c/mirror_zfs
1
0
Fork 0
mirror of https://git.proxmox.com/git/mirror_zfs.git synced 2024-12-29 12:29:35 +03:00
Code Issues Packages Projects Releases Wiki Activity
a10e552b99
mirror_zfs/include/os/linux/zfs/sys/zpl.h
Brian Atkinson a10e552b99
Adding Direct IO Support 
Adding O_DIRECT support to ZFS to bypass the ARC for writes/reads.

O_DIRECT support in ZFS will always ensure there is coherency between
buffered and O_DIRECT IO requests. This ensures that all IO requests,
whether buffered or direct, will see the same file contents at all
times. Just as in other FS's , O_DIRECT does not imply O_SYNC. While
data is written directly to VDEV disks, metadata will not be synced
until the associated  TXG is synced.
For both O_DIRECT read and write request the offset and request sizes,
at a minimum, must be PAGE_SIZE aligned. In the event they are not,
then EINVAL is returned unless the direct property is set to always (see
below).

For O_DIRECT writes:
The request also must be block aligned (recordsize) or the write
request will take the normal (buffered) write path. In the event that
request is block aligned and a cached copy of the buffer in the ARC,
then it will be discarded from the ARC forcing all further reads to
retrieve the data from disk.

For O_DIRECT reads:
The only alignment restrictions are PAGE_SIZE alignment. In the event
that the requested data is in buffered (in the ARC) it will just be
copied from the ARC into the user buffer.

For both O_DIRECT writes and reads the O_DIRECT flag will be ignored in
the event that file contents are mmap'ed. In this case, all requests
that are at least PAGE_SIZE aligned will just fall back to the buffered
paths. If the request however is not PAGE_SIZE aligned, EINVAL will
be returned as always regardless if the file's contents are mmap'ed.

Since O_DIRECT writes go through the normal ZIO pipeline, the
following operations are supported just as with normal buffered writes:
Checksum
Compression
Encryption
Erasure Coding
There is one caveat for the data integrity of O_DIRECT writes that is
distinct for each of the OS's supported by ZFS.
FreeBSD - FreeBSD is able to place user pages under write protection so
          any data in the user buffers and written directly down to the
	  VDEV disks is guaranteed to not change. There is no concern
	  with data integrity and O_DIRECT writes.
Linux - Linux is not able to place anonymous user pages under write
        protection. Because of this, if the user decides to manipulate
	the page contents while the write operation is occurring, data
	integrity can not be guaranteed. However, there is a module
	parameter `zfs_vdev_direct_write_verify` that controls the
	if a O_DIRECT writes that can occur to a top-level VDEV before
	a checksum verify is run before the contents of the I/O buffer
        are committed to disk. In the event of a checksum verification
	failure the write will return EIO. The number of O_DIRECT write
	checksum verification errors can be observed by doing
	`zpool status -d`, which will list all verification errors that
	have occurred on a top-level VDEV. Along with `zpool status`, a
	ZED event will be issues as `dio_verify` when a checksum
	verification error occurs.

ZVOLs and dedup is not currently supported with Direct I/O.

A new dataset property `direct` has been added with the following 3
allowable values:
disabled - Accepts O_DIRECT flag, but silently ignores it and treats
	   the request as a buffered IO request.
standard - Follows the alignment restrictions  outlined above for
	   write/read IO requests when the O_DIRECT flag is used.
always   - Treats every write/read IO request as though it passed
           O_DIRECT and will do O_DIRECT if the alignment restrictions
	   are met otherwise will redirect through the ARC. This
	   property will not allow a request to fail.

There is also a module parameter zfs_dio_enabled that can be used to
force all reads and writes through the ARC. By setting this module
parameter to 0, it mimics as if the  direct dataset property is set to
disabled.

