OpenCloudOS-Kernel/fs/9p/vfs_inode_dotl.c

1004 lines
24 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* linux/fs/9p/vfs_inode_dotl.c
*
* This file contains vfs inode ops for the 9P2000.L protocol.
*
* Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com>
* Copyright (C) 2002 by Ron Minnich <rminnich@lanl.gov>
*/
#include <linux/module.h>
#include <linux/errno.h>
#include <linux/fs.h>
#include <linux/file.h>
#include <linux/pagemap.h>
#include <linux/stat.h>
#include <linux/string.h>
#include <linux/inet.h>
#include <linux/namei.h>
#include <linux/idr.h>
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/xattr.h>
#include <linux/posix_acl.h>
#include <net/9p/9p.h>
#include <net/9p/client.h>
#include "v9fs.h"
#include "v9fs_vfs.h"
#include "fid.h"
#include "cache.h"
#include "xattr.h"
#include "acl.h"
static int
v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, umode_t omode,
dev_t rdev);
/**
* v9fs_get_fsgid_for_create - Helper function to get the gid for creating a
* new file system object. This checks the S_ISGID to determine the owning
* group of the new file system object.
*/
static kgid_t v9fs_get_fsgid_for_create(struct inode *dir_inode)
{
BUG_ON(dir_inode == NULL);
if (dir_inode->i_mode & S_ISGID) {
/* set_gid bit is set.*/
return dir_inode->i_gid;
}
return current_fsgid();
}
static int v9fs_test_inode_dotl(struct inode *inode, void *data)
{
struct v9fs_inode *v9inode = V9FS_I(inode);
struct p9_stat_dotl *st = (struct p9_stat_dotl *)data;
/* don't match inode of different type */
if ((inode->i_mode & S_IFMT) != (st->st_mode & S_IFMT))
return 0;
if (inode->i_generation != st->st_gen)
return 0;
/* compare qid details */
if (memcmp(&v9inode->qid.version,
&st->qid.version, sizeof(v9inode->qid.version)))
return 0;
if (v9inode->qid.type != st->qid.type)
return 0;
if (v9inode->qid.path != st->qid.path)
return 0;
return 1;
}
/* Always get a new inode */
static int v9fs_test_new_inode_dotl(struct inode *inode, void *data)
{
return 0;
}
static int v9fs_set_inode_dotl(struct inode *inode, void *data)
{
struct v9fs_inode *v9inode = V9FS_I(inode);
struct p9_stat_dotl *st = (struct p9_stat_dotl *)data;
memcpy(&v9inode->qid, &st->qid, sizeof(st->qid));
inode->i_generation = st->st_gen;
return 0;
}
static struct inode *v9fs_qid_iget_dotl(struct super_block *sb,
struct p9_qid *qid,
struct p9_fid *fid,
struct p9_stat_dotl *st,
int new)
{
int retval;
unsigned long i_ino;
struct inode *inode;
struct v9fs_session_info *v9ses = sb->s_fs_info;
int (*test)(struct inode *, void *);
if (new)
test = v9fs_test_new_inode_dotl;
else
test = v9fs_test_inode_dotl;
i_ino = v9fs_qid2ino(qid);
inode = iget5_locked(sb, i_ino, test, v9fs_set_inode_dotl, st);
if (!inode)
return ERR_PTR(-ENOMEM);
if (!(inode->i_state & I_NEW))
return inode;
/*
* initialize the inode with the stat info
* FIXME!! we may need support for stale inodes
* later.
*/
inode->i_ino = i_ino;
retval = v9fs_init_inode(v9ses, inode,
st->st_mode, new_decode_dev(st->st_rdev));
if (retval)
goto error;
9p: use inode->i_lock to protect i_size_write() under 32-bit Use inode->i_lock to protect i_size_write(), else i_size_read() in generic_fillattr() may loop infinitely in read_seqcount_begin() when multiple processes invoke v9fs_vfs_getattr() or v9fs_vfs_getattr_dotl() simultaneously under 32-bit SMP environment, and a soft lockup will be triggered as show below: watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [stat:2217] Modules linked in: CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system PC is at generic_fillattr+0x104/0x108 LR is at 0xec497f00 pc : [<802b8898>] lr : [<ec497f00>] psr: 200c0013 sp : ec497e20 ip : ed608030 fp : ec497e3c r10: 00000000 r9 : ec497f00 r8 : ed608030 r7 : ec497ebc r6 : ec497f00 r5 : ee5c1550 r4 : ee005780 r3 : 0000052d r2 : 00000000 r1 : ec497f00 r0 : ed608030 Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: ac48006a DAC: 00000051 CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system Backtrace: [<8010d974>] (dump_backtrace) from [<8010dc88>] (show_stack+0x20/0x24) [<8010dc68>] (show_stack) from [<80a1d194>] (dump_stack+0xb0/0xdc) [<80a1d0e4>] (dump_stack) from [<80109f34>] (show_regs+0x1c/0x20) [<80109f18>] (show_regs) from [<801d0a80>] (watchdog_timer_fn+0x280/0x2f8) [<801d0800>] (watchdog_timer_fn) from [<80198658>] (__hrtimer_run_queues+0x18c/0x380) [<801984cc>] (__hrtimer_run_queues) from [<80198e60>] (hrtimer_run_queues+0xb8/0xf0) [<80198da8>] (hrtimer_run_queues) from [<801973e8>] (run_local_timers+0x28/0x64) [<801973c0>] (run_local_timers) from [<80197460>] (update_process_times+0x3c/0x6c) [<80197424>] (update_process_times) from [<801ab2b8>] (tick_nohz_handler+0xe0/0x1bc) [<801ab1d8>] (tick_nohz_handler) from [<80843050>] (arch_timer_handler_virt+0x38/0x48) [<80843018>] (arch_timer_handler_virt) from [<80180a64>] (handle_percpu_devid_irq+0x8c/0x240) [<801809d8>] (handle_percpu_devid_irq) from [<8017ac20>] (generic_handle_irq+0x34/0x44) [<8017abec>] (generic_handle_irq) from [<8017b344>] (__handle_domain_irq+0x6c/0xc4) [<8017b2d8>] (__handle_domain_irq) from [<801022e0>] (gic_handle_irq+0x4c/0x88) [<80102294>] (gic_handle_irq) from [<80101a30>] (__irq_svc+0x70/0x98) [<802b8794>] (generic_fillattr) from [<8056b284>] (v9fs_vfs_getattr_dotl+0x74/0xa4) [<8056b210>] (v9fs_vfs_getattr_dotl) from [<802b8904>] (vfs_getattr_nosec+0x68/0x7c) [<802b889c>] (vfs_getattr_nosec) from [<802b895c>] (vfs_getattr+0x44/0x48) [<802b8918>] (vfs_getattr) from [<802b8a74>] (vfs_statx+0x9c/0xec) [<802b89d8>] (vfs_statx) from [<802b9428>] (sys_lstat64+0x48/0x78) [<802b93e0>] (sys_lstat64) from [<80101000>] (ret_fast_syscall+0x0/0x28) [dominique.martinet@cea.fr: updated comment to not refer to a function in another subsystem] Link: http://lkml.kernel.org/r/20190124063514.8571-2-houtao1@huawei.com Cc: stable@vger.kernel.org Fixes: 7549ae3e81cc ("9p: Use the i_size_[read, write]() macros instead of using inode->i_size directly.") Reported-by: Xing Gaopeng <xingaopeng@huawei.com> Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
2019-01-24 14:35:13 +08:00
v9fs_stat2inode_dotl(st, inode, 0);
v9fs_cache_inode_get_cookie(inode);
retval = v9fs_get_acl(inode, fid);
if (retval)
goto error;
unlock_new_inode(inode);
return inode;
error:
iget_failed(inode);
return ERR_PTR(retval);
}
struct inode *
v9fs_inode_from_fid_dotl(struct v9fs_session_info *v9ses, struct p9_fid *fid,
struct super_block *sb, int new)
{
struct p9_stat_dotl *st;
struct inode *inode = NULL;
st = p9_client_getattr_dotl(fid, P9_STATS_BASIC | P9_STATS_GEN);
if (IS_ERR(st))
return ERR_CAST(st);
inode = v9fs_qid_iget_dotl(sb, &st->qid, fid, st, new);
kfree(st);
return inode;
}
struct dotl_openflag_map {
int open_flag;
int dotl_flag;
};
static int v9fs_mapped_dotl_flags(int flags)
{
int i;
int rflags = 0;
struct dotl_openflag_map dotl_oflag_map[] = {
{ O_CREAT, P9_DOTL_CREATE },
{ O_EXCL, P9_DOTL_EXCL },
{ O_NOCTTY, P9_DOTL_NOCTTY },
{ O_APPEND, P9_DOTL_APPEND },
{ O_NONBLOCK, P9_DOTL_NONBLOCK },
{ O_DSYNC, P9_DOTL_DSYNC },
{ FASYNC, P9_DOTL_FASYNC },
{ O_DIRECT, P9_DOTL_DIRECT },
{ O_LARGEFILE, P9_DOTL_LARGEFILE },
{ O_DIRECTORY, P9_DOTL_DIRECTORY },
{ O_NOFOLLOW, P9_DOTL_NOFOLLOW },
{ O_NOATIME, P9_DOTL_NOATIME },
{ O_CLOEXEC, P9_DOTL_CLOEXEC },
{ O_SYNC, P9_DOTL_SYNC},
};
for (i = 0; i < ARRAY_SIZE(dotl_oflag_map); i++) {
if (flags & dotl_oflag_map[i].open_flag)
rflags |= dotl_oflag_map[i].dotl_flag;
}
return rflags;
}
/**
* v9fs_open_to_dotl_flags- convert Linux specific open flags to
* plan 9 open flag.
* @flags: flags to convert
*/
int v9fs_open_to_dotl_flags(int flags)
{
int rflags = 0;
/*
* We have same bits for P9_DOTL_READONLY, P9_DOTL_WRONLY
* and P9_DOTL_NOACCESS
*/
rflags |= flags & O_ACCMODE;
rflags |= v9fs_mapped_dotl_flags(flags);
return rflags;
}
/**
* v9fs_vfs_create_dotl - VFS hook to create files for 9P2000.L protocol.
