2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
#include <linux/ceph/ceph_debug.h>
|
2009-10-07 02:31:07 +08:00
|
|
|
|
|
|
|
|
#include <linux/backing-dev.h>
|
2010-07-01 12:34:01 +08:00
|
|
|
#include <linux/ctype.h>
|
2009-10-07 02:31:07 +08:00
|
|
|
#include <linux/fs.h>
|
|
|
|
|
#include <linux/inet.h>
|
|
|
|
|
#include <linux/in6.h>
|
|
|
|
|
#include <linux/module.h>
|
|
|
|
|
#include <linux/mount.h>
|
|
|
|
|
#include <linux/parser.h>
|
|
|
|
|
#include <linux/sched.h>
|
|
|
|
|
#include <linux/seq_file.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
|
|
|
#include <linux/slab.h>
|
2009-10-07 02:31:07 +08:00
|
|
|
#include <linux/statfs.h>
|
|
|
|
|
#include <linux/string.h>
|
|
|
|
|
|
|
|
|
|
#include "super.h"
|
2010-04-07 06:14:15 +08:00
|
|
|
#include "mds_client.h"
|
2013-08-22 05:29:54 +08:00
|
|
|
#include "cache.h"
|
2010-04-07 06:14:15 +08:00
|
|
|
|
2012-07-31 07:23:22 +08:00
|
|
|
#include <linux/ceph/ceph_features.h>
|
2010-04-07 06:14:15 +08:00
|
|
|
#include <linux/ceph/decode.h>
|
|
|
|
|
#include <linux/ceph/mon_client.h>
|
|
|
|
|
#include <linux/ceph/auth.h>
|
|
|
|
|
#include <linux/ceph/debugfs.h>
|
2009-10-07 02:31:07 +08:00
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* Ceph superblock operations
|
|
|
|
|
*
|
|
|
|
|
* Handle the basics of mounting, unmounting.
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* super ops
|
|
|
|
|
*/
|
|
|
|
|
static void ceph_put_super(struct super_block *s)
|
|
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_fs_client *fsc = ceph_sb_to_client(s);
|
2009-10-07 02:31:07 +08:00
|
|
|
|
|
|
|
|
dout("put_super\n");
|
2010-04-07 06:14:15 +08:00
|
|
|
ceph_mdsc_close_sessions(fsc->mdsc);
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf)
|
|
|
|
|
{
|
2015-03-18 06:25:59 +08:00
|
|
|
struct ceph_fs_client *fsc = ceph_inode_to_client(d_inode(dentry));
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_monmap *monmap = fsc->client->monc.monmap;
|
2009-10-07 02:31:07 +08:00
|
|
|
struct ceph_statfs st;
|
|
|
|
|
u64 fsid;
|
|
|
|
|
int err;
|
2017-08-16 22:19:27 +08:00
|
|
|
u64 data_pool;
|
|
|
|
|
|
|
|
|
|
if (fsc->mdsc->mdsmap->m_num_data_pg_pools == 1) {
|
|
|
|
|
data_pool = fsc->mdsc->mdsmap->m_data_pg_pools[0];
|
|
|
|
|
} else {
|
|
|
|
|
data_pool = CEPH_NOPOOL;
|
|
|
|
|
}
|
2009-10-07 02:31:07 +08:00
|
|
|
|
|
|
|
|
dout("statfs\n");
|
2017-08-16 22:19:27 +08:00
|
|
|
err = ceph_monc_do_statfs(&fsc->client->monc, data_pool, &st);
|
2009-10-07 02:31:07 +08:00
|
|
|
if (err < 0)
|
|
|
|
|
return err;
|
|
|
|
|
|
|
|
|
|
/* fill in kstatfs */
|
|
|
|
|
buf->f_type = CEPH_SUPER_MAGIC; /* ?? */
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* express utilization in terms of large blocks to avoid
|
|
|
|
|
* overflow on 32-bit machines.
|
2013-02-23 07:31:00 +08:00
|
|
|
*
|
|
|
|
|
* NOTE: for the time being, we make bsize == frsize to humor
|
|
|
|
|
* not-yet-ancient versions of glibc that are broken.
|
|
|
|
|
* Someday, we will probably want to report a real block
|
|
|
|
|
* size... whatever that may mean for a network file system!
|
2009-10-07 02:31:07 +08:00
|
|
|
*/
|
|
|
|
|
buf->f_bsize = 1 << CEPH_BLOCK_SHIFT;
|
2013-02-23 07:31:00 +08:00
|
|
|
buf->f_frsize = 1 << CEPH_BLOCK_SHIFT;
|
2009-10-07 02:31:07 +08:00
|
|
|
buf->f_blocks = le64_to_cpu(st.kb) >> (CEPH_BLOCK_SHIFT-10);
|
2011-07-27 02:26:54 +08:00
|
|
|
buf->f_bfree = le64_to_cpu(st.kb_avail) >> (CEPH_BLOCK_SHIFT-10);
|
2009-10-07 02:31:07 +08:00
|
|
|
buf->f_bavail = le64_to_cpu(st.kb_avail) >> (CEPH_BLOCK_SHIFT-10);
|
|
|
|
|
|
|
|
|
|
buf->f_files = le64_to_cpu(st.num_objects);
|
|
|
|
|
buf->f_ffree = -1;
|
2010-06-02 03:51:12 +08:00
|
|
|
buf->f_namelen = NAME_MAX;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2017-10-23 22:58:40 +08:00
|
|
|
/* Must convert the fsid, for consistent values across arches */
|
|
|
|
|
fsid = le64_to_cpu(*(__le64 *)(&monmap->fsid)) ^
|
|
|
|
|
le64_to_cpu(*((__le64 *)&monmap->fsid + 1));
|
2009-10-07 02:31:07 +08:00
|
|
|
buf->f_fsid.val[0] = fsid & 0xffffffff;
|
|
|
|
|
buf->f_fsid.val[1] = fsid >> 32;
|
|
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
2010-07-31 00:38:13 +08:00
|
|
|
static int ceph_sync_fs(struct super_block *sb, int wait)
|
2009-10-07 02:31:07 +08:00
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
|
2010-07-31 00:38:13 +08:00
|
|
|
|
|
|
|
|
if (!wait) {
|
|
|
|
|
dout("sync_fs (non-blocking)\n");
|
2010-04-07 06:14:15 +08:00
|
|
|
ceph_flush_dirty_caps(fsc->mdsc);
|
2010-07-31 00:38:13 +08:00
|
|
|
dout("sync_fs (non-blocking) done\n");
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
dout("sync_fs (blocking)\n");
|
2010-04-07 06:14:15 +08:00
|
|
|
ceph_osdc_sync(&fsc->client->osdc);
|
|
|
|
|
ceph_mdsc_sync(fsc->mdsc);
|
2010-07-31 00:38:13 +08:00
|
|
|
dout("sync_fs (blocking) done\n");
|
2009-10-07 02:31:07 +08:00
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* mount options
|
|
|
|
|
*/
|
|
|
|
|
enum {
|
|
|
|
|
Opt_wsize,
|
|
|
|
|
Opt_rsize,
|
2011-08-04 23:03:44 +08:00
|
|
|
Opt_rasize,
|
2009-10-07 02:31:07 +08:00
|
|
|
Opt_caps_wanted_delay_min,
|
|
|
|
|
Opt_caps_wanted_delay_max,
|
|
|
|
|
Opt_readdir_max_entries,
|
2010-05-15 04:06:30 +08:00
|
|
|
Opt_readdir_max_bytes,
|
2009-12-19 05:51:57 +08:00
|
|
|
Opt_congestion_kb,
|
2009-10-28 01:19:28 +08:00
|
|
|
Opt_last_int,
|
2009-10-07 02:31:07 +08:00
|
|
|
/* int args above */
|
|
|
|
|
Opt_snapdirname,
|
2016-07-08 11:25:38 +08:00
|
|
|
Opt_mds_namespace,
|
2017-06-27 11:57:56 +08:00
|
|
|
Opt_fscache_uniq,
|
2009-10-28 01:19:28 +08:00
|
|
|
Opt_last_string,
|
2009-10-07 02:31:07 +08:00
|
|
|
/* string args above */
|
|
|
|
|
Opt_dirstat,
|
|
|
|
|
Opt_nodirstat,
|
|
|
|
|
Opt_rbytes,
|
|
|
|
|
Opt_norbytes,
|
2012-02-15 21:43:54 +08:00
|
|
|
Opt_asyncreaddir,
|
2009-10-07 02:31:07 +08:00
|
|
|
Opt_noasyncreaddir,
|
2012-01-11 01:12:55 +08:00
|
|
|
Opt_dcache,
|
|
|
|
|
Opt_nodcache,
|
2011-01-22 08:44:03 +08:00
|
|
|
Opt_ino32,
|
2012-02-15 21:43:54 +08:00
|
|
|
Opt_noino32,
|
2013-08-22 05:29:54 +08:00
|
|
|
Opt_fscache,
|
2014-02-17 02:05:29 +08:00
|
|
|
Opt_nofscache,
|
2015-04-27 15:33:28 +08:00
|
|
|
Opt_poolperm,
|
|
|
|
|
Opt_nopoolperm,
|
2016-11-10 16:02:06 +08:00
|
|
|
Opt_require_active_mds,
|
|
|
|
|
Opt_norequire_active_mds,
|
2014-02-17 02:05:29 +08:00
|
|
|
#ifdef CONFIG_CEPH_FS_POSIX_ACL
|
|
|
|
|
Opt_acl,
|
|
|
|
|
#endif
|
2015-04-27 15:33:28 +08:00
|
|
|
Opt_noacl,
|
2009-10-07 02:31:07 +08:00
|
|
|
};
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
static match_table_t fsopt_tokens = {
|
2009-10-07 02:31:07 +08:00
|
|
|
{Opt_wsize, "wsize=%d"},
|
|
|
|
|
{Opt_rsize, "rsize=%d"},
|
2011-08-04 23:03:44 +08:00
|
|
|
{Opt_rasize, "rasize=%d"},
|
2009-10-07 02:31:07 +08:00
|
|
|
{Opt_caps_wanted_delay_min, "caps_wanted_delay_min=%d"},
|
|
|
|
|
{Opt_caps_wanted_delay_max, "caps_wanted_delay_max=%d"},
|
|
|
|
|
{Opt_readdir_max_entries, "readdir_max_entries=%d"},
|
2010-05-15 04:06:30 +08:00
|
|
|
{Opt_readdir_max_bytes, "readdir_max_bytes=%d"},
|
2009-12-19 05:51:57 +08:00
|
|
|
{Opt_congestion_kb, "write_congestion_kb=%d"},
|
2009-10-07 02:31:07 +08:00
|
|
|
/* int args above */
|
|
|
|
|
{Opt_snapdirname, "snapdirname=%s"},
|
2016-07-08 11:25:38 +08:00
|
|
|
{Opt_mds_namespace, "mds_namespace=%s"},
|
2017-06-27 11:57:56 +08:00
|
|
|
{Opt_fscache_uniq, "fsc=%s"},
|
2009-10-07 02:31:07 +08:00
|
|
|
/* string args above */
|
|
|
|
|
{Opt_dirstat, "dirstat"},
|
|
|
|
|
{Opt_nodirstat, "nodirstat"},
|
|
|
|
|
{Opt_rbytes, "rbytes"},
|
|
|
|
|
{Opt_norbytes, "norbytes"},
|
2012-02-15 21:43:54 +08:00
|
|
|
{Opt_asyncreaddir, "asyncreaddir"},
|
2009-10-07 02:31:07 +08:00
|
|
|
{Opt_noasyncreaddir, "noasyncreaddir"},
|
2012-01-11 01:12:55 +08:00
|
|
|
{Opt_dcache, "dcache"},
|
|
|
|
|
{Opt_nodcache, "nodcache"},
|
2011-01-22 08:44:03 +08:00
|
|
|
{Opt_ino32, "ino32"},
|
2012-02-15 21:43:54 +08:00
