Re: [PATCH] fs/nfs/inode.c: adjust code alignment

[Steve Dickson <SteveD@redhat.com>]

Hello,

On 05/08/13 10:59, Myklebust, Trond wrote:
> On Mon, 2013-08-05 at 16:47 +0200, Julia Lawall wrote:
>> From: Julia Lawall <Julia.Lawall@lip6.fr>
>>
>> Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr>
>>
>> ---
>>
>> This patch adjusts the code so that the alignment matches the current
>> semantics.  I have no idea if it is the intended semantics, though.  Should
>> the call to nfs_setsecurity also be under the else?
>>
> 
>>  fs/nfs/inode.c |    2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
>> index af6e806..d8ad685 100644
>> --- a/fs/nfs/inode.c
>> +++ b/fs/nfs/inode.c
>> @@ -463,7 +463,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh
>> *fh, struct nfs_fattr *fattr, st
>>                 unlock_new_inode(inode);
>>         } else
>>                 nfs_refresh_inode(inode, fattr);
>> -               nfs_setsecurity(inode, fattr, label);
>> +       nfs_setsecurity(inode, fattr, label);
This call to nfs_setsecurity() is not needed. The security only needs
to be set when the i-node is created... 

steved.

>>         dprintk("NFS: nfs_fhget(%s/%Ld fh_crc=0x%08x ct=%d)\n",
>>                 inode->i_sb->s_id,
>>                 (long long)NFS_FILEID(inode),
> 
> Hi Julia,
> 
> Thanks for pointing this out! Given that the 'then' clause of the if
> statement already calls nfs_setsecurity before unlocking the inode, I
> suspect that the above _should_ really be part of the 'else' clause. 
> 
> That said, I can't see that calling nfs_setsecurity twice on the inode
> can cause any unintended side-effects, so I suggest that we rather queue
> the patch up for inclusion in 3.12.
> Steve and Dave, any comments?
>