* @dir: directory inode that is being created
* @dentry: dentry that is being deleted
* @omode: create permissions
*
*/
static int
v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, umode_t omode,
bool excl)
{
return v9fs_vfs_mknod_dotl(dir, dentry, omode, 0);
}
static int
v9fs_vfs_atomic_open_dotl(struct inode *dir, struct dentry *dentry,
struct file *file, unsigned flags, umode_t omode)
{
int err = 0;
kgid_t gid;
umode_t mode;
const unsigned char *name = NULL;
struct p9_qid qid;
struct inode *inode;
struct p9_fid *fid = NULL;
struct v9fs_inode *v9inode;
struct p9_fid *dfid, *ofid, *inode_fid;
struct v9fs_session_info *v9ses;
struct posix_acl *pacl = NULL, *dacl = NULL;
struct dentry *res = NULL;
if (d_in_lookup(dentry)) {
res = v9fs_vfs_lookup(dir, dentry, 0);
if (IS_ERR(res))
return PTR_ERR(res);
if (res)
dentry = res;
}
/* Only creates */
if (!(flags & O_CREAT) || d_really_is_positive(dentry))
return finish_no_open(file, res);
v9ses = v9fs_inode2v9ses(dir);
name = dentry->d_name.name;
p9_debug(P9_DEBUG_VFS, "name:%s flags:0x%x mode:0x%hx\n",
name, flags, omode);
dfid = v9fs_parent_fid(dentry);
if (IS_ERR(dfid)) {
err = PTR_ERR(dfid);
p9_debug(P9_DEBUG_VFS, "fid lookup failed %d\n", err);
goto out;
}
/* clone a fid to use for creation */
ofid = clone_fid(dfid);
if (IS_ERR(ofid)) {
err = PTR_ERR(ofid);
p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err);
goto out;
}
gid = v9fs_get_fsgid_for_create(dir);
mode = omode;
/* Update mode based on ACL value */
err = v9fs_acl_mode(dir, &mode, &dacl, &pacl);
if (err) {
p9_debug(P9_DEBUG_VFS, "Failed to get acl values in creat %d\n",
err);
goto error;
}
err = p9_client_create_dotl(ofid, name, v9fs_open_to_dotl_flags(flags),
mode, gid, &qid);
if (err < 0) {
p9_debug(P9_DEBUG_VFS, "p9_client_open_dotl failed in creat %d\n",
err);
goto error;
}
v9fs_invalidate_inode_attr(dir);
/* instantiate inode and assign the unopened fid to the dentry */
fid = p9_client_walk(dfid, 1, &name, 1);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
p9_client_clunk(dfid);
if (IS_ERR(fid)) {
err = PTR_ERR(fid);
p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err);
fid = NULL;
goto error;
}
inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n", err);
goto error;
}
/* Now set the ACL based on the default value */
v9fs_set_create_acl(inode, fid, dacl, pacl);
v9fs_fid_add(dentry, fid);
d_instantiate(dentry, inode);
v9inode = V9FS_I(inode);
mutex_lock(&v9inode->v_mutex);
if ((v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) &&
!v9inode->writeback_fid &&
((flags & O_ACCMODE) != O_RDONLY)) {
/*
* clone a fid and add it to writeback_fid
* we do it during open time instead of
* page dirty time via write_begin/page_mkwrite
* because we want write after unlink usecase
* to work.
*/
inode_fid = v9fs_writeback_fid(dentry);
if (IS_ERR(inode_fid)) {
err = PTR_ERR(inode_fid);
mutex_unlock(&v9inode->v_mutex);
goto err_clunk_old_fid;
}
v9inode->writeback_fid = (void *) inode_fid;
}
mutex_unlock(&v9inode->v_mutex);
/* Since we are opening a file, assign the open fid to the file */
err = finish_open(file, dentry, generic_file_open);
if (err)
goto err_clunk_old_fid;
file->private_data = ofid;
if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE)
v9fs_cache_inode_set_cookie(inode, file);
v9fs_open_fid_add(inode, ofid);
file->f_mode |= FMODE_CREATED;
out:
v9fs_put_acl(dacl, pacl);
dput(res);
return err;
error:
if (fid)
p9_client_clunk(fid);
err_clunk_old_fid:
if (ofid)
p9_client_clunk(ofid);
goto out;
}
/**
* v9fs_vfs_mkdir_dotl - VFS mkdir hook to create a directory
* @dir: inode that is being unlinked
* @dentry: dentry that is being unlinked
* @omode: mode for new directory
*
*/
static int v9fs_vfs_mkdir_dotl(struct inode *dir,
struct dentry *dentry, umode_t omode)
{
int err;
struct v9fs_session_info *v9ses;
struct p9_fid *fid = NULL, *dfid = NULL;
kgid_t gid;
const unsigned char *name;
umode_t mode;
struct inode *inode;
struct p9_qid qid;
struct posix_acl *dacl = NULL, *pacl = NULL;
p9_debug(P9_DEBUG_VFS, "name %pd\n", dentry);
err = 0;
v9ses = v9fs_inode2v9ses(dir);
omode |= S_IFDIR;
if (dir->i_mode & S_ISGID)
omode |= S_ISGID;
dfid = v9fs_parent_fid(dentry);
if (IS_ERR(dfid)) {
err = PTR_ERR(dfid);
p9_debug(P9_DEBUG_VFS, "fid lookup failed %d\n", err);
dfid = NULL;
goto error;
}
gid = v9fs_get_fsgid_for_create(dir);
mode = omode;
/* Update mode based on ACL value */
err = v9fs_acl_mode(dir, &mode, &dacl, &pacl);
if (err) {
p9_debug(P9_DEBUG_VFS, "Failed to get acl values in mkdir %d\n",
err);
goto error;
}
name = dentry->d_name.name;
err = p9_client_mkdir_dotl(dfid, name, mode, gid, &qid);
if (err < 0)
goto error;
fid = p9_client_walk(dfid, 1, &name, 1);
if (IS_ERR(fid)) {
err = PTR_ERR(fid);
p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n",
err);
fid = NULL;
goto error;
}
/* instantiate inode and assign the unopened fid to the dentry */
if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) {
inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n",
err);
goto error;
}
v9fs_fid_add(dentry, fid);
v9fs_set_create_acl(inode, fid, dacl, pacl);
d_instantiate(dentry, inode);
fid = NULL;
err = 0;
} else {
/*
* Not in cached mode. No need to populate
* inode with stat. We need to get an inode
* so that we can set the acl with dentry
*/
inode = v9fs_get_inode(dir->i_sb, mode, 0);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
goto error;
}
v9fs_set_create_acl(inode, fid, dacl, pacl);
d_instantiate(dentry, inode);
}
inc_nlink(dir);
v9fs_invalidate_inode_attr(dir);
error:
if (fid)
p9_client_clunk(fid);
v9fs_put_acl(dacl, pacl);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
p9_client_clunk(dfid);
return err;
}
static int
statx: Add a system call to make enhanced file info available Add a system call to make extended file information available, including file creation and some attribute flags where available through the underlying filesystem. The getattr inode operation is altered to take two additional arguments: a u32 request_mask and an unsigned int flags that indicate the synchronisation mode. This change is propagated to the vfs_getattr*() function. Functions like vfs_stat() are now inline wrappers around new functions vfs_statx() and vfs_statx_fd() to reduce stack usage. ======== OVERVIEW ======== The idea was initially proposed as a set of xattrs that could be retrieved with getxattr(), but the general preference proved to be for a new syscall with an extended stat structure. A number of requests were gathered for features to be included. The following have been included: (1) Make the fields a consistent size on all arches and make them large. (2) Spare space, request flags and information flags are provided for future expansion. (3) Better support for the y2038 problem [Arnd Bergmann] (tv_sec is an __s64). (4) Creation time: The SMB protocol carries the creation time, which could be exported by Samba, which will in turn help CIFS make use of FS-Cache as that can be used for coherency data (stx_btime). This is also specified in NFSv4 as a recommended attribute and could be exported by NFSD [Steve French]. (5) Lightweight stat: Ask for just those details of interest, and allow a netfs (such as NFS) to approximate anything not of interest, possibly without going to the server [Trond Myklebust, Ulrich Drepper, Andreas Dilger] (AT_STATX_DONT_SYNC). (6) Heavyweight stat: Force a netfs to go to the server, even if it thinks its cached attributes are up to date [Trond Myklebust] (AT_STATX_FORCE_SYNC). And the following have been left out for future extension: (7) Data version number: Could be used by userspace NFS servers [Aneesh Kumar]. Can also be used to modify fill_post_wcc() in NFSD which retrieves i_version directly, but has just called vfs_getattr(). It could get it from the kstat struct if it used vfs_xgetattr() instead. (There's disagreement on the exact semantics of a single field, since not all filesystems do this the same way). (8) BSD stat compatibility: Including more fields from the BSD stat such as creation time (st_btime) and inode generation number (st_gen) [Jeremy Allison, Bernd Schubert]. (9) Inode generation number: Useful for FUSE and userspace NFS servers [Bernd Schubert]. (This was asked for but later deemed unnecessary with the open-by-handle capability available and caused disagreement as to whether it's a security hole or not). (10) Extra coherency data may be useful in making backups [Andreas Dilger]. (No particular data were offered, but things like last backup timestamp, the data version number and the DOS archive bit would come into this category). (11) Allow the filesystem to indicate what it can/cannot provide: A filesystem can now say it doesn't support a standard stat feature if that isn't available, so if, for instance, inode numbers or UIDs don't exist or are fabricated locally... (This requires a separate system call - I have an fsinfo() call idea for this). (12) Store a 16-byte volume ID in the superblock that can be returned in struct xstat [Steve French]. (Deferred to fsinfo). (13) Include granularity fields in the time data to indicate the granularity of each of the times (NFSv4 time_delta) [Steve French]. (Deferred to fsinfo). (14) FS_IOC_GETFLAGS value. These could be translated to BSD's st_flags. Note that the Linux IOC flags are a mess and filesystems such as Ext4 define flags that aren't in linux/fs.h, so translation in the kernel may be a necessity (or, possibly, we provide the filesystem type too). (Some attributes are made available in stx_attributes, but the general feeling was that the IOC flags were to ext[234]-specific and shouldn't be exposed through statx this way). (15) Mask of features available on file (eg: ACLs, seclabel) [Brad Boyer, Michael Kerrisk]. (Deferred, probably to fsinfo. Finding out if there's an ACL or seclabal might require extra filesystem operations). (16) Femtosecond-resolution timestamps [Dave Chinner]. (A __reserved field has been left in the statx_timestamp struct for this - if there proves to be a need). (17) A set multiple attributes syscall to go with this. =============== NEW SYSTEM CALL =============== The new system call is: int ret = statx(int dfd, const char *filename, unsigned int flags, unsigned int mask, struct statx *buffer); The dfd, filename and flags parameters indicate the file to query, in a similar way to fstatat(). There is no equivalent of lstat() as that can be emulated with statx() by passing