-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 5/27/14, 12:10 PM, Chris Mason wrote:
> On 05/26/2014 09:35 PM, Jeff Mahoney wrote:
>> We are currently allocating space_info objects in an array when
>> we allocate space_info. When a user does something like:
>>
>> # btrfs balance start -mconvert
On 05/26/2014 09:35 PM, Jeff Mahoney wrote:
> We are currently allocating space_info objects in an array when we
> allocate space_info. When a user does something like:
>
> # btrfs balance start -mconvert=raid1 -dconvert=raid1 /mnt
> # btrfs balance start -mconvert=single -dconvert=single /mnt -f
On Tue, May 27, 2014 at 08:14:11AM -0400, Jeff Mahoney wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> On 5/27/14, 7:26 AM, David Sterba wrote:
> > A different lockdep chain, from test btrfs/014:
>
> Are you sure this one was with the new patch? We shouldn't be holding
> groups_sem a
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 5/27/14, 7:26 AM, David Sterba wrote:
> A different lockdep chain, from test btrfs/014:
Are you sure this one was with the new patch? We shouldn't be holding
groups_sem anymore in btrfs_remove_block_group while performing the
kobject ops.
- -Jeff
A different lockdep chain, from test btrfs/014:
[ 229.850151] ==
[ 229.850788] [ INFO: possible circular locking dependency detected ]
[ 229.850788] 3.15.0-rc7-default+ #146 Tainted: GW
[ 229.850788] --
We are currently allocating space_info objects in an array when we
allocate space_info. When a user does something like:
# btrfs balance start -mconvert=raid1 -dconvert=raid1 /mnt
# btrfs balance start -mconvert=single -dconvert=single /mnt -f
# btrfs balance start -mconvert=raid1 -dconvert=raid1
