From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrew Morton Subject: + mm-add-page_cache_readahead_unbounded.patch added to -mm tree Date: Tue, 14 Apr 2020 18:18:00 -0700 Message-ID: <20200415011800.CI2eyn5j1%akpm@linux-foundation.org> References: <20200412004155.1a8f4e081b4e03ef5903abb5@linux-foundation.org> Reply-To: linux-kernel@vger.kernel.org Return-path: Received: from mail.kernel.org ([198.145.29.99]:51346 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2392439AbgDOBSC (ORCPT ); Tue, 14 Apr 2020 21:18:02 -0400 In-Reply-To: <20200412004155.1a8f4e081b4e03ef5903abb5@linux-foundation.org> Sender: mm-commits-owner@vger.kernel.org List-Id: mm-commits@vger.kernel.org To: darrick.wong@oracle.com, dchinner@redhat.com, ebiggers@google.com, gaoxiang25@huawei.com, hch@lst.de, jaegeuk@kernel.org, jhubbard@nvidia.com, joseph.qi@linux.alibaba.com, junxiao.bi@oracle.com, mhocko@suse.com, mm-commits@vger.kernel.org, william.kucharski@oracle.com, willy@infradead.org, xiyou.wangcong@gmail.com, yuchao0@huawei.com, ziy@nvidia.com The patch titled Subject: mm: add page_cache_readahead_unbounded has been added to the -mm tree. Its filename is mm-add-page_cache_readahead_unbounded.patch This patch should soon appear at http://ozlabs.org/~akpm/mmots/broken-out/mm-add-page_cache_readahead_unbounded.patch and later at http://ozlabs.org/~akpm/mmotm/broken-out/mm-add-page_cache_readahead_unbounded.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next and is updated there every 3-4 working days ------------------------------------------------------ From: "Matthew Wilcox (Oracle)" Subject: mm: add page_cache_readahead_unbounded ext4 and f2fs have duplicated the guts of the readahead code so they can read past i_size. Instead, separate out the guts of the readahead code so they can call it directly. Link: http://lkml.kernel.org/r/20200414150233.24495-14-willy@infradead.org Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: Christoph Hellwig Reviewed-by: William Kucharski Reviewed-by: Eric Biggers Tested-by: Eric Biggers Cc: Chao Yu Cc: Cong Wang Cc: Darrick J. Wong Cc: Dave Chinner Cc: Gao Xiang Cc: Jaegeuk Kim Cc: John Hubbard Cc: Joseph Qi Cc: Junxiao Bi Cc: Michal Hocko Cc: Zi Yan Signed-off-by: Andrew Morton --- fs/ext4/verity.c | 35 +------------------ fs/f2fs/data.c | 2 - fs/f2fs/f2fs.h | 3 - fs/f2fs/verity.c | 35 +------------------ include/linux/pagemap.h | 3 + mm/readahead.c | 68 ++++++++++++++++++++++++++------------ 6 files changed, 55 insertions(+), 91 deletions(-) --- a/fs/ext4/verity.c~mm-add-page_cache_readahead_unbounded +++ a/fs/ext4/verity.c @@ -342,37 +342,6 @@ static int ext4_get_verity_descriptor(st return desc_size; } -/* - * Prefetch some pages from the file's Merkle tree. - * - * This is basically a stripped-down version of __do_page_cache_readahead() - * which works on pages past i_size. - */ -static void ext4_merkle_tree_readahead(struct address_space *mapping, - pgoff_t start_index, unsigned long count) -{ - LIST_HEAD(pages); - unsigned int nr_pages = 0; - struct page *page; - pgoff_t index; - struct blk_plug plug; - - for (index = start_index; index < start_index + count; index++) { - page = xa_load(&mapping->i_pages, index); - if (!page || xa_is_value(page)) { - page = __page_cache_alloc(readahead_gfp_mask(mapping)); - if (!page) - break; - page->index = index; - list_add(&page->lru, &pages); - nr_pages++; - } - } - blk_start_plug(&plug); - ext4_mpage_readpages(mapping, &pages, NULL, nr_pages, true); - blk_finish_plug(&plug); -} - static struct page *ext4_read_merkle_tree_page(struct inode *inode, pgoff_t index, unsigned long num_ra_pages) @@ -386,8 +355,8 @@ static struct page *ext4_read_merkle_tre if (page) put_page(page); else if (num_ra_pages > 1) - ext4_merkle_tree_readahead(inode->i_mapping, index, - num_ra_pages); + page_cache_readahead_unbounded(inode->i_mapping, NULL, + index, num_ra_pages, 