Re: [PATCH v2] Btrfs: don't update atime on RO subvolumes

[Liu Bo <liubo2009@cn.fujitsu.com>]

On 06/15/2012 05:18 PM, Alexander Block wrote:

> On Fri, Jun 15, 2012 at 10:56 AM, Liu Bo <liubo2009@cn.fujitsu.com> wrote:
>>> diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
>>> index f6ab6f5..4d0ceed 100644
>>> --- a/fs/btrfs/inode.c
>>> +++ b/fs/btrfs/inode.c
>>> @@ -4478,6 +4478,11 @@ int btrfs_dirty_inode(struct inode *inode)
>>>  static int btrfs_update_time(struct inode *inode, struct timespec *now,
>>>                            int flags)
>>>  {
>>> +     struct btrfs_root *root = BTRFS_I(inode)->root;
>>> +
>>> +     if (btrfs_root_readonly(root))
>>> +             return -EROFS;
>>> +
>>
>> It also needs to acquire root->fs_info->subvol_sem, doesn't it?
>>
>> thanks,
>> liubo
>>
> 
> Normally yes I think. But does it matter at this point? If the flags are
> modified at the same time as we access them, wouldn't it be still
> random which value (old/new) we get, even with subvol_sem?
> 
> (sorry for the double mail Liu...as always forgot Reply All)
> 


I find other callers use btrfs_root_readonly without subvol_sem either,
anyway, I'm ok with this patch.

thanks,
liubo