mutex: Add support for wound/wait style locks
authorMaarten Lankhorst <[email protected]>
Mon, 24 Jun 2013 08:30:04 +0000 (10:30 +0200)
committerIngo Molnar <[email protected]>
Wed, 26 Jun 2013 10:10:56 +0000 (12:10 +0200)
commit040a0a37100563754bb1fee6ff6427420bcfa609
tree153e76b27ef95f57ad2055d42b8bb07cb1f8d493
parenta41b56efa70e060f650aeb54740aaf52044a1ead
mutex: Add support for wound/wait style locks

Wound/wait mutexes are used when other multiple lock
acquisitions of a similar type can be done in an arbitrary
order. The deadlock handling used here is called wait/wound in
the RDBMS literature: The older tasks waits until it can acquire
the contended lock. The younger tasks needs to back off and drop
all the locks it is currently holding, i.e. the younger task is
wounded.

For full documentation please read Documentation/ww-mutex-design.txt.

References: https://lwn.net/Articles/548909/
Signed-off-by: Maarten Lankhorst <[email protected]>
Acked-by: Daniel Vetter <[email protected]>
Acked-by: Rob Clark <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: Linus Torvalds <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
Documentation/ww-mutex-design.txt [new file with mode: 0644]
include/linux/mutex-debug.h
include/linux/mutex.h
kernel/mutex.c
lib/debug_locks.c