|
|
|
{Opt_noino32, "noino32"},
|
2013-08-22 05:29:54 +08:00
|
|
|
{Opt_fscache, "fsc"},
|
|
|
|
|
{Opt_nofscache, "nofsc"},
|
2015-04-27 15:33:28 +08:00
|
|
|
{Opt_poolperm, "poolperm"},
|
|
|
|
|
{Opt_nopoolperm, "nopoolperm"},
|
2016-11-10 16:02:06 +08:00
|
|
|
{Opt_require_active_mds, "require_active_mds"},
|
|
|
|
|
{Opt_norequire_active_mds, "norequire_active_mds"},
|
2014-02-17 02:05:29 +08:00
|
|
|
#ifdef CONFIG_CEPH_FS_POSIX_ACL
|
|
|
|
|
{Opt_acl, "acl"},
|
|
|
|
|
#endif
|
|
|
|
|
{Opt_noacl, "noacl"},
|
2009-10-07 02:31:07 +08:00
|
|
|
{-1, NULL}
|
|
|
|
|
};
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
static int parse_fsopt_token(char *c, void *private)
|
2010-07-01 12:34:01 +08:00
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_mount_options *fsopt = private;
|
|
|
|
|
substring_t argstr[MAX_OPT_ARGS];
|
|
|
|
|
int token, intval, ret;
|
|
|
|
|
|
|
|
|
|
token = match_token((char *)c, fsopt_tokens, argstr);
|
|
|
|
|
if (token < 0)
|
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
|
|
if (token < Opt_last_int) {
|
|
|
|
|
ret = match_int(&argstr[0], &intval);
|
|
|
|
|
if (ret < 0) {
|
|
|
|
|
pr_err("bad mount option arg (not int) "
|
|
|
|
|
"at '%s'\n", c);
|
|
|
|
|
return ret;
|
2010-07-01 12:34:01 +08:00
|
|
|
}
|
2010-04-07 06:14:15 +08:00
|
|
|
dout("got int token %d val %d\n", token, intval);
|
|
|
|
|
} else if (token > Opt_last_int && token < Opt_last_string) {
|
|
|
|
|
dout("got string token %d val %s\n", token,
|
|
|
|
|
argstr[0].from);
|
|
|
|
|
} else {
|
|
|
|
|
dout("got token %d\n", token);
|
2010-07-01 12:34:01 +08:00
|
|
|
}
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
switch (token) {
|
|
|
|
|
case Opt_snapdirname:
|
|
|
|
|
kfree(fsopt->snapdir_name);
|
|
|
|
|
fsopt->snapdir_name = kstrndup(argstr[0].from,
|
|
|
|
|
argstr[0].to-argstr[0].from,
|
|
|
|
|
GFP_KERNEL);
|
|
|
|
|
if (!fsopt->snapdir_name)
|
|
|
|
|
return -ENOMEM;
|
|
|
|
|
break;
|
2016-03-30 17:18:34 +08:00
|
|
|
case Opt_mds_namespace:
|
2016-07-08 11:25:38 +08:00
|
|
|
fsopt->mds_namespace = kstrndup(argstr[0].from,
|
|
|
|
|
argstr[0].to-argstr[0].from,
|
|
|
|
|
GFP_KERNEL);
|
|
|
|
|
if (!fsopt->mds_namespace)
|
|
|
|
|
return -ENOMEM;
|
2016-03-30 17:18:34 +08:00
|
|
|
break;
|
2017-06-27 11:57:56 +08:00
|
|
|
case Opt_fscache_uniq:
|
|
|
|
|
fsopt->fscache_uniq = kstrndup(argstr[0].from,
|
|
|
|
|
argstr[0].to-argstr[0].from,
|
|
|
|
|
GFP_KERNEL);
|
|
|
|
|
if (!fsopt->fscache_uniq)
|
|
|
|
|
return -ENOMEM;
|
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_FSCACHE;
|
|
|
|
|
break;
|
2016-07-08 11:25:38 +08:00
|
|
|
/* misc */
|
2010-04-07 06:14:15 +08:00
|
|
|
case Opt_wsize:
|
2017-07-11 17:34:46 +08:00
|
|
|
if (intval < PAGE_SIZE || intval > CEPH_MAX_WRITE_SIZE)
|
|
|
|
|
return -EINVAL;
|
|
|
|
|
fsopt->wsize = ALIGN(intval, PAGE_SIZE);
|
2010-04-07 06:14:15 +08:00
|
|
|
break;
|
|
|
|
|
case Opt_rsize:
|
2017-07-11 15:56:09 +08:00
|
|
|
if (intval < PAGE_SIZE || intval > CEPH_MAX_READ_SIZE)
|
|
|
|
|
return -EINVAL;
|
|
|
|
|
fsopt->rsize = ALIGN(intval, PAGE_SIZE);
|
2010-04-07 06:14:15 +08:00
|
|
|
break;
|
2011-08-04 23:03:44 +08:00
|
|
|
case Opt_rasize:
|
2017-07-11 18:49:44 +08:00
|
|
|
if (intval < 0)
|
|
|
|
|
return -EINVAL;
|
|
|
|
|
fsopt->rasize = ALIGN(intval + PAGE_SIZE - 1, PAGE_SIZE);
|
2011-08-04 23:03:44 +08:00
|
|
|
break;
|
2010-04-07 06:14:15 +08:00
|
|
|
case Opt_caps_wanted_delay_min:
|
2017-07-11 18:49:44 +08:00
|
|
|
if (intval < 1)
|
|
|
|
|
return -EINVAL;
|
2010-04-07 06:14:15 +08:00
|
|
|
fsopt->caps_wanted_delay_min = intval;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_caps_wanted_delay_max:
|
2017-07-11 18:49:44 +08:00
|
|
|
if (intval < 1)
|
|
|
|
|
return -EINVAL;
|
2010-04-07 06:14:15 +08:00
|
|
|
fsopt->caps_wanted_delay_max = intval;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_readdir_max_entries:
|
2017-07-11 18:49:44 +08:00
|
|
|
if (intval < 1)
|
|
|
|
|
return -EINVAL;
|
2010-04-07 06:14:15 +08:00
|
|
|
fsopt->max_readdir = intval;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_readdir_max_bytes:
|
2017-07-11 18:49:44 +08:00
|
|
|
if (intval < PAGE_SIZE && intval != 0)
|
|
|
|
|
return -EINVAL;
|
2010-04-07 06:14:15 +08:00
|
|
|
fsopt->max_readdir_bytes = intval;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_congestion_kb:
|
2017-07-11 18:49:44 +08:00
|
|
|
if (intval < 1024) /* at least 1M */
|
|
|
|
|
return -EINVAL;
|
2010-04-07 06:14:15 +08:00
|
|
|
fsopt->congestion_kb = intval;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_dirstat:
|
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_DIRSTAT;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_nodirstat:
|
|
|
|
|
fsopt->flags &= ~CEPH_MOUNT_OPT_DIRSTAT;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_rbytes:
|
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_RBYTES;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_norbytes:
|
|
|
|
|
fsopt->flags &= ~CEPH_MOUNT_OPT_RBYTES;
|
|
|
|
|
break;
|
2012-02-15 21:43:54 +08:00
|
|
|
case Opt_asyncreaddir:
|
|
|
|
|
fsopt->flags &= ~CEPH_MOUNT_OPT_NOASYNCREADDIR;
|
|
|
|
|
break;
|
2010-04-07 06:14:15 +08:00
|
|
|
case Opt_noasyncreaddir:
|
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_NOASYNCREADDIR;
|
|
|
|
|
break;
|
2012-01-11 01:12:55 +08:00
|
|
|
case Opt_dcache:
|
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_DCACHE;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_nodcache:
|
|
|
|
|
fsopt->flags &= ~CEPH_MOUNT_OPT_DCACHE;
|
|
|
|
|
break;
|
2011-01-22 08:44:03 +08:00
|
|
|
case Opt_ino32:
|
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_INO32;
|
|
|
|
|
break;
|
2012-02-15 21:43:54 +08:00
|
|
|
case Opt_noino32:
|
|
|
|
|
fsopt->flags &= ~CEPH_MOUNT_OPT_INO32;
|
|
|
|
|
break;
|
2013-08-22 05:29:54 +08:00
|
|
|
case Opt_fscache:
|
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_FSCACHE;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_nofscache:
|
|
|
|
|
fsopt->flags &= ~CEPH_MOUNT_OPT_FSCACHE;
|
|
|
|
|
break;
|
2015-04-27 15:33:28 +08:00
|
|
|
case Opt_poolperm:
|
|
|
|
|
fsopt->flags &= ~CEPH_MOUNT_OPT_NOPOOLPERM;
|
|
|
|
|
printk ("pool perm");
|
|
|
|
|
break;
|
|
|
|
|
case Opt_nopoolperm:
|
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_NOPOOLPERM;
|
|
|
|
|
break;
|
2016-11-10 16:02:06 +08:00
|
|
|
case Opt_require_active_mds:
|
|
|
|
|
fsopt->flags &= ~CEPH_MOUNT_OPT_MOUNTWAIT;
|
|
|
|
|
break;
|
|
|
|
|
case Opt_norequire_active_mds:
|
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_MOUNTWAIT;
|
|
|
|
|
break;
|
2014-02-17 02:05:29 +08:00
|
|
|
#ifdef CONFIG_CEPH_FS_POSIX_ACL
|
|
|
|
|
case Opt_acl:
|
|
|
|
|
fsopt->sb_flags |= MS_POSIXACL;
|
|
|
|
|
break;
|
|
|
|
|
#endif
|
|
|
|
|
case Opt_noacl:
|
|
|
|
|
fsopt->sb_flags &= ~MS_POSIXACL;
|
|
|
|
|
break;
|
2010-04-07 06:14:15 +08:00
|
|
|
default:
|
|
|
|
|
BUG_ON(token);
|
|
|
|
|
}
|
|
|
|
|
return 0;
|
2010-07-01 12:34:01 +08:00
|
|
|
}
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
static void destroy_mount_options(struct ceph_mount_options *args)
|
2009-10-07 02:31:07 +08:00
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
dout("destroy_mount_options %p\n", args);
|
|
|
|
|
kfree(args->snapdir_name);
|
2016-07-08 11:25:38 +08:00
|
|
|
kfree(args->mds_namespace);
|
2016-04-21 11:09:55 +08:00
|
|
|
kfree(args->server_path);
|
2017-06-27 11:57:56 +08:00
|
|
|
kfree(args->fscache_uniq);
|
2010-04-07 06:14:15 +08:00
|
|
|
kfree(args);
|
|
|
|
|
}
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
static int strcmp_null(const char *s1, const char *s2)
|
|
|
|
|
{
|
|
|
|
|
if (!s1 && !s2)
|
|
|
|
|
return 0;
|
|
|
|
|
if (s1 && !s2)
|
|
|
|
|
return -1;
|
|
|
|
|
if (!s1 && s2)
|
|
|
|
|
return 1;
|
|
|
|
|
return strcmp(s1, s2);
|
|
|
|
|
}
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
static int compare_mount_options(struct ceph_mount_options *new_fsopt,
