Hello,
1/3 is a bugfix and it fixes userland (not kernel) data corruption
with vfio (and in general device driver) page pinning done with
get_user_pages. More testing of it under any load is welcome (also not
necessarily a page pinning load using vfio).
Along with the above I'm sending also 2/3 and 3/3 but those are not
meant to be merged upstream quickly and they're very low priority and
furthermore 2/3 is not zero risk and it didn't get enough testing
yet. Queuing 2/3 in -mm to give it more exposure should be ok
though. 2/3 is only suitable for merging at the very opening of merge
window anyway.
Andrea Arcangeli (3):
mm: thp: calculate the mapcount correctly for THP pages during WP
faults
mm: thp: microoptimize compound_mapcount()
mm: thp: split_huge_pmd_address() comment improvement
include/linux/mm.h | 12 +++++++--
include/linux/swap.h | 8 +++---
mm/huge_memory.c | 73 ++++++++++++++++++++++++++++++++++++++++++++--------
mm/memory.c | 22 ++++++++++------
mm/swapfile.c | 13 +++++-----
5 files changed, 98 insertions(+), 30 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/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>