Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756715AbbBHR4C (ORCPT ); Sun, 8 Feb 2015 12:56:02 -0500 Received: from e28smtp08.in.ibm.com ([122.248.162.8]:47057 "EHLO e28smtp08.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752060AbbBHRz7 (ORCPT ); Sun, 8 Feb 2015 12:55:59 -0500 Message-ID: <54D7A3A4.1030609@linux.vnet.ibm.com> Date: Sun, 08 Feb 2015 23:27:56 +0530 From: Raghavendra K T Organization: IBM User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7 MIME-Version: 1.0 To: Sasha Levin CC: tglx@linutronix.de, mingo@redhat.com, hpa@zytor.com, peterz@infradead.org, torvalds@linux-foundation.org, konrad.wilk@oracle.com, pbonzini@redhat.com, paulmck@linux.vnet.ibm.com, waiman.long@hp.com, davej@redhat.com, oleg@redhat.com, x86@kernel.org, jeremy@goop.org, paul.gortmaker@windriver.com, ak@linux.intel.com, jasowang@redhat.com, linux-kernel@vger.kernel.org, kvm@vger.kernel.org, virtualization@lists.linux-foundation.org, xen-devel@lists.xenproject.org, riel@redhat.com, borntraeger@de.ibm.com, akpm@linux-foundation.org, a.ryabinin@samsung.com Subject: Re: [PATCH] x86 spinlock: Fix memory corruption on completing completions References: <1423234148-13886-1-git-send-email-raghavendra.kt@linux.vnet.ibm.com> <54D50E84.2060703@oracle.com> In-Reply-To: <54D50E84.2060703@oracle.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit X-TM-AS-MML: disable X-Content-Scanned: Fidelis XPS MAILER x-cbid: 15020817-0029-0000-0000-0000045FD64E Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3966 Lines: 100 On 02/07/2015 12:27 AM, Sasha Levin wrote: > On 02/06/2015 09:49 AM, Raghavendra K T wrote: >> Paravirt spinlock clears slowpath flag after doing unlock. >> As explained by Linus currently it does: >> prev = *lock; >> add_smp(&lock->tickets.head, TICKET_LOCK_INC); >> >> /* add_smp() is a full mb() */ >> >> if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG)) >> __ticket_unlock_slowpath(lock, prev); >> >> >> which is *exactly* the kind of things you cannot do with spinlocks, >> because after you've done the "add_smp()" and released the spinlock >> for the fast-path, you can't access the spinlock any more. Exactly >> because a fast-path lock might come in, and release the whole data >> structure. >> >> Linus suggested that we should not do any writes to lock after unlock(), >> and we can move slowpath clearing to fastpath lock. >> >> However it brings additional case to be handled, viz., slowpath still >> could be set when somebody does arch_trylock. Handle that too by ignoring >> slowpath flag during lock availability check. >> >> Reported-by: Sasha Levin >> Suggested-by: Linus Torvalds >> Signed-off-by: Raghavendra K T > > With this patch, my VMs lock up quickly after boot with: Tried to reproduce the hang myself, and there seems to be still a barrier (or logic I miss). Looking closely below, unlock_kick got missed though we see that SLOWPATH_FLAG is still set: /me goes back to look closely (gdb) bt #0 native_halt () at ./arch/x86/include/asm/irqflags.h:55 #1 0xffffffff81037c27 in halt () at ./arch/x86/include/asm/paravirt.h:116 #2 kvm_lock_spinning (lock=0xffff88023ffe8240, want=52504) at arch/x86/kernel/kvm.c:786 #3 0xffffffff81037251 in __raw_callee_save_kvm_lock_spinning () #4 0xffff88023fc0edb0 in ?? () #5 0x0000000000000000 in ?? () (gdb) p *(arch_spinlock_t *)0xffff88023ffe8240 $1 = {{head_tail = 3441806612, tickets = {head = 52500, tail = 52517}}} (gdb) t 2 [Switching to thread 2 (Thread 2)] #0 native_halt () at ./arch/x86/include/asm/irqflags.h:55 55 } (gdb) bt #0 native_halt () at ./arch/x86/include/asm/irqflags.h:55 #1 0xffffffff81037c27 in halt () at ./arch/x86/include/asm/paravirt.h:116 #2 kvm_lock_spinning (lock=0xffff88023ffe8240, want=52502) at arch/x86/kernel/kvm.c:786 #3 0xffffffff81037251 in __raw_callee_save_kvm_lock_spinning () #4 0x0000000000000246 in irq_stack_union () #5 0x0000000000080750 in ?? () #6 0x0000000000020000 in ?? () #7 0x0000000000000004 in irq_stack_union () #8 0x000000000000cd16 in nmi_print_seq () Cannot access memory at address 0xbfc0 (gdb) t 3 [Switching to thread 3 (Thread 3)] #0 native_halt () at ./arch/x86/include/asm/irqflags.h:55 55 } (gdb) bt #0 native_halt () at ./arch/x86/include/asm/irqflags.h:55 #1 0xffffffff81037c27 in halt () at ./arch/x86/include/asm/paravirt.h:116 #2 kvm_lock_spinning (lock=0xffff88023ffe8240, want=52512) at arch/x86/kernel/kvm.c:786 #3 0xffffffff81037251 in __raw_callee_save_kvm_lock_spinning () #4 0xffff88023fc8edb0 in ?? () #5 0x0000000000000000 in ?? () [...] //other threads with similar output (gdb) t 8 [Switching to thread 8 (Thread 8)] #0 native_halt () at ./arch/x86/include/asm/irqflags.h:55 55 } (gdb) bt #0 native_halt () at ./arch/x86/include/asm/irqflags.h:55 #1 0xffffffff81037c27 in halt () at ./arch/x86/include/asm/paravirt.h:116 #2 kvm_lock_spinning (lock=0xffff88023ffe8240, want=52500) at arch/x86/kernel/kvm.c:786 #3 0xffffffff81037251 in __raw_callee_save_kvm_lock_spinning () #4 0xffff88023fdcedb0 in ?? () #5 0x0000000000000000 in ?? () -- 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/