From patchwork Mon Jan 28 21:24:34 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Dennis Zhou X-Patchwork-Id: 10784817 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 B38A613B5 for ; Mon, 28 Jan 2019 21:25:28 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id A15D52B279 for ; Mon, 28 Jan 2019 21:25:28 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 953982B288; Mon, 28 Jan 2019 21:25:28 +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=-7.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, 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 D80C02B279 for ; Mon, 28 Jan 2019 21:25:27 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728486AbfA1VZX (ORCPT ); Mon, 28 Jan 2019 16:25:23 -0500 Received: from mail-yw1-f67.google.com ([209.85.161.67]:37090 "EHLO mail-yw1-f67.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728316AbfA1VY4 (ORCPT ); Mon, 28 Jan 2019 16:24:56 -0500 Received: by mail-yw1-f67.google.com with SMTP id h193so7358275ywc.4; Mon, 28 Jan 2019 13:24:56 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=OpMPEOdbVcWt4t/2TFv/NEECu9N5O2HbNXitx/LNP7k=; b=YDiERm27/Vo14tpG3VYqNdE0FXV3tw0SEehD2b0hzu6ExoNgbgXFcUx8v+3KSaVPJC eUb79wZ7TKOBrHbNj2ZQnjbZD2VojuKIUBLfwjI3ZVRwetRz5hq5mLoxjV8q6/qC+LA9 L2PjFM1p86zz/SYS8PvYzDiMTFuHfNUgu7DNRfXgWHAi3GngZd7thhiGk32+8trRZ2YS f/ZruV4pyYVA7/IafCs2uvSPz4gwuIuuPX2owLx/AGKtaa3B3GD5eJQzQ/Ueute5gB1R DwuZUoHyJ63AG+M8HEYwtLW9Q98OMqgSjfl0vev0Bt3PBETW7ttf3nthuHRd48b0cIND j4IQ== X-Gm-Message-State: AJcUukfm56ZhBFqLriSOasCCNbmRuoMivuDj2ZSYkrrN7UUcX5Xz8UZv qNAqSW5wI4f5e4S15HQ+GHU= X-Google-Smtp-Source: ALg8bN4TlsfktrxYLCaEyhNx98IE9Q0yEcHlTEAGh9XyZj0y1KqBrakMwg1rKEmOTKXc3ZUIW/WZeQ== X-Received: by 2002:a81:378b:: with SMTP id e133mr22220086ywa.79.1548710695992; Mon, 28 Jan 2019 13:24:55 -0800 (PST) Received: from dennisz-mbp.thefacebook.com ([199.201.65.135]) by smtp.gmail.com with ESMTPSA id a7sm24170415ywb.18.2019.01.28.13.24.54 (version=TLS1_2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Mon, 28 Jan 2019 13:24:55 -0800 (PST) From: Dennis Zhou To: David Sterba , Josef Bacik , Chris Mason , Omar Sandoval , Nick Terrell Cc: kernel-team@fb.com, linux-btrfs@vger.kernel.org, linux-kernel@vger.kernel.org, Dennis Zhou Subject: [PATCH 08/11] btrfs: plumb level through the compression interface Date: Mon, 28 Jan 2019 16:24:34 -0500 Message-Id: <20190128212437.11597-9-dennis@kernel.org> X-Mailer: git-send-email 2.13.5 In-Reply-To: <20190128212437.11597-1-dennis@kernel.org> References: <20190128212437.11597-1-dennis@kernel.org> Sender: linux-btrfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-btrfs@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Zlib compression supports multiple levels, but doesn't require changing in how a workspace itself is created and managed. Zstd introduces a different memory requirement such that higher levels of compression require more memory. This requires changes in how the alloc()/get() methods work for zstd. This pach plumbs compression level through the interface as a parameter in preparation for zstd compression levels. This gives the compression types opportunity to create/manage based on the compression level. Signed-off-by: Dennis Zhou Reviewed-by: Nikolay Borisov Reviewed-by: Josef Bacik --- fs/btrfs/compression.c | 31 ++++++++++++++++--------------- fs/btrfs/compression.h | 7 ++++--- fs/btrfs/lzo.c | 6 +++--- fs/btrfs/zlib.c | 7 ++++--- fs/btrfs/zstd.c | 6 +++--- 5 files changed, 30 insertions(+), 27 deletions(-) diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index ab694760ffdb..e509071eaa69 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -744,9 +744,9 @@ static void heuristic_cleanup_workspace_manager(void) btrfs_cleanup_workspace_manager(&heuristic_wsm); } -static struct list_head *heuristic_get_workspace(void) +static struct list_head *heuristic_get_workspace(unsigned int level) { - return btrfs_get_workspace(&heuristic_wsm); + return btrfs_get_workspace(&heuristic_wsm, level); } static void heuristic_put_workspace(struct list_head *ws) @@ -766,7 +766,7 @@ static void free_heuristic_ws(struct list_head *ws) kfree(workspace); } -static struct list_head *alloc_heuristic_ws(void) +static struct list_head *alloc_heuristic_ws(unsigned int level) { struct heuristic_ws *ws; @@ -825,7 +825,7 @@ void btrfs_init_workspace_manager(struct workspace_manager *wsm, * Preallocate one workspace for each compression type so * we can guarantee forward progress in the worst case */ - workspace = wsm->ops->alloc_workspace(); + workspace = wsm->ops->alloc_workspace(0); if (IS_ERR(workspace)) { pr_warn("BTRFS: cannot preallocate compression workspace, will try later\n"); } else { @@ -853,7 +853,8 @@ void btrfs_cleanup_workspace_manager(struct workspace_manager *wsman) * Preallocation makes a forward progress guarantees and we do not return * errors. */ -struct list_head *btrfs_get_workspace(struct workspace_manager *wsm) +struct list_head *btrfs_get_workspace(struct workspace_manager *wsm, + unsigned int level) { struct list_head *workspace; int cpus = num_online_cpus(); @@ -899,7 +900,7 @@ struct list_head *btrfs_get_workspace(struct workspace_manager *wsm) * context of btrfs_compress_bio/btrfs_compress_pages */ nofs_flag = memalloc_nofs_save(); - workspace = wsm->ops->alloc_workspace(); + workspace = wsm->ops->alloc_workspace(level); memalloc_nofs_restore(nofs_flag); if (IS_ERR(workspace)) { @@ -930,9 +931,9 @@ struct list_head *btrfs_get_workspace(struct workspace_manager *wsm) return workspace; } -static struct list_head *get_workspace(int type) +static struct list_head *get_workspace(int type, int level) { - return btrfs_compress_op[type]->get_workspace(); + return btrfs_compress_op[type]->get_workspace(level); } /* @@ -1003,12 +1004,13 @@ int btrfs_compress_pages(unsigned int type_level, struct address_space *mapping, unsigned long *total_out) { int type = BTRFS_COMPRESS_TYPE(type_level); + int level = BTRFS_COMPRESS_LEVEL(type_level); struct list_head *workspace; int ret; - workspace = get_workspace(type); + workspace = get_workspace(type, level); - btrfs_compress_op[type]->set_level(workspace, type_level); + btrfs_compress_op[type]->set_level(workspace, level); ret = btrfs_compress_op[type]->compress_pages(workspace, mapping, start, pages, out_pages, @@ -1037,7 +1039,7 @@ static int btrfs_decompress_bio(struct compressed_bio *cb) int ret; int type = cb->compress_type; - workspace = get_workspace(type); + workspace = get_workspace(type, 0); ret = btrfs_compress_op[type]->decompress_bio(workspace, cb); put_workspace(type, workspace); @@ -1055,13 +1057,12 @@ int btrfs_decompress(int type, unsigned char *data_in, struct page *dest_page, struct list_head *workspace; int ret; - workspace = get_workspace(type); - + workspace = get_workspace(type, 0); ret = btrfs_compress_op[type]->decompress(workspace, data_in, dest_page, start_byte, srclen, destlen); - put_workspace(type, workspace); + return ret; } @@ -1489,7 +1490,7 @@ static void heuristic_collect_sample(struct inode *inode, u64 start, u64 end, */ int btrfs_compress_heuristic(struct inode *inode, u64 start, u64 end) { - struct list_head *ws_list = get_workspace(0); + struct list_head *ws_list = get_workspace(0, 0); struct heuristic_ws *ws; u32 i; u8 