Hello, I'd like to attend this years LSF. I've been doing various file system work for the last 3 years, most of that is btrfs. To make btrfs more stable and more completely, * I build up btrfs's error handling framework which works similar to what ext4_std_error does. * I place much energy on btrfs's fsync performance and improve it to be an acceptable number. * I get snapshot-aware defrag project to work which addresses serious defragment problems among multiply snapshots. * I add nocow and compress control on per file basis to btrfs. * I have a large amount of bugfixes and cleanups. I'd like to discuss the following topic, Btrfs needs to reserve metadata space for updating its core structure Btree, sinc we cannot predict how much we'll use precisely, now btrfs has to be pessimestic and reserve metadata space as the worst case for each update. It'b be a serious problem except delalloc writes, delalloc writes needs to reserve metadata space, which can be a problem since btrfs will have to keep reserved space till end io work finishes. This can be even more worse when there are multiply writes in flight, so we have lots of space reserved to be used in writing back dirty pages and end io work, the delay time between file write and writeback is various. As it's always 'the worst case', it's inevitable to waste metadata space due to the current design. So I'd like to talk with people to try to work out some potential solutions. thanks, liubo -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
