aboutsummaryrefslogtreecommitdiff
path: root/fs/ceph
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ceph')
-rw-r--r--fs/ceph/Kconfig26
-rw-r--r--fs/ceph/Makefile25
-rw-r--r--fs/ceph/acl.c194
-rw-r--r--fs/ceph/addr.c673
-rw-r--r--fs/ceph/cache.c402
-rw-r--r--fs/ceph/cache.h182
-rw-r--r--fs/ceph/caps.c1065
-rw-r--r--fs/ceph/debugfs.c27
-rw-r--r--fs/ceph/dir.c612
-rw-r--r--fs/ceph/export.c266
-rw-r--r--fs/ceph/file.c1019
-rw-r--r--fs/ceph/inode.c892
-rw-r--r--fs/ceph/ioctl.c180
-rw-r--r--fs/ceph/ioctl.h60
-rw-r--r--fs/ceph/locks.c271
-rw-r--r--fs/ceph/mds_client.c801
-rw-r--r--fs/ceph/mds_client.h67
-rw-r--r--fs/ceph/mdsmap.c62
-rw-r--r--fs/ceph/snap.c88
-rw-r--r--fs/ceph/strings.c7
-rw-r--r--fs/ceph/super.c274
-rw-r--r--fs/ceph/super.h288
-rw-r--r--fs/ceph/xattr.c581
23 files changed, 5547 insertions, 2515 deletions
diff --git a/fs/ceph/Kconfig b/fs/ceph/Kconfig
index 9eb134ea6eb..264e9bf83ff 100644
--- a/fs/ceph/Kconfig
+++ b/fs/ceph/Kconfig
@@ -1,6 +1,6 @@
config CEPH_FS
- tristate "Ceph distributed file system (EXPERIMENTAL)"
- depends on INET && EXPERIMENTAL
+ tristate "Ceph distributed file system"
+ depends on INET
select CEPH_LIB
select LIBCRC32C
select CRYPTO_AES
@@ -16,3 +16,25 @@ config CEPH_FS
If unsure, say N.
+if CEPH_FS
+config CEPH_FSCACHE
+ bool "Enable Ceph client caching support"
+ depends on CEPH_FS=m && FSCACHE || CEPH_FS=y && FSCACHE=y
+ help
+ Choose Y here to enable persistent, read-only local
+ caching support for Ceph clients using FS-Cache
+
+endif
+
+config CEPH_FS_POSIX_ACL
+ bool "Ceph POSIX Access Control Lists"
+ depends on CEPH_FS
+ select FS_POSIX_ACL
+ help
+ POSIX Access Control Lists (ACLs) support permissions for users and
+ groups beyond the owner/group/world scheme.
+
+ To learn more about Access Control Lists, visit the POSIX ACLs for
+ Linux website <http://acl.bestbits.at/>.
+
+ If you don't know what Access Control Lists are, say N
diff --git a/fs/ceph/Makefile b/fs/ceph/Makefile
index 9e6c4f2e8ff..85a4230b9bf 100644
--- a/fs/ceph/Makefile
+++ b/fs/ceph/Makefile
@@ -2,31 +2,12 @@
# Makefile for CEPH filesystem.
#
-ifneq ($(KERNELRELEASE),)
-
obj-$(CONFIG_CEPH_FS) += ceph.o
-ceph-objs := super.o inode.o dir.o file.o locks.o addr.o ioctl.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 \
mds_client.o mdsmap.o strings.o ceph_frag.o \
debugfs.o
-else
-#Otherwise we were called directly from the command
-# line; invoke the kernel build system.
-
-KERNELDIR ?= /lib/modules/$(shell uname -r)/build
-PWD := $(shell pwd)
-
-default: all
-
-all:
- $(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_FS=m modules
-
-modules_install:
- $(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_FS=m modules_install
-
-clean:
- $(MAKE) -C $(KERNELDIR) M=$(PWD) clean
-
-endif
+ceph-$(CONFIG_CEPH_FSCACHE) += cache.o
+ceph-$(CONFIG_CEPH_FS_POSIX_ACL) += acl.o
diff --git a/fs/ceph/acl.c b/fs/ceph/acl.c
new file mode 100644
index 00000000000..469f2e8657e
--- /dev/null
+++ b/fs/ceph/acl.c
@@ -0,0 +1,194 @@
+/*
+ * linux/fs/ceph/acl.c
+ *
+ * Copyright (C) 2013 Guangliang Zhao, <lucienchao@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public
+ * License v2 as published by the Free Software Foundation.
+ *
+ * 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, write to the
+ * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ * Boston, MA 021110-1307, USA.
+ */
+
+#include <linux/ceph/ceph_debug.h>
+#include <linux/fs.h>
+#include <linux/string.h>
+#include <linux/xattr.h>
+#include <linux/posix_acl_xattr.h>
+#include <linux/posix_acl.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+
+#include "super.h"
+
+static inline void ceph_set_cached_acl(struct inode *inode,
+ int type, struct posix_acl *acl)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+
+ spin_lock(&ci->i_ceph_lock);
+ if (__ceph_caps_issued_mask(ci, CEPH_CAP_XATTR_SHARED, 0))
+ set_cached_acl(inode, type, acl);
+ spin_unlock(&ci->i_ceph_lock);
+}
+
+static inline struct posix_acl *ceph_get_cached_acl(struct inode *inode,
+ int type)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ struct posix_acl *acl = ACL_NOT_CACHED;
+
+ spin_lock(&ci->i_ceph_lock);
+ if (__ceph_caps_issued_mask(ci, CEPH_CAP_XATTR_SHARED, 0))
+ acl = get_cached_acl(inode, type);
+ spin_unlock(&ci->i_ceph_lock);
+
+ return acl;
+}
+
+struct posix_acl *ceph_get_acl(struct inode *inode, int type)
+{
+ int size;
+ const char *name;
+ char *value = NULL;
+ struct posix_acl *acl;
+
+ switch (type) {
+ case ACL_TYPE_ACCESS:
+ name = POSIX_ACL_XATTR_ACCESS;
+ break;
+ case ACL_TYPE_DEFAULT:
+ name = POSIX_ACL_XATTR_DEFAULT;
+ break;
+ default:
+ BUG();
+ }
+
+ size = __ceph_getxattr(inode, name, "", 0);
+ if (size > 0) {
+ value = kzalloc(size, GFP_NOFS);
+ if (!value)
+ return ERR_PTR(-ENOMEM);
+ size = __ceph_getxattr(inode, name, value, size);
+ }
+
+ if (size > 0)
+ acl = posix_acl_from_xattr(&init_user_ns, value, size);
+ else if (size == -ERANGE || size == -ENODATA || size == 0)
+ acl = NULL;
+ else
+ acl = ERR_PTR(-EIO);
+
+ kfree(value);
+
+ if (!IS_ERR(acl))
+ ceph_set_cached_acl(inode, type, acl);
+
+ return acl;
+}
+
+int ceph_set_acl(struct inode *inode, struct posix_acl *acl, int type)
+{
+ int ret = 0, size = 0;
+ const char *name = NULL;
+ char *value = NULL;
+ struct iattr newattrs;
+ umode_t new_mode = inode->i_mode, old_mode = inode->i_mode;
+ struct dentry *dentry;
+
+ switch (type) {
+ case ACL_TYPE_ACCESS:
+ name = POSIX_ACL_XATTR_ACCESS;
+ if (acl) {
+ ret = posix_acl_equiv_mode(acl, &new_mode);
+ if (ret < 0)
+ goto out;
+ if (ret == 0)
+ acl = NULL;
+ }
+ break;
+ case ACL_TYPE_DEFAULT:
+ if (!S_ISDIR(inode->i_mode)) {
+ ret = acl ? -EINVAL : 0;
+ goto out;
+ }
+ name = POSIX_ACL_XATTR_DEFAULT;
+ break;
+ default:
+ ret = -EINVAL;
+ goto out;
+ }
+
+ if (acl) {
+ size = posix_acl_xattr_size(acl->a_count);
+ value = kmalloc(size, GFP_NOFS);
+ if (!value) {
+ ret = -ENOMEM;
+ goto out;
+ }
+
+ ret = posix_acl_to_xattr(&init_user_ns, acl, value, size);
+ if (ret < 0)
+ goto out_free;
+ }
+
+ dentry = d_find_alias(inode);
+ if (new_mode != old_mode) {
+ newattrs.ia_mode = new_mode;
+ newattrs.ia_valid = ATTR_MODE;
+ ret = ceph_setattr(dentry, &newattrs);
+ if (ret)
+ goto out_dput;
+ }
+
+ ret = __ceph_setxattr(dentry, name, value, size, 0);
+ if (ret) {
+ if (new_mode != old_mode) {
+ newattrs.ia_mode = old_mode;
+ newattrs.ia_valid = ATTR_MODE;
+ ceph_setattr(dentry, &newattrs);
+ }
+ goto out_dput;
+ }
+
+ ceph_set_cached_acl(inode, type, acl);
+
+out_dput:
+ dput(dentry);
+out_free:
+ kfree(value);
+out:
+ return ret;
+}
+
+int ceph_init_acl(struct dentry *dentry, struct inode *inode, struct inode *dir)
+{
+ struct posix_acl *default_acl, *acl;
+ int error;
+
+ error = posix_acl_create(dir, &inode->i_mode, &default_acl, &acl);
+ if (error)
+ return error;
+
+ if (!default_acl && !acl)
+ cache_no_acl(inode);
+
+ if (default_acl) {
+ error = ceph_set_acl(inode, default_acl, ACL_TYPE_DEFAULT);
+ posix_acl_release(default_acl);
+ }
+ if (acl) {
+ if (!error)
+ error = ceph_set_acl(inode, acl, ACL_TYPE_ACCESS);
+ posix_acl_release(acl);
+ }
+ return error;
+}
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index e9c874abc9e..90b3954d48e 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -11,6 +11,7 @@
#include "super.h"
#include "mds_client.h"
+#include "cache.h"
#include <linux/ceph/osd_client.h>
/*
@@ -24,7 +25,7 @@
* context needs to be associated with the osd write during writeback.
*
* Similarly, struct ceph_inode_info maintains a set of counters to
- * count dirty pages on the inode. In the absense of snapshots,
+ * count dirty pages on the inode. In the absence of snapshots,
* i_wrbuffer_ref == i_wrbuffer_ref_head == the dirty page count.
*
* When a snapshot is taken (that is, when the client receives
@@ -54,7 +55,12 @@
(CONGESTION_ON_THRESH(congestion_kb) - \
(CONGESTION_ON_THRESH(congestion_kb) >> 2))
-
+static inline struct ceph_snap_context *page_snap_context(struct page *page)
+{
+ if (PagePrivate(page))
+ return (void *)page->private;
+ return NULL;
+}
/*
* Dirty a page. Optimistically adjust accounting, on the assumption
@@ -65,15 +71,16 @@ static int ceph_set_page_dirty(struct page *page)
struct address_space *mapping = page->mapping;
struct inode *inode;
struct ceph_inode_info *ci;
- int undo = 0;
struct ceph_snap_context *snapc;
+ int ret;
if (unlikely(!mapping))
return !TestSetPageDirty(page);
- if (TestSetPageDirty(page)) {
+ if (PageDirty(page)) {
dout("%p set_page_dirty %p idx %lu -- already dirty\n",
mapping->host, page, page->index);
+ BUG_ON(!PagePrivate(page));
return 0;
}
@@ -87,12 +94,12 @@ static int ceph_set_page_dirty(struct page *page)
snapc = ceph_get_snap_context(ci->i_snap_realm->cached_context);
/* dirty the head */
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (ci->i_head_snapc == NULL)
ci->i_head_snapc = ceph_get_snap_context(snapc);
++ci->i_wrbuffer_ref_head;
if (ci->i_wrbuffer_ref == 0)
- igrab(inode);
+ ihold(inode);
++ci->i_wrbuffer_ref;
dout("%p set_page_dirty %p idx %lu head %d/%d -> %d/%d "
"snapc %p seq %lld (%d snaps)\n",
@@ -100,37 +107,21 @@ static int ceph_set_page_dirty(struct page *page)
ci->i_wrbuffer_ref-1, ci->i_wrbuffer_ref_head-1,
ci->i_wrbuffer_ref, ci->i_wrbuffer_ref_head,
snapc, snapc->seq, snapc->num_snaps);
- spin_unlock(&inode->i_lock);
-
- /* now adjust page */
- spin_lock_irq(&mapping->tree_lock);
- if (page->mapping) { /* Race with truncate? */
- WARN_ON_ONCE(!PageUptodate(page));
- account_page_dirtied(page, page->mapping);
- radix_tree_tag_set(&mapping->page_tree,
- page_index(page), PAGECACHE_TAG_DIRTY);
-
- /*
- * Reference snap context in page->private. Also set
- * PagePrivate so that we get invalidatepage callback.
- */
- page->private = (unsigned long)snapc;
- SetPagePrivate(page);
- } else {
- dout("ANON set_page_dirty %p (raced truncate?)\n", page);
- undo = 1;
- }
-
- spin_unlock_irq(&mapping->tree_lock);
+ spin_unlock(&ci->i_ceph_lock);
- if (undo)
- /* whoops, we failed to dirty the page */
- ceph_put_wrbuffer_cap_refs(ci, 1, snapc);
+ /*
+ * Reference snap context in page->private. Also set
+ * PagePrivate so that we get invalidatepage callback.
+ */
+ BUG_ON(PagePrivate(page));
+ page->private = (unsigned long)snapc;
+ SetPagePrivate(page);
- __mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
+ ret = __set_page_dirty_nobuffers(page);
+ WARN_ON(!PageLocked(page));
+ WARN_ON(!page->mapping);
- BUG_ON(!PageDirty(page));
- return 1;
+ return ret;
}
/*
@@ -138,18 +129,26 @@ static int ceph_set_page_dirty(struct page *page)
* dirty page counters appropriately. Only called if there is private
* data on the page.
*/
-static void ceph_invalidatepage(struct page *page, unsigned long offset)
+static void ceph_invalidatepage(struct page *page, unsigned int offset,
+ unsigned int length)
{
struct inode *inode;
struct ceph_inode_info *ci;
- struct ceph_snap_context *snapc = (void *)page->private;
-
- BUG_ON(!PageLocked(page));
- BUG_ON(!page->private);
- BUG_ON(!PagePrivate(page));
- BUG_ON(!page->mapping);
+ struct ceph_snap_context *snapc = page_snap_context(page);
inode = page->mapping->host;
+ ci = ceph_inode(inode);
+
+ if (offset != 0 || length != PAGE_CACHE_SIZE) {
+ dout("%p invalidatepage %p idx %lu partial dirty page %u~%u\n",
+ inode, page, page->index, offset, length);
+ return;
+ }
+
+ ceph_invalidate_fscache_page(inode, page);
+
+ if (!PagePrivate(page))
+ return;
/*
* We can get non-dirty pages here due to races between
@@ -159,32 +158,28 @@ static void ceph_invalidatepage(struct page *page, unsigned long offset)
if (!PageDirty(page))
pr_err("%p invalidatepage %p page not dirty\n", inode, page);
- if (offset == 0)
- ClearPageChecked(page);
+ ClearPageChecked(page);
- ci = ceph_inode(inode);
- if (offset == 0) {
- dout("%p invalidatepage %p idx %lu full dirty page %lu\n",
- inode, page, page->index, offset);
- ceph_put_wrbuffer_cap_refs(ci, 1, snapc);
- ceph_put_snap_context(snapc);
- page->private = 0;
- ClearPagePrivate(page);
- } else {
- dout("%p invalidatepage %p idx %lu partial dirty page\n",
- inode, page, page->index);
- }
+ dout("%p invalidatepage %p idx %lu full dirty page\n",
+ inode, page, page->index);
+
+ ceph_put_wrbuffer_cap_refs(ci, 1, snapc);
+ ceph_put_snap_context(snapc);
+ page->private = 0;
+ ClearPagePrivate(page);
}
-/* just a sanity check */
static int ceph_releasepage(struct page *page, gfp_t g)
{
struct inode *inode = page->mapping ? page->mapping->host : NULL;
dout("%p releasepage %p idx %lu\n", inode, page, page->index);
WARN_ON(PageDirty(page));
- WARN_ON(page->private);
- WARN_ON(PagePrivate(page));
- return 0;
+
+ /* Can we release the page from the cache? */
+ if (!ceph_release_fscache_page(page, g))
+ return 0;
+
+ return !PagePrivate(page);
}
/*
@@ -192,29 +187,39 @@ static int ceph_releasepage(struct page *page, gfp_t g)
*/
static int readpage_nounlock(struct file *filp, struct page *page)
{
- struct inode *inode = filp->f_dentry->d_inode;
+ struct inode *inode = file_inode(filp);
struct ceph_inode_info *ci = ceph_inode(inode);
- struct ceph_osd_client *osdc =
+ struct ceph_osd_client *osdc =
&ceph_inode_to_client(inode)->client->osdc;
int err = 0;
u64 len = PAGE_CACHE_SIZE;
+ err = ceph_readpage_from_fscache(inode, page);
+
+ if (err == 0)
+ goto out;
+
dout("readpage inode %p file %p page %p index %lu\n",
inode, filp, page, page->index);
err = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout,
- page->index << PAGE_CACHE_SHIFT, &len,
+ (u64) page_offset(page), &len,
ci->i_truncate_seq, ci->i_truncate_size,
- &page, 1);
+ &page, 1, 0);
if (err == -ENOENT)
err = 0;
if (err < 0) {
SetPageError(page);
+ ceph_fscache_readpage_cancel(inode, page);
goto out;
- } else if (err < PAGE_CACHE_SIZE) {
+ }
+ if (err < PAGE_CACHE_SIZE)
/* zero fill remainder of page */
zero_user_segment(page, err, PAGE_CACHE_SIZE);
- }
+ else
+ flush_dcache_page(page);
+
SetPageUptodate(page);
+ ceph_readpage_to_fscache(inode, page);
out:
return err < 0 ? err : 0;
@@ -228,102 +233,180 @@ static int ceph_readpage(struct file *filp, struct page *page)
}
/*
- * Build a vector of contiguous pages from the provided page list.
+ * Finish an async read(ahead) op.
