Hi, here's one suggestion for discussion at LSF/VM summit: It seems to me that the current writeback code has outlived it's days. The code is rather complex as various tweaks accumulated over years, filesystems workaround MAX_WRITEBACK_PAGES limitation to get better performance, inode writeback is suboptimal for filesystems doing delayed allocation (we end up writing inodes twice), sync writeback is (at least in theory) livelockable. So I think it would be worthwhile to sit down and get to some agreement on how do we improve this situation. Honza -- Jan Kara <jack@xxxxxxx> SUSE Labs, CR -- 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
