Re: [PATCH] fs: dlm: Protect IPV6 field access by CONFIG_IPV6

[Alexander Aring]

Hi,

On Thu, Nov 18, 2021 at 3:26 AM Geert Uytterhoeven <geert@xxxxxxxxxxxxxx> wrote:
>
> If CONFIG_IPV6=n:
>
>     In file included from fs/dlm/lowcomms.c:46:
>     fs/dlm/lowcomms.c: In function ‘lowcomms_error_report’:
>     ./include/net/sock.h:386:34: error: ‘struct sock_common’ has no member named ‘skc_v6_daddr’; did you mean ‘skc_daddr’?
>       386 | #define sk_v6_daddr  __sk_common.skc_v6_daddr
>           |                                  ^~~~~~~~~~~~
>     ./include/linux/printk.h:422:19: note: in expansion of macro ‘sk_v6_daddr’
>       422 |   _p_func(_fmt, ##__VA_ARGS__);    \
>           |                   ^~~~~~~~~~~
>     ./include/linux/printk.h:450:26: note: in expansion of macro ‘printk_index_wrap’
>       450 | #define printk(fmt, ...) printk_index_wrap(_printk, fmt, ##__VA_ARGS__)
>           |                          ^~~~~~~~~~~~~~~~~
>     ./include/linux/printk.h:644:3: note: in expansion of macro ‘printk’
>       644 |   printk(fmt, ##__VA_ARGS__);    \
>           |   ^~~~~~
>     fs/dlm/lowcomms.c:612:3: note: in expansion of macro ‘printk_ratelimited’
>       612 |   printk_ratelimited(KERN_ERR "dlm: node %d: socket error "
>           |   ^~~~~~~~~~~~~~~~~~
>
> Fix this by protecting the code that accesses IPV6-only fields by a
> check for CONFIG_IPV6.
>
> Reported-by: Randy Dunlap <rdunlap@xxxxxxxxxxxxx>
> Fixes: 4c3d90570bcc2b33 ("fs: dlm: don't call kernel_getpeername() in error_report()")
> Signed-off-by: Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>
> ---

Thanks, but the issue has already been fixed in the same way [0].

- Alex

[0] https://git.kernel.org/pub/scm/linux/kernel/git/teigland/linux-dlm.git/commit/?h=next&id=1b9beda83e27a0c2cd75d1cb743c297c7b36c844