aboutsummaryrefslogtreecommitdiffstats
path: root/fs/ceph
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ceph')
-rw-r--r--fs/ceph/Makefile2
-rw-r--r--fs/ceph/addr.c63
-rw-r--r--fs/ceph/cache.c117
-rw-r--r--fs/ceph/caps.c128
-rw-r--r--fs/ceph/debugfs.c8
-rw-r--r--fs/ceph/dir.c205
-rw-r--r--fs/ceph/file.c327
-rw-r--r--fs/ceph/inode.c36
-rw-r--r--fs/ceph/ioctl.c13
-rw-r--r--fs/ceph/locks.c20
-rw-r--r--fs/ceph/mds_client.c87
-rw-r--r--fs/ceph/mds_client.h4
-rw-r--r--fs/ceph/quota.c361
-rw-r--r--fs/ceph/snap.c2
-rw-r--r--fs/ceph/super.c50
-rw-r--r--fs/ceph/super.h42
-rw-r--r--fs/ceph/xattr.c70
17 files changed, 1102 insertions, 433 deletions
diff --git a/fs/ceph/Makefile b/fs/ceph/Makefile
index 174f5709e508..a699e320393f 100644
--- a/fs/ceph/Makefile
+++ b/fs/ceph/Makefile
@@ -6,7 +6,7 @@
obj-$(CONFIG_CEPH_FS) += ceph.o
ceph-y := super.o inode.o dir.o file.o locks.o addr.o ioctl.o \
- export.o caps.o snap.o xattr.o \
+ export.o caps.o snap.o xattr.o quota.o \
mds_client.o mdsmap.o strings.o ceph_frag.o \
debugfs.o
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index b4336b42ce3b..5f7ad3d0df2e 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -15,6 +15,7 @@
#include "mds_client.h"
#include "cache.h"
#include <linux/ceph/osd_client.h>
+#include <linux/ceph/striper.h>
/*
* Ceph address space ops.
@@ -438,7 +439,7 @@ static int ceph_readpages(struct file *file, struct address_space *mapping,
{
struct inode *inode = file_inode(file);
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
- struct ceph_file_info *ci = file->private_data;
+ struct ceph_file_info *fi = file->private_data;
struct ceph_rw_context *rw_ctx;
int rc = 0;
int max = 0;
@@ -452,7 +453,7 @@ static int ceph_readpages(struct file *file, struct address_space *mapping,
if (rc == 0)
goto out;
- rw_ctx = ceph_find_rw_context(ci);
+ rw_ctx = ceph_find_rw_context(fi);
max = fsc->mount_options->rsize >> PAGE_SHIFT;
dout("readpages %p file %p ctx %p nr_pages %d max %d\n",
inode, file, rw_ctx, nr_pages, max);
@@ -800,7 +801,7 @@ static int ceph_writepages_start(struct address_space *mapping,
struct ceph_osd_request *req = NULL;
struct ceph_writeback_ctl ceph_wbc;
bool should_loop, range_whole = false;
- bool stop, done = false;
+ bool done = false;
dout("writepages_start %p (mode=%s)\n", inode,
wbc->sync_mode == WB_SYNC_NONE ? "NONE" :
@@ -856,7 +857,7 @@ retry:
* in that range can be associated with newer snapc.
* They are not writeable until we write all dirty pages
* associated with 'snapc' get written */
- if (index > 0 || wbc->sync_mode != WB_SYNC_NONE)
+ if (index > 0)
should_loop = true;
dout(" non-head snapc, range whole\n");
}
@@ -864,8 +865,7 @@ retry:
ceph_put_snap_context(last_snapc);
last_snapc = snapc;
- stop = false;
- while (!stop && index <= end) {
+ while (!done && index <= end) {
int num_ops = 0, op_idx;
unsigned i, pvec_pages, max_pages, locked_pages = 0;
struct page **pages = NULL, **data_pages;
@@ -898,16 +898,30 @@ get_more_pages:
unlock_page(page);
continue;
}
- if (strip_unit_end && (page->index > strip_unit_end)) {
- dout("end of strip unit %p\n", page);
+ /* only if matching snap context */
+ pgsnapc = page_snap_context(page);
+ if (pgsnapc != snapc) {
+ dout("page snapc %p %lld != oldest %p %lld\n",
+ pgsnapc, pgsnapc->seq, snapc, snapc->seq);
+ if (!should_loop &&
+ !ceph_wbc.head_snapc &&
+ wbc->sync_mode != WB_SYNC_NONE)
+ should_loop = true;
unlock_page(page);
- break;
+ continue;
}
if (page_offset(page) >= ceph_wbc.i_size) {
dout("%p page eof %llu\n",
page, ceph_wbc.i_size);
- /* not done if range_cyclic */
- stop = true;
+ if (ceph_wbc.size_stable ||
+ page_offset(page) >= i_size_read(inode))
+ mapping->a_ops->invalidatepage(page,
+ 0, PAGE_SIZE);
+ unlock_page(page);
+ continue;
+ }
+ if (strip_unit_end && (page->index > strip_unit_end)) {
+ dout("end of strip unit %p\n", page);
unlock_page(page);
break;
}
@@ -921,15 +935,6 @@ get_more_pages:
wait_on_page_writeback(page);
}
- /* only if matching snap context */
- pgsnapc = page_snap_context(page);
- if (pgsnapc != snapc) {
- dout("page snapc %p %lld != oldest %p %lld\n",
- pgsnapc, pgsnapc->seq, snapc, snapc->seq);
- unlock_page(page);
- continue;
- }
-
if (!clear_page_dirty_for_io(page)) {
dout("%p !clear_page_dirty_for_io\n", page);
unlock_page(page);
@@ -945,19 +950,15 @@ get_more_pages:
if (locked_pages == 0) {
u64 objnum;
u64 objoff;
+ u32 xlen;
/* prepare async write request */
offset = (u64)page_offset(page);
- len = wsize;
-
- rc = ceph_calc_file_object_mapping(&ci->i_layout,
- offset, len,
- &objnum, &objoff,
- &len);
- if (rc < 0) {
- unlock_page(page);
- break;
- }
+ ceph_calc_file_object_mapping(&ci->i_layout,
+ offset, wsize,
+ &objnum, &objoff,
+ &xlen);
+ len = xlen;
num_ops = 1;
strip_unit_end = page->index +
@@ -1146,7 +1147,7 @@ new_request:
* we tagged for writeback prior to entering this loop.
*/
if (wbc->nr_to_write <= 0 && wbc->sync_mode == WB_SYNC_NONE)
- done = stop = true;
+ done = true;
release_pvec_pages:
dout("pagevec_release on %d pages (%p)\n", (int)pvec.nr,
diff --git a/fs/ceph/cache.c b/fs/ceph/cache.c
index a3ab265d3215..bb524c880b1e 100644
--- a/fs/ceph/cache.c
+++ b/fs/ceph/cache.c
@@ -27,7 +27,6 @@
struct ceph_aux_inode {
u64 version;
struct timespec mtime;
- loff_t size;
};
struct fscache_netfs ceph_cache_netfs = {
@@ -41,37 +40,18 @@ static LIST_HEAD(ceph_fscache_list);
struct ceph_fscache_entry {
struct list_head list;
struct fscache_cookie *fscache;
- struct ceph_fsid fsid;
size_t uniq_len;
+ /* The following members must be last */
+ struct ceph_fsid fsid;
char uniquifier[0];
};
-static uint16_t ceph_fscache_session_get_key(const void *cookie_netfs_data,
- void *buffer, uint16_t maxbuf)
-{
- const struct ceph_fs_client* fsc = cookie_netfs_data;
- const char *fscache_uniq = fsc->mount_options->fscache_uniq;
- uint16_t fsid_len, uniq_len;
-
- fsid_len = sizeof(fsc->client->fsid);
- uniq_len = fscache_uniq ? strlen(fscache_uniq) : 0;
- if (fsid_len + uniq_len > maxbuf)
- return 0;
-
- memcpy(buffer, &fsc->client->fsid, fsid_len);
- if (uniq_len)
- memcpy(buffer + fsid_len, fscache_uniq, uniq_len);
-
- return fsid_len + uniq_len;
-}
-
static const struct fscache_cookie_def ceph_fscache_fsid_object_def = {
.name = "CEPH.fsid",
.type = FSCACHE_COOKIE_TYPE_INDEX,
- .get_key = ceph_fscache_session_get_key,
};
-int ceph_fscache_register(void)
+int __init ceph_fscache_register(void)
{
return fscache_register_netfs(&ceph_cache_netfs);
}
@@ -110,16 +90,19 @@ int ceph_fscache_register_fs(struct ceph_fs_client* fsc)
goto out_unlock;
}
+ memcpy(&ent->fsid, fsid, sizeof(*fsid));
+ if (uniq_len > 0) {
+ memcpy(&ent->uniquifier, fscache_uniq, uniq_len);
+ ent->uniq_len = uniq_len;
+ }
+
fsc->fscache = fscache_acquire_cookie(ceph_cache_netfs.primary_index,
&ceph_fscache_fsid_object_def,
- fsc, true);
+ &ent->fsid, sizeof(ent->fsid) + uniq_len,
+ NULL, 0,
+ fsc, 0, true);
if (fsc->fscache) {
- memcpy(&ent->fsid, fsid, sizeof(*fsid));
- if (uniq_len > 0) {
- memcpy(&ent->uniquifier, fscache_uniq, uniq_len);
- ent->uniq_len = uniq_len;
- }
ent->fscache = fsc->fscache;
list_add_tail(&ent->list, &ceph_fscache_list);
} else {
@@ -133,73 +116,32 @@ out_unlock:
return err;
}
-static uint16_t ceph_fscache_inode_get_key(const void *cookie_netfs_data,
- void *buffer, uint16_t maxbuf)
-{
- const struct ceph_inode_info* ci = cookie_netfs_data;
- uint16_t klen;
-
- /* use ceph virtual inode (id + snapshot) */
- klen = sizeof(ci->i_vino);
- if (klen > maxbuf)
- return 0;
-
- memcpy(buffer, &ci->i_vino, klen);
- return klen;
-}
-
-static uint16_t ceph_fscache_inode_get_aux(const void *cookie_netfs_data,
- void *buffer, uint16_t bufmax)
-{
- struct ceph_aux_inode aux;
- const struct ceph_inode_info* ci = cookie_netfs_data;
- const struct inode* inode = &ci->vfs_inode;
-
- memset(&aux, 0, sizeof(aux));
- aux.version = ci->i_version;
- aux.mtime = inode->i_mtime;
- aux.size = i_size_read(inode);
-
- memcpy(buffer, &aux, sizeof(aux));
-
- return sizeof(aux);
-}
-
-static void ceph_fscache_inode_get_attr(const void *cookie_netfs_data,
- uint64_t *size)
-{
- const struct ceph_inode_info* ci = cookie_netfs_data;
- *size = i_size_read(&ci->vfs_inode);
-}
-
static enum fscache_checkaux ceph_fscache_inode_check_aux(
- void *cookie_netfs_data, const void *data, uint16_t dlen)
+ void *cookie_netfs_data, const void *data, uint16_t dlen,
+ loff_t object_size)
{
struct ceph_aux_inode aux;
struct ceph_inode_info* ci = cookie_netfs_data;
struct inode* inode = &ci->vfs_inode;
- if (dlen != sizeof(aux))
+ if (dlen != sizeof(aux) ||
+ i_size_read(inode) != object_size)
return FSCACHE_CHECKAUX_OBSOLETE;
memset(&aux, 0, sizeof(aux));
aux.version = ci->i_version;
aux.mtime = inode->i_mtime;
- aux.size = i_size_read(inode);
if (memcmp(data, &aux, sizeof(aux)) != 0)
return FSCACHE_CHECKAUX_OBSOLETE;
- dout("ceph inode 0x%p cached okay", ci);
+ dout("ceph inode 0x%p cached okay\n", ci);
return FSCACHE_CHECKAUX_OKAY;
}
static const struct fscache_cookie_def ceph_fscache_inode_object_def = {
.name = "CEPH.inode",
.type = FSCACHE_COOKIE_TYPE_DATAFILE,
- .get_key = ceph_fscache_inode_get_key,
- .get_attr = ceph_fscache_inode_get_attr,
- .get_aux = ceph_fscache_inode_get_aux,
.check_aux = ceph_fscache_inode_check_aux,
};
@@ -207,6 +149,7 @@ void ceph_fscache_register_inode_cookie(struct inode *inode)
{
struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+ struct ceph_aux_inode aux;
/* No caching for filesystem */
if (!fsc->fscache)
@@ -218,9 +161,14 @@ void ceph_fscache_register_inode_cookie(struct inode *inode)
inode_lock_nested(inode, I_MUTEX_CHILD);
if (!ci->fscache) {
+ memset(&aux, 0, sizeof(aux));
+ aux.version = ci->i_version;
+ aux.mtime = inode->i_mtime;
ci->fscache = fscache_acquire_cookie(fsc->fscache,
- &ceph_fscache_inode_object_def,
- ci, false);
+ &ceph_fscache_inode_object_def,
+ &ci->i_vino, sizeof(ci->i_vino),
+ &aux, sizeof(aux),
+ ci, i_size_read(inode), false);
}
inode_unlock(inode);
}
@@ -235,7 +183,7 @@ void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci)
ci->fscache = NULL;
fscache_uncache_all_inode_pages(cookie, &ci->vfs_inode);
- fscache_relinquish_cookie(cookie, 0);
+ fscache_relinquish_cookie(cookie, &ci->i_vino, false);
}
static bool ceph_fscache_can_enable(void *data)
@@ -254,11 +202,11 @@ void ceph_fscache_file_set_cookie(struct inode *inode, struct file *filp)
if (inode_is_open_for_write(inode)) {
dout("fscache_file_set_cookie %p %p disabling cache\n",
inode, filp);
- fscache_disable_cookie(ci->fscache, false);
+ fscache_disable_cookie(ci->fscache, &ci->i_vino, false);
fscache_uncache_all_inode_pages(ci->fscache, inode);
} else {
- fscache_enable_cookie(ci->fscache, ceph_fscache_can_enable,
- inode);
+ fscache_enable_cookie(ci->fscache, &ci->i_vino, i_size_read(inode),
+ ceph_fscache_can_enable, inode);