AT_SYMLINK_NOFOLLOW in flags. There is also no equivalent of fstat() as that can be emulated by passing a NULL filename to statx() with the fd of interest in dfd. Whether or not statx() synchronises the attributes with the backing store can be controlled by OR'ing a value into the flags argument (this typically only affects network filesystems): (1) AT_STATX_SYNC_AS_STAT tells statx() to behave as stat() does in this respect. (2) AT_STATX_FORCE_SYNC will require a network filesystem to synchronise its attributes with the server - which might require data writeback to occur to get the timestamps correct. (3) AT_STATX_DONT_SYNC will suppress synchronisation with the server in a network filesystem. The resulting values should be considered approximate. mask is a bitmask indicating the fields in struct statx that are of interest to the caller. The user should set this to STATX_BASIC_STATS to get the basic set returned by stat(). It should be noted that asking for more information may entail extra I/O operations. buffer points to the destination for the data. This must be 256 bytes in size. ====================== MAIN ATTRIBUTES RECORD ====================== The following structures are defined in which to return the main attribute set: struct statx_timestamp { __s64 tv_sec; __s32 tv_nsec; __s32 __reserved; }; struct statx { __u32 stx_mask; __u32 stx_blksize; __u64 stx_attributes; __u32 stx_nlink; __u32 stx_uid; __u32 stx_gid; __u16 stx_mode; __u16 __spare0[1]; __u64 stx_ino; __u64 stx_size; __u64 stx_blocks; __u64 __spare1[1]; struct statx_timestamp stx_atime; struct statx_timestamp stx_btime; struct statx_timestamp stx_ctime; struct statx_timestamp stx_mtime; __u32 stx_rdev_major; __u32 stx_rdev_minor; __u32 stx_dev_major; __u32 stx_dev_minor; __u64 __spare2[14]; }; The defined bits in request_mask and stx_mask are: STATX_TYPE Want/got stx_mode & S_IFMT STATX_MODE Want/got stx_mode & ~S_IFMT STATX_NLINK Want/got stx_nlink STATX_UID Want/got stx_uid STATX_GID Want/got stx_gid STATX_ATIME Want/got stx_atime{,_ns} STATX_MTIME Want/got stx_mtime{,_ns} STATX_CTIME Want/got stx_ctime{,_ns} STATX_INO Want/got stx_ino STATX_SIZE Want/got stx_size STATX_BLOCKS Want/got stx_blocks STATX_BASIC_STATS [The stuff in the normal stat struct] STATX_BTIME Want/got stx_btime{,_ns} STATX_ALL [All currently available stuff] stx_btime is the file creation time, stx_mask is a bitmask indicating the data provided and __spares*[] are where as-yet undefined fields can be placed. Time fields are structures with separate seconds and nanoseconds fields plus a reserved field in case we want to add even finer resolution. Note that times will be negative if before 1970; in such a case, the nanosecond fields will also be negative if not zero. The bits defined in the stx_attributes field convey information about a file, how it is accessed, where it is and what it does. The following attributes map to FS_*_FL flags and are the same numerical value: STATX_ATTR_COMPRESSED File is compressed by the fs STATX_ATTR_IMMUTABLE File is marked immutable STATX_ATTR_APPEND File is append-only STATX_ATTR_NODUMP File is not to be dumped STATX_ATTR_ENCRYPTED File requires key to decrypt in fs Within the kernel, the supported flags are listed by: KSTAT_ATTR_FS_IOC_FLAGS [Are any other IOC flags of sufficient general interest to be exposed through this interface?] New flags include: STATX_ATTR_AUTOMOUNT Object is an automount trigger These are for the use of GUI tools that might want to mark files specially, depending on what they are. Fields in struct statx come in a number of classes: (0) stx_dev_*, stx_blksize. These are local system information and are always available. (1) stx_mode, stx_nlinks, stx_uid, stx_gid, stx_[amc]time, stx_ino, stx_size, stx_blocks. These will be returned whether the caller asks for them or not. The corresponding bits in stx_mask will be set to indicate whether they actually have valid values. If the caller didn't ask for them, then they may be approximated. For example, NFS won't waste any time updating them from the server, unless as a byproduct of updating something requested. If the values don't actually exist for the underlying object (such as UID or GID on a DOS file), then the bit won't be set in the stx_mask, even if the caller asked for the value. In such a case, the returned value will be a fabrication. Note that there are instances where the type might not be valid, for instance Windows reparse points. (2) stx_rdev_*. This will be set only if stx_mode indicates we're looking at a blockdev or a chardev, otherwise will be 0. (3) stx_btime. Similar to (1), except this will be set to 0 if it doesn't exist. ======= TESTING ======= The following test program can be used to test the statx system call: samples/statx/test-statx.c Just compile and run, passing it paths to the files you want to examine. The file is built automatically if CONFIG_SAMPLES is enabled. Here's some example output. Firstly, an NFS directory that crosses to another FSID. Note that the AUTOMOUNT attribute is set because transiting this directory will cause d_automount to be invoked by the VFS. [root@andromeda ~]# /tmp/test-statx -A /warthog/data statx(/warthog/data) = 0 results=7ff Size: 4096 Blocks: 8 IO Block: 1048576 directory Device: 00:26 Inode: 1703937 Links: 125 Access: (3777/drwxrwxrwx) Uid: 0 Gid: 4041 Access: 2016-11-24 09:02:12.219699527+0000 Modify: 2016-11-17 10:44:36.225653653+0000 Change: 2016-11-17 10:44:36.225653653+0000 Attributes: 0000000000001000 (-------- -------- -------- -------- -------- -------- ---m---- --------) Secondly, the result of automounting on that directory. [root@andromeda ~]# /tmp/test-statx /warthog/data statx(/warthog/data) = 0 results=7ff Size: 4096 Blocks: 8 IO Block: 1048576 directory Device: 00:27 Inode: 2 Links: 125 Access: (3777/drwxrwxrwx) Uid: 0 Gid: 4041 Access: 2016-11-24 09:02:12.219699527+0000 Modify: 2016-11-17 10:44:36.225653653+0000 Change: 2016-11-17 10:44:36.225653653+0000 Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2017-02-01 00:46:22 +08:00
v9fs_vfs_getattr_dotl(const struct path *path, struct kstat *stat,
u32 request_mask, unsigned int flags)
{
statx: Add a system call to make enhanced file info available Add a system call to make extended file information available, including file creation and some attribute flags where available through the underlying filesystem. The getattr inode operation is altered to take two additional arguments: a u32 request_mask and an unsigned int flags that indicate the synchronisation mode. This change is propagated to the vfs_getattr*() function. Functions like vfs_stat() are now inline wrappers around new functions vfs_statx() and vfs_statx_fd() to reduce stack usage. ======== OVERVIEW ======== The idea was initially proposed as a set of xattrs that could be retrieved with getxattr(), but the general preference proved to be for a new syscall with an extended stat structure. A number of requests were gathered for features to be included. The following have been included: (1) Make the fields a consistent size on all arches and make them large. (2) Spare space, request flags and information flags are provided for future expansion. (3) Better support for the y2038 problem [Arnd Bergmann] (tv_sec is an __s64). (4) Creation time: The SMB protocol carries the creation time, which could be exported by Samba, which will in turn help CIFS make use of FS-Cache as that can be used for coherency data (stx_btime). This is also specified in NFSv4 as a recommended attribute and could be exported by NFSD [Steve French]. (5) Lightweight stat: Ask for just those details of interest, and allow a netfs (such as NFS) to approximate anything not of interest, possibly without going to the server [Trond Myklebust, Ulrich Drepper, Andreas Dilger] (AT_STATX_DONT_SYNC). (6) Heavyweight stat: Force a netfs to go to the server, even if it thinks its cached attributes are up to date [Trond Myklebust] (AT_STATX_FORCE_SYNC). And the following have been left out for future extension: (7) Data version number: Could be used by userspace NFS servers [Aneesh Kumar]. Can also be used to modify fill_post_wcc() in NFSD which retrieves i_version directly, but has just called vfs_getattr(). It could get it from the kstat struct if it used vfs_xgetattr() instead. (There's disagreement on the exact semantics of a single field, since not all filesystems do this the same way). (8) BSD stat compatibility: Including more fields from the BSD stat such as creation time (st_btime) and inode generation number (st_gen) [Jeremy Allison, Bernd Schubert]. (9) Inode generation number: Useful for FUSE and userspace NFS servers [Bernd Schubert]. (This was asked for but later deemed unnecessary with the open-by-handle capability available and caused disagreement as to whether it's a security hole or not). (10) Extra coherency data may be useful in making backups [Andreas Dilger]. (No particular data were offered, but things like last backup timestamp, the data version number and the DOS archive bit would come into this category). (11) Allow the filesystem to indicate what it can/cannot provide: A filesystem can now say it doesn't support a standard stat feature if that isn't available, so if, for instance, inode numbers or UIDs don't exist or are fabricated locally... (This requires a separate system call - I have an fsinfo() call idea for this). (12) Store a 16-byte volume ID in the superblock that can be returned in struct xstat [Steve French]. (Deferred to fsinfo). (13) Include granularity fields in the time data to indicate the granularity of each of the times (NFSv4 time_delta) [Steve French]. (Deferred to fsinfo). (14) FS_IOC_GETFLAGS value. These could be translated to BSD's st_flags. Note that the Linux IOC flags are a mess and filesystems such as Ext4 define flags that aren't in linux/fs.h, so translation in the kernel may be a necessity (or, possibly, we provide the filesystem type too). (Some attributes are made available in stx_attributes, but the general feeling was that the IOC flags were to ext[234]-specific and shouldn't be exposed through statx this way). (15) Mask of features available on file (eg: ACLs, seclabel) [Brad Boyer, Michael Kerrisk]. (Deferred, probably to fsinfo. Finding out if there's an ACL or seclabal might require extra filesystem operations). (16) Femtosecond-resolution timestamps [Dave Chinner]. (A __reserved field has been left in