From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756562AbcIUPuH (ORCPT ); Wed, 21 Sep 2016 11:50:07 -0400 Received: from mail-lf0-f53.google.com ([209.85.215.53]:32815 "EHLO mail-lf0-f53.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755699AbcIUPuE (ORCPT ); Wed, 21 Sep 2016 11:50:04 -0400 From: Roman Pen Cc: Roman Pen , Andy Lutomirski , Josh Poimboeuf , Borislav Petkov , Brian Gerst , Denys Vlasenko , "H . Peter Anvin" , Peter Zijlstra , Thomas Gleixner , Ingo Molnar , Tejun Heo , x86@kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH 1/2] x86/dumpstack: on oops do not rewind stack for kthread Date: Wed, 21 Sep 2016 17:43:49 +0200 Message-Id: <20160921154350.13128-1-roman.penyaev@profitbricks.com> X-Mailer: git-send-email 2.9.3 To: unlisted-recipients:; (no To-header on input) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org kthread uses stack and keeps completion structure on it to be woken up on vfork_done completion. In commit 2deb4be28 Andy Lutomirski rewinds the stack unconditionally and further completion of task->vfork_done for any kthread leads to stack corruption (or infinite spin on attempt to spin lock on garbage memory). Signed-off-by: Roman Pen Cc: Andy Lutomirski Cc: Josh Poimboeuf Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Tejun Heo Cc: x86@kernel.org Cc: linux-kernel@vger.kernel.org --- arch/x86/kernel/dumpstack.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index e0648f7..74be764 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -250,9 +250,14 @@ void oops_end(unsigned long flags, struct pt_regs *regs, int signr) /* * We're not going to return, but we might be on an IST stack or * have very little stack space left. Rewind the stack and kill - * the task. + * the task. But kthread is a special case, since kthread uses + * stack to keep completion structure to be woken on vfork_done + * completion. */ - rewind_stack_do_exit(signr); + if (current->flags & PF_KTHREAD) + do_exit(signr); + else + rewind_stack_do_exit(signr); } NOKPROBE_SYMBOL(oops_end); -- 2.9.3