|
|
|
|
|
struct ceph_options *new_opt,
|
|
|
|
|
struct ceph_fs_client *fsc)
|
|
|
|
|
{
|
|
|
|
|
struct ceph_mount_options *fsopt1 = new_fsopt;
|
|
|
|
|
struct ceph_mount_options *fsopt2 = fsc->mount_options;
|
|
|
|
|
int ofs = offsetof(struct ceph_mount_options, snapdir_name);
|
|
|
|
|
int ret;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
ret = memcmp(fsopt1, fsopt2, ofs);
|
|
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
|
|
ret = strcmp_null(fsopt1->snapdir_name, fsopt2->snapdir_name);
|
2016-07-08 11:25:38 +08:00
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
ret = strcmp_null(fsopt1->mds_namespace, fsopt2->mds_namespace);
|
2010-04-07 06:14:15 +08:00
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
2016-04-21 11:09:55 +08:00
|
|
|
ret = strcmp_null(fsopt1->server_path, fsopt2->server_path);
|
2017-06-27 11:57:56 +08:00
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
ret = strcmp_null(fsopt1->fscache_uniq, fsopt2->fscache_uniq);
|
2016-04-21 11:09:55 +08:00
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
return ceph_compare_options(new_opt, fsc->client);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int parse_mount_options(struct ceph_mount_options **pfsopt,
|
|
|
|
|
struct ceph_options **popt,
|
|
|
|
|
int flags, char *options,
|
2016-04-21 11:09:55 +08:00
|
|
|
const char *dev_name)
|
2010-04-07 06:14:15 +08:00
|
|
|
{
|
|
|
|
|
struct ceph_mount_options *fsopt;
|
|
|
|
|
const char *dev_name_end;
|
ceph: let path portion of mount "device" be optional
A recent change to /sbin/mountall causes any trailing '/' character
in the "device" (or fs_spec) field in /etc/fstab to be stripped. As
a result, an entry for a ceph mount that intends to mount the root
of the name space ends up with now path portion, and the ceph mount
option processing code rejects this.
That is, an entry in /etc/fstab like:
cephserver:port:/ /mnt ceph defaults 0 0
provides to the ceph code just "cephserver:port:" as the "device,"
and that gets rejected.
Although this is a bug in /sbin/mountall, we can have the ceph mount
code support an empty/nonexistent path, interpreting it to mean the
root of the name space.
RFC 5952 offers recommendations for how to express IPv6 addresses,
and recommends the usage found in RFC 3986 (which specifies the
format for URI's) for representing both IPv4 and IPv6 addresses that
include port numbers. (See in particular the definition of
"authority" found in the Appendix of RFC 3986.)
According to those standards, no host specification will ever
contain a '/' character. As a result, it is sufficient to scan a
provided "device" from an /etc/fstab entry for the first '/'
character, and if it's found, treat that as the beginning of the
path. If no '/' character is present, we can treat the entire
string as the monitor host specification(s), and assume the path
to be the root of the name space. We'll still require a ':' to
separate the host portion from the (possibly empty) path portion.
This means that we can more formally define how ceph will interpret
the "device" it's provided when processing a mount request:
"device" will look like:
<server_spec>[,<server_spec>...]:[<path>]
where
<server_spec> is <ip>[:<port>]
<path> is optional, but if present must begin with '/'
This addresses http://tracker.newdream.net/issues/2919
Signed-off-by: Alex Elder <elder@inktank.com>
Reviewed-by: Dan Mick <dan.mick@inktank.com>
2012-08-10 01:33:26 +08:00
|
|
|
int err;
|
|
|
|
|
|
|
|
|
|
if (!dev_name || !*dev_name)
|
|
|
|
|
return -EINVAL;
|
2010-04-07 06:14:15 +08:00
|
|
|
|
|
|
|
|
fsopt = kzalloc(sizeof(*fsopt), GFP_KERNEL);
|
|
|
|
|
if (!fsopt)
|
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
|
|
dout("parse_mount_options %p, dev_name '%s'\n", fsopt, dev_name);
|
|
|
|
|
|
2011-08-23 03:49:23 +08:00
|
|
|
fsopt->sb_flags = flags;
|
|
|
|
|
fsopt->flags = CEPH_MOUNT_OPT_DEFAULT;
|
2010-04-07 06:14:15 +08:00
|
|
|
|
2017-07-11 17:34:46 +08:00
|
|
|
fsopt->wsize = CEPH_MAX_WRITE_SIZE;
|
2017-07-11 15:56:09 +08:00
|
|
|
fsopt->rsize = CEPH_MAX_READ_SIZE;
|
2011-08-23 03:49:23 +08:00
|
|
|
fsopt->rasize = CEPH_RASIZE_DEFAULT;
|
|
|
|
|
fsopt->snapdir_name = kstrdup(CEPH_SNAPDIRNAME_DEFAULT, GFP_KERNEL);
|
2015-03-22 00:54:58 +08:00
|
|
|
if (!fsopt->snapdir_name) {
|
|
|
|
|
err = -ENOMEM;
|
|
|
|
|
goto out;
|
|
|
|
|
}
|
|
|
|
|
|
2011-01-18 23:59:40 +08:00
|
|
|
fsopt->caps_wanted_delay_min = CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT;
|
|
|
|
|
fsopt->caps_wanted_delay_max = CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT;
|
2011-08-23 03:49:23 +08:00
|
|
|
fsopt->max_readdir = CEPH_MAX_READDIR_DEFAULT;
|
|
|
|
|
fsopt->max_readdir_bytes = CEPH_MAX_READDIR_BYTES_DEFAULT;
|
|
|
|
|
fsopt->congestion_kb = default_congestion_kb();
|
|
|
|
|
|
ceph: let path portion of mount "device" be optional
A recent change to /sbin/mountall causes any trailing '/' character
in the "device" (or fs_spec) field in /etc/fstab to be stripped. As
a result, an entry for a ceph mount that intends to mount the root
of the name space ends up with now path portion, and the ceph mount
option processing code rejects this.
That is, an entry in /etc/fstab like:
cephserver:port:/ /mnt ceph defaults 0 0
provides to the ceph code just "cephserver:port:" as the "device,"
and that gets rejected.
Although this is a bug in /sbin/mountall, we can have the ceph mount
code support an empty/nonexistent path, interpreting it to mean the
root of the name space.
RFC 5952 offers recommendations for how to express IPv6 addresses,
and recommends the usage found in RFC 3986 (which specifies the
format for URI's) for representing both IPv4 and IPv6 addresses that
include port numbers. (See in particular the definition of
"authority" found in the Appendix of RFC 3986.)
According to those standards, no host specification will ever
contain a '/' character. As a result, it is sufficient to scan a
provided "device" from an /etc/fstab entry for the first '/'
character, and if it's found, treat that as the beginning of the
path. If no '/' character is present, we can treat the entire
string as the monitor host specification(s), and assume the path
to be the root of the name space. We'll still require a ':' to
separate the host portion from the (possibly empty) path portion.
This means that we can more formally define how ceph will interpret
the "device" it's provided when processing a mount request:
"device" will look like:
<server_spec>[,<server_spec>...]:[<path>]
where
<server_spec> is <ip>[:<port>]
<path> is optional, but if present must begin with '/'
This addresses http://tracker.newdream.net/issues/2919
Signed-off-by: Alex Elder <elder@inktank.com>
Reviewed-by: Dan Mick <dan.mick@inktank.com>
2012-08-10 01:33:26 +08:00
|
|
|
/*
|
|
|
|
|
* Distinguish the server list from the path in "dev_name".
|
|
|
|
|
* Internally we do not include the leading '/' in the path.
|
|
|
|
|
*
|
|
|
|
|
* "dev_name" will look like:
|
|
|
|
|
* <server_spec>[,<server_spec>...]:[<path>]
|
|
|
|
|
* where
|
|
|
|
|
* <server_spec> is <ip>[:<port>]
|
|
|
|
|
* <path> is optional, but if present must begin with '/'
|
|
|
|
|
*/
|
|
|
|
|
dev_name_end = strchr(dev_name, '/');
|
|
|
|
|
if (dev_name_end) {
|
2016-09-14 14:53:05 +08:00
|
|
|
if (strlen(dev_name_end) > 1) {
|
|
|
|
|
fsopt->server_path = kstrdup(dev_name_end, GFP_KERNEL);
|
|
|
|
|
if (!fsopt->server_path) {
|
|
|
|
|
err = -ENOMEM;
|
|
|
|
|
goto out;
|
|
|
|
|
}
|
2016-04-21 11:09:55 +08:00
|
|
|
}
|
ceph: let path portion of mount "device" be optional
A recent change to /sbin/mountall causes any trailing '/' character
in the "device" (or fs_spec) field in /etc/fstab to be stripped. As
a result, an entry for a ceph mount that intends to mount the root
of the name space ends up with now path portion, and the ceph mount
option processing code rejects this.
