2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
#include <linux/ceph/ceph_debug.h>
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
#include <linux/backing-dev.h>
|
2010-07-01 06:34:01 +02:00
|
|
|
#include <linux/ctype.h>
|
2009-10-06 20:31:07 +02:00
|
|
|
#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 09:04:11 +01:00
|
|
|
#include <linux/slab.h>
|
2009-10-06 20:31:07 +02:00
|
|
|
#include <linux/statfs.h>
|
|
|
|
#include <linux/string.h>
|
|
|
|
|
|
|
|
#include "super.h"
|
2010-04-07 00:14:15 +02:00
|
|
|
#include "mds_client.h"
|
|
|
|
|
|
|
|
#include <linux/ceph/decode.h>
|
|
|
|
#include <linux/ceph/mon_client.h>
|
|
|
|
#include <linux/ceph/auth.h>
|
|
|
|
#include <linux/ceph/debugfs.h>
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Ceph superblock operations
|
|
|
|
*
|
|
|
|
* Handle the basics of mounting, unmounting.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* super ops
|
|
|
|
*/
|
|
|
|
static void ceph_put_super(struct super_block *s)
|
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_fs_client *fsc = ceph_sb_to_client(s);
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
dout("put_super\n");
|
2010-04-07 00:14:15 +02:00
|
|
|
ceph_mdsc_close_sessions(fsc->mdsc);
|
2010-05-05 01:14:46 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* ensure we release the bdi before put_anon_super releases
|
|
|
|
* the device name.
|
|
|
|
*/
|
2010-04-07 00:14:15 +02:00
|
|
|
if (s->s_bdi == &fsc->backing_dev_info) {
|
|
|
|
bdi_unregister(&fsc->backing_dev_info);
|
2010-05-05 01:14:46 +02:00
|
|
|
s->s_bdi = NULL;
|
|
|
|
}
|
|
|
|
|
2009-10-06 20:31:07 +02:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf)
|
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_fs_client *fsc = ceph_inode_to_client(dentry->d_inode);
|
|
|
|
struct ceph_monmap *monmap = fsc->client->monc.monmap;
|
2009-10-06 20:31:07 +02:00
|
|
|
struct ceph_statfs st;
|
|
|
|
u64 fsid;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
dout("statfs\n");
|
2010-04-07 00:14:15 +02:00
|
|
|
err = ceph_monc_do_statfs(&fsc->client->monc, &st);
|
2009-10-06 20:31:07 +02:00
|
|
|
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.
|
|
|
|
*/
|
|
|
|
buf->f_bsize = 1 << CEPH_BLOCK_SHIFT;
|
|
|
|
buf->f_blocks = le64_to_cpu(st.kb) >> (CEPH_BLOCK_SHIFT-10);
|
2011-07-26 20:26:54 +02:00
|
|
|
buf->f_bfree = le64_to_cpu(st.kb_avail) >> (CEPH_BLOCK_SHIFT-10);
|
2009-10-06 20:31:07 +02:00
|
|
|
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;
|
2010-06-01 21:51:12 +02:00
|
|
|
buf->f_namelen = NAME_MAX;
|
2009-10-06 20:31:07 +02:00
|
|
|
buf->f_frsize = PAGE_CACHE_SIZE;
|
|
|
|
|
|
|
|
/* leave fsid little-endian, regardless of host endianness */
|
|
|
|
fsid = *(u64 *)(&monmap->fsid) ^ *((u64 *)&monmap->fsid + 1);
|
|
|
|
buf->f_fsid.val[0] = fsid & 0xffffffff;
|
|
|
|
buf->f_fsid.val[1] = fsid >> 32;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2010-07-30 18:38:13 +02:00
|
|
|
static int ceph_sync_fs(struct super_block *sb, int wait)
|
2009-10-06 20:31:07 +02:00
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
|
2010-07-30 18:38:13 +02:00
|
|
|
|
|
|
|
if (!wait) {
|
|
|
|
dout("sync_fs (non-blocking)\n");
|
2010-04-07 00:14:15 +02:00
|
|
|
ceph_flush_dirty_caps(fsc->mdsc);
|
2010-07-30 18:38:13 +02:00
|
|
|
dout("sync_fs (non-blocking) done\n");
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
dout("sync_fs (blocking)\n");
|
2010-04-07 00:14:15 +02:00
|
|
|
ceph_osdc_sync(&fsc->client->osdc);
|
|
|
|
ceph_mdsc_sync(fsc->mdsc);
|
2010-07-30 18:38:13 +02:00
|
|
|
dout("sync_fs (blocking) done\n");
|
2009-10-06 20:31:07 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* mount options
|
|
|
|
*/
|
|
|
|
enum {
|
|
|
|
Opt_wsize,
|
|
|
|
Opt_rsize,
|
2011-08-04 17:03:44 +02:00
|
|
|
Opt_rasize,
|
2009-10-06 20:31:07 +02:00
|
|
|
Opt_caps_wanted_delay_min,
|
|
|
|
Opt_caps_wanted_delay_max,
|
2010-04-30 01:38:32 +02:00
|
|
|
Opt_cap_release_safety,
|
2009-10-06 20:31:07 +02:00
|
|
|
Opt_readdir_max_entries,
|
2010-05-14 22:06:30 +02:00
|
|
|
Opt_readdir_max_bytes,
|
2009-12-18 22:51:57 +01:00
|
|
|
Opt_congestion_kb,
|
2009-10-27 18:19:28 +01:00
|
|
|
Opt_last_int,
|
2009-10-06 20:31:07 +02:00
|
|
|
/* int args above */
|
|
|
|
Opt_snapdirname,
|
2009-10-27 18:19:28 +01:00
|
|
|
Opt_last_string,
|
2009-10-06 20:31:07 +02:00
|
|
|
/* string args above */
|
|
|
|
Opt_dirstat,
|
|
|
