From patchwork Fri Apr 1 06:35:02 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Qu Wenruo X-Patchwork-Id: 8720441 Return-Path: X-Original-To: patchwork-linux-btrfs@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork1.web.kernel.org (Postfix) with ESMTP id 12AC69F36E for ; Fri, 1 Apr 2016 06:35:47 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id BACD0203AE for ; Fri, 1 Apr 2016 06:35:45 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 6286D203AD for ; Fri, 1 Apr 2016 06:35:44 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758312AbcDAGfi (ORCPT ); Fri, 1 Apr 2016 02:35:38 -0400 Received: from cn.fujitsu.com ([222.73.24.84]:62526 "EHLO song.cn.fujitsu.com" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1758254AbcDAGfh (ORCPT ); Fri, 1 Apr 2016 02:35:37 -0400 X-IronPort-AV: E=Sophos;i="5.20,367,1444665600"; d="scan'208";a="419980" Received: from unknown (HELO cn.fujitsu.com) ([10.167.250.3]) by song.cn.fujitsu.com with ESMTP; 01 Apr 2016 14:35:22 +0800 Received: from localhost.localdomain (unknown [10.167.226.34]) by cn.fujitsu.com (Postfix) with ESMTP id 8266C405640B; Fri, 1 Apr 2016 14:35:19 +0800 (CST) From: Qu Wenruo To: linux-btrfs@vger.kernel.org Cc: Wang Xiaoguang Subject: [PATCH v10 11/21] btrfs: dedupe: Add ioctl for inband dedupelication Date: Fri, 1 Apr 2016 14:35:02 +0800 Message-Id: <1459492512-31435-12-git-send-email-quwenruo@cn.fujitsu.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1459492512-31435-1-git-send-email-quwenruo@cn.fujitsu.com> References: <1459492512-31435-1-git-send-email-quwenruo@cn.fujitsu.com> MIME-Version: 1.0 X-yoursite-MailScanner-ID: 8266C405640B.AD859 X-yoursite-MailScanner: Found to be clean X-yoursite-MailScanner-From: quwenruo@cn.fujitsu.com X-Spam-Status: No, score=-7.9 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_HI, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=ham version=3.3.1 Sender: linux-btrfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-btrfs@vger.kernel.org X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP From: Wang Xiaoguang Add ioctl interface for inband dedupelication, which includes: 1) enable 2) disable 3) status And a pseudo RO compat flag, to imply that btrfs now supports inband dedup. However we don't add any ondisk format change, it's just a pseudo RO compat flag. All these ioctl interface are state-less, which means caller don't need to bother previous dedupe state before calling them, and only need to care the final desired state. For example, if user want to enable dedupe with specified block size and limit, just fill the ioctl structure and call enable ioctl. No need to check if dedupe is already running. These ioctls will handle things like re-configure or disable quite well. Signed-off-by: Qu Wenruo Signed-off-by: Wang Xiaoguang --- fs/btrfs/ctree.h | 1 + fs/btrfs/dedupe.c | 48 ++++++++++++++++++++++++++++++++ fs/btrfs/dedupe.h | 15 ++++++++++ fs/btrfs/disk-io.c | 3 ++ fs/btrfs/ioctl.c | 68 ++++++++++++++++++++++++++++++++++++++++++++++ fs/btrfs/sysfs.c | 2 ++ include/uapi/linux/btrfs.h | 23 ++++++++++++++++ 7 files changed, 160 insertions(+) diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 022ab61..85044bf 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -508,6 +508,7 @@ struct btrfs_super_block { * ones specified below then we will fail to mount */ #define BTRFS_FEATURE_COMPAT_RO_FREE_SPACE_TREE (1ULL << 0) +#define BTRFS_FEATURE_COMPAT_RO_DEDUPE (1ULL << 1) #define BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF (1ULL << 0) #define BTRFS_FEATURE_INCOMPAT_DEFAULT_SUBVOL (1ULL << 1) diff --git a/fs/btrfs/dedupe.c b/fs/btrfs/dedupe.c index bdaea3a..cfb7fea 100644 --- a/fs/btrfs/dedupe.c +++ b/fs/btrfs/dedupe.c @@ -41,6 +41,33 @@ static inline struct inmem_hash *inmem_alloc_hash(u16 type) GFP_NOFS); } +void btrfs_dedupe_status(struct btrfs_fs_info *fs_info, + struct btrfs_ioctl_dedupe_args *dargs) +{ + struct