summaryrefslogtreecommitdiff
path: root/include/linux/rwsem-spinlock.h
diff options
context:
space:
mode:
authorKirill Tkhai <ktkhai@virtuozzo.com>2017-06-19 21:02:12 +0300
committerIngo Molnar <mingo@kernel.org>2017-08-10 13:28:55 +0300
commit0aa1125fa8bc5e5f98317156728fa4d0293561a5 (patch)
tree800ddf02b2c633b9021db2c5742a781abb6ea903 /include/linux/rwsem-spinlock.h
parent50972fe78f24f1cd0b9d7bbf1f87d2be9e4f412e (diff)
downloadlinux-0aa1125fa8bc5e5f98317156728fa4d0293561a5.tar.xz
locking/rwsem-spinlock: Add killable versions of __down_read()
Rename __down_read() in __down_read_common() and teach it to abort waiting in case of pending signals and killable state argument passed. Note, that we shouldn't wake anybody up in EINTR path, as: We check for signal_pending_state() after (!waiter.task) test and under spinlock. So, current task wasn't able to be woken up. It may be in two cases: a writer is owner of the sem, or a writer is a first waiter of the sem. If a writer is owner of the sem, no one else may work with it in parallel. It will wake somebody, when it call up_write() or downgrade_write(). If a writer is the first waiter, it will be woken up, when the last active reader releases the sem, and sem->count became 0. Also note, that set_current_state() may be moved down to schedule() (after !waiter.task check), as all assignments in this type of semaphore (including wake_up), occur under spinlock, so we can't miss anything. Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: arnd@arndb.de Cc: avagin@virtuozzo.com Cc: davem@davemloft.net Cc: fenghua.yu@intel.com Cc: gorcunov@virtuozzo.com Cc: heiko.carstens@de.ibm.com Cc: hpa@zytor.com Cc: ink@jurassic.park.msu.ru Cc: mattst88@gmail.com Cc: rth@twiddle.net Cc: schwidefsky@de.ibm.com Cc: tony.luck@intel.com Link: http://lkml.kernel.org/r/149789533283.9059.9829416940494747182.stgit@localhost.localdomain Signed-off-by: Ingo Molnar <mingo@kernel.org>
Diffstat (limited to 'include/linux/rwsem-spinlock.h')
-rw-r--r--include/linux/rwsem-spinlock.h1
1 files changed, 1 insertions, 0 deletions
diff --git a/include/linux/rwsem-spinlock.h b/include/linux/rwsem-spinlock.h
index ae0528b834cd..e784761a4443 100644
--- a/include/linux/rwsem-spinlock.h
+++ b/include/linux/rwsem-spinlock.h
@@ -32,6 +32,7 @@ struct rw_semaphore {
#define RWSEM_UNLOCKED_VALUE 0x00000000
extern void __down_read(struct rw_semaphore *sem);
+extern int __must_check __down_read_killable(struct rw_semaphore *sem);
extern int __down_read_trylock(struct rw_semaphore *sem);
extern void __down_write(struct rw_semaphore *sem);
extern int __must_check __down_write_killable(struct rw_semaphore *sem);