if (fscache_cookie_enabled(ci->fscache)) {
dout("fscache_file_set_cookie %p %p enabling cache\n",
inode, filp);
@@ -351,7 +299,8 @@ void ceph_readpage_to_fscache(struct inode *inode, struct page *page)
if (!cache_valid(ci))
return;
- ret = fscache_write_page(ci->fscache, page, GFP_KERNEL);
+ ret = fscache_write_page(ci->fscache, page, i_size_read(inode),
+ GFP_KERNEL);
if (ret)
fscache_uncache_page(ci->fscache, page);
}
@@ -385,7 +334,7 @@ void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc)
WARN_ON_ONCE(!found);
mutex_unlock(&ceph_fscache_lock);
- __fscache_relinquish_cookie(fsc->fscache, 0);
+ __fscache_relinquish_cookie(fsc->fscache, NULL, false);
}
fsc->fscache = NULL;
}
@@ -402,7 +351,7 @@ void ceph_fscache_revalidate_cookie(struct ceph_inode_info *ci)
* truncate while the caller holds CEPH_CAP_FILE_RD */
mutex_lock(&ci->i_truncate_mutex);
if (!cache_valid(ci)) {
- if (fscache_check_consistency(ci->fscache))
+ if (fscache_check_consistency(ci->fscache, &ci->i_vino))
fscache_invalidate(ci->fscache);
spin_lock(&ci->i_ceph_lock);
ci->i_fscache_gen = ci->i_rdcache_gen;
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 0e5bd3e3344e..23dbfae16156 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -184,36 +184,54 @@ int ceph_reserve_caps(struct ceph_mds_client *mdsc,
mdsc->caps_avail_count);
spin_unlock(&mdsc->caps_list_lock);
- for (i = have; i < need; i++) {
-retry:
+ for (i = have; i < need; ) {
cap = kmem_cache_alloc(ceph_cap_cachep, GFP_NOFS);
- if (!cap) {
- if (!trimmed) {
- for (j = 0; j < mdsc->max_sessions; j++) {
- s = __ceph_lookup_mds_session(mdsc, j);
- if (!s)
- continue;
- mutex_unlock(&mdsc->mutex);
+ if (cap) {
+ list_add(&cap->caps_item, &newcaps);
+ alloc++;
+ i++;
+ continue;
+ }
- mutex_lock(&s->s_mutex);
- max_caps = s->s_nr_caps - (need - i);
- ceph_trim_caps(mdsc, s, max_caps);
- mutex_unlock(&s->s_mutex);
+ if (!trimmed) {
+ for (j = 0; j < mdsc->max_sessions; j++) {
+ s = __ceph_lookup_mds_session(mdsc, j);
+ if (!s)
+ continue;
+ mutex_unlock(&mdsc->mutex);
- ceph_put_mds_session(s);
- mutex_lock(&mdsc->mutex);
- }
- trimmed = true;
- goto retry;
- } else {
- pr_warn("reserve caps ctx=%p ENOMEM "
- "need=%d got=%d\n",
- ctx, need, have + alloc);
- goto out_nomem;
+ mutex_lock(&s->s_mutex);
+ max_caps = s->s_nr_caps - (need - i);
+ ceph_trim_caps(mdsc, s, max_caps);
+ mutex_unlock(&s->s_mutex);
+
+ ceph_put_mds_session(s);
+ mutex_lock(&mdsc->mutex);
}
+ trimmed = true;
+
+ spin_lock(&mdsc->caps_list_lock);
+ if (mdsc->caps_avail_count) {
+ int more_have;
+ if (mdsc->caps_avail_count >= need - i)
+ more_have = need - i;
+ else
+ more_have = mdsc->caps_avail_count;
+
+ i += more_have;
+ have += more_have;
+ mdsc->caps_avail_count -= more_have;
+ mdsc->caps_reserve_count += more_have;
+
+ }
+ spin_unlock(&mdsc->caps_list_lock);
+
+ continue;
}
- list_add(&cap->caps_item, &newcaps);
- alloc++;
+
+ pr_warn("reserve caps ctx=%p ENOMEM need=%d got=%d\n",
+ ctx, need, have + alloc);
+ goto out_nomem;
}
BUG_ON(have + alloc != need);
@@ -234,16 +252,28 @@ retry:
return 0;
out_nomem:
+
+ spin_lock(&mdsc->caps_list_lock);
+ mdsc->caps_avail_count += have;
+ mdsc->caps_reserve_count -= have;
+
while (!list_empty(&newcaps)) {
cap = list_first_entry(&newcaps,
struct ceph_cap, caps_item);
list_del(&cap->caps_item);
- kmem_cache_free(ceph_cap_cachep, cap);
+
+ /* Keep some preallocated caps around (ceph_min_count), to
+ * avoid lots of free/alloc churn. */
+ if (mdsc->caps_avail_count >=
+ mdsc->caps_reserve_count + mdsc->caps_min_count) {
+ kmem_cache_free(ceph_cap_cachep, cap);
+ } else {
+ mdsc->caps_avail_count++;
+ mdsc->caps_total_count++;
+ list_add(&cap->caps_item, &mdsc->caps_list);
+ }
}
- spin_lock(&mdsc->caps_list_lock);
- mdsc->caps_avail_count += have;
- mdsc->caps_reserve_count -= have;
BUG_ON(mdsc->caps_total_count != mdsc->caps_use_count +
mdsc->caps_reserve_count +
mdsc->caps_avail_count);
@@ -254,12 +284,26 @@ out_nomem:
int ceph_unreserve_caps(struct ceph_mds_client *mdsc,
struct ceph_cap_reservation *ctx)
{
+ int i;
+ struct ceph_cap *cap;
+
dout("unreserve caps ctx=%p count=%d\n", ctx, ctx->count);
if (ctx->count) {
spin_lock(&mdsc->caps_list_lock);
BUG_ON(mdsc->caps_reserve_count < ctx->count);
mdsc->caps_reserve_count -= ctx->count;
- mdsc->caps_avail_count += ctx->count;
+ if (mdsc->caps_avail_count >=
+ mdsc->caps_reserve_count + mdsc->caps_min_count) {
+ mdsc->caps_total_count -= ctx->count;
+ for (i = 0; i < ctx->count; i++) {
+ cap = list_first_entry(&mdsc->caps_list,
+ struct ceph_cap, caps_item);
+ list_del(&cap->caps_item);
+ kmem_cache_free(ceph_cap_cachep, cap);
+ }
+ } else {
+ mdsc->caps_avail_count += ctx->count;
+ }
ctx->count = 0;
dout("unreserve caps %d = %d used + %d resv + %d avail\n",
mdsc->caps_total_count, mdsc->caps_use_count,
@@ -285,7 +329,23 @@ struct ceph_cap *ceph_get_cap(struct ceph_mds_client *mdsc,
mdsc->caps_use_count++;
mdsc->caps_total_count++;
spin_unlock(&mdsc->caps_list_lock);
+ } else {
+ spin_lock(&mdsc->caps_list_lock);
+ if (mdsc->caps_avail_count) {
+ BUG_ON(list_empty(&mdsc->caps_list));
+
+ mdsc->caps_avail_count--;
+ mdsc->caps_use_count++;
+ cap = list_first_entry(&mdsc->caps_list,
+ struct ceph_cap, caps_item);
+ list_del(&cap->caps_item);
+
+ BUG_ON(mdsc->caps_total_count != mdsc->caps_use_count +
+ mdsc->caps_reserve_count + mdsc->caps_avail_count);
+ }
+ spin_unlock(&mdsc->caps_list_lock);
}
+
return cap;
}
@@ -341,6 +401,8 @@ void ceph_reservation_status(struct ceph_fs_client *fsc,
{
struct ceph_mds_client *mdsc = fsc->mdsc;
+ spin_lock(&mdsc->caps_list_lock);
+
if (total)
*total = mdsc->caps_total_count;
if (avail)
@@ -351,6 +413,8 @@ void ceph_reservation_status(struct ceph_fs_client *fsc,
*reserved = mdsc->caps_reserve_count;
if (min)
*min = mdsc->caps_min_count;
+
+ spin_unlock(&mdsc->caps_list_lock);
}
/*
@@ -639,9 +703,11 @@ void ceph_add_cap(struct inode *inode,
}
spin_lock(&realm->inodes_with_caps_lock);
- ci->i_snap_realm = realm;
list_add(&ci->i_snap_realm_item,
&realm->inodes_with_caps);
+ ci->i_snap_realm = realm;
+ if (realm->ino == ci->i_vino.ino)
+ realm->inode = inode;
spin_unlock(&realm->inodes_with_caps_lock);
if (oldrealm)
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index 644def813754..abdf98deeec4 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -260,7 +260,7 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
goto out;
fsc->debugfs_mdsmap = debugfs_create_file("mdsmap",
- 0600,
+ 0400,
fsc->client->debugfs_dir,
fsc,
&mdsmap_show_fops);
@@ -268,7 +268,7 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
goto out;
fsc->debugfs_mds_sessions = debugfs_create_file("mds_sessions",
- 0600,
+ 0400,
fsc->client->debugfs_dir,
fsc,
&mds_sessions_show_fops);
@@ -276,7 +276,7 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
goto out;
fsc->debugfs_mdsc = debugfs_create_file("mdsc",
- 0600,
+ 0400,
fsc->client->debugfs_dir,
fsc,
&mdsc_show_fops);
@@ -292,7 +292,7 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
goto out;
fsc->debugfs_dentry_lru = debugfs_create_file("dentry_lru",
- 0600,
+ 0400,
fsc->client->debugfs_dir,
fsc,
&dentry_lru_show_fops);
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index f1d9c6cc0491..1a78dd6f8bf2 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -2,7 +2,6 @@
#include <linux/ceph/ceph_debug.h>
#include <linux/spinlock.h>
-#include <linux/fs_struct.h>
#include <linux/namei.h>
#include <linux/slab.h>
#include <linux/sched.h>
@@ -102,18 +101,18 @@ static int fpos_cmp(loff_t l, loff_t r)
* regardless of what dir changes take place on the
* server.
*/
-static int note_last_dentry(struct ceph_file_info *fi, const char *name,
+static int note_last_dentry(struct ceph_dir_file_info *dfi, const char *name,
int len, unsigned next_offset)
{
char *buf = kmalloc(len+1, GFP_KERNEL);
if (!buf)
return -ENOMEM;
- kfree(fi->last_name);
- fi->last_name = buf;
- memcpy(fi->last_name, name, len);
- fi->last_name[len] = 0;
- fi->next_offset = next_offset;
- dout("note_last_dentry '%s'\n", fi->last_name);
+ kfree(dfi->last_name);
+ dfi->last_name = buf;
+ memcpy(dfi->last_name, name, len);
+ dfi->last_name[len] = 0;
+ dfi->next_offset = next_offset;
+ dout("note_last_dentry '%s'\n", dfi->last_name);
return 0;
}
@@ -175,7 +174,7 @@ __dcache_find_get_entry(struct dentry *parent, u64 idx,
static int __dcache_readdir(struct file *file, struct dir_context *ctx,
int shared_gen)
{
- struct ceph_file_info *fi = file->private_data;
+ struct ceph_dir_file_info *dfi = file->private_data;
struct dentry *parent = file->f_path.dentry;
struct inode *dir = d_inode(parent);
struct dentry *dentry, *last = NULL;
@@ -222,7 +221,7 @@ static int __dcache_readdir(struct file *file, struct dir_context *ctx,
bool emit_dentry = false;
dentry = __dcache_find_get_entry(parent, idx++, &cache_ctl);
if (!dentry) {
- fi->flags |= CEPH_F_ATEND;
+ dfi->file_info.flags |= CEPH_F_ATEND;
err = 0;
break;
}
@@ -273,33 +272,33 @@ out:
if (last) {
int ret;
di = ceph_dentry(last);
- ret = note_last_dentry(fi, last->d_name.name, last->d_name.len,
+ ret = note_last_dentry(dfi, last->d_name.name, last->d_name.len,
fpos_off(di->offset) + 1);
if (ret < 0)
err = ret;
dput(last);
/* last_name no longer match cache index */
- if (fi->readdir_cache_idx >= 0) {
- fi->readdir_cache_idx = -1;
- fi->dir_release_count = 0;
+ if (dfi->readdir_cache_idx >= 0) {
+ dfi->readdir_cache_idx = -1;
+ dfi->dir_release_count = 0;
}
}
return err;
}
-static bool need_send_readdir(struct ceph_file_info *fi, loff_t pos)
+static bool need_send_readdir(struct ceph_dir_file_info *dfi, loff_t pos)
{
- if (!fi->last_readdir)
+ if (!dfi->last_readdir)
return true;
if (is_hash_order(pos))
- return !ceph_frag_contains_value(fi->frag, fpos_hash(pos));
+ return !ceph_frag_contains_value(dfi->frag, fpos_hash(pos));
else
- return fi->frag != fpos_frag(pos);
+ return dfi->frag != fpos_frag(pos);
}
static int ceph_readdir(struct file *file, struct dir_context *ctx)
{
- struct ceph_file_info *fi = file->private_data;
+ struct ceph_dir_file_info *dfi = file->private_data;
struct inode *inode = file_inode(file);
struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
@@ -310,7 +309,7 @@ static int ceph_readdir(struct file *file, struct dir_context *ctx)
struct ceph_mds_reply_info_parsed *rinfo;
dout("readdir %p file %p pos %llx\n", inode, file, ctx->pos);
- if (fi->flags & CEPH_F_ATEND)
+ if (dfi->file_info.flags & CEPH_F_ATEND)
return 0;
/* always start with . and .. */
@@ -351,15 +350,15 @@ static int ceph_readdir(struct file *file, struct dir_context *ctx)
/* proceed with a normal readdir */
more:
/* do we have the correct frag content buffered? */
- if (need_send_readdir(fi, ctx->pos)) {
+ if (need_send_readdir(dfi, ctx->pos)) {
struct ceph_mds_request *req;
int op = ceph_snap(inode) == CEPH_SNAPDIR ?
