Re: [RFC PATCH] security: <linux/lsm_hooks.h>: fix all kernel-doc warnings
From: Randy Dunlap
Date: Tue Feb 18 2020 - 14:30:56 EST
On 2/18/20 6:03 AM, Stephen Smalley wrote:
> On 2/16/20 2:08 AM, Randy Dunlap wrote:
>> From: Randy Dunlap <rdunlap@xxxxxxxxxxxxx>
>>
>> Fix all kernel-doc warnings in <linux/lsm_hooks.h>.
>> Fixes the following warnings:
>>
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'quotactl' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'quota_on' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'sb_free_mnt_opts' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'sb_eat_lsm_opts' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'sb_kern_mount' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'sb_show_options' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'sb_add_mnt_opt' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'd_instantiate' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'getprocattr' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'setprocattr' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'locked_down' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'perf_event_open' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'perf_event_alloc' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'perf_event_free' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'perf_event_read' not described in 'security_list_options'
>> ../include/linux/lsm_hooks.h:1830: warning: Function parameter or member 'perf_event_write' not described in 'security_list_options'
>>
>> Signed-off-by: Randy Dunlap <rdunlap@xxxxxxxxxxxxx>
>> Cc: John Johansen <john.johansen@xxxxxxxxxxxxx>
>> Cc: Kees Cook <keescook@xxxxxxxxxxxx>
>> Cc: Micah Morton <mortonm@xxxxxxxxxxxx>
>> Cc: James Morris <jmorris@xxxxxxxxx>
>> Cc: "Serge E. Hallyn" <serge@xxxxxxxxxx>
>> Cc: linux-security-module@xxxxxxxxxxxxxxx
>> Cc: Paul Moore <paul@xxxxxxxxxxxxxx>
>> Cc: Stephen Smalley <sds@xxxxxxxxxxxxx>
>> Cc: Eric Paris <eparis@xxxxxxxxxxxxxx>
>> Cc: Casey Schaufler <casey@xxxxxxxxxxxxxxxx>
>> Cc: Kentaro Takeda <takedakn@xxxxxxxxxxxxx>
>> Cc: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx>
>> ---
>> Notes:
>> a. The location for some of these might need to be modified.
>> b. 'locked_down' was just missing a final ':'.
>> c. Added a new section: Security hooks for perf events.
>>
>> Â include/linux/lsm_hooks.h |ÂÂ 36 +++++++++++++++++++++++++++++++++++-
>> Â 1 file changed, 35 insertions(+), 1 deletion(-)
>>
>> --- lnx-56-rc1.orig/include/linux/lsm_hooks.h
>> +++ lnx-56-rc1/include/linux/lsm_hooks.h
>> @@ -136,6 +140,10 @@
>> ÂÂ *ÂÂÂ @sb superblock being remounted
>> ÂÂ *ÂÂÂ @data contains the filesystem-specific data.
>> ÂÂ *ÂÂÂ Return 0 if permission is granted.
>> + * @sb_kern_mount:
>> + *ÂÂÂÂ Mount this @sb if allowed by permissions.
>> + * @sb_show_options:
>> + *ÂÂÂÂ Show (print on @m) mount options for this @sb.
>> ÂÂ * @sb_umount:
>> ÂÂ *ÂÂÂ Check permission before the @mnt file system is unmounted.
>> ÂÂ *ÂÂÂ @mnt contains the mounted file system.
>
> Thanks for doing this. Note that some of the existing kernel-doc comments for these hooks include a separate line describing each parameter (not just embedded in the function description) and a line describing the return value. Is that optional for kernel-doc? Obviously what you have added here is an improvement, just wondering whether it suffices or needs further augmentation.
Hi Stephen,
The additional kernel-doc comments that you refer to are obviously Good to Have,
but they are not required. I didn't feel comfortable or qualified to add
all of that info, but if anyone wants to help/contribute, please do so.
thanks.
--
~Randy