|
Opt_nodirstat,
|
|
|
|
Opt_rbytes,
|
|
|
|
Opt_norbytes,
|
|
|
|
Opt_noasyncreaddir,
|
2011-01-22 01:44:03 +01:00
|
|
|
Opt_ino32,
|
2009-10-06 20:31:07 +02:00
|
|
|
};
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
static match_table_t fsopt_tokens = {
|
2009-10-06 20:31:07 +02:00
|
|
|
{Opt_wsize, "wsize=%d"},
|
|
|
|
{Opt_rsize, "rsize=%d"},
|
2011-08-04 17:03:44 +02:00
|
|
|
{Opt_rasize, "rasize=%d"},
|
2009-10-06 20:31:07 +02:00
|
|
|
{Opt_caps_wanted_delay_min, "caps_wanted_delay_min=%d"},
|
|
|
|
{Opt_caps_wanted_delay_max, "caps_wanted_delay_max=%d"},
|
2010-04-30 01:38:32 +02:00
|
|
|
{Opt_cap_release_safety, "cap_release_safety=%d"},
|
2009-10-06 20:31:07 +02:00
|
|
|
{Opt_readdir_max_entries, "readdir_max_entries=%d"},
|
2010-05-14 22:06:30 +02:00
|
|
|
{Opt_readdir_max_bytes, "readdir_max_bytes=%d"},
|
2009-12-18 22:51:57 +01:00
|
|
|
{Opt_congestion_kb, "write_congestion_kb=%d"},
|
2009-10-06 20:31:07 +02:00
|
|
|
/* int args above */
|
|
|
|
{Opt_snapdirname, "snapdirname=%s"},
|
|
|
|
/* string args above */
|
|
|
|
{Opt_dirstat, "dirstat"},
|
|
|
|
{Opt_nodirstat, "nodirstat"},
|
|
|
|
{Opt_rbytes, "rbytes"},
|
|
|
|
{Opt_norbytes, "norbytes"},
|
|
|
|
{Opt_noasyncreaddir, "noasyncreaddir"},
|
2011-01-22 01:44:03 +01:00
|
|
|
{Opt_ino32, "ino32"},
|
2009-10-06 20:31:07 +02:00
|
|
|
{-1, NULL}
|
|
|
|
};
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
static int parse_fsopt_token(char *c, void *private)
|
2010-07-01 06:34:01 +02:00
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
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 06:34:01 +02:00
|
|
|
}
|
2010-04-07 00:14:15 +02:00
|
|
|
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 06:34:01 +02:00
|
|
|
}
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
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;
|
|
|
|
|
|
|
|
/* misc */
|
|
|
|
case Opt_wsize:
|
|
|
|
fsopt->wsize = intval;
|
|
|
|
break;
|
|
|
|
case Opt_rsize:
|
|
|
|
fsopt->rsize = intval;
|
|
|
|
break;
|
2011-08-04 17:03:44 +02:00
|
|
|
case Opt_rasize:
|
|
|
|
fsopt->rasize = intval;
|
|
|
|
break;
|
2010-04-07 00:14:15 +02:00
|
|
|
case Opt_caps_wanted_delay_min:
|
|
|
|
fsopt->caps_wanted_delay_min = intval;
|
|
|
|
break;
|
|
|
|
case Opt_caps_wanted_delay_max:
|
|
|
|
fsopt->caps_wanted_delay_max = intval;
|
|
|
|
break;
|
|
|
|
case Opt_readdir_max_entries:
|
|
|
|
fsopt->max_readdir = intval;
|
|
|
|
break;
|
|
|
|
case Opt_readdir_max_bytes:
|
|
|
|
fsopt->max_readdir_bytes = intval;
|
|
|
|
break;
|
|
|
|
case Opt_congestion_kb:
|
|
|
|
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;
|
|
|
|
case Opt_noasyncreaddir:
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_NOASYNCREADDIR;
|
|
|
|
break;
|
2011-01-22 01:44:03 +01:00
|
|
|
case Opt_ino32:
|
|
|
|
fsopt->flags |= CEPH_MOUNT_OPT_INO32;
|
|
|
|
break;
|
2010-04-07 00:14:15 +02:00
|
|
|
default:
|
|
|
|
BUG_ON(token);
|
|
|
|
}
|
|
|
|
return 0;
|
2010-07-01 06:34:01 +02:00
|
|
|
}
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
static void destroy_mount_options(struct ceph_mount_options *args)
|
2009-10-06 20:31:07 +02:00
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
dout("destroy_mount_options %p\n", args);
|
|
|
|
kfree(args->snapdir_name);
|
|
|
|
kfree(args);
|
|
|
|
}
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
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);
|
|
|
|
}
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
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;
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
ret = memcmp(fsopt1, fsopt2, ofs);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
ret = strcmp_null(fsopt1->snapdir_name, fsopt2->snapdir_name);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
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,
|
|
|
|
const char *dev_name,
|
|
|
|
const char **path)
|
|
|
|
{
|
|
|
|
struct ceph_mount_options *fsopt;
|
|
|
|
const char *dev_name_end;
|
|
|
|
int err = -ENOMEM;
|
|
|
|
|
|
|
|
fsopt = kzalloc(sizeof(*fsopt), GFP_KERNEL);
|
|
|
|
if (!fsopt)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
dout("parse_mount_options %p, dev_name '%s'\n", fsopt, dev_name);
|
|
|
|
|
2011-08-22 21:49:23 +02:00
|
|
|
fsopt->sb_flags = flags;
|
|
|
|
fsopt->flags = CEPH_MOUNT_OPT_DEFAULT;
|
2010-04-07 00:14:15 +02:00
|
|
|
|
2011-08-22 21:49:23 +02:00
|
|
|