CEPH_MDS_OP_LSSNAP : CEPH_MDS_OP_READDIR;
/* discard old result, if any */
- if (fi->last_readdir) {
- ceph_mdsc_put_request(fi->last_readdir);
- fi->last_readdir = NULL;
+ if (dfi->last_readdir) {
+ ceph_mdsc_put_request(dfi->last_readdir);
+ dfi->last_readdir = NULL;
}
if (is_hash_order(ctx->pos)) {
@@ -373,7 +372,7 @@ more:
}
dout("readdir fetching %llx.%llx frag %x offset '%s'\n",
- ceph_vinop(inode), frag, fi->last_name);
+ ceph_vinop(inode), frag, dfi->last_name);
req = ceph_mdsc_create_request(mdsc, op, USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
@@ -389,8 +388,8 @@ more:
__set_bit(CEPH_MDS_R_DIRECT_IS_HASH, &req->r_req_flags);
req->r_inode_drop = CEPH_CAP_FILE_EXCL;
}
- if (fi->last_name) {
- req->r_path2 = kstrdup(fi->last_name, GFP_KERNEL);
+ if (dfi->last_name) {
+ req->r_path2 = kstrdup(dfi->last_name, GFP_KERNEL);
if (!req->r_path2) {
ceph_mdsc_put_request(req);
return -ENOMEM;
@@ -400,10 +399,10 @@ more:
cpu_to_le32(fpos_hash(ctx->pos));
}
- req->r_dir_release_cnt = fi->dir_release_count;
- req->r_dir_ordered_cnt = fi->dir_ordered_count;
- req->r_readdir_cache_idx = fi->readdir_cache_idx;
- req->r_readdir_offset = fi->next_offset;
+ req->r_dir_release_cnt = dfi->dir_release_count;
+ req->r_dir_ordered_cnt = dfi->dir_ordered_count;
+ req->r_readdir_cache_idx = dfi->readdir_cache_idx;
+ req->r_readdir_offset = dfi->next_offset;
req->r_args.readdir.frag = cpu_to_le32(frag);
req->r_args.readdir.flags =
cpu_to_le16(CEPH_READDIR_REPLY_BITFLAGS);
@@ -427,35 +426,35 @@ more:
if (le32_to_cpu(rinfo->dir_dir->frag) != frag) {
frag = le32_to_cpu(rinfo->dir_dir->frag);
if (!rinfo->hash_order) {
- fi->next_offset = req->r_readdir_offset;
+ dfi->next_offset = req->r_readdir_offset;
/* adjust ctx->pos to beginning of frag */
ctx->pos = ceph_make_fpos(frag,
- fi->next_offset,
+ dfi->next_offset,
false);
}
}
- fi->frag = frag;
- fi->last_readdir = req;
+ dfi->frag = frag;
+ dfi->last_readdir = req;
if (test_bit(CEPH_MDS_R_DID_PREPOPULATE, &req->r_req_flags)) {
- fi->readdir_cache_idx = req->r_readdir_cache_idx;
- if (fi->readdir_cache_idx < 0) {
+ dfi->readdir_cache_idx = req->r_readdir_cache_idx;
+ if (dfi->readdir_cache_idx < 0) {
/* preclude from marking dir ordered */
- fi->dir_ordered_count = 0;
+ dfi->dir_ordered_count = 0;
} else if (ceph_frag_is_leftmost(frag) &&
- fi->next_offset == 2) {
+ dfi->next_offset == 2) {
/* note dir version at start of readdir so
* we can tell if any dentries get dropped */
- fi->dir_release_count = req->r_dir_release_cnt;
- fi->dir_ordered_count = req->r_dir_ordered_cnt;
+ dfi->dir_release_count = req->r_dir_release_cnt;
+ dfi->dir_ordered_count = req->r_dir_ordered_cnt;
}
} else {
- dout("readdir !did_prepopulate");
+ dout("readdir !did_prepopulate\n");
/* disable readdir cache */
- fi->readdir_cache_idx = -1;
+ dfi->readdir_cache_idx = -1;
/* preclude from marking dir complete */
- fi->dir_release_count = 0;
+ dfi->dir_release_count = 0;
}
/* note next offset and last dentry name */
@@ -464,19 +463,19 @@ more:
rinfo->dir_entries + (rinfo->dir_nr-1);
unsigned next_offset = req->r_reply_info.dir_end ?
2 : (fpos_off(rde->offset) + 1);
- err = note_last_dentry(fi, rde->name, rde->name_len,
+ err = note_last_dentry(dfi, rde->name, rde->name_len,
next_offset);
if (err)
return err;
} else if (req->r_reply_info.dir_end) {
- fi->next_offset = 2;
+ dfi->next_offset = 2;
/* keep last name */
}
}
- rinfo = &fi->last_readdir->r_reply_info;
+ rinfo = &dfi->last_readdir->r_reply_info;
dout("readdir frag %x num %d pos %llx chunk first %llx\n",
- fi->frag, rinfo->dir_nr, ctx->pos,
+ dfi->frag, rinfo->dir_nr, ctx->pos,
rinfo->dir_nr ? rinfo->dir_entries[0].offset : 0LL);
i = 0;
@@ -520,52 +519,55 @@ more:
ctx->pos++;
}
- ceph_mdsc_put_request(fi->last_readdir);
- fi->last_readdir = NULL;
+ ceph_mdsc_put_request(dfi->last_readdir);
+ dfi->last_readdir = NULL;
- if (fi->next_offset > 2) {
- frag = fi->frag;
+ if (dfi->next_offset > 2) {
+ frag = dfi->frag;
goto more;
}
/* more frags? */
- if (!ceph_frag_is_rightmost(fi->frag)) {
- frag = ceph_frag_next(fi->frag);
+ if (!ceph_frag_is_rightmost(dfi->frag)) {
+ frag = ceph_frag_next(dfi->frag);
if (is_hash_order(ctx->pos)) {
loff_t new_pos = ceph_make_fpos(ceph_frag_value(frag),
- fi->next_offset, true);
+ dfi->next_offset, true);
if (new_pos > ctx->pos)
ctx->pos = new_pos;
/* keep last_name */
} else {
- ctx->pos = ceph_make_fpos(frag, fi->next_offset, false);
- kfree(fi->last_name);
- fi->last_name = NULL;
+ ctx->pos = ceph_make_fpos(frag, dfi->next_offset,
+ false);
+ kfree(dfi->last_name);
+ dfi->last_name = NULL;
}
dout("readdir next frag is %x\n", frag);
goto more;
}
- fi->flags |= CEPH_F_ATEND;
+ dfi->file_info.flags |= CEPH_F_ATEND;
/*
* if dir_release_count still matches the dir, no dentries
* were released during the whole readdir, and we should have
* the complete dir contents in our cache.
*/
- if (atomic64_read(&ci->i_release_count) == fi->dir_release_count) {
+ if (atomic64_read(&ci->i_release_count) ==
+ dfi->dir_release_count) {
spin_lock(&ci->i_ceph_lock);
- if (fi->dir_ordered_count == atomic64_read(&ci->i_ordered_count)) {
+ if (dfi->dir_ordered_count ==
+ atomic64_read(&ci->i_ordered_count)) {
dout(" marking %p complete and ordered\n", inode);
/* use i_size to track number of entries in
* readdir cache */
- BUG_ON(fi->readdir_cache_idx < 0);
- i_size_write(inode, fi->readdir_cache_idx *
+ BUG_ON(dfi->readdir_cache_idx < 0);
+ i_size_write(inode, dfi->readdir_cache_idx *
sizeof(struct dentry*));
} else {
dout(" marking %p complete\n", inode);
}
- __ceph_dir_set_complete(ci, fi->dir_release_count,
- fi->dir_ordered_count);
+ __ceph_dir_set_complete(ci, dfi->dir_release_count,
+ dfi->dir_ordered_count);
spin_unlock(&ci->i_ceph_lock);
}
@@ -573,25 +575,25 @@ more:
return 0;
}
-static void reset_readdir(struct ceph_file_info *fi)
+static void reset_readdir(struct ceph_dir_file_info *dfi)
{
- if (fi->last_readdir) {
- ceph_mdsc_put_request(fi->last_readdir);
- fi->last_readdir = NULL;
+ if (dfi->last_readdir) {
+ ceph_mdsc_put_request(dfi->last_readdir);
+ dfi->last_readdir = NULL;
}
- kfree(fi->last_name);
- fi->last_name = NULL;
- fi->dir_release_count = 0;
- fi->readdir_cache_idx = -1;
- fi->next_offset = 2; /* compensate for . and .. */
- fi->flags &= ~CEPH_F_ATEND;
+ kfree(dfi->last_name);
+ dfi->last_name = NULL;
+ dfi->dir_release_count = 0;
+ dfi->readdir_cache_idx = -1;
+ dfi->next_offset = 2; /* compensate for . and .. */
+ dfi->file_info.flags &= ~CEPH_F_ATEND;
}
/*
* discard buffered readdir content on seekdir(0), or seek to new frag,
* or seek prior to current chunk
*/
-static bool need_reset_readdir(struct ceph_file_info *fi, loff_t new_pos)
+static bool need_reset_readdir(struct ceph_dir_file_info *dfi, loff_t new_pos)
{
struct ceph_mds_reply_info_parsed *rinfo;
loff_t chunk_offset;
@@ -600,10 +602,10 @@ static bool need_reset_readdir(struct ceph_file_info *fi, loff_t new_pos)
if (is_hash_order(new_pos)) {
/* no need to reset last_name for a forward seek when
* dentries are sotred in hash order */
- } else if (fi->frag != fpos_frag(new_pos)) {
+ } else if (dfi->frag != fpos_frag(new_pos)) {
return true;
}
- rinfo = fi->last_readdir ? &fi->last_readdir->r_reply_info : NULL;
+ rinfo = dfi->last_readdir ? &dfi->last_readdir->r_reply_info : NULL;
if (!rinfo || !rinfo->dir_nr)
return true;
chunk_offset = rinfo->dir_entries[0].offset;
@@ -613,7 +615,7 @@ static bool need_reset_readdir(struct ceph_file_info *fi, loff_t new_pos)
static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int whence)
{
- struct ceph_file_info *fi = file->private_data;
+ struct ceph_dir_file_info *dfi = file->private_data;
struct inode *inode = file->f_mapping->host;
loff_t retval;
@@ -631,20 +633,20 @@ static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int whence)
}
if (offset >= 0) {
- if (need_reset_readdir(fi, offset)) {
+ if (need_reset_readdir(dfi, offset)) {
dout("dir_llseek dropping %p content\n", file);
- reset_readdir(fi);
+ reset_readdir(dfi);
} else if (is_hash_order(offset) && offset > file->f_pos) {
/* for hash offset, we don't know if a forward seek
* is within same frag */
- fi->dir_release_count = 0;
- fi->readdir_cache_idx = -1;
+ dfi->dir_release_count = 0;
+ dfi->readdir_cache_idx = -1;
}
if (offset != file->f_pos) {
file->f_pos = offset;
file->f_version = 0;
- fi->flags &= ~CEPH_F_ATEND;
+ dfi->file_info.flags &= ~CEPH_F_ATEND;
}
retval = offset;
}
@@ -825,6 +827,9 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry,
if (ceph_snap(dir) != CEPH_NOSNAP)
return -EROFS;
+ if (ceph_quota_is_max_files_exceeded(dir))
+ return -EDQUOT;
+
err = ceph_pre_init_acls(dir, &mode, &acls);
if (err < 0)
return err;
@@ -878,6 +883,9 @@ static int ceph_symlink(struct inode *dir, struct dentry *dentry,
if (ceph_snap(dir) != CEPH_NOSNAP)
return -EROFS;
+ if (ceph_quota_is_max_files_exceeded(dir))
+ return -EDQUOT;
+
dout("symlink in dir %p dentry %p to '%s'\n", dir, dentry, dest);
req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SYMLINK, USE_AUTH_MDS);
if (IS_ERR(req)) {
@@ -927,6 +935,12 @@ static int ceph_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
goto out;
}
+ if (op == CEPH_MDS_OP_MKDIR &&
+ ceph_quota_is_max_files_exceeded(dir)) {
+ err = -EDQUOT;
+ goto out;
+ }
+
mode |= S_IFDIR;
err = ceph_pre_init_acls(dir, &mode, &acls);
if (err < 0)
@@ -1066,6 +1080,11 @@ static int ceph_rename(struct inode *old_dir, struct dentry *old_dentry,
else
return -EROFS;
}
+ /* don't allow cross-quota renames */
+ if ((old_dir != new_dir) &&
+ (!ceph_quota_is_same_realm(old_dir, new_dir)))
+ return -EXDEV;
+
dout("rename dir %p dentry %p to dir %p dentry %p\n",
old_dir, old_dentry, new_dir, new_dentry);
req = ceph_mdsc_create_request(mdsc, op, USE_AUTH_MDS);
@@ -1352,7 +1371,7 @@ static void ceph_d_prune(struct dentry *dentry)
static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
loff_t *ppos)
{
- struct ceph_file_info *cf = file->private_data;
+ struct ceph_dir_file_info *dfi = file->private_data;
struct inode *inode = file_inode(file);
struct ceph_inode_info *ci = ceph_inode(inode);
int left;
@@ -1361,12 +1380,12 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
if (!ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), DIRSTAT))
return -EISDIR;
- if (!cf->dir_info) {
- cf->dir_info = kmalloc(bufsize, GFP_KERNEL);
- if (!cf->dir_info)
+ if (!dfi->dir_info) {
+ dfi->dir_info = kmalloc(bufsize, GFP_KERNEL);
+ if (!dfi->dir_info)
return -ENOMEM;
- cf->dir_info_len =
- snprintf(cf->dir_info, bufsize,
+ dfi->dir_info_len =
+ snprintf(dfi->dir_info, bufsize,
"entries: %20lld\n"
" files: %20lld\n"
" subdirs: %20lld\n"
@@ -1386,10 +1405,10 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
(long)ci->i_rctime.tv_nsec);
}
- if (*ppos >= cf->dir_info_len)
+ if (*ppos >= dfi->dir_info_len)
return 0;
- size = min_t(unsigned, size, cf->dir_info_len-*ppos);
- left = copy_to_user(buf, cf->dir_info + *ppos, size);
+ size = min_t(unsigned, size, dfi->dir_info_len-*ppos);
+ left = copy_to_user(buf, dfi->dir_info + *ppos, size);
if (left == size)
return -EFAULT;
*ppos += (size - left);
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 6639926eed4e..cf0e45b10121 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -30,6 +30,8 @@ static __le32 ceph_flags_sys2wire(u32 flags)
break;
}
+ flags &= ~O_ACCMODE;
+
#define ceph_sys2wire(a) if (flags & a) { wire_flags |= CEPH_##a; flags &= ~a; }
ceph_sys2wire(O_CREAT);
@@ -41,7 +43,7 @@ static __le32 ceph_flags_sys2wire(u32 flags)
#undef ceph_sys2wire
if (flags)
- dout("unused open flags: %x", flags);
+ dout("unused open flags: %x\n", flags);
return cpu_to_le32(wire_flags);
}
@@ -68,69 +70,104 @@ static __le32 ceph_flags_sys2wire(u32 flags)
*/
/*
- * Calculate the length sum of direct io vectors that can
- * be combined into one page vector.
