[PATCH v1 1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break

[PATCH v1 1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break

[Andy Shevchenko]

Clang is not happy about set but (in some cases) unused variable:

fs/nfsd/export.c:1027:17: error: variable 'inode' set but not used [-Werror,-Wunused-but-set-variable]

since it's used as a parameter to dprintk() which might be configured
a no-op. To avoid uglifying code with the specific ifdeffery just mark
the variable __maybe_unused.

The commit [1], which introduced this behaviour, is quite old and hence
the Fixes tag points to the first of Git era.

Link: https://git.kernel.org/pub/scm/linux/kernel/git/history/history.git/commit/?id=0431923fb7a1 [1]
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 fs/nfsd/export.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
index 9d55512d0cc9..2a1499f2ad19 100644
--- a/fs/nfsd/export.c
+++ b/fs/nfsd/export.c
@@ -1024,7 +1024,7 @@ exp_rootfh(struct net *net, struct auth_domain *clp, char *name,
 {
 	struct svc_export	*exp;
 	struct path		path;
-	struct inode		*inode;
+	struct inode		*inode __maybe_unused;
 	struct svc_fh		fh;
 	int			err;
 	struct nfsd_net		*nn = net_generic(net, nfsd_net_id);
-- 
2.50.1

Re: [PATCH v1 1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break

[Chuck Lever]

On 11/13/25 3:31 AM, Andy Shevchenko wrote:
> Clang is not happy about set but (in some cases) unused variable:
> 
> fs/nfsd/export.c:1027:17: error: variable 'inode' set but not used [-Werror,-Wunused-but-set-variable]
> 
> since it's used as a parameter to dprintk() which might be configured
> a no-op. To avoid uglifying code with the specific ifdeffery just mark
> the variable __maybe_unused.
> 
> The commit [1], which introduced this behaviour, is quite old and hence
> the Fixes tag points to the first of Git era.
> 
> Link: https://git.kernel.org/pub/scm/linux/kernel/git/history/history.git/commit/?id=0431923fb7a1 [1]
> Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  fs/nfsd/export.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c
> index 9d55512d0cc9..2a1499f2ad19 100644
> --- a/fs/nfsd/export.c
> +++ b/fs/nfsd/export.c
> @@ -1024,7 +1024,7 @@ exp_rootfh(struct net *net, struct auth_domain *clp, char *name,
>  {
>  	struct svc_export	*exp;
>  	struct path		path;
> -	struct inode		*inode;
> +	struct inode		*inode __maybe_unused;

We typically use "#ifdef CONFIG_SUNRPC_DEBUG" instead for this purpose.
I don't think that would be terribly out of place.


>  	struct svc_fh		fh;
>  	int			err;
>  	struct nfsd_net		*nn = net_generic(net, nfsd_net_id);


-- 
Chuck Lever

Re: [PATCH v1 1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break

[Chuck Lever]

From: Chuck Lever <chuck.lever@oracle.com>

On Thu, 13 Nov 2025 09:31:31 +0100, Andy Shevchenko wrote:
> Clang is not happy about set but (in some cases) unused variable:
> 
> fs/nfsd/export.c:1027:17: error: variable 'inode' set but not used [-Werror,-Wunused-but-set-variable]
> 
> since it's used as a parameter to dprintk() which might be configured
> a no-op. To avoid uglifying code with the specific ifdeffery just mark
> the variable __maybe_unused.
> 
> [...]

Applied to nfsd-testing, thanks!

[1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break
      commit: 56e9f88b25abf08de6f2b1bfbbb2ddc4e6622d1e

--
Chuck Lever

Re: [PATCH v1 1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break

[Andy Shevchenko]

On Mon, Nov 17, 2025 at 08:49:29AM -0500, Chuck Lever wrote:
> On Thu, 13 Nov 2025 09:31:31 +0100, Andy Shevchenko wrote:
> > Clang is not happy about set but (in some cases) unused variable:
> > 
> > fs/nfsd/export.c:1027:17: error: variable 'inode' set but not used [-Werror,-Wunused-but-set-variable]
> > 
> > since it's used as a parameter to dprintk() which might be configured
> > a no-op. To avoid uglifying code with the specific ifdeffery just mark
> > the variable __maybe_unused.