*/
-static struct page **page_vector_from_list(struct list_head *page_list,
- unsigned *nr_pages)
+static void finish_read(struct ceph_osd_request *req, struct ceph_msg *msg)
{
- struct page **pages;
- struct page *page;
- int next_index, contig_pages = 0;
+ struct inode *inode = req->r_inode;
+ struct ceph_osd_data *osd_data;
+ int rc = req->r_result;
+ int bytes = le32_to_cpu(msg->hdr.data_len);
+ int num_pages;
+ int i;
- /* build page vector */
- pages = kmalloc(sizeof(*pages) * *nr_pages, GFP_NOFS);
- if (!pages)
- return ERR_PTR(-ENOMEM);
+ dout("finish_read %p req %p rc %d bytes %d\n", inode, req, rc, bytes);
- BUG_ON(list_empty(page_list));
- next_index = list_entry(page_list->prev, struct page, lru)->index;
- list_for_each_entry_reverse(page, page_list, lru) {
- if (page->index == next_index) {
- dout("readpages page %d %p\n", contig_pages, page);
- pages[contig_pages] = page;
- contig_pages++;
- next_index++;
- } else {
- break;
+ /* unlock all pages, zeroing any data we didn't read */
+ osd_data = osd_req_op_extent_osd_data(req, 0);
+ BUG_ON(osd_data->type != CEPH_OSD_DATA_TYPE_PAGES);
+ num_pages = calc_pages_for((u64)osd_data->alignment,
+ (u64)osd_data->length);
+ for (i = 0; i < num_pages; i++) {
+ struct page *page = osd_data->pages[i];
+
+ if (rc < 0)
+ goto unlock;
+ if (bytes < (int)PAGE_CACHE_SIZE) {
+ /* zero (remainder of) page */
+ int s = bytes < 0 ? 0 : bytes;
+ zero_user_segment(page, s, PAGE_CACHE_SIZE);
}
+ dout("finish_read %p uptodate %p idx %lu\n", inode, page,
+ page->index);
+ flush_dcache_page(page);
+ SetPageUptodate(page);
+ ceph_readpage_to_fscache(inode, page);
+unlock:
+ unlock_page(page);
+ page_cache_release(page);
+ bytes -= PAGE_CACHE_SIZE;
}
- *nr_pages = contig_pages;
- return pages;
+ kfree(osd_data->pages);
+}
+
+static void ceph_unlock_page_vector(struct page **pages, int num_pages)
+{
+ int i;
+
+ for (i = 0; i < num_pages; i++)
+ unlock_page(pages[i]);
}
/*
- * Read multiple pages. Leave pages we don't read + unlock in page_list;
- * the caller (VM) cleans them up.
+ * start an async read(ahead) operation. return nr_pages we submitted
+ * a read for on success, or negative error code.
*/
-static int ceph_readpages(struct file *file, struct address_space *mapping,
- struct list_head *page_list, unsigned nr_pages)
+static int start_read(struct inode *inode, struct list_head *page_list, int max)
{
- struct inode *inode = file->f_dentry->d_inode;
- struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_osd_client *osdc =
&ceph_inode_to_client(inode)->client->osdc;
- int rc = 0;
- struct page **pages;
- loff_t offset;
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ struct page *page = list_entry(page_list->prev, struct page, lru);
+ struct ceph_vino vino;
+ struct ceph_osd_request *req;
+ u64 off;
u64 len;
+ int i;
+ struct page **pages;
+ pgoff_t next_index;
+ int nr_pages = 0;
+ int ret;
- dout("readpages %p file %p nr_pages %d\n",
- inode, file, nr_pages);
-
- pages = page_vector_from_list(page_list, &nr_pages);
- if (IS_ERR(pages))
- return PTR_ERR(pages);
+ off = (u64) page_offset(page);
- /* guess read extent */
- offset = pages[0]->index << PAGE_CACHE_SHIFT;
+ /* count pages */
+ next_index = page->index;
+ list_for_each_entry_reverse(page, page_list, lru) {
+ if (page->index != next_index)
+ break;
+ nr_pages++;
+ next_index++;
+ if (max && nr_pages == max)
+ break;
+ }
len = nr_pages << PAGE_CACHE_SHIFT;
- rc = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout,
- offset, &len,
- ci->i_truncate_seq, ci->i_truncate_size,
- pages, nr_pages);
- if (rc == -ENOENT)
- rc = 0;
- if (rc < 0)
- goto out;
-
- for (; !list_empty(page_list) && len > 0;
- rc -= PAGE_CACHE_SIZE, len -= PAGE_CACHE_SIZE) {
- struct page *page =
- list_entry(page_list->prev, struct page, lru);
+ dout("start_read %p nr_pages %d is %lld~%lld\n", inode, nr_pages,
+ off, len);
+ vino = ceph_vino(inode);
+ req = ceph_osdc_new_request(osdc, &ci->i_layout, vino, off, &len,
+ 1, CEPH_OSD_OP_READ,
+ CEPH_OSD_FLAG_READ, NULL,
+ ci->i_truncate_seq, ci->i_truncate_size,
+ false);
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+ /* build page vector */
+ nr_pages = calc_pages_for(0, len);
+ pages = kmalloc(sizeof(*pages) * nr_pages, GFP_NOFS);
+ ret = -ENOMEM;
+ if (!pages)
+ goto out;
+ for (i = 0; i < nr_pages; ++i) {
+ page = list_entry(page_list->prev, struct page, lru);
+ BUG_ON(PageLocked(page));
list_del(&page->lru);
- if (rc < (int)PAGE_CACHE_SIZE) {
- /* zero (remainder of) page */
- int s = rc < 0 ? 0 : rc;
- zero_user_segment(page, s, PAGE_CACHE_SIZE);
- }
-
- if (add_to_page_cache_lru(page, mapping, page->index,
+ dout("start_read %p adding %p idx %lu\n", inode, page,
+ page->index);
+ if (add_to_page_cache_lru(page, &inode->i_data, page->index,
GFP_NOFS)) {
+ ceph_fscache_uncache_page(inode, page);
page_cache_release(page);
- dout("readpages %p add_to_page_cache failed %p\n",
+ dout("start_read %p add_to_page_cache failed %p\n",
inode, page);
- continue;
+ nr_pages = i;
+ goto out_pages;
}
- dout("readpages %p adding %p idx %lu\n", inode, page,
- page->index);
- flush_dcache_page(page);
- SetPageUptodate(page);
- unlock_page(page);
- page_cache_release(page);
+ pages[i] = page;
}
- rc = 0;
+ osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0, false, false);
+ req->r_callback = finish_read;
+ req->r_inode = inode;
+
+ ceph_osdc_build_request(req, off, NULL, vino.snap, NULL);
+
+ dout("start_read %p starting %p %lld~%lld\n", inode, req, off, len);
+ ret = ceph_osdc_start_request(osdc, req, false);
+ if (ret < 0)
+ goto out_pages;
+ ceph_osdc_put_request(req);
+ return nr_pages;
+out_pages:
+ ceph_unlock_page_vector(pages, nr_pages);
+ ceph_release_page_vector(pages, nr_pages);
out:
- kfree(pages);
+ ceph_osdc_put_request(req);
+ return ret;
+}
+
+
+/*
+ * Read multiple pages. Leave pages we don't read + unlock in page_list;
+ * the caller (VM) cleans them up.
+ */
+static int ceph_readpages(struct file *file, struct address_space *mapping,
+ struct list_head *page_list, unsigned nr_pages)
+{
+ struct inode *inode = file_inode(file);
+ struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+ int rc = 0;
+ int max = 0;
+
+ rc = ceph_readpages_from_fscache(mapping->host, mapping, page_list,
+ &nr_pages);
+
+ if (rc == 0)
+ goto out;
+
+ if (fsc->mount_options->rsize >= PAGE_CACHE_SIZE)
+ max = (fsc->mount_options->rsize + PAGE_CACHE_SIZE - 1)
+ >> PAGE_SHIFT;
+
+ dout("readpages %p file %p nr_pages %d max %d\n", inode,
+ file, nr_pages,
+ max);
+ while (!list_empty(page_list)) {
+ rc = start_read(inode, page_list, max);
+ if (rc < 0)
+ goto out;
+ BUG_ON(rc == 0);
+ }
+out:
+ ceph_fscache_readpages_cancel(inode, page_list);
+
+ dout("readpages %p file %p ret %d\n", inode, file, rc);
return rc;
}
@@ -338,7 +421,7 @@ static struct ceph_snap_context *get_oldest_context(struct inode *inode,
struct ceph_snap_context *snapc = NULL;
struct ceph_cap_snap *capsnap = NULL;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
dout(" cap_snap %p snapc %p has %d dirty pages\n", capsnap,
capsnap->context, capsnap->dirty_pages);
@@ -354,7 +437,7 @@ static struct ceph_snap_context *get_oldest_context(struct inode *inode,
dout(" head snapc %p has %d dirty pages\n",
snapc, ci->i_wrbuffer_ref_head);
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return snapc;
}
@@ -370,13 +453,12 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
struct ceph_inode_info *ci;
struct ceph_fs_client *fsc;
struct ceph_osd_client *osdc;
- loff_t page_off = page->index << PAGE_CACHE_SHIFT;
- int len = PAGE_CACHE_SIZE;
- loff_t i_size;
- int err = 0;
struct ceph_snap_context *snapc, *oldest;
- u64 snap_size = 0;
+ loff_t page_off = page_offset(page);
long writeback_stat;
+ u64 truncate_size, snap_size = 0;
+ u32 truncate_seq;
+ int err = 0, len = PAGE_CACHE_SIZE;
dout("writepage %p idx %lu\n", page, page->index);
@@ -390,7 +472,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
osdc = &fsc->client->osdc;
/* verify this is a writeable snap context */
- snapc = (void *)page->private;
+ snapc = page_snap_context(page);
if (snapc == NULL) {
dout("writepage %p page %p not dirty?\n", inode, page);
goto out;
@@ -398,7 +480,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
oldest = get_oldest_context(inode, &snap_size);
if (snapc->seq > oldest->seq) {
dout("writepage %p page %p snapc %p not writeable - noop\n",
- inode, page, (void *)page->private);
+ inode, page, snapc);
/* we should only noop if called by kswapd */
WARN_ON((current->flags & PF_MEMALLOC) == 0);
ceph_put_snap_context(oldest);
@@ -406,13 +488,20 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
}
ceph_put_snap_context(oldest);
+ spin_lock(&ci->i_ceph_lock);
+ truncate_seq = ci->i_truncate_seq;
+ truncate_size = ci->i_truncate_size;
+ if (!snap_size)
+ snap_size = i_size_read(inode);
+ spin_unlock(&ci->i_ceph_lock);
+
/* is this a partial page at end of file? */
- if (snap_size)
- i_size = snap_size;
- else
- i_size = i_size_read(inode);
- if (i_size < page_off + len)
- len = i_size - page_off;
+ if (page_off >= snap_size) {
+ dout("%p page eof %llu\n", page, snap_size);
+ goto out;
+ }
+ if (snap_size < page_off + len)
+ len = snap_size - page_off;
dout("writepage %p page %p index %lu on %llu~%u snapc %p\n",
inode, page, page->index, page_off, len, snapc);
@@ -422,13 +511,14 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
CONGESTION_ON_THRESH(fsc->mount_options->congestion_kb))
set_bdi_congested(&fsc->backing_dev_info, BLK_RW_ASYNC);
+ ceph_readpage_to_fscache(inode, page);
+
set_page_writeback(page);
err = ceph_osdc_writepages(osdc, ceph_vino(inode),
&ci->i_layout, snapc,
page_off, len,
- ci->i_truncate_seq, ci->i_truncate_size,
- &inode->i_mtime,
- &page, 1, 0, 0, true);
+ truncate_seq, truncate_size,
+ &inode->i_mtime, &page, 1);
if (err < 0) {
dout("writepage setting page/mapping error %d %p\n", err, page);
SetPageError(page);
@@ -453,7 +543,7 @@ static int ceph_writepage(struct page *page, struct writeback_control *wbc)
int err;
struct inode *inode = page->mapping->host;
BUG_ON(!inode);
- igrab(inode);
+ ihold(inode);
err = writepage_nounlock(page, wbc);
unlock_page(page);
iput(inode);
@@ -478,7 +568,6 @@ static void ceph_release_pages(struct page **pages, int num)
pagevec_release(&pvec);
}
-
/*
* async writeback completion handler.
*
@@ -489,27 +578,24 @@ static void writepages_finish(struct ceph_osd_request *req,
struct ceph_msg *msg)
{
struct inode *inode = req->r_inode;
- struct ceph_osd_reply_head *replyhead;
- struct ceph_osd_op *op;
struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_osd_data *osd_data;
unsigned wrote;
struct page *page;
+ int num_pages;
int i;
struct ceph_snap_context *snapc = req->r_snapc;
struct address_space *mapping = inode->i_mapping;
- __s32 rc = -EIO;
- u64 bytes = 0;
+ int rc = req->r_result;
+ u64 bytes = req->r_ops[0].extent.length;
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
long writeback_stat;
unsigned issued = ceph_caps_issued(ci);
- /* parse reply */
- replyhead = msg->front.iov_base;
- WARN_ON(le32_to_cpu(replyhead->num_ops) == 0);
- op = (void *)(replyhead + 1);
- rc = le32_to_cpu(replyhead->result);
- bytes = le64_to_cpu(op->extent.length);
-
+ osd_data = osd_req_op_extent_osd_data(req, 0);
+ BUG_ON(osd_data->type != CEPH_OSD_DATA_TYPE_PAGES);
+ num_pages = calc_pages_for((u64)osd_data->alignment,
+ (u64)osd_data->length);
if (rc >= 0) {
/*
* Assume we wrote the pages we originally sent. The
@@ -517,7 +603,7 @@ static void writepages_finish(struct ceph_osd_request *req,
* raced with a truncation and was adjusted at the osd,
* so don't believe the reply.
*/
- wrote = req->r_num_pages;
+ wrote = num_pages;
} else {
wrote = 0;
mapping_set_error(mapping, rc);
@@ -526,8 +612,8 @@ static void writepages_finish(struct ceph_osd_request *req,
inode, rc, bytes, wrote);
/* clean all pages */
- for (i = 0; i < req->r_num_pages; i++) {
- page = req->r_pages[i];
+ for (i = 0; i < num_pages; i++) {
+ page = osd_data->pages[i];
BUG_ON(!page);
WARN_ON(!PageUptodate(page));
@@ -538,7 +624,7 @@ static void writepages_finish(struct ceph_osd_request *req,
clear_bdi_congested(&fsc->backing_dev_info,
BLK_RW_ASYNC);
- ceph_put_snap_context((void *)page->private);
+ ceph_put_snap_context(page_snap_context(page));
page->private = 0;
ClearPagePrivate(page);
dout("unlocking %d %p\n", i, page);
@@ -556,35 +642,18 @@ static void writepages_finish(struct ceph_osd_request *req,
unlock_page(page);
}
dout("%p wrote+cleaned %d pages\n", inode, wrote);
- ceph_put_wrbuffer_cap_refs(ci, req->r_num_pages, snapc);
+ ceph_put_wrbuffer_cap_refs(ci, num_pages, snapc);
- ceph_release_pages(req->r_pages, req->r_num_pages);
- if (req->r_pages_from_pool)
- mempool_free(req->r_pages,
+ ceph_release_pages(osd_data->pages, num_pages);
+ if (osd_data->pages_from_pool)
+ mempool_free(osd_data->pages,
ceph_sb_to_client(inode->i_sb)->wb_pagevec_pool);
else
- kfree(req->r_pages);
+ kfree(osd_data->pages);
ceph_osdc_put_request(req);
}
/*
- * allocate a page vec, either directly, or if necessary, via a the
- * mempool. we avoid the mempool if we can because req->r_num_pages
- * may be less than the maximum write size.
- */
-static void alloc_page_vec(struct ceph_fs_client *fsc,
- struct ceph_osd_request *req)
-{
- req->r_pages = kmalloc(sizeof(struct page *) * req->r_num_pages,
- GFP_NOFS);
- if (!req->r_pages) {
- req->r_pages = mempool_alloc(fsc->wb_pagevec_pool, GFP_NOFS);
- req->r_pages_from_pool = 1;
- WARN_ON(!req->r_pages);
- }
-}
-
-/*
* initiate async writeback
*/
static int ceph_writepages_start(struct address_space *mapping,
@@ -592,7 +661,8 @@ static int ceph_writepages_start(struct address_space *mapping,
{
struct inode *inode = mapping->host;
struct ceph_inode_info *ci = ceph_inode(inode);
- struct ceph_fs_client *fsc;
+ struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+ struct ceph_vino vino = ceph_vino(inode);
pgoff_t index, start, end;
int range_whole = 0;
int should_loop = 1;
@@ -604,24 +674,24 @@ static int ceph_writepages_start(struct address_space *mapping,
unsigned wsize = 1 << inode->i_blkbits;
struct ceph_osd_request *req = NULL;
int do_sync;
- u64 snap_size = 0;
+ u64 truncate_size, snap_size;
+ u32 truncate_seq;
/*
* Include a 'sync' in the OSD request if this is a data
* integrity write (e.g., O_SYNC write or fsync()), or if our
* cap is being revoked.