+ * How many pages to get in one call to iov_iter_get_pages(). This
+ * determines the size of the on-stack array used as a buffer.
*/
-static size_t dio_get_pagev_size(const struct iov_iter *it)
+#define ITER_GET_BVECS_PAGES 64
+
+static ssize_t __iter_get_bvecs(struct iov_iter *iter, size_t maxsize,
+ struct bio_vec *bvecs)
{
- const struct iovec *iov = it->iov;
- const struct iovec *iovend = iov + it->nr_segs;
- size_t size;
-
- size = iov->iov_len - it->iov_offset;
- /*
- * An iov can be page vectored when both the current tail
- * and the next base are page aligned.
- */
- while (PAGE_ALIGNED((iov->iov_base + iov->iov_len)) &&
- (++iov < iovend && PAGE_ALIGNED((iov->iov_base)))) {
- size += iov->iov_len;
- }
- dout("dio_get_pagevlen len = %zu\n", size);
- return size;
+ size_t size = 0;
+ int bvec_idx = 0;
+
+ if (maxsize > iov_iter_count(iter))
+ maxsize = iov_iter_count(iter);
+
+ while (size < maxsize) {
+ struct page *pages[ITER_GET_BVECS_PAGES];
+ ssize_t bytes;
+ size_t start;
+ int idx = 0;
+
+ bytes = iov_iter_get_pages(iter, pages, maxsize - size,
+ ITER_GET_BVECS_PAGES, &start);
+ if (bytes < 0)
+ return size ?: bytes;
+
+ iov_iter_advance(iter, bytes);
+ size += bytes;
+
+ for ( ; bytes; idx++, bvec_idx++) {
+ struct bio_vec bv = {
+ .bv_page = pages[idx],
+ .bv_len = min_t(int, bytes, PAGE_SIZE - start),
+ .bv_offset = start,
+ };
+
+ bvecs[bvec_idx] = bv;
+ bytes -= bv.bv_len;
+ start = 0;
+ }
+ }
+
+ return size;
}
/*
- * Allocate a page vector based on (@it, @nbytes).
- * The return value is the tuple describing a page vector,
- * that is (@pages, @page_align, @num_pages).
+ * iov_iter_get_pages() only considers one iov_iter segment, no matter
+ * what maxsize or maxpages are given. For ITER_BVEC that is a single
+ * page.
+ *
+ * Attempt to get up to @maxsize bytes worth of pages from @iter.
+ * Return the number of bytes in the created bio_vec array, or an error.
*/
-static struct page **
-dio_get_pages_alloc(const struct iov_iter *it, size_t nbytes,
- size_t *page_align, int *num_pages)
+static ssize_t iter_get_bvecs_alloc(struct iov_iter *iter, size_t maxsize,
+ struct bio_vec **bvecs, int *num_bvecs)
{
- struct iov_iter tmp_it = *it;
- size_t align;
- struct page **pages;
- int ret = 0, idx, npages;
+ struct bio_vec *bv;
+ size_t orig_count = iov_iter_count(iter);
+ ssize_t bytes;
+ int npages;
- align = (unsigned long)(it->iov->iov_base + it->iov_offset) &
- (PAGE_SIZE - 1);
- npages = calc_pages_for(align, nbytes);
- pages = kvmalloc(sizeof(*pages) * npages, GFP_KERNEL);
- if (!pages)
- return ERR_PTR(-ENOMEM);
+ iov_iter_truncate(iter, maxsize);
+ npages = iov_iter_npages(iter, INT_MAX);
+ iov_iter_reexpand(iter, orig_count);
- for (idx = 0; idx < npages; ) {
- size_t start;
- ret = iov_iter_get_pages(&tmp_it, pages + idx, nbytes,
- npages - idx, &start);
- if (ret < 0)
- goto fail;
+ /*
+ * __iter_get_bvecs() may populate only part of the array -- zero it
+ * out.
+ */
+ bv = kvmalloc_array(npages, sizeof(*bv), GFP_KERNEL | __GFP_ZERO);
+ if (!bv)
+ return -ENOMEM;
- iov_iter_advance(&tmp_it, ret);
- nbytes -= ret;
- idx += (ret + start + PAGE_SIZE - 1) / PAGE_SIZE;
+ bytes = __iter_get_bvecs(iter, maxsize, bv);
+ if (bytes < 0) {
+ /*
+ * No pages were pinned -- just free the array.
+ */
+ kvfree(bv);
+ return bytes;
}
- BUG_ON(nbytes != 0);
- *num_pages = npages;
- *page_align = align;
- dout("dio_get_pages_alloc: got %d pages align %zu\n", npages, align);
- return pages;
-fail:
- ceph_put_page_vector(pages, idx, false);
- return ERR_PTR(ret);
+ *bvecs = bv;
+ *num_bvecs = npages;
+ return bytes;
+}
+
+static void put_bvecs(struct bio_vec *bvecs, int num_bvecs, bool should_dirty)
+{
+ int i;
+
+ for (i = 0; i < num_bvecs; i++) {
+ if (bvecs[i].bv_page) {
+ if (should_dirty)
+ set_page_dirty_lock(bvecs[i].bv_page);
+ put_page(bvecs[i].bv_page);
+ }
+ }
+ kvfree(bvecs);
}
/*
@@ -159,13 +196,50 @@ out:
return req;
}
+static int ceph_init_file_info(struct inode *inode, struct file *file,
+ int fmode, bool isdir)
+{
+ struct ceph_file_info *fi;
+
+ dout("%s %p %p 0%o (%s)\n", __func__, inode, file,
+ inode->i_mode, isdir ? "dir" : "regular");
+ BUG_ON(inode->i_fop->release != ceph_release);
+
+ if (isdir) {
+ struct ceph_dir_file_info *dfi =
+ kmem_cache_zalloc(ceph_dir_file_cachep, GFP_KERNEL);
+ if (!dfi) {
+ ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */
+ return -ENOMEM;
+ }
+
+ file->private_data = dfi;
+ fi = &dfi->file_info;
+ dfi->next_offset = 2;
+ dfi->readdir_cache_idx = -1;
+ } else {
+ fi = kmem_cache_zalloc(ceph_file_cachep, GFP_KERNEL);
+ if (!fi) {
+ ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */
+ return -ENOMEM;
+ }
+
+ file->private_data = fi;
+ }
+
+ fi->fmode = fmode;
+ spin_lock_init(&fi->rw_contexts_lock);
+ INIT_LIST_HEAD(&fi->rw_contexts);
+
+ return 0;
+}
+
/*
* initialize private struct file data.
