Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752522Ab3JKEgA (ORCPT ); Fri, 11 Oct 2013 00:36:00 -0400 Received: from mga02.intel.com ([134.134.136.20]:44433 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751040Ab3JKEf7 (ORCPT ); Fri, 11 Oct 2013 00:35:59 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.90,1077,1371106800"; d="scan'208";a="409224292" Date: Fri, 11 Oct 2013 12:35:53 +0800 From: Fengguang Wu To: Ingo Molnar Cc: Jakub Jelinek , Richard Henderson , Peter Zijlstra , Oleg Nesterov , Linus Torvalds , Linux Kernel Mailing List Subject: Re: [PATCH, -v3] compiler/gcc4: Add quirk for 'asm goto' miscompilation bug Message-ID: <20131011043553.GA9015@localhost> References: <20131009190231.GI13848@laptop.programming.kicks-ass.net> <20131009190851.GX30970@tucnak.zalov.cz> <20131010062238.GB9853@gmail.com> <20131010065104.GY30970@tucnak.zalov.cz> <20131010080457.GC21875@tucnak.zalov.cz> <20131010082430.GA20577@gmail.com> <20131010083107.GZ30970@tucnak.zalov.cz> <5256B3E2.4060401@twiddle.net> <20131010142703.GI30970@tucnak.zalov.cz> <20131010151219.GC5167@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20131010151219.GC5167@gmail.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 8906 Lines: 143 On Thu, Oct 10, 2013 at 05:12:19PM +0200, Ingo Molnar wrote: > > * Jakub Jelinek wrote: > > > On Thu, Oct 10, 2013 at 07:04:18AM -0700, Richard Henderson wrote: > > > On 10/10/2013 01:31 AM, Jakub Jelinek wrote: > > > > Also, for the bitops patch, you probably want an asm_volatile_goto variant. > > > > > > Why? Asm without output (which asm goto must be) are automatically volatile. > > > > You're right. > > Indeed. Something like the updated version below? > > [ I kept the asm_volatile_goto() naming to make sure the implicit volatile > is visible. ] Ingo, I applied this patch (as commit c015a7f0b746) on top of tip/sched/core 3354781a2184 "sched/numa: Reflow task_numa_group() to avoid a compiler warning" The below results show that it does fixed the many oops: /kernel/i386-randconfig-i005-1008/c015a7f0b746d0a7e18b82876bd2b693722479d2 +--------------------------------------------------------------+--------------+--------------+ | | 3354781a2184 | c015a7f0b746 | +--------------------------------------------------------------+--------------+--------------+ | WARNING:CPU:PID:at_fs/sysfs/dir.c:sysfs_add_one() | 9 | | | WARNING:CPU:PID:at_lib/kobject.c:kobject_add_internal() | 9 | | | WARNING:CPU:PID:at_drivers/base/driver.c:driver_unregister() | 9 | | | WARNING:CPU:PID:at_lib/list_debug.c:__list_add() | 9 | | | kernel_BUG_at_kernel/timer.c | 9 | | | invalid_opcode:PREEMPT_SMP | 9 | | | Kernel_panic-not_syncing:Fatal_exception_in_interrupt | 9 | | | good_boots | 0 | 9 | +--------------------------------------------------------------+--------------+--------------+ /kernel/i386-randconfig-r4-1009/c015a7f0b746d0a7e18b82876bd2b693722479d2 +--------------------------------------------------------------+--------------+--------------+ | | 3354781a2184 | c015a7f0b746 | +--------------------------------------------------------------+--------------+--------------+ | BUG:kernel_early_hang_without_any_printk_output | 2 | 3 | | WARNING:CPU:PID:at_fs/sysfs/dir.c:sysfs_add_one() | 7 | | | WARNING:CPU:PID:at_lib/kobject.c:kobject_add_internal() | 7 | | | WARNING:CPU:PID:at_drivers/base/driver.c:driver_unregister() | 7 | | | WARNING:CPU:PID:at_lib/debugobjects.c:debug_print_object() | 7 | | | INFO:trying_to_register_non-static_key | 7 | | | kernel_BUG_at_kernel/timer.c | 7 | | | invalid_opcode:SMP | 7 | | | Kernel_panic-not_syncing:Fatal_exception_in_interrupt | 7 | | | good_boots | 0 | 6 | +--------------------------------------------------------------+--------------+--------------+ /kernel/i386-randconfig-r8-1005/c015a7f0b746d0a7e18b82876bd2b693722479d2 +---------------------------------------------------------------+--------------+--------------+ | | 3354781a2184 | c015a7f0b746 | +---------------------------------------------------------------+--------------+--------------+ | BUG:unable_to_handle_kernel_NULL_pointer_dereference_at | 1 | | | Oops:DEBUG_PAGEALLOC | 9 | | | kernel_BUG_at_kernel/cred.c | 1 | | | invalid_opcode:DEBUG_PAGEALLOC | 1 | | | Kernel_panic-not_syncing:CRED:put_cred_rcu()sees_with_usage | 4 | | | BUG:unable_to_handle_kernel_paging_request_at | 3 | | | BUG:unable_to_handle_kernel_NULL_pointer_dereference_at(null) | 5 | | | good_boots | 0 | 9 | +---------------------------------------------------------------+--------------+--------------+ However it seems that it leads to or increases the chances of early boot hang: /kernel/x86_64-randconfig-a1-1009/c015a7f0b746d0a7e18b82876bd2b693722479d2 +-------------------------------------------------+--------------+--------------+ | | 3354781a2184 | c015a7f0b746 | +-------------------------------------------------+--------------+--------------+ | good_boots | 9 | 8 | | BUG:kernel_early_hang_without_any_printk_output | 0 | 1 | +-------------------------------------------------+--------------+--------------+ /kernel/x86_64-randconfig-a5-1009/c015a7f0b746d0a7e18b82876bd2b693722479d2 +-------------------------------------------------+--------------+--------------+ | | 3354781a2184 | c015a7f0b746 | +-------------------------------------------------+--------------+--------------+ | good_boots | 9 | 8 | | BUG:kernel_early_hang_without_any_printk_output | 0 | 1 | +-------------------------------------------------+--------------+--------------+ /kernel/i386-randconfig-r1-1009/c015a7f0b746d0a7e18b82876bd2b693722479d2 +-------------------------------------------------+--------------+--------------+ | | 3354781a2184 | c015a7f0b746 | +-------------------------------------------------+--------------+--------------+ | BUG:kernel_early_hang_without_any_printk_output | 9 | 9 | +-------------------------------------------------+--------------+--------------+ /kernel/i386-randconfig-r7-1009/c015a7f0b746d0a7e18b82876bd2b693722479d2 +--------------------------------------------------+--------------+--------------+ | | 3354781a2184 | c015a7f0b746 | +--------------------------------------------------+--------------+--------------+ | BUG:soft_lockup-CPU_stuck_for_s | 1 | | | INFO:task_blocked_for_more_than_seconds | 8 | 8 | | Kernel_panic-not_syncing:hung_task:blocked_tasks | 8 | 8 | | BUG:kernel_early_hang_without_any_printk_output | 0 | 1 | +--------------------------------------------------+--------------+--------------+ /kernel/i386-randconfig-r5-1009/c015a7f0b746d0a7e18b82876bd2b693722479d2 +-------------------------------------------------+--------------+--------------+ | | 3354781a2184 | c015a7f0b746 | +-------------------------------------------------+--------------+--------------+ | BUG:kernel_early_hang_without_any_printk_output | 9 | 9 | +-------------------------------------------------+--------------+--------------+ /kernel/i386-randconfig-i008-1008/c015a7f0b746d0a7e18b82876bd2b693722479d2 +-------------------------------------------------+--------------+--------------+ | | 3354781a2184 | c015a7f0b746 | +-------------------------------------------------+--------------+--------------+ | BUG:kernel_early_hang_without_any_printk_output | 9 | 9 | +-------------------------------------------------+--------------+--------------+ /kernel/i386-randconfig-i006-1008/c015a7f0b746d0a7e18b82876bd2b693722479d2 +-------------------------------------------------+--------------+--------------+ | | 3354781a2184 | c015a7f0b746 | +-------------------------------------------------+--------------+--------------+ | good_boots | 9 | 7 | | BUG:kernel_early_hang_without_any_printk_output | 0 | 2 | +-------------------------------------------------+--------------+--------------+ Thanks, Fengguang -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/