the statx_timestamp struct for this - if there proves to be a need). (17) A set multiple attributes syscall to go with this. =============== NEW SYSTEM CALL =============== The new system call is: int ret = statx(int dfd, const char *filename, unsigned int flags, unsigned int mask, struct statx *buffer); The dfd, filename and flags parameters indicate the file to query, in a similar way to fstatat(). There is no equivalent of lstat() as that can be emulated with statx() by passing AT_SYMLINK_NOFOLLOW in flags. There is also no equivalent of fstat() as that can be emulated by passing a NULL filename to statx() with the fd of interest in dfd. Whether or not statx() synchronises the attributes with the backing store can be controlled by OR'ing a value into the flags argument (this typically only affects network filesystems): (1) AT_STATX_SYNC_AS_STAT tells statx() to behave as stat() does in this respect. (2) AT_STATX_FORCE_SYNC will require a network filesystem to synchronise its attributes with the server - which might require data writeback to occur to get the timestamps correct. (3) AT_STATX_DONT_SYNC will suppress synchronisation with the server in a network filesystem. The resulting values should be considered approximate. mask is a bitmask indicating the fields in struct statx that are of interest to the caller. The user should set this to STATX_BASIC_STATS to get the basic set returned by stat(). It should be noted that asking for more information may entail extra I/O operations. buffer points to the destination for the data. This must be 256 bytes in size. ====================== MAIN ATTRIBUTES RECORD ====================== The following structures are defined in which to return the main attribute set: struct statx_timestamp { __s64 tv_sec; __s32 tv_nsec; __s32 __reserved; }; struct statx { __u32 stx_mask; __u32 stx_blksize; __u64 stx_attributes; __u32 stx_nlink; __u32 stx_uid; __u32 stx_gid; __u16 stx_mode; __u16 __spare0[1]; __u64 stx_ino; __u64 stx_size; __u64 stx_blocks; __u64 __spare1[1]; struct statx_timestamp stx_atime; struct statx_timestamp stx_btime; struct statx_timestamp stx_ctime; struct statx_timestamp stx_mtime; __u32 stx_rdev_major; __u32 stx_rdev_minor; __u32 stx_dev_major; __u32 stx_dev_minor; __u64 __spare2[14]; }; The defined bits in request_mask and stx_mask are: STATX_TYPE Want/got stx_mode & S_IFMT STATX_MODE Want/got stx_mode & ~S_IFMT STATX_NLINK Want/got stx_nlink STATX_UID Want/got stx_uid STATX_GID Want/got stx_gid STATX_ATIME Want/got stx_atime{,_ns} STATX_MTIME Want/got stx_mtime{,_ns} STATX_CTIME Want/got stx_ctime{,_ns} STATX_INO Want/got stx_ino STATX_SIZE Want/got stx_size STATX_BLOCKS Want/got stx_blocks STATX_BASIC_STATS [The stuff in the normal stat struct] STATX_BTIME Want/got stx_btime{,_ns} STATX_ALL [All currently available stuff] stx_btime is the file creation time, stx_mask is a bitmask indicating the data provided and __spares*[] are where as-yet undefined fields can be placed. Time fields are structures with separate seconds and nanoseconds fields plus a reserved field in case we want to add even finer resolution. Note that times will be negative if before 1970; in such a case, the nanosecond fields will also be negative if not zero. The bits defined in the stx_attributes field convey information about a file, how it is accessed, where it is and what it does. The following attributes map to FS_*_FL flags and are the same numerical value: STATX_ATTR_COMPRESSED File is compressed by the fs STATX_ATTR_IMMUTABLE File is marked immutable STATX_ATTR_APPEND File is append-only STATX_ATTR_NODUMP File is not to be dumped STATX_ATTR_ENCRYPTED File requires key to decrypt in fs Within the kernel, the supported flags are listed by: KSTAT_ATTR_FS_IOC_FLAGS [Are any other IOC flags of sufficient general interest to be exposed through this interface?] New flags include: STATX_ATTR_AUTOMOUNT Object is an automount trigger These are for the use of GUI tools that might want to mark files specially, depending on what they are. Fields in struct statx come in a number of classes: (0) stx_dev_*, stx_blksize. These are local system information and are always available. (1) stx_mode, stx_nlinks, stx_uid, stx_gid, stx_[amc]time, stx_ino, stx_size, stx_blocks. These will be returned whether the caller asks for them or not. The corresponding bits in stx_mask will be set to indicate whether they actually have valid values. If the caller didn't ask for them, then they may be approximated. For example, NFS won't waste any time updating them from the server, unless as a byproduct of updating something requested. If the values don't actually exist for the underlying object (such as UID or GID on a DOS file), then the bit won't be set in the stx_mask, even if the caller asked for the value. In such a case, the returned value will be a fabrication. Note that there are instances where the type might not be valid, for instance Windows reparse points. (2) stx_rdev_*. This will be set only if stx_mode indicates we're looking at a blockdev or a chardev, otherwise will be 0. (3) stx_btime. Similar to (1), except this will be set to 0 if it doesn't exist. ======= TESTING ======= The following test program can be used to test the statx system call: samples/statx/test-statx.c Just compile and run, passing it paths to the files you want to examine. The file is built automatically if CONFIG_SAMPLES is enabled. Here's some example output. Firstly, an NFS directory that crosses to another FSID. Note that the AUTOMOUNT attribute is set because transiting this directory will cause d_automount to be invoked by the VFS. [root@andromeda ~]# /tmp/test-statx -A /warthog/data statx(/warthog/data) = 0 results=7ff Size: 4096 Blocks: 8 IO Block: 1048576 directory Device: 00:26 Inode: 1703937 Links: 125 Access: (3777/drwxrwxrwx) Uid: 0 Gid: 4041 Access: 2016-11-24 09:02:12.219699527+0000 Modify: 2016-11-17 10:44:36.225653653+0000 Change: 2016-11-17 10:44:36.225653653+0000 Attributes: 0000000000001000 (-------- -------- -------- -------- -------- -------- ---m---- --------) Secondly, the result of automounting on that directory. [root@andromeda ~]# /tmp/test-statx /warthog/data statx(/warthog/data) = 0 results=7ff Size: 4096 Blocks: 8 IO Block: 1048576 directory Device: 00:27 Inode: 2 Links: 125 Access: (3777/drwxrwxrwx) Uid: 0 Gid: 4041 Access: 2016-11-24 09:02:12.219699527+0000 Modify: 2016-11-17 10:44:36.225653653+0000 Change: 2016-11-17 10:44:36.225653653+0000 Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2017-02-01 00:46:22 +08:00
struct dentry *dentry = path->dentry;
struct v9fs_session_info *v9ses;
struct p9_fid *fid;
struct p9_stat_dotl *st;
p9_debug(P9_DEBUG_VFS, "dentry: %p\n", dentry);
v9ses = v9fs_dentry2v9ses(dentry);
if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) {
generic_fillattr(d_inode(dentry), stat);
return 0;
}
fid = v9fs_fid_lookup(dentry);
if (IS_ERR(fid))
return PTR_ERR(fid);
/* Ask for all the fields in stat structure. Server will return
* whatever it supports
*/
st = p9_client_getattr_dotl(fid, P9_STATS_ALL);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
p9_client_clunk(fid);
if (IS_ERR(st))
return PTR_ERR(st);
9p: use inode->i_lock to protect i_size_write() under 32-bit Use inode->i_lock to protect i_size_write(), else i_size_read() in generic_fillattr() may loop infinitely in read_seqcount_begin() when multiple processes invoke v9fs_vfs_getattr() or v9fs_vfs_getattr_dotl() simultaneously under 32-bit SMP environment, and a soft lockup will be triggered as show below: watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [stat:2217] Modules linked in: CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system PC is at generic_fillattr+0x104/0x108 LR is at 0xec497f00 pc : [<802b8898>] lr : [<ec497f00>] psr: 200c0013 sp : ec497e20 ip : ed608030 fp : ec497e3c r10: 00000000 r9 : ec497f00 r8 : ed608030 r7 : ec497ebc r6 : ec497f00 r5 : ee5c1550 r4 : ee005780 r3 : 0000052d r2 : 00000000 r1 : ec497f00 r0 : ed608030 Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: ac48006a DAC: 00000051 CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system Backtrace: [<8010d974>] (dump_backtrace) from [<8010dc88>] (show_stack+0x20/0x24) [<8010dc68>] (show_stack) from [<80a1d194>] (dump_stack+0xb0/0xdc) [<80a1d0e4>] (dump_stack) from [<80109f34>] (show_regs+0x1c/0x20) [<80109f18>] (show_regs) from [<801d0a80>] (watchdog_timer_fn+0x280/0x2f8) [<801d0800>] (watchdog_timer_fn) from [<80198658>] (__hrtimer_run_queues+0x18c/0x380) [<801984cc>] (__hrtimer_run_queues) from [<80198e60>] (hrtimer_run_queues+0xb8/0xf0) [<80198da8>] (hrtimer_run_queues) from [<801973e8>] (run_local_timers+0x28/0x64) [<801973c0>] (run_local_timers) from [<80197460>] (update_process_times+0x3c/0x6c) [<80197424>] (update_process_times) from [<801ab2b8>] (tick_nohz_handler+0xe0/0x1bc) [<801ab1d8>] (tick_nohz_handler) from [<80843050>] (arch_timer_handler_virt+0x38/0x48) [<80843018>] (arch_timer_handler_virt) from [<80180a64>] (handle_percpu_devid_irq+0x8c/0x240) [<801809d8>] (handle_percpu_devid_irq) from [<8017ac20>] (generic_handle_irq+0x34/0x44) [<8017abec>] (generic_handle_irq) from [<8017b344>] (__handle_domain_irq+0x6c/0xc4) [<8017b2d8>] (__handle_domain_irq) from [<801022e0>] (gic_handle_irq+0x4c/0x88) [<80102294>] (gic_handle_irq) from [<80101a30>] (__irq_svc+0x70/0x98) [<802b8794>] (generic_fillattr) from [<8056b284>] (v9fs_vfs_getattr_dotl+0x74/0xa4) [<8056b210>] (v9fs_vfs_getattr_dotl) from [<802b8904>] (vfs_getattr_nosec+0x68/0x7c) [<802b889c>] (vfs_getattr_nosec) from [<802b895c>] (vfs_getattr+0x44/0x48) [<802b8918>] (vfs_getattr) from [<802b8a74>] (vfs_statx+0x9c/0xec) [<802b89d8>] (vfs_statx) from [<802b9428>] (sys_lstat64+0x48/0x78) [<802b93e0>] (sys_lstat64) from [<80101000>] (ret_fast_syscall+0x0/0x28) [dominique.martinet@cea.fr: updated comment to not refer to a function in another subsystem] Link: http://lkml.kernel.org/r/20190124063514.8571-2-houtao1@huawei.com Cc: stable@vger.kernel.org Fixes: 7549ae3e81cc ("9p: Use the i_size_[read, write]() macros instead of using inode->i_size directly.") Reported-by: Xing Gaopeng <xingaopeng@huawei.com> Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
2019-01-24 14:35:13 +08:00
v9fs_stat2inode_dotl(st, d_inode(dentry), 0);
generic_fillattr(d_inode(dentry), stat);
/* Change block size to what the server returned */
stat->blksize = st->st_blksize;
kfree(st);
return 0;
}
/*
* Attribute flags.
