mce: Stop using array-index-based RCU primitives
authorPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Mon, 20 Apr 2015 01:16:02 +0000 (18:16 -0700)
committerPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Wed, 27 May 2015 19:56:16 +0000 (12:56 -0700)
commite90328b87eeff91574c18815bd6991b03bd6ecc0
tree4c5b7d82d6160ab7931fa4bbb79e371c0b909bbe
parent7d0ae8086b828311250c6afdf800b568ac9bd693
mce: Stop using array-index-based RCU primitives

Because mce is arch-specific x86 code, there is little or no
performance benefit of using rcu_dereference_index_check() over using
smp_load_acquire().  It also turns out that mce is the only place that
array-index-based RCU is used, and it would be convenient to drop
this portion of the RCU API.

This patch therefore changes rcu_dereference_index_check() uses to
smp_load_acquire(), but keeping the lockdep diagnostics, and also
changes rcu_access_index() uses to READ_ONCE().

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: linux-edac@vger.kernel.org
Cc: Tony Luck <tony.luck@intel.com>
Acked-by: Borislav Petkov <bp@suse.de>
arch/x86/kernel/cpu/mcheck/mce.c
This page took 0.027071 seconds and 5 git commands to generate.