That is, an entry in /etc/fstab like:
cephserver:port:/ /mnt ceph defaults 0 0
provides to the ceph code just "cephserver:port:" as the "device,"
and that gets rejected.
Although this is a bug in /sbin/mountall, we can have the ceph mount
code support an empty/nonexistent path, interpreting it to mean the
root of the name space.
RFC 5952 offers recommendations for how to express IPv6 addresses,
and recommends the usage found in RFC 3986 (which specifies the
format for URI's) for representing both IPv4 and IPv6 addresses that
include port numbers. (See in particular the definition of
"authority" found in the Appendix of RFC 3986.)
According to those standards, no host specification will ever
contain a '/' character. As a result, it is sufficient to scan a
provided "device" from an /etc/fstab entry for the first '/'
character, and if it's found, treat that as the beginning of the
path. If no '/' character is present, we can treat the entire
string as the monitor host specification(s), and assume the path
to be the root of the name space. We'll still require a ':' to
separate the host portion from the (possibly empty) path portion.
This means that we can more formally define how ceph will interpret
the "device" it's provided when processing a mount request:
"device" will look like:
<server_spec>[,<server_spec>...]:[<path>]
where
<server_spec> is <ip>[:<port>]
<path> is optional, but if present must begin with '/'
This addresses http://tracker.newdream.net/issues/2919
Signed-off-by: Alex Elder <elder@inktank.com>
Reviewed-by: Dan Mick <dan.mick@inktank.com>
2012-08-10 01:33:26 +08:00
|
|
|
} else {
|
|
|
|
|
dev_name_end = dev_name + strlen(dev_name);
|
|
|
|
|
}
|
2011-08-23 03:49:23 +08:00
|
|
|
err = -EINVAL;
|
ceph: let path portion of mount "device" be optional
A recent change to /sbin/mountall causes any trailing '/' character
in the "device" (or fs_spec) field in /etc/fstab to be stripped. As
a result, an entry for a ceph mount that intends to mount the root
of the name space ends up with now path portion, and the ceph mount
option processing code rejects this.
That is, an entry in /etc/fstab like:
cephserver:port:/ /mnt ceph defaults 0 0
provides to the ceph code just "cephserver:port:" as the "device,"
and that gets rejected.
Although this is a bug in /sbin/mountall, we can have the ceph mount
code support an empty/nonexistent path, interpreting it to mean the
root of the name space.
RFC 5952 offers recommendations for how to express IPv6 addresses,
and recommends the usage found in RFC 3986 (which specifies the
format for URI's) for representing both IPv4 and IPv6 addresses that
include port numbers. (See in particular the definition of
"authority" found in the Appendix of RFC 3986.)
According to those standards, no host specification will ever
contain a '/' character. As a result, it is sufficient to scan a
provided "device" from an /etc/fstab entry for the first '/'
character, and if it's found, treat that as the beginning of the
path. If no '/' character is present, we can treat the entire
string as the monitor host specification(s), and assume the path
to be the root of the name space. We'll still require a ':' to
separate the host portion from the (possibly empty) path portion.
This means that we can more formally define how ceph will interpret
the "device" it's provided when processing a mount request:
"device" will look like:
<server_spec>[,<server_spec>...]:[<path>]
where
<server_spec> is <ip>[:<port>]
<path> is optional, but if present must begin with '/'
This addresses http://tracker.newdream.net/issues/2919
Signed-off-by: Alex Elder <elder@inktank.com>
Reviewed-by: Dan Mick <dan.mick@inktank.com>
2012-08-10 01:33:26 +08:00
|
|
|
dev_name_end--; /* back up to ':' separator */
|
2013-07-02 06:33:39 +08:00
|
|
|
if (dev_name_end < dev_name || *dev_name_end != ':') {
|
ceph: let path portion of mount "device" be optional
A recent change to /sbin/mountall causes any trailing '/' character
in the "device" (or fs_spec) field in /etc/fstab to be stripped. As
a result, an entry for a ceph mount that intends to mount the root
of the name space ends up with now path portion, and the ceph mount
option processing code rejects this.
That is, an entry in /etc/fstab like:
cephserver:port:/ /mnt ceph defaults 0 0
provides to the ceph code just "cephserver:port:" as the "device,"
and that gets rejected.
Although this is a bug in /sbin/mountall, we can have the ceph mount
code support an empty/nonexistent path, interpreting it to mean the
root of the name space.
RFC 5952 offers recommendations for how to express IPv6 addresses,
and recommends the usage found in RFC 3986 (which specifies the
format for URI's) for representing both IPv4 and IPv6 addresses that
include port numbers. (See in particular the definition of
"authority" found in the Appendix of RFC 3986.)
According to those standards, no host specification will ever
contain a '/' character. As a result, it is sufficient to scan a
provided "device" from an /etc/fstab entry for the first '/'
character, and if it's found, treat that as the beginning of the
path. If no '/' character is present, we can treat the entire
string as the monitor host specification(s), and assume the path
to be the root of the name space. We'll still require a ':' to
separate the host portion from the (possibly empty) path portion.
This means that we can more formally define how ceph will interpret
the "device" it's provided when processing a mount request:
"device" will look like:
<server_spec>[,<server_spec>...]:[<path>]
where
<server_spec> is <ip>[:<port>]
<path> is optional, but if present must begin with '/'
This addresses http://tracker.newdream.net/issues/2919
Signed-off-by: Alex Elder <elder@inktank.com>
Reviewed-by: Dan Mick <dan.mick@inktank.com>
2012-08-10 01:33:26 +08:00
|
|
|
pr_err("device name is missing path (no : separator in %s)\n",
|
2011-08-23 03:49:23 +08:00
|
|
|
dev_name);
|
|
|
|
|
goto out;
|
|
|
|
|
}
|
2010-04-07 06:14:15 +08:00
|
|
|
dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name);
|
2016-04-21 11:09:55 +08:00
|
|
|
if (fsopt->server_path)
|
|
|
|
|
dout("server path '%s'\n", fsopt->server_path);
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2012-01-25 00:08:36 +08:00
|
|
|
*popt = ceph_parse_options(options, dev_name, dev_name_end,
|
2010-04-07 06:14:15 +08:00
|
|
|
parse_fsopt_token, (void *)fsopt);
|
2012-01-25 00:08:36 +08:00
|
|
|
if (IS_ERR(*popt)) {
|
|
|
|
|
err = PTR_ERR(*popt);
|
2010-04-07 06:14:15 +08:00
|
|
|
goto out;
|
2012-01-25 00:08:36 +08:00
|
|
|
}
|
2010-04-07 06:14:15 +08:00
|
|
|
|
|
|
|
|
/* success */
|
|
|
|
|
*pfsopt = fsopt;
|
|
|
|
|
return 0;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2009-10-27 13:07:53 +08:00
|
|
|
out:
|
2010-04-07 06:14:15 +08:00
|
|
|
destroy_mount_options(fsopt);
|
|
|
|
|
return err;
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
/**
|
|
|
|
|
* ceph_show_options - Show mount options in /proc/mounts
|
|
|
|
|
* @m: seq_file to write to
|
2011-12-09 10:32:45 +08:00
|
|
|
* @root: root of that (sub)tree
|
2010-04-07 06:14:15 +08:00
|
|
|
*/
|
2011-12-09 10:32:45 +08:00
|
|
|
static int ceph_show_options(struct seq_file *m, struct dentry *root)
|
2009-10-07 02:31:07 +08:00
|
|
|
{
|
2011-12-09 10:32:45 +08:00
|
|
|
struct ceph_fs_client *fsc = ceph_sb_to_client(root->d_sb);
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_mount_options *fsopt = fsc->mount_options;
|
2015-03-26 02:02:16 +08:00
|
|
|
size_t pos;
|
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
|
|
/* a comma between MNT/MS and client options */
|
|
|
|
|
seq_putc(m, ',');
|
|
|
|
|
pos = m->count;
|
|
|
|
|
|
|
|
|
|
ret = ceph_print_client_options(m, fsc->client);
|
|
|
|
|
if (ret)
|
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
|
|
/* retract our comma if no client options */
|
|
|
|
|
if (m->count == pos)
|
|
|
|
|
m->count--;
|
2010-04-07 06:14:15 +08:00
|
|
|
|
|
|
|
|
if (fsopt->flags & CEPH_MOUNT_OPT_DIRSTAT)
|
|
|
|
|
seq_puts(m, ",dirstat");
|
2016-01-25 10:44:33 +08:00
|
|
|
if ((fsopt->flags & CEPH_MOUNT_OPT_RBYTES))
|
|
|
|
|
seq_puts(m, ",rbytes");
|
2010-04-07 06:14:15 +08:00
|
|
|
if (fsopt->flags & CEPH_MOUNT_OPT_NOASYNCREADDIR)
|
|
|
|
|
seq_puts(m, ",noasyncreaddir");
|
2015-03-26 02:10:09 +08:00
|
|
|
if ((fsopt->flags & CEPH_MOUNT_OPT_DCACHE) == 0)
|
2012-01-11 01:12:55 +08:00
|
|
|
seq_puts(m, ",nodcache");
|
2017-06-27 11:57:56 +08:00
|
|
|
if (fsopt->flags & CEPH_MOUNT_OPT_FSCACHE) {
|
|
|
|
|
if (fsopt->fscache_uniq)
|
|
|
|
|
seq_printf(m, ",fsc=%s", fsopt->fscache_uniq);
|
|
|
|
|
else
|
|
|
|
|
seq_puts(m, ",fsc");
|
|
|
|
|
}
|
2015-04-27 15:33:28 +08:00
|
|
|
if (fsopt->flags & CEPH_MOUNT_OPT_NOPOOLPERM)
|
|
|
|
|
seq_puts(m, ",nopoolperm");
|
2010-04-07 06:14:15 +08:00
|
|
|
|
2014-02-17 02:05:29 +08:00
|
|
|
#ifdef CONFIG_CEPH_FS_POSIX_ACL
|
|
|
|
|
if (fsopt->sb_flags & MS_POSIXACL)
|
|
|
|
|
seq_puts(m, ",acl");
|
|
|
|
|
else
|
|
|
|
|
seq_puts(m, ",noacl");
|
|
|
|
|
#endif
|
|
|
|
|
|
2016-07-08 11:25:38 +08:00
|
|
|
if (fsopt->mds_namespace)
|
|
|
|
|
seq_printf(m, ",mds_namespace=%s", fsopt->mds_namespace);
|
2010-04-07 06:14:15 +08:00
|
|
|
if (fsopt->wsize)
|
|
|
|
|
seq_printf(m, ",wsize=%d", fsopt->wsize);
|
2017-07-11 15:56:09 +08:00
|
|
|
if (fsopt->rsize != CEPH_MAX_READ_SIZE)
|
2010-04-07 06:14:15 +08:00
|
|
|
seq_printf(m, ",rsize=%d", fsopt->rsize);
|
2011-08-04 23:03:44 +08:00
|
|
|
if (fsopt->rasize != CEPH_RASIZE_DEFAULT)
|
2011-12-02 00:06:52 +08:00
|
|
|
seq_printf(m, ",rasize=%d", fsopt->rasize);
|
2010-04-07 06:14:15 +08:00
|
|
|
if (fsopt->congestion_kb != default_congestion_kb())
|
|
|
|
|
seq_printf(m, ",write_congestion_kb=%d", fsopt->congestion_kb);
|
|
|
|
|
if (fsopt->caps_wanted_delay_min != CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT)
|
|
|
|
|
seq_printf(m, ",caps_wanted_delay_min=%d",
|
|
|
|
|
fsopt->caps_wanted_delay_min);
|
|
|
|
|
if (fsopt->caps_wanted_delay_max != CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT)
|
|
|
|
|
seq_printf(m, ",caps_wanted_delay_max=%d",
|
|
|
|
|
fsopt->caps_wanted_delay_max);
|
|
|
|
|
if (fsopt->max_readdir != CEPH_MAX_READDIR_DEFAULT)
|
|
|
|
|
seq_printf(m, ",readdir_max_entries=%d", fsopt->max_readdir);
|
|
|
|
|
if (fsopt->max_readdir_bytes != CEPH_MAX_READDIR_BYTES_DEFAULT)
|
|
|
|
|
seq_printf(m, ",readdir_max_bytes=%d", fsopt->max_readdir_bytes);
|
|
|
|
|
if (strcmp(fsopt->snapdir_name, CEPH_SNAPDIRNAME_DEFAULT))
|
fs: create and use seq_show_option for escaping
Many file systems that implement the show_options hook fail to correctly
escape their output which could lead to unescaped characters (e.g. new
lines) leaking into /proc/mounts and /proc/[pid]/mountinfo files. This
could lead to confusion, spoofed entries (resulting in things like
systemd issuing false d-bus "mount" notifications), and who knows what
else. This looks like it would only be the root user stepping on
themselves, but it's possible weird things could happen in containers or
in other situations with delegated mount privileges.