*/
- do_sync = wbc->sync_mode == WB_SYNC_ALL;
- if (ceph_caps_revoking(ci, CEPH_CAP_FILE_BUFFER))
+ if ((wbc->sync_mode == WB_SYNC_ALL) ||
+ ceph_caps_revoking(ci, CEPH_CAP_FILE_BUFFER))
do_sync = 1;
dout("writepages_start %p dosync=%d (mode=%s)\n",
inode, do_sync,
wbc->sync_mode == WB_SYNC_NONE ? "NONE" :
(wbc->sync_mode == WB_SYNC_ALL ? "ALL" : "HOLD"));
- fsc = ceph_inode_to_client(inode);
if (fsc->mount_state == CEPH_MOUNT_SHUTDOWN) {
- pr_warning("writepage_start %p on forced umount\n", inode);
+ pr_warn("writepage_start %p on forced umount\n", inode);
return -EIO; /* we're in a forced umount, don't write! */
}
if (fsc->mount_options->wsize && fsc->mount_options->wsize < wsize)
@@ -650,6 +720,7 @@ static int ceph_writepages_start(struct address_space *mapping,
retry:
/* find oldest snap context with dirty data */
ceph_put_snap_context(snapc);
+ snap_size = 0;
snapc = get_oldest_context(inode, &snap_size);
if (!snapc) {
/* hmm, why does writepages get called when there
@@ -657,8 +728,18 @@ retry:
dout(" no snap context with dirty data?\n");
goto out;
}
+ if (snap_size == 0)
+ snap_size = i_size_read(inode);
dout(" oldest snapc is %p seq %lld (%d snaps)\n",
snapc, snapc->seq, snapc->num_snaps);
+
+ spin_lock(&ci->i_ceph_lock);
+ truncate_seq = ci->i_truncate_seq;
+ truncate_size = ci->i_truncate_size;
+ if (!snap_size)
+ snap_size = i_size_read(inode);
+ spin_unlock(&ci->i_ceph_lock);
+
if (last_snapc && snapc != last_snapc) {
/* if we switched to a newer snapc, restart our scan at the
* start of the original file range. */
@@ -669,15 +750,16 @@ retry:
last_snapc = snapc;
while (!done && index <= end) {
+ int num_ops = do_sync ? 2 : 1;
unsigned i;
int first;
pgoff_t next;
int pvec_pages, locked_pages;
+ struct page **pages = NULL;
+ mempool_t *pool = NULL; /* Becomes non-null if mempool used */
struct page *page;
int want;
u64 offset, len;
- struct ceph_osd_request_head *reqhead;
- struct ceph_osd_op *op;
long writeback_stat;
next = 0;
@@ -726,11 +808,8 @@ get_more_pages:
dout("waiting on writeback %p\n", page);
wait_on_page_writeback(page);
}
- if ((snap_size && page_offset(page) > snap_size) ||
- (!snap_size &&
- page_offset(page) > i_size_read(inode))) {
- dout("%p page eof %llu\n", page, snap_size ?
- snap_size : i_size_read(inode));
+ if (page_offset(page) >= snap_size) {
+ dout("%p page eof %llu\n", page, snap_size);
done = 1;
unlock_page(page);
break;
@@ -742,7 +821,7 @@ get_more_pages:
}
/* only if matching snap context */
- pgsnapc = (void *)page->private;
+ pgsnapc = page_snap_context(page);
if (pgsnapc->seq > snapc->seq) {
dout("page snapc %p %lld > oldest %p %lld\n",
pgsnapc, pgsnapc->seq, snapc, snapc->seq);
@@ -758,28 +837,42 @@ get_more_pages:
break;
}
- /* ok */
+ /*
+ * We have something to write. If this is
+ * the first locked page this time through,
+ * allocate an osd request and a page array
+ * that it will use.
+ */
if (locked_pages == 0) {
+ BUG_ON(pages);
/* prepare async write request */
- offset = (unsigned long long)page->index
- << PAGE_CACHE_SHIFT;
+ offset = (u64)page_offset(page);
len = wsize;
req = ceph_osdc_new_request(&fsc->client->osdc,
- &ci->i_layout,
- ceph_vino(inode),
- offset, &len,
- CEPH_OSD_OP_WRITE,
- CEPH_OSD_FLAG_WRITE |
- CEPH_OSD_FLAG_ONDISK,
- snapc, do_sync,
- ci->i_truncate_seq,
- ci->i_truncate_size,
- &inode->i_mtime, true, 1);
- max_pages = req->r_num_pages;
-
- alloc_page_vec(fsc, req);
+ &ci->i_layout, vino,
+ offset, &len, num_ops,
+ CEPH_OSD_OP_WRITE,
+ CEPH_OSD_FLAG_WRITE |
+ CEPH_OSD_FLAG_ONDISK,
+ snapc, truncate_seq,
+ truncate_size, true);
+ if (IS_ERR(req)) {
+ rc = PTR_ERR(req);
+ unlock_page(page);
+ break;
+ }
+
req->r_callback = writepages_finish;
req->r_inode = inode;
+
+ max_pages = calc_pages_for(0, (u64)len);
+ pages = kmalloc(max_pages * sizeof (*pages),
+ GFP_NOFS);
+ if (!pages) {
+ pool = fsc->wb_pagevec_pool;
+ pages = mempool_alloc(pool, GFP_NOFS);
+ BUG_ON(!pages);
+ }
}
/* note position of first page in pvec */
@@ -797,7 +890,7 @@ get_more_pages:
}
set_page_writeback(page);
- req->r_pages[locked_pages] = page;
+ pages[locked_pages] = page;
locked_pages++;
next = page->index + 1;
}
@@ -826,22 +919,30 @@ get_more_pages:
pvec.nr -= i-first;
}
- /* submit the write */
- offset = req->r_pages[0]->index << PAGE_CACHE_SHIFT;
- len = min((snap_size ? snap_size : i_size_read(inode)) - offset,
+ /* Format the osd request message and submit the write */
+
+ offset = page_offset(pages[0]);
+ len = min(snap_size - offset,
(u64)locked_pages << PAGE_CACHE_SHIFT);
dout("writepages got %d pages at %llu~%llu\n",
locked_pages, offset, len);
- /* revise final length, page count */
- req->r_num_pages = locked_pages;
- reqhead = req->r_request->front.iov_base;
- op = (void *)(reqhead + 1);
- op->extent.length = cpu_to_le64(len);
- op->payload_len = cpu_to_le32(len);
- req->r_request->hdr.data_len = cpu_to_le32(len);
+ osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0,
+ !!pool, false);
+
+ pages = NULL; /* request message now owns the pages array */
+ pool = NULL;
+
+ /* Update the write op length in case we changed it */
- ceph_osdc_start_request(&fsc->client->osdc, req, true);
+ osd_req_op_extent_update(req, 0, len);
+
+ vino = ceph_vino(inode);
+ ceph_osdc_build_request(req, offset, snapc, vino.snap,
+ &inode->i_mtime);
+
+ rc = ceph_osdc_start_request(&fsc->client->osdc, req, true);
+ BUG_ON(rc);
req = NULL;
/* continue? */
@@ -873,8 +974,6 @@ release_pvec_pages:
out:
if (req)
ceph_osdc_put_request(req);
- if (rc > 0)
- rc = 0; /* vfs expects us to return 0 */
ceph_put_snap_context(snapc);
dout("writepages done, rc = %d\n", rc);
return rc;
@@ -907,7 +1006,7 @@ static int ceph_update_writeable_page(struct file *file,
loff_t pos, unsigned len,
struct page *page)
{
- struct inode *inode = file->f_dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
loff_t page_off = pos & PAGE_CACHE_MASK;
@@ -925,7 +1024,7 @@ retry_locked:
BUG_ON(!ci->i_snap_realm);
down_read(&mdsc->snap_rwsem);
BUG_ON(!ci->i_snap_realm->cached_context);
- snapc = (void *)page->private;
+ snapc = page_snap_context(page);
if (snapc && snapc != ci->i_head_snapc) {
/*
* this page is already dirty in another (older) snap
@@ -1016,7 +1115,7 @@ static int ceph_write_begin(struct file *file, struct address_space *mapping,
loff_t pos, unsigned len, unsigned flags,
struct page **pagep, void **fsdata)
{
- struct inode *inode = file->f_dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct page *page;
pgoff_t index = pos >> PAGE_CACHE_SHIFT;
int r;
@@ -1046,7 +1145,7 @@ static int ceph_write_end(struct file *file, struct address_space *mapping,
loff_t pos, unsigned len, unsigned copied,
struct page *page, void *fsdata)
{
- struct inode *inode = file->f_dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
struct ceph_mds_client *mdsc = fsc->mdsc;
unsigned from = pos & (PAGE_CACHE_SIZE - 1);
@@ -1085,8 +1184,8 @@ static int ceph_write_end(struct file *file, struct address_space *mapping,
* never get called.
*/
static ssize_t ceph_direct_io(int rw, struct kiocb *iocb,
- const struct iovec *iov,
- loff_t pos, unsigned long nr_segs)
+ struct iov_iter *iter,
+ loff_t pos)
{
WARN_ON(1);
return -EINVAL;
@@ -1109,27 +1208,83 @@ const struct address_space_operations ceph_aops = {
/*
* vm ops
*/
+static int ceph_filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+ struct inode *inode = file_inode(vma->vm_file);
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_file_info *fi = vma->vm_file->private_data;
+ loff_t off = vmf->pgoff << PAGE_CACHE_SHIFT;
+ int want, got, ret;
+
+ dout("filemap_fault %p %llx.%llx %llu~%zd trying to get caps\n",
+ inode, ceph_vinop(inode), off, (size_t)PAGE_CACHE_SIZE);
+ if (fi->fmode & CEPH_FILE_MODE_LAZY)
+ want = CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO;
+ else
+ want = CEPH_CAP_FILE_CACHE;
+ while (1) {
+ got = 0;
+ ret = ceph_get_caps(ci, CEPH_CAP_FILE_RD, want, &got, -1);
+ if (ret == 0)
+ break;
+ if (ret != -ERESTARTSYS) {
+ WARN_ON(1);
+ return VM_FAULT_SIGBUS;
+ }
+ }
+ dout("filemap_fault %p %llu~%zd got cap refs on %s\n",
+ inode, off, (size_t)PAGE_CACHE_SIZE, ceph_cap_string(got));
+
+ ret = filemap_fault(vma, vmf);
+
+ dout("filemap_fault %p %llu~%zd dropping cap refs on %s ret %d\n",
+ inode, off, (size_t)PAGE_CACHE_SIZE, ceph_cap_string(got), ret);
+ ceph_put_cap_refs(ci, got);
+
+ return ret;
+}
/*
* Reuse write_begin here for simplicity.
*/
static int ceph_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
{
- struct inode *inode = vma->vm_file->f_dentry->d_inode;
- struct page *page = vmf->page;
+ struct inode *inode = file_inode(vma->vm_file);
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_file_info *fi = vma->vm_file->private_data;
struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
- loff_t off = page->index << PAGE_CACHE_SHIFT;
- loff_t size, len;
- int ret;
+ struct page *page = vmf->page;
+ loff_t off = page_offset(page);
+ loff_t size = i_size_read(inode);
+ size_t len;
+ int want, got, ret;
- size = i_size_read(inode);
if (off + PAGE_CACHE_SIZE <= size)
len = PAGE_CACHE_SIZE;
else
len = size & ~PAGE_CACHE_MASK;
- dout("page_mkwrite %p %llu~%llu page %p idx %lu\n", inode,
- off, len, page, page->index);
+ dout("page_mkwrite %p %llx.%llx %llu~%zd getting caps i_size %llu\n",
+ inode, ceph_vinop(inode), off, len, size);
+ if (fi->fmode & CEPH_FILE_MODE_LAZY)
+ want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO;
+ else
+ want = CEPH_CAP_FILE_BUFFER;
+ while (1) {
+ got = 0;
+ ret = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, &got, off + len);
+ if (ret == 0)
+ break;
+ if (ret != -ERESTARTSYS) {
+ WARN_ON(1);
+ return VM_FAULT_SIGBUS;
+ }
+ }
+ dout("page_mkwrite %p %llu~%zd got cap refs on %s\n",
+ inode, off, len, ceph_cap_string(got));
+
+ /* Update time before taking page lock */
+ file_update_time(vma->vm_file);
lock_page(page);
@@ -1151,15 +1306,28 @@ static int ceph_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
ret = VM_FAULT_SIGBUS;
}
out:
- dout("page_mkwrite %p %llu~%llu = %d\n", inode, off, len, ret);
- if (ret != VM_FAULT_LOCKED)
+ if (ret != VM_FAULT_LOCKED) {
unlock_page(page);
+ } else {
+ int dirty;
+ spin_lock(&ci->i_ceph_lock);
+ dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
+ spin_unlock(&ci->i_ceph_lock);
+ if (dirty)
+ __mark_inode_dirty(inode, dirty);
+ }
+
+ dout("page_mkwrite %p %llu~%zd dropping cap refs on %s ret %d\n",
+ inode, off, len, ceph_cap_string(got), ret);
+ ceph_put_cap_refs(ci, got);
+
return ret;
}
static struct vm_operations_struct ceph_vmops = {
- .fault = filemap_fault,
+ .fault = ceph_filemap_fault,
.page_mkwrite = ceph_page_mkwrite,
+ .remap_pages = generic_file_remap_pages,
};
int ceph_mmap(struct file *file, struct vm_area_struct *vma)
@@ -1170,6 +1338,5 @@ int ceph_mmap(struct file *file, struct vm_area_struct *vma)
return -ENOEXEC;
file_accessed(file);
vma->vm_ops = &ceph_vmops;
- vma->vm_flags |= VM_CAN_NONLINEAR;
return 0;
}
diff --git a/fs/ceph/cache.c b/fs/ceph/cache.c
new file mode 100644
index 00000000000..834f9f3723f
--- /dev/null
+++ b/fs/ceph/cache.c
@@ -0,0 +1,402 @@
+/*
+ * Ceph cache definitions.
+ *
+ * Copyright (C) 2013 by Adfin Solutions, Inc. All Rights Reserved.
+ * Written by Milosz Tanski (milosz@adfin.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation.
+ *
+ * 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, write to:
+ * Free Software Foundation
+ * 51 Franklin Street, Fifth Floor
+ * Boston, MA 02111-1301 USA
+ *
+ */
+
+#include "super.h"
+#include "cache.h"
+
+struct ceph_aux_inode {
+ struct timespec mtime;
+ loff_t size;
+};
+
+struct fscache_netfs ceph_cache_netfs = {
+ .name = "ceph",
+ .version = 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;
+ uint16_t klen;
+
+ klen = sizeof(fsc->client->fsid);
+ if (klen > maxbuf)
+ return 0;
+
+ memcpy(buffer, &fsc->client->fsid, klen);
+ return klen;
+}
+
+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)
+{
+ return fscache_register_netfs(&ceph_cache_netfs);
+}
+
+void ceph_fscache_unregister(void)
+{
+ fscache_unregister_netfs(&ceph_cache_netfs);
+}
+
+int ceph_fscache_register_fs(struct ceph_fs_client* fsc)
+{
+ fsc->fscache = fscache_acquire_cookie(ceph_cache_netfs.primary_index,
+ &ceph_fscache_fsid_object_def,
+ fsc, true);
+
+ if (fsc->fscache == NULL) {
+ pr_err("Unable to resgister fsid: %p fscache cookie", fsc);
+ return 0;
+ }
+
+ fsc->revalidate_wq = alloc_workqueue("ceph-revalidate", 0, 1);
+ if (fsc->revalidate_wq == NULL)
+ return -ENOMEM;
+
+ return 0;
+}
+
+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 + snaphot) */
+ 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.mtime = inode->i_mtime;
+ aux.size = inode->i_size;
+
+ 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;
+ const struct inode* inode = &ci->vfs_inode;
+
+ *size = inode->i_size;
+}
+
+static enum fscache_checkaux ceph_fscache_inode_check_aux(
+ void *cookie_netfs_data, const void *data, uint16_t dlen)
+{
+ struct ceph_aux_inode aux;
+ struct ceph_inode_info* ci = cookie_netfs_data;
+ struct inode* inode = &ci->vfs_inode;
+
+ if (dlen != sizeof(aux))
+ return FSCACHE_CHECKAUX_OBSOLETE;
+
+ memset(&aux, 0, sizeof(aux));
+ aux.mtime = inode->i_mtime;
+ aux.size = inode->i_size;
+
+ if (memcmp(data, &aux, sizeof(aux)) != 0)
+ return FSCACHE_CHECKAUX_OBSOLETE;
+
+ dout("ceph inode 0x%p cached okay", ci);
+ return FSCACHE_CHECKAUX_OKAY;
+}
+
+static void ceph_fscache_inode_now_uncached(void* cookie_netfs_data)
+{
+ struct ceph_inode_info* ci = cookie_netfs_data;
+ struct pagevec pvec;
+ pgoff_t first;
+ int loop, nr_pages;
+
+ pagevec_init(&pvec, 0);
+ first = 0;
+
+ dout("ceph inode 0x%p now uncached", ci);
+
+ while (1) {
+ nr_pages = pagevec_lookup(&pvec, ci->vfs_inode.i_mapping, first,
+ PAGEVEC_SIZE - pagevec_count(&pvec));
+
+ if (!nr_pages)
+ break;
+
+ for (loop = 0; loop < nr_pages; loop++)
+ ClearPageFsCache(pvec.pages[loop]);
+
+ first = pvec.pages[nr_pages - 1]->index + 1;
+
+ pvec.nr = nr_pages;
+ pagevec_release(&pvec);
+ cond_resched();
+ }
+}
+
+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,
+ .now_uncached = ceph_fscache_inode_now_uncached,
+};
+
+void ceph_fscache_register_inode_cookie(struct ceph_fs_client* fsc,
+ struct ceph_inode_info* ci)
+{
+ struct inode* inode = &ci->vfs_inode;
+
+ /* No caching for filesystem */
+ if (fsc->fscache == NULL)
+ return;
+
+ /* Only cache for regular files that are read only */
+ if ((ci->vfs_inode.i_mode & S_IFREG) == 0)
+ return;
+
+ /* Avoid multiple racing open requests */
+ mutex_lock(&inode->i_mutex);
+
+ if (ci->fscache)
+ goto done;
+
+ ci->fscache = fscache_acquire_cookie(fsc->fscache,
+ &ceph_fscache_inode_object_def,
+ ci, true);
+ fscache_check_consistency(ci->fscache);
+done:
+ mutex_unlock(&inode->i_mutex);
+
+}
+
+void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci)
+{
+ struct fscache_cookie* cookie;
+
+ if ((cookie = ci->fscache) == NULL)
+ return;
+
+ ci->fscache = NULL;
+
+ fscache_uncache_all_inode_pages(cookie, &ci->vfs_inode);
+ fscache_relinquish_cookie(cookie, 0);
+}
+
+static void ceph_vfs_readpage_complete(struct page *page, void *data, int error)
+{
+ if (!error)
+ SetPageUptodate(page);
+}
+
+static void ceph_vfs_readpage_complete_unlock(struct page *page, void *data, int error)
+{
+ if (!error)
+ SetPageUptodate(page);
+
+ unlock_page(page);
+}
+
+static inline int cache_valid(struct ceph_inode_info *ci)
+{
+ return ((ceph_caps_issued(ci) & CEPH_CAP_FILE_CACHE) &&
+ (ci->i_fscache_gen == ci->i_rdcache_gen));
+}
+
+
+/* Atempt to read from the fscache,
+ *
+ * This function is called from the readpage_nounlock context. DO NOT attempt to
+ * unlock the page here (or in the callback).