[...]

> Applied to nfsd-testing, thanks!
> 
> [1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break
>       commit: 56e9f88b25abf08de6f2b1bfbbb2ddc4e6622d1e

Thanks, but still no appearance in Linux Next and problem seems to be present.

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v1 1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break

[Chuck Lever]

On 11/27/25 2:50 AM, Andy Shevchenko wrote:
> On Mon, Nov 17, 2025 at 08:49:29AM -0500, Chuck Lever wrote:
>> On Thu, 13 Nov 2025 09:31:31 +0100, Andy Shevchenko wrote:
>>> Clang is not happy about set but (in some cases) unused variable:
>>>
>>> fs/nfsd/export.c:1027:17: error: variable 'inode' set but not used [-Werror,-Wunused-but-set-variable]
>>>
>>> since it's used as a parameter to dprintk() which might be configured
>>> a no-op. To avoid uglifying code with the specific ifdeffery just mark
>>> the variable __maybe_unused.
> 
> [...]
> 
>> Applied to nfsd-testing, thanks!
>>
>> [1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break
>>       commit: 56e9f88b25abf08de6f2b1bfbbb2ddc4e6622d1e
> 
> Thanks, but still no appearance in Linux Next and problem seems to be present.
> 

The usual practice is to keep patches in nfsd-testing for four
weeks to allow NFSD and community CI processes to work, and to
enable extended review before it is merged. Both the community
CI processes (eg, zero-day bots) and the availability of
reviewers are not something I have control over.

It will be available for upstream merge after December 11. You
seem to be suggesting there is a sense of urgency so I will
direct it towards v6.20-rc as soon as it is merge-ready.


-- 
Chuck Lever

Re: [PATCH v1 1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break

[Andy Shevchenko]

On Thu, Nov 27, 2025 at 11:20:16AM -0500, Chuck Lever wrote:
> On 11/27/25 2:50 AM, Andy Shevchenko wrote:
> > On Mon, Nov 17, 2025 at 08:49:29AM -0500, Chuck Lever wrote:
> >> On Thu, 13 Nov 2025 09:31:31 +0100, Andy Shevchenko wrote:
> >>> Clang is not happy about set but (in some cases) unused variable:
> >>>
> >>> fs/nfsd/export.c:1027:17: error: variable 'inode' set but not used [-Werror,-Wunused-but-set-variable]
> >>>
> >>> since it's used as a parameter to dprintk() which might be configured
> >>> a no-op. To avoid uglifying code with the specific ifdeffery just mark
> >>> the variable __maybe_unused.

[...]

> >> Applied to nfsd-testing, thanks!
> >>
> >> [1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break
> >>       commit: 56e9f88b25abf08de6f2b1bfbbb2ddc4e6622d1e
> > 
> > Thanks, but still no appearance in Linux Next and problem seems to be present.
> > 
> 
> The usual practice is to keep patches in nfsd-testing for four
> weeks to allow NFSD and community CI processes to work, and to
> enable extended review before it is merged. Both the community
> CI processes (eg, zero-day bots) and the availability of
> reviewers are not something I have control over.
> 
> It will be available for upstream merge after December 11. You
> seem to be suggesting there is a sense of urgency so I will
> direct it towards v6.20-rc as soon as it is merge-ready.

Since it's (not so critical TBH, but still) a build breakage I supposed this to
go via the respective -fixes path. But okay, your call.

So far I will keep a patch locally to remember to annoy you :-) if it isn't applied

TL;DR: Thanks for the explanation.

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v1 1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break

[Chuck Lever]

