From patchwork Thu Aug 9 02:54:09 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Rashmica Gupta X-Patchwork-Id: 10560841 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 91828174A for ; Thu, 9 Aug 2018 02:54:39 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 7DF832ABD8 for ; Thu, 9 Aug 2018 02:54:39 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 701432ABE6; Thu, 9 Aug 2018 02:54:39 +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=-3.0 required=2.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FROM,MAILING_LIST_MULTI,RCVD_IN_DNSWL_NONE autolearn=ham version=3.3.1 Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 9F86C2ABD8 for ; Thu, 9 Aug 2018 02:54:38 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 876F76B0005; Wed, 8 Aug 2018 22:54:37 -0400 (EDT) Delivered-To: linux-mm-outgoing@kvack.org Received: by kanga.kvack.org (Postfix, from userid 40) id 84C266B0006; Wed, 8 Aug 2018 22:54:37 -0400 (EDT) X-Original-To: int-list-linux-mm@kvack.org X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 7396F6B0007; Wed, 8 Aug 2018 22:54:37 -0400 (EDT) X-Original-To: linux-mm@kvack.org X-Delivered-To: linux-mm@kvack.org Received: from mail-pf1-f200.google.com (mail-pf1-f200.google.com [209.85.210.200]) by kanga.kvack.org (Postfix) with ESMTP id 326876B0005 for ; Wed, 8 Aug 2018 22:54:37 -0400 (EDT) Received: by mail-pf1-f200.google.com with SMTP id a23-v6so2456302pfo.23 for ; Wed, 08 Aug 2018 19:54:37 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:dkim-signature:from:to:cc:subject:date :message-id; bh=tFwl8V00znn4BRd+60VMf7GTR2h8kaA8lcYmqGSB/Rw=; b=KxcBMI4zPOeZuHYjFTX4T9YteGH0M4SMxmyGEpXFqoT2P5xYC+SDnlM9K7CnsXupwq qyiQzxq2i/37gArPPTvqD5+Gl3lXG39/4oGNBQePlQ9eTMDahMr2BFbw2U05wi5VgONa xJuWGuYqQ2MlNKzbxJbEaX65kSKBkd6cNMsS0x/Jr0dZQkWnB1WP/gEPtfCB1+sxdDFL aCWVOa5GXNSmygFdpJDB9INbCpJTaxBC9XtLKvyqbc3h6kfgyYpU9mHFn5AyXjRWIRqg wJOoeCFJ2c0XT7pX+Cf3h5hLWK/xOYg3oBmh1zpXp4DmoJh6w8A8kI/0wJDnT7EGxLdb KSZQ== X-Gm-Message-State: AOUpUlGOiOGwObj0q3C/YOnvk2XOW0uZ5jNTGuoQ2zYw9xOuj/yP4tvd LskuDgWPQx1SJLZS+aHKu5Xtb5cY+bnqs8x57QFnfpRD5UZQJg0iFh7OkmvjLvcn3ptjVzFoEv5 LZ1la+k2sxiVZhYfexuvpBHZc6gCyMwdH/dFzbERrFM2hUVhkVIZECAAqbpBIoYmwsWWRAI3DHp tJOHNNUHkjNh9E4H+Wz+GkHJVf+SqAY/A35g2dz5AHr/+YdMtLgUVMMhxJLioHordxOBkAns3Ch QcdGG8/wIdaBfUJTrJySL9h5OgC8NHyy3Lqg9X4wpQcKqLuxqU/jXDmyWO0Pxf33VQ5yA+6RLva OvJRhEcuWe6brpizMwUhQRzmwxB6rPvj3y7f5G2ntllcWF0QTGp7LX7Aqoe46Com7UtppwdpR6/ e X-Received: by 2002:a63:181e:: with SMTP id y30-v6mr318389pgl.82.1533783276826; Wed, 08 Aug 2018 19:54:36 -0700 (PDT) X-Received: by 2002:a63:181e:: with SMTP id y30-v6mr318308pgl.82.1533783274871; Wed, 08 Aug 2018 19:54:34 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1533783274; cv=none; d=google.com; s=arc-20160816; b=umI/xbv7PtFUspoZ2ZtJZHJHkZTqej9Qk4bnLErgZvHdOU54xUFz2WTmUyTjiWmAuX gXaE+Sw0X4yHOfn/Y/nflN+G3JhYFcANsdoEf0Rbp56osoAaxLO+HGs1JCp8Q85AJlvn KHIeaSkgDXpE8Q7yocQxe+YyirmHA8lFinpuO0L8Pcxnn00JWQlxoHTnOkl6gkset1wF 19yMXAt4893lk6XG9j6AUT1Y/Fc0pSljMA4RGXjkW8MN/iB4JSzkgrCeu7/PAwkgT3C7 TZTYggmbJGGhxGdMxoJNkZAg8WYi+q9HC9B3xlBWpI/RXfN4CLKbsJEhoMUVXVjfO5nE Vvig== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=message-id:date:subject:cc:to:from:dkim-signature :arc-authentication-results; bh=tFwl8V00znn4BRd+60VMf7GTR2h8kaA8lcYmqGSB/Rw=; b=s5+obGAulmNaKNb10L9rD5aIJ/wQ/y3iwpYBhHtzeVKbEW2VrfLdjLCwHuriDar6en HCL/CVzAMdftV/9JuXEOMGvCROXLzIgmjXaMsR5mbyposeus/DBc0NvLU2sTstUHZ2f1 ODbjm/riU6iXg6iXBL8o0FHg8EaaTIwdeTrYqrdkBIoUs0WYRWXEIM/ZlPJcHU23E3qq eXhTxeZ4rbqspHTcpH0PfLrlsEtdKIIHVYDFcNsguhxJRg02rhn4VpySFEZVP2S5kgRE E2czgCs+PCsrtbXxMfDcr0HBjqb5DqUqzMAZgR5se4zDl3Fefh/+9j48e7evF3lj0xio Tu1g== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@gmail.com header.s=20161025 header.b=qU7CS+0f; spf=pass (google.com: domain of rashmica.g@gmail.com designates 209.85.220.65 as permitted