* if we fail, clean up by dropping fmode reference on the ceph_inode
*/
static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
{
- struct ceph_file_info *cf;
int ret = 0;
switch (inode->i_mode & S_IFMT) {
@@ -173,22 +247,10 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
ceph_fscache_register_inode_cookie(inode);
ceph_fscache_file_set_cookie(inode, file);
case S_IFDIR:
- dout("init_file %p %p 0%o (regular)\n", inode, file,
- inode->i_mode);
- cf = kmem_cache_zalloc(ceph_file_cachep, GFP_KERNEL);
- if (!cf) {
- ceph_put_fmode(ceph_inode(inode), fmode); /* clean up */
- return -ENOMEM;
- }
- cf->fmode = fmode;
-
- spin_lock_init(&cf->rw_contexts_lock);
- INIT_LIST_HEAD(&cf->rw_contexts);
-
- cf->next_offset = 2;
- cf->readdir_cache_idx = -1;
- file->private_data = cf;
- BUG_ON(inode->i_fop->release != ceph_release);
+ ret = ceph_init_file_info(inode, file, fmode,
+ S_ISDIR(inode->i_mode));
+ if (ret)
+ return ret;
break;
case S_IFLNK:
@@ -278,11 +340,11 @@ int ceph_open(struct inode *inode, struct file *file)
struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
struct ceph_mds_client *mdsc = fsc->mdsc;
struct ceph_mds_request *req;
- struct ceph_file_info *cf = file->private_data;
+ struct ceph_file_info *fi = file->private_data;
int err;
int flags, fmode, wanted;
- if (cf) {
+ if (fi) {
dout("open file %p is already opened\n", file);
return 0;
}
@@ -375,7 +437,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
struct ceph_mds_request *req;
struct dentry *dn;
struct ceph_acls_info acls = {};
- int mask;
+ int mask;
int err;
dout("atomic_open %p dentry %p '%pd' %s flags %d mode 0%o\n",
@@ -386,6 +448,8 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
return -ENAMETOOLONG;
if (flags & O_CREAT) {
+ if (ceph_quota_is_max_files_exceeded(dir))
+ return -EDQUOT;
err = ceph_pre_init_acls(dir, &mode, &acls);
if (err < 0)
return err;
@@ -460,16 +524,27 @@ out_acl:
int ceph_release(struct inode *inode, struct file *file)
{
struct ceph_inode_info *ci = ceph_inode(inode);
- struct ceph_file_info *cf = file->private_data;
- dout("release inode %p file %p\n", inode, file);
- ceph_put_fmode(ci, cf->fmode);
- if (cf->last_readdir)
- ceph_mdsc_put_request(cf->last_readdir);
- kfree(cf->last_name);
- kfree(cf->dir_info);
- WARN_ON(!list_empty(&cf->rw_contexts));
- kmem_cache_free(ceph_file_cachep, cf);
+ if (S_ISDIR(inode->i_mode)) {
+ struct ceph_dir_file_info *dfi = file->private_data;
+ dout("release inode %p dir file %p\n", inode, file);
+ WARN_ON(!list_empty(&dfi->file_info.rw_contexts));
+
+ ceph_put_fmode(ci, dfi->file_info.fmode);
+
+ if (dfi->last_readdir)
+ ceph_mdsc_put_request(dfi->last_readdir);
+ kfree(dfi->last_name);
+ kfree(dfi->dir_info);
+ kmem_cache_free(ceph_dir_file_cachep, dfi);
+ } else {
+ struct ceph_file_info *fi = file->private_data;
+ dout("release inode %p regular file %p\n", inode, file);
+ WARN_ON(!list_empty(&fi->rw_contexts));
+
+ ceph_put_fmode(ci, fi->fmode);
+ kmem_cache_free(ceph_file_cachep, fi);
+ }
/* wake up anyone waiting for caps on this inode */
wake_up_all(&ci->i_cap_wq);
@@ -640,7 +715,8 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *to,
struct ceph_aio_request {
struct kiocb *iocb;
size_t total_len;
- int write;
+ bool write;
+ bool should_dirty;
int error;
struct list_head osd_reqs;
unsigned num_reqs;
@@ -705,11 +781,12 @@ static void ceph_aio_complete_req(struct ceph_osd_request *req)
struct inode *inode = req->r_inode;
struct ceph_aio_request *aio_req = req->r_priv;
struct ceph_osd_data *osd_data = osd_req_op_extent_osd_data(req, 0);
- int num_pages = calc_pages_for((u64)osd_data->alignment,
- osd_data->length);
- dout("ceph_aio_complete_req %p rc %d bytes %llu\n",
- inode, rc, osd_data->length);
+ BUG_ON(osd_data->type != CEPH_OSD_DATA_TYPE_BVECS);
+ BUG_ON(!osd_data->num_bvecs);
+
+ dout("ceph_aio_complete_req %p rc %d bytes %u\n",
+ inode, rc, osd_data->bvec_pos.iter.bi_size);
if (rc == -EOLDSNAPC) {
struct ceph_aio_work *aio_work;
@@ -727,9 +804,10 @@ static void ceph_aio_complete_req(struct ceph_osd_request *req)
} else if (!aio_req->write) {
if (rc == -ENOENT)
rc = 0;
- if (rc >= 0 && osd_data->length > rc) {
- int zoff = osd_data->alignment + rc;
- int zlen = osd_data->length - rc;
+ if (rc >= 0 && osd_data->bvec_pos.iter.bi_size > rc) {
+ struct iov_iter i;
+ int zlen = osd_data->bvec_pos.iter.bi_size - rc;
+
/*
* If read is satisfied by single OSD request,
* it can pass EOF. Otherwise read is within
@@ -744,13 +822,16 @@ static void ceph_aio_complete_req(struct ceph_osd_request *req)
aio_req->total_len = rc + zlen;
}
- if (zlen > 0)
- ceph_zero_page_vector_range(zoff, zlen,
- osd_data->pages);
+ iov_iter_bvec(&i, ITER_BVEC, osd_data->bvec_pos.bvecs,
+ osd_data->num_bvecs,
+ osd_data->bvec_pos.iter.bi_size);
+ iov_iter_advance(&i, rc);
+ iov_iter_zero(zlen, &i);
}
}
- ceph_put_page_vector(osd_data->pages, num_pages, !aio_req->write);
+ put_bvecs(osd_data->bvec_pos.bvecs, osd_data->num_bvecs,
+ aio_req->should_dirty);
ceph_osdc_put_request(req);
if (rc < 0)
@@ -838,7 +919,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
struct ceph_vino vino;
struct ceph_osd_request *req;
- struct page **pages;
+ struct bio_vec *bvecs;
struct ceph_aio_request *aio_req = NULL;
int num_pages = 0;
int flags;
@@ -847,6 +928,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
size_t count = iov_iter_count(iter);
loff_t pos = iocb->ki_pos;
bool write = iov_iter_rw(iter) == WRITE;
+ bool should_dirty = !write && iter_is_iovec(iter);
if (write && ceph_snap(file_inode(file)) != CEPH_NOSNAP)
return -EROFS;
@@ -872,10 +954,14 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
}
while (iov_iter_count(iter) > 0) {
- u64 size = dio_get_pagev_size(iter);
- size_t start = 0;
+ u64 size = iov_iter_count(iter);
ssize_t len;
+ if (write)
+ size = min_t(u64, size, fsc->mount_options->wsize);
+ else
+ size = min_t(u64, size, fsc->mount_options->rsize);
+
vino = ceph_vino(inode);
req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
vino, pos, &size, 0,
@@ -891,18 +977,14 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
break;
}
- if (write)
- size = min_t(u64, size, fsc->mount_options->wsize);
- else
- size = min_t(u64, size, fsc->mount_options->rsize);
-
- len = size;
- pages = dio_get_pages_alloc(iter, len, &start, &num_pages);
- if (IS_ERR(pages)) {
+ len = iter_get_bvecs_alloc(iter, size, &bvecs, &num_pages);
+ if (len < 0) {
ceph_osdc_put_request(req);
- ret = PTR_ERR(pages);
+ ret = len;
break;
}
+ if (len != size)
+ osd_req_op_extent_update(req, 0, len);
/*
* To simplify error handling, allow AIO when IO within i_size
@@ -914,6 +996,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
if (aio_req) {
aio_req->iocb = iocb;
aio_req->write = write;
+ aio_req->should_dirty = should_dirty;
INIT_LIST_HEAD(&aio_req->osd_reqs);
if (write) {
aio_req->mtime = mtime;
@@ -934,8 +1017,7 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
req->r_mtime = mtime;
}
- osd_req_op_extent_osd_data_pages(req, 0, pages, len, start,
- false, false);
+ osd_req_op_extent_osd_data_bvecs(req, 0, bvecs, num_pages, len);
if (aio_req) {
aio_req->total_len += len;
@@ -948,7 +1030,6 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
list_add_tail(&req->r_unsafe_item, &aio_req->osd_reqs);
pos += len;
- iov_iter_advance(iter, len);
continue;
}
@@ -961,25 +1042,26 @@ ceph_direct_read_write(struct kiocb *iocb, struct iov_iter *iter,
if (ret == -ENOENT)
ret = 0;
if (ret >= 0 && ret < len && pos + ret < size) {
+ struct iov_iter i;
int zlen = min_t(size_t, len - ret,
size - pos - ret);
- ceph_zero_page_vector_range(start + ret, zlen,
- pages);
+
+ iov_iter_bvec(&i, ITER_BVEC, bvecs, num_pages,
+ len);
+ iov_iter_advance(&i, ret);
+ iov_iter_zero(zlen, &i);
ret += zlen;
}
if (ret >= 0)
len = ret;
}
- ceph_put_page_vector(pages, num_pages, !write);
-
+ put_bvecs(bvecs, num_pages, should_dirty);
ceph_osdc_put_request(req);
if (ret < 0)
break;
pos += len;
- iov_iter_advance(iter, len);
-
if (!write && pos >= size)
break;
@@ -1335,6 +1417,11 @@ retry_snap:
pos = iocb->ki_pos;
count = iov_iter_count(from);
+ if (ceph_quota_is_max_bytes_exceeded(inode, pos + count)) {
+ err = -EDQUOT;
+ goto out;
+ }
+
err = file_remove_privs(file);
if (err)
goto out;
@@ -1416,6 +1503,7 @@ retry_snap:
if (written >= 0) {
int dirty;
+
spin_lock(&ci->i_ceph_lock);
ci->i_inline_version = CEPH_INLINE_NONE;
dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR,
@@ -1423,6 +1511,8 @@ retry_snap:
spin_unlock(&ci->i_ceph_lock);
if (dirty)
__mark_inode_dirty(inode, dirty);
+ if (ceph_quota_is_max_bytes_approaching(inode, iocb->ki_pos))
+ ceph_check_caps(ci, CHECK_CAPS_NODELAY, NULL);
}
dout("aio_write %p %llx.%llx %llu~%u dropping cap refs on %s\n",
@@ -1665,6 +1755,12 @@ static long ceph_fallocate(struct file *file, int mode,
goto unlock;
}
+ if (!(mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE)) &&
+ ceph_quota_is_max_bytes_exceeded(inode, offset + length)) {
+ ret = -EDQUOT;
+ goto unlock;
+ }
+
if (ceph_osdmap_flag(osdc, CEPH_OSDMAP_FULL) &&
!(mode & FALLOC_FL_PUNCH_HOLE)) {
ret = -ENOSPC;
@@ -1713,6 +1809,9 @@ static long ceph_fallocate(struct file *file, int mode,
spin_unlock(&ci->i_ceph_lock);
if (dirty)
__mark_inode_dirty(inode, dirty);
+ if ((endoff > size) &&
+ ceph_quota_is_max_bytes_approaching(inode, endoff))
+ ceph_check_caps(ci, CHECK_CAPS_NODELAY, NULL);
}
ceph_put_cap_refs(ci, got);
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index c6ec5aa46100..ae056927080d 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -441,6 +441,9 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
atomic64_set(&ci->i_complete_seq[1], 0);
ci->i_symlink = NULL;
+ ci->i_max_bytes = 0;
+ ci->i_max_files = 0;
+
memset(&ci->i_dir_layout, 0, sizeof(ci->i_dir_layout));
RCU_INIT_POINTER(ci->i_layout.pool_ns, NULL);
@@ -536,6 +539,9 @@ void ceph_destroy_inode(struct inode *inode)
ceph_queue_caps_release(inode);
+ if (__ceph_has_any_quota(ci))
+ ceph_adjust_quota_realms_count(inode, false);
+
/*
* we may still have a snap_realm reference if there are stray
* caps in i_snap_caps.
@@ -548,6 +554,9 @@ void ceph_destroy_inode(struct inode *inode)
dout(" dropping residual ref to snap realm %p\n", realm);
spin_lock(&realm->inodes_with_caps_lock);
list_del_init(&ci->i_snap_realm_item);
+ ci->i_snap_realm = NULL;
+ if (realm->ino == ci->i_vino.ino)
+ realm->inode = NULL;
spin_unlock(&realm->inodes_with_caps_lock);
ceph_put_snap_realm(mdsc, realm);
}
@@ -660,13 +669,15 @@ void ceph_fill_file_time(struct inode *inode, int issued,
CEPH_CAP_FILE_BUFFER|
CEPH_CAP_AUTH_EXCL|
CEPH_CAP_XATTR_EXCL)) {
- if (timespec_compare(ctime, &inode->i_ctime) > 0) {
+ if (ci->i_version == 0 ||
+ timespec_compare(ctime, &inode->i_ctime) > 0) {
dout("ctime %ld.%09ld -> %ld.%09ld inc w/ cap\n",
inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec,
ctime->tv_sec, ctime->tv_nsec);
inode->i_ctime = *ctime;
}
- if (ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) > 0) {
+ if (ci->i_version == 0 ||
+ ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) > 0) {
/* the MDS did a utimes() */
dout("mtime %ld.%09ld -> %ld.%09ld "
"tw %d -> %d\n",
@@ -786,10 +797,11 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
new_issued = ~issued & le32_to_cpu(info->cap.caps);
/* update inode */
- ci->i_version = le64_to_cpu(info->version);
inode->i_rdev = le32_to_cpu(info->rdev);
inode->i_blkbits = fls(le32_to_cpu(info->layout.fl_stripe_unit)) - 1;
+ __ceph_update_quota(ci, iinfo->max_bytes, iinfo->max_files);
+
if ((new_version || (new_issued & CEPH_CAP_AUTH_SHARED)) &&
(issued & CEPH_CAP_AUTH_EXCL) == 0) {
inode->i_mode = le32_to_cpu(info->mode);
@@ -857,6 +869,9 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
xattr_blob = NULL;
}
+ /* finally update i_version */
+ ci->i_version = le64_to_cpu(info->version);
+
inode->i_mapping->a_ops = &ceph_aops;
switch (inode->i_mode & S_IFMT) {
@@ -1867,20 +1882,9 @@ retry:
* possibly truncate them.. so write AND block!
