From patchwork Tue May 30 23:42:53 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Verma, Vishal L" X-Patchwork-Id: 9755623 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 7EED3602BF for ; Tue, 30 May 2017 23:46:10 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 679EB2623C for ; Tue, 30 May 2017 23:46:10 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 48B96283F5; Tue, 30 May 2017 23:46:10 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=2.0 tests=BAYES_00,RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 61AF92623C for ; Tue, 30 May 2017 23:46:08 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750913AbdE3XqH (ORCPT ); Tue, 30 May 2017 19:46:07 -0400 Received: from mga14.intel.com ([192.55.52.115]:5586 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750902AbdE3XqH (ORCPT ); Tue, 30 May 2017 19:46:07 -0400 Received: from fmsmga004.fm.intel.com ([10.253.24.48]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 30 May 2017 16:46:06 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.38,420,1491289200"; d="scan'208";a="268249565" Received: from omniknight.lm.intel.com ([10.232.112.53]) by fmsmga004.fm.intel.com with ESMTP; 30 May 2017 16:46:06 -0700 From: Vishal Verma To: Cc: Vishal Verma , linux-nvdimm@lists.01.org, Dan Williams , "Rafael J . Wysocki" Subject: [RFC PATCH] acpi: change memory allocations to GFP_NOIO Date: Tue, 30 May 2017 17:42:53 -0600 Message-Id: <20170530234253.16510-1-vishal.l.verma@intel.com> X-Mailer: git-send-email 2.9.3 Sender: linux-acpi-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-acpi@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP With the ACPI NFIT 'DSM' methods, acpi can be called from IO paths. Specifically, the DSM to clear media errors is called during writes, so that we can provide a writes-fix-errors model. However it is easy to imagine a scenario like: - write through the nvdimm driver - acpi allocation - writeback, causes more IO through the nvdimm driver - deadlock Making the acpi allocations GPF_NOIO would ensure that it doesn't trigger writeback, and avoids the above deadlock. Cc: Cc: Cc: Dan Williams Cc: Rafael J. Wysocki Signed-off-by: Vishal Verma --- Is this too invasive a change? ACPI code in general appears to be good about error-checking allocations, and in our scenario above, a failed allocation would just mean the attempt to issue the DSM would fail, and the IO would simply fail. include/acpi/platform/aclinuxex.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/include/acpi/platform/aclinuxex.h b/include/acpi/platform/aclinuxex.h index efdff52..9bd54ec 100644 --- a/include/acpi/platform/aclinuxex.h +++ b/include/acpi/platform/aclinuxex.h @@ -83,12 +83,12 @@ acpi_status acpi_os_terminate(void); */ static inline void *acpi_os_allocate(acpi_size size) { - return kmalloc(size, irqs_disabled()? GFP_ATOMIC : GFP_KERNEL); + return kmalloc(size, irqs_disabled()? GFP_ATOMIC : GFP_NOIO); } static inline void *acpi_os_allocate_zeroed(acpi_size size) { - return kzalloc(size, irqs_disabled()? GFP_ATOMIC : GFP_KERNEL); + return kzalloc(size, irqs_disabled()? GFP_ATOMIC : GFP_NOIO); } static inline void acpi_os_free(void *memory) @@ -99,7 +99,7 @@ static inline void acpi_os_free(void *memory) static inline void *acpi_os_acquire_object(acpi_cache_t * cache) { return kmem_cache_zalloc(cache, - irqs_disabled()? GFP_ATOMIC : GFP_KERNEL); + irqs_disabled()? GFP_ATOMIC : GFP_NOIO); } static inline acpi_thread_id acpi_os_get_thread_id(void)