From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932408AbaIENqR (ORCPT ); Fri, 5 Sep 2014 09:46:17 -0400 Received: from mx1.redhat.com ([209.132.183.28]:25033 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757015AbaIENqP (ORCPT ); Fri, 5 Sep 2014 09:46:15 -0400 Date: Fri, 5 Sep 2014 15:43:44 +0200 From: Oleg Nesterov To: Suresh Siddha , "H. Peter Anvin" Cc: Fenghua Yu , Ingo Molnar , Linus Torvalds , x86@kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH v2 2/5] x86, fpu: don't abuse ->has_fpu in __kernel_fpu_{begin,end}() Message-ID: <20140905134344.GA14249@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140905134325.GA14228@redhat.com> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Now that we have in_kernel_fpu we can remove __thread_clear_has_fpu() in __kernel_fpu_begin(). And this allows to replace the asymmetrical and nontrivial use_eager_fpu+tsk_used_math check in kernel_fpu_end() with the same __thread_has_fpu(). The logic becomes really simple; if _begin() does save() then _end() needs restore(), this is controlled by __thread_has_fpu(). Otherwise they do clts/stts unless use_eager_fpu(). Not only this makes begin/end symmetrical and imo more understandable, potentially this allows to change irq_fpu_usable() to avoid all other checks except "in_kernel_fpu". Also, with this patch __kernel_fpu_end() does restore_fpu_checking() and WARNs if it fails instead of math_state_restore(). I think this looks better because we no longer need __thread_fpu_begin(), and it would be better to report the failure in this case. Signed-off-by: Oleg Nesterov --- arch/x86/kernel/i387.c | 19 ++++++------------- 1 files changed, 6 insertions(+), 13 deletions(-) diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 8fb8868..19dd36d 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -82,9 +82,7 @@ void __kernel_fpu_begin(void) /* FIXME: race with math_state_restore()-like code */ if (__thread_has_fpu(me)) { - __thread_clear_has_fpu(me); __save_init_fpu(me); - /* We do 'stts()' in __kernel_fpu_end() */ } else if (!use_eager_fpu()) { this_cpu_write(fpu_owner_task, NULL); clts(); @@ -94,17 +92,12 @@ EXPORT_SYMBOL(__kernel_fpu_begin); void __kernel_fpu_end(void) { - if (use_eager_fpu()) { - /* - * For eager fpu, most the time, tsk_used_math() is true. - * Restore the user math as we are done with the kernel usage. - * At few instances during thread exit, signal handling etc, - * tsk_used_math() is false. Those few places will take proper - * actions, so we don't need to restore the math here. - */ - if (likely(tsk_used_math(current))) - math_state_restore(); - } else { + struct task_struct *me = current; + + if (__thread_has_fpu(me)) { + if (WARN_ON(restore_fpu_checking(me))) + drop_init_fpu(me); + } else if (!use_eager_fpu()) { stts(); } -- 1.5.5.1