*/
#define P9_ATTR_MODE (1 << 0)
#define P9_ATTR_UID (1 << 1)
#define P9_ATTR_GID (1 << 2)
#define P9_ATTR_SIZE (1 << 3)
#define P9_ATTR_ATIME (1 << 4)
#define P9_ATTR_MTIME (1 << 5)
#define P9_ATTR_CTIME (1 << 6)
#define P9_ATTR_ATIME_SET (1 << 7)
#define P9_ATTR_MTIME_SET (1 << 8)
struct dotl_iattr_map {
int iattr_valid;
int p9_iattr_valid;
};
static int v9fs_mapped_iattr_valid(int iattr_valid)
{
int i;
int p9_iattr_valid = 0;
struct dotl_iattr_map dotl_iattr_map[] = {
{ ATTR_MODE, P9_ATTR_MODE },
{ ATTR_UID, P9_ATTR_UID },
{ ATTR_GID, P9_ATTR_GID },
{ ATTR_SIZE, P9_ATTR_SIZE },
{ ATTR_ATIME, P9_ATTR_ATIME },
{ ATTR_MTIME, P9_ATTR_MTIME },
{ ATTR_CTIME, P9_ATTR_CTIME },
{ ATTR_ATIME_SET, P9_ATTR_ATIME_SET },
{ ATTR_MTIME_SET, P9_ATTR_MTIME_SET },
};
for (i = 0; i < ARRAY_SIZE(dotl_iattr_map); i++) {
if (iattr_valid & dotl_iattr_map[i].iattr_valid)
p9_iattr_valid |= dotl_iattr_map[i].p9_iattr_valid;
}
return p9_iattr_valid;
}
/**
* v9fs_vfs_setattr_dotl - set file metadata
* @dentry: file whose metadata to set
* @iattr: metadata assignment structure
*
*/
int v9fs_vfs_setattr_dotl(struct dentry *dentry, struct iattr *iattr)
{
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
int retval, use_dentry = 0;
struct p9_fid *fid = NULL;
struct p9_iattr_dotl p9attr;
struct inode *inode = d_inode(dentry);
p9_debug(P9_DEBUG_VFS, "\n");
retval = setattr_prepare(dentry, iattr);
if (retval)
return retval;
p9attr.valid = v9fs_mapped_iattr_valid(iattr->ia_valid);
p9attr.mode = iattr->ia_mode;
p9attr.uid = iattr->ia_uid;
p9attr.gid = iattr->ia_gid;
p9attr.size = iattr->ia_size;
p9attr.atime_sec = iattr->ia_atime.tv_sec;
p9attr.atime_nsec = iattr->ia_atime.tv_nsec;
p9attr.mtime_sec = iattr->ia_mtime.tv_sec;
p9attr.mtime_nsec = iattr->ia_mtime.tv_nsec;
if (iattr->ia_valid & ATTR_FILE) {
fid = iattr->ia_file->private_data;
WARN_ON(!fid);
}
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
if (!fid) {
fid = v9fs_fid_lookup(dentry);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
use_dentry = 1;
}
if (IS_ERR(fid))
return PTR_ERR(fid);
/* Write all dirty data */
if (S_ISREG(inode->i_mode))
filemap_write_and_wait(inode->i_mapping);
retval = p9_client_setattr(fid, &p9attr);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
if (retval < 0) {
if (use_dentry)
p9_client_clunk(fid);
return retval;
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
}
if ((iattr->ia_valid & ATTR_SIZE) &&
iattr->ia_size != i_size_read(inode))
truncate_setsize(inode, iattr->ia_size);
v9fs_invalidate_inode_attr(inode);
setattr_copy(inode, iattr);
mark_inode_dirty(inode);
if (iattr->ia_valid & ATTR_MODE) {
/* We also want to update ACL when we update mode bits */
retval = v9fs_acl_chmod(inode, fid);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
if (retval < 0) {
if (use_dentry)
p9_client_clunk(fid);
return retval;
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
}
}
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
if (use_dentry)
p9_client_clunk(fid);
return 0;
}
/**
* v9fs_stat2inode_dotl - populate an inode structure with stat info
* @stat: stat structure
* @inode: inode to populate
9p: use inode->i_lock to protect i_size_write() under 32-bit Use inode->i_lock to protect i_size_write(), else i_size_read() in generic_fillattr() may loop infinitely in read_seqcount_begin() when multiple processes invoke v9fs_vfs_getattr() or v9fs_vfs_getattr_dotl() simultaneously under 32-bit SMP environment, and a soft lockup will be triggered as show below: watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [stat:2217] Modules linked in: CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system PC is at generic_fillattr+0x104/0x108 LR is at 0xec497f00 pc : [<802b8898>] lr : [<ec497f00>] psr: 200c0013 sp : ec497e20 ip : ed608030 fp : ec497e3c r10: 00000000 r9 : ec497f00 r8 : ed608030 r7 : ec497ebc r6 : ec497f00 r5 : ee5c1550 r4 : ee005780 r3 : 0000052d r2 : 00000000 r1 : ec497f00 r0 : ed608030 Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: ac48006a DAC: 00000051 CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system Backtrace: [<8010d974>] (dump_backtrace) from [<8010dc88>] (show_stack+0x20/0x24) [<8010dc68>] (show_stack) from [<80a1d194>] (dump_stack+0xb0/0xdc) [<80a1d0e4>] (dump_stack) from [<80109f34>] (show_regs+0x1c/0x20) [<80109f18>] (show_regs) from [<801d0a80>] (watchdog_timer_fn+0x280/0x2f8) [<801d0800>] (watchdog_timer_fn) from [<80198658>] (__hrtimer_run_queues+0x18c/0x380) [<801984cc>] (__hrtimer_run_queues) from [<80198e60>] (hrtimer_run_queues+0xb8/0xf0) [<80198da8>] (hrtimer_run_queues) from [<801973e8>] (run_local_timers+0x28/0x64) [<801973c0>] (run_local_timers) from [<80197460>] (update_process_times+0x3c/0x6c) [<80197424>] (update_process_times) from [<801ab2b8>] (tick_nohz_handler+0xe0/0x1bc) [<801ab1d8>] (tick_nohz_handler) from [<80843050>] (arch_timer_handler_virt+0x38/0x48) [<80843018>] (arch_timer_handler_virt) from [<80180a64>] (handle_percpu_devid_irq+0x8c/0x240) [<801809d8>] (handle_percpu_devid_irq) from [<8017ac20>] (generic_handle_irq+0x34/0x44) [<8017abec>] (generic_handle_irq) from [<8017b344>] (__handle_domain_irq+0x6c/0xc4) [<8017b2d8>] (__handle_domain_irq) from [<801022e0>] (gic_handle_irq+0x4c/0x88) [<80102294>] (gic_handle_irq) from [<80101a30>] (__irq_svc+0x70/0x98) [<802b8794>] (generic_fillattr) from [<8056b284>] (v9fs_vfs_getattr_dotl+0x74/0xa4) [<8056b210>] (v9fs_vfs_getattr_dotl) from [<802b8904>] (vfs_getattr_nosec+0x68/0x7c) [<802b889c>] (vfs_getattr_nosec) from [<802b895c>] (vfs_getattr+0x44/0x48) [<802b8918>] (vfs_getattr) from [<802b8a74>] (vfs_statx+0x9c/0xec) [<802b89d8>] (vfs_statx) from [<802b9428>] (sys_lstat64+0x48/0x78) [<802b93e0>] (sys_lstat64) from [<80101000>] (ret_fast_syscall+0x0/0x28) [dominique.martinet@cea.fr: updated comment to not refer to a function in another subsystem] Link: http://lkml.kernel.org/r/20190124063514.8571-2-houtao1@huawei.com Cc: stable@vger.kernel.org Fixes: 7549ae3e81cc ("9p: Use the i_size_[read, write]() macros instead of using inode->i_size directly.") Reported-by: Xing Gaopeng <xingaopeng@huawei.com> Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
2019-01-24 14:35:13 +08:00
* @flags: ctrl flags (e.g. V9FS_STAT2INODE_KEEP_ISIZE)
*
*/
void
9p: use inode->i_lock to protect i_size_write() under 32-bit Use inode->i_lock to protect i_size_write(), else i_size_read() in generic_fillattr() may loop infinitely in read_seqcount_begin() when multiple processes invoke v9fs_vfs_getattr() or v9fs_vfs_getattr_dotl() simultaneously under 32-bit SMP environment, and a soft lockup will be triggered as show below: watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [stat:2217] Modules linked in: CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system PC is at generic_fillattr+0x104/0x108 LR is at 0xec497f00 pc : [<802b8898>] lr : [<ec497f00>] psr: 200c0013 sp : ec497e20 ip : ed608030 fp : ec497e3c r10: 00000000 r9 : ec497f00 r8 : ed608030 r7 : ec497ebc r6 : ec497f00 r5 : ee5c1550 r4 : ee005780 r3 : 0000052d r2 : 00000000 r1 : ec497f00 r0 : ed608030 Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: ac48006a DAC: 00000051 CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system Backtrace: [<8010d974>] (dump_backtrace) from [<8010dc88>] (show_stack+0x20/0x24) [<8010dc68>] (show_stack) from [<80a1d194>] (dump_stack+0xb0/0xdc) [<80a1d0e4>] (dump_stack) from [<80109f34>] (show_regs+0x1c/0x20) [<80109f18>] (show_regs) from [<801d0a80>] (watchdog_timer_fn+0x280/0x2f8) [<801d0800>] (watchdog_timer_fn) from [<80198658>] (__hrtimer_run_queues+0x18c/0x380) [<801984cc>] (__hrtimer_run_queues) from [<80198e60>] (hrtimer_run_queues+0xb8/0xf0) [<80198da8>] (hrtimer_run_queues) from [<801973e8>] (run_local_timers+0x28/0x64) [<801973c0>] (run_local_timers) from [<80197460>] (update_process_times+0x3c/0x6c) [<80197424>] (update_process_times) from [<801ab2b8>] (tick_nohz_handler+0xe0/0x1bc) [<801ab1d8>] (tick_nohz_handler) from [<80843050>] (arch_timer_handler_virt+0x38/0x48) [<80843018>] (arch_timer_handler_virt) from [<80180a64>] (handle_percpu_devid_irq+0x8c/0x240) [<801809d8>] (handle_percpu_devid_irq) from [<8017ac20>] (generic_handle_irq+0x34/0x44) [<8017abec>] (generic_handle_irq) from [<8017b344>] (__handle_domain_irq+0x6c/0xc4) [<8017b2d8>] (__handle_domain_irq) from [<801022e0>] (gic_handle_irq+0x4c/0x88) [<80102294>] (gic_handle_irq) from [<80101a30>] (__irq_svc+0x70/0x98) [<802b8794>] (generic_fillattr) from [<8056b284>] (v9fs_vfs_getattr_dotl+0x74/0xa4) [<8056b210>] (v9fs_vfs_getattr_dotl) from [<802b8904>] (vfs_getattr_nosec+0x68/0x7c) [<802b889c>] (vfs_getattr_nosec) from [<802b895c>] (vfs_getattr+0x44/0x48) [<802b8918>] (vfs_getattr) from [<802b8a74>] (vfs_statx+0x9c/0xec) [<802b89d8>] (vfs_statx) from [<802b9428>] (sys_lstat64+0x48/0x78) [<802b93e0>] (sys_lstat64) from [<80101000>] (ret_fast_syscall+0x0/0x28) [dominique.martinet@cea.fr: updated comment to not refer to a function in another subsystem] Link: http://lkml.kernel.org/r/20190124063514.8571-2-houtao1@huawei.com Cc: stable@vger.kernel.org Fixes: 7549ae3e81cc ("9p: Use the i_size_[read, write]() macros instead of using inode->i_size directly.") Reported-by: Xing Gaopeng <xingaopeng@huawei.com> Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
