From patchwork Fri Mar 29 15:27:44 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Dan Williams X-Patchwork-Id: 10877465 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 A3A521874 for ; Fri, 29 Mar 2019 15:40:26 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 8FF9924B5B for ; Fri, 29 Mar 2019 15:40:26 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 842BA28779; Fri, 29 Mar 2019 15:40:26 +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=-2.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, RCVD_IN_DNSWL_NONE autolearn=ham version=3.3.1 Received: from ml01.01.org (ml01.01.org [198.145.21.10]) (using TLSv1.2 with cipher DHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id CF369288AE for ; Fri, 29 Mar 2019 15:40:25 +0000 (UTC) Received: from [127.0.0.1] (localhost [IPv6:::1]) by ml01.01.org (Postfix) with ESMTP id C90C8211EC669; Fri, 29 Mar 2019 08:40:25 -0700 (PDT) X-Original-To: linux-nvdimm@lists.01.org Delivered-To: linux-nvdimm@lists.01.org Received-SPF: Pass (sender SPF authorized) identity=mailfrom; client-ip=134.134.136.126; helo=mga18.intel.com; envelope-from=dan.j.williams@intel.com; receiver=linux-nvdimm@lists.01.org Received: from mga18.intel.com (mga18.intel.com [134.134.136.126]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ml01.01.org (Postfix) with ESMTPS id 1F6E0211EA9EC for ; Fri, 29 Mar 2019 08:40:24 -0700 (PDT) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga002.fm.intel.com ([10.253.24.26]) by orsmga106.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 29 Mar 2019 08:40:24 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.60,284,1549958400"; d="scan'208";a="156424704" Received: from dwillia2-desk3.jf.intel.com (HELO dwillia2-desk3.amr.corp.intel.com) ([10.54.39.16]) by fmsmga002.fm.intel.com with ESMTP; 29 Mar 2019 08:40:23 -0700 Subject: [PATCH 4/6] lib/genalloc: Introduce chunk owners From: Dan Williams To: akpm@linux-foundation.org Date: Fri, 29 Mar 2019 08:27:44 -0700 Message-ID: <155387326480.2443841.4506646109682116121.stgit@dwillia2-desk3.amr.corp.intel.com> In-Reply-To: <155387324370.2443841.574715745262628837.stgit@dwillia2-desk3.amr.corp.intel.com> References: <155387324370.2443841.574715745262628837.stgit@dwillia2-desk3.amr.corp.intel.com> User-Agent: StGit/0.18-2-gc94f MIME-Version: 1.0 X-BeenThere: linux-nvdimm@lists.01.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: "Linux-nvdimm developer list." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: linux-nvdimm@lists.01.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, =?utf-8?b?SsOpcsO0bWU=?= Glisse , linux-pci@vger.kernel.org, Bjorn Helgaas , Christoph Hellwig Errors-To: linux-nvdimm-bounces@lists.01.org Sender: "Linux-nvdimm" X-Virus-Scanned: ClamAV using ClamSMTP The p2pdma facility enables a provider to publish a pool of dma addresses for a consumer to allocate. A genpool is used internally by p2pdma to collect dma resources, 'chunks', to be handed out to consumers. Whenever a consumer allocates a resource it needs to pin the 'struct dev_pagemap' instance that backs the chunk selected by pci_alloc_p2pmem(). Currently that reference is taken globally on the entire provider device. That sets up a lifetime mismatch whereby the p2pdma core needs to maintain hacks to make sure the percpu_ref is not released twice. This lifetime mismatch also stands in the way of a fix to devm_memremap_pages() whereby devm_memremap_pages_release() must wait for the percpu_ref ->release() callback to complete before it can proceed to teardown pages. So, towards fixing this situation, introduce the ability to store a 'chunk owner' at gen_pool_add() time, and a facility to retrieve the owner at gen_pool_{alloc,free}() time. For p2pdma this will be used to store and recall individual dev_pagemap reference counter instances per-chunk. Cc: Logan Gunthorpe Cc: Ira Weiny Cc: Bjorn Helgaas Cc: "Jérôme Glisse" Cc: Christoph Hellwig Signed-off-by: Dan Williams --- include/linux/genalloc.h | 55 +++++++++++++++++++++++++++++++++++++++++----- lib/genalloc.c | 51 +++++++++++++++++++++---------------------- 2 files changed, 74 insertions(+), 32 deletions(-) diff --git a/include/linux/genalloc.h b/include/linux/genalloc.h index dd0a452373e7..a337313e064f 100644 --- a/include/linux/genalloc.h +++ b/include/linux/genalloc.h @@ -75,6 +75,7 @@ struct gen_pool_chunk { struct list_head next_chunk; /* next chunk in pool */ atomic_long_t avail; phys_addr_t phys_addr; /* physical starting address of memory chunk */ + void *owner; /* private data to retrieve at alloc time */ unsigned long start_addr; /* start address of memory chunk */ unsigned long end_addr; /* end address of memory chunk (inclusive) */ unsigned long bits[0]; /* bitmap for allocating memory chunk */ @@ -96,8 +97,15 @@ struct genpool_data_fixed { extern struct gen_pool *gen_pool_create(int, int); extern phys_addr_t gen_pool_virt_to_phys(struct gen_pool *pool, unsigned long); -extern int gen_pool_add_virt(struct gen_pool *, unsigned long, phys_addr_t, - size_t, int); +extern int gen_pool_add_owner(struct gen_pool *, unsigned long, phys_addr_t, + size_t, int, void *); + +static inline int gen_pool_add_virt(struct gen_pool *pool, unsigned long addr, + phys_addr_t phys, size_t size, int nid) +{ + return gen_pool_add_owner(pool, addr, phys, size, nid, NULL); +} + /** * gen_pool_add - add a new chunk of special memory to the pool * @pool: pool to add new memory chunk to @@ -116,12 +124,47 @@ static inline int gen_pool_add(struct gen_pool *pool, unsigned long addr, return gen_pool_add_virt(pool, addr, -1, size, nid); } extern void gen_pool_destroy(struct gen_pool *); -extern unsigned long gen_pool_alloc(struct gen_pool *, size_t); -extern unsigned long gen_pool_alloc_algo(struct gen_pool *, size_t, - genpool_algo_t algo, void *data); +unsigned long gen_pool_alloc_algo_owner(struct gen_pool *pool, size_t size, + genpool_algo_t algo, void *data, void **owner); + +static inline unsigned long gen_pool_alloc_owner(struct gen_pool *pool, + size_t size, void **owner) +{ + return gen_pool_alloc_algo_owner(pool, size, pool->algo, pool->data, + owner); +} + +static inline unsigned long gen_pool_alloc_algo(struct gen_pool *pool, + size_t size, genpool_algo_t algo, void *data) +{ + return gen_pool_alloc_algo_owner(pool, size, algo, data, NULL); +} + +/** + * gen_pool_alloc - allocate special memory from the pool + * @pool: pool to allocate from + * @size: number of bytes to allocate from the pool + * + * Allocate the requested number of bytes from the specified pool. + * Uses the pool allocation function (with first-fit algorithm by default). + * Can not be used in NMI handler on architectures without + * NMI-safe cmpxchg implementation. + */ +static inline unsigned long gen_pool_alloc(struct gen_pool *pool, size_t size) +{ + return gen_pool_alloc_algo(pool, size, pool->algo, pool->data); +} + extern void *gen_pool_dma_alloc(struct gen_pool *pool, size_t size, dma_addr_t *dma); -extern void gen_pool_free(struct gen_pool *, unsigned long, size_t); +extern void gen_pool_free_owner(struct gen_pool *pool, unsigned long addr, + size_t size, void **owner); +static inline void gen_pool_free(struct gen_pool *pool, unsigned long addr, + size_t size) +{ + gen_pool_free_owner(pool, addr, size, NULL); +} + extern void gen_pool_for_each_chunk(struct gen_pool *, void (*)(struct gen_pool *, struct gen_pool_chunk *, void *), void *); extern size_t gen_pool_avail(struct gen_pool *); diff --git a/lib/genalloc.c b/lib/genalloc.c index 7e85d1e37a6e..770c769d7cb7 100644 --- a/lib/genalloc.c +++ b/lib/genalloc.c @@ -168,20 +168,21 @@ struct gen_pool *gen_pool_create(int min_alloc_order, int nid) EXPORT_SYMBOL(gen_pool_create); /** - * gen_pool_add_virt - add a new chunk of special memory to the pool + * gen_pool_add_owner- add a new chunk of special memory to the pool * @pool: pool to add new memory chunk to * @virt: virtual starting address of memory chunk to add to pool * @phys: physical starting address of memory chunk to add to pool * @size: size in bytes of the memory chunk to add to pool * @nid: node id of the node the chunk structure and bitmap should be * allocated on, or -1 + * @owner: private data the publisher would like to recall at alloc time * * Add a new chunk of special memory to the specified pool. * * Returns 0 on success or a -ve errno on failure. */ -int gen_pool_add_virt(struct gen_pool *pool, unsigned long virt, phys_addr_t phys, - size_t size, int nid) +int gen_pool_add_owner(struct gen_pool *pool, unsigned long