sender) smtp.mailfrom=rashmica.g@gmail.com; dmarc=pass (p=NONE sp=QUARANTINE dis=NONE) header.from=gmail.com Received: from mail-sor-f65.google.com (mail-sor-f65.google.com. [209.85.220.65]) by mx.google.com with SMTPS id p28-v6sor1697625pfd.84.2018.08.08.19.54.34 for (Google Transport Security); Wed, 08 Aug 2018 19:54:34 -0700 (PDT) Received-SPF: pass (google.com: domain of rashmica.g@gmail.com designates 209.85.220.65 as permitted sender) client-ip=209.85.220.65; Authentication-Results: mx.google.com; dkim=pass header.i=@gmail.com header.s=20161025 header.b=qU7CS+0f; spf=pass (google.com: domain of rashmica.g@gmail.com designates 209.85.220.65 as permitted sender) smtp.mailfrom=rashmica.g@gmail.com; dmarc=pass (p=NONE sp=QUARANTINE dis=NONE) header.from=gmail.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:cc:subject:date:message-id; bh=tFwl8V00znn4BRd+60VMf7GTR2h8kaA8lcYmqGSB/Rw=; b=qU7CS+0fAa1sGX3ya47y9enOICbUYETce6ITJO+RX+Cp8QYa4h3+YX6ahK751u2xte 8CXCOrxApwKdVKia3JGvuIJfGjxUDv3+wvL3+4fHjBF6WDbqTFkzajnBYsO588yp0fJd dd72Osgzr7TXdedniMWMMG4krjNiXAwlJz3GUt1zjF8iccRh/XjEi31Tn6Ctk6nXx3aD xs9L7J+1V32O39a45Sqi3dYaOg4Z0ZwK4qXU75y15tyW0y8XBiIOM35AiFQVYlC6gVzI CKHpsyKX/xjU1wZhqN7GhLvKaFXIcQ9iV8D41CRxFFCKwxWhR/FPkUO7BqSqAah6iCqn /rSg== X-Google-Smtp-Source: AA+uWPwgEXWKSnTIwAe+GM/lG6p4rPNk1Nqt57jrbLI68G++3mE2natSennJjzIvdhjNkR/Akw7HKA== X-Received: by 2002:a62:ce05:: with SMTP id y5-v6mr320390pfg.95.1533783274295; Wed, 08 Aug 2018 19:54:34 -0700 (PDT) Received: from rashmica.ozlabs.ibm.com ([122.99.82.10]) by smtp.gmail.com with ESMTPSA id o10-v6sm9796705pfk.76.2018.08.08.19.54.27 (version=TLS1_2 cipher=ECDHE-RSA-CHACHA20-POLY1305 bits=256/256); Wed, 08 Aug 2018 19:54:33 -0700 (PDT) From: Rashmica Gupta To: toshi.kani@hpe.com, tglx@linutronix.de, akpm@linux-foundation.org, bp@suse.de, brijesh.singh@amd.com, thomas.lendacky@amd.com, jglisse@redhat.com, gregkh@linuxfoundation.org, baiyaowei@cmss.chinamobile.com, dan.j.williams@intel.com, mhocko@suse.com, iamjoonsoo.kim@lge.com, vbabka@suse.cz, malat@debian.org, bhelgaas@google.com, osalvador@techadventures.net, yasu.isimatu@gmail.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org, rppt@linux.vnet.ibm.com Cc: Rashmica Gupta Subject: [PATCH v3] resource: Merge resources on a node when hot-adding memory Date: Thu, 9 Aug 2018 12:54:09 +1000 Message-Id: <20180809025409.31552-1-rashmica.g@gmail.com> X-Mailer: git-send-email 2.14.4 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: X-Virus-Scanned: ClamAV using ClamSMTP When hot-removing memory release_mem_region_adjustable() splits iomem resources if they are not the exact size of the memory being hot-deleted. Adding this memory back to the kernel adds a new resource. Eg a node has memory 0x0 - 0xfffffffff. Offlining and hot-removing 1GB from 0xf40000000 results in the single resource 0x0-0xfffffffff being split into two resources: 0x0-0xf3fffffff and 0xf80000000-0xfffffffff. When we hot-add the memory back we now have three resources: 0x0-0xf3fffffff, 0xf40000000-0xf7fffffff, and 0xf80000000-0xfffffffff. Now if we try to remove some memory that overlaps these resources, like 2GB from 0xf40000000, release_mem_region_adjustable() fails as it expects the chunk of memory to be within the boundaries of a single resource. This patch adds a function request_resource_and_merge(). This is called instead of request_resource_conflict() when registering a resource in add_memory(). It calls request_resource_conflict() and if hot-removing is enabled (if it isn't we won't get resource fragmentation) we attempt to merge contiguous resources on the node. Signed-off-by: Rashmica Gupta Acked-by: Vlastimil Babka Reviewed-by: Mike Rapoport --- v2->v3: Update Xen balloon, make the commit msg and a comment clearer, and changed '>' to '>=' when comparing the end of a resource and the end of a node. v1->v2: Only attempt to