+ */
+int ceph_readpage_from_fscache(struct inode *inode, struct page *page)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ int ret;
+
+ if (!cache_valid(ci))
+ return -ENOBUFS;
+
+ ret = fscache_read_or_alloc_page(ci->fscache, page,
+ ceph_vfs_readpage_complete, NULL,
+ GFP_KERNEL);
+
+ switch (ret) {
+ case 0: /* Page found */
+ dout("page read submitted\n");
+ return 0;
+ case -ENOBUFS: /* Pages were not found, and can't be */
+ case -ENODATA: /* Pages were not found */
+ dout("page/inode not in cache\n");
+ return ret;
+ default:
+ dout("%s: unknown error ret = %i\n", __func__, ret);
+ return ret;
+ }
+}
+
+int ceph_readpages_from_fscache(struct inode *inode,
+ struct address_space *mapping,
+ struct list_head *pages,
+ unsigned *nr_pages)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ int ret;
+
+ if (!cache_valid(ci))
+ return -ENOBUFS;
+
+ ret = fscache_read_or_alloc_pages(ci->fscache, mapping, pages, nr_pages,
+ ceph_vfs_readpage_complete_unlock,
+ NULL, mapping_gfp_mask(mapping));
+
+ switch (ret) {
+ case 0: /* All pages found */
+ dout("all-page read submitted\n");
+ return 0;
+ case -ENOBUFS: /* Some pages were not found, and can't be */
+ case -ENODATA: /* some pages were not found */
+ dout("page/inode not in cache\n");
+ return ret;
+ default:
+ dout("%s: unknown error ret = %i\n", __func__, ret);
+ return ret;
+ }
+}
+
+void ceph_readpage_to_fscache(struct inode *inode, struct page *page)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ int ret;
+
+ if (!PageFsCache(page))
+ return;
+
+ if (!cache_valid(ci))
+ return;
+
+ ret = fscache_write_page(ci->fscache, page, GFP_KERNEL);
+ if (ret)
+ fscache_uncache_page(ci->fscache, page);
+}
+
+void ceph_invalidate_fscache_page(struct inode* inode, struct page *page)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+
+ if (!PageFsCache(page))
+ return;
+
+ fscache_wait_on_page_write(ci->fscache, page);
+ fscache_uncache_page(ci->fscache, page);
+}
+
+void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc)
+{
+ if (fsc->revalidate_wq)
+ destroy_workqueue(fsc->revalidate_wq);
+
+ fscache_relinquish_cookie(fsc->fscache, 0);
+ fsc->fscache = NULL;
+}
+
+static void ceph_revalidate_work(struct work_struct *work)
+{
+ int issued;
+ u32 orig_gen;
+ struct ceph_inode_info *ci = container_of(work, struct ceph_inode_info,
+ i_revalidate_work);
+ struct inode *inode = &ci->vfs_inode;
+
+ spin_lock(&ci->i_ceph_lock);
+ issued = __ceph_caps_issued(ci, NULL);
+ orig_gen = ci->i_rdcache_gen;
+ spin_unlock(&ci->i_ceph_lock);
+
+ if (!(issued & CEPH_CAP_FILE_CACHE)) {
+ dout("revalidate_work lost cache before validation %p\n",
+ inode);
+ goto out;
+ }
+
+ if (!fscache_check_consistency(ci->fscache))
+ fscache_invalidate(ci->fscache);
+
+ spin_lock(&ci->i_ceph_lock);
+ /* Update the new valid generation (backwards sanity check too) */
+ if (orig_gen > ci->i_fscache_gen) {
+ ci->i_fscache_gen = orig_gen;
+ }
+ spin_unlock(&ci->i_ceph_lock);
+
+out:
+ iput(&ci->vfs_inode);
+}
+
+void ceph_queue_revalidate(struct inode *inode)
+{
+ struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
+ struct ceph_inode_info *ci = ceph_inode(inode);
+
+ if (fsc->revalidate_wq == NULL || ci->fscache == NULL)
+ return;
+
+ ihold(inode);
+
+ if (queue_work(ceph_sb_to_client(inode->i_sb)->revalidate_wq,
+ &ci->i_revalidate_work)) {
+ dout("ceph_queue_revalidate %p\n", inode);
+ } else {
+ dout("ceph_queue_revalidate %p failed\n)", inode);
+ iput(inode);
+ }
+}
+
+void ceph_fscache_inode_init(struct ceph_inode_info *ci)
+{
+ ci->fscache = NULL;
+ /* The first load is verifed cookie open time */
+ ci->i_fscache_gen = 1;
+ INIT_WORK(&ci->i_revalidate_work, ceph_revalidate_work);
+}
diff --git a/fs/ceph/cache.h b/fs/ceph/cache.h
new file mode 100644
index 00000000000..5ac591bd012
--- /dev/null
+++ b/fs/ceph/cache.h
@@ -0,0 +1,182 @@
+/*
+ * Ceph cache definitions.
+ *
+ * Copyright (C) 2013 by Adfin Solutions, Inc. All Rights Reserved.
+ * Written by Milosz Tanski (milosz@adfin.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation.
+ *
+ * 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, write to:
+ * Free Software Foundation
+ * 51 Franklin Street, Fifth Floor
+ * Boston, MA 02111-1301 USA
+ *
+ */
+
+#ifndef _CEPH_CACHE_H
+#define _CEPH_CACHE_H
+
+#ifdef CONFIG_CEPH_FSCACHE
+
+extern struct fscache_netfs ceph_cache_netfs;
+
+int ceph_fscache_register(void);
+void ceph_fscache_unregister(void);
+
+int ceph_fscache_register_fs(struct ceph_fs_client* fsc);
+void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc);
+
+void ceph_fscache_inode_init(struct ceph_inode_info *ci);
+void ceph_fscache_register_inode_cookie(struct ceph_fs_client* fsc,
+ struct ceph_inode_info* ci);
+void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci);
+
+int ceph_readpage_from_fscache(struct inode *inode, struct page *page);
+int ceph_readpages_from_fscache(struct inode *inode,
+ struct address_space *mapping,
+ struct list_head *pages,
+ unsigned *nr_pages);
+void ceph_readpage_to_fscache(struct inode *inode, struct page *page);
+void ceph_invalidate_fscache_page(struct inode* inode, struct page *page);
+void ceph_queue_revalidate(struct inode *inode);
+
+static inline void ceph_fscache_update_objectsize(struct inode *inode)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ fscache_attr_changed(ci->fscache);
+}
+
+static inline void ceph_fscache_invalidate(struct inode *inode)
+{
+ fscache_invalidate(ceph_inode(inode)->fscache);
+}
+
+static inline void ceph_fscache_uncache_page(struct inode *inode,
+ struct page *page)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ return fscache_uncache_page(ci->fscache, page);
+}
+
+static inline int ceph_release_fscache_page(struct page *page, gfp_t gfp)
+{
+ struct inode* inode = page->mapping->host;
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ return fscache_maybe_release_page(ci->fscache, page, gfp);
+}
+
+static inline void ceph_fscache_readpage_cancel(struct inode *inode,
+ struct page *page)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ if (fscache_cookie_valid(ci->fscache) && PageFsCache(page))
+ __fscache_uncache_page(ci->fscache, page);
+}
+
+static inline void ceph_fscache_readpages_cancel(struct inode *inode,
+ struct list_head *pages)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ return fscache_readpages_cancel(ci->fscache, pages);
+}
+
+#else
+
+static inline int ceph_fscache_register(void)
+{
+ return 0;
+}
+
+static inline void ceph_fscache_unregister(void)
+{
+}
+
+static inline int ceph_fscache_register_fs(struct ceph_fs_client* fsc)
+{
+ return 0;
+}
+
+static inline void ceph_fscache_unregister_fs(struct ceph_fs_client* fsc)
+{
+}
+
+static inline void ceph_fscache_inode_init(struct ceph_inode_info *ci)
+{
+}
+
+static inline void ceph_fscache_register_inode_cookie(struct ceph_fs_client* parent_fsc,
+ struct ceph_inode_info* ci)
+{
+}
+
+static inline void ceph_fscache_uncache_page(struct inode *inode,
+ struct page *pages)
+{
+}
+
+static inline int ceph_readpage_from_fscache(struct inode* inode,
+ struct page *page)
+{
+ return -ENOBUFS;
+}
+
+static inline int ceph_readpages_from_fscache(struct inode *inode,
+ struct address_space *mapping,
+ struct list_head *pages,
+ unsigned *nr_pages)
+{
+ return -ENOBUFS;
+}
+
+static inline void ceph_readpage_to_fscache(struct inode *inode,
+ struct page *page)
+{
+}
+
+static inline void ceph_fscache_update_objectsize(struct inode *inode)
+{
+}
+
+static inline void ceph_fscache_invalidate(struct inode *inode)
+{
+}
+
+static inline void ceph_invalidate_fscache_page(struct inode *inode,
+ struct page *page)
+{
+}
+
+static inline void ceph_fscache_unregister_inode_cookie(struct ceph_inode_info* ci)
+{
+}
+
+static inline int ceph_release_fscache_page(struct page *page, gfp_t gfp)
+{
+ return 1;
+}
+
+static inline void ceph_fscache_readpage_cancel(struct inode *inode,
+ struct page *page)
+{
+}
+
+static inline void ceph_fscache_readpages_cancel(struct inode *inode,
+ struct list_head *pages)
+{
+}
+
+static inline void ceph_queue_revalidate(struct inode *inode)
+{
+}
+
+#endif
+
+#endif
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 98ab13e2b71..1fde164b74b 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -10,6 +10,7 @@
#include "super.h"
#include "mds_client.h"
+#include "cache.h"
#include <linux/ceph/decode.h>
#include <linux/ceph/messenger.h>
@@ -147,7 +148,7 @@ void ceph_adjust_min_caps(struct ceph_mds_client *mdsc, int delta)
spin_unlock(&mdsc->caps_list_lock);
}
-int ceph_reserve_caps(struct ceph_mds_client *mdsc,
+void ceph_reserve_caps(struct ceph_mds_client *mdsc,
struct ceph_cap_reservation *ctx, int need)
{
int i;
@@ -155,7 +156,6 @@ int ceph_reserve_caps(struct ceph_mds_client *mdsc,
int have;
int alloc = 0;
LIST_HEAD(newcaps);
- int ret = 0;
dout("reserve caps ctx=%p need=%d\n", ctx, need);
@@ -174,14 +174,15 @@ int ceph_reserve_caps(struct ceph_mds_client *mdsc,
for (i = have; i < need; i++) {
cap = kmem_cache_alloc(ceph_cap_cachep, GFP_NOFS);
- if (!cap) {
- ret = -ENOMEM;
- goto out_alloc_count;
- }
+ if (!cap)
+ break;
list_add(&cap->caps_item, &newcaps);
alloc++;
}
- BUG_ON(have + alloc != need);
+ /* we didn't manage to reserve as much as we needed */
+ if (have + alloc != need)
+ pr_warn("reserve caps ctx=%p ENOMEM need=%d got=%d\n",
+ ctx, need, have + alloc);
spin_lock(&mdsc->caps_list_lock);
mdsc->caps_total_count += alloc;
@@ -197,13 +198,6 @@ int ceph_reserve_caps(struct ceph_mds_client *mdsc,
dout("reserve caps ctx=%p %d = %d used + %d resv + %d avail\n",
ctx, mdsc->caps_total_count, mdsc->caps_use_count,
mdsc->caps_reserve_count, mdsc->caps_avail_count);
- return 0;
-
-out_alloc_count:
- /* we didn't manage to reserve as much as we needed */
- pr_warning("reserve caps ctx=%p ENOMEM need=%d got=%d\n",
- ctx, need, have);
- return ret;
}
int ceph_unreserve_caps(struct ceph_mds_client *mdsc,
@@ -227,8 +221,8 @@ int ceph_unreserve_caps(struct ceph_mds_client *mdsc,
return 0;
}
-static struct ceph_cap *get_cap(struct ceph_mds_client *mdsc,
- struct ceph_cap_reservation *ctx)
+struct ceph_cap *ceph_get_cap(struct ceph_mds_client *mdsc,
+ struct ceph_cap_reservation *ctx)
{
struct ceph_cap *cap = NULL;
@@ -236,8 +230,10 @@ static struct ceph_cap *get_cap(struct ceph_mds_client *mdsc,
if (!ctx) {
cap = kmem_cache_alloc(ceph_cap_cachep, GFP_NOFS);
if (cap) {
+ spin_lock(&mdsc->caps_list_lock);
mdsc->caps_use_count++;
mdsc->caps_total_count++;
+ spin_unlock(&mdsc->caps_list_lock);
}
return cap;
}
@@ -309,7 +305,7 @@ void ceph_reservation_status(struct ceph_fs_client *fsc,
/*
* Find ceph_cap for given mds, if any.
*
- * Called with i_lock held.
+ * Called with i_ceph_lock held.
*/
static struct ceph_cap *__get_cap_for_mds(struct ceph_inode_info *ci, int mds)
{
@@ -332,9 +328,9 @@ struct ceph_cap *ceph_get_cap_for_mds(struct ceph_inode_info *ci, int mds)
{
struct ceph_cap *cap;
- spin_lock(&ci->vfs_inode.i_lock);
+ spin_lock(&ci->i_ceph_lock);
cap = __get_cap_for_mds(ci, mds);
- spin_unlock(&ci->vfs_inode.i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return cap;
}
@@ -361,15 +357,16 @@ static int __ceph_get_cap_mds(struct ceph_inode_info *ci)
int ceph_get_cap_mds(struct inode *inode)
{
+ struct ceph_inode_info *ci = ceph_inode(inode);
int mds;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
mds = __ceph_get_cap_mds(ceph_inode(inode));
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return mds;
}
/*
- * Called under i_lock.
+ * Called under i_ceph_lock.
*/
static void __insert_cap_node(struct ceph_inode_info *ci,
struct ceph_cap *new)
@@ -415,7 +412,7 @@ static void __cap_set_timeouts(struct ceph_mds_client *mdsc,
*
* If I_FLUSH is set, leave the inode at the front of the list.
*
- * Caller holds i_lock
+ * Caller holds i_ceph_lock
* -> we take mdsc->cap_delay_lock
*/
static void __cap_delay_requeue(struct ceph_mds_client *mdsc,
@@ -457,7 +454,7 @@ static void __cap_delay_requeue_front(struct ceph_mds_client *mdsc,
/*
* Cancel delayed work on cap.
*
- * Caller must hold i_lock.
+ * Caller must hold i_ceph_lock.
*/
static void __cap_delay_cancel(struct ceph_mds_client *mdsc,
struct ceph_inode_info *ci)
@@ -483,11 +480,12 @@ static void __check_cap_issue(struct ceph_inode_info *ci, struct ceph_cap *cap,
* i_rdcache_gen.
*/
if ((issued & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) &&
- (had & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0)
+ (had & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0) {
ci->i_rdcache_gen++;
+ }
/*
- * if we are newly issued FILE_SHARED, clear I_COMPLETE; we
+ * if we are newly issued FILE_SHARED, mark dir not complete; we
* don't know what happened to this directory while we didn't
* have the cap.
*/
@@ -496,7 +494,7 @@ static void __check_cap_issue(struct ceph_inode_info *ci, struct ceph_cap *cap,
ci->i_shared_gen++;
if (S_ISDIR(ci->vfs_inode.i_mode)) {
dout(" marking %p NOT complete\n", &ci->vfs_inode);
- ci->i_ceph_flags &= ~CEPH_I_COMPLETE;
+ __ceph_dir_clear_complete(ci);
}
}
}
@@ -510,15 +508,14 @@ static void __check_cap_issue(struct ceph_inode_info *ci, struct ceph_cap *cap,
* it is < 0. (This is so we can atomically add the cap and add an
* open file reference to it.)
*/
-int ceph_add_cap(struct inode *inode,
- struct ceph_mds_session *session, u64 cap_id,
- int fmode, unsigned issued, unsigned wanted,
- unsigned seq, unsigned mseq, u64 realmino, int flags,
- struct ceph_cap_reservation *caps_reservation)
+void ceph_add_cap(struct inode *inode,
+ struct ceph_mds_session *session, u64 cap_id,
+ int fmode, unsigned issued, unsigned wanted,
+ unsigned seq, unsigned mseq, u64 realmino, int flags,
+ struct ceph_cap **new_cap)
{
struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
struct ceph_inode_info *ci = ceph_inode(inode);
- struct ceph_cap *new_cap = NULL;
struct ceph_cap *cap;
int mds = session->s_mds;
int actual_wanted;
@@ -533,42 +530,44 @@ int ceph_add_cap(struct inode *inode,
if (fmode >= 0)
wanted |= ceph_caps_for_mode(fmode);
-retry:
- spin_lock(&inode->i_lock);
cap = __get_cap_for_mds(ci, mds);
if (!cap) {
- if (new_cap) {
- cap = new_cap;
- new_cap = NULL;
- } else {
- spin_unlock(&inode->i_lock);
- new_cap = get_cap(mdsc, caps_reservation);
- if (new_cap == NULL)
- return -ENOMEM;
- goto retry;
- }
+ cap = *new_cap;
+ *new_cap = NULL;
cap->issued = 0;
cap->implemented = 0;
cap->mds = mds;
cap->mds_wanted = 0;
+ cap->mseq = 0;
cap->ci = ci;
__insert_cap_node(ci, cap);
- /* clear out old exporting info? (i.e. on cap import) */
- if (ci->i_cap_exporting_mds == mds) {
- ci->i_cap_exporting_issued = 0;
- ci->i_cap_exporting_mseq = 0;
- ci->i_cap_exporting_mds = -1;
- }
-
/* add to session cap list */
cap->session = session;
spin_lock(&session->s_cap_lock);
list_add_tail(&cap->session_caps, &session->s_caps);
session->s_nr_caps++;
spin_unlock(&session->s_cap_lock);
+ } else {
+ /*
+ * auth mds of the inode changed. we received the cap export
+ * message, but still haven't received the cap import message.