On 11/27/25 11:55 AM, Andy Shevchenko wrote:
> On Thu, Nov 27, 2025 at 11:20:16AM -0500, Chuck Lever wrote:
>> On 11/27/25 2:50 AM, Andy Shevchenko wrote:
>>> On Mon, Nov 17, 2025 at 08:49:29AM -0500, Chuck Lever wrote:
>>>> On Thu, 13 Nov 2025 09:31:31 +0100, Andy Shevchenko wrote:
>>>>> Clang is not happy about set but (in some cases) unused variable:
>>>>>
>>>>> fs/nfsd/export.c:1027:17: error: variable 'inode' set but not used [-Werror,-Wunused-but-set-variable]
>>>>>
>>>>> since it's used as a parameter to dprintk() which might be configured
>>>>> a no-op. To avoid uglifying code with the specific ifdeffery just mark
>>>>> the variable __maybe_unused.
> 
> [...]
> 
>>>> Applied to nfsd-testing, thanks!
>>>>
>>>> [1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break
>>>>       commit: 56e9f88b25abf08de6f2b1bfbbb2ddc4e6622d1e
>>>
>>> Thanks, but still no appearance in Linux Next and problem seems to be present.
>>>
>>
>> The usual practice is to keep patches in nfsd-testing for four
>> weeks to allow NFSD and community CI processes to work, and to
>> enable extended review before it is merged. Both the community
>> CI processes (eg, zero-day bots) and the availability of
>> reviewers are not something I have control over.
>>
>> It will be available for upstream merge after December 11. You
>> seem to be suggesting there is a sense of urgency so I will
>> direct it towards v6.20-rc as soon as it is merge-ready.

Oops:

s/v6.20-rc/v6.19-rc/


> Since it's (not so critical TBH, but still) a build breakage I supposed this to
> go via the respective -fixes path.

Yes, what I meant above was I will submit it just after the
v6.19 merge window closes in a few weeks.


> But okay, your call.

It's just a build warning, but I know such issues affect the
Fedora and Red Hat kernel build pipelines, as they enable the
"warning => error" compile option.

However, those distributions enable SunRPC debugging, which
means they won't see it. So I think this problem is not likely
to be pervasive.


-- 
Chuck Lever

Re: [PATCH v1 1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break

[Andy Shevchenko]

On Thu, Nov 27, 2025 at 12:08:09PM -0500, Chuck Lever wrote:
> On 11/27/25 11:55 AM, Andy Shevchenko wrote:
> > On Thu, Nov 27, 2025 at 11:20:16AM -0500, Chuck Lever wrote:
> >> On 11/27/25 2:50 AM, Andy Shevchenko wrote:
> >>> On Mon, Nov 17, 2025 at 08:49:29AM -0500, Chuck Lever wrote:
> >>>> On Thu, 13 Nov 2025 09:31:31 +0100, Andy Shevchenko wrote:
> >>>>> Clang is not happy about set but (in some cases) unused variable:
> >>>>>
> >>>>> fs/nfsd/export.c:1027:17: error: variable 'inode' set but not used [-Werror,-Wunused-but-set-variable]
> >>>>>
> >>>>> since it's used as a parameter to dprintk() which might be configured
> >>>>> a no-op. To avoid uglifying code with the specific ifdeffery just mark
> >>>>> the variable __maybe_unused.

[...]

> >>>> Applied to nfsd-testing, thanks!
> >>>>
> >>>> [1/1] nfsd: Mark variable __maybe_unused to avoid W=1 build break
> >>>>       commit: 56e9f88b25abf08de6f2b1bfbbb2ddc4e6622d1e
> >>>
> >>> Thanks, but still no appearance in Linux Next and problem seems to be present.
> >>>
> >>
> >> The usual practice is to keep patches in nfsd-testing for four
> >> weeks to allow NFSD and community CI processes to work, and to
> >> enable extended review before it is merged. Both the community
> >> CI processes (eg, zero-day bots) and the availability of
> >> reviewers are not something I have control over.
> >>
> >> It will be available for upstream merge after December 11. You
> >> seem to be suggesting there is a sense of urgency so I will
> >> direct it towards v6.20-rc as soon as it is merge-ready.
> 
> Oops:
> 
> s/v6.20-rc/v6.19-rc/
> 
> 
> > Since it's (not so critical TBH, but still) a build breakage I supposed this to
> > go via the respective -fixes path.
> 
> Yes, what I meant above was I will submit it just after the
> v6.19 merge window closes in a few weeks.

Ah, that's wonderful, thanks!

> > But okay, your call.
> 
> It's just a build warning, but I know such issues affect the
> Fedora and Red Hat kernel build pipelines, as they enable the
> "warning => error" compile option.
> 
> However, those distributions enable SunRPC debugging, which
> means they won't see it. So I think this problem is not likely
> to be pervasive.

-- 
With Best Regards,
Andy Shevchenko