Here's an example using overlay with setuid fusermount trusting the
contents of /proc/mounts (via the /etc/mtab symlink). Imagine the use
of "sudo" is something more sneaky:
$ BASE="ovl"
$ MNT="$BASE/mnt"
$ LOW="$BASE/lower"
$ UP="$BASE/upper"
$ WORK="$BASE/work/ 0 0
none /proc fuse.pwn user_id=1000"
$ mkdir -p "$LOW" "$UP" "$WORK"
$ sudo mount -t overlay -o "lowerdir=$LOW,upperdir=$UP,workdir=$WORK" none /mnt
$ cat /proc/mounts
none /root/ovl/mnt overlay rw,relatime,lowerdir=ovl/lower,upperdir=ovl/upper,workdir=ovl/work/ 0 0
none /proc fuse.pwn user_id=1000 0 0
$ fusermount -u /proc
$ cat /proc/mounts
cat: /proc/mounts: No such file or directory
This fixes the problem by adding new seq_show_option and
seq_show_option_n helpers, and updating the vulnerable show_option
handlers to use them as needed. Some, like SELinux, need to be open
coded due to unusual existing escape mechanisms.
[akpm@linux-foundation.org: add lost chunk, per Kees]
[keescook@chromium.org: seq_show_option should be using const parameters]
Signed-off-by: Kees Cook <keescook@chromium.org>
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Acked-by: Jan Kara <jack@suse.com>
Acked-by: Paul Moore <paul@paul-moore.com>
Cc: J. R. Okajima <hooanon05g@gmail.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2015-09-05 06:44:57 +08:00
|
|
|
seq_show_option(m, "snapdirname", fsopt->snapdir_name);
|
2015-03-26 02:02:16 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
return 0;
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
2010-04-07 06:14:15 +08:00
|
|
|
* handle any mon messages the standard library doesn't understand.
|
|
|
|
|
* return error if we don't either.
|
2009-10-07 02:31:07 +08:00
|
|
|
*/
|
2010-04-07 06:14:15 +08:00
|
|
|
static int extra_mon_dispatch(struct ceph_client *client, struct ceph_msg *msg)
|
2009-10-07 02:31:07 +08:00
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_fs_client *fsc = client->private;
|
|
|
|
|
int type = le16_to_cpu(msg->hdr.type);
|
|
|
|
|
|
|
|
|
|
switch (type) {
|
|
|
|
|
case CEPH_MSG_MDS_MAP:
|
2016-07-08 11:25:38 +08:00
|
|
|
ceph_mdsc_handle_mdsmap(fsc->mdsc, msg);
|
|
|
|
|
return 0;
|
|
|
|
|
case CEPH_MSG_FS_MAP_USER:
|
|
|
|
|
ceph_mdsc_handle_fsmap(fsc->mdsc, msg);
|
2010-04-07 06:14:15 +08:00
|
|
|
return 0;
|
|
|
|
|
default:
|
|
|
|
|
return -1;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* create a new fs client
|
|
|
|
|
*/
|
2011-09-24 02:53:30 +08:00
|
|
|
static struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt,
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_options *opt)
|
|
|
|
|
{
|
|
|
|
|
struct ceph_fs_client *fsc;
|
2013-04-01 23:48:40 +08:00
|
|
|
int page_count;
|
|
|
|
|
size_t size;
|
2009-10-07 02:31:07 +08:00
|
|
|
int err = -ENOMEM;
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
fsc = kzalloc(sizeof(*fsc), GFP_KERNEL);
|
|
|
|
|
if (!fsc)
|
2009-10-07 02:31:07 +08:00
|
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
|
|
2017-03-04 01:16:07 +08:00
|
|
|
fsc->client = ceph_create_client(opt, fsc);
|
2010-04-07 06:14:15 +08:00
|
|
|
if (IS_ERR(fsc->client)) {
|
|
|
|
|
err = PTR_ERR(fsc->client);
|
|
|
|
|
goto fail;
|
|
|
|
|
}
|
|
|
|
|
fsc->client->extra_mon_dispatch = extra_mon_dispatch;
|
2016-07-08 11:25:38 +08:00
|
|
|
|
2017-08-21 02:22:02 +08:00
|
|
|
if (!fsopt->mds_namespace) {
|
2016-07-08 11:25:38 +08:00
|
|
|
ceph_monc_want_map(&fsc->client->monc, CEPH_SUB_MDSMAP,
|
|
|
|
|
0, true);
|
|
|
|
|
} else {
|
|
|
|
|
ceph_monc_want_map(&fsc->client->monc, CEPH_SUB_FSMAP,
|
|
|
|
|
0, false);
|
|
|
|
|
}
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
fsc->mount_options = fsopt;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
fsc->sb = NULL;
|
|
|
|
|
fsc->mount_state = CEPH_MOUNT_MOUNTING;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
atomic_long_set(&fsc->writeback_count, 0);
|
2009-10-07 02:31:07 +08:00
|
|
|
|
|
|
|
|
err = -ENOMEM;
|
2011-01-03 21:49:45 +08:00
|
|
|
/*
|
|
|
|
|
* The number of concurrent works can be high but they don't need
|
|
|
|
|
* to be processed in parallel, limit concurrency.
|
|
|
|
|
*/
|
|
|
|
|
fsc->wb_wq = alloc_workqueue("ceph-writeback", 0, 1);
|
2017-08-21 02:22:02 +08:00
|
|
|
if (!fsc->wb_wq)
|
2017-04-12 18:24:33 +08:00
|
|
|
goto fail_client;
|
2011-01-03 21:49:45 +08:00
|
|
|
fsc->pg_inv_wq = alloc_workqueue("ceph-pg-invalid", 0, 1);
|
2017-08-21 02:22:02 +08:00
|
|
|
if (!fsc->pg_inv_wq)
|
2009-10-07 02:31:07 +08:00
|
|
|
goto fail_wb_wq;
|
2011-01-03 21:49:45 +08:00
|
|
|
fsc->trunc_wq = alloc_workqueue("ceph-trunc", 0, 1);
|
2017-08-21 02:22:02 +08:00
|
|
|
if (!fsc->trunc_wq)
|
2009-10-07 02:31:07 +08:00
|
|
|
goto fail_pg_inv_wq;
|
|
|
|
|
|
2009-11-19 07:08:44 +08:00
|
|
|
/* set up mempools */
|
|
|
|
|
err = -ENOMEM;
|
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros
PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time
ago with promise that one day it will be possible to implement page
cache with bigger chunks than PAGE_SIZE.
This promise never materialized. And unlikely will.
We have many places where PAGE_CACHE_SIZE assumed to be equal to
PAGE_SIZE. And it's constant source of confusion on whether
PAGE_CACHE_* or PAGE_* constant should be used in a particular case,
especially on the border between fs and mm.
Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much
breakage to be doable.
Let's stop pretending that pages in page cache are special. They are
not.
The changes are pretty straight-forward:
- <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN};
- page_cache_get() -> get_page();
- page_cache_release() -> put_page();
This patch contains automated changes generated with coccinelle using
script below. For some reason, coccinelle doesn't patch header files.
I've called spatch for them manually.
The only adjustment after coccinelle is revert of changes to
PAGE_CAHCE_ALIGN definition: we are going to drop it later.