fsopt->rsize = CEPH_RSIZE_DEFAULT;
|
|
|
|
fsopt->rasize = CEPH_RASIZE_DEFAULT;
|
|
|
|
fsopt->snapdir_name = kstrdup(CEPH_SNAPDIRNAME_DEFAULT, GFP_KERNEL);
|
2011-01-18 16:59:40 +01:00
|
|
|
fsopt->caps_wanted_delay_min = CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT;
|
|
|
|
fsopt->caps_wanted_delay_max = CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT;
|
2011-08-22 21:49:23 +02:00
|
|
|
fsopt->cap_release_safety = CEPH_CAP_RELEASE_SAFETY_DEFAULT;
|
|
|
|
fsopt->max_readdir = CEPH_MAX_READDIR_DEFAULT;
|
|
|
|
fsopt->max_readdir_bytes = CEPH_MAX_READDIR_BYTES_DEFAULT;
|
|
|
|
fsopt->congestion_kb = default_congestion_kb();
|
|
|
|
|
|
|
|
/* ip1[:port1][,ip2[:port2]...]:/subdir/in/fs */
|
|
|
|
err = -EINVAL;
|
|
|
|
if (!dev_name)
|
|
|
|
goto out;
|
|
|
|
*path = strstr(dev_name, ":/");
|
|
|
|
if (*path == NULL) {
|
|
|
|
pr_err("device name is missing path (no :/ in %s)\n",
|
|
|
|
dev_name);
|
|
|
|
goto out;
|
|
|
|
}
|
2010-04-07 00:14:15 +02:00
|
|
|
dev_name_end = *path;
|
|
|
|
dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name);
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
/* path on server */
|
|
|
|
*path += 2;
|
|
|
|
dout("server path '%s'\n", *path);
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
err = ceph_parse_options(popt, options, dev_name, dev_name_end,
|
|
|
|
parse_fsopt_token, (void *)fsopt);
|
|
|
|
if (err)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
/* success */
|
|
|
|
*pfsopt = fsopt;
|
|
|
|
return 0;
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2009-10-27 06:07:53 +01:00
|
|
|
out:
|
2010-04-07 00:14:15 +02:00
|
|
|
destroy_mount_options(fsopt);
|
|
|
|
return err;
|
2009-10-06 20:31:07 +02:00
|
|
|
}
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
/**
|
|
|
|
* ceph_show_options - Show mount options in /proc/mounts
|
|
|
|
* @m: seq_file to write to
|
|
|
|
* @mnt: mount descriptor
|
|
|
|
*/
|
|
|
|
static int ceph_show_options(struct seq_file *m, struct vfsmount *mnt)
|
2009-10-06 20:31:07 +02:00
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_fs_client *fsc = ceph_sb_to_client(mnt->mnt_sb);
|
|
|
|
struct ceph_mount_options *fsopt = fsc->mount_options;
|
|
|
|
struct ceph_options *opt = fsc->client->options;
|
|
|
|
|
|
|
|
if (opt->flags & CEPH_OPT_FSID)
|
|
|
|
seq_printf(m, ",fsid=%pU", &opt->fsid);
|
|
|
|
if (opt->flags & CEPH_OPT_NOSHARE)
|
|
|
|
seq_puts(m, ",noshare");
|
|
|
|
if (opt->flags & CEPH_OPT_NOCRC)
|
|
|
|
seq_puts(m, ",nocrc");
|
|
|
|
|
|
|
|
if (opt->name)
|
|
|
|
seq_printf(m, ",name=%s", opt->name);
|
2011-03-26 00:32:57 +01:00
|
|
|
if (opt->key)
|
2010-04-07 00:14:15 +02:00
|
|
|
seq_puts(m, ",secret=<hidden>");
|
|
|
|
|
|
|
|
if (opt->mount_timeout != CEPH_MOUNT_TIMEOUT_DEFAULT)
|
|
|
|
seq_printf(m, ",mount_timeout=%d", opt->mount_timeout);
|
|
|
|
if (opt->osd_idle_ttl != CEPH_OSD_IDLE_TTL_DEFAULT)
|
|
|
|
seq_printf(m, ",osd_idle_ttl=%d", opt->osd_idle_ttl);
|
|
|
|
if (opt->osd_timeout != CEPH_OSD_TIMEOUT_DEFAULT)
|
|
|
|
seq_printf(m, ",osdtimeout=%d", opt->osd_timeout);
|
|
|
|
if (opt->osd_keepalive_timeout != CEPH_OSD_KEEPALIVE_DEFAULT)
|
|
|
|
seq_printf(m, ",osdkeepalivetimeout=%d",
|
|
|
|
opt->osd_keepalive_timeout);
|
|
|
|
|
|
|
|
if (fsopt->flags & CEPH_MOUNT_OPT_DIRSTAT)
|
|
|
|
seq_puts(m, ",dirstat");
|
|
|
|
if ((fsopt->flags & CEPH_MOUNT_OPT_RBYTES) == 0)
|
|
|
|
seq_puts(m, ",norbytes");
|
|
|
|
if (fsopt->flags & CEPH_MOUNT_OPT_NOASYNCREADDIR)
|
|
|
|
seq_puts(m, ",noasyncreaddir");
|
|
|
|
|
|
|
|
if (fsopt->wsize)
|
|
|
|
seq_printf(m, ",wsize=%d", fsopt->wsize);
|
2011-03-10 22:33:26 +01:00
|
|
|
if (fsopt->rsize != CEPH_RSIZE_DEFAULT)
|
2010-04-07 00:14:15 +02:00
|
|
|
seq_printf(m, ",rsize=%d", fsopt->rsize);
|
2011-08-04 17:03:44 +02:00
|
|
|
if (fsopt->rasize != CEPH_RASIZE_DEFAULT)
|
|
|
|
seq_printf(m, ",rasize=%d", fsopt->rsize);
|
2010-04-07 00:14:15 +02:00
|
|
|
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->cap_release_safety != CEPH_CAP_RELEASE_SAFETY_DEFAULT)
|
|
|
|
seq_printf(m, ",cap_release_safety=%d",
|
|
|
|
fsopt->cap_release_safety);
|
|
|
|
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))
|
|
|
|
seq_printf(m, ",snapdirname=%s", fsopt->snapdir_name);
|
|
|
|
return 0;
|
2009-10-06 20:31:07 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2010-04-07 00:14:15 +02:00
|
|
|
* handle any mon messages the standard library doesn't understand.