*/
if (ci->i_wrbuffer_ref_head < ci->i_wrbuffer_ref) {
- struct ceph_cap_snap *capsnap;
- to = ci->i_truncate_size;
- list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
- // MDS should have revoked Frw caps
- WARN_ON_ONCE(capsnap->writing);
- if (capsnap->dirty_pages && capsnap->size > to)
- to = capsnap->size;
- }
spin_unlock(&ci->i_ceph_lock);
dout("__do_pending_vmtruncate %p flushing snaps first\n",
inode);
-
- truncate_pagecache(inode, to);
-
filemap_write_and_wait_range(&inode->i_data, 0,
inode->i_sb->s_maxbytes);
goto retry;
@@ -2152,6 +2156,10 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
if (err != 0)
return err;
+ if ((attr->ia_valid & ATTR_SIZE) &&
+ ceph_quota_is_max_bytes_exceeded(inode, attr->ia_size))
+ return -EDQUOT;
+
err = __ceph_setattr(inode, attr);
if (err >= 0 && (attr->ia_valid & ATTR_MODE))
diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c
index 851aa69ec8f0..c90f03beb15d 100644
--- a/fs/ceph/ioctl.c
+++ b/fs/ceph/ioctl.c
@@ -5,7 +5,7 @@
#include "super.h"
#include "mds_client.h"
#include "ioctl.h"
-
+#include <linux/ceph/striper.h>
/*
* ioctls
@@ -185,7 +185,7 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg)
&ceph_sb_to_client(inode->i_sb)->client->osdc;
struct ceph_object_locator oloc;
CEPH_DEFINE_OID_ONSTACK(oid);
- u64 len = 1, olen;
+ u32 xlen;
u64 tmp;
struct ceph_pg pgid;
int r;
@@ -195,13 +195,8 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg)
return -EFAULT;
down_read(&osdc->lock);
- r = ceph_calc_file_object_mapping(&ci->i_layout, dl.file_offset, len,
- &dl.object_no, &dl.object_offset,
- &olen);
- if (r < 0) {
- up_read(&osdc->lock);
- return -EIO;
- }
+ ceph_calc_file_object_mapping(&ci->i_layout, dl.file_offset, 1,
+ &dl.object_no, &dl.object_offset, &xlen);
dl.file_offset -= dl.object_offset;
dl.object_size = ci->i_layout.object_size;
dl.block_size = ci->i_layout.stripe_unit;
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index 9e66f69ee8a5..9dae2ec7e1fa 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -95,7 +95,7 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct inode *inode,
owner = secure_addr(fl->fl_owner);
dout("ceph_lock_message: rule: %d, op: %d, owner: %llx, pid: %llu, "
- "start: %llu, length: %llu, wait: %d, type: %d", (int)lock_type,
+ "start: %llu, length: %llu, wait: %d, type: %d\n", (int)lock_type,
(int)operation, owner, (u64)fl->fl_pid, fl->fl_start, length,
wait, fl->fl_type);
@@ -132,7 +132,7 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct inode *inode,
}
ceph_mdsc_put_request(req);
dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, "
- "length: %llu, wait: %d, type: %d, err code %d", (int)lock_type,
+ "length: %llu, wait: %d, type: %d, err code %d\n", (int)lock_type,
(int)operation, (u64)fl->fl_pid, fl->fl_start,
length, wait, fl->fl_type, err);
return err;
@@ -226,7 +226,7 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
if (__mandatory_lock(file->f_mapping->host) && fl->fl_type != F_UNLCK)
return -ENOLCK;
- dout("ceph_lock, fl_owner: %p", fl->fl_owner);
+ dout("ceph_lock, fl_owner: %p\n", fl->fl_owner);
/* set wait bit as appropriate, then make command as Ceph expects it*/
if (IS_GETLK(cmd))
@@ -264,7 +264,7 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
err = ceph_lock_message(CEPH_LOCK_FCNTL, op, inode, lock_cmd, wait, fl);
if (!err) {
if (op == CEPH_MDS_OP_SETFILELOCK) {
- dout("mds locked, locking locally");
+ dout("mds locked, locking locally\n");
err = posix_lock_file(file, fl, NULL);
if (err) {
/* undo! This should only happen if
@@ -272,7 +272,7 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
* deadlock. */
ceph_lock_message(CEPH_LOCK_FCNTL, op, inode,
CEPH_LOCK_UNLOCK, 0, fl);
- dout("got %d on posix_lock_file, undid lock",
+ dout("got %d on posix_lock_file, undid lock\n",
err);
}
}
@@ -294,7 +294,7 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
if (fl->fl_type & LOCK_MAND)
return -EOPNOTSUPP;
- dout("ceph_flock, fl_file: %p", fl->fl_file);
+ dout("ceph_flock, fl_file: %p\n", fl->fl_file);
spin_lock(&ci->i_ceph_lock);
if (ci->i_ceph_flags & CEPH_I_ERROR_FILELOCK) {
@@ -329,7 +329,7 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
ceph_lock_message(CEPH_LOCK_FLOCK,
CEPH_MDS_OP_SETFILELOCK,
inode, CEPH_LOCK_UNLOCK, 0, fl);
- dout("got %d on locks_lock_file_wait, undid lock", err);
+ dout("got %d on locks_lock_file_wait, undid lock\n", err);
}
}
return err;
@@ -356,7 +356,7 @@ void ceph_count_locks(struct inode *inode, int *fcntl_count, int *flock_count)
++(*flock_count);
spin_unlock(&ctx->flc_lock);
}
- dout("counted %d flock locks and %d fcntl locks",
+ dout("counted %d flock locks and %d fcntl locks\n",
*flock_count, *fcntl_count);
}
@@ -384,7 +384,7 @@ static int lock_to_ceph_filelock(struct file_lock *lock,
cephlock->type = CEPH_LOCK_UNLOCK;
break;
default:
- dout("Have unknown lock type %d", lock->fl_type);
+ dout("Have unknown lock type %d\n", lock->fl_type);
err = -EINVAL;
}
@@ -407,7 +407,7 @@ int ceph_encode_locks_to_buffer(struct inode *inode,
int seen_flock = 0;
int l = 0;
- dout("encoding %d flock and %d fcntl locks", num_flock_locks,
+ dout("encoding %d flock and %d fcntl locks\n", num_flock_locks,
num_fcntl_locks);
if (!ctx)
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 2e8f90f96540..5ece2e6ad154 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -100,6 +100,26 @@ static int parse_reply_info_in(void **p, void *end,
} else
info->inline_version = CEPH_INLINE_NONE;
+ if (features & CEPH_FEATURE_MDS_QUOTA) {
+ u8 struct_v, struct_compat;
+ u32 struct_len;
+
+ /*
+ * both struct_v and struct_compat are expected to be >= 1
+ */
+ ceph_decode_8_safe(p, end, struct_v, bad);
+ ceph_decode_8_safe(p, end, struct_compat, bad);
+ if (!struct_v || !struct_compat)
+ goto bad;
+ ceph_decode_32_safe(p, end, struct_len, bad);
+ ceph_decode_need(p, end, struct_len, bad);
+ ceph_decode_64_safe(p, end, info->max_bytes, bad);
+ ceph_decode_64_safe(p, end, info->max_files, bad);
+ } else {
+ info->max_bytes = 0;
+ info->max_files = 0;
+ }
+
info->pool_ns_len = 0;
info->pool_ns_data = NULL;
if (features & CEPH_FEATURE_FS_FILE_LAYOUT_V2) {
@@ -384,7 +404,7 @@ static struct ceph_mds_session *get_session(struct ceph_mds_session *s)
refcount_read(&s->s_ref)-1, refcount_read(&s->s_ref));
return s;
} else {
- dout("mdsc get_session %p 0 -- FAIL", s);
+ dout("mdsc get_session %p 0 -- FAIL\n", s);
return NULL;
}
}
@@ -419,9 +439,10 @@ struct ceph_mds_session *__ceph_lookup_mds_session(struct ceph_mds_client *mdsc,
static bool __have_session(struct ceph_mds_client *mdsc, int mds)
{
- if (mds >= mdsc->max_sessions)
+ if (mds >= mdsc->max_sessions || !mdsc->sessions[mds])
return false;
- return mdsc->sessions[mds];
+ else
+ return true;
}
static int __verify_registered_session(struct ceph_mds_client *mdsc,
@@ -448,6 +469,25 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
s = kzalloc(sizeof(*s), GFP_NOFS);
if (!s)
return ERR_PTR(-ENOMEM);
+
+ if (mds >= mdsc->max_sessions) {
+ int newmax = 1 << get_count_order(mds + 1);
+ struct ceph_mds_session **sa;
+
+ dout("%s: realloc to %d\n", __func__, newmax);
+ sa = kcalloc(newmax, sizeof(void *), GFP_NOFS);
+ if (!sa)
+ goto fail_realloc;
+ if (mdsc->sessions) {
+ memcpy(sa, mdsc->sessions,
+ mdsc->max_sessions * sizeof(void *));
+ kfree(mdsc->sessions);
+ }
+ mdsc->sessions = sa;
+ mdsc->max_sessions = newmax;
+ }
+
+ dout("%s: mds%d\n", __func__, mds);
s->s_mdsc = mdsc;
s->s_mds = mds;
s->s_state = CEPH_MDS_SESSION_NEW;
@@ -476,23 +516,6 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
INIT_LIST_HEAD(&s->s_cap_releases);
INIT_LIST_HEAD(&s->s_cap_flushing);
- dout("register_session mds%d\n", mds);
- if (mds >= mdsc->max_sessions) {
- int newmax = 1 << get_count_order(mds+1);
- struct ceph_mds_session **sa;
-
- dout("register_session realloc to %d\n", newmax);
- sa = kcalloc(newmax, sizeof(void *), GFP_NOFS);
- if (!sa)
- goto fail_realloc;
- if (mdsc->sessions) {
- memcpy(sa, mdsc->sessions,
- mdsc->max_sessions * sizeof(void *));
- kfree(mdsc->sessions);
- }
- mdsc->sessions = sa;
- mdsc->max_sessions = newmax;
- }
mdsc->sessions[mds] = s;
atomic_inc(&mdsc->num_sessions);
refcount_inc(&s->s_ref); /* one ref to sessions[], one to caller */
@@ -2531,10 +2554,10 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
* Otherwise we just have to return an ESTALE
*/
if (result == -ESTALE) {
- dout("got ESTALE on request %llu", req->r_tid);
+ dout("got ESTALE on request %llu\n", req->r_tid);
req->r_resend_mds = -1;
if (req->r_direct_mode != USE_AUTH_MDS) {
- dout("not using auth, setting for that now");
+ dout("not using auth, setting for that now\n");
req->r_direct_mode = USE_AUTH_MDS;
__do_request(mdsc, req);
mutex_unlock(&mdsc->mutex);
@@ -2542,13 +2565,13 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
} else {
int mds = __choose_mds(mdsc, req);
if (mds >= 0 && mds != req->r_session->s_mds) {
- dout("but auth changed, so resending");
+ dout("but auth changed, so resending\n");
__do_request(mdsc, req);
mutex_unlock(&mdsc->mutex);
goto out;
}
}
- dout("have to return ESTALE on request %llu", req->r_tid);
+ dout("have to return ESTALE on request %llu\n", req->r_tid);
}
@@ -3470,13 +3493,12 @@ void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session,
}
/*
- * drop all leases (and dentry refs) in preparation for umount
+ * lock unlock sessions, to wait ongoing session activities
*/
-static void drop_leases(struct ceph_mds_client *mdsc)
+static void lock_unlock_sessions(struct ceph_mds_client *mdsc)
{
int i;
- dout("drop_leases\n");
mutex_lock(&mdsc->mutex);
for (i = 0; i < mdsc->max_sessions; i++) {
struct ceph_mds_session *s = __ceph_lookup_mds_session(mdsc, i);
@@ -3572,7 +3594,6 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc)
if (!mdsc)
return -ENOMEM;
mdsc->fsc = fsc;
- fsc->mdsc = mdsc;
mutex_init(&mdsc->mutex);
mdsc->mdsmap = kzalloc(sizeof(*mdsc->mdsmap), GFP_NOFS);
if (!mdsc->mdsmap) {
@@ -3580,6 +3601,7 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc)
return -ENOMEM;
}
+ fsc->mdsc = mdsc;
init_completion(&mdsc->safe_umount_waiters);
init_waitqueue_head(&mdsc->session_close_wq);
INIT_LIST_HEAD(&mdsc->waiting_for_map);
@@ -3587,6 +3609,7 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc)
atomic_set(&mdsc->num_sessions, 0);
mdsc->max_sessions = 0;
mdsc->stopping = 0;
+ atomic64_set(&mdsc->quotarealms_count, 0);
mdsc->last_snap_seq = 0;
init_rwsem(&mdsc->snap_rwsem);
mdsc->snap_realms = RB_ROOT;
@@ -3660,7 +3683,7 @@ void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc)
dout("pre_umount\n");
mdsc->stopping = 1;
- drop_leases(mdsc);
+ lock_unlock_sessions(mdsc);
ceph_flush_dirty_caps(mdsc);
wait_requests(mdsc);
@@ -3858,6 +3881,9 @@ void ceph_mdsc_destroy(struct ceph_fs_client *fsc)
struct ceph_mds_client *mdsc = fsc->mdsc;
dout("mdsc_destroy %p\n", mdsc);
+ if (!mdsc)
+ return;
+
/* flush out any connection work with references to us */
ceph_msgr_flush();
@@ -4077,6 +4103,9 @@ static void dispatch(struct ceph_connection *con, struct ceph_msg *msg)
case CEPH_MSG_CLIENT_LEASE:
handle_lease(mdsc, s, msg);
break;
+ case CEPH_MSG_CLIENT_QUOTA:
+ ceph_handle_quota(mdsc, s, msg);
+ break;
default:
pr_err("received unknown message type %d %s\n", type,
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index 71e3b783ee6f..2ec3b5b35067 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -49,6 +49,8 @@ struct ceph_mds_reply_info_in {
char *inline_data;
u32 pool_ns_len;
char *pool_ns_data;
+ u64 max_bytes;
+ u64 max_files;
};
struct ceph_mds_reply_dir_entry {
@@ -312,6 +314,8 @@ struct ceph_mds_client {
int max_sessions; /* len of s_mds_sessions */
int stopping; /* true if shutting down */
+ atomic64_t quotarealms_count; /* # realms with quota */
+
/*
* snap_rwsem will cover cap linkage into snaprealms, and
* realm snap contexts. (later, we can do per-realm snap
diff --git a/fs/ceph/quota.c b/fs/ceph/quota.c
new file mode 100644
index 000000000000..242bfa5c0539
--- /dev/null
+++ b/fs/ceph/quota.c
@@ -0,0 +1,361 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * quota.c - CephFS quota
+ *
+ * Copyright (C) 2017-2018 SUSE
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/statfs.h>
+
+#include "super.h"
+#include "mds_client.h"
+
+void ceph_adjust_quota_realms_count(struct inode *inode, bool inc)
+{
+ struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
+ if (inc)
+ atomic64_inc(&mdsc->quotarealms_count);
+ else
+ atomic64_dec(&mdsc->quotarealms_count);
+}
+
+static inline bool ceph_has_realms_with_quotas(struct inode *inode)
+{
+ struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
+ return atomic64_read(&mdsc->quotarealms_count) > 0;
+}
+
+void ceph_handle_quota(struct ceph_mds_client *mdsc,
+ struct ceph_mds_session *session,
+ struct ceph_msg *msg)
+{
+ struct super_block *sb = mdsc->fsc->sb;
+ struct ceph_mds_quota *h = msg->front.iov_base;
+ struct ceph_vino vino;
+ struct inode *inode;
+ struct ceph_inode_info *ci;
+
+ if (msg->front.iov_len != sizeof(*h)) {
+ pr_err("%s corrupt message mds%d len %d\n", __func__,
+ session->s_mds, (int)msg->front.iov_len);
+ ceph_msg_dump(msg);
+ return;
+ }
+
+ /* increment msg sequence number */
+ mutex_lock(&session->s_mutex);
+ session->s_seq++;
+ mutex_unlock(&session->s_mutex);
+
+ /* lookup inode */
+ vino.ino = le64_to_cpu(h->ino);
+ vino.snap = CEPH_NOSNAP;
+ inode = ceph_find_inode(sb, vino);
+ if (!inode) {
+ pr_warn("Failed to find inode %llu\n", vino.ino);
+ return;
+ }
+ ci = ceph_inode(inode);
+
+ spin_lock(&ci->i_ceph_lock);
+ ci->i_rbytes = le64_to_cpu(h->rbytes);
+ ci->i_rfiles = le64_to_cpu(h->rfiles);
+ ci->i_rsubdirs = le64_to_cpu(h->rsubdirs);
+ __ceph_update_quota(ci, le64_to_cpu(h->max_bytes),
+ le64_to_cpu(h->max_files));
+ spin_unlock(&ci->i_ceph_lock);
+
+ iput(inode);
+}
+
+/*
+ * This function walks through the snaprealm for an inode and returns the
+ * ceph_snap_realm for the first snaprealm that has quotas set (either max_files
+ * or max_bytes). If the root is reached, return the root ceph_snap_realm
+ * instead.