There are few places in the code where coccinelle didn't reach. I'll
fix them manually in a separate patch. Comments and documentation also
will be addressed with the separate patch.
virtual patch
@@
expression E;
@@
- E << (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
expression E;
@@
- E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
@@
- PAGE_CACHE_SHIFT
+ PAGE_SHIFT
@@
@@
- PAGE_CACHE_SIZE
+ PAGE_SIZE
@@
@@
- PAGE_CACHE_MASK
+ PAGE_MASK
@@
expression E;
@@
- PAGE_CACHE_ALIGN(E)
+ PAGE_ALIGN(E)
@@
expression E;
@@
- page_cache_get(E)
+ get_page(E)
@@
expression E;
@@
- page_cache_release(E)
+ put_page(E)
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 20:29:47 +08:00
|
|
|
page_count = fsc->mount_options->wsize >> PAGE_SHIFT;
|
2013-04-01 23:48:40 +08:00
|
|
|
size = sizeof (struct page *) * (page_count ? page_count : 1);
|
|
|
|
|
fsc->wb_pagevec_pool = mempool_create_kmalloc_pool(10, size);
|
2010-04-07 06:14:15 +08:00
|
|
|
if (!fsc->wb_pagevec_pool)
|
2009-11-19 07:08:44 +08:00
|
|
|
goto fail_trunc_wq;
|
|
|
|
|
|
2010-02-18 02:02:43 +08:00
|
|
|
/* caps */
|
2010-04-07 06:14:15 +08:00
|
|
|
fsc->min_caps = fsopt->max_readdir;
|
|
|
|
|
|
|
|
|
|
return fsc;
|
2009-11-19 07:08:44 +08:00
|
|
|
|
2009-10-07 02:31:07 +08:00
|
|
|
fail_trunc_wq:
|
2010-04-07 06:14:15 +08:00
|
|
|
destroy_workqueue(fsc->trunc_wq);
|
2009-10-07 02:31:07 +08:00
|
|
|
fail_pg_inv_wq:
|
2010-04-07 06:14:15 +08:00
|
|
|
destroy_workqueue(fsc->pg_inv_wq);
|
2009-10-07 02:31:07 +08:00
|
|
|
fail_wb_wq:
|
2010-04-07 06:14:15 +08:00
|
|
|
destroy_workqueue(fsc->wb_wq);
|
|
|
|
|
fail_client:
|
|
|
|
|
ceph_destroy_client(fsc->client);
|
2009-10-07 02:31:07 +08:00
|
|
|
fail:
|
2010-04-07 06:14:15 +08:00
|
|
|
kfree(fsc);
|
2009-10-07 02:31:07 +08:00
|
|
|
return ERR_PTR(err);
|
|
|
|
|
}
|
|
|
|
|
|
2011-09-24 02:53:30 +08:00
|
|
|
static void destroy_fs_client(struct ceph_fs_client *fsc)
|
2009-10-07 02:31:07 +08:00
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
dout("destroy_fs_client %p\n", fsc);
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
destroy_workqueue(fsc->wb_wq);
|
|
|
|
|
destroy_workqueue(fsc->pg_inv_wq);
|
|
|
|
|
destroy_workqueue(fsc->trunc_wq);
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
mempool_destroy(fsc->wb_pagevec_pool);
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
destroy_mount_options(fsc->mount_options);
|
2010-05-05 07:14:46 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
ceph_destroy_client(fsc->client);
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
kfree(fsc);
|
|
|
|
|
dout("destroy_fs_client %p done\n", fsc);
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
|
|
|
|
|
2009-11-19 08:50:41 +08:00
|
|
|
/*
|
2010-04-07 06:14:15 +08:00
|
|
|
* caches
|
2009-11-19 08:50:41 +08:00
|
|
|
*/
|
2010-04-07 06:14:15 +08:00
|
|
|
struct kmem_cache *ceph_inode_cachep;
|
|
|
|
|
struct kmem_cache *ceph_cap_cachep;
|
2015-06-10 17:26:13 +08:00
|
|
|
struct kmem_cache *ceph_cap_flush_cachep;
|
2010-04-07 06:14:15 +08:00
|
|
|
struct kmem_cache *ceph_dentry_cachep;
|
|
|
|
|
struct kmem_cache *ceph_file_cachep;
|
|
|
|
|
|
|
|
|
|
static void ceph_inode_init_once(void *foo)
|
2009-11-19 08:50:41 +08:00
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_inode_info *ci = foo;
|
|
|
|
|
inode_init_once(&ci->vfs_inode);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int __init init_caches(void)
|
|
|
|
|
{
|
2013-08-22 05:29:54 +08:00
|
|
|
int error = -ENOMEM;
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
ceph_inode_cachep = kmem_cache_create("ceph_inode_info",
|
|
|
|
|
sizeof(struct ceph_inode_info),
|
|
|
|
|
__alignof__(struct ceph_inode_info),
|
2016-01-15 07:18:21 +08:00
|
|
|
SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD|
|
|
|
|
|
SLAB_ACCOUNT, ceph_inode_init_once);
|
2017-08-21 02:22:02 +08:00
|
|
|
if (!ceph_inode_cachep)
|
2010-04-07 06:14:15 +08:00
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
|
|
ceph_cap_cachep = KMEM_CACHE(ceph_cap,
|
|
|
|
|
SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
|
2017-08-21 02:22:02 +08:00
|
|
|
if (!ceph_cap_cachep)
|
2010-04-07 06:14:15 +08:00
|
|
|
goto bad_cap;
|
2015-06-10 17:26:13 +08:00
|
|
|
ceph_cap_flush_cachep = KMEM_CACHE(ceph_cap_flush,
|
|
|
|
|
SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
|
2017-08-21 02:22:02 +08:00
|
|
|
if (!ceph_cap_flush_cachep)
|
2015-06-10 17:26:13 +08:00
|
|
|
goto bad_cap_flush;
|
2010-04-07 06:14:15 +08:00
|
|
|
|
|
|
|
|
ceph_dentry_cachep = KMEM_CACHE(ceph_dentry_info,
|
|
|
|
|
SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
|
2017-08-21 02:22:02 +08:00
|
|
|
if (!ceph_dentry_cachep)
|
2010-04-07 06:14:15 +08:00
|
|
|
goto bad_dentry;
|
|
|
|
|
|
2016-07-26 01:12:13 +08:00
|
|
|
ceph_file_cachep = KMEM_CACHE(ceph_file_info, SLAB_MEM_SPREAD);
|
|
|
|
|
|
2017-08-21 02:22:02 +08:00
|
|
|
if (!ceph_file_cachep)
|
2010-04-07 06:14:15 +08:00
|
|
|
goto bad_file;
|
|
|
|
|
|
2013-08-22 05:29:54 +08:00
|
|
|
if ((error = ceph_fscache_register()))
|
|
|
|
|
goto bad_file;
|
2010-04-07 06:14:15 +08:00
|
|
|
|
2013-08-22 05:29:54 +08:00
|
|
|
return 0;
|
2010-04-07 06:14:15 +08:00
|
|
|
bad_file:
|
|
|
|
|
kmem_cache_destroy(ceph_dentry_cachep);
|
|
|
|
|
bad_dentry:
|
2015-06-10 17:26:13 +08:00
|
|
|
kmem_cache_destroy(ceph_cap_flush_cachep);
|
|
|
|
|
bad_cap_flush:
|
2010-04-07 06:14:15 +08:00
|
|
|
kmem_cache_destroy(ceph_cap_cachep);
|
|
|
|
|
bad_cap:
|
|
|
|
|
kmem_cache_destroy(ceph_inode_cachep);
|
2013-08-22 05:29:54 +08:00
|
|
|
return error;
|
2009-11-19 08:50:41 +08:00
|
|
|
}
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
static void destroy_caches(void)
|
|
|
|
|
{
|
2012-09-26 09:33:07 +08:00
|
|
|
/*
|
|
|
|
|
* Make sure all delayed rcu free inodes are flushed before we
|
|
|
|
|
* destroy cache.
|
|
|
|
|
*/
|
|
|
|
|
rcu_barrier();
|
2013-08-22 05:29:54 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
kmem_cache_destroy(ceph_inode_cachep);
|
|
|
|
|
kmem_cache_destroy(ceph_cap_cachep);
|
2015-06-10 17:26:13 +08:00
|
|
|
kmem_cache_destroy(ceph_cap_flush_cachep);
|
2010-04-07 06:14:15 +08:00
|
|
|
kmem_cache_destroy(ceph_dentry_cachep);
|
|
|
|
|
kmem_cache_destroy(ceph_file_cachep);
|
2013-08-22 05:29:54 +08:00
|
|
|
|
|
|
|
|
ceph_fscache_unregister();
|
2010-04-07 06:14:15 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
2009-10-07 02:31:07 +08:00
|
|
|
/*
|
2010-04-07 06:14:15 +08:00
|
|
|
* ceph_umount_begin - initiate forced umount. Tear down down the
|
|
|
|
|
* mount, skipping steps that may hang while waiting for server(s).
|
2009-10-07 02:31:07 +08:00
|
|
|
*/
|
2010-04-07 06:14:15 +08:00
|
|
|
static void ceph_umount_begin(struct super_block *sb)
|
2009-10-07 02:31:07 +08:00
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
|
|
|
|
|
|
|
|
|
|
dout("ceph_umount_begin - starting forced umount\n");
|
|
|
|
|
if (!fsc)
|
|
|
|
|
return;
|
|
|
|
|
fsc->mount_state = CEPH_MOUNT_SHUTDOWN;
|
2015-07-01 16:27:46 +08:00
|
|
|
ceph_mdsc_force_umount(fsc->mdsc);
|
2010-04-07 06:14:15 +08:00
|
|
|
return;
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
static const struct super_operations ceph_super_ops = {
|
|
|
|
|
.alloc_inode = ceph_alloc_inode,
|
|
|
|
|
.destroy_inode = ceph_destroy_inode,
|
|
|
|
|
.write_inode = ceph_write_inode,
|
2013-09-20 19:55:31 +08:00
|
|
|
.drop_inode = ceph_drop_inode,
|
2010-04-07 06:14:15 +08:00
|
|
|
.sync_fs = ceph_sync_fs,
|
|
|
|
|
.put_super = ceph_put_super,
|
|
|
|
|
.show_options = ceph_show_options,
|
|
|
|
|
.statfs = ceph_statfs,
|
|
|
|
|
.umount_begin = ceph_umount_begin,
|
|
|
|
|
};
|
|
|
|
|
|
2009-10-07 02:31:07 +08:00
|
|
|
/*
|
|
|
|
|
* Bootstrap mount by opening the root directory. Note the mount
|
|
|
|
|
* @started time from caller, and time out if this takes too long.