+ * handle_cap_export() updated the new auth MDS' cap.
+ *
+ * "ceph_seq_cmp(seq, cap->seq) <= 0" means we are processing
+ * a message that was send before the cap import message. So
+ * don't remove caps.
+ */
+ if (ceph_seq_cmp(seq, cap->seq) <= 0) {
+ WARN_ON(cap != ci->i_auth_cap);
+ WARN_ON(cap->cap_id != cap_id);
+ seq = cap->seq;
+ mseq = cap->mseq;
+ issued |= cap->issued;
+ flags |= CEPH_CAP_FLAG_AUTH;
+ }
}
if (!ci->i_snap_realm) {
@@ -607,10 +606,15 @@ retry:
__cap_delay_requeue(mdsc, ci);
}
- if (flags & CEPH_CAP_FLAG_AUTH)
- ci->i_auth_cap = cap;
- else if (ci->i_auth_cap == cap)
- ci->i_auth_cap = NULL;
+ if (flags & CEPH_CAP_FLAG_AUTH) {
+ if (ci->i_auth_cap == NULL ||
+ ceph_seq_cmp(ci->i_auth_cap->mseq, mseq) < 0) {
+ ci->i_auth_cap = cap;
+ cap->mds_wanted = wanted;
+ }
+ } else {
+ WARN_ON(ci->i_auth_cap == cap);
+ }
dout("add_cap inode %p (%llx.%llx) cap %p %s now %s seq %d mds%d\n",
inode, ceph_vinop(inode), cap, ceph_cap_string(issued),
@@ -618,7 +622,10 @@ retry:
cap->cap_id = cap_id;
cap->issued = issued;
cap->implemented |= issued;
- cap->mds_wanted |= wanted;
+ if (ceph_seq_cmp(mseq, cap->mseq) > 0)
+ cap->mds_wanted = wanted;
+ else
+ cap->mds_wanted |= wanted;
cap->seq = seq;
cap->issue_seq = seq;
cap->mseq = mseq;
@@ -626,9 +633,6 @@ retry:
if (fmode >= 0)
__ceph_get_fmode(ci, fmode);
- spin_unlock(&inode->i_lock);
- wake_up_all(&ci->i_cap_wq);
- return 0;
}
/*
@@ -641,10 +645,10 @@ static int __cap_is_valid(struct ceph_cap *cap)
unsigned long ttl;
u32 gen;
- spin_lock(&cap->session->s_cap_lock);
+ spin_lock(&cap->session->s_gen_ttl_lock);
gen = cap->session->s_cap_gen;
ttl = cap->session->s_cap_ttl;
- spin_unlock(&cap->session->s_cap_lock);
+ spin_unlock(&cap->session->s_gen_ttl_lock);
if (cap->cap_gen < gen || time_after_eq(jiffies, ttl)) {
dout("__cap_is_valid %p cap %p issued %s "
@@ -663,7 +667,7 @@ static int __cap_is_valid(struct ceph_cap *cap)
*/
int __ceph_caps_issued(struct ceph_inode_info *ci, int *implemented)
{
- int have = ci->i_snap_caps | ci->i_cap_exporting_issued;
+ int have = ci->i_snap_caps;
struct ceph_cap *cap;
struct rb_node *p;
@@ -679,6 +683,15 @@ int __ceph_caps_issued(struct ceph_inode_info *ci, int *implemented)
if (implemented)
*implemented |= cap->implemented;
}
+ /*
+ * exclude caps issued by non-auth MDS, but are been revoking
+ * by the auth MDS. The non-auth MDS should be revoking/exporting
+ * these caps, but the message is delayed.
+ */
+ if (ci->i_auth_cap) {
+ cap = ci->i_auth_cap;
+ have &= ~cap->implemented | cap->issued;
+ }
return have;
}
@@ -765,7 +778,7 @@ int __ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask, int touch)
if (touch) {
struct rb_node *q;
- /* touch this + preceeding caps */
+ /* touch this + preceding caps */
__touch_cap(cap);
for (q = rb_first(&ci->i_caps); q != p;
q = rb_next(q)) {
@@ -786,23 +799,29 @@ int __ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask, int touch)
/*
* Return true if mask caps are currently being revoked by an MDS.
*/
-int ceph_caps_revoking(struct ceph_inode_info *ci, int mask)
+int __ceph_caps_revoking_other(struct ceph_inode_info *ci,
+ struct ceph_cap *ocap, int mask)
{
- struct inode *inode = &ci->vfs_inode;
struct ceph_cap *cap;
struct rb_node *p;
- int ret = 0;
- spin_lock(&inode->i_lock);
for (p = rb_first(&ci->i_caps); p; p = rb_next(p)) {
cap = rb_entry(p, struct ceph_cap, ci_node);
- if (__cap_is_valid(cap) &&
- (cap->implemented & ~cap->issued & mask)) {
- ret = 1;
- break;
- }
+ if (cap != ocap &&
+ (cap->implemented & ~cap->issued & mask))
+ return 1;
}
- spin_unlock(&inode->i_lock);
+ return 0;
+}
+
+int ceph_caps_revoking(struct ceph_inode_info *ci, int mask)
+{
+ struct inode *inode = &ci->vfs_inode;
+ int ret;
+
+ spin_lock(&ci->i_ceph_lock);
+ ret = __ceph_caps_revoking_other(ci, NULL, mask);
+ spin_unlock(&ci->i_ceph_lock);
dout("ceph_caps_revoking %p %s = %d\n", inode,
ceph_cap_string(mask), ret);
return ret;
@@ -819,7 +838,7 @@ int __ceph_caps_used(struct ceph_inode_info *ci)
used |= CEPH_CAP_FILE_CACHE;
if (ci->i_wr_ref)
used |= CEPH_CAP_FILE_WR;
- if (ci->i_wrbuffer_ref)
+ if (ci->i_wb_ref || ci->i_wrbuffer_ref)
used |= CEPH_CAP_FILE_BUFFER;
return used;
}
@@ -850,26 +869,41 @@ int __ceph_caps_mds_wanted(struct ceph_inode_info *ci)
cap = rb_entry(p, struct ceph_cap, ci_node);
if (!__cap_is_valid(cap))
continue;
- mds_wanted |= cap->mds_wanted;
+ if (cap == ci->i_auth_cap)
+ mds_wanted |= cap->mds_wanted;
+ else
+ mds_wanted |= (cap->mds_wanted & ~CEPH_CAP_ANY_FILE_WR);
}
return mds_wanted;
}
/*
- * called under i_lock
+ * called under i_ceph_lock
*/
static int __ceph_is_any_caps(struct ceph_inode_info *ci)
{
- return !RB_EMPTY_ROOT(&ci->i_caps) || ci->i_cap_exporting_mds >= 0;
+ return !RB_EMPTY_ROOT(&ci->i_caps);
+}
+
+int ceph_is_any_caps(struct inode *inode)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ int ret;
+
+ spin_lock(&ci->i_ceph_lock);
+ ret = __ceph_is_any_caps(ci);
+ spin_unlock(&ci->i_ceph_lock);
+
+ return ret;
}
/*
* Remove a cap. Take steps to deal with a racing iterate_session_caps.
*
- * caller should hold i_lock.
+ * caller should hold i_ceph_lock.
* caller will not hold session s_mutex if called from destroy_inode.
*/
-void __ceph_remove_cap(struct ceph_cap *cap)
+void __ceph_remove_cap(struct ceph_cap *cap, bool queue_release)
{
struct ceph_mds_session *session = cap->session;
struct ceph_inode_info *ci = cap->ci;
@@ -881,6 +915,16 @@ void __ceph_remove_cap(struct ceph_cap *cap)
/* remove from session list */
spin_lock(&session->s_cap_lock);
+ /*
+ * s_cap_reconnect is protected by s_cap_lock. no one changes
+ * s_cap_gen while session is in the reconnect state.
+ */
+ if (queue_release &&
+ (!session->s_cap_reconnect ||
+ cap->cap_gen == session->s_cap_gen))
+ __queue_cap_release(session, ci->i_vino.ino, cap->cap_id,
+ cap->mseq, cap->issue_seq);
+
if (session->s_cap_iterator == cap) {
/* not yet, we are iterating over this very cap */
dout("__ceph_remove_cap delaying %p removal from session %p\n",
@@ -928,7 +972,7 @@ static int send_cap_msg(struct ceph_mds_session *session,
u64 size, u64 max_size,
struct timespec *mtime, struct timespec *atime,
u64 time_warp_seq,
- uid_t uid, gid_t gid, mode_t mode,
+ kuid_t uid, kgid_t gid, umode_t mode,
u64 xattr_version,
struct ceph_buffer *xattrs_buf,
u64 follows)
@@ -944,7 +988,7 @@ static int send_cap_msg(struct ceph_mds_session *session,
seq, issue_seq, mseq, follows, size, max_size,
xattr_version, xattrs_buf ? (int)xattrs_buf->vec.iov_len : 0);
- msg = ceph_msg_new(CEPH_MSG_CLIENT_CAPS, sizeof(*fc), GFP_NOFS);
+ msg = ceph_msg_new(CEPH_MSG_CLIENT_CAPS, sizeof(*fc), GFP_NOFS, false);
if (!msg)
return -ENOMEM;
@@ -972,8 +1016,8 @@ static int send_cap_msg(struct ceph_mds_session *session,
ceph_encode_timespec(&fc->atime, atime);
fc->time_warp_seq = cpu_to_le32(time_warp_seq);
- fc->uid = cpu_to_le32(uid);
- fc->gid = cpu_to_le32(gid);
+ fc->uid = cpu_to_le32(from_kuid(&init_user_ns, uid));
+ fc->gid = cpu_to_le32(from_kgid(&init_user_ns, gid));
fc->mode = cpu_to_le32(mode);
fc->xattr_version = cpu_to_le64(xattr_version);
@@ -987,15 +1031,14 @@ static int send_cap_msg(struct ceph_mds_session *session,
return 0;
}
-static void __queue_cap_release(struct ceph_mds_session *session,
- u64 ino, u64 cap_id, u32 migrate_seq,
- u32 issue_seq)
+void __queue_cap_release(struct ceph_mds_session *session,
+ u64 ino, u64 cap_id, u32 migrate_seq,
+ u32 issue_seq)
{
struct ceph_msg *msg;
struct ceph_mds_cap_release *head;
struct ceph_mds_cap_item *item;
- spin_lock(&session->s_cap_lock);
BUG_ON(!session->s_num_cap_releases);
msg = list_first_entry(&session->s_cap_releases,
struct ceph_msg, list_head);
@@ -1005,7 +1048,7 @@ static void __queue_cap_release(struct ceph_mds_session *session,
BUG_ON(msg->front.iov_len + sizeof(*item) > PAGE_CACHE_SIZE);
head = msg->front.iov_base;
- head->num = cpu_to_le32(le32_to_cpu(head->num) + 1);
+ le32_add_cpu(&head->num, 1);
item = msg->front.iov_base + msg->front.iov_len;
item->ino = cpu_to_le64(ino);
item->cap_id = cpu_to_le64(cap_id);
@@ -1024,12 +1067,11 @@ static void __queue_cap_release(struct ceph_mds_session *session,
(int)CEPH_CAPS_PER_RELEASE,
(int)msg->front.iov_len);
}
- spin_unlock(&session->s_cap_lock);
}
/*
* Queue cap releases when an inode is dropped from our cache. Since
- * inode is about to be destroyed, there is no need for i_lock.
+ * inode is about to be destroyed, there is no need for i_ceph_lock.
*/
void ceph_queue_caps_release(struct inode *inode)
{
@@ -1039,18 +1081,14 @@ void ceph_queue_caps_release(struct inode *inode)
p = rb_first(&ci->i_caps);
while (p) {
struct ceph_cap *cap = rb_entry(p, struct ceph_cap, ci_node);
- struct ceph_mds_session *session = cap->session;
-
- __queue_cap_release(session, ceph_ino(inode), cap->cap_id,
- cap->mseq, cap->issue_seq);
p = rb_next(p);
- __ceph_remove_cap(cap);
+ __ceph_remove_cap(cap, true);
}
}
/*
* Send a cap msg on the given inode. Update our caps state, then
- * drop i_lock and send the message.
+ * drop i_ceph_lock and send the message.
*
* Make note of max_size reported/requested from mds, revoked caps
* that have now been implemented.
@@ -1062,13 +1100,13 @@ void ceph_queue_caps_release(struct inode *inode)
* Return non-zero if delayed release, or we experienced an error
* such that the caller should requeue + retry later.
*
- * called with i_lock, then drops it.
+ * called with i_ceph_lock, then drops it.
* caller should hold snap_rwsem (read), s_mutex.
*/
static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
int op, int used, int want, int retain, int flushing,
unsigned *pflush_tid)
- __releases(cap->ci->vfs_inode->i_lock)
+ __releases(cap->ci->i_ceph_lock)
{
struct ceph_inode_info *ci = cap->ci;
struct inode *inode = &ci->vfs_inode;
@@ -1078,9 +1116,9 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
u64 size, max_size;
struct timespec mtime, atime;
int wake = 0;
- mode_t mode;
- uid_t uid;
- gid_t gid;
+ umode_t mode;
+ kuid_t uid;
+ kgid_t gid;
struct ceph_mds_session *session;
u64 xattr_version = 0;
struct ceph_buffer *xattr_blob = NULL;
@@ -1171,7 +1209,7 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
xattr_version = ci->i_xattrs.version;
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
ret = send_cap_msg(session, ceph_vino(inode).ino, cap_id,
op, keep, want, flushing, seq, flush_tid, issue_seq, mseq,
@@ -1199,13 +1237,13 @@ static int __send_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap,
* Unless @again is true, skip cap_snaps that were already sent to
* the MDS (i.e., during this session).
*
- * Called under i_lock. Takes s_mutex as needed.
+ * Called under i_ceph_lock. Takes s_mutex as needed.
*/
void __ceph_flush_snaps(struct ceph_inode_info *ci,
struct ceph_mds_session **psession,
int again)
- __releases(ci->vfs_inode->i_lock)
- __acquires(ci->vfs_inode->i_lock)
+ __releases(ci->i_ceph_lock)
+ __acquires(ci->i_ceph_lock)
{
struct inode *inode = &ci->vfs_inode;
int mds;
@@ -1262,7 +1300,7 @@ retry:
session = NULL;
}
if (!session) {
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
mutex_lock(&mdsc->mutex);
session = __ceph_lookup_mds_session(mdsc, mds);
mutex_unlock(&mdsc->mutex);
@@ -1276,7 +1314,7 @@ retry:
* deletion or migration. retry, and we'll
* get a better @mds value next time.
*/
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
goto retry;
}
@@ -1286,7 +1324,7 @@ retry:
list_del_init(&capsnap->flushing_item);
list_add_tail(&capsnap->flushing_item,
&session->s_cap_snaps_flushing);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
dout("flush_snaps %p cap_snap %p follows %lld tid %llu\n",
inode, capsnap, capsnap->follows, capsnap->flush_tid);
@@ -1303,7 +1341,7 @@ retry:
next_follows = capsnap->follows + 1;
ceph_put_cap_snap(capsnap);
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
goto retry;
}
@@ -1323,18 +1361,17 @@ out:
static void ceph_flush_snaps(struct ceph_inode_info *ci)
{
- struct inode *inode = &ci->vfs_inode;
-
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
__ceph_flush_snaps(ci, NULL, 0);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
}
/*
- * Mark caps dirty. If inode is newly dirty, add to the global dirty
- * list.
+ * Mark caps dirty. If inode is newly dirty, return the dirty flags.
+ * Caller is then responsible for calling __mark_inode_dirty with the
+ * returned flags value.
*/
-void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
+int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
{
struct ceph_mds_client *mdsc =
ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
@@ -1350,14 +1387,15 @@ void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
if (!ci->i_head_snapc)
ci->i_head_snapc = ceph_get_snap_context(
ci->i_snap_realm->cached_context);
- dout(" inode %p now dirty snapc %p\n", &ci->vfs_inode,
- ci->i_head_snapc);
+ dout(" inode %p now dirty snapc %p auth cap %p\n",
+ &ci->vfs_inode, ci->i_head_snapc, ci->i_auth_cap);
+ WARN_ON(!ci->i_auth_cap);
BUG_ON(!list_empty(&ci->i_dirty_item));
spin_lock(&mdsc->cap_dirty_lock);
list_add(&ci->i_dirty_item, &mdsc->cap_dirty);
spin_unlock(&mdsc->cap_dirty_lock);
if (ci->i_flushing_caps == 0) {
- igrab(inode);
+ ihold(inode);
dirty |= I_DIRTY_SYNC;
}
}
@@ -1365,16 +1403,15 @@ void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
if (((was | ci->i_flushing_caps) & CEPH_CAP_FILE_BUFFER) &&
(mask & CEPH_CAP_FILE_BUFFER))
dirty |= I_DIRTY_DATASYNC;
- if (dirty)
- __mark_inode_dirty(inode, dirty);
__cap_delay_requeue(mdsc, ci);
+ return dirty;
}
/*
* Add dirty inode to the flushing list. Assigned a seq number so we
* can wait for caps to flush without starving.
*
- * Called under i_lock.
+ * Called under i_ceph_lock.