|
|
|
|
* return error if we don't either.
|
2009-10-06 20:31:07 +02:00
|
|
|
*/
|
2010-04-07 00:14:15 +02:00
|
|
|
static int extra_mon_dispatch(struct ceph_client *client, struct ceph_msg *msg)
|
2009-10-06 20:31:07 +02:00
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_fs_client *fsc = client->private;
|
|
|
|
int type = le16_to_cpu(msg->hdr.type);
|
|
|
|
|
|
|
|
switch (type) {
|
|
|
|
case CEPH_MSG_MDS_MAP:
|
|
|
|
ceph_mdsc_handle_map(fsc->mdsc, msg);
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
default:
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* create a new fs client
|
|
|
|
*/
|
2011-09-23 20:53:30 +02:00
|
|
|
static struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt,
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_options *opt)
|
|
|
|
{
|
|
|
|
struct ceph_fs_client *fsc;
|
2011-08-09 18:41:59 +02:00
|
|
|
const unsigned supported_features =
|
|
|
|
CEPH_FEATURE_FLOCK |
|
|
|
|
CEPH_FEATURE_DIRLAYOUTHASH;
|
|
|
|
const unsigned required_features = 0;
|
2009-10-06 20:31:07 +02:00
|
|
|
int err = -ENOMEM;
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
fsc = kzalloc(sizeof(*fsc), GFP_KERNEL);
|
|
|
|
if (!fsc)
|
2009-10-06 20:31:07 +02:00
|
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
|
2011-08-09 18:41:59 +02:00
|
|
|
fsc->client = ceph_create_client(opt, fsc, supported_features,
|
|
|
|
required_features);
|
2010-04-07 00:14:15 +02:00
|
|
|
if (IS_ERR(fsc->client)) {
|
|
|
|
err = PTR_ERR(fsc->client);
|
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
fsc->client->extra_mon_dispatch = extra_mon_dispatch;
|
|
|
|
fsc->client->monc.want_mdsmap = 1;
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
fsc->mount_options = fsopt;
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
fsc->sb = NULL;
|
|
|
|
fsc->mount_state = CEPH_MOUNT_MOUNTING;
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
atomic_long_set(&fsc->writeback_count, 0);
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
err = bdi_init(&fsc->backing_dev_info);
|
2009-11-02 18:32:47 +01:00
|
|
|
if (err < 0)
|
2010-04-07 00:14:15 +02:00
|
|
|
goto fail_client;
|
2009-11-02 18:32:47 +01:00
|
|
|
|
2009-10-06 20:31:07 +02:00
|
|
|
err = -ENOMEM;
|
2011-01-03 14:49:45 +01:00
|
|
|
/*
|
|
|
|
* 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);
|
2010-04-07 00:14:15 +02:00
|
|
|
if (fsc->wb_wq == NULL)
|
2009-11-02 18:32:47 +01:00
|
|
|
goto fail_bdi;
|
2011-01-03 14:49:45 +01:00
|
|
|
fsc->pg_inv_wq = alloc_workqueue("ceph-pg-invalid", 0, 1);
|
2010-04-07 00:14:15 +02:00
|
|
|
if (fsc->pg_inv_wq == NULL)
|
2009-10-06 20:31:07 +02:00
|
|
|
goto fail_wb_wq;
|
2011-01-03 14:49:45 +01:00
|
|
|
fsc->trunc_wq = alloc_workqueue("ceph-trunc", 0, 1);
|
2010-04-07 00:14:15 +02:00
|
|
|
if (fsc->trunc_wq == NULL)
|
2009-10-06 20:31:07 +02:00
|
|
|
goto fail_pg_inv_wq;
|
|
|
|
|
2009-11-19 00:08:44 +01:00
|
|
|
/* set up mempools */
|
|
|
|
err = -ENOMEM;
|
2010-04-07 00:14:15 +02:00
|
|
|
fsc->wb_pagevec_pool = mempool_create_kmalloc_pool(10,
|
|
|
|
fsc->mount_options->wsize >> PAGE_CACHE_SHIFT);
|
|
|
|
if (!fsc->wb_pagevec_pool)
|
2009-11-19 00:08:44 +01:00
|
|
|
goto fail_trunc_wq;
|
|
|
|
|
2010-02-17 19:02:43 +01:00
|
|
|
/* caps */
|
2010-04-07 00:14:15 +02:00
|
|
|
fsc->min_caps = fsopt->max_readdir;
|
|
|
|
|
|
|
|
return fsc;
|
2009-11-19 00:08:44 +01:00
|
|
|
|
2009-10-06 20:31:07 +02:00
|
|
|
fail_trunc_wq:
|
2010-04-07 00:14:15 +02:00
|
|
|
destroy_workqueue(fsc->trunc_wq);
|
2009-10-06 20:31:07 +02:00
|
|
|
fail_pg_inv_wq:
|
2010-04-07 00:14:15 +02:00
|
|
|
destroy_workqueue(fsc->pg_inv_wq);
|
2009-10-06 20:31:07 +02:00
|
|
|
fail_wb_wq:
|
2010-04-07 00:14:15 +02:00
|
|
|
destroy_workqueue(fsc->wb_wq);
|
2009-11-02 18:32:47 +01:00
|