0); page = read_mapping_page(inode->i_mapping, index, NULL); } return page; --- a/fs/f2fs/data.c~mm-add-page_cache_readahead_unbounded +++ a/fs/f2fs/data.c @@ -2177,7 +2177,7 @@ out: * use ->readpage() or do the necessary surgery to decouple ->readpages() * from read-ahead. */ -int f2fs_mpage_readpages(struct address_space *mapping, +static int f2fs_mpage_readpages(struct address_space *mapping, struct list_head *pages, struct page *page, unsigned nr_pages, bool is_readahead) { --- a/fs/f2fs/f2fs.h~mm-add-page_cache_readahead_unbounded +++ a/fs/f2fs/f2fs.h @@ -3373,9 +3373,6 @@ int f2fs_reserve_new_block(struct dnode_ int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index); int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *from); int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index); -int f2fs_mpage_readpages(struct address_space *mapping, - struct list_head *pages, struct page *page, - unsigned nr_pages, bool is_readahead); struct page *f2fs_get_read_data_page(struct inode *inode, pgoff_t index, int op_flags, bool for_write); struct page *f2fs_find_data_page(struct inode *inode, pgoff_t index); --- a/fs/f2fs/verity.c~mm-add-page_cache_readahead_unbounded +++ a/fs/f2fs/verity.c @@ -222,37 +222,6 @@ static int f2fs_get_verity_descriptor(st return size; } -/* - * Prefetch some pages from the file's Merkle tree. - * - * This is basically a stripped-down version of __do_page_cache_readahead() - * which works on pages past i_size. - */ -static void f2fs_merkle_tree_readahead(struct address_space *mapping, - pgoff_t start_index, unsigned long count) -{ - LIST_HEAD(pages); - unsigned int nr_pages = 0; - struct page *page; - pgoff_t index; - struct blk_plug plug; - - for (index = start_index; index < start_index + count; index++) { - page = xa_load(&mapping->i_pages, index); - if (!page || xa_is_value(page)) { - page = __page_cache_alloc(readahead_gfp_mask(mapping)); - if (!page) - break; - page->index = index; - list_add(&page->lru, &pages); - nr_pages++; - } - } - blk_start_plug(&plug); - f2fs_mpage_readpages(mapping, &pages, NULL, nr_pages, true); - blk_finish_plug(&plug); -} - static struct page *f2fs_read_merkle_tree_page(struct inode *inode, pgoff_t index, unsigned long num_ra_pages) @@ -266,8 +235,8 @@ static struct page *f2fs_read_merkle_tre if (page) put_page(page); else if (num_ra_pages > 1) - f2fs_merkle_tree_readahead(inode->i_mapping, index, - num_ra_pages); + page_cache_readahead_unbounded(inode->i_mapping, NULL, + index, num_ra_pages, 0); page = read_mapping_page(inode->i_mapping, index, NULL); } return page; --- a/include/linux/pagemap.h~mm-add-page_cache_readahead_unbounded +++ a/include/linux/pagemap.h @@ -622,6 +622,9 @@ void page_cache_sync_readahead(struct ad void page_cache_async_readahead(struct address_space *, struct file_ra_state *, struct file *, struct page *, pgoff_t index, unsigned long req_count); +void page_cache_readahead_unbounded(struct address_space *, struct file *, + pgoff_t index, unsigned long nr_to_read, + unsigned long lookahead_count); /* * Like add_to_page_cache_locked, but used to add newly allocated pages: --- a/mm/readahead.c~mm-add-page_cache_readahead_unbounded +++ a/mm/readahead.c @@ -156,37 +156,34 @@ out: rac->_index++; } -/* - * __do_page_cache_readahead() actually reads a chunk of disk. It allocates - * the pages first, then submits them for I/O. This avoids the very bad - * behaviour which would occur if page allocations are causing VM writeback. - * We really don't want to intermingle reads and writes like that. +/** + * page_cache_readahead_unbounded - Start unchecked readahead. + * @mapping: File address space. + * @file: This instance of the open file; used for authentication. + * @index: First page index to read. + * @nr_to_read: The number of pages to