Hello,
A little bit of everything really, with the last two patches being
the most interesting ones.
Patches 1-3 cleanup a bit of ww mutex code.
Patch 4 isolates osq code.
Patch 5 uses the brand new READ/ASSIGN_ONCE primitives.
Patch 6 is a performance patch and gets rid of barrier calls when
polling for the (osq) lock.
More details obviously in the individual patches. Applies on today's -tip.
Please consider for 3.20, thanks!
Changes since v1 (https://lkml.org/lkml/2014/12/28/22):
o drop patch 1 (pickd up by paulmck)
o drop patch 2 (no can do, thanks a lot gcc)
o improve changelogs.
Davidlohr Bueso (6):
locking/mutex: Checking the stamp is ww only
locking/mutex: Move mcs related comments to proper location
locking/mutex: Introduce ww_mutex_set_context_slowpath
locking/mcs: Better differentiate between mcs variants
locking: Use [READ,ASSIGN]_ONCE() for non-scalar types
locking/osq: No need for load/acquire when acquire-polling
include/linux/osq_lock.h | 12 ++-
kernel/Kconfig.locks | 4 +
kernel/locking/Makefile | 3 +-
kernel/locking/mcs_spinlock.c | 208 ------------------------------------------
kernel/locking/mcs_spinlock.h | 22 +----
kernel/locking/mutex.c | 66 +++++++-------
kernel/locking/osq_lock.c | 203 +++++++++++++++++++++++++++++++++++++++++
kernel/locking/rwsem-xadd.c | 4 +-
8 files changed, 258 insertions(+), 264 deletions(-)
delete mode 100644 kernel/locking/mcs_spinlock.c
create mode 100644 kernel/locking/osq_lock.c
--
2.1.2
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