2019-01-24 14:35:13 +08:00
v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode,
unsigned int flags)
{
umode_t mode;
struct v9fs_inode *v9inode = V9FS_I(inode);
if ((stat->st_result_mask & P9_STATS_BASIC) == P9_STATS_BASIC) {
inode->i_atime.tv_sec = stat->st_atime_sec;
inode->i_atime.tv_nsec = stat->st_atime_nsec;
inode->i_mtime.tv_sec = stat->st_mtime_sec;
inode->i_mtime.tv_nsec = stat->st_mtime_nsec;
inode->i_ctime.tv_sec = stat->st_ctime_sec;
inode->i_ctime.tv_nsec = stat->st_ctime_nsec;
inode->i_uid = stat->st_uid;
inode->i_gid = stat->st_gid;
set_nlink(inode, stat->st_nlink);
mode = stat->st_mode & S_IALLUGO;
mode |= inode->i_mode & ~S_IALLUGO;
inode->i_mode = mode;
9p: use inode->i_lock to protect i_size_write() under 32-bit Use inode->i_lock to protect i_size_write(), else i_size_read() in generic_fillattr() may loop infinitely in read_seqcount_begin() when multiple processes invoke v9fs_vfs_getattr() or v9fs_vfs_getattr_dotl() simultaneously under 32-bit SMP environment, and a soft lockup will be triggered as show below: watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [stat:2217] Modules linked in: CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system PC is at generic_fillattr+0x104/0x108 LR is at 0xec497f00 pc : [<802b8898>] lr : [<ec497f00>] psr: 200c0013 sp : ec497e20 ip : ed608030 fp : ec497e3c r10: 00000000 r9 : ec497f00 r8 : ed608030 r7 : ec497ebc r6 : ec497f00 r5 : ee5c1550 r4 : ee005780 r3 : 0000052d r2 : 00000000 r1 : ec497f00 r0 : ed608030 Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: ac48006a DAC: 00000051 CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system Backtrace: [<8010d974>] (dump_backtrace) from [<8010dc88>] (show_stack+0x20/0x24) [<8010dc68>] (show_stack) from [<80a1d194>] (dump_stack+0xb0/0xdc) [<80a1d0e4>] (dump_stack) from [<80109f34>] (show_regs+0x1c/0x20) [<80109f18>] (show_regs) from [<801d0a80>] (watchdog_timer_fn+0x280/0x2f8) [<801d0800>] (watchdog_timer_fn) from [<80198658>] (__hrtimer_run_queues+0x18c/0x380) [<801984cc>] (__hrtimer_run_queues) from [<80198e60>] (hrtimer_run_queues+0xb8/0xf0) [<80198da8>] (hrtimer_run_queues) from [<801973e8>] (run_local_timers+0x28/0x64) [<801973c0>] (run_local_timers) from [<80197460>] (update_process_times+0x3c/0x6c) [<80197424>] (update_process_times) from [<801ab2b8>] (tick_nohz_handler+0xe0/0x1bc) [<801ab1d8>] (tick_nohz_handler) from [<80843050>] (arch_timer_handler_virt+0x38/0x48) [<80843018>] (arch_timer_handler_virt) from [<80180a64>] (handle_percpu_devid_irq+0x8c/0x240) [<801809d8>] (handle_percpu_devid_irq) from [<8017ac20>] (generic_handle_irq+0x34/0x44) [<8017abec>] (generic_handle_irq) from [<8017b344>] (__handle_domain_irq+0x6c/0xc4) [<8017b2d8>] (__handle_domain_irq) from [<801022e0>] (gic_handle_irq+0x4c/0x88) [<80102294>] (gic_handle_irq) from [<80101a30>] (__irq_svc+0x70/0x98) [<802b8794>] (generic_fillattr) from [<8056b284>] (v9fs_vfs_getattr_dotl+0x74/0xa4) [<8056b210>] (v9fs_vfs_getattr_dotl) from [<802b8904>] (vfs_getattr_nosec+0x68/0x7c) [<802b889c>] (vfs_getattr_nosec) from [<802b895c>] (vfs_getattr+0x44/0x48) [<802b8918>] (vfs_getattr) from [<802b8a74>] (vfs_statx+0x9c/0xec) [<802b89d8>] (vfs_statx) from [<802b9428>] (sys_lstat64+0x48/0x78) [<802b93e0>] (sys_lstat64) from [<80101000>] (ret_fast_syscall+0x0/0x28) [dominique.martinet@cea.fr: updated comment to not refer to a function in another subsystem] Link: http://lkml.kernel.org/r/20190124063514.8571-2-houtao1@huawei.com Cc: stable@vger.kernel.org Fixes: 7549ae3e81cc ("9p: Use the i_size_[read, write]() macros instead of using inode->i_size directly.") Reported-by: Xing Gaopeng <xingaopeng@huawei.com> Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
2019-01-24 14:35:13 +08:00
if (!(flags & V9FS_STAT2INODE_KEEP_ISIZE))
v9fs_i_size_write(inode, stat->st_size);
inode->i_blocks = stat->st_blocks;
} else {
if (stat->st_result_mask & P9_STATS_ATIME) {
inode->i_atime.tv_sec = stat->st_atime_sec;
inode->i_atime.tv_nsec = stat->st_atime_nsec;
}
if (stat->st_result_mask & P9_STATS_MTIME) {
inode->i_mtime.tv_sec = stat->st_mtime_sec;
inode->i_mtime.tv_nsec = stat->st_mtime_nsec;
}
if (stat->st_result_mask & P9_STATS_CTIME) {
inode->i_ctime.tv_sec = stat->st_ctime_sec;
inode->i_ctime.tv_nsec = stat->st_ctime_nsec;
}
if (stat->st_result_mask & P9_STATS_UID)
inode->i_uid = stat->st_uid;
if (stat->st_result_mask & P9_STATS_GID)
inode->i_gid = stat->st_gid;
if (stat->st_result_mask & P9_STATS_NLINK)
set_nlink(inode, stat->st_nlink);
if (stat->st_result_mask & P9_STATS_MODE) {
inode->i_mode = stat->st_mode;
if ((S_ISBLK(inode->i_mode)) ||
(S_ISCHR(inode->i_mode)))
init_special_inode(inode, inode->i_mode,
inode->i_rdev);
}
if (stat->st_result_mask & P9_STATS_RDEV)
inode->i_rdev = new_decode_dev(stat->st_rdev);
9p: use inode->i_lock to protect i_size_write() under 32-bit Use inode->i_lock to protect i_size_write(), else i_size_read() in generic_fillattr() may loop infinitely in read_seqcount_begin() when multiple processes invoke v9fs_vfs_getattr() or v9fs_vfs_getattr_dotl() simultaneously under 32-bit SMP environment, and a soft lockup will be triggered as show below: watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [stat:2217] Modules linked in: CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system PC is at generic_fillattr+0x104/0x108 LR is at 0xec497f00 pc : [<802b8898>] lr : [<ec497f00>] psr: 200c0013 sp : ec497e20 ip : ed608030 fp : ec497e3c r10: 00000000 r9 : ec497f00 r8 : ed608030 r7 : ec497ebc r6 : ec497f00 r5 : ee5c1550 r4 : ee005780 r3 : 0000052d r2 : 00000000 r1 : ec497f00 r0 : ed608030 Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: ac48006a DAC: 00000051 CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system Backtrace: [<8010d974>] (dump_backtrace) from [<8010dc88>] (show_stack+0x20/0x24) [<8010dc68>] (show_stack) from [<80a1d194>] (dump_stack+0xb0/0xdc) [<80a1d0e4>] (dump_stack) from [<80109f34>] (show_regs+0x1c/0x20) [<80109f18>] (show_regs) from [<801d0a80>] (watchdog_timer_fn+0x280/0x2f8) [<801d0800>] (watchdog_timer_fn) from [<80198658>] (__hrtimer_run_queues+0x18c/0x380) [<801984cc>] (__hrtimer_run_queues) from [<80198e60>] (hrtimer_run_queues+0xb8/0xf0) [<80198da8>] (hrtimer_run_queues) from [<801973e8>] (run_local_timers+0x28/0x64) [<801973c0>] (run_local_timers) from [<80197460>] (update_process_times+0x3c/0x6c) [<80197424>] (update_process_times) from [<801ab2b8>] (tick_nohz_handler+0xe0/0x1bc) [<801ab1d8>] (tick_nohz_handler) from [<80843050>] (arch_timer_handler_virt+0x38/0x48) [<80843018>] (arch_timer_handler_virt) from [<80180a64>] (handle_percpu_devid_irq+0x8c/0x240) [<801809d8>] (handle_percpu_devid_irq) from [<8017ac20>] (generic_handle_irq+0x34/0x44) [<8017abec>] (generic_handle_irq) from [<8017b344>] (__handle_domain_irq+0x6c/0xc4) [<8017b2d8>] (__handle_domain_irq) from [<801022e0>] (gic_handle_irq+0x4c/0x88) [<80102294>] (gic_handle_irq) from [<80101a30>] (__irq_svc+0x70/0x98) [<802b8794>] (generic_fillattr) from [<8056b284>] (v9fs_vfs_getattr_dotl+0x74/0xa4) [<8056b210>] (v9fs_vfs_getattr_dotl) from [<802b8904>] (vfs_getattr_nosec+0x68/0x7c) [<802b889c>] (vfs_getattr_nosec) from [<802b895c>] (vfs_getattr+0x44/0x48) [<802b8918>] (vfs_getattr) from [<802b8a74>] (vfs_statx+0x9c/0xec) [<802b89d8>] (vfs_statx) from [<802b9428>] (sys_lstat64+0x48/0x78) [<802b93e0>] (sys_lstat64) from [<80101000>] (ret_fast_syscall+0x0/0x28) [dominique.martinet@cea.fr: updated comment to not refer to a function in another subsystem] Link: http://lkml.kernel.org/r/20190124063514.8571-2-houtao1@huawei.com Cc: stable@vger.kernel.org Fixes: 7549ae3e81cc ("9p: Use the i_size_[read, write]() macros instead of using inode->i_size directly.") Reported-by: Xing Gaopeng <xingaopeng@huawei.com> Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
2019-01-24 14:35:13 +08:00
if (!(flags & V9FS_STAT2INODE_KEEP_ISIZE) &&
stat->st_result_mask & P9_STATS_SIZE)
v9fs_i_size_write(inode, stat->st_size);
if (stat->st_result_mask & P9_STATS_BLOCKS)
inode->i_blocks = stat->st_blocks;
}
if (stat->st_result_mask & P9_STATS_GEN)
inode->i_generation = stat->st_gen;
/* Currently we don't support P9_STATS_BTIME and P9_STATS_DATA_VERSION
* because the inode structure does not have fields for them.
