From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:35595) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1b6Eft-0002mF-63 for qemu-devel@nongnu.org; Fri, 27 May 2016 06:07:29 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1b6Efn-0007Qq-Ko for qemu-devel@nongnu.org; Fri, 27 May 2016 06:07:25 -0400 Received: from mx1.redhat.com ([209.132.183.28]:57338) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1b6Efn-0007QW-FF for qemu-devel@nongnu.org; Fri, 27 May 2016 06:07:19 -0400 From: Paolo Bonzini Date: Fri, 27 May 2016 12:06:32 +0200 Message-Id: <1464343604-517-20-git-send-email-pbonzini@redhat.com> In-Reply-To: <1464343604-517-1-git-send-email-pbonzini@redhat.com> References: <1464343604-517-1-git-send-email-pbonzini@redhat.com> Subject: [Qemu-devel] [PULL 19/31] atomics: do not emit consume barrier for atomic_rcu_read List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: "Emilio G. Cota" From: "Emilio G. Cota" Currently we emit a consume-load in atomic_rcu_read. Because of limitations in current compilers, this is overkill for non-Alpha hosts and it is only useful to make Thread Sanitizer work. This patch leaves the consume-load in atomic_rcu_read when compiling with Thread Sanitizer enabled, and resorts to a relaxed load + smp_read_barrier_depends otherwise. On an RMO host architecture, such as aarch64, the performance improvement of this change is easily measurable. For instance, qht-bench performs an atomic_rcu_read on every lookup. Performance before and after applying this patch: $ tests/qht-bench -d 5 -n 1 Before: 9.78 MT/s After: 10.96 MT/s Signed-off-by: Emilio G. Cota Message-Id: <1464120374-8950-4-git-send-email-cota@braap.org> Signed-off-by: Paolo Bonzini --- include/qemu/atomic.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/include/qemu/atomic.h b/include/qemu/atomic.h index 96db6e9..7a59096 100644 --- a/include/qemu/atomic.h +++ b/include/qemu/atomic.h @@ -67,13 +67,23 @@ __atomic_store(ptr, &_val, __ATOMIC_RELAXED); \ } while(0) -/* Atomic RCU operations imply weak memory barriers */ +/* See above: most compilers currently treat consume and acquire the + * same, but this slows down atomic_rcu_read unnecessarily. + */ +#ifdef __SANITIZE_THREAD__ +#define atomic_rcu_read__nocheck(ptr, valptr) \ + __atomic_load(ptr, valptr, __ATOMIC_CONSUME); +#else +#define atomic_rcu_read__nocheck(ptr, valptr) \ + __atomic_load(ptr, valptr, __ATOMIC_RELAXED); \ + smp_read_barrier_depends(); +#endif #define atomic_rcu_read(ptr) \ ({ \ QEMU_BUILD_BUG_ON(sizeof(*ptr) > sizeof(void *)); \ typeof(*ptr) _val; \ - __atomic_load(ptr, &_val, __ATOMIC_CONSUME); \ + atomic_rcu_read__nocheck(ptr, &_val); \ _val; \ }) -- 2.5.5