*/
static int __mark_caps_flushing(struct inode *inode,
struct ceph_mds_session *session)
@@ -1422,16 +1459,16 @@ static int try_nonblocking_invalidate(struct inode *inode)
struct ceph_inode_info *ci = ceph_inode(inode);
u32 invalidating_gen = ci->i_rdcache_gen;
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
invalidate_mapping_pages(&inode->i_data, 0, -1);
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (inode->i_data.nrpages == 0 &&
invalidating_gen == ci->i_rdcache_gen) {
/* success. */
dout("try_nonblocking_invalidate %p success\n", inode);
- ci->i_rdcache_gen = 0;
- ci->i_rdcache_revoking = 0;
+ /* save any racing async invalidate some trouble */
+ ci->i_rdcache_revoking = ci->i_rdcache_gen - 1;
return 0;
}
dout("try_nonblocking_invalidate %p failed\n", inode);
@@ -1456,7 +1493,7 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags,
struct ceph_mds_client *mdsc = fsc->mdsc;
struct inode *inode = &ci->vfs_inode;
struct ceph_cap *cap;
- int file_wanted, used;
+ int file_wanted, used, cap_used;
int took_snap_rwsem = 0; /* true if mdsc->snap_rwsem held */
int issued, implemented, want, retain, revoking, flushing = 0;
int mds = -1; /* keep track of how far we've gone through i_caps list
@@ -1471,7 +1508,7 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags,
if (mdsc->stopping)
is_delayed = 1;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (ci->i_ceph_flags & CEPH_I_FLUSH)
flags |= CHECK_CAPS_FLUSH;
@@ -1481,7 +1518,7 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags,
__ceph_flush_snaps(ci, &session, 0);
goto retry_locked;
retry:
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
retry_locked:
file_wanted = __ceph_caps_file_wanted(ci);
used = __ceph_caps_used(ci);
@@ -1559,10 +1596,16 @@ retry_locked:
/* NOTE: no side-effects allowed, until we take s_mutex */
+ cap_used = used;
+ if (ci->i_auth_cap && cap != ci->i_auth_cap)
+ cap_used &= ~ci->i_auth_cap->issued;
+
revoking = cap->implemented & ~cap->issued;
- if (revoking)
- dout(" mds%d revoking %s\n", cap->mds,
- ceph_cap_string(revoking));
+ dout(" mds%d cap %p used %s issued %s implemented %s revoking %s\n",
+ cap->mds, cap, ceph_cap_string(cap->issued),
+ ceph_cap_string(cap_used),
+ ceph_cap_string(cap->implemented),
+ ceph_cap_string(revoking));
if (cap == ci->i_auth_cap &&
(cap->issued & CEPH_CAP_FILE_WR)) {
@@ -1588,7 +1631,7 @@ retry_locked:
}
/* completed revocation? going down and there are no caps? */
- if (revoking && (revoking & used) == 0) {
+ if (revoking && (revoking & cap_used) == 0) {
dout("completed revocation of %s\n",
ceph_cap_string(cap->implemented & ~cap->issued));
goto ack;
@@ -1634,7 +1677,7 @@ ack:
if (mutex_trylock(&session->s_mutex) == 0) {
dout("inverting session/ino locks on %p\n",
session);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
if (took_snap_rwsem) {
up_read(&mdsc->snap_rwsem);
took_snap_rwsem = 0;
@@ -1648,7 +1691,7 @@ ack:
if (down_read_trylock(&mdsc->snap_rwsem) == 0) {
dout("inverting snap/in locks on %p\n",
inode);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
down_read(&mdsc->snap_rwsem);
took_snap_rwsem = 1;
goto retry;
@@ -1658,14 +1701,16 @@ ack:
if (cap == ci->i_auth_cap && ci->i_dirty_caps)
flushing = __mark_caps_flushing(inode, session);
+ else
+ flushing = 0;
mds = cap->mds; /* remember mds, so we don't repeat */
sent++;
- /* __send_cap drops i_lock */
- delayed += __send_cap(mdsc, cap, CEPH_CAP_OP_UPDATE, used, want,
- retain, flushing, NULL);
- goto retry; /* retake i_lock and restart our cap scan. */
+ /* __send_cap drops i_ceph_lock */
+ delayed += __send_cap(mdsc, cap, CEPH_CAP_OP_UPDATE, cap_used,
+ want, retain, flushing, NULL);
+ goto retry; /* retake i_ceph_lock and restart our cap scan. */
}
/*
@@ -1679,7 +1724,7 @@ ack:
else if (!is_delayed || force_requeue)
__cap_delay_requeue(mdsc, ci);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
if (queue_invalidate)
ceph_queue_invalidate(inode);
@@ -1693,16 +1738,15 @@ ack:
/*
* Try to flush dirty caps back to the auth mds.
*/
-static int try_flush_caps(struct inode *inode, struct ceph_mds_session *session,
- unsigned *flush_tid)
+static int try_flush_caps(struct inode *inode, unsigned *flush_tid)
{
struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
struct ceph_inode_info *ci = ceph_inode(inode);
- int unlock_session = session ? 0 : 1;
int flushing = 0;
+ struct ceph_mds_session *session = NULL;
retry:
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (ci->i_ceph_flags & CEPH_I_NOFLUSH) {
dout("try_flush_caps skipping %p I_NOFLUSH set\n", inode);
goto out;
@@ -1713,32 +1757,33 @@ retry:
int want = __ceph_caps_wanted(ci);
int delayed;
- if (!session) {
- spin_unlock(&inode->i_lock);
+ if (!session || session != cap->session) {
+ spin_unlock(&ci->i_ceph_lock);
+ if (session)
+ mutex_unlock(&session->s_mutex);
session = cap->session;
mutex_lock(&session->s_mutex);
goto retry;
}
- BUG_ON(session != cap->session);
if (cap->session->s_state < CEPH_MDS_SESSION_OPEN)
goto out;
flushing = __mark_caps_flushing(inode, session);
- /* __send_cap drops i_lock */
+ /* __send_cap drops i_ceph_lock */
delayed = __send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH, used, want,
cap->issued | cap->implemented, flushing,
flush_tid);
if (!delayed)
goto out_unlocked;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
__cap_delay_requeue(mdsc, ci);
}
out:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
out_unlocked:
- if (session && unlock_session)
+ if (session)
mutex_unlock(&session->s_mutex);
return flushing;
}
@@ -1751,7 +1796,7 @@ static int caps_are_flushed(struct inode *inode, unsigned tid)
struct ceph_inode_info *ci = ceph_inode(inode);
int i, ret = 1;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
for (i = 0; i < CEPH_CAP_BITS; i++)
if ((ci->i_flushing_caps & (1 << i)) &&
ci->i_cap_flush_tid[i] <= tid) {
@@ -1759,7 +1804,7 @@ static int caps_are_flushed(struct inode *inode, unsigned tid)
ret = 0;
break;
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return ret;
}
@@ -1807,7 +1852,7 @@ out:
spin_unlock(&ci->i_unsafe_lock);
}
-int ceph_fsync(struct file *file, int datasync)
+int ceph_fsync(struct file *file, loff_t start, loff_t end, int datasync)
{
struct inode *inode = file->f_mapping->host;
struct ceph_inode_info *ci = ceph_inode(inode);
@@ -1818,11 +1863,12 @@ int ceph_fsync(struct file *file, int datasync)
dout("fsync %p%s\n", inode, datasync ? " datasync" : "");
sync_write_wait(inode);
- ret = filemap_write_and_wait(inode->i_mapping);
+ ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
if (ret < 0)
return ret;
+ mutex_lock(&inode->i_mutex);
- dirty = try_flush_caps(inode, NULL, &flush_tid);
+ dirty = try_flush_caps(inode, &flush_tid);
dout("fsync dirty caps are %s\n", ceph_cap_string(dirty));
/*
@@ -1837,6 +1883,7 @@ int ceph_fsync(struct file *file, int datasync)
}
dout("fsync %p%s done\n", inode, datasync ? " datasync" : "");
+ mutex_unlock(&inode->i_mutex);
return ret;
}
@@ -1856,7 +1903,7 @@ int ceph_write_inode(struct inode *inode, struct writeback_control *wbc)
dout("write_inode %p wait=%d\n", inode, wait);
if (wait) {
- dirty = try_flush_caps(inode, NULL, &flush_tid);
+ dirty = try_flush_caps(inode, &flush_tid);
if (dirty)
err = wait_event_interruptible(ci->i_cap_wq,
caps_are_flushed(inode, flush_tid));
@@ -1864,10 +1911,10 @@ int ceph_write_inode(struct inode *inode, struct writeback_control *wbc)
struct ceph_mds_client *mdsc =
ceph_sb_to_client(inode->i_sb)->mdsc;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (__ceph_caps_dirty(ci))
__cap_delay_requeue_front(mdsc, ci);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
}
return err;
}
@@ -1890,7 +1937,7 @@ static void kick_flushing_capsnaps(struct ceph_mds_client *mdsc,
struct inode *inode = &ci->vfs_inode;
struct ceph_cap *cap;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
cap = ci->i_auth_cap;
if (cap && cap->session == session) {
dout("kick_flushing_caps %p cap %p capsnap %p\n", inode,
@@ -1900,7 +1947,7 @@ static void kick_flushing_capsnaps(struct ceph_mds_client *mdsc,
pr_err("%p auth cap %p not mds%d ???\n", inode,
cap, session->s_mds);
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
}
}
@@ -1917,7 +1964,7 @@ void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc,
struct ceph_cap *cap;
int delayed = 0;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
cap = ci->i_auth_cap;
if (cap && cap->session == session) {
dout("kick_flushing_caps %p cap %p %s\n", inode,
@@ -1928,15 +1975,51 @@ void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc,
cap->issued | cap->implemented,
ci->i_flushing_caps, NULL);
if (delayed) {
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
__cap_delay_requeue(mdsc, ci);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
}
} else {
pr_err("%p auth cap %p not mds%d ???\n", inode,
cap, session->s_mds);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
+ }
+ }
+}
+
+static void kick_flushing_inode_caps(struct ceph_mds_client *mdsc,
+ struct ceph_mds_session *session,
+ struct inode *inode)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_cap *cap;
+ int delayed = 0;
+
+ spin_lock(&ci->i_ceph_lock);
+ cap = ci->i_auth_cap;
+ dout("kick_flushing_inode_caps %p flushing %s flush_seq %lld\n", inode,
+ ceph_cap_string(ci->i_flushing_caps), ci->i_cap_flush_seq);
+
+ __ceph_flush_snaps(ci, &session, 1);
+
+ if (ci->i_flushing_caps) {
+ spin_lock(&mdsc->cap_dirty_lock);
+ list_move_tail(&ci->i_flushing_item,
+ &cap->session->s_cap_flushing);
+ spin_unlock(&mdsc->cap_dirty_lock);
+
+ delayed = __send_cap(mdsc, cap, CEPH_CAP_OP_FLUSH,
+ __ceph_caps_used(ci),
+ __ceph_caps_wanted(ci),
+ cap->issued | cap->implemented,
+ ci->i_flushing_caps, NULL);
+ if (delayed) {
+ spin_lock(&ci->i_ceph_lock);
+ __cap_delay_requeue(mdsc, ci);
+ spin_unlock(&ci->i_ceph_lock);
}
+ } else {
+ spin_unlock(&ci->i_ceph_lock);
}
}
@@ -1945,7 +2028,7 @@ void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc,
* Take references to capabilities we hold, so that we don't release
* them to the MDS prematurely.
*
- * Protected by i_lock.
+ * Protected by i_ceph_lock.
*/
static void __take_cap_refs(struct ceph_inode_info *ci, int got)
{
@@ -1958,11 +2041,11 @@ static void __take_cap_refs(struct ceph_inode_info *ci, int got)
if (got & CEPH_CAP_FILE_WR)
ci->i_wr_ref++;
if (got & CEPH_CAP_FILE_BUFFER) {
- if (ci->i_wrbuffer_ref == 0)
- igrab(&ci->vfs_inode);
- ci->i_wrbuffer_ref++;
- dout("__take_cap_refs %p wrbuffer %d -> %d (?)\n",
- &ci->vfs_inode, ci->i_wrbuffer_ref-1, ci->i_wrbuffer_ref);
+ if (ci->i_wb_ref == 0)
+ ihold(&ci->vfs_inode);
+ ci->i_wb_ref++;
+ dout("__take_cap_refs %p wb %d -> %d (?)\n",
+ &ci->vfs_inode, ci->i_wb_ref-1, ci->i_wb_ref);
}
}
@@ -1983,7 +2066,7 @@ static int try_get_cap_refs(struct ceph_inode_info *ci, int need, int want,
dout("get_cap_refs %p need %s want %s\n", inode,
ceph_cap_string(need), ceph_cap_string(want));
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
/* make sure file is actually open */
file_wanted = __ceph_caps_file_wanted(ci);
@@ -1995,11 +2078,20 @@ static int try_get_cap_refs(struct ceph_inode_info *ci, int need, int want,
goto out;
}
- if (need & CEPH_CAP_FILE_WR) {
+ /* finish pending truncate */
+ while (ci->i_truncate_pending) {
+ spin_unlock(&ci->i_ceph_lock);
+ __ceph_do_pending_vmtruncate(inode);
+ spin_lock(&ci->i_ceph_lock);
+ }
+
+ have = __ceph_caps_issued(ci, &implemented);
+
+ if (have & need & CEPH_CAP_FILE_WR) {
if (endoff >= 0 && endoff > (loff_t)ci->i_max_size) {
dout("get_cap_refs %p endoff %llu > maxsize %llu\n",
inode, endoff, ci->i_max_size);
- if (endoff > ci->i_wanted_max_size) {
+ if (endoff > ci->i_requested_max_size) {
*check_max = 1;
ret = 1;
}
@@ -2014,13 +2106,6 @@ static int try_get_cap_refs(struct ceph_inode_info *ci, int need, int want,
goto out;
}
}
- have = __ceph_caps_issued(ci, &implemented);
-
- /*
- * disallow writes while a truncate is pending
- */
- if (ci->i_truncate_pending)
- have &= ~CEPH_CAP_FILE_WR;
if ((have & need) == need) {
/*
@@ -2044,7 +2129,7 @@ static int try_get_cap_refs(struct ceph_inode_info *ci, int need, int want,
ceph_cap_string(have), ceph_cap_string(need));
}
out:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
dout("get_cap_refs %p ret %d got %s\n", inode,
ret, ceph_cap_string(*got));
return ret;
@@ -2061,16 +2146,19 @@ static void check_max_size(struct inode *inode, loff_t endoff)
int check = 0;
/* do we need to explicitly request a larger max_size? */
- spin_lock(&inode->i_lock);
- if ((endoff >= ci->i_max_size ||
- endoff > (inode->i_size << 1)) &&
- endoff > ci->i_wanted_max_size) {
+ spin_lock(&ci->i_ceph_lock);
+ if (endoff >= ci->i_max_size && endoff > ci->i_wanted_max_size) {
dout("write %p at large endoff %llu, req max_size\n",
inode, endoff);
ci->i_wanted_max_size = endoff;
- check = 1;
}
- spin_unlock(&inode->i_lock);
+ /* duplicate ceph_check_caps()'s logic */
+ if (ci->i_auth_cap &&
+ (ci->i_auth_cap->issued & CEPH_CAP_FILE_WR) &&
+ ci->i_wanted_max_size > ci->i_max_size &&
+ ci->i_wanted_max_size > ci->i_requested_max_size)
+ check = 1;
+ spin_unlock(&ci->i_ceph_lock);
if (check)
ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
}
@@ -2107,9 +2195,9 @@ retry:
*/
void ceph_get_cap_refs(struct ceph_inode_info *ci, int caps)
{
- spin_lock(&ci->vfs_inode.i_lock);
+ spin_lock(&ci->i_ceph_lock);
__take_cap_refs(ci, caps);
- spin_unlock(&ci->vfs_inode.i_lock);
+ spin_unlock(&ci->i_ceph_lock);
}
/*
@@ -2127,7 +2215,7 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
int last = 0, put = 0, flushsnaps = 0, wake = 0;
struct ceph_cap_snap *capsnap;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (had & CEPH_CAP_PIN)
--ci->i_pin_ref;
if (had & CEPH_CAP_FILE_RD)
@@ -2137,12 +2225,12 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
if (--ci->i_rdcache_ref == 0)
last++;
if (had & CEPH_CAP_FILE_BUFFER) {
- if (--ci->i_wrbuffer_ref == 0) {
+ if (--ci->i_wb_ref == 0) {
last++;
put++;
}
- dout("put_cap_refs %p wrbuffer %d -> %d (?)\n",
- inode, ci->i_wrbuffer_ref+1, ci->i_wrbuffer_ref);
+ dout("put_cap_refs %p wb %d -> %d (?)\n",
+ inode, ci->i_wb_ref+1, ci->i_wb_ref);
}
if (had & CEPH_CAP_FILE_WR)
if (--ci->i_wr_ref == 0) {
@@ -2160,7 +2248,7 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
}
}
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
dout("put_cap_refs %p had %s%s%s\n", inode, ceph_cap_string(had),
last ? " last" : "", put ? " put" : "");
@@ -2192,7 +2280,7 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
int found = 0;
struct ceph_cap_snap *capsnap = NULL;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
ci->i_wrbuffer_ref -= nr;
last = !ci->i_wrbuffer_ref;
@@ -2241,7 +2329,7 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
}
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
if (last) {
ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
@@ -2255,42 +2343,88 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
}
/*
+ * Invalidate unlinked inode's aliases, so we can drop the inode ASAP.
+ */
+static void invalidate_aliases(struct inode *inode)
+{
+ struct dentry *dn, *prev = NULL;
+
+ dout("invalidate_aliases inode %p\n", inode);
+ d_prune_aliases(inode);
+ /*
+ * For non-directory inode, d_find_alias() only returns
+ * hashed dentry. After calling d_invalidate(), the
+ * dentry becomes unhashed.
+ *
+ * For directory inode, d_find_alias() can return
+ * unhashed dentry. But directory inode should have
+ * one alias at most.
+ */
+ while ((dn = d_find_alias(inode))) {
+ if (dn == prev) {
+ dput(dn);
+ break;
+ }
+ d_invalidate(dn);
+ if (prev)
+ dput(prev);
+ prev = dn;
+ }
+ if (prev)
+ dput(prev);
+}
+
+/*
* Handle a cap GRANT message from the MDS. (Note that a GRANT may
* actually be a revocation if it specifies a smaller cap set.)
*
- * caller holds s_mutex and i_lock, we drop both.
- *
- * return value:
- * 0 - ok
- * 1 - check_caps on auth cap only (writeback)
- * 2 - check_caps (ack revoke)
+ * caller holds s_mutex and i_ceph_lock, we drop both.