*/
v9inode->cache_validity &= ~V9FS_INO_INVALID_ATTR;
}
static int
v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry,
const char *symname)
{
int err;
kgid_t gid;
const unsigned char *name;
struct p9_qid qid;
struct inode *inode;
struct p9_fid *dfid;
struct p9_fid *fid = NULL;
struct v9fs_session_info *v9ses;
name = dentry->d_name.name;
p9_debug(P9_DEBUG_VFS, "%lu,%s,%s\n", dir->i_ino, name, symname);
v9ses = v9fs_inode2v9ses(dir);
dfid = v9fs_parent_fid(dentry);
if (IS_ERR(dfid)) {
err = PTR_ERR(dfid);
p9_debug(P9_DEBUG_VFS, "fid lookup failed %d\n", err);
return err;
}
gid = v9fs_get_fsgid_for_create(dir);
/* Server doesn't alter fid on TSYMLINK. Hence no need to clone it. */
err = p9_client_symlink(dfid, name, symname, gid, &qid);
if (err < 0) {
p9_debug(P9_DEBUG_VFS, "p9_client_symlink failed %d\n", err);
goto error;
}
v9fs_invalidate_inode_attr(dir);
if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) {
/* Now walk from the parent so we can get an unopened fid. */
fid = p9_client_walk(dfid, 1, &name, 1);
if (IS_ERR(fid)) {
err = PTR_ERR(fid);
p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n",
err);
fid = NULL;
goto error;
}
/* instantiate inode and assign the unopened fid to dentry */
inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n",
err);
goto error;
}
v9fs_fid_add(dentry, fid);
d_instantiate(dentry, inode);
fid = NULL;
err = 0;
} else {
/* Not in cached mode. No need to populate inode with stat */
inode = v9fs_get_inode(dir->i_sb, S_IFLNK, 0);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
goto error;
}
d_instantiate(dentry, inode);
}
error:
if (fid)
p9_client_clunk(fid);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
p9_client_clunk(dfid);
return err;
}
/**
* v9fs_vfs_link_dotl - create a hardlink for dotl
* @old_dentry: dentry for file to link to
* @dir: inode destination for new link
* @dentry: dentry for link
*
*/
static int
v9fs_vfs_link_dotl(struct dentry *old_dentry, struct inode *dir,
struct dentry *dentry)
{
int err;
struct p9_fid *dfid, *oldfid;
struct v9fs_session_info *v9ses;
p9_debug(P9_DEBUG_VFS, "dir ino: %lu, old_name: %pd, new_name: %pd\n",
dir->i_ino, old_dentry, dentry);
v9ses = v9fs_inode2v9ses(dir);
dfid = v9fs_parent_fid(dentry);
if (IS_ERR(dfid))
return PTR_ERR(dfid);
oldfid = v9fs_fid_lookup(old_dentry);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
if (IS_ERR(oldfid)) {
p9_client_clunk(dfid);
return PTR_ERR(oldfid);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
}
err = p9_client_link(dfid, oldfid, dentry->d_name.name);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
p9_client_clunk(dfid);
p9_client_clunk(oldfid);
if (err < 0) {
p9_debug(P9_DEBUG_VFS, "p9_client_link failed %d\n", err);
return err;
}
v9fs_invalidate_inode_attr(dir);
if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) {
/* Get the latest stat info from server. */
struct p9_fid *fid;
fid = v9fs_fid_lookup(old_dentry);
if (IS_ERR(fid))
return PTR_ERR(fid);
v9fs_refresh_inode_dotl(fid, d_inode(old_dentry));
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
p9_client_clunk(fid);
}
ihold(d_inode(old_dentry));
d_instantiate(dentry, d_inode(old_dentry));
return err;
}
/**
* v9fs_vfs_mknod_dotl - create a special file
* @dir: inode destination for new link
* @dentry: dentry for file
* @omode: mode for creation
* @rdev: device associated with special file
*
*/
static int
v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, umode_t omode,
dev_t rdev)
{
int err;
kgid_t gid;
const unsigned char *name;
umode_t mode;
struct v9fs_session_info *v9ses;
struct p9_fid *fid = NULL, *dfid = NULL;
struct inode *inode;
struct p9_qid qid;
struct posix_acl *dacl = NULL, *pacl = NULL;
p9_debug(P9_DEBUG_VFS, " %lu,%pd mode: %hx MAJOR: %u MINOR: %u\n",
dir->i_ino, dentry, omode,
MAJOR(rdev), MINOR(rdev));
v9ses = v9fs_inode2v9ses(dir);
dfid = v9fs_parent_fid(dentry);
if (IS_ERR(dfid)) {
err = PTR_ERR(dfid);
p9_debug(P9_DEBUG_VFS, "fid lookup failed %d\n", err);
dfid = NULL;
goto error;
}
gid = v9fs_get_fsgid_for_create(dir);
mode = omode;
/* Update mode based on ACL value */
err = v9fs_acl_mode(dir, &mode, &dacl, &pacl);
if (err) {
p9_debug(P9_DEBUG_VFS, "Failed to get acl values in mknod %d\n",
err);
goto error;
}
name = dentry->d_name.name;
err = p9_client_mknod_dotl(dfid, name, mode, rdev, gid, &qid);
if (err < 0)
goto error;
v9fs_invalidate_inode_attr(dir);
fid = p9_client_walk(dfid, 1, &name, 1);
if (IS_ERR(fid)) {
err = PTR_ERR(fid);
p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n",
err);
fid = NULL;
goto error;
}
/* instantiate inode and assign the unopened fid to the dentry */
if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) {
inode = v9fs_get_new_inode_from_fid(v9ses, fid, dir->i_sb);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
p9_debug(P9_DEBUG_VFS, "inode creation failed %d\n",
err);
goto error;
}
v9fs_set_create_acl(inode, fid, dacl, pacl);
v9fs_fid_add(dentry, fid);
d_instantiate(dentry, inode);
fid = NULL;
err = 0;
} else {
/*
* Not in cached mode. No need to populate inode with stat.