|
|
fail_bdi:
|
2010-04-07 00:14:15 +02:00
|
|
|
bdi_destroy(&fsc->backing_dev_info);
|
|
|
|
fail_client:
|
|
|
|
ceph_destroy_client(fsc->client);
|
2009-10-06 20:31:07 +02:00
|
|
|
fail:
|
2010-04-07 00:14:15 +02:00
|
|
|
kfree(fsc);
|
2009-10-06 20:31:07 +02:00
|
|
|
return ERR_PTR(err);
|
|
|
|
}
|
|
|
|
|
2011-09-23 20:53:30 +02:00
|
|
|
static void destroy_fs_client(struct ceph_fs_client *fsc)
|
2009-10-06 20:31:07 +02:00
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
dout("destroy_fs_client %p\n", fsc);
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
destroy_workqueue(fsc->wb_wq);
|
|
|
|
destroy_workqueue(fsc->pg_inv_wq);
|
|
|
|
destroy_workqueue(fsc->trunc_wq);
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
bdi_destroy(&fsc->backing_dev_info);
|
2010-05-29 18:41:23 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
mempool_destroy(fsc->wb_pagevec_pool);
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
destroy_mount_options(fsc->mount_options);
|
2010-05-05 01:14:46 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
ceph_fs_debugfs_cleanup(fsc);
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
ceph_destroy_client(fsc->client);
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
kfree(fsc);
|
|
|
|
dout("destroy_fs_client %p done\n", fsc);
|
2009-10-06 20:31:07 +02:00
|
|
|
}
|
|
|
|
|
2009-11-19 01:50:41 +01:00
|
|
|
/*
|
2010-04-07 00:14:15 +02:00
|
|
|
* caches
|
2009-11-19 01:50:41 +01:00
|
|
|
*/
|
2010-04-07 00:14:15 +02:00
|
|
|
struct kmem_cache *ceph_inode_cachep;
|
|
|
|
struct kmem_cache *ceph_cap_cachep;
|
|
|
|
struct kmem_cache *ceph_dentry_cachep;
|
|
|
|
struct kmem_cache *ceph_file_cachep;
|
|
|
|
|
|
|
|
static void ceph_inode_init_once(void *foo)
|
2009-11-19 01:50:41 +01:00
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_inode_info *ci = foo;
|
|
|
|
inode_init_once(&ci->vfs_inode);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int __init init_caches(void)
|
|
|
|
{
|
|
|
|
ceph_inode_cachep = kmem_cache_create("ceph_inode_info",
|
|
|
|
sizeof(struct ceph_inode_info),
|
|
|
|
__alignof__(struct ceph_inode_info),
|
|
|
|
(SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD),
|
|
|
|
ceph_inode_init_once);
|
|
|
|
if (ceph_inode_cachep == NULL)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
ceph_cap_cachep = KMEM_CACHE(ceph_cap,
|
|
|
|
SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
|
|
|
|
if (ceph_cap_cachep == NULL)
|
|
|
|
goto bad_cap;
|
|
|
|
|
|
|
|
ceph_dentry_cachep = KMEM_CACHE(ceph_dentry_info,
|
|
|
|
SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
|
|
|
|
if (ceph_dentry_cachep == NULL)
|
|
|
|
goto bad_dentry;
|
|
|
|
|
|
|
|
ceph_file_cachep = KMEM_CACHE(ceph_file_info,
|
|
|
|
SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
|
|
|
|
if (ceph_file_cachep == NULL)
|
|
|
|
goto bad_file;
|
|
|
|
|
2009-11-19 01:50:41 +01:00
|
|
|
return 0;
|
2010-04-07 00:14:15 +02:00
|
|
|
|
|
|
|
bad_file:
|
|
|
|
kmem_cache_destroy(ceph_dentry_cachep);
|
|
|
|
bad_dentry:
|
|
|
|
kmem_cache_destroy(ceph_cap_cachep);
|
|
|
|
bad_cap:
|
|
|
|
kmem_cache_destroy(ceph_inode_cachep);
|
|
|
|
return -ENOMEM;
|
2009-11-19 01:50:41 +01:00
|
|
|
}
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
static void destroy_caches(void)
|
|
|
|
{
|
|
|
|
kmem_cache_destroy(ceph_inode_cachep);
|
|
|
|
kmem_cache_destroy(ceph_cap_cachep);
|
|
|
|
kmem_cache_destroy(ceph_dentry_cachep);
|
|
|
|
kmem_cache_destroy(ceph_file_cachep);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2009-10-06 20:31:07 +02:00
|
|
|
/*
|
2010-04-07 00:14:15 +02:00
|
|
|
* ceph_umount_begin - initiate forced umount. Tear down down the
|
|
|
|
* mount, skipping steps that may hang while waiting for server(s).