*/
-static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
+static void handle_cap_grant(struct ceph_mds_client *mdsc,
+ struct inode *inode, struct ceph_mds_caps *grant,
+ void *snaptrace, int snaptrace_len,
+ struct ceph_buffer *xattr_buf,
struct ceph_mds_session *session,
- struct ceph_cap *cap,
- struct ceph_buffer *xattr_buf)
- __releases(inode->i_lock)
+ struct ceph_cap *cap, int issued)
+ __releases(ci->i_ceph_lock)
{
struct ceph_inode_info *ci = ceph_inode(inode);
int mds = session->s_mds;
- unsigned seq = le32_to_cpu(grant->seq);
- unsigned issue_seq = le32_to_cpu(grant->issue_seq);
+ int seq = le32_to_cpu(grant->seq);
int newcaps = le32_to_cpu(grant->caps);
- int issued, implemented, used, wanted, dirty;
+ int used, wanted, dirty;
u64 size = le64_to_cpu(grant->size);
u64 max_size = le64_to_cpu(grant->max_size);
struct timespec mtime, atime, ctime;
int check_caps = 0;
- int wake = 0;
- int writeback = 0;
- int revoked_rdcache = 0;
- int queue_invalidate = 0;
-
- dout("handle_cap_grant inode %p cap %p mds%d seq %u/%u %s\n",
- inode, cap, mds, seq, issue_seq, ceph_cap_string(newcaps));
+ bool wake = 0;
+ bool writeback = 0;
+ bool queue_trunc = 0;
+ bool queue_invalidate = 0;
+ bool queue_revalidate = 0;
+ bool deleted_inode = 0;
+
+ dout("handle_cap_grant inode %p cap %p mds%d seq %d %s\n",
+ inode, cap, mds, seq, ceph_cap_string(newcaps));
dout(" size %llu max_size %llu, i_size %llu\n", size, max_size,
inode->i_size);
+
+ /*
+ * auth mds of the inode changed. we received the cap export message,
+ * but still haven't received the cap import message. handle_cap_export
+ * updated the new auth MDS' cap.
+ *
+ * "ceph_seq_cmp(seq, cap->seq) <= 0" means we are processing a message
+ * that was sent before the cap import message. So don't remove caps.
+ */
+ if (ceph_seq_cmp(seq, cap->seq) <= 0) {
+ WARN_ON(cap != ci->i_auth_cap);
+ WARN_ON(cap->cap_id != le64_to_cpu(grant->cap_id));
+ seq = cap->seq;
+ newcaps |= cap->issued;
+ }
+
/*
* If CACHE is being revoked, and we have no dirty buffers,
* try to invalidate (once). (If there are dirty buffers, we
@@ -2299,9 +2433,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
if (((cap->issued & ~newcaps) & CEPH_CAP_FILE_CACHE) &&
(newcaps & CEPH_CAP_FILE_LAZYIO) == 0 &&
!ci->i_wrbuffer_ref) {
- if (try_nonblocking_invalidate(inode) == 0) {
- revoked_rdcache = 1;
- } else {
+ if (try_nonblocking_invalidate(inode)) {
/* there were locked pages.. invalidate later
in a separate thread. */
if (ci->i_rdcache_revoking != ci->i_rdcache_gen) {
@@ -2309,27 +2441,33 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
ci->i_rdcache_revoking = ci->i_rdcache_gen;
}
}
+
+ ceph_fscache_invalidate(inode);
}
/* side effects now are allowed */
-
- issued = __ceph_caps_issued(ci, &implemented);
- issued |= implemented | __ceph_caps_dirty(ci);
-
cap->cap_gen = session->s_cap_gen;
+ cap->seq = seq;
__check_cap_issue(ci, cap, newcaps);
- if ((issued & CEPH_CAP_AUTH_EXCL) == 0) {
+ if ((newcaps & CEPH_CAP_AUTH_SHARED) &&
+ (issued & CEPH_CAP_AUTH_EXCL) == 0) {
inode->i_mode = le32_to_cpu(grant->mode);
- inode->i_uid = le32_to_cpu(grant->uid);
- inode->i_gid = le32_to_cpu(grant->gid);
+ inode->i_uid = make_kuid(&init_user_ns, le32_to_cpu(grant->uid));
+ inode->i_gid = make_kgid(&init_user_ns, le32_to_cpu(grant->gid));
dout("%p mode 0%o uid.gid %d.%d\n", inode, inode->i_mode,
- inode->i_uid, inode->i_gid);
+ from_kuid(&init_user_ns, inode->i_uid),
+ from_kgid(&init_user_ns, inode->i_gid));
}
- if ((issued & CEPH_CAP_LINK_EXCL) == 0)
- inode->i_nlink = le32_to_cpu(grant->nlink);
+ if ((newcaps & CEPH_CAP_AUTH_SHARED) &&
+ (issued & CEPH_CAP_LINK_EXCL) == 0) {
+ set_nlink(inode, le32_to_cpu(grant->nlink));
+ if (inode->i_nlink == 0 &&
+ (newcaps & (CEPH_CAP_LINK_SHARED | CEPH_CAP_LINK_EXCL)))
+ deleted_inode = 1;
+ }
if ((issued & CEPH_CAP_XATTR_EXCL) == 0 && grant->xattr_len) {
int len = le32_to_cpu(grant->xattr_len);
@@ -2342,29 +2480,44 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
ceph_buffer_put(ci->i_xattrs.blob);
ci->i_xattrs.blob = ceph_buffer_get(xattr_buf);
ci->i_xattrs.version = version;
+ ceph_forget_all_cached_acls(inode);
}
}
- /* size/ctime/mtime/atime? */
- ceph_fill_file_size(inode, issued,
- le32_to_cpu(grant->truncate_seq),
- le64_to_cpu(grant->truncate_size), size);
- ceph_decode_timespec(&mtime, &grant->mtime);
- ceph_decode_timespec(&atime, &grant->atime);
- ceph_decode_timespec(&ctime, &grant->ctime);
- ceph_fill_file_time(inode, issued,
- le32_to_cpu(grant->time_warp_seq), &ctime, &mtime,
- &atime);
-
- /* max size increase? */
- if (max_size != ci->i_max_size) {
- dout("max_size %lld -> %llu\n", ci->i_max_size, max_size);
- ci->i_max_size = max_size;
- if (max_size >= ci->i_wanted_max_size) {
- ci->i_wanted_max_size = 0; /* reset */
- ci->i_requested_max_size = 0;
+ /* Do we need to revalidate our fscache cookie. Don't bother on the
+ * first cache cap as we already validate at cookie creation time. */
+ if ((issued & CEPH_CAP_FILE_CACHE) && ci->i_rdcache_gen > 1)
+ queue_revalidate = 1;
+
+ if (newcaps & CEPH_CAP_ANY_RD) {
+ /* ctime/mtime/atime? */
+ ceph_decode_timespec(&mtime, &grant->mtime);
+ ceph_decode_timespec(&atime, &grant->atime);
+ ceph_decode_timespec(&ctime, &grant->ctime);
+ ceph_fill_file_time(inode, issued,
+ le32_to_cpu(grant->time_warp_seq),
+ &ctime, &mtime, &atime);
+ }
+
+ if (newcaps & (CEPH_CAP_ANY_FILE_RD | CEPH_CAP_ANY_FILE_WR)) {
+ /* file layout may have changed */
+ ci->i_layout = grant->layout;
+ /* size/truncate_seq? */
+ queue_trunc = ceph_fill_file_size(inode, issued,
+ le32_to_cpu(grant->truncate_seq),
+ le64_to_cpu(grant->truncate_size),
+ size);
+ /* max size increase? */
+ if (ci->i_auth_cap == cap && max_size != ci->i_max_size) {
+ dout("max_size %lld -> %llu\n",
+ ci->i_max_size, max_size);
+ ci->i_max_size = max_size;
+ if (max_size >= ci->i_wanted_max_size) {
+ ci->i_wanted_max_size = 0; /* reset */
+ ci->i_requested_max_size = 0;
+ }
+ wake = 1;
}
- wake = 1;
}
/* check cap bits */
@@ -2379,15 +2532,11 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
dout("mds wanted %s -> %s\n",
ceph_cap_string(le32_to_cpu(grant->wanted)),
ceph_cap_string(wanted));
- grant->wanted = cpu_to_le32(wanted);
+ /* imported cap may not have correct mds_wanted */
+ if (le32_to_cpu(grant->op) == CEPH_CAP_OP_IMPORT)
+ check_caps = 1;
}
- cap->seq = seq;
- cap->issue_seq = issue_seq;
-
- /* file layout may have changed */
- ci->i_layout = grant->layout;
-
/* revocation, grant, or no-op? */
if (cap->issued & ~newcaps) {
int revoking = cap->issued & ~newcaps;
@@ -2414,6 +2563,11 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
} else {
dout("grant: %s -> %s\n", ceph_cap_string(cap->issued),
ceph_cap_string(newcaps));
+ /* non-auth MDS is revoking the newly grant caps ? */
+ if (cap == ci->i_auth_cap &&
+ __ceph_caps_revoking_other(ci, cap, newcaps))
+ check_caps = 2;
+
cap->issued = newcaps;
cap->implemented |= newcaps; /* add bits only, to
* avoid stepping on a
@@ -2422,7 +2576,25 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
}
BUG_ON(cap->issued & ~cap->implemented);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
+
+ if (le32_to_cpu(grant->op) == CEPH_CAP_OP_IMPORT) {
+ down_write(&mdsc->snap_rwsem);
+ ceph_update_snap_trace(mdsc, snaptrace,
+ snaptrace + snaptrace_len, false);
+ downgrade_write(&mdsc->snap_rwsem);
+ kick_flushing_inode_caps(mdsc, session, inode);
+ up_read(&mdsc->snap_rwsem);
+ if (newcaps & ~issued)
+ wake = 1;
+ }
+
+ if (queue_trunc) {
+ ceph_queue_vmtruncate(inode);
+ ceph_queue_revalidate(inode);
+ } else if (queue_revalidate)
+ ceph_queue_revalidate(inode);
+
if (writeback)
/*
* queue inode for writeback: we can't actually call
@@ -2432,6 +2604,8 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
ceph_queue_writeback(inode);
if (queue_invalidate)
ceph_queue_invalidate(inode);
+ if (deleted_inode)
+ invalidate_aliases(inode);
if (wake)
wake_up_all(&ci->i_cap_wq);
@@ -2452,7 +2626,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid,
struct ceph_mds_caps *m,
struct ceph_mds_session *session,
struct ceph_cap *cap)
- __releases(inode->i_lock)
+ __releases(ci->i_ceph_lock)
{
struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
@@ -2508,7 +2682,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid,
wake_up_all(&ci->i_cap_wq);
out:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
if (drop)
iput(inode);
}
@@ -2531,7 +2705,7 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid,
dout("handle_cap_flushsnap_ack inode %p ci %p mds%d follows %lld\n",
inode, ci, session->s_mds, follows);
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
if (capsnap->follows == follows) {
if (capsnap->flush_tid != flush_tid) {
@@ -2554,7 +2728,7 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid,
capsnap, capsnap->follows);
}
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
if (drop)
iput(inode);
}
@@ -2567,7 +2741,7 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid,
static void handle_cap_trunc(struct inode *inode,
struct ceph_mds_caps *trunc,
struct ceph_mds_session *session)
- __releases(inode->i_lock)
+ __releases(ci->i_ceph_lock)
{
struct ceph_inode_info *ci = ceph_inode(inode);
int mds = session->s_mds;
@@ -2586,10 +2760,12 @@ static void handle_cap_trunc(struct inode *inode,
inode, mds, seq, truncate_size, truncate_seq);
queue_trunc = ceph_fill_file_size(inode, issued,
truncate_seq, truncate_size, size);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
- if (queue_trunc)
+ if (queue_trunc) {
ceph_queue_vmtruncate(inode);
+ ceph_fscache_invalidate(inode);
+ }
}
/*
@@ -2601,96 +2777,200 @@ static void handle_cap_trunc(struct inode *inode,
* caller holds s_mutex
*/
static void handle_cap_export(struct inode *inode, struct ceph_mds_caps *ex,
- struct ceph_mds_session *session,
- int *open_target_sessions)
+ struct ceph_mds_cap_peer *ph,
+ struct ceph_mds_session *session)
{
+ struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
+ struct ceph_mds_session *tsession = NULL;
+ struct ceph_cap *cap, *tcap, *new_cap = NULL;
struct ceph_inode_info *ci = ceph_inode(inode);
- int mds = session->s_mds;
+ u64 t_cap_id;
unsigned mseq = le32_to_cpu(ex->migrate_seq);
- struct ceph_cap *cap = NULL, *t;
- struct rb_node *p;
- int remember = 1;
+ unsigned t_seq, t_mseq;
+ int target, issued;
+ int mds = session->s_mds;
- dout("handle_cap_export inode %p ci %p mds%d mseq %d\n",
- inode, ci, mds, mseq);
+ if (ph) {
+ t_cap_id = le64_to_cpu(ph->cap_id);
+ t_seq = le32_to_cpu(ph->seq);
+ t_mseq = le32_to_cpu(ph->mseq);
+ target = le32_to_cpu(ph->mds);
+ } else {
+ t_cap_id = t_seq = t_mseq = 0;
+ target = -1;
+ }
- spin_lock(&inode->i_lock);
+ dout("handle_cap_export inode %p ci %p mds%d mseq %d target %d\n",
+ inode, ci, mds, mseq, target);
+retry:
+ spin_lock(&ci->i_ceph_lock);
+ cap = __get_cap_for_mds(ci, mds);
+ if (!cap || cap->cap_id != le64_to_cpu(ex->cap_id))
+ goto out_unlock;
- /* make sure we haven't seen a higher mseq */
- for (p = rb_first(&ci->i_caps); p; p = rb_next(p)) {
- t = rb_entry(p, struct ceph_cap, ci_node);
- if (ceph_seq_cmp(t->mseq, mseq) > 0) {
- dout(" higher mseq on cap from mds%d\n",
- t->session->s_mds);
- remember = 0;
+ if (target < 0) {
+ __ceph_remove_cap(cap, false);
+ goto out_unlock;
+ }
+
+ /*
+ * now we know we haven't received the cap import message yet
+ * because the exported cap still exist.
+ */
+
+ issued = cap->issued;
+ WARN_ON(issued != cap->implemented);
+
+ tcap = __get_cap_for_mds(ci, target);
+ if (tcap) {
+ /* already have caps from the target */
+ if (tcap->cap_id != t_cap_id ||
+ ceph_seq_cmp(tcap->seq, t_seq) < 0) {
+ dout(" updating import cap %p mds%d\n", tcap, target);
+ tcap->cap_id = t_cap_id;
+ tcap->seq = t_seq - 1;
+ tcap->issue_seq = t_seq - 1;
+ tcap->mseq = t_mseq;
+ tcap->issued |= issued;
+ tcap->implemented |= issued;
+ if (cap == ci->i_auth_cap)
+ ci->i_auth_cap = tcap;
+ if (ci->i_flushing_caps && ci->i_auth_cap == tcap) {
+ spin_lock(&mdsc->cap_dirty_lock);
+ list_move_tail(&ci->i_flushing_item,
+ &tcap->session->s_cap_flushing);
+ spin_unlock(&mdsc->cap_dirty_lock);
+ }
}
- if (t->session->s_mds == mds)
- cap = t;
+ __ceph_remove_cap(cap, false);
+ goto out_unlock;
+ } else if (tsession) {
+ /* add placeholder for the export tagert */
+ int flag = (cap == ci->i_auth_cap) ? CEPH_CAP_FLAG_AUTH : 0;
+ ceph_add_cap(inode, tsession, t_cap_id, -1, issued, 0,
+ t_seq - 1, t_mseq, (u64)-1, flag, &new_cap);
+
+ __ceph_remove_cap(cap, false);
+ goto out_unlock;
}
- if (cap) {
- if (remember) {
- /* make note */
- ci->i_cap_exporting_mds = mds;
- ci->i_cap_exporting_mseq = mseq;
- ci->i_cap_exporting_issued = cap->issued;
+ spin_unlock(&ci->i_ceph_lock);
+ mutex_unlock(&session->s_mutex);
- /*
- * make sure we have open sessions with all possible
- * export targets, so that we get the matching IMPORT
- */
- *open_target_sessions = 1;
+ /* open target session */
+ tsession = ceph_mdsc_open_export_target_session(mdsc, target);
+ if (!IS_ERR(tsession)) {
+ if (mds > target) {
+ mutex_lock(&session->s_mutex);
+ mutex_lock_nested(&tsession->s_mutex,
+ SINGLE_DEPTH_NESTING);
+ } else {
+ mutex_lock(&tsession->s_mutex);
+ mutex_lock_nested(&session->s_mutex,
+ SINGLE_DEPTH_NESTING);
}
- __ceph_remove_cap(cap);
+ ceph_add_cap_releases(mdsc, tsession);
+ new_cap = ceph_get_cap(mdsc, NULL);
+ } else {
+ WARN_ON(1);
+ tsession = NULL;
+ target = -1;
}
- /* else, we already released it */
+ goto retry;
- spin_unlock(&inode->i_lock);
+out_unlock:
+ spin_unlock(&ci->i_ceph_lock);
+ mutex_unlock(&session->s_mutex);
+ if (tsession) {
+ mutex_unlock(&tsession->s_mutex);
+ ceph_put_mds_session(tsession);
+ }
+ if (new_cap)
+ ceph_put_cap(mdsc, new_cap);
}
/*
- * Handle cap IMPORT. If there are temp bits from an older EXPORT,
- * clean them up.
+ * Handle cap IMPORT.
*
- * caller holds s_mutex.