merge resources if hot-remove is enabled. drivers/xen/balloon.c | 3 +- include/linux/ioport.h | 2 + include/linux/memory_hotplug.h | 2 +- kernel/resource.c | 120 +++++++++++++++++++++++++++++++++++++++++ mm/memory_hotplug.c | 22 ++++---- 5 files changed, 136 insertions(+), 13 deletions(-) diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c index 065f0b607373..9b972b37b0da 100644 --- a/drivers/xen/balloon.c +++ b/drivers/xen/balloon.c @@ -401,7 +401,8 @@ static enum bp_state reserve_additional_memory(void) * callers drop the mutex before trying again. */ mutex_unlock(&balloon_mutex); - rc = add_memory_resource(nid, resource, memhp_auto_online); + rc = add_memory_resource(nid, resource->start, resource_size(resource), + memhp_auto_online); mutex_lock(&balloon_mutex); if (rc) { diff --git a/include/linux/ioport.h b/include/linux/ioport.h index da0ebaec25f0..f5b93a711e86 100644 --- a/include/linux/ioport.h +++ b/include/linux/ioport.h @@ -189,6 +189,8 @@ extern int allocate_resource(struct resource *root, struct resource *new, resource_size_t, resource_size_t), void *alignf_data); +extern struct resource *request_resource_and_merge(struct resource *parent, + struct resource *new, int nid); struct resource *lookup_resource(struct resource *root, resource_size_t start); int adjust_resource(struct resource *res, resource_size_t start, resource_size_t size); diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h index 4e9828cda7a2..9c00f97c8cc6 100644 --- a/include/linux/memory_hotplug.h +++ b/include/linux/memory_hotplug.h @@ -322,7 +322,7 @@ static inline void remove_memory(int nid, u64 start, u64 size) {} extern int walk_memory_range(unsigned long start_pfn, unsigned long end_pfn, void *arg, int (*func)(struct memory_block *, void *)); extern int add_memory(int nid, u64 start, u64 size); -extern int add_memory_resource(int nid, struct resource *resource, bool online); +extern int add_memory_resource(int nid, u64 start, u64 size, bool online); extern int arch_add_memory(int nid, u64 start, u64 size, struct vmem_altmap *altmap, bool want_memblock); extern void move_pfn_range_to_zone(struct zone *zone, unsigned long start_pfn, diff --git a/kernel/resource.c b/kernel/resource.c index 30e1bc68503b..a31d3f5bccb7 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -1621,3 +1621,123 @@ static int __init strict_iomem(char *str) } __setup("iomem=", strict_iomem); + +#ifdef CONFIG_MEMORY_HOTPLUG +#ifdef CONFIG_MEMORY_HOTREMOVE +/* + * Attempt to merge resource and it's sibling + */ +static int merge_resources(struct resource *res) +{ + struct resource *next; + struct resource *tmp; + uint64_t size; + int ret = -EINVAL; + + next = res->sibling; + + /* + * Not sure how to handle two different children. So only attempt + * to merge two resources if neither have children, only one has a + * child or if both have the same child. + */ + if ((res->child && next->child) && (res->child != next->child)) + return ret; + + if (res->end + 1 != next->start) + return ret; + + if (res->flags != next->flags) + return ret; + + /* Update sibling and child of resource */ + res->sibling = next->sibling; + tmp = res->child; + if (!res->child) + res->child = next->child; + + size = next->end - res->start + 1; + ret = __adjust_resource(res, res->start, size); + if (ret) { + /* Failed so restore resource to original state */ + res->sibling = next; + res->child = tmp; + return ret; + } + + free_resource(next); + + return ret; +} + +/* + * Attempt to merge resources on the node + */ +static void merge_node_resources(int nid, struct resource *parent) +{ + struct resource *res; + uint64_t start_addr; + uint64_t end_addr; + int ret; + + start_addr = node_start_pfn(nid) << PAGE_SHIFT; + end_addr = node_end_pfn(nid) << PAGE_SHIFT; + + write_lock(&resource_lock); + + /* Get the first resource */ + res = parent->child; + + while (res) { + /* Check that the resource is within the node */ + if (res->start < start_addr) { + res = res->sibling; + continue; + } + /* Exit if sibling resource is past end of node */ + if (res->sibling->end >= end_addr) + break; + + ret = merge_resources(res); + if (!ret) + continue; + res = res->sibling; + } + write_unlock(&resource_lock); +} +#endif /* CONFIG_MEMORY_HOTREMOVE */ + +/** + * request_resource_and_merge() - request an I/O or memory resource for hot-add + * @parent: parent resource descriptor + * @new: resource descriptor desired by caller + * @nid: node id of the node we want the resource on + * + * If no conflict resource then attempt to merge resources on the node. + * + * This is intended to cleanup the fragmentation of resources that occurs when + * hot-removing memory (see release_mem_region_adjustable). If hot-removing is + * not enabled then there is no point trying to merge resources. + * + * Note that the inability to merge resources is not an error. + * + * Return: NULL for successful request of resource and conflict resource if + * there was a conflict. + */ +struct resource *request_resource_and_merge(struct resource *parent, + struct resource *new, int nid) +{ + struct resource *conflict; + + conflict = request_resource_conflict(parent, new); + + if (conflict) + return conflict; + +#ifdef CONFIG_MEMORY_HOTREMOVE + merge_node_resources(nid, parent); +#endif /* CONFIG_MEMORY_HOTREMOVE */ + + return NULL; +} +#endif /* CONFIG_MEMORY_HOTPLUG */ diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 7deb49f69e27..2e342f5ce322 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -97,7 +97,7 @@ void mem_hotplug_done(void) } /* add this memory to iomem resource */ -static struct resource *register_memory_resource(u64 start, u64 size) +static struct resource *register_memory_resource(int nid, u64 start, u64 size) { struct resource *res, *conflict; res = kzalloc(sizeof(struct resource), GFP_KERNEL); @@ -108,7 +108,7 @@ static struct resource *register_memory_resource(u64 start, u64 size) res->start = start; res->end = start + size - 1; res->flags = IORESOURCE_SYSTEM_RAM | IORESOURCE_BUSY; - conflict = request_resource_conflict(&iomem_resource, res); + conflict = request_resource_and_merge(&iomem_resource, res, nid); if (conflict) { if (conflict->desc == IORES_DESC_DEVICE_PRIVATE_MEMORY) { pr_debug("Device unaddressable memory block " @@ -122,11 +122,15 @@ static struct resource *register_memory_resource(u64 start, u64 size) return res; } -static void release_memory_resource(struct resource *res) +static void release_memory_resource(struct resource *res, u64 start, u64 size) { if (!res) return; +#ifdef CONFIG_MEMORY_HOTREMOVE + release_mem_region_adjustable(&iomem_resource, start, size); +#else release_resource(res); +#endif kfree(res); return; } @@ -1096,17 +1100,13 @@ static int online_memory_block(struct memory_block *mem, void *arg) } /* we are OK calling __meminit stuff here - we have CONFIG_MEMORY_HOTPLUG */ -int __ref add_memory_resource(int nid, struct resource *res, bool online) +int __ref add_memory_resource(int nid, u64 start, u64 size, bool online) { - u64 start, size; pg_data_t *pgdat = NULL; bool new_pgdat; bool new_node; int ret; - start = res->start; - size = resource_size(res); - ret = check_hotplug_memory_range(start, size); if (ret) return ret; @@ -1195,13 +1195,13 @@ int __ref add_memory(int nid, u64 start, u64 size) struct resource *res; int ret; - res = register_memory_resource(start, size); + res = register_memory_resource(nid, start, size); if (IS_ERR(res)) return PTR_ERR(res); - ret = add_memory_resource(nid, res, memhp_auto_online); + ret = add_memory_resource(nid, start, size, memhp_auto_online); if (ret < 0) - release_memory_resource(res); + release_memory_resource(res, start, size); return ret; } EXPORT_SYMBOL_GPL(add_memory);