|
2009-10-06 20:31:07 +02:00
|
|
|
*/
|
2010-04-07 00:14:15 +02:00
|
|
|
static void ceph_umount_begin(struct super_block *sb)
|
2009-10-06 20:31:07 +02:00
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
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;
|
|
|
|
return;
|
2009-10-06 20:31:07 +02:00
|
|
|
}
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
static const struct super_operations ceph_super_ops = {
|
|
|
|
.alloc_inode = ceph_alloc_inode,
|
|
|
|
.destroy_inode = ceph_destroy_inode,
|
|
|
|
.write_inode = ceph_write_inode,
|
|
|
|
.sync_fs = ceph_sync_fs,
|
|
|
|
.put_super = ceph_put_super,
|
|
|
|
.show_options = ceph_show_options,
|
|
|
|
.statfs = ceph_statfs,
|
|
|
|
.umount_begin = ceph_umount_begin,
|
|
|
|
};
|
|
|
|
|
2009-10-06 20:31:07 +02:00
|
|
|
/*
|
|
|
|
* 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 00:14:15 +02:00
|
|
|
static struct dentry *open_root_dentry(struct ceph_fs_client *fsc,
|
2009-10-06 20:31:07 +02:00
|
|
|
const char *path,
|
|
|
|
unsigned long started)
|
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_mds_client *mdsc = fsc->mdsc;
|
2009-10-06 20:31:07 +02:00
|
|
|
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 12:01:14 +02:00
|
|
|
return ERR_CAST(req);
|
2009-10-06 20:31:07 +02:00
|
|
|
req->r_path1 = kstrdup(path, GFP_NOFS);
|
|
|
|
req->r_ino1.ino = CEPH_INO_ROOT;
|
|
|
|
req->r_ino1.snap = CEPH_NOSNAP;
|
|
|
|
req->r_started = started;
|
2010-04-07 00:14:15 +02:00
|
|
|
req->r_timeout = fsc->client->options->mount_timeout * HZ;
|
2009-10-06 20:31:07 +02:00
|
|
|
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) {
|
|
|
|
dout("open_root_inode success\n");
|
|
|
|
if (ceph_ino(req->r_target_inode) == CEPH_INO_ROOT &&
|
2010-04-07 00:14:15 +02:00
|
|
|
fsc->sb->s_root == NULL)
|
2009-10-06 20:31:07 +02:00
|
|
|
root = d_alloc_root(req->r_target_inode);
|
|
|
|
else
|
|
|
|
root = d_obtain_alias(req->r_target_inode);
|
|
|
|
req->r_target_inode = NULL;
|
|
|
|
dout("open_root_inode success, root dentry is %p\n", root);
|
|
|
|
} else {
|
|
|
|
root = ERR_PTR(err);
|
|
|
|
}
|
|
|
|
ceph_mdsc_put_request(req);
|
|
|
|
return root;
|
|
|
|
}
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
|
|
|
|
|
|
|
|
|
2009-10-06 20:31:07 +02:00
|
|
|
/*
|
|
|
|
* mount: join the ceph cluster, and open root directory.
|
|
|
|
*/
|
2010-07-26 14:17:55 +02:00
|
|
|
static struct dentry *ceph_real_mount(struct ceph_fs_client *fsc,
|
2009-10-06 20:31:07 +02:00
|
|
|
const char *path)
|
|
|
|
{
|
|
|
|
int err;
|
|
|
|
unsigned long started = jiffies; /* note the start time */
|
|
|
|
struct dentry *root;
|
2010-04-07 00:14:15 +02:00
|
|
|
int first = 0; /* first vfsmount for this super_block */
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
dout("mount start\n");
|
2010-04-07 00:14:15 +02:00
|
|
|
mutex_lock(&fsc->client->mount_mutex);
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
err = __ceph_open_session(fsc->client, started);
|
2009-10-06 20:31:07 +02:00
|
|
|
if (err < 0)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
dout("mount opening root\n");
|
2010-04-07 00:14:15 +02:00
|
|
|
root = open_root_dentry(fsc, "", started);
|
2009-10-06 20:31:07 +02:00
|
|
|
if (IS_ERR(root)) {
|
|
|
|
err = PTR_ERR(root);
|
|
|
|
goto out;
|
|
|
|
}
|
2010-04-07 00:14:15 +02:00
|
|
|
if (fsc->sb->s_root) {
|
2009-10-06 20:31:07 +02:00
|
|
|
dput(root);
|
2010-04-07 00:14:15 +02:00
|
|
|
} else {
|
|
|
|
fsc->sb->s_root = root;
|
|
|
|
first = 1;
|
|
|
|
|
|
|
|
err = ceph_fs_debugfs_init(fsc);
|
|
|
|
if (err < 0)
|
|
|
|
goto fail;
|
|
|
|
}
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
if (path[0] == 0) {
|
|
|
|
dget(root);
|
|
|
|
} else {
|
|
|
|
dout("mount opening base mountpoint\n");
|
2010-04-07 00:14:15 +02:00
|
|
|
root = open_root_dentry(fsc, path, started);
|
2009-10-06 20:31:07 +02:00
|
|
|
if (IS_ERR(root)) {
|
|
|
|
err = PTR_ERR(root);
|
2010-04-07 00:14:15 +02:00
|
|
|
goto fail;
|
2009-10-06 20:31:07 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
fsc->mount_state = CEPH_MOUNT_MOUNTED;
|
2009-10-06 20:31:07 +02:00
|
|
|
dout("mount success\n");
|
2010-07-26 14:17:55 +02:00
|
|
|
mutex_unlock(&fsc->client->mount_mutex);