virt, phys_addr_t phys, + size_t size, int nid, void *owner) { struct gen_pool_chunk *chunk; int nbits = size >> pool->min_alloc_order; @@ -195,6 +196,7 @@ int gen_pool_add_virt(struct gen_pool *pool, unsigned long virt, phys_addr_t phy chunk->phys_addr = phys; chunk->start_addr = virt; chunk->end_addr = virt + size - 1; + chunk->owner = owner; atomic_long_set(&chunk->avail, size); spin_lock(&pool->lock); @@ -203,7 +205,7 @@ int gen_pool_add_virt(struct gen_pool *pool, unsigned long virt, phys_addr_t phy return 0; } -EXPORT_SYMBOL(gen_pool_add_virt); +EXPORT_SYMBOL(gen_pool_add_owner); /** * gen_pool_virt_to_phys - return the physical address of memory @@ -260,35 +262,20 @@ void gen_pool_destroy(struct gen_pool *pool) EXPORT_SYMBOL(gen_pool_destroy); /** - * gen_pool_alloc - allocate special memory from the pool - * @pool: pool to allocate from - * @size: number of bytes to allocate from the pool - * - * Allocate the requested number of bytes from the specified pool. - * Uses the pool allocation function (with first-fit algorithm by default). - * Can not be used in NMI handler on architectures without - * NMI-safe cmpxchg implementation. - */ -unsigned long gen_pool_alloc(struct gen_pool *pool, size_t size) -{ - return gen_pool_alloc_algo(pool, size, pool->algo, pool->data); -} -EXPORT_SYMBOL(gen_pool_alloc); - -/** - * gen_pool_alloc_algo - allocate special memory from the pool + * gen_pool_alloc_algo_owner - allocate special memory from the pool * @pool: pool to allocate from * @size: number of bytes to allocate from the pool * @algo: algorithm passed from caller * @data: data passed to algorithm + * @owner: optionally retrieve the chunk owner * * Allocate the requested number of bytes from the specified pool. * Uses the pool allocation function (with first-fit algorithm by default). * Can not be used in NMI handler on architectures without * NMI-safe cmpxchg implementation. */ -unsigned long gen_pool_alloc_algo(struct gen_pool *pool, size_t size, - genpool_algo_t algo, void *data) +unsigned long gen_pool_alloc_algo_owner(struct gen_pool *pool, size_t size, + genpool_algo_t algo, void *data, void **owner) { struct gen_pool_chunk *chunk; unsigned long addr = 0; @@ -299,6 +286,9 @@ unsigned long gen_pool_alloc_algo(struct gen_pool *pool, size_t size, BUG_ON(in_nmi()); #endif + if (owner) + *owner = NULL; + if (size == 0) return 0; @@ -326,12 +316,14 @@ unsigned long gen_pool_alloc_algo(struct gen_pool *pool, size_t size, addr = chunk->start_addr + ((unsigned long)start_bit << order); size = nbits << order; atomic_long_sub(size, &chunk->avail); + if (owner) + *owner = chunk->owner; break; } rcu_read_unlock(); return addr; } -EXPORT_SYMBOL(gen_pool_alloc_algo); +EXPORT_SYMBOL(gen_pool_alloc_algo_owner); /** * gen_pool_dma_alloc - allocate special memory from the pool for DMA usage @@ -367,12 +359,14 @@ EXPORT_SYMBOL(gen_pool_dma_alloc); * @pool: pool to free to * @addr: starting address of memory to free back to pool * @size: size in bytes of memory to free + * @owner: private data stashed at gen_pool_add() time * * Free previously allocated special memory back to the specified * pool. Can not be used in NMI handler on architectures without * NMI-safe cmpxchg implementation. */ -void gen_pool_free(struct gen_pool *pool, unsigned long addr, size_t size) +void gen_pool_free_owner(struct gen_pool *pool, unsigned long addr, size_t size, + void **owner) { struct gen_pool_chunk *chunk; int order = pool->min_alloc_order; @@ -382,6 +376,9 @@ void gen_pool_free(struct gen_pool *pool, unsigned long addr, size_t size) BUG_ON(in_nmi()); #endif + if (owner) + *owner = NULL; + nbits = (size + (1UL << order) - 1) >> order; rcu_read_lock(); list_for_each_entry_rcu(chunk, &pool->chunks, next_chunk) { @@ -392,6 +389,8 @@ void gen_pool_free(struct gen_pool *pool, unsigned long addr, size_t size) BUG_ON(remain); size = nbits << order; atomic_long_add(size, &chunk->avail); + if (owner) + *owner = chunk->owner; rcu_read_unlock(); return; } @@ -399,7 +398,7 @@ void gen_pool_free(struct gen_pool *pool, unsigned long addr, size_t size) rcu_read_unlock(); BUG(); } -EXPORT_SYMBOL(gen_pool_free); +EXPORT_SYMBOL(gen_pool_free_owner); /** * gen_pool_for_each_chunk - call func for every chunk of generic memory pool