From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752711AbaLCWcJ (ORCPT ); Wed, 3 Dec 2014 17:32:09 -0500 Received: from e06smtp11.uk.ibm.com ([195.75.94.107]:39624 "EHLO e06smtp11.uk.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751898AbaLCWa2 (ORCPT ); Wed, 3 Dec 2014 17:30:28 -0500 From: Christian Borntraeger To: linux-kernel@vger.kernel.org Cc: linux-arch@vger.kernel.org, paulmck@linux.vnet.ibm.com, torvalds@linux-foundation.org, Christian Borntraeger Subject: [PATCH 3/9] x86/spinlock: Replace ACCESS_ONCE with READ_ONCE Date: Wed, 3 Dec 2014 23:30:15 +0100 Message-Id: <1417645821-54731-4-git-send-email-borntraeger@de.ibm.com> X-Mailer: git-send-email 1.9.3 In-Reply-To: <1417645821-54731-1-git-send-email-borntraeger@de.ibm.com> References: <1417645821-54731-1-git-send-email-borntraeger@de.ibm.com> X-TM-AS-MML: disable X-Content-Scanned: Fidelis XPS MAILER x-cbid: 14120322-0005-0000-0000-0000025484E1 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org ACCESS_ONCE does not work reliably on non-scalar types. For example gcc 4.6 and 4.7 might remove the volatile tag for such accesses during the SRA (scalar replacement of aggregates) step (https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58145) Change the spinlock code to replace ACCESS_ONCE with READ_ONCE. Signed-off-by: Christian Borntraeger --- arch/x86/include/asm/spinlock.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index 9295016..12a69b4 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -92,7 +92,7 @@ static __always_inline void arch_spin_lock(arch_spinlock_t *lock) unsigned count = SPIN_THRESHOLD; do { - if (ACCESS_ONCE(lock->tickets.head) == inc.tail) + if (READ_ONCE(lock->tickets.head) == inc.tail) goto out; cpu_relax(); } while (--count); @@ -105,7 +105,7 @@ static __always_inline int arch_spin_trylock(arch_spinlock_t *lock) { arch_spinlock_t old, new; - old.tickets = ACCESS_ONCE(lock->tickets); + old.tickets = READ_ONCE(lock->tickets); if (old.tickets.head != (old.tickets.tail & ~TICKET_SLOWPATH_FLAG)) return 0; @@ -162,14 +162,14 @@ static __always_inline void arch_spin_unlock(arch_spinlock_t *lock) static inline int arch_spin_is_locked(arch_spinlock_t *lock) { - struct __raw_tickets tmp = ACCESS_ONCE(lock->tickets); + struct __raw_tickets tmp = READ_ONCE(lock->tickets); return tmp.tail != tmp.head; } static inline int arch_spin_is_contended(arch_spinlock_t *lock) { - struct __raw_tickets tmp = ACCESS_ONCE(lock->tickets); + struct __raw_tickets tmp = READ_ONCE(lock->tickets); return (__ticket_t)(tmp.tail - tmp.head) > TICKET_LOCK_INC; } -- 1.9.3