|
|
|
|
return root;
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
out:
|
2010-04-07 00:14:15 +02:00
|
|
|
mutex_unlock(&fsc->client->mount_mutex);
|
2010-07-26 14:17:55 +02:00
|
|
|
return ERR_PTR(err);
|
2010-04-07 00:14:15 +02:00
|
|
|
|
|
|
|
fail:
|
|
|
|
if (first) {
|
|
|
|
dput(fsc->sb->s_root);
|
|
|
|
fsc->sb->s_root = NULL;
|
|
|
|
}
|
|
|
|
goto out;
|
2009-10-06 20:31:07 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static int ceph_set_super(struct super_block *s, void *data)
|
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_fs_client *fsc = data;
|
2009-10-06 20:31:07 +02:00
|
|
|
int ret;
|
|
|
|
|
|
|
|
dout("set_super %p data %p\n", s, data);
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
s->s_flags = fsc->mount_options->sb_flags;
|
2009-10-06 20:31:07 +02:00
|
|
|
s->s_maxbytes = 1ULL << 40; /* temp value until we get mdsmap */
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
s->s_fs_info = fsc;
|
|
|
|
fsc->sb = s;
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
s->s_op = &ceph_super_ops;
|
|
|
|
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 00:14:15 +02:00
|
|
|
fsc->sb = NULL;
|
2009-10-06 20:31:07 +02:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* share superblock if same fs AND options
|
|
|
|
*/
|
|
|
|
static int ceph_compare_super(struct super_block *sb, void *data)
|
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
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);
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
dout("ceph_compare_super %p\n", sb);
|
2010-04-07 00:14:15 +02:00
|
|
|
|
|
|
|
if (compare_mount_options(fsopt, opt, other)) {
|
|
|
|
dout("monitor(s)/mount options don't match\n");
|
|
|
|
return 0;
|
2009-10-06 20:31:07 +02:00
|
|
|
}
|
2010-04-07 00:14:15 +02:00
|
|
|
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) {
|
2009-10-06 20:31:07 +02:00
|
|
|
dout("flags differ\n");
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* construct our own bdi so we can control readahead, etc.
|
|
|
|
*/
|
2010-06-10 17:13:58 +02:00
|
|
|
static atomic_long_t bdi_seq = ATOMIC_LONG_INIT(0);
|
2010-05-05 01:39:35 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
static int ceph_register_bdi(struct super_block *sb,
|
|
|
|
struct ceph_fs_client *fsc)
|
2009-10-06 20:31:07 +02:00
|
|
|
{
|
|
|
|
int err;
|
|
|
|
|
2011-08-04 17:03:44 +02:00
|
|
|
/* set ra_pages based on rasize mount option? */
|
|
|
|
if (fsc->mount_options->rasize >= PAGE_CACHE_SIZE)
|
2010-04-07 00:14:15 +02:00
|
|
|
fsc->backing_dev_info.ra_pages =
|
2011-08-04 17:03:44 +02:00
|
|
|
(fsc->mount_options->rasize + PAGE_CACHE_SIZE - 1)
|
2009-10-06 20:31:07 +02:00
|
|
|
>> PAGE_SHIFT;
|
2011-07-22 20:12:28 +02:00
|
|
|
else
|
|
|
|
fsc->backing_dev_info.ra_pages =
|
|
|
|
default_backing_dev_info.ra_pages;
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
err = bdi_register(&fsc->backing_dev_info, NULL, "ceph-%d",
|
2010-05-05 01:39:35 +02:00
|
|
|
atomic_long_inc_return(&bdi_seq));
|
2010-05-05 01:14:46 +02:00
|
|
|
if (!err)
|
2010-04-07 00:14:15 +02:00
|
|
|
sb->s_bdi = &fsc->backing_dev_info;
|
2009-10-06 20:31:07 +02:00
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2010-07-26 14:17:55 +02:00
|
|
|
static struct dentry *ceph_mount(struct file_system_type *fs_type,
|
|
|
|
int flags, const char *dev_name, void *data)
|
2009-10-06 20:31:07 +02:00
|
|
|
{
|
|
|
|
struct super_block *sb;
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_fs_client *fsc;
|
2010-07-26 14:17:55 +02:00
|
|
|
struct dentry *res;
|
2009-10-06 20:31:07 +02:00
|
|
|
int err;
|
|
|
|
int (*compare_super)(struct super_block *, void *) = ceph_compare_super;
|
2009-11-04 20:40:05 +01:00
|
|
|
const char *path = NULL;
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_mount_options *fsopt = NULL;
|
|
|
|
struct ceph_options *opt = NULL;
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-07-26 14:17:55 +02:00
|
|
|
dout("ceph_mount\n");
|
2010-04-07 00:14:15 +02:00
|
|
|
err = parse_mount_options(&fsopt, &opt, flags, data, dev_name, &path);
|
2010-07-26 14:17:55 +02:00
|
|
|
if (err < 0) {
|
|
|
|
res = ERR_PTR(err);
|
2009-10-27 19:50:50 +01:00