|
|
|
|
|
*/
|
2010-04-07 06:14:15 +08:00
|
|
|
static struct dentry *open_root_dentry(struct ceph_fs_client *fsc,
|
2009-10-07 02:31:07 +08:00
|
|
|
const char *path,
|
|
|
|
|
unsigned long started)
|
|
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_mds_client *mdsc = fsc->mdsc;
|
2009-10-07 02:31:07 +08:00
|
|
|
struct ceph_mds_request *req = NULL;
|
|
|
|
|
int err;
|
|
|
|
|
struct dentry *root;
|
|
|
|
|
|
|
|
|
|
/* open dir */
|
|
|
|
|
dout("open_root_inode opening '%s'\n", path);
|
|
|
|
|
req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_GETATTR, USE_ANY_MDS);
|
|
|
|
|
if (IS_ERR(req))
|
2010-05-22 18:01:14 +08:00
|
|
|
return ERR_CAST(req);
|
2009-10-07 02:31:07 +08:00
|
|
|
req->r_path1 = kstrdup(path, GFP_NOFS);
|
2015-03-22 00:54:58 +08:00
|
|
|
if (!req->r_path1) {
|
|
|
|
|
root = ERR_PTR(-ENOMEM);
|
|
|
|
|
goto out;
|
|
|
|
|
}
|
|
|
|
|
|
2009-10-07 02:31:07 +08:00
|
|
|
req->r_ino1.ino = CEPH_INO_ROOT;
|
|
|
|
|
req->r_ino1.snap = CEPH_NOSNAP;
|
|
|
|
|
req->r_started = started;
|
2015-05-15 17:02:17 +08:00
|
|
|
req->r_timeout = fsc->client->options->mount_timeout;
|
2009-10-07 02:31:07 +08:00
|
|
|
req->r_args.getattr.mask = cpu_to_le32(CEPH_STAT_CAP_INODE);
|
|
|
|
|
req->r_num_caps = 2;
|
|
|
|
|
err = ceph_mdsc_do_request(mdsc, NULL, req);
|
|
|
|
|
if (err == 0) {
|
2012-01-10 05:34:32 +08:00
|
|
|
struct inode *inode = req->r_target_inode;
|
|
|
|
|
req->r_target_inode = NULL;
|
2009-10-07 02:31:07 +08:00
|
|
|
dout("open_root_inode success\n");
|
2016-09-14 14:53:05 +08:00
|
|
|
root = d_make_root(inode);
|
|
|
|
|
if (!root) {
|
|
|
|
|
root = ERR_PTR(-ENOMEM);
|
|
|
|
|
goto out;
|
2011-11-12 01:48:08 +08:00
|
|
|
}
|
2009-10-07 02:31:07 +08:00
|
|
|
dout("open_root_inode success, root dentry is %p\n", root);
|
|
|
|
|
} else {
|
|
|
|
|
root = ERR_PTR(err);
|
|
|
|
|
}
|
2012-01-10 05:34:32 +08:00
|
|
|
out:
|
2009-10-07 02:31:07 +08:00
|
|
|
ceph_mdsc_put_request(req);
|
|
|
|
|
return root;
|
|
|
|
|
}
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2009-10-07 02:31:07 +08:00
|
|
|
/*
|
|
|
|
|
* mount: join the ceph cluster, and open root directory.
|
|
|
|
|
*/
|
2016-04-21 11:09:55 +08:00
|
|
|
static struct dentry *ceph_real_mount(struct ceph_fs_client *fsc)
|
2009-10-07 02:31:07 +08:00
|
|
|
{
|
|
|
|
|
int err;
|
|
|
|
|
unsigned long started = jiffies; /* note the start time */
|
|
|
|
|
struct dentry *root;
|
2010-04-07 06:14:15 +08:00
|
|
|
int first = 0; /* first vfsmount for this super_block */
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2016-03-12 13:20:48 +08:00
|
|
|
dout("mount start %p\n", fsc);
|
2010-04-07 06:14:15 +08:00
|
|
|
mutex_lock(&fsc->client->mount_mutex);
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2016-03-12 13:20:48 +08:00
|
|
|
if (!fsc->sb->s_root) {
|
2016-09-14 14:53:05 +08:00
|
|
|
const char *path;
|
2016-03-12 13:20:48 +08:00
|
|
|
err = __ceph_open_session(fsc->client, started);
|
|
|
|
|
if (err < 0)
|
|
|
|
|
goto out;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2017-06-27 11:57:56 +08:00
|
|
|
/* setup fscache */
|
|
|
|
|
if (fsc->mount_options->flags & CEPH_MOUNT_OPT_FSCACHE) {
|
|
|
|
|
err = ceph_fscache_register_fs(fsc);
|
|
|
|
|
if (err < 0)
|
|
|
|
|
goto out;
|
|
|
|
|
}
|
|
|
|
|
|
2016-09-14 14:53:05 +08:00
|
|
|
if (!fsc->mount_options->server_path) {
|
|
|
|
|
path = "";
|
|
|
|
|
dout("mount opening path \\t\n");
|
|
|
|
|
} else {
|
|
|
|
|
path = fsc->mount_options->server_path + 1;
|
|
|
|
|
dout("mount opening path %s\n", path);
|
|
|
|
|
}
|
|
|
|
|
root = open_root_dentry(fsc, path, started);
|
2016-03-12 13:20:48 +08:00
|
|
|
if (IS_ERR(root)) {
|
|
|
|
|
err = PTR_ERR(root);
|
|
|
|
|
goto out;
|
|
|
|
|
}
|
2016-09-14 14:53:05 +08:00
|
|
|
fsc->sb->s_root = dget(root);
|
2010-04-07 06:14:15 +08:00
|
|
|
first = 1;
|
|
|
|
|
|
|
|
|
|
err = ceph_fs_debugfs_init(fsc);
|
|
|
|
|
if (err < 0)
|
|
|
|
|
goto fail;
|
2016-10-13 23:15:37 +08:00
|
|
|
} else {
|
|
|
|
|
root = dget(fsc->sb->s_root);
|
2010-04-07 06:14:15 +08:00
|
|
|
}
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
fsc->mount_state = CEPH_MOUNT_MOUNTED;
|
2009-10-07 02:31:07 +08:00
|
|
|
dout("mount success\n");
|
2010-07-26 20:17:55 +08:00
|
|
|
mutex_unlock(&fsc->client->mount_mutex);
|
|
|
|
|
return root;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
fail:
|
|
|
|
|
if (first) {
|
|
|
|
|
dput(fsc->sb->s_root);
|
|
|
|
|
fsc->sb->s_root = NULL;
|
|
|
|
|
}
|
2016-03-12 13:20:48 +08:00
|
|
|
out:
|
|
|
|
|
mutex_unlock(&fsc->client->mount_mutex);
|
|
|
|
|
return ERR_PTR(err);
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int ceph_set_super(struct super_block *s, void *data)
|
|
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_fs_client *fsc = data;
|
2009-10-07 02:31:07 +08:00
|
|
|
int ret;
|
|
|
|
|
|
|
|
|
|
dout("set_super %p data %p\n", s, data);
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
s->s_flags = fsc->mount_options->sb_flags;
|
2009-10-07 02:31:07 +08:00
|
|
|
s->s_maxbytes = 1ULL << 40; /* temp value until we get mdsmap */
|
|
|
|
|
|
2013-11-11 15:18:03 +08:00
|
|
|
s->s_xattr = ceph_xattr_handlers;
|
2010-04-07 06:14:15 +08:00
|
|
|
s->s_fs_info = fsc;
|
|
|
|
|
fsc->sb = s;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
|
|
|
|
s->s_op = &ceph_super_ops;
|
2016-10-29 09:52:50 +08:00
|
|
|
s->s_d_op = &ceph_dentry_ops;
|
2009-10-07 02:31:07 +08:00
|
|
|
s->s_export_op = &ceph_export_ops;
|
|
|
|
|
|
|
|
|
|
s->s_time_gran = 1000; /* 1000 ns == 1 us */
|
|
|
|
|
|
|
|
|
|
ret = set_anon_super(s, NULL); /* what is that second arg for? */
|
|
|
|
|
if (ret != 0)
|
|
|
|
|
goto fail;
|
|
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
|
|
fail:
|
|
|
|
|
s->s_fs_info = NULL;
|
2010-04-07 06:14:15 +08:00
|
|
|
fsc->sb = NULL;
|
2009-10-07 02:31:07 +08:00
|
|
|
return ret;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* share superblock if same fs AND options
|
|
|
|
|
*/
|
|
|
|
|
static int ceph_compare_super(struct super_block *sb, void *data)
|
|
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_fs_client *new = data;
|
|
|
|
|
struct ceph_mount_options *fsopt = new->mount_options;
|
|
|
|
|
struct ceph_options *opt = new->client->options;
|
|
|
|
|
struct ceph_fs_client *other = ceph_sb_to_client(sb);
|
2009-10-07 02:31:07 +08:00
|
|
|
|
|
|
|
|
dout("ceph_compare_super %p\n", sb);
|
2010-04-07 06:14:15 +08:00
|
|
|
|
|
|
|
|
if (compare_mount_options(fsopt, opt, other)) {
|
|
|
|
|
dout("monitor(s)/mount options don't match\n");
|
|
|
|
|
return 0;
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
2010-04-07 06:14:15 +08:00
|
|
|
if ((opt->flags & CEPH_OPT_FSID) &&
|
|
|
|
|
ceph_fsid_compare(&opt->fsid, &other->client->fsid)) {
|
|
|
|
|
dout("fsid doesn't match\n");
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
if (fsopt->sb_flags != other->mount_options->sb_flags) {
|
2009-10-07 02:31:07 +08:00
|
|
|
dout("flags differ\n");
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
return 1;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/*
|
|
|
|
|
* construct our own bdi so we can control readahead, etc.
