Re: [PATCH 2/2] Btrfs: use helper for logical resolve

[Liu Bo <bo.li.liu@oracle.com>]

On 08/16/2012 01:10 PM, Jan Schmidt wrote:
> On Thu, August 16, 2012 at 06:42 (+0200), Liu Bo wrote:
>> We already have a helper, iterate_inodes_from_logical(), for logical resolve,
>> so just use it.
>>
>> Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
>> ---
>>  fs/btrfs/ioctl.c |   20 +++-----------------
>>  1 files changed, 3 insertions(+), 17 deletions(-)
>>
>> diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
>> index 405b279..f1ff21a 100644
>> --- a/fs/btrfs/ioctl.c
>> +++ b/fs/btrfs/ioctl.c
[...]
>> -	extent_item_pos = loi->logical - key.objectid;
>> -	ret = iterate_extent_inodes(root->fs_info, key.objectid,
>> -					extent_item_pos, 0, build_ino_list,
>> -					inodes);
>> -
>> +	path->search_commit_root = 0;
> 
> I came across that search_commit_root as well when re-reading
> iterate_inodes_from_logical. Wouldn't explicitly set the default value here,
> though. We don't do it before each and every call to btrfs_search_slot, either.
> 

ok, will remove the redundant default set.

>> +	ret = iterate_inodes_from_logical(loi->logical, root->fs_info, path,
>> +					  build_ino_list, inodes);
> 
> To maintain the current behavior, we should patch -EINVAL to -ENOENT here.
> 

oops, I should have noticed it.

Thanks for the comments, Jan :)

thanks,
liubo