Add forward declaration of struct seq_file before using it in
function prototype.

Fixes: 9cc46516ddf4 ("userns: Add a knob to disable setgroups on a per user namespace basis")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

diff --git a/include/linux/user_namespace.h b/include/linux/user_namespace.h
index a0bb6d012137..5c7aa940b1aa 100644
--- a/include/linux/user_namespace.h
+++ b/include/linux/user_namespace.h
@@ -16,6 +16,8 @@
 #define UID_GID_MAP_MAX_BASE_EXTENTS 5
 #define UID_GID_MAP_MAX_EXTENTS 340
 
+struct seq_file;
+
 struct uid_gid_extent {
 	u32 first;
 	u32 lower_first;
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Herbert Xu <herbert@gondor.apana.org.au> writes:

> Add forward declaration of struct seq_file before using it in
> function prototype.

I don't see anything wrong with doing that.

However:
> Fixes: 9cc46516ddf4 ("userns: Add a knob to disable setgroups on a per
> user namespace basis")

That description seems to be lacking something.  The code as has been
fine since 2014 and there is no conditional compilation involved.

Did some other change cause a declaration of seq_file not to be present?
Did user_namespace.h get included somewhere new and that new location
doesn't wind up declaring seq_file?

Especially since fixes tags tend to get code back ported I would
appreciate a bit more context on why this is a problem in 2025 when
it hasn't been a problem for the last decade.

Thank you,
Eric Biederman

> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
>
> diff --git a/include/linux/user_namespace.h b/include/linux/user_namespace.h
> index a0bb6d012137..5c7aa940b1aa 100644
> --- a/include/linux/user_namespace.h
> +++ b/include/linux/user_namespace.h
> @@ -16,6 +16,8 @@
>  #define UID_GID_MAP_MAX_BASE_EXTENTS 5
>  #define UID_GID_MAP_MAX_EXTENTS 340
>  
> +struct seq_file;
> +
>  struct uid_gid_extent {
>  	u32 first;
>  	u32 lower_first;

On Fri, May 30, 2025 at 08:27:14AM -0500, Eric W. Biederman wrote:
>
> That description seems to be lacking something.  The code as has been
> fine since 2014 and there is no conditional compilation involved.

I'm happy to drop the fixes header.  There is no need to backport
this.

> Did some other change cause a declaration of seq_file not to be present?
> Did user_namespace.h get included somewhere new and that new location
> doesn't wind up declaring seq_file?

The seq_file forward declaration is currently being provided by
linux/sched.h.  I tried removing it and found this issue.

Cheers,
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Herbert Xu <herbert@gondor.apana.org.au> writes:

> On Fri, May 30, 2025 at 08:27:14AM -0500, Eric W. Biederman wrote:
>>
>> That description seems to be lacking something.  The code as has been
>> fine since 2014 and there is no conditional compilation involved.
>
> I'm happy to drop the fixes header.  There is no need to backport
> this.

The fixes tag isn't wrong.

>> Did some other change cause a declaration of seq_file not to be present?
>> Did user_namespace.h get included somewhere new and that new location
>> doesn't wind up declaring seq_file?
>
> The seq_file forward declaration is currently being provided by
> linux/sched.h.  I tried removing it and found this issue.

Thanks.

That is the interesting information that I would appreciate to see
in the change log.

If someone has to play code archaeologist and has to look at the
history, this is exactly the information that is needed to understand
why the change was made.

Eric