This is a resend of version 3, rebased to v3.2-rc2. In addition to my own tests - results in 3/5 - Wu Fengguang also ran tests of his own in combination with the IO-less dirty throttling series, the results of which can be found here: http://article.gmane.org/gmane.comp.file-systems.ext4/28795 http://article.gmane.org/gmane.linux.kernel.mm/69648 Per-zone dirty limits try to distribute page cache pages allocated for writing across zones in proportion to the individual zone sizes, to reduce the likelihood of reclaim having to write back individual pages from the LRU lists in order to make progress. Please consider merging into 3.3. fs/btrfs/file.c | 2 +- include/linux/gfp.h | 4 +- include/linux/mmzone.h | 6 + include/linux/swap.h | 1 + include/linux/writeback.h | 1 + mm/filemap.c | 5 +- mm/page-writeback.c | 290 +++++++++++++++++++++++++++++---------------- mm/page_alloc.c | 48 ++++++++ 8 files changed, 251 insertions(+), 106 deletions(-) -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>