+ *
+ * Note that the caller is responsible for calling ceph_put_snap_realm() on the
+ * returned realm.
+ */
+static struct ceph_snap_realm *get_quota_realm(struct ceph_mds_client *mdsc,
+ struct inode *inode)
+{
+ struct ceph_inode_info *ci = NULL;
+ struct ceph_snap_realm *realm, *next;
+ struct inode *in;
+ bool has_quota;
+
+ if (ceph_snap(inode) != CEPH_NOSNAP)
+ return NULL;
+
+ realm = ceph_inode(inode)->i_snap_realm;
+ if (realm)
+ ceph_get_snap_realm(mdsc, realm);
+ else
+ pr_err_ratelimited("get_quota_realm: ino (%llx.%llx) "
+ "null i_snap_realm\n", ceph_vinop(inode));
+ while (realm) {
+ spin_lock(&realm->inodes_with_caps_lock);
+ in = realm->inode ? igrab(realm->inode) : NULL;
+ spin_unlock(&realm->inodes_with_caps_lock);
+ if (!in)
+ break;
+
+ ci = ceph_inode(in);
+ has_quota = __ceph_has_any_quota(ci);
+ iput(in);
+
+ next = realm->parent;
+ if (has_quota || !next)
+ return realm;
+
+ ceph_get_snap_realm(mdsc, next);
+ ceph_put_snap_realm(mdsc, realm);
+ realm = next;
+ }
+ if (realm)
+ ceph_put_snap_realm(mdsc, realm);
+
+ return NULL;
+}
+
+bool ceph_quota_is_same_realm(struct inode *old, struct inode *new)
+{
+ struct ceph_mds_client *mdsc = ceph_inode_to_client(old)->mdsc;
+ struct ceph_snap_realm *old_realm, *new_realm;
+ bool is_same;
+
+ down_read(&mdsc->snap_rwsem);
+ old_realm = get_quota_realm(mdsc, old);
+ new_realm = get_quota_realm(mdsc, new);
+ is_same = (old_realm == new_realm);
+ up_read(&mdsc->snap_rwsem);
+
+ if (old_realm)
+ ceph_put_snap_realm(mdsc, old_realm);
+ if (new_realm)
+ ceph_put_snap_realm(mdsc, new_realm);
+
+ return is_same;
+}
+
+enum quota_check_op {
+ QUOTA_CHECK_MAX_FILES_OP, /* check quota max_files limit */
+ QUOTA_CHECK_MAX_BYTES_OP, /* check quota max_files limit */
+ QUOTA_CHECK_MAX_BYTES_APPROACHING_OP /* check if quota max_files
+ limit is approaching */
+};
+
+/*
+ * check_quota_exceeded() will walk up the snaprealm hierarchy and, for each
+ * realm, it will execute quota check operation defined by the 'op' parameter.
+ * The snaprealm walk is interrupted if the quota check detects that the quota
+ * is exceeded or if the root inode is reached.
+ */
+static bool check_quota_exceeded(struct inode *inode, enum quota_check_op op,
+ loff_t delta)
+{
+ struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
+ struct ceph_inode_info *ci;
+ struct ceph_snap_realm *realm, *next;
+ struct inode *in;
+ u64 max, rvalue;
+ bool exceeded = false;
+
+ if (ceph_snap(inode) != CEPH_NOSNAP)
+ return false;
+
+ down_read(&mdsc->snap_rwsem);
+ realm = ceph_inode(inode)->i_snap_realm;
+ if (realm)
+ ceph_get_snap_realm(mdsc, realm);
+ else
+ pr_err_ratelimited("check_quota_exceeded: ino (%llx.%llx) "
+ "null i_snap_realm\n", ceph_vinop(inode));
+ while (realm) {
+ spin_lock(&realm->inodes_with_caps_lock);
+ in = realm->inode ? igrab(realm->inode) : NULL;
+ spin_unlock(&realm->inodes_with_caps_lock);
+ if (!in)
+ break;
+
+ ci = ceph_inode(in);
+ spin_lock(&ci->i_ceph_lock);
+ if (op == QUOTA_CHECK_MAX_FILES_OP) {
+ max = ci->i_max_files;
+ rvalue = ci->i_rfiles + ci->i_rsubdirs;
+ } else {
+ max = ci->i_max_bytes;
+ rvalue = ci->i_rbytes;
+ }
+ spin_unlock(&ci->i_ceph_lock);
+ switch (op) {
+ case QUOTA_CHECK_MAX_FILES_OP:
+ exceeded = (max && (rvalue >= max));
+ break;
+ case QUOTA_CHECK_MAX_BYTES_OP:
+ exceeded = (max && (rvalue + delta > max));
+ break;
+ case QUOTA_CHECK_MAX_BYTES_APPROACHING_OP:
+ if (max) {
+ if (rvalue >= max)
+ exceeded = true;
+ else {
+ /*
+ * when we're writing more that 1/16th
+ * of the available space
+ */
+ exceeded =
+ (((max - rvalue) >> 4) < delta);
+ }
+ }
+ break;
+ default:
+ /* Shouldn't happen */
+ pr_warn("Invalid quota check op (%d)\n", op);
+ exceeded = true; /* Just break the loop */
+ }
+ iput(in);
+
+ next = realm->parent;
+ if (exceeded || !next)
+ break;
+ ceph_get_snap_realm(mdsc, next);
+ ceph_put_snap_realm(mdsc, realm);
+ realm = next;
+ }
+ ceph_put_snap_realm(mdsc, realm);
+ up_read(&mdsc->snap_rwsem);
+
+ return exceeded;
+}
+
+/*
+ * ceph_quota_is_max_files_exceeded - check if we can create a new file
+ * @inode: directory where a new file is being created
+ *
+ * This functions returns true is max_files quota allows a new file to be
+ * created. It is necessary to walk through the snaprealm hierarchy (until the
+ * FS root) to check all realms with quotas set.
+ */
+bool ceph_quota_is_max_files_exceeded(struct inode *inode)
+{
+ if (!ceph_has_realms_with_quotas(inode))
+ return false;
+
+ WARN_ON(!S_ISDIR(inode->i_mode));
+
+ return check_quota_exceeded(inode, QUOTA_CHECK_MAX_FILES_OP, 0);
+}
+
+/*
+ * ceph_quota_is_max_bytes_exceeded - check if we can write to a file
+ * @inode: inode being written
+ * @newsize: new size if write succeeds
+ *
+ * This functions returns true is max_bytes quota allows a file size to reach
+ * @newsize; it returns false otherwise.
+ */
+bool ceph_quota_is_max_bytes_exceeded(struct inode *inode, loff_t newsize)
+{
+ loff_t size = i_size_read(inode);
+
+ if (!ceph_has_realms_with_quotas(inode))
+ return false;
+
+ /* return immediately if we're decreasing file size */
+ if (newsize <= size)
+ return false;
+
+ return check_quota_exceeded(inode, QUOTA_CHECK_MAX_BYTES_OP, (newsize - size));
+}
+
+/*
+ * ceph_quota_is_max_bytes_approaching - check if we're reaching max_bytes
+ * @inode: inode being written
+ * @newsize: new size if write succeeds
+ *
+ * This function returns true if the new file size @newsize will be consuming
+ * more than 1/16th of the available quota space; it returns false otherwise.
+ */
+bool ceph_quota_is_max_bytes_approaching(struct inode *inode, loff_t newsize)
+{
+ loff_t size = ceph_inode(inode)->i_reported_size;
+
+ if (!ceph_has_realms_with_quotas(inode))
+ return false;
+
+ /* return immediately if we're decreasing file size */
+ if (newsize <= size)
+ return false;
+
+ return check_quota_exceeded(inode, QUOTA_CHECK_MAX_BYTES_APPROACHING_OP,
+ (newsize - size));
+}
+
+/*
+ * ceph_quota_update_statfs - if root has quota update statfs with quota status
+ * @fsc: filesystem client instance
+ * @buf: statfs to update
+ *
+ * If the mounted filesystem root has max_bytes quota set, update the filesystem
+ * statistics with the quota status.
+ *
+ * This function returns true if the stats have been updated, false otherwise.
+ */
+bool ceph_quota_update_statfs(struct ceph_fs_client *fsc, struct kstatfs *buf)
+{
+ struct ceph_mds_client *mdsc = fsc->mdsc;
+ struct ceph_inode_info *ci;
+ struct ceph_snap_realm *realm;
+ struct inode *in;
+ u64 total = 0, used, free;
+ bool is_updated = false;
+
+ down_read(&mdsc->snap_rwsem);
+ realm = get_quota_realm(mdsc, d_inode(fsc->sb->s_root));
+ up_read(&mdsc->snap_rwsem);
+ if (!realm)
+ return false;
+
+ spin_lock(&realm->inodes_with_caps_lock);
+ in = realm->inode ? igrab(realm->inode) : NULL;
+ spin_unlock(&realm->inodes_with_caps_lock);
+ if (in) {
+ ci = ceph_inode(in);
+ spin_lock(&ci->i_ceph_lock);
+ if (ci->i_max_bytes) {
+ total = ci->i_max_bytes >> CEPH_BLOCK_SHIFT;
+ used = ci->i_rbytes >> CEPH_BLOCK_SHIFT;
+ /* It is possible for a quota to be exceeded.
+ * Report 'zero' in that case
+ */
+ free = total > used ? total - used : 0;
+ }
+ spin_unlock(&ci->i_ceph_lock);
+ if (total) {
+ buf->f_blocks = total;
+ buf->f_bfree = free;
+ buf->f_bavail = free;
+ is_updated = true;
+ }
+ iput(in);
+ }
+ ceph_put_snap_realm(mdsc, realm);
+
+ return is_updated;
+}
+
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index 07cf95e6413d..041c27ea8de1 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -931,6 +931,8 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc,
list_add(&ci->i_snap_realm_item,
&realm->inodes_with_caps);
ci->i_snap_realm = realm;
+ if (realm->ino == ci->i_vino.ino)
+ realm->inode = inode;
spin_unlock(&realm->inodes_with_caps_lock);
spin_unlock(&ci->i_ceph_lock);
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index fb2bc9c15a23..b33082e6878f 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -76,9 +76,18 @@ static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf)
*/
buf->f_bsize = 1 << CEPH_BLOCK_SHIFT;
buf->f_frsize = 1 << CEPH_BLOCK_SHIFT;
- buf->f_blocks = le64_to_cpu(st.kb) >> (CEPH_BLOCK_SHIFT-10);
- buf->f_bfree = le64_to_cpu(st.kb_avail) >> (CEPH_BLOCK_SHIFT-10);
- buf->f_bavail = le64_to_cpu(st.kb_avail) >> (CEPH_BLOCK_SHIFT-10);
+
+ /*
+ * By default use root quota for stats; fallback to overall filesystem
+ * usage if using 'noquotadf' mount option or if the root dir doesn't
+ * have max_bytes quota set.