btrfs_dedupe_info *dedupe_info = fs_info->dedupe_info; + + if (!fs_info->dedupe_enabled || !dedupe_info) { + dargs->status = 0; + dargs->blocksize = 0; + dargs->backend = 0; + dargs->hash_type = 0; + dargs->limit_nr = 0; + dargs->current_nr = 0; + return; + } + mutex_lock(&dedupe_info->lock); + dargs->status = 1; + dargs->blocksize = dedupe_info->blocksize; + dargs->backend = dedupe_info->backend; + dargs->hash_type = dedupe_info->hash_type; + dargs->limit_nr = dedupe_info->limit_nr; + dargs->limit_mem = dedupe_info->limit_nr * + (sizeof(struct inmem_hash) + + btrfs_dedupe_sizes[dedupe_info->hash_type]); + dargs->current_nr = dedupe_info->current_nr; + mutex_unlock(&dedupe_info->lock); +} + static int init_dedupe_info(struct btrfs_dedupe_info **ret_info, u16 type, u16 backend, u64 blocksize, u64 limit) { @@ -371,6 +398,27 @@ static void inmem_destroy(struct btrfs_dedupe_info *dedupe_info) mutex_unlock(&dedupe_info->lock); } +int btrfs_dedupe_cleanup(struct btrfs_fs_info *fs_info) +{ + struct btrfs_dedupe_info *dedupe_info; + + fs_info->dedupe_enabled = 0; + /* same as disable */ + smp_wmb(); + dedupe_info = fs_info->dedupe_info; + fs_info->dedupe_info = NULL; + + if (!dedupe_info) + return 0; + + if (dedupe_info->backend == BTRFS_DEDUPE_BACKEND_INMEMORY) + inmem_destroy(dedupe_info); + + crypto_free_shash(dedupe_info->dedupe_driver); + kfree(dedupe_info); + return 0; +} + int btrfs_dedupe_disable(struct btrfs_fs_info *fs_info) { struct btrfs_dedupe_info *dedupe_info; diff --git a/fs/btrfs/dedupe.h b/fs/btrfs/dedupe.h index e5d0d34..f5d2b45 100644 --- a/fs/btrfs/dedupe.h +++ b/fs/btrfs/dedupe.h @@ -103,6 +103,15 @@ static inline struct btrfs_dedupe_hash *btrfs_dedupe_alloc_hash(u16 type) int btrfs_dedupe_enable(struct btrfs_fs_info *fs_info, u16 type, u16 backend, u64 blocksize, u64 limit_nr, u64 limit_mem); + + /* + * Get inband dedupe info + * Since it needs to access different backends' hash size, which + * is not exported, we need such simple function. + */ +void btrfs_dedupe_status(struct btrfs_fs_info *fs_info, + struct btrfs_ioctl_dedupe_args *dargs); + /* * Disable dedupe and invalidate all its dedupe data. * Called at dedupe disable time. @@ -110,6 +119,12 @@ int btrfs_dedupe_enable(struct btrfs_fs_info *fs_info, u16 type, u16 backend, int btrfs_dedupe_disable(struct btrfs_fs_info *fs_info); /* + * Cleanup current btrfs_dedupe_info + * Called in umount time + */ +int btrfs_dedupe_cleanup(struct btrfs_fs_info *fs_info); + +/* * Calculate hash for dedup. * Caller must ensure [start, start + dedupe_bs) has valid data. */ diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 3cf4c11..ed6a6fd 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -51,6 +51,7 @@ #include "sysfs.h" #include "qgroup.h" #include "compression.h" +#include "dedupe.h" #ifdef CONFIG_X86 #include @@ -3884,6 +3885,8 @@ void close_ctree(struct btrfs_root *root) btrfs_free_qgroup_config(fs_info); + btrfs_dedupe_cleanup(fs_info); + if (percpu_counter_sum(&fs_info->delalloc_bytes)) { btrfs_info(fs_info, "at unmount delalloc count %lld", percpu_counter_sum(&fs_info->delalloc_bytes)); diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 053e677..f659ed5 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -61,6 +61,7 @@ #include "qgroup.h" #include "tree-log.h" #include "compression.h" +#include "dedupe.h" #ifdef CONFIG_64BIT /* If we have a 32-bit userspace and 64-bit kernel, then the UAPI @@ -3206,6 +3207,69 @@ ssize_t btrfs_dedupe_file_range(struct file *src_file, u64 loff, u64 olen, return olen; } +static long btrfs_ioctl_dedupe_ctl(struct btrfs_root *root, void __user *args) +{ + struct btrfs_ioctl_dedupe_args *dargs; + struct btrfs_fs_info *fs_info = root->fs_info; + int ret; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + dargs = memdup_user(args, sizeof(*dargs)); + if (IS_ERR(dargs)) { + ret = PTR_ERR(dargs); + return ret; + } + + if (dargs->cmd >= BTRFS_DEDUPE_CTL_LAST) { + ret = -EINVAL; + goto out; + } + switch (dargs->cmd) { + case BTRFS_DEDUPE_CTL_ENABLE: + mutex_lock(&fs_info->dedupe_ioctl_lock); + ret = btrfs_dedupe_enable(fs_info, dargs->hash_type, + dargs->backend, dargs->blocksize, + dargs->limit_nr, dargs->limit_mem); + mutex_unlock(&fs_info->dedupe_ioctl_lock); + if (ret < 0) + break; + + /* Also copy the result to caller for further use */ + btrfs_dedupe_status(fs_info, dargs); + if (copy_to_user(args, dargs, sizeof(*dargs))) + ret = -EFAULT; + else + ret = 0; + break; + case BTRFS_DEDUPE_CTL_DISABLE: + mutex_lock(&fs_info->dedupe_ioctl_lock); + ret = btrfs_dedupe_disable(fs_info); + mutex_unlock(&fs_info->dedupe_ioctl_lock); + break; + case BTRFS_DEDUPE_CTL_STATUS: + mutex_lock(&fs_info->dedupe_ioctl_lock); + btrfs_dedupe_status(fs_info, dargs); + mutex_unlock(&fs_info->dedupe_ioctl_lock); + if (copy_to_user(args, dargs, sizeof(*dargs))) + ret = -EFAULT; + else + ret = 0; + break; + default: + /* + * Use this return value to inform progs that kernel + * doesn't support such new command. + */ + ret = -EOPNOTSUPP; + break; + } +out: + kfree(dargs); + return ret; +} + static int clone_finish_inode_update(struct btrfs_trans_handle *trans, struct inode *inode, u64 endoff, @@ -5542,6 +5606,10 @@ long btrfs_ioctl(struct file *file, unsigned int return btrfs_ioctl_get_fslabel(file, argp); case BTRFS_IOC_SET_FSLABEL: return btrfs_ioctl_set_fslabel(file, argp); +#ifdef CONFIG_BTRFS_DEBUG + case BTRFS_IOC_DEDUPE_CTL: + return btrfs_ioctl_dedupe_ctl(root, argp); +#endif case BTRFS_IOC_GET_SUPPORTED_FEATURES: return btrfs_ioctl_get_supported_features(argp); case BTRFS_IOC_GET_FEATURES: diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c index 539e7b5..18686d1 100644 --- a/fs/btrfs/sysfs.c +++ b/fs/btrfs/sysfs.c @@ -203,6 +203,7 @@ BTRFS_FEAT_ATTR_INCOMPAT(raid56, RAID56); BTRFS_FEAT_ATTR_INCOMPAT(skinny_metadata, SKINNY_METADATA); BTRFS_FEAT_ATTR_INCOMPAT(no_holes, NO_HOLES); BTRFS_FEAT_ATTR_COMPAT_RO(free_space_tree, FREE_SPACE_TREE); +BTRFS_FEAT_ATTR_COMPAT_RO(dedupe, DEDUPE); static struct attribute *btrfs_supported_feature_attrs[] = { BTRFS_FEAT_ATTR_PTR(mixed_backref), @@ -215,6 +216,7 @@ static struct attribute *btrfs_supported_feature_attrs[] = { BTRFS_FEAT_ATTR_PTR(skinny_metadata), BTRFS_FEAT_ATTR_PTR(no_holes), BTRFS_FEAT_ATTR_PTR(free_space_tree), + BTRFS_FEAT_ATTR_PTR(dedupe), NULL }; diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h index dea8931..de48414 100644 --- a/include/uapi/linux/btrfs.h +++ b/include/uapi/linux/btrfs.h @@ -445,6 +445,27 @@ struct btrfs_ioctl_get_dev_stats { __u64 unused[128 - 2 - BTRFS_DEV_STAT_VALUES_MAX]; /* pad to 1k */ }; +/* + * de-duplication control modes + * For re-config, re-enable will handle it + */ +#define BTRFS_DEDUPE_CTL_ENABLE 1 +#define BTRFS_DEDUPE_CTL_DISABLE 2 +#define BTRFS_DEDUPE_CTL_STATUS 3 +#define BTRFS_DEDUPE_CTL_LAST 4 +struct btrfs_ioctl_dedupe_args { + __u16 cmd; /* In: command(see above macro) */ + __u64 blocksize; /* In/Out: For enable/status */ + __u64 limit_nr; /* In/Out: For enable/status */ + __u64 limit_mem; /* In/Out: For enable/status */ + __u64 current_nr; /* Out: For status output */ + __u16 backend; /* In/Out: For enable/status */ + __u16 hash_type; /* In/Out: For enable/status */ + u8 status; /* Out: For status output */ + /* pad to 512 bytes */ + u8 __unused[473]; +}; + #define BTRFS_QUOTA_CTL_ENABLE 1 #define BTRFS_QUOTA_CTL_DISABLE 2 #define BTRFS_QUOTA_CTL_RESCAN__NOTUSED 3 @@ -653,6 +674,8 @@ static inline char *btrfs_err_str(enum btrfs_err_code err_code) struct btrfs_ioctl_dev_replace_args) #define BTRFS_IOC_FILE_EXTENT_SAME _IOWR(BTRFS_IOCTL_MAGIC, 54, \ struct btrfs_ioctl_same_args) +#define BTRFS_IOC_DEDUPE_CTL _IOWR(BTRFS_IOCTL_MAGIC, 55, \ + struct btrfs_ioctl_dedupe_args) #define BTRFS_IOC_GET_FEATURES _IOR(BTRFS_IOCTL_MAGIC, 57, \ struct btrfs_ioctl_feature_flags) #define BTRFS_IOC_SET_FEATURES _IOW(BTRFS_IOCTL_MAGIC, 57, \