locks/nfsd: allocate file lock outside of spinlock
authorArnd Bergmann <arnd@arndb.de>
Wed, 27 Oct 2010 13:46:08 +0000 (15:46 +0200)
committerArnd Bergmann <arnd@arndb.de>
Wed, 27 Oct 2010 19:41:50 +0000 (21:41 +0200)
commitc5b1f0d92c36851aca09ac6c7c0c4f9690ac14f3
treec8aa4ad65aea3b97292135a4c23d512e6071dc8d
parenta282a1fa6b23bd21ba0b86e53ed2a316b001836f
locks/nfsd: allocate file lock outside of spinlock

As suggested by Christoph Hellwig, this moves allocation
of new file locks out of generic_setlease into the
callers, nfs4_open_delegation and fcntl_setlease in order
to allow GFP_KERNEL allocations when lock_flocks has
become a spinlock.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: J. Bruce Fields <bfields@redhat.com>
fs/locks.c
fs/nfsd/nfs4state.c
include/linux/fs.h