|
|
|
|
|
*/
|
2010-06-10 23:13:58 +08:00
|
|
|
static atomic_long_t bdi_seq = ATOMIC_LONG_INIT(0);
|
2010-05-05 07:39:35 +08:00
|
|
|
|
2017-04-12 18:24:33 +08:00
|
|
|
static int ceph_setup_bdi(struct super_block *sb, struct ceph_fs_client *fsc)
|
2009-10-07 02:31:07 +08:00
|
|
|
{
|
|
|
|
|
int err;
|
|
|
|
|
|
2017-04-12 18:24:33 +08:00
|
|
|
err = super_setup_bdi_name(sb, "ceph-%ld",
|
|
|
|
|
atomic_long_inc_return(&bdi_seq));
|
|
|
|
|
if (err)
|
|
|
|
|
return err;
|
|
|
|
|
|
2011-08-04 23:03:44 +08:00
|
|
|
/* set ra_pages based on rasize mount option? */
|
2017-07-11 18:49:44 +08:00
|
|
|
sb->s_bdi->ra_pages = fsc->mount_options->rasize >> PAGE_SHIFT;
|
2011-07-23 02:12:28 +08:00
|
|
|
|
2017-07-11 15:56:09 +08:00
|
|
|
/* set io_pages based on max osd read size */
|
|
|
|
|
sb->s_bdi->io_pages = fsc->mount_options->rsize >> PAGE_SHIFT;
|
2017-01-10 21:17:56 +08:00
|
|
|
|
2017-04-12 18:24:33 +08:00
|
|
|
return 0;
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
|
|
|
|
|
2010-07-26 20:17:55 +08:00
|
|
|
static struct dentry *ceph_mount(struct file_system_type *fs_type,
|
|
|
|
|
int flags, const char *dev_name, void *data)
|
2009-10-07 02:31:07 +08:00
|
|
|
{
|
|
|
|
|
struct super_block *sb;
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_fs_client *fsc;
|
2010-07-26 20:17:55 +08:00
|
|
|
struct dentry *res;
|
2009-10-07 02:31:07 +08:00
|
|
|
int err;
|
|
|
|
|
int (*compare_super)(struct super_block *, void *) = ceph_compare_super;
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_mount_options *fsopt = NULL;
|
|
|
|
|
struct ceph_options *opt = NULL;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-07-26 20:17:55 +08:00
|
|
|
dout("ceph_mount\n");
|
2014-02-17 02:05:29 +08:00
|
|
|
|
|
|
|
|
#ifdef CONFIG_CEPH_FS_POSIX_ACL
|
|
|
|
|
flags |= MS_POSIXACL;
|
|
|
|
|
#endif
|
2016-04-21 11:09:55 +08:00
|
|
|
err = parse_mount_options(&fsopt, &opt, flags, data, dev_name);
|
2010-07-26 20:17:55 +08:00
|
|
|
if (err < 0) {
|
|
|
|
|
res = ERR_PTR(err);
|
2009-10-28 02:50:50 +08:00
|
|
|
goto out_final;
|
2010-07-26 20:17:55 +08:00
|
|
|
}
|
2009-10-07 02:31:07 +08:00
|
|
|
|
|
|
|
|
/* create client (which we may/may not use) */
|
2010-04-07 06:14:15 +08:00
|
|
|
fsc = create_fs_client(fsopt, opt);
|
|
|
|
|
if (IS_ERR(fsc)) {
|
2010-07-26 20:17:55 +08:00
|
|
|
res = ERR_CAST(fsc);
|
2011-08-23 03:49:41 +08:00
|
|
|
destroy_mount_options(fsopt);
|
|
|
|
|
ceph_destroy_options(opt);
|
2009-10-28 02:50:50 +08:00
|
|
|
goto out_final;
|
|
|
|
|
}
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
err = ceph_mdsc_init(fsc);
|
2010-07-26 20:17:55 +08:00
|
|
|
if (err < 0) {
|
|
|
|
|
res = ERR_PTR(err);
|
2010-04-07 06:14:15 +08:00
|
|
|
goto out;
|
2010-07-26 20:17:55 +08:00
|
|
|
}
|
2010-04-07 06:14:15 +08:00
|
|
|
|
|
|
|
|
if (ceph_test_opt(fsc->client, NOSHARE))
|
2009-10-07 02:31:07 +08:00
|
|
|
compare_super = NULL;
|
2012-06-25 19:55:37 +08:00
|
|
|
sb = sget(fs_type, compare_super, ceph_set_super, flags, fsc);
|
2009-10-07 02:31:07 +08:00
|
|
|
if (IS_ERR(sb)) {
|
2010-07-26 20:17:55 +08:00
|
|
|
res = ERR_CAST(sb);
|
2009-10-07 02:31:07 +08:00
|
|
|
goto out;
|
|
|
|
|
}
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
if (ceph_sb_to_client(sb) != fsc) {
|
|
|
|
|
ceph_mdsc_destroy(fsc);
|
|
|
|
|
destroy_fs_client(fsc);
|
|
|
|
|
fsc = ceph_sb_to_client(sb);
|
|
|
|
|
dout("get_sb got existing client %p\n", fsc);
|
2009-10-07 02:31:07 +08:00
|
|
|
} else {
|
2010-04-07 06:14:15 +08:00
|
|
|
dout("get_sb using new client %p\n", fsc);
|
2017-04-12 18:24:33 +08:00
|
|
|
err = ceph_setup_bdi(sb, fsc);
|
2010-07-26 20:17:55 +08:00
|
|
|
if (err < 0) {
|
|
|
|
|
res = ERR_PTR(err);
|
2009-10-07 02:31:07 +08:00
|
|
|
goto out_splat;
|
2010-07-26 20:17:55 +08:00
|
|
|
}
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
|
|
|
|
|
2016-04-21 11:09:55 +08:00
|
|
|
res = ceph_real_mount(fsc);
|
2010-07-26 20:17:55 +08:00
|
|
|
if (IS_ERR(res))
|
2009-10-07 02:31:07 +08:00
|
|
|
goto out_splat;
|
2010-07-26 20:17:55 +08:00
|
|
|
dout("root %p inode %p ino %llx.%llx\n", res,
|
2015-03-18 06:25:59 +08:00
|
|
|
d_inode(res), ceph_vinop(d_inode(res)));
|
2010-07-26 20:17:55 +08:00
|
|
|
return res;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
|
|
|
|
out_splat:
|
2010-04-07 06:14:15 +08:00
|
|
|
ceph_mdsc_close_sessions(fsc->mdsc);
|
2010-03-22 07:22:29 +08:00
|
|
|
deactivate_locked_super(sb);
|
2009-10-07 02:31:07 +08:00
|
|
|
goto out_final;
|
|
|
|
|
|
|
|
|
|
out:
|
2010-04-07 06:14:15 +08:00
|
|
|
ceph_mdsc_destroy(fsc);
|
|
|
|
|
destroy_fs_client(fsc);
|
2009-10-07 02:31:07 +08:00
|
|
|
out_final:
|
2010-07-26 20:17:55 +08:00
|
|
|
dout("ceph_mount fail %ld\n", PTR_ERR(res));
|
|
|
|
|
return res;
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void ceph_kill_sb(struct super_block *s)
|
|
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
struct ceph_fs_client *fsc = ceph_sb_to_client(s);
|
2015-01-14 17:42:38 +08:00
|
|
|
dev_t dev = s->s_dev;
|
|
|
|
|
|
2009-10-07 02:31:07 +08:00
|
|
|
dout("kill_sb %p\n", s);
|
2015-01-14 17:42:38 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
ceph_mdsc_pre_umount(fsc->mdsc);
|
2015-01-14 17:42:38 +08:00
|
|
|
generic_shutdown_super(s);
|
2017-06-22 16:26:34 +08:00
|
|
|
|
|
|
|
|
fsc->client->extra_mon_dispatch = NULL;
|
|
|
|
|
ceph_fs_debugfs_cleanup(fsc);
|
|
|
|
|
|
2017-06-27 11:57:56 +08:00
|
|
|
ceph_fscache_unregister_fs(fsc);
|
|
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
ceph_mdsc_destroy(fsc);
|
2015-01-14 17:42:38 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
destroy_fs_client(fsc);
|
2015-01-14 17:42:38 +08:00
|
|
|
free_anon_bdev(dev);
|
2009-10-07 02:31:07 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static struct file_system_type ceph_fs_type = {
|
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
|
.name = "ceph",
|
2010-07-26 20:17:55 +08:00
|
|
|
.mount = ceph_mount,
|
2009-10-07 02:31:07 +08:00
|
|
|
.kill_sb = ceph_kill_sb,
|
|
|
|
|
.fs_flags = FS_RENAME_DOES_D_MOVE,
|
|
|
|
|
};
|
2013-03-03 11:39:14 +08:00
|
|
|
MODULE_ALIAS_FS("ceph");
|
2009-10-07 02:31:07 +08:00
|
|
|
|
|
|
|
|
static int __init init_ceph(void)
|
|
|
|
|
{
|
2010-04-07 06:14:15 +08:00
|
|
|
int ret = init_caches();
|
2009-10-07 02:31:07 +08:00
|
|
|
if (ret)
|
2010-04-07 06:14:15 +08:00
|
|
|
goto out;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2014-03-09 23:16:40 +08:00
|
|
|
ceph_flock_init();
|
2012-01-24 05:49:28 +08:00
|
|
|
ceph_xattr_init();
|
2009-10-07 02:31:07 +08:00
|
|
|
ret = register_filesystem(&ceph_fs_type);
|
|
|
|
|
if (ret)
|
2016-02-16 22:00:24 +08:00
|
|
|
goto out_xattr;
|
2009-10-07 02:31:07 +08:00
|
|
|
|
2010-04-07 06:14:15 +08:00
|
|
|
pr_info("loaded (mds proto %d)\n", CEPH_MDSC_PROTOCOL);
|
|
|
|
|
|
2009-10-07 02:31:07 +08:00
|
|
|
return 0;
|
|
|
|
|
|
2014-11-06 15:09:41 +08:00
|
|
|
out_xattr:
|
2012-01-24 05:49:28 +08:00
|
|
|
ceph_xattr_exit();
|
2009-10-07 02:31:07 +08:00
|
|
|
destroy_caches();
|
|
|
|
|
out:
|
|
|
|
|
return ret;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void __exit exit_ceph(void)
|
|
|
|
|
{
|
|
|
|
|
dout("exit_ceph\n");
|
|
|
|
|
unregister_filesystem(&ceph_fs_type);
|
2012-01-24 05:49:28 +08:00
|
|
|
ceph_xattr_exit();
|
2009-10-07 02:31:07 +08:00
|
|
|
destroy_caches();
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
module_init(init_ceph);
|
|
|
|
|
module_exit(exit_ceph);
|
|
|
|
|
|
|
|
|
|
MODULE_AUTHOR("Sage Weil <sage@newdream.net>");
|
|
|
|
|
MODULE_AUTHOR("Yehuda Sadeh <yehuda@hq.newdream.net>");
|
|
|
|
|
MODULE_AUTHOR("Patience Warnick <patience@newdream.net>");
|
|
|
|
|
MODULE_DESCRIPTION("Ceph filesystem for Linux");
|
|
|
|
|
MODULE_LICENSE("GPL");
|