 0xFFFFF;
> >  }
> >  
> > -static inline u8 vop_hsize(const vop_header *vh)
> > +static inline u8 __maybe_unused vop_hsize(const vop_header *vh)
> >  {
> >  	return (__le32_to_cpu((*vh)[1]) >> 8) & 0xFF;
> >  }
> >  
> > -static inline u8 vop_vsize(const vop_header *vh)
> > +static inline u8 __maybe_unused vop_vsize(const vop_header *vh)
> >  {
> >  	return __le32_to_cpu((*vh)[1]) & 0xFF;
> >  }
> > @@ -436,12 +436,12 @@ static inline u32 vop_jpeg_size(const vop_header *vh)
> >  	return __le32_to_cpu((*vh)[4]) & 0xFFFFF;
> >  }
> >  
> > -static inline u32 vop_sec(const vop_header *vh)
> > +static inline u32 __maybe_unused vop_sec(const vop_header *vh)
> >  {
> >  	return __le32_to_cpu((*vh)[5]);
> >  }
> >  
> > -static inline u32 vop_usec(const vop_header *vh)
> > +static inline __maybe_unused u32 vop_usec(const vop_header *vh)
> >  {
> >  	return __le32_to_cpu((*vh)[6]);
> >  }
> > -- 
> > 2.33.1
> > 
> >   


Thanks,
Mauro