Implement a proper rw_tryupgrade

Current rw_tryupgrade does rw_exit and then rw_tryenter(RW_RWITER), and then
does rw_enter(RW_READER) if it fails. This violate the assumption that
rw_tryupgrade should be atomic and could cause extra contention or even lock
inversion.

This patch we implement a proper rw_tryupgrade. For rwsem-spinlock, we take
the spinlock to check rwsem->count and rwsem->wait_list. For normal rwsem, we
use cmpxchg on rwsem->count to change the value from single reader to single
writer.

Signed-off-by: Chunwei Chen <david.chen@osnexus.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Tim Chase <tim@chase2k.com>
Closes zfsonlinux/zfs#4692
Closes #554
This commit is contained in:
Chunwei Chen
2016-05-25 16:35:42 -07:00
committed by Brian Behlendorf
parent c60a51b640
commit f58040c0fc
5 changed files with 147 additions and 13 deletions
+25
View File
@@ -39,6 +39,7 @@ AC_DEFUN([SPL_AC_CONFIG_KERNEL], [
SPL_AC_2ARGS_ZLIB_DEFLATE_WORKSPACESIZE
SPL_AC_SHRINK_CONTROL_STRUCT
SPL_AC_RWSEM_SPINLOCK_IS_RAW
SPL_AC_RWSEM_ACTIVITY
SPL_AC_SCHED_RT_HEADER
SPL_AC_2ARGS_VFS_GETATTR
SPL_AC_USLEEP_RANGE
@@ -1316,6 +1317,30 @@ AC_DEFUN([SPL_AC_RWSEM_SPINLOCK_IS_RAW], [
EXTRA_KCFLAGS="$tmp_flags"
])
dnl #
dnl # 3.16 API Change
dnl #
dnl # rwsem-spinlock "->activity" changed to "->count"
dnl #
AC_DEFUN([SPL_AC_RWSEM_ACTIVITY], [
AC_MSG_CHECKING([whether struct rw_semaphore has member activity])
tmp_flags="$EXTRA_KCFLAGS"
EXTRA_KCFLAGS="-Werror"
SPL_LINUX_TRY_COMPILE([
#include <linux/rwsem.h>
],[
struct rw_semaphore dummy_semaphore __attribute__ ((unused));
dummy_semaphore.activity = 0;
],[
AC_MSG_RESULT(yes)
AC_DEFINE(HAVE_RWSEM_ACTIVITY, 1,
[struct rw_semaphore has member activity])
],[
AC_MSG_RESULT(no)
])
EXTRA_KCFLAGS="$tmp_flags"
])
dnl #
dnl # 3.9 API change,
dnl # Moved things from linux/sched.h to linux/sched/rt.h