On 06/28/2013 10:11 AM, Jensen wrote:
> it looks good , but it is better that if moving the following two sentence to
> before 'ocfs2_journal_dirty(handle, fe_bh);' sentence.
> 1.ocfs2_journal_dirty(handle, prev_bg_bh);
> 2.ocfs2_journal_dirty(handle, bg_bh);
Thanks for your review, but are you
it looks good , but it is better that if moving the following two sentence to
before 'ocfs2_journal_dirty(handle, fe_bh);' sentence.
1.ocfs2_journal_dirty(handle, prev_bg_bh);
2.ocfs2_journal_dirty(handle, bg_bh);
because if it fail, it maybe commit by jbd2 journal.
Thanks,
Jensen
On 2013/6/20
On 06/27/2013 06:13 AM, Andrew Morton wrote:
> On Thu, 20 Jun 2013 13:29:33 +0800 Jeff Liu wrote:
>
>> From: Jie Liu
>>
>> In ocfs2_relink_block_group(), we roll back all those changes if
>> notify intent to modify buffers for metadata update failed even
>> if the relevant buffer has not yet be
On Thu, 20 Jun 2013 13:29:33 +0800 Jeff Liu wrote:
> From: Jie Liu
>
> In ocfs2_relink_block_group(), we roll back all those changes if
> notify intent to modify buffers for metadata update failed even
> if the relevant buffer has not yet been modified/got dirty at that
> point, that are not qu
On 2013/6/20 13:29, Jeff Liu wrote:
> From: Jie Liu
>
> In ocfs2_relink_block_group(), we roll back all those changes if
> notify intent to modify buffers for metadata update failed even
> if the relevant buffer has not yet been modified/got dirty at that
> point, that are not quite right because
From: Jie Liu
In ocfs2_relink_block_group(), we roll back all those changes if
notify intent to modify buffers for metadata update failed even
if the relevant buffer has not yet been modified/got dirty at that
point, that are not quite right because of:
- None buffer has been modified/dirty if f