read. + * @lookahead_size: Where to start the next readahead. + * + * This function is for filesystems to call when they want to start + * readahead beyond a file's stated i_size. This is almost certainly + * not the function you want to call. Use page_cache_async_readahead() + * or page_cache_sync_readahead() instead. + * + * Context: File is referenced by caller. Mutexes may be held by caller. + * May sleep, but will not reenter filesystem to reclaim memory. */ -void __do_page_cache_readahead(struct address_space *mapping, - struct file *filp, pgoff_t index, unsigned long nr_to_read, +void page_cache_readahead_unbounded(struct address_space *mapping, + struct file *file, pgoff_t index, unsigned long nr_to_read, unsigned long lookahead_size) { - struct inode *inode = mapping->host; LIST_HEAD(page_pool); - loff_t isize = i_size_read(inode); gfp_t gfp_mask = readahead_gfp_mask(mapping); struct readahead_control rac = { .mapping = mapping, - .file = filp, + .file = file, ._index = index, }; unsigned long i; - pgoff_t end_index; /* The last page we want to read */ - - if (isize == 0) - return; - - end_index = (isize - 1) >> PAGE_SHIFT; - if (index > end_index) - return; - /* Don't read past the page containing the last byte of the file */ - if (nr_to_read > end_index - index) - nr_to_read = end_index - index + 1; /* * Preallocate as many pages as we will need. @@ -230,6 +227,35 @@ void __do_page_cache_readahead(struct ad */ read_pages(&rac, &page_pool, false); } +EXPORT_SYMBOL_GPL(page_cache_readahead_unbounded); + +/* + * __do_page_cache_readahead() actually reads a chunk of disk. It allocates + * the pages first, then submits them for I/O. This avoids the very bad + * behaviour which would occur if page allocations are causing VM writeback. + * We really don't want to intermingle reads and writes like that. + */ +void __do_page_cache_readahead(struct address_space *mapping, + struct file *file, pgoff_t index, unsigned long nr_to_read, + unsigned long lookahead_size) +{ + struct inode *inode = mapping->host; + loff_t isize = i_size_read(inode); + pgoff_t end_index; /* The last page we want to read */ + + if (isize == 0) + return; + + end_index = (isize - 1) >> PAGE_SHIFT; + if (index > end_index) + return; + /* Don't read past the page containing the last byte of the file */ + if (nr_to_read > end_index - index) + nr_to_read = end_index - index + 1; + + page_cache_readahead_unbounded(mapping, file, index, nr_to_read, + lookahead_size); +} /* * Chunk the readahead into 2 megabyte units, so that we don't pin too much _ Patches currently in -mm which might be from willy@infradead.org are mm-move-readahead-prototypes-from-mmh.patch mm-return-void-from-various-readahead-functions.patch mm-ignore-return-value-of-readpages.patch mm-move-readahead-nr_pages-check-into-read_pages.patch mm-add-new-readahead_control-api.patch mm-use-readahead_control-to-pass-arguments.patch mm-rename-various-offset-parameters-to-index.patch mm-rename-readahead-loop-variable-to-i.patch mm-remove-page_offset-from-readahead-loop.patch mm-put-readahead-pages-in-cache-earlier.patch mm-add-readahead-address-space-operation.patch mm-move-end_index-check-out-of-readahead-loop.patch mm-add-page_cache_readahead_unbounded.patch mm-document-why-we-dont-set-pagereadahead.patch mm-use-memalloc_nofs_save-in-readahead-path.patch fs-convert-mpage_readpages-to-mpage_readahead.patch btrfs-convert-from-readpages-to-readahead.patch erofs-convert-uncompressed-files-from-readpages-to-readahead.patch erofs-convert-compressed-files-from-readpages-to-readahead.patch ext4-convert-from-readpages-to-readahead.patch ext4-pass-the-inode-to-ext4_mpage_readpages.patch f2fs-convert-from-readpages-to-readahead.patch f2fs-pass-the-inode-to-f2fs_mpage_readpages.patch fuse-convert-from-readpages-to-readahead.patch iomap-convert-from-readpages-to-readahead.patch