On Tue, Apr 11, 2017 at 09:44:03AM +0800, Qu Wenruo wrote: > > I get that we cannot easily avoid using the extent_changeset, so we'll > > end up one way or another, the stack conservation has slight preference. > > Yes, I understand dynamic allocation can complicate the error handler. > > But the allocation inside btrfs_qgroup_reserve_data() could reduce the > impact. And bring less compact to qgroup disabled case. > (So btrfs_qgroup_reserve_data() accepts struct extent_changeset ** other > than struct extent_changeset *) > > The code to be modified should be at the same level as current patch. > Mostly change extent_changeset_release() to extent_changeset_free(). > > > So overall the impact is not that huge, unless current error handlers > already have some problems. > > Personally speaking I'm OK either way. > > Do I need to build a temporary branch/patchset for your evaluation? Yes please. -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