Reviewed-by: Brian Behlendorf <behlendorf@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tony Hutter <hutter2@llnl.gov>
Signed-off-by: Brian Atkinson <batkinson@lanl.gov>
Co-authored-by: Mark Maybee <mark.maybee@delphix.com>
Co-authored-by: Matt Macy <mmacy@FreeBSD.org>
Co-authored-by: Brian Behlendorf <behlendorf@llnl.gov>
Closes #10018
2024-09-14 13:47:59 -07:00

297 lines
9.1 KiB
C
Raw Blame History

/*
* CDDL HEADER START
*
* The contents of this file are subject to the terms of the
* Common Development and Distribution License (the "License").
* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
* or https://opensource.org/licenses/CDDL-1.0.
* See the License for the specific language governing permissions
* and limitations under the License.
*
* When distributing Covered Code, include this CDDL HEADER in each
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
* If applicable, add the following below this CDDL HEADER, with the
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*/
/*
* Copyright (c) 2011, Lawrence Livermore National Security, LLC.
*/
#ifndef _SYS_ZPL_H
#define _SYS_ZPL_H
#include <sys/mntent.h>
#include <sys/vfs.h>
#include <linux/aio.h>
#include <linux/dcache_compat.h>
#include <linux/exportfs.h>
#include <linux/falloc.h>
#include <linux/parser.h>
#include <linux/vfs_compat.h>
#include <linux/writeback.h>
#include <linux/xattr_compat.h>
/* zpl_inode.c */
extern void zpl_vap_init(vattr_t *vap, struct inode *dir,
umode_t mode, cred_t *cr, zidmap_t *mnt_ns);
extern const struct inode_operations zpl_inode_operations;
#ifdef HAVE_RENAME2_OPERATIONS_WRAPPER
extern const struct inode_operations_wrapper zpl_dir_inode_operations;
#else
extern const struct inode_operations zpl_dir_inode_operations;
#endif
extern const struct inode_operations zpl_symlink_inode_operations;
extern const struct inode_operations zpl_special_inode_operations;
/* zpl_file.c */
extern const struct address_space_operations zpl_address_space_operations;
#ifdef HAVE_VFS_FILE_OPERATIONS_EXTEND
extern const struct file_operations_extend zpl_file_operations;
#else
extern const struct file_operations zpl_file_operations;
#endif
extern const struct file_operations zpl_dir_file_operations;
/* zpl_super.c */
extern void zpl_prune_sb(uint64_t nr_to_scan, void *arg);
extern const struct super_operations zpl_super_operations;
extern const struct export_operations zpl_export_operations;
extern struct file_system_type zpl_fs_type;
/* zpl_xattr.c */
extern ssize_t zpl_xattr_list(struct dentry *dentry, char *buf, size_t size);
extern int zpl_xattr_security_init(struct inode *ip, struct inode *dip,
const struct qstr *qstr);
#if defined(CONFIG_FS_POSIX_ACL)
#if defined(HAVE_SET_ACL)
#if defined(HAVE_SET_ACL_IDMAP_DENTRY)
extern int zpl_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
struct posix_acl *acl, int type);
#elif defined(HAVE_SET_ACL_USERNS)
extern int zpl_set_acl(struct user_namespace *userns, struct inode *ip,
struct posix_acl *acl, int type);
#elif defined(HAVE_SET_ACL_USERNS_DENTRY_ARG2)
extern int zpl_set_acl(struct user_namespace *userns, struct dentry *dentry,