|
|
|
goto out_final;
|
2010-07-26 14:17:55 +02:00
|
|
|
}
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
/* create client (which we may/may not use) */
|
2010-04-07 00:14:15 +02:00
|
|
|
fsc = create_fs_client(fsopt, opt);
|
|
|
|
if (IS_ERR(fsc)) {
|
2010-07-26 14:17:55 +02:00
|
|
|
res = ERR_CAST(fsc);
|
2011-08-22 21:49:41 +02:00
|
|
|
destroy_mount_options(fsopt);
|
|
|
|
ceph_destroy_options(opt);
|
2009-10-27 19:50:50 +01:00
|
|
|
goto out_final;
|
|
|
|
}
|
2009-10-06 20:31:07 +02:00
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
err = ceph_mdsc_init(fsc);
|
2010-07-26 14:17:55 +02:00
|
|
|
if (err < 0) {
|
|
|
|
res = ERR_PTR(err);
|
2010-04-07 00:14:15 +02:00
|
|
|
goto out;
|
2010-07-26 14:17:55 +02:00
|
|
|
}
|
2010-04-07 00:14:15 +02:00
|
|
|
|
|
|
|
if (ceph_test_opt(fsc->client, NOSHARE))
|
2009-10-06 20:31:07 +02:00
|
|
|
compare_super = NULL;
|
2010-04-07 00:14:15 +02:00
|
|
|
sb = sget(fs_type, compare_super, ceph_set_super, fsc);
|
2009-10-06 20:31:07 +02:00
|
|
|
if (IS_ERR(sb)) {
|
2010-07-26 14:17:55 +02:00
|
|
|
res = ERR_CAST(sb);
|
2009-10-06 20:31:07 +02:00
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
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);
|
2009-10-06 20:31:07 +02:00
|
|
|
} else {
|
2010-04-07 00:14:15 +02:00
|
|
|
dout("get_sb using new client %p\n", fsc);
|
|
|
|
err = ceph_register_bdi(sb, fsc);
|
2010-07-26 14:17:55 +02:00
|
|
|
if (err < 0) {
|
|
|
|
res = ERR_PTR(err);
|
2009-10-06 20:31:07 +02:00
|
|
|
goto out_splat;
|
2010-07-26 14:17:55 +02:00
|
|
|
}
|
2009-10-06 20:31:07 +02:00
|
|
|
}
|
|
|
|
|
2010-07-26 14:17:55 +02:00
|
|
|
res = ceph_real_mount(fsc, path);
|
|
|
|
if (IS_ERR(res))
|
2009-10-06 20:31:07 +02:00
|
|
|
goto out_splat;
|
2010-07-26 14:17:55 +02:00
|
|
|
dout("root %p inode %p ino %llx.%llx\n", res,
|
|
|
|
res->d_inode, ceph_vinop(res->d_inode));
|
|
|
|
return res;
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
out_splat:
|
2010-04-07 00:14:15 +02:00
|
|
|
ceph_mdsc_close_sessions(fsc->mdsc);
|
2010-03-22 00:22:29 +01:00
|
|
|
deactivate_locked_super(sb);
|
2009-10-06 20:31:07 +02:00
|
|
|
goto out_final;
|
|
|
|
|
|
|
|
out:
|
2010-04-07 00:14:15 +02:00
|
|
|
ceph_mdsc_destroy(fsc);
|
|
|
|
destroy_fs_client(fsc);
|
2009-10-06 20:31:07 +02:00
|
|
|
out_final:
|
2010-07-26 14:17:55 +02:00
|
|
|
dout("ceph_mount fail %ld\n", PTR_ERR(res));
|
|
|
|
return res;
|
2009-10-06 20:31:07 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static void ceph_kill_sb(struct super_block *s)
|
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
struct ceph_fs_client *fsc = ceph_sb_to_client(s);
|
2009-10-06 20:31:07 +02:00
|
|
|
dout("kill_sb %p\n", s);
|
2010-04-07 00:14:15 +02:00
|
|
|
ceph_mdsc_pre_umount(fsc->mdsc);
|
2009-10-06 20:31:07 +02:00
|
|
|
kill_anon_super(s); /* will call put_super after sb is r/o */
|
2010-04-07 00:14:15 +02:00
|
|
|
ceph_mdsc_destroy(fsc);
|
|
|
|
destroy_fs_client(fsc);
|
2009-10-06 20:31:07 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct file_system_type ceph_fs_type = {
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.name = "ceph",
|
2010-07-26 14:17:55 +02:00
|
|
|
.mount = ceph_mount,
|
2009-10-06 20:31:07 +02:00
|
|
|
.kill_sb = ceph_kill_sb,
|
|
|
|
.fs_flags = FS_RENAME_DOES_D_MOVE,
|
|
|
|
};
|
|
|
|
|
|
|
|
#define _STRINGIFY(x) #x
|
|
|
|
#define STRINGIFY(x) _STRINGIFY(x)
|
|
|
|
|
|
|
|
static int __init init_ceph(void)
|
|
|
|
{
|
2010-04-07 00:14:15 +02:00
|
|
|
int ret = init_caches();
|
2009-10-06 20:31:07 +02:00
|
|
|
if (ret)
|
2010-04-07 00:14:15 +02:00
|
|
|
goto out;
|
2009-10-06 20:31:07 +02:00
|
|
|
|
|
|
|
ret = register_filesystem(&ceph_fs_type);
|
|
|
|
if (ret)
|
|
|
|
goto out_icache;
|
|
|
|
|
2010-04-07 00:14:15 +02:00
|
|
|
pr_info("loaded (mds proto %d)\n", CEPH_MDSC_PROTOCOL);
|
|
|
|
|
2009-10-06 20:31:07 +02:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
out_icache:
|
|
|
|
destroy_caches();
|
|
|
|
out:
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __exit exit_ceph(void)
|
|
|
|
{
|
|
|
|
dout("exit_ceph\n");
|
|
|
|
unregister_filesystem(&ceph_fs_type);
|
|
|
|
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");
|