On 04/21/2013 01:57 AM, Theodore Ts'o wrote:
On Sat, Apr 20, 2013 at 11:18:17PM +0200, Bernd Schubert wrote:Alex, Andrew, did you notice the patch Ted just sent? ("ext4: mark all metadata I/O with REQ_META")This patch was sent to fix another issue that was brought up at Linux Storage, Filesystem, and MM workshop. I did bring up this issue with Mel Gorman while at LSF/MM, and as a result, tThe mm folks are going to look into making mark_page_accessed() do the right thing, or
Yeah, I know that REQ_META is a hint/flag for the IO scheduler.
perhaps provide us with new interface. The problem with forcing the page to be marked as activated is this would cause a TLB flush, which would be pointless since this these buddy bitmap pages aren't actually mapped in anywhere.
I just thought we can (mis)use that flag and and add another information to the page that it holds meta data. The mm system then could use that flag and evict those pages with a lower priority compared to other pages. I'm curious about outcome of the mm folks. Please let me know if there is anything I can help with.
Thanks, Bernd -- 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/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>