On Mon, Jun 17, 2019 at 05:06:22PM +0800, Qu Wenruo wrote: > And for a btrfs specific question, why we have packed attribute for > btrfs_key? > I see no specific reason to make a CPU native structure packed, not to > mention we already have btrfs_disk_key.
For that one there's no reason to use packed and we can go further and reorder the members so that the offset is right after objectid, ie. both at aligned addresses. I have a patch for that but I still need to validate that.