struct posix_acl *acl, int type);
#else
extern int zpl_set_acl(struct inode *ip, struct posix_acl *acl, int type);
#endif /* HAVE_SET_ACL_USERNS */
#endif /* HAVE_SET_ACL */
#if defined(HAVE_GET_ACL_RCU) || defined(HAVE_GET_INODE_ACL)
extern struct posix_acl *zpl_get_acl(struct inode *ip, int type, bool rcu);
#elif defined(HAVE_GET_ACL)
extern struct posix_acl *zpl_get_acl(struct inode *ip, int type);
#endif
extern int zpl_init_acl(struct inode *ip, struct inode *dir);
extern int zpl_chmod_acl(struct inode *ip);
#else
static inline int
zpl_init_acl(struct inode *ip, struct inode *dir)
{
return (0);
}
static inline int
zpl_chmod_acl(struct inode *ip)
{
return (0);
}
#endif /* CONFIG_FS_POSIX_ACL */
extern xattr_handler_t *zpl_xattr_handlers[];
/* zpl_ctldir.c */
extern const struct file_operations zpl_fops_root;
extern const struct inode_operations zpl_ops_root;
extern const struct file_operations zpl_fops_snapdir;
extern const struct inode_operations zpl_ops_snapdir;
extern const struct file_operations zpl_fops_shares;
extern const struct inode_operations zpl_ops_shares;
#if defined(HAVE_VFS_ITERATE) || defined(HAVE_VFS_ITERATE_SHARED)
#define ZPL_DIR_CONTEXT_INIT(_dirent, _actor, _pos) { \
.actor = _actor, \
.pos = _pos, \
}
typedef struct dir_context zpl_dir_context_t;
#define zpl_dir_emit dir_emit
#define zpl_dir_emit_dot dir_emit_dot
#define zpl_dir_emit_dotdot dir_emit_dotdot
#define zpl_dir_emit_dots dir_emit_dots
#else
typedef struct zpl_dir_context {
void *dirent;
const filldir_t actor;
loff_t pos;
} zpl_dir_context_t;
#define ZPL_DIR_CONTEXT_INIT(_dirent, _actor, _pos) { \
.dirent = _dirent, \
.actor = _actor, \
.pos = _pos, \
}
static inline bool
zpl_dir_emit(zpl_dir_context_t *ctx, const char *name, int namelen,
uint64_t ino, unsigned type)
{
return (!ctx->actor(ctx->dirent, name, namelen, ctx->pos, ino, type));
}
static inline bool
zpl_dir_emit_dot(struct file *file, zpl_dir_context_t *ctx)
{
return (ctx->actor(ctx->dirent, ".", 1, ctx->pos,
file_inode(file)->i_ino, DT_DIR) == 0);
}
static inline bool
zpl_dir_emit_dotdot(struct file *file, zpl_dir_context_t *ctx)
{
return (ctx->actor(ctx->dirent, "..", 2, ctx->pos,
parent_ino(file_dentry(file)), DT_DIR) == 0);
}
static inline bool
zpl_dir_emit_dots(struct file *file, zpl_dir_context_t *ctx)
{
if (ctx->pos == 0) {
if (!zpl_dir_emit_dot(file, ctx))
return (false);
ctx->pos = 1;
}
if (ctx->pos == 1) {
if (!zpl_dir_emit_dotdot(file, ctx))
return (false);
ctx->pos = 2;
}
return (true);
}
#endif /* HAVE_VFS_ITERATE */
/* zpl_file_range.c */
/* handlers for file_operations of the same name */
extern ssize_t zpl_copy_file_range(struct file *src_file, loff_t src_off,
struct file *dst_file, loff_t dst_off, size_t len, unsigned int flags);
extern loff_t zpl_remap_file_range(struct file *src_file, loff_t src_off,
struct file *dst_file, loff_t dst_off, loff_t len, unsigned int flags);
extern int zpl_clone_file_range(struct file *src_file, loff_t src_off,
struct file *dst_file, loff_t dst_off, uint64_t len);
extern int zpl_dedupe_file_range(struct file *src_file, loff_t src_off,
struct file *dst_file, loff_t dst_off, uint64_t len);
