From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932082AbbCYJNj (ORCPT ); Wed, 25 Mar 2015 05:13:39 -0400 Received: from terminus.zytor.com ([198.137.202.10]:37429 "EHLO terminus.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751849AbbCYJNf (ORCPT ); Wed, 25 Mar 2015 05:13:35 -0400 Date: Wed, 25 Mar 2015 02:13:22 -0700 From: tip-bot for Andy Lutomirski Message-ID: Cc: luto@kernel.org, linux-kernel@vger.kernel.org, tj@kernel.org, hpa@zytor.com, jkosina@suse.cz, mingo@kernel.org, dvlasenk@redhat.com, tiwai@suse.de, tglx@linutronix.de, stefan.seyfried@googlemail.com Reply-To: stefan.seyfried@googlemail.com, tglx@linutronix.de, tiwai@suse.de, dvlasenk@redhat.com, mingo@kernel.org, jkosina@suse.cz, luto@kernel.org, linux-kernel@vger.kernel.org, tj@kernel.org, hpa@zytor.com In-Reply-To: <189320d42b4d671df78c10555976bb10af1ffc75.1427137498.git.luto@kernel.org> References: <189320d42b4d671df78c10555976bb10af1ffc75.1427137498.git.luto@kernel.org> To: linux-tip-commits@vger.kernel.org Subject: [tip:x86/asm] x86/asm/entry: Check for syscall exit work with IRQs disabled Git-Commit-ID: b3494a4ab20f6bdf74cdf2badf7918bb65ee8a00 X-Mailer: tip-git-log-daemon Robot-ID: Robot-Unsubscribe: Contact to get blacklisted from these emails MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=UTF-8 Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Commit-ID: b3494a4ab20f6bdf74cdf2badf7918bb65ee8a00 Gitweb: http://git.kernel.org/tip/b3494a4ab20f6bdf74cdf2badf7918bb65ee8a00 Author: Andy Lutomirski AuthorDate: Mon, 23 Mar 2015 12:32:54 -0700 Committer: Ingo Molnar CommitDate: Tue, 24 Mar 2015 21:08:28 +0100 x86/asm/entry: Check for syscall exit work with IRQs disabled We currently have a race: if we're preempted during syscall exit, we can fail to process syscall return work that is queued up while we're preempted in ret_from_sys_call after checking ti.flags. Fix it by disabling interrupts before checking ti.flags. Reported-by: Stefan Seyfried Reported-by: Takashi Iwai Signed-off-by: Andy Lutomirski Acked-by: Denys Vlasenko Cc: Jiri Kosina Cc: Tejun Heo Fixes: 96b6352c1271 ("x86_64, entry: Remove the syscall exit audit") Link: http://lkml.kernel.org/r/189320d42b4d671df78c10555976bb10af1ffc75.1427137498.git.luto@kernel.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/entry_64.S | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 1d74d16..2babb39 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -364,12 +364,21 @@ system_call_fastpath: * Has incomplete stack frame and undefined top of stack. */ ret_from_sys_call: - testl $_TIF_ALLWORK_MASK,TI_flags+THREAD_INFO(%rsp,RIP-ARGOFFSET) - jnz int_ret_from_sys_call_fixup /* Go the the slow path */ - LOCKDEP_SYS_EXIT DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF + + /* + * We must check ti flags with interrupts (or at least preemption) + * off because we must *never* return to userspace without + * processing exit work that is enqueued if we're preempted here. + * In particular, returning to userspace with any of the one-shot + * flags (TIF_NOTIFY_RESUME, TIF_USER_RETURN_NOTIFY, etc) set is + * very bad. + */ + testl $_TIF_ALLWORK_MASK,TI_flags+THREAD_INFO(%rsp,RIP-ARGOFFSET) + jnz int_ret_from_sys_call_fixup /* Go the the slow path */ + CFI_REMEMBER_STATE /* * sysretq will re-enable interrupts: @@ -386,7 +395,7 @@ ret_from_sys_call: int_ret_from_sys_call_fixup: FIXUP_TOP_OF_STACK %r11, -ARGOFFSET - jmp int_ret_from_sys_call + jmp int_ret_from_sys_call_irqs_off /* Do syscall tracing */ tracesys: @@ -432,6 +441,7 @@ tracesys_phase2: GLOBAL(int_ret_from_sys_call) DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF +int_ret_from_sys_call_irqs_off: movl $_TIF_ALLWORK_MASK,%edi /* edi: mask to check */ GLOBAL(int_with_check)