Hello, It is my understanding that currently, the only difference between subvolumes and snapshots are that snapshots share a root with an existing tree, while subvolumes start off empty. There is an interesting use case difference, though: because a subvolume by definition cannot share data with other subvolumes, that frees it up to being mounted with different options (e.g. nodatasum, nodatacow). The reason I'm wondering is that I just switched my home directory to btrfs, and one of my machine is often used for rebuilding RPMs (for initial testing, before getting it built on a build server). The data is never kept for a long time, and so checksumming and COW would just slow things down. Would this be possible, and if so, is it planned for implementation by 1.0? Thanks, -- Michel Alexandre Salim -- 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