byte; diff --git a/fs/btrfs/compression.h b/fs/btrfs/compression.h index 05342ad081d6..e3627139bc5c 100644 --- a/fs/btrfs/compression.h +++ b/fs/btrfs/compression.h @@ -114,7 +114,8 @@ struct workspace_manager { void btrfs_init_workspace_manager(struct workspace_manager *wsm, const struct btrfs_compress_op *ops); -struct list_head *btrfs_get_workspace(struct workspace_manager *wsm); +struct list_head *btrfs_get_workspace(struct workspace_manager *wsm, + unsigned int level); void btrfs_put_workspace(struct workspace_manager *wsm, struct list_head *ws); void btrfs_cleanup_workspace_manager(struct workspace_manager *wsm); @@ -123,11 +124,11 @@ struct btrfs_compress_op { void (*cleanup_workspace_manager)(void); - struct list_head *(*get_workspace)(void); + struct list_head *(*get_workspace)(unsigned int level); void (*put_workspace)(struct list_head *ws); - struct list_head *(*alloc_workspace)(void); + struct list_head *(*alloc_workspace)(unsigned int level); void (*free_workspace)(struct list_head *workspace); diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c index f0837b2c8e94..f132af45a924 100644 --- a/fs/btrfs/lzo.c +++ b/fs/btrfs/lzo.c @@ -73,9 +73,9 @@ static void lzo_cleanup_workspace_manager(void) btrfs_cleanup_workspace_manager(&wsm); } -static struct list_head *lzo_get_workspace(void) +static struct list_head *lzo_get_workspace(unsigned int level) { - return btrfs_get_workspace(&wsm); + return btrfs_get_workspace(&wsm, level); } static void lzo_put_workspace(struct list_head *ws) @@ -93,7 +93,7 @@ static void lzo_free_workspace(struct list_head *ws) kfree(workspace); } -static struct list_head *lzo_alloc_workspace(void) +static struct list_head *lzo_alloc_workspace(unsigned int level) { struct workspace *workspace; diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c index 04687bf692e3..e2173d0c4fd3 100644 --- a/fs/btrfs/zlib.c +++ b/fs/btrfs/zlib.c @@ -39,9 +39,9 @@ static void zlib_cleanup_workspace_manager(void) btrfs_cleanup_workspace_manager(&wsm); } -static struct list_head *zlib_get_workspace(void) +static struct list_head *zlib_get_workspace(unsigned int level) { - return btrfs_get_workspace(&wsm); + return btrfs_get_workspace(&wsm, level); } static void zlib_put_workspace(struct list_head *ws) @@ -58,7 +58,7 @@ static void zlib_free_workspace(struct list_head *ws) kfree(workspace); } -static struct list_head *zlib_alloc_workspace(void) +static struct list_head *zlib_alloc_workspace(unsigned int level) { struct workspace *workspace; int workspacesize; @@ -71,6 +71,7 @@ static struct list_head *zlib_alloc_workspace(void) zlib_inflate_workspacesize()); workspace->strm.workspace = kvmalloc(workspacesize, GFP_KERNEL); workspace->buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + workspace->level = level; if (!workspace->strm.workspace || !workspace->buf) goto fail; diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c index b06eaf171be7..404101864220 100644 --- a/fs/btrfs/zstd.c +++ b/fs/btrfs/zstd.c @@ -53,9 +53,9 @@ static void zstd_cleanup_workspace_manager(void) btrfs_cleanup_workspace_manager(&wsm); } -static struct list_head *zstd_get_workspace(void) +static struct list_head *zstd_get_workspace(unsigned int level) { - return btrfs_get_workspace(&wsm); + return btrfs_get_workspace(&wsm, level); } static void zstd_put_workspace(struct list_head *ws) @@ -72,7 +72,7 @@ static void zstd_free_workspace(struct list_head *ws) kfree(workspace); } -static struct list_head *zstd_alloc_workspace(void) +static struct list_head *zstd_alloc_workspace(unsigned int level) { ZSTD_parameters params = zstd_get_btrfs_parameters(ZSTD_BTRFS_MAX_INPUT);