+ */
+ if (ceph_test_mount_opt(fsc, NOQUOTADF) ||
+ !ceph_quota_update_statfs(fsc, buf)) {
+ buf->f_blocks = le64_to_cpu(st.kb) >> (CEPH_BLOCK_SHIFT-10);
+ buf->f_bfree = le64_to_cpu(st.kb_avail) >> (CEPH_BLOCK_SHIFT-10);
+ 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;
@@ -151,6 +160,8 @@ enum {
Opt_acl,
#endif
Opt_noacl,
+ Opt_quotadf,
+ Opt_noquotadf,
};
static match_table_t fsopt_tokens = {
@@ -187,6 +198,8 @@ static match_table_t fsopt_tokens = {
{Opt_acl, "acl"},
#endif
{Opt_noacl, "noacl"},
+ {Opt_quotadf, "quotadf"},
+ {Opt_noquotadf, "noquotadf"},
{-1, NULL}
};
@@ -314,13 +327,16 @@ static int parse_fsopt_token(char *c, void *private)
break;
case Opt_fscache:
fsopt->flags |= CEPH_MOUNT_OPT_FSCACHE;
+ kfree(fsopt->fscache_uniq);
+ fsopt->fscache_uniq = NULL;
break;
case Opt_nofscache:
fsopt->flags &= ~CEPH_MOUNT_OPT_FSCACHE;
+ kfree(fsopt->fscache_uniq);
+ fsopt->fscache_uniq = NULL;
break;
case Opt_poolperm:
fsopt->flags &= ~CEPH_MOUNT_OPT_NOPOOLPERM;
- printk ("pool perm");
break;
case Opt_nopoolperm:
fsopt->flags |= CEPH_MOUNT_OPT_NOPOOLPERM;
@@ -331,6 +347,12 @@ static int parse_fsopt_token(char *c, void *private)
case Opt_norequire_active_mds:
fsopt->flags |= CEPH_MOUNT_OPT_MOUNTWAIT;
break;
+ case Opt_quotadf:
+ fsopt->flags &= ~CEPH_MOUNT_OPT_NOQUOTADF;
+ break;
+ case Opt_noquotadf:
+ fsopt->flags |= CEPH_MOUNT_OPT_NOQUOTADF;
+ break;
#ifdef CONFIG_CEPH_FS_POSIX_ACL
case Opt_acl:
fsopt->sb_flags |= SB_POSIXACL;
@@ -513,13 +535,12 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root)
if ((fsopt->flags & CEPH_MOUNT_OPT_DCACHE) == 0)
seq_puts(m, ",nodcache");
if (fsopt->flags & CEPH_MOUNT_OPT_FSCACHE) {
- if (fsopt->fscache_uniq)
- seq_printf(m, ",fsc=%s", fsopt->fscache_uniq);
- else
- seq_puts(m, ",fsc");
+ seq_show_option(m, "fsc", fsopt->fscache_uniq);
}
if (fsopt->flags & CEPH_MOUNT_OPT_NOPOOLPERM)
seq_puts(m, ",nopoolperm");
+ if (fsopt->flags & CEPH_MOUNT_OPT_NOQUOTADF)
+ seq_puts(m, ",noquotadf");
#ifdef CONFIG_CEPH_FS_POSIX_ACL
if (fsopt->sb_flags & SB_POSIXACL)
@@ -529,7 +550,7 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root)
#endif
if (fsopt->mds_namespace)
- seq_printf(m, ",mds_namespace=%s", fsopt->mds_namespace);
+ seq_show_option(m, "mds_namespace", fsopt->mds_namespace);
if (fsopt->wsize)
seq_printf(m, ",wsize=%d", fsopt->wsize);
if (fsopt->rsize != CEPH_MAX_READ_SIZE)
@@ -679,6 +700,7 @@ struct kmem_cache *ceph_cap_cachep;
struct kmem_cache *ceph_cap_flush_cachep;
struct kmem_cache *ceph_dentry_cachep;
struct kmem_cache *ceph_file_cachep;
+struct kmem_cache *ceph_dir_file_cachep;
static void ceph_inode_init_once(void *foo)
{
@@ -698,8 +720,7 @@ static int __init init_caches(void)
if (!ceph_inode_cachep)
return -ENOMEM;
- ceph_cap_cachep = KMEM_CACHE(ceph_cap,
- SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD);
+ ceph_cap_cachep = KMEM_CACHE(ceph_cap, SLAB_MEM_SPREAD);
if (!ceph_cap_cachep)
goto bad_cap;
ceph_cap_flush_cachep = KMEM_CACHE(ceph_cap_flush,
@@ -716,6 +737,10 @@ static int __init init_caches(void)
if (!ceph_file_cachep)
goto bad_file;
+ ceph_dir_file_cachep = KMEM_CACHE(ceph_dir_file_info, SLAB_MEM_SPREAD);
+ if (!ceph_dir_file_cachep)
+ goto bad_dir_file;
+
error = ceph_fscache_register();
if (error)
goto bad_fscache;
@@ -723,6 +748,8 @@ static int __init init_caches(void)
return 0;
bad_fscache:
+ kmem_cache_destroy(ceph_dir_file_cachep);
+bad_dir_file:
kmem_cache_destroy(ceph_file_cachep);
bad_file:
kmem_cache_destroy(ceph_dentry_cachep);
@@ -748,6 +775,7 @@ static void destroy_caches(void)
kmem_cache_destroy(ceph_cap_flush_cachep);
kmem_cache_destroy(ceph_dentry_cachep);
kmem_cache_destroy(ceph_file_cachep);
+ kmem_cache_destroy(ceph_dir_file_cachep);
ceph_fscache_unregister();
}
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index 1c2086e0fec2..a7077a0c989f 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -39,6 +39,7 @@
#define CEPH_MOUNT_OPT_FSCACHE (1<<10) /* use fscache */
#define CEPH_MOUNT_OPT_NOPOOLPERM (1<<11) /* no pool permission check */
#define CEPH_MOUNT_OPT_MOUNTWAIT (1<<12) /* mount waits if no mds is up */
+#define CEPH_MOUNT_OPT_NOQUOTADF (1<<13) /* no root dir quota in statfs */
#define CEPH_MOUNT_OPT_DEFAULT CEPH_MOUNT_OPT_DCACHE
@@ -310,6 +311,9 @@ struct ceph_inode_info {
u64 i_rbytes, i_rfiles, i_rsubdirs;
u64 i_files, i_subdirs;
+ /* quotas */
+ u64 i_max_bytes, i_max_files;
+
struct rb_root i_fragtree;
int i_fragtree_nsplits;
struct mutex i_fragtree_mutex;
@@ -671,6 +675,10 @@ struct ceph_file_info {
spinlock_t rw_contexts_lock;
struct list_head rw_contexts;
+};
+
+struct ceph_dir_file_info {
+ struct ceph_file_info file_info;
/* readdir: position within the dir */
u32 frag;
@@ -748,6 +756,7 @@ struct ceph_readdir_cache_control {
*/
struct ceph_snap_realm {
u64 ino;
+ struct inode *inode;
atomic_t nref;
struct rb_node node;
@@ -1066,4 +1075,37 @@ extern int ceph_locks_to_pagelist(struct ceph_filelock *flocks,
extern int ceph_fs_debugfs_init(struct ceph_fs_client *client);
extern void ceph_fs_debugfs_cleanup(struct ceph_fs_client *client);
+/* quota.c */
+static inline bool __ceph_has_any_quota(struct ceph_inode_info *ci)
+{
+ return ci->i_max_files || ci->i_max_bytes;
+}
+
+extern void ceph_adjust_quota_realms_count(struct inode *inode, bool inc);
+
+static inline void __ceph_update_quota(struct ceph_inode_info *ci,
+ u64 max_bytes, u64 max_files)
+{
+ bool had_quota, has_quota;
+ had_quota = __ceph_has_any_quota(ci);
+ ci->i_max_bytes = max_bytes;
+ ci->i_max_files = max_files;
+ has_quota = __ceph_has_any_quota(ci);
+
+ if (had_quota != has_quota)
+ ceph_adjust_quota_realms_count(&ci->vfs_inode, has_quota);
+}
+
+extern void ceph_handle_quota(struct ceph_mds_client *mdsc,
+ struct ceph_mds_session *session,
+ struct ceph_msg *msg);
+extern bool ceph_quota_is_max_files_exceeded(struct inode *inode);
+extern bool ceph_quota_is_same_realm(struct inode *old, struct inode *new);
+extern bool ceph_quota_is_max_bytes_exceeded(struct inode *inode,
+ loff_t newlen);
+extern bool ceph_quota_is_max_bytes_approaching(struct inode *inode,
+ loff_t newlen);
+extern bool ceph_quota_update_statfs(struct ceph_fs_client *fsc,
+ struct kstatfs *buf);
+
#endif /* _FS_CEPH_SUPER_H */
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index e1c4e0b12b4c..315f7e63e7cc 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -224,6 +224,39 @@ static size_t ceph_vxattrcb_dir_rctime(struct ceph_inode_info *ci, char *val,
(long)ci->i_rctime.tv_nsec);
}
+/* quotas */
+
+static bool ceph_vxattrcb_quota_exists(struct ceph_inode_info *ci)
+{
+ bool ret = false;
+ spin_lock(&ci->i_ceph_lock);
+ if ((ci->i_max_files || ci->i_max_bytes) &&
+ ci->i_vino.snap == CEPH_NOSNAP &&
+ ci->i_snap_realm &&
+ ci->i_snap_realm->ino == ci->i_vino.ino)
+ ret = true;
+ spin_unlock(&ci->i_ceph_lock);
+ return ret;
+}
+
+static size_t ceph_vxattrcb_quota(struct ceph_inode_info *ci, char *val,
+ size_t size)
+{
+ return snprintf(val, size, "max_bytes=%llu max_files=%llu",
+ ci->i_max_bytes, ci->i_max_files);
+}
+
+static size_t ceph_vxattrcb_quota_max_bytes(struct ceph_inode_info *ci,
+ char *val, size_t size)
+{
+ return snprintf(val, size, "%llu", ci->i_max_bytes);
+}
+
+static size_t ceph_vxattrcb_quota_max_files(struct ceph_inode_info *ci,
+ char *val, size_t size)
+{
+ return snprintf(val, size, "%llu", ci->i_max_files);
+}
#define CEPH_XATTR_NAME(_type, _name) XATTR_CEPH_PREFIX #_type "." #_name
#define CEPH_XATTR_NAME2(_type, _name, _name2) \
@@ -247,6 +280,15 @@ static size_t ceph_vxattrcb_dir_rctime(struct ceph_inode_info *ci, char *val,
.hidden = true, \
.exists_cb = ceph_vxattrcb_layout_exists, \
}
+#define XATTR_QUOTA_FIELD(_type, _name) \
+ { \
+ .name = CEPH_XATTR_NAME(_type, _name), \
+ .name_size = sizeof(CEPH_XATTR_NAME(_type, _name)), \
+ .getxattr_cb = ceph_vxattrcb_ ## _type ## _ ## _name, \
+ .readonly = false, \
+ .hidden = true, \
+ .exists_cb = ceph_vxattrcb_quota_exists, \
+ }
static struct ceph_vxattr ceph_dir_vxattrs[] = {
{
@@ -270,6 +312,16 @@ static struct ceph_vxattr ceph_dir_vxattrs[] = {
XATTR_NAME_CEPH(dir, rsubdirs),
XATTR_NAME_CEPH(dir, rbytes),
XATTR_NAME_CEPH(dir, rctime),
+ {
+ .name = "ceph.quota",
+ .name_size = sizeof("ceph.quota"),
+ .getxattr_cb = ceph_vxattrcb_quota,
+ .readonly = false,
+ .hidden = true,
+ .exists_cb = ceph_vxattrcb_quota_exists,
+ },
+ XATTR_QUOTA_FIELD(quota, max_bytes),
+ XATTR_QUOTA_FIELD(quota, max_files),
{ .name = NULL, 0 } /* Required table terminator */
};
static size_t ceph_dir_vxattrs_name_size; /* total size of all names */
@@ -964,14 +1016,19 @@ int __ceph_setxattr(struct inode *inode, const char *name,
char *newval = NULL;
struct ceph_inode_xattr *xattr = NULL;
int required_blob_size;
+ bool check_realm = false;
bool lock_snap_rwsem = false;
if (ceph_snap(inode) != CEPH_NOSNAP)
return -EROFS;
vxattr = ceph_match_vxattr(inode, name);
- if (vxattr && vxattr->readonly)
- return -EOPNOTSUPP;
+ if (vxattr) {
+ if (vxattr->readonly)
+ return -EOPNOTSUPP;
+ if (value && !strncmp(vxattr->name, "ceph.quota", 10))
+ check_realm = true;
+ }
/* pass any unhandled ceph.* xattrs through to the MDS */
if (!strncmp(name, XATTR_CEPH_PREFIX, XATTR_CEPH_PREFIX_LEN))
@@ -1065,6 +1122,15 @@ do_sync_unlocked:
err = -EBUSY;
} else {
err = ceph_sync_setxattr(inode, name, value, size, flags);
+ if (err >= 0 && check_realm) {
+ /* check if snaprealm was created for quota inode */
+ spin_lock(&ci->i_ceph_lock);
+ if ((ci->i_max_files || ci->i_max_bytes) &&
+ !(ci->i_snap_realm &&
+ ci->i_snap_realm->ino == ci->i_vino.ino))
+ err = -EOPNOTSUPP;
+ spin_unlock(&ci->i_ceph_lock);
+ }
}
out:
ceph_free_cap_flush(prealloc_cf);
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.