/* compat for FICLONE/FICLONERANGE/FIDEDUPERANGE ioctls */
typedef struct {
int64_t fcr_src_fd;
uint64_t fcr_src_offset;
uint64_t fcr_src_length;
uint64_t fcr_dest_offset;
} zfs_ioc_compat_file_clone_range_t;
typedef struct {
int64_t fdri_dest_fd;
uint64_t fdri_dest_offset;
uint64_t fdri_bytes_deduped;
int32_t fdri_status;
uint32_t fdri_reserved;
} zfs_ioc_compat_dedupe_range_info_t;
typedef struct {
uint64_t fdr_src_offset;
uint64_t fdr_src_length;
uint16_t fdr_dest_count;
uint16_t fdr_reserved1;
uint32_t fdr_reserved2;
zfs_ioc_compat_dedupe_range_info_t fdr_info[];
} zfs_ioc_compat_dedupe_range_t;
#define ZFS_IOC_COMPAT_FICLONE _IOW(0x94, 9, int)
#define ZFS_IOC_COMPAT_FICLONERANGE \
_IOW(0x94, 13, zfs_ioc_compat_file_clone_range_t)
#define ZFS_IOC_COMPAT_FIDEDUPERANGE \
_IOWR(0x94, 54, zfs_ioc_compat_dedupe_range_t)
extern long zpl_ioctl_ficlone(struct file *filp, void *arg);
extern long zpl_ioctl_ficlonerange(struct file *filp, void *arg);
extern long zpl_ioctl_fideduperange(struct file *filp, void *arg);
#if defined(HAVE_INODE_TIMESTAMP_TRUNCATE)
#define zpl_inode_timestamp_truncate(ts, ip) timestamp_truncate(ts, ip)
#elif defined(HAVE_INODE_TIMESPEC64_TIMES)
#define zpl_inode_timestamp_truncate(ts, ip) \
timespec64_trunc(ts, (ip)->i_sb->s_time_gran)
#else
#define zpl_inode_timestamp_truncate(ts, ip) \
timespec_trunc(ts, (ip)->i_sb->s_time_gran)
#endif
#if defined(HAVE_INODE_OWNER_OR_CAPABLE)
#define zpl_inode_owner_or_capable(ns, ip) inode_owner_or_capable(ip)
#elif defined(HAVE_INODE_OWNER_OR_CAPABLE_USERNS)
#define zpl_inode_owner_or_capable(ns, ip) inode_owner_or_capable(ns, ip)
#elif defined(HAVE_INODE_OWNER_OR_CAPABLE_IDMAP)
#define zpl_inode_owner_or_capable(idmap, ip) inode_owner_or_capable(idmap, ip)
#else
#error "Unsupported kernel"
#endif
#if defined(HAVE_SETATTR_PREPARE_USERNS) || defined(HAVE_SETATTR_PREPARE_IDMAP)
#define zpl_setattr_prepare(ns, dentry, ia) setattr_prepare(ns, dentry, ia)
#else
/*
* Use kernel-provided version, or our own from
* linux/vfs_compat.h
*/
#define zpl_setattr_prepare(ns, dentry, ia) setattr_prepare(dentry, ia)
#endif
#ifdef HAVE_INODE_GET_CTIME
#define zpl_inode_get_ctime(ip) inode_get_ctime(ip)
#else
#define zpl_inode_get_ctime(ip) (ip->i_ctime)
#endif
#ifdef HAVE_INODE_SET_CTIME_TO_TS
#define zpl_inode_set_ctime_to_ts(ip, ts) inode_set_ctime_to_ts(ip, ts)
#else
#define zpl_inode_set_ctime_to_ts(ip, ts) (ip->i_ctime = ts)
#endif
#ifdef HAVE_INODE_GET_ATIME
#define zpl_inode_get_atime(ip) inode_get_atime(ip)
#else
#define zpl_inode_get_atime(ip) (ip->i_atime)
#endif
#ifdef HAVE_INODE_SET_ATIME_TO_TS
#define zpl_inode_set_atime_to_ts(ip, ts) inode_set_atime_to_ts(ip, ts)
#else
#define zpl_inode_set_atime_to_ts(ip, ts) (ip->i_atime = ts)
#endif
#ifdef HAVE_INODE_GET_MTIME
#define zpl_inode_get_mtime(ip) inode_get_mtime(ip)
#else
#define zpl_inode_get_mtime(ip) (ip->i_mtime)
#endif
#ifdef HAVE_INODE_SET_MTIME_TO_TS
#define zpl_inode_set_mtime_to_ts(ip, ts) inode_set_mtime_to_ts(ip, ts)
#else
#define zpl_inode_set_mtime_to_ts(ip, ts) (ip->i_mtime = ts)
#endif
#endif /* _SYS_ZPL_H */
Reference in New Issue View Git Blame Copy Permalink