Finally got the regressions in the whiteout compaction and btree locking changes fixed - those have now been merged into bcache-dev. We should hopefully be _finally_ ready to merge in foreground btree node merging, once it's had a bit more testing to very no new or exposed regressions. Besides the memcpy() bug in the whiteout compaction code, the other bug was in the new btree_iter_traverse_error() code - it wasn't reinitializing sorted_iters in a retry path, causing it to go into an infinite loop - but I don't think anyone besides me was seeing this. After btree node merging goes in, then next thing I'm working on is more performance improvements related to btree node writes and further improvements to handling of whiteouts. I'm planning on writing up a longer post explaining what's changing, but for now suffice it to say when I'm finished we ought to be beating the _pants_ off ext4/xfs/btrfs across the board. -- To unsubscribe from this list: send the line "unsubscribe linux-bcache" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
