From patchwork Thu Sep 29 00:10:11 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Verma, Vishal L" X-Patchwork-Id: 9355435 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 AEA516077B for ; Thu, 29 Sep 2016 00:13:05 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 9E21929352 for ; Thu, 29 Sep 2016 00:13:05 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 923982964D; Thu, 29 Sep 2016 00:13:05 +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=unavailable 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 25DAE29352 for ; Thu, 29 Sep 2016 00:13:05 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750812AbcI2AND (ORCPT ); Wed, 28 Sep 2016 20:13:03 -0400 Received: from mga03.intel.com ([134.134.136.65]:17263 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753231AbcI2ANC (ORCPT ); Wed, 28 Sep 2016 20:13:02 -0400 Received: from orsmga004.jf.intel.com ([10.7.209.38]) by orsmga103.jf.intel.com with ESMTP; 28 Sep 2016 17:11:06 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.30,412,1470726000"; d="scan'208";a="14817502" Received: from omniknight.lm.intel.com ([10.232.112.53]) by orsmga004.jf.intel.com with ESMTP; 28 Sep 2016 17:11:03 -0700 From: Vishal Verma To: Cc: Dan Williams , Ross Zwisler , Linda Knippers , , "Rafael J. Wysocki" , Vishal Verma Subject: [PATCH 3/3] libnvdimm: clear the internal poison_list when clearing badblocks Date: Wed, 28 Sep 2016 18:10:11 -0600 Message-Id: <1475107811-8880-4-git-send-email-vishal.l.verma@intel.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1475107811-8880-1-git-send-email-vishal.l.verma@intel.com> References: <1475107811-8880-1-git-send-email-vishal.l.verma@intel.com> 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 nvdimm_clear_poison cleared the user-visible badblocks, and sent commands to the NVDIMM to clear the areas marked as 'poison', but it neglected to clear the same areas from the internal poison_list which is used to marshal ARS results before sorting them by namespace. As a result, once on-demand ARS functionality was added: 37b137f nfit, libnvdimm: allow an ARS scrub to be triggered on demand A scrub triggered from either sysfs or an MCE was found to be adding stale entries that had been cleared from gendisk->badblocks, but were still present in nvdimm_bus->poison_list. This adds the missing step of clearing poison_list entries when clearing poison, so that it is in sync with badblocks. Fixes: 37b137f nfit, libnvdimm: allow an ARS scrub to be triggered on demand Cc: Dan Williams Signed-off-by: Vishal Verma --- drivers/nvdimm/bus.c | 2 ++ drivers/nvdimm/core.c | 73 ++++++++++++++++++++++++++++++++++++++++++++--- include/linux/libnvdimm.h | 2 ++ 3 files changed, 73 insertions(+), 4 deletions(-) diff --git a/drivers/nvdimm/bus.c b/drivers/nvdimm/bus.c index 935866f..a8b6949 100644 --- a/drivers/nvdimm/bus.c +++ b/drivers/nvdimm/bus.c @@ -217,6 +217,8 @@ long nvdimm_clear_poison(struct device *dev, phys_addr_t phys, return rc; if (cmd_rc < 0) return cmd_rc; + + nvdimm_clear_from_poison_list(nvdimm_bus, phys, len); return clear_err.cleared; } EXPORT_SYMBOL_GPL(nvdimm_clear_poison); diff --git a/drivers/nvdimm/core.c b/drivers/nvdimm/core.c index 715583f..42e40db 100644 --- a/drivers/nvdimm/core.c +++ b/drivers/nvdimm/core.c @@ -541,11 +541,12 @@ void nvdimm_badblocks_populate(struct nd_region *nd_region, } EXPORT_SYMBOL_GPL(nvdimm_badblocks_populate); -static int add_poison(struct nvdimm_bus *nvdimm_bus, u64 addr, u64 length) +static int add_poison(struct nvdimm_bus *nvdimm_bus, u64 addr, u64 length, + gfp_t flags) { struct nd_poison *pl; - pl = kzalloc(sizeof(*pl), GFP_KERNEL); + pl = kzalloc(sizeof(*pl), flags); if (!pl) return -ENOMEM; @@ -561,7 +562,7 @@ static int bus_add_poison(struct nvdimm_bus *nvdimm_bus, u64 addr, u64 length) struct nd_poison *pl; if (list_empty(&nvdimm_bus->poison_list)) - return add_poison(nvdimm_bus, addr, length); + return add_poison(nvdimm_bus, addr, length, GFP_KERNEL); /* * There is a chance this is a duplicate, check for those first. @@ -581,7 +582,7 @@ static int bus_add_poison(struct nvdimm_bus *nvdimm_bus, u64 addr, u64 length) * as any overlapping ranges will get resolved when the list is consumed * and converted to badblocks */ - return add_poison(nvdimm_bus, addr, length); + return add_poison(nvdimm_bus, addr, length, GFP_KERNEL); } int nvdimm_bus_add_poison(struct nvdimm_bus *nvdimm_bus, u64 addr, u64 length) @@ -596,6 +597,70 @@ int nvdimm_bus_add_poison(struct nvdimm_bus *nvdimm_bus, u64 addr, u64 length) } EXPORT_SYMBOL_GPL(nvdimm_bus_add_poison); +void nvdimm_clear_from_poison_list(struct nvdimm_bus *nvdimm_bus, + phys_addr_t start, unsigned int len) +{ + struct list_head *poison_list = &nvdimm_bus->poison_list; + u64 clr_end = start + len - 1; + struct nd_poison *pl, *next; + + nvdimm_bus_lock(&nvdimm_bus->dev); + WARN_ON_ONCE(list_empty(poison_list)); + + /* + * [start, clr_end] is the poison interval being cleared. + * [pl->start, pl_end] is the poison_list entry we're comparing + * the above interval against. The poison list entry may need + * to be modified (update either start or length), deleted, or + * split into two based on the overlap characteristics + */ + + list_for_each_entry_safe(pl, next, poison_list, list) { + u64 pl_end = pl->start + pl->length - 1; + + /* Skip intervals with no intersection */ + if (pl_end < start) + continue; + if (pl->start > clr_end) + continue; + /* Delete completely overlapped poison entries */ + if ((pl->start >= start) && (pl_end <= clr_end)) { + list_del(&pl->list); + kfree(pl); + continue; + } + /* Adjust start point of partially cleared entries */ + if ((start <= pl->start) && (clr_end > pl->start)) { + pl->length -= clr_end - pl->start + 1; + pl->start = clr_end + 1; + continue; + } + /* Adjust pl->length for partial clearing at the tail end */ + if ((pl->start < start) && (pl_end <= clr_end)) { + /* pl->start remains the same */ + pl->length = start - pl->start; + continue; + } + /* + * If clearing in the middle of an entry, we split it into + * two by modifying the current entry to represent one half of + * the split, and adding a new entry for the second half. + */ + if ((pl->start < start) && (pl_end > clr_end)) { + u64 new_start = clr_end + 1; + u64 new_len = pl_end - new_start + 1; + + /* Add new entry covering the right half */ + add_poison(nvdimm_bus, new_start, new_len, GFP_NOIO); + /* Adjust this entry to cover the left half */ + pl->length = start - pl->start; + continue; + } + } + nvdimm_bus_unlock(&nvdimm_bus->dev); +} +EXPORT_SYMBOL_GPL(nvdimm_clear_from_poison_list); + #ifdef CONFIG_BLK_DEV_INTEGRITY int nd_integrity_init(struct gendisk *disk, unsigned long meta_size) { diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h index b519e13..bbfce62 100644 --- a/include/linux/libnvdimm.h +++ b/include/linux/libnvdimm.h @@ -129,6 +129,8 @@ static inline struct nd_blk_region_desc *to_blk_region_desc( } int nvdimm_bus_add_poison(struct nvdimm_bus *nvdimm_bus, u64 addr, u64 length); +void nvdimm_clear_from_poison_list(struct nvdimm_bus *nvdimm_bus, + phys_addr_t start, unsigned int len); struct nvdimm_bus *nvdimm_bus_register(struct device *parent, struct nvdimm_bus_descriptor *nfit_desc); void nvdimm_bus_unregister(struct nvdimm_bus *nvdimm_bus);