* socket syscall returns a fd, so we need instantiate
*/
inode = v9fs_get_inode(dir->i_sb, mode, rdev);
if (IS_ERR(inode)) {
err = PTR_ERR(inode);
goto error;
}
v9fs_set_create_acl(inode, fid, dacl, pacl);
d_instantiate(dentry, inode);
}
error:
if (fid)
p9_client_clunk(fid);
v9fs_put_acl(dacl, pacl);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
p9_client_clunk(dfid);
return err;
}
/**
* v9fs_vfs_get_link_dotl - follow a symlink path
* @dentry: dentry for symlink
* @inode: inode for symlink
* @done: destructor for return value
*/
static const char *
v9fs_vfs_get_link_dotl(struct dentry *dentry,
struct inode *inode,
struct delayed_call *done)
{
struct p9_fid *fid;
char *target;
int retval;
if (!dentry)
return ERR_PTR(-ECHILD);
p9_debug(P9_DEBUG_VFS, "%pd\n", dentry);
fid = v9fs_fid_lookup(dentry);
if (IS_ERR(fid))
return ERR_CAST(fid);
retval = p9_client_readlink(fid, &target);
9p: add refcount to p9_fid struct Fix race issue in fid contention. Eric's and Greg's patch offer a mechanism to fix open-unlink-f*syscall bug in 9p. But there is race issue in fid parallel accesses. As Greg's patch stores all of fids from opened files into according inode, so all the lookup fid ops can retrieve fid from inode preferentially. But there is no mechanism to handle the fid contention issue. For example, there are two threads get the same fid in the same time and one of them clunk the fid before the other thread ready to discard the fid. In this scenario, it will lead to some fatal problems, even kernel core dump. I introduce a mechanism to fix this race issue. A counter field introduced into p9_fid struct to store the reference counter to the fid. When a fid is allocated from the inode or dentry, the counter will increase, and will decrease at the end of its occupation. It is guaranteed that the fid won't be clunked before the reference counter go down to 0, then we can avoid the clunked fid to be used. tests: race issue test from the old test case: for file in {01..50}; do touch f.${file}; done seq 1 1000 | xargs -n 1 -P 50 -I{} cat f.* > /dev/null open-unlink-f*syscall test: I have tested for f*syscall include: ftruncate fstat fchown fchmod faccessat. Link: http://lkml.kernel.org/r/20200923141146.90046-5-jianyong.wu@arm.com Fixes: 478ba09edc1f ("fs/9p: search open fids first") Signed-off-by: Jianyong Wu <jianyong.wu@arm.com> Signed-off-by: Dominique Martinet <asmadeus@codewreck.org>
2020-09-23 22:11:46 +08:00
p9_client_clunk(fid);
if (retval)
return ERR_PTR(retval);
set_delayed_call(done, kfree_link, target);
return target;
}
int v9fs_refresh_inode_dotl(struct p9_fid *fid, struct inode *inode)
{
struct p9_stat_dotl *st;
struct v9fs_session_info *v9ses;
9p: use inode->i_lock to protect i_size_write() under 32-bit Use inode->i_lock to protect i_size_write(), else i_size_read() in generic_fillattr() may loop infinitely in read_seqcount_begin() when multiple processes invoke v9fs_vfs_getattr() or v9fs_vfs_getattr_dotl() simultaneously under 32-bit SMP environment, and a soft lockup will be triggered as show below: watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [stat:2217] Modules linked in: CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system PC is at generic_fillattr+0x104/0x108 LR is at 0xec497f00 pc : [<802b8898>] lr : [<ec497f00>] psr: 200c0013 sp : ec497e20 ip : ed608030 fp : ec497e3c r10: 00000000 r9 : ec497f00 r8 : ed608030 r7 : ec497ebc r6 : ec497f00 r5 : ee5c1550 r4 : ee005780 r3 : 0000052d r2 : 00000000 r1 : ec497f00 r0 : ed608030 Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: ac48006a DAC: 00000051 CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system Backtrace: [<8010d974>] (dump_backtrace) from [<8010dc88>] (show_stack+0x20/0x24) [<8010dc68>] (show_stack) from [<80a1d194>] (dump_stack+0xb0/0xdc) [<80a1d0e4>] (dump_stack) from [<80109f34>] (show_regs+0x1c/0x20) [<80109f18>] (show_regs) from [<801d0a80>] (watchdog_timer_fn+0x280/0x2f8) [<801d0800>] (watchdog_timer_fn) from [<80198658>] (__hrtimer_run_queues+0x18c/0x380) [<801984cc>] (__hrtimer_run_queues) from [<80198e60>] (hrtimer_run_queues+0xb8/0xf0) [<80198da8>] (hrtimer_run_queues) from [<801973e8>] (run_local_timers+0x28/0x64) [<801973c0>] (run_local_timers) from [<80197460>] (update_process_times+0x3c/0x6c) [<80197424>] (update_process_times) from [<801ab2b8>] (tick_nohz_handler+0xe0/0x1bc) [<801ab1d8>] (tick_nohz_handler) from [<80843050>] (arch_timer_handler_virt+0x38/0x48) [<80843018>] (arch_timer_handler_virt) from [<80180a64>] (handle_percpu_devid_irq+0x8c/0x240) [<801809d8>] (handle_percpu_devid_irq) from [<8017ac20>] (generic_handle_irq+0x34/0x44) [<8017abec>] (generic_handle_irq) from [<8017b344>] (__handle_domain_irq+0x6c/0xc4) [<8017b2d8>] (__handle_domain_irq) from [<801022e0>] (gic_handle_irq+0x4c/0x88) [<80102294>] (gic_handle_irq) from [<80101a30>] (__irq_svc+0x70/0x98) [<802b8794>] (generic_fillattr) from [<8056b284>] (v9fs_vfs_getattr_dotl+0x74/0xa4) [<8056b210>] (v9fs_vfs_getattr_dotl) from [<802b8904>] (vfs_getattr_nosec+0x68/0x7c) [<802b889c>] (vfs_getattr_nosec) from [<802b895c>] (vfs_getattr+0x44/0x48) [<802b8918>] (vfs_getattr) from [<802b8a74>] (vfs_statx+0x9c/0xec) [<802b89d8>] (vfs_statx) from [<802b9428>] (sys_lstat64+0x48/0x78) [<802b93e0>] (sys_lstat64) from [<80101000>] (ret_fast_syscall+0x0/0x28) [dominique.martinet@cea.fr: updated comment to not refer to a function in another subsystem] Link: http://lkml.kernel.org/r/20190124063514.8571-2-houtao1@huawei.com Cc: stable@vger.kernel.org Fixes: 7549ae3e81cc ("9p: Use the i_size_[read, write]() macros instead of using inode->i_size directly.") Reported-by: Xing Gaopeng <xingaopeng@huawei.com> Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
2019-01-24 14:35:13 +08:00
unsigned int flags;
v9ses = v9fs_inode2v9ses(inode);
st = p9_client_getattr_dotl(fid, P9_STATS_ALL);
if (IS_ERR(st))
return PTR_ERR(st);
/*
* Don't update inode if the file type is different
*/
if ((inode->i_mode & S_IFMT) != (st->st_mode & S_IFMT))
goto out;
/*
* We don't want to refresh inode->i_size,
* because we may have cached data
*/
9p: use inode->i_lock to protect i_size_write() under 32-bit Use inode->i_lock to protect i_size_write(), else i_size_read() in generic_fillattr() may loop infinitely in read_seqcount_begin() when multiple processes invoke v9fs_vfs_getattr() or v9fs_vfs_getattr_dotl() simultaneously under 32-bit SMP environment, and a soft lockup will be triggered as show below: watchdog: BUG: soft lockup - CPU#5 stuck for 22s! [stat:2217] Modules linked in: CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system PC is at generic_fillattr+0x104/0x108 LR is at 0xec497f00 pc : [<802b8898>] lr : [<ec497f00>] psr: 200c0013 sp : ec497e20 ip : ed608030 fp : ec497e3c r10: 00000000 r9 : ec497f00 r8 : ed608030 r7 : ec497ebc r6 : ec497f00 r5 : ee5c1550 r4 : ee005780 r3 : 0000052d r2 : 00000000 r1 : ec497f00 r0 : ed608030 Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none Control: 10c5387d Table: ac48006a DAC: 00000051 CPU: 5 PID: 2217 Comm: stat Not tainted 5.0.0-rc1-00005-g7f702faf5a9e #4 Hardware name: Generic DT based system Backtrace: [<8010d974>] (dump_backtrace) from [<8010dc88>] (show_stack+0x20/0x24) [<8010dc68>] (show_stack) from [<80a1d194>] (dump_stack+0xb0/0xdc) [<80a1d0e4>] (dump_stack) from [<80109f34>] (show_regs+0x1c/0x20) [<80109f18>] (show_regs) from [<801d0a80>] (watchdog_timer_fn+0x280/0x2f8) [<801d0800>] (watchdog_timer_fn) from [<80198658>] (__hrtimer_run_queues+0x18c/0x380) [<801984cc>] (__hrtimer_run_queues) from [<80198e60>] (hrtimer_run_queues+0xb8/0xf0) [<80198da8>] (hrtimer_run_queues) from [<801973e8>] (run_local_timers+0x28/0x64) [<801973c0>] (run_local_timers) from [<80197460>] (update_process_times+0x3c/0x6c) [<80197424>] (update_process_times) from [<801ab2b8>] (tick_nohz_handler+0xe0/0x1bc) [<801ab1d8>] (tick_nohz_handler) from [<80843050>] (arch_timer_handler_virt+0x38/0x48) [<80843018>] (arch_timer_handler_virt) from [<80180a64>] (handle_percpu_devid_irq+0x8c/0x240) [<801809d8>] (handle_percpu_devid_irq) from [<8017ac20>] (generic_handle_irq+0x34/0x44) [<8017abec>] (generic_handle_irq) from [<8017b344>] (__handle_domain_irq+0x6c/0xc4) [<8017b2d8>] (__handle_domain_irq) from [<801022e0>] (gic_handle_irq+0x4c/0x88) [<80102294>] (gic_handle_irq) from [<80101a30>] (__irq_svc+0x70/0x98) [<802b8794>] (generic_fillattr) from [<8056b284>] (v9fs_vfs_getattr_dotl+0x74/0xa4) [<8056b210>] (v9fs_vfs_getattr_dotl) from [<802b8904>] (vfs_getattr_nosec+0x68/0x7c) [<802b889c>] (vfs_getattr_nosec) from [<802b895c>] (vfs_getattr+0x44/0x48) [<802b8918>] (vfs_getattr) from [<802b8a74>] (vfs_statx+0x9c/0xec) [<802b89d8>] (vfs_statx) from [<802b9428>] (sys_lstat64+0x48/0x78) [<802b93e0>] (sys_lstat64) from [<80101000>] (ret_fast_syscall+0x0/0x28) [dominique.martinet@cea.fr: updated comment to not refer to a function in another subsystem] Link: http://lkml.kernel.org/r/20190124063514.8571-2-houtao1@huawei.com Cc: stable@vger.kernel.org Fixes: 7549ae3e81cc ("9p: Use the i_size_[read, write]() macros instead of using inode->i_size directly.") Reported-by: Xing Gaopeng <xingaopeng@huawei.com> Signed-off-by: Hou Tao <houtao1@huawei.com> Signed-off-by: Dominique Martinet <dominique.martinet@cea.fr>
2019-01-24 14:35:13 +08:00
flags = (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) ?
V9FS_STAT2INODE_KEEP_ISIZE : 0;
v9fs_stat2inode_dotl(st, inode, flags);
out:
kfree(st);
return 0;
}
const struct inode_operations v9fs_dir_inode_operations_dotl = {
.create = v9fs_vfs_create_dotl,
.atomic_open = v9fs_vfs_atomic_open_dotl,
.lookup = v9fs_vfs_lookup,
.link = v9fs_vfs_link_dotl,
.symlink = v9fs_vfs_symlink_dotl,
.unlink = v9fs_vfs_unlink,
.mkdir = v9fs_vfs_mkdir_dotl,
.rmdir = v9fs_vfs_rmdir,
.mknod = v9fs_vfs_mknod_dotl,
.rename = v9fs_vfs_rename,
.getattr = v9fs_vfs_getattr_dotl,
.setattr = v9fs_vfs_setattr_dotl,
.listxattr = v9fs_listxattr,
.get_acl = v9fs_iop_get_acl,
};
const struct inode_operations v9fs_file_inode_operations_dotl = {
.getattr = v9fs_vfs_getattr_dotl,
.setattr = v9fs_vfs_setattr_dotl,
.listxattr = v9fs_listxattr,
.get_acl = v9fs_iop_get_acl,
};
const struct inode_operations v9fs_symlink_inode_operations_dotl = {
.get_link = v9fs_vfs_get_link_dotl,
.getattr = v9fs_vfs_getattr_dotl,
.setattr = v9fs_vfs_setattr_dotl,
.listxattr = v9fs_listxattr,
};