+ * caller holds s_mutex. acquires i_ceph_lock
*/
static void handle_cap_import(struct ceph_mds_client *mdsc,
struct inode *inode, struct ceph_mds_caps *im,
+ struct ceph_mds_cap_peer *ph,
struct ceph_mds_session *session,
- void *snaptrace, int snaptrace_len)
+ struct ceph_cap **target_cap, int *old_issued)
+ __acquires(ci->i_ceph_lock)
{
struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_cap *cap, *ocap, *new_cap = NULL;
int mds = session->s_mds;
- unsigned issued = le32_to_cpu(im->caps);
+ int issued;
+ unsigned caps = le32_to_cpu(im->caps);
unsigned wanted = le32_to_cpu(im->wanted);
unsigned seq = le32_to_cpu(im->seq);
unsigned mseq = le32_to_cpu(im->migrate_seq);
u64 realmino = le64_to_cpu(im->realm);
u64 cap_id = le64_to_cpu(im->cap_id);
+ u64 p_cap_id;
+ int peer;
+
+ if (ph) {
+ p_cap_id = le64_to_cpu(ph->cap_id);
+ peer = le32_to_cpu(ph->mds);
+ } else {
+ p_cap_id = 0;
+ peer = -1;
+ }
+
+ dout("handle_cap_import inode %p ci %p mds%d mseq %d peer %d\n",
+ inode, ci, mds, mseq, peer);
- if (ci->i_cap_exporting_mds >= 0 &&
- ceph_seq_cmp(ci->i_cap_exporting_mseq, mseq) < 0) {
- dout("handle_cap_import inode %p ci %p mds%d mseq %d"
- " - cleared exporting from mds%d\n",
- inode, ci, mds, mseq,
- ci->i_cap_exporting_mds);
- ci->i_cap_exporting_issued = 0;
- ci->i_cap_exporting_mseq = 0;
- ci->i_cap_exporting_mds = -1;
+retry:
+ spin_lock(&ci->i_ceph_lock);
+ cap = __get_cap_for_mds(ci, mds);
+ if (!cap) {
+ if (!new_cap) {
+ spin_unlock(&ci->i_ceph_lock);
+ new_cap = ceph_get_cap(mdsc, NULL);
+ goto retry;
+ }
+ cap = new_cap;
} else {
- dout("handle_cap_import inode %p ci %p mds%d mseq %d\n",
- inode, ci, mds, mseq);
+ if (new_cap) {
+ ceph_put_cap(mdsc, new_cap);
+ new_cap = NULL;
+ }
+ }
+
+ __ceph_caps_issued(ci, &issued);
+ issued |= __ceph_caps_dirty(ci);
+
+ ceph_add_cap(inode, session, cap_id, -1, caps, wanted, seq, mseq,
+ realmino, CEPH_CAP_FLAG_AUTH, &new_cap);
+
+ ocap = peer >= 0 ? __get_cap_for_mds(ci, peer) : NULL;
+ if (ocap && ocap->cap_id == p_cap_id) {
+ dout(" remove export cap %p mds%d flags %d\n",
+ ocap, peer, ph->flags);
+ if ((ph->flags & CEPH_CAP_FLAG_AUTH) &&
+ (ocap->seq != le32_to_cpu(ph->seq) ||
+ ocap->mseq != le32_to_cpu(ph->mseq))) {
+ pr_err("handle_cap_import: mismatched seq/mseq: "
+ "ino (%llx.%llx) mds%d seq %d mseq %d "
+ "importer mds%d has peer seq %d mseq %d\n",
+ ceph_vinop(inode), peer, ocap->seq,
+ ocap->mseq, mds, le32_to_cpu(ph->seq),
+ le32_to_cpu(ph->mseq));
+ }
+ __ceph_remove_cap(ocap, (ph->flags & CEPH_CAP_FLAG_RELEASE));
}
- down_write(&mdsc->snap_rwsem);
- ceph_update_snap_trace(mdsc, snaptrace, snaptrace+snaptrace_len,
- false);
- downgrade_write(&mdsc->snap_rwsem);
- ceph_add_cap(inode, session, cap_id, -1,
- issued, wanted, seq, mseq, realmino, CEPH_CAP_FLAG_AUTH,
- NULL /* no caps context */);
- try_flush_caps(inode, session, NULL);
- up_read(&mdsc->snap_rwsem);
+ /* make sure we re-request max_size, if necessary */
+ ci->i_wanted_max_size = 0;
+ ci->i_requested_max_size = 0;
+
+ *old_issued = issued;
+ *target_cap = cap;
}
/*
@@ -2705,10 +2985,12 @@ void ceph_handle_caps(struct ceph_mds_session *session,
struct ceph_mds_client *mdsc = session->s_mdsc;
struct super_block *sb = mdsc->fsc->sb;
struct inode *inode;
+ struct ceph_inode_info *ci;
struct ceph_cap *cap;
struct ceph_mds_caps *h;
+ struct ceph_mds_cap_peer *peer = NULL;
int mds = session->s_mds;
- int op;
+ int op, issued;
u32 seq, mseq;
struct ceph_vino vino;
u64 cap_id;
@@ -2717,12 +2999,13 @@ void ceph_handle_caps(struct ceph_mds_session *session,
void *snaptrace;
size_t snaptrace_len;
void *flock;
+ void *end;
u32 flock_len;
- int open_target_sessions = 0;
dout("handle_caps from mds%d\n", mds);
/* decode */
+ end = msg->front.iov_base + msg->front.iov_len;
tid = le64_to_cpu(msg->hdr.tid);
if (msg->front.iov_len < sizeof(*h))
goto bad;
@@ -2740,32 +3023,50 @@ void ceph_handle_caps(struct ceph_mds_session *session,
snaptrace_len = le32_to_cpu(h->snap_trace_len);
if (le16_to_cpu(msg->hdr.version) >= 2) {
- void *p, *end;
-
- p = snaptrace + snaptrace_len;
- end = msg->front.iov_base + msg->front.iov_len;
+ void *p = snaptrace + snaptrace_len;
ceph_decode_32_safe(&p, end, flock_len, bad);
+ if (p + flock_len > end)
+ goto bad;
flock = p;
} else {
flock = NULL;
flock_len = 0;
}
+ if (le16_to_cpu(msg->hdr.version) >= 3) {
+ if (op == CEPH_CAP_OP_IMPORT) {
+ void *p = flock + flock_len;
+ if (p + sizeof(*peer) > end)
+ goto bad;
+ peer = p;
+ } else if (op == CEPH_CAP_OP_EXPORT) {
+ /* recorded in unused fields */
+ peer = (void *)&h->size;
+ }
+ }
+
mutex_lock(&session->s_mutex);
session->s_seq++;
dout(" mds%d seq %lld cap seq %u\n", session->s_mds, session->s_seq,
(unsigned)seq);
+ if (op == CEPH_CAP_OP_IMPORT)
+ ceph_add_cap_releases(mdsc, session);
+
/* lookup ino */
inode = ceph_find_inode(sb, vino);
+ ci = ceph_inode(inode);
dout(" op %s ino %llx.%llx inode %p\n", ceph_cap_op_name(op), vino.ino,
vino.snap, inode);
if (!inode) {
dout(" i don't have ino %llx\n", vino.ino);
- if (op == CEPH_CAP_OP_IMPORT)
+ if (op == CEPH_CAP_OP_IMPORT) {
+ spin_lock(&session->s_cap_lock);
__queue_cap_release(session, vino.ino, cap_id,
mseq, seq);
+ spin_unlock(&session->s_cap_lock);
+ }
goto flush_cap_releases;
}
@@ -2776,32 +3077,35 @@ void ceph_handle_caps(struct ceph_mds_session *session,
goto done;
case CEPH_CAP_OP_EXPORT:
- handle_cap_export(inode, h, session, &open_target_sessions);
- goto done;
+ handle_cap_export(inode, h, peer, session);
+ goto done_unlocked;
case CEPH_CAP_OP_IMPORT:
- handle_cap_import(mdsc, inode, h, session,
- snaptrace, snaptrace_len);
- ceph_check_caps(ceph_inode(inode), CHECK_CAPS_NODELAY,
- session);
+ handle_cap_import(mdsc, inode, h, peer, session,
+ &cap, &issued);
+ handle_cap_grant(mdsc, inode, h, snaptrace, snaptrace_len,
+ msg->middle, session, cap, issued);
goto done_unlocked;
}
/* the rest require a cap */
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
cap = __get_cap_for_mds(ceph_inode(inode), mds);
if (!cap) {
dout(" no cap on %p ino %llx.%llx from mds%d\n",
inode, ceph_ino(inode), ceph_snap(inode), mds);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
goto flush_cap_releases;
}
- /* note that each of these drops i_lock for us */
+ /* note that each of these drops i_ceph_lock for us */
switch (op) {
case CEPH_CAP_OP_REVOKE:
case CEPH_CAP_OP_GRANT:
- handle_cap_grant(inode, h, session, cap, msg->middle);
+ __ceph_caps_issued(ci, &issued);
+ issued |= __ceph_caps_dirty(ci);
+ handle_cap_grant(mdsc, inode, h, NULL, 0, msg->middle,
+ session, cap, issued);
goto done_unlocked;
case CEPH_CAP_OP_FLUSH_ACK:
@@ -2813,7 +3117,7 @@ void ceph_handle_caps(struct ceph_mds_session *session,
break;
default:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
pr_err("ceph_handle_caps: unknown cap op %d %s\n", op,
ceph_cap_op_name(op));
}
@@ -2834,8 +3138,6 @@ done:
done_unlocked:
if (inode)
iput(inode);
- if (open_target_sessions)
- ceph_mdsc_open_export_target_sessions(mdsc, session);
return;
bad:
@@ -2876,47 +3178,24 @@ void ceph_check_delayed_caps(struct ceph_mds_client *mdsc)
*/
void ceph_flush_dirty_caps(struct ceph_mds_client *mdsc)
{
- struct ceph_inode_info *ci, *nci = NULL;
- struct inode *inode, *ninode = NULL;
- struct list_head *p, *n;
+ struct ceph_inode_info *ci;
+ struct inode *inode;
dout("flush_dirty_caps\n");
spin_lock(&mdsc->cap_dirty_lock);
- list_for_each_safe(p, n, &mdsc->cap_dirty) {
- if (nci) {
- ci = nci;
- inode = ninode;
- ci->i_ceph_flags &= ~CEPH_I_NOFLUSH;
- dout("flush_dirty_caps inode %p (was next inode)\n",
- inode);
- } else {
- ci = list_entry(p, struct ceph_inode_info,
- i_dirty_item);
- inode = igrab(&ci->vfs_inode);
- BUG_ON(!inode);
- dout("flush_dirty_caps inode %p\n", inode);
- }
- if (n != &mdsc->cap_dirty) {
- nci = list_entry(n, struct ceph_inode_info,
- i_dirty_item);
- ninode = igrab(&nci->vfs_inode);
- BUG_ON(!ninode);
- nci->i_ceph_flags |= CEPH_I_NOFLUSH;
- dout("flush_dirty_caps next inode %p, noflush\n",
- ninode);
- } else {
- nci = NULL;
- ninode = NULL;
- }
+ while (!list_empty(&mdsc->cap_dirty)) {
+ ci = list_first_entry(&mdsc->cap_dirty, struct ceph_inode_info,
+ i_dirty_item);
+ inode = &ci->vfs_inode;
+ ihold(inode);
+ dout("flush_dirty_caps %p\n", inode);
spin_unlock(&mdsc->cap_dirty_lock);
- if (inode) {
- ceph_check_caps(ci, CHECK_CAPS_NODELAY|CHECK_CAPS_FLUSH,
- NULL);
- iput(inode);
- }
+ ceph_check_caps(ci, CHECK_CAPS_NODELAY|CHECK_CAPS_FLUSH, NULL);
+ iput(inode);
spin_lock(&mdsc->cap_dirty_lock);
}
spin_unlock(&mdsc->cap_dirty_lock);
+ dout("flush_dirty_caps done\n");
}
/*
@@ -2929,13 +3208,13 @@ void ceph_put_fmode(struct ceph_inode_info *ci, int fmode)
struct inode *inode = &ci->vfs_inode;
int last = 0;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
dout("put_fmode %p fmode %d %d -> %d\n", inode, fmode,
ci->i_nr_by_mode[fmode], ci->i_nr_by_mode[fmode]-1);
BUG_ON(ci->i_nr_by_mode[fmode] == 0);
if (--ci->i_nr_by_mode[fmode] == 0)
last++;
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
if (last && ci->i_vino.snap == CEPH_NOSNAP)
ceph_check_caps(ci, 0, NULL);
@@ -2958,7 +3237,7 @@ int ceph_encode_inode_release(void **p, struct inode *inode,
int used, dirty;
int ret = 0;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
used = __ceph_caps_used(ci);
dirty = __ceph_caps_dirty(ci);
@@ -2976,21 +3255,19 @@ int ceph_encode_inode_release(void **p, struct inode *inode,
(cap->issued & unless) == 0)) {
if ((cap->issued & drop) &&
(cap->issued & unless) == 0) {
- dout("encode_inode_release %p cap %p %s -> "
- "%s\n", inode, cap,
+ int wanted = __ceph_caps_wanted(ci);
+ if ((ci->i_ceph_flags & CEPH_I_NODELAY) == 0)
+ wanted |= cap->mds_wanted;
+ dout("encode_inode_release %p cap %p "
+ "%s -> %s, wanted %s -> %s\n", inode, cap,
ceph_cap_string(cap->issued),
- ceph_cap_string(cap->issued & ~drop));
+ ceph_cap_string(cap->issued & ~drop),
+ ceph_cap_string(cap->mds_wanted),
+ ceph_cap_string(wanted));
+
cap->issued &= ~drop;
cap->implemented &= ~drop;
- if (ci->i_ceph_flags & CEPH_I_NODELAY) {
- int wanted = __ceph_caps_wanted(ci);
- dout(" wanted %s -> %s (act %s)\n",
- ceph_cap_string(cap->mds_wanted),
- ceph_cap_string(cap->mds_wanted &
- ~wanted),
- ceph_cap_string(wanted));
- cap->mds_wanted &= wanted;
- }
+ cap->mds_wanted = wanted;
} else {
dout("encode_inode_release %p cap %p %s"
" (force)\n", inode, cap,
@@ -3002,7 +3279,7 @@ int ceph_encode_inode_release(void **p, struct inode *inode,
rel->seq = cpu_to_le32(cap->seq);
rel->issue_seq = cpu_to_le32(cap->issue_seq),
rel->mseq = cpu_to_le32(cap->mseq);
- rel->caps = cpu_to_le32(cap->issued);
+ rel->caps = cpu_to_le32(cap->implemented);
rel->wanted = cpu_to_le32(cap->mds_wanted);
rel->dname_len = 0;
rel->dname_seq = 0;
@@ -3013,7 +3290,7 @@ int ceph_encode_inode_release(void **p, struct inode *inode,
inode, cap, ceph_cap_string(cap->issued));
}
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return ret;
}
@@ -3028,7 +3305,7 @@ int ceph_encode_dentry_release(void **p, struct dentry *dentry,
/*
* force an record for the directory caps if we have a dentry lease.
- * this is racy (can't take i_lock and d_lock together), but it
+ * this is racy (can't take i_ceph_lock and d_lock together), but it
* doesn't have to be perfect; the mds will revoke anything we don't
* release.
*/
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index 7ae1b3d55b5..5a743ac141a 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -60,17 +60,20 @@ static int mdsc_show(struct seq_file *s, void *p)
for (rp = rb_first(&mdsc->request_tree); rp; rp = rb_next(rp)) {
req = rb_entry(rp, struct ceph_mds_request, r_node);
- if (req->r_request)
- seq_printf(s, "%lld\tmds%d\t", req->r_tid, req->r_mds);
- else
+ if (req->r_request && req->r_session)
+ seq_printf(s, "%lld\tmds%d\t", req->r_tid,
+ req->r_session->s_mds);
+ else if (!req->r_request)
seq_printf(s, "%lld\t(no request)\t", req->r_tid);
+ else
+ seq_printf(s, "%lld\t(no session)\t", req->r_tid);
seq_printf(s, "%s", ceph_mds_op_name(req->r_op));
if (req->r_got_unsafe)
- seq_printf(s, "\t(unsafe)");
+ seq_puts(s, "\t(unsafe)");
else
- seq_printf(s, "\t");
+ seq_puts(s, "\t");
if (req->r_inode) {
seq_printf(s, " #%llx", ceph_ino(req->r_inode));
@@ -90,6 +93,8 @@ static int mdsc_show(struct seq_file *s, void *p)
} else if (req->r_path1) {
seq_printf(s, " #%llx/%s", req->r_ino1.ino,
req->r_path1);
+ } else {
+ seq_printf(s, " #%llx", req->r_ino1.ino);
}
if (req->r_old_dentry) {
@@ -99,7 +104,8 @@ static int mdsc_show(struct seq_file *s, void *p)
path = NULL;
spin_lock(&req->r_old_dentry->d_lock);
seq_printf(s, " #%llx/%.*s (%s)",
- ceph_ino(req->r_old_dentry->d_parent->d_inode),
+ req->r_old_dentry_dir ?
+ ceph_ino(req->r_old_dentry_dir) : 0,
req->r_old_dentry->d_name.len,
req->r_old_dentry->d_name.name,
path ? path : "");
@@ -113,7 +119,7 @@ static int mdsc_show(struct seq_file *s, void *p)
seq_printf(s, " %s", req->r_path2);
}
- seq_printf(s, "\n");
+ seq_puts(s, "\n");
}
mutex_unlock(&mdsc->mutex);
@@ -198,6 +204,7 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
int err = -ENOMEM;
dout("ceph_fs_debugfs_init\n");
+ BUG_ON(!fsc->client->debugfs_dir);
fsc->debugfs_congestion_kb =
debugfs_create_file("writeback_congestion_kb",
0600,
@@ -207,8 +214,6 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
if (!fsc->debugfs_congestion_kb)
goto out;
- dout("a\n");
-
snprintf(name, sizeof(name), "../../bdi/%s",
dev_name(fsc->backing_dev_info.dev));
fsc->debugfs_bdi =
@@ -218,7 +223,6 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
if (!fsc->debugfs_bdi)
goto out;
- dout("b\n");
fsc->debugfs_mdsmap = debugfs_create_file("mdsmap",
0600,
fsc->client->debugfs_dir,
@@ -227,7 +231,6 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
if (!fsc->debugfs_mdsmap)
goto out;
- dout("ca\n");
fsc->debugfs_mdsc = debugfs_create_file("mdsc",
0600,
fsc->client->debugfs_dir,
@@ -236,7 +239,6 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
if (!fsc->debugfs_mdsc)
goto out;
- dout("da\n");
fsc->debugfs_caps = debugfs_create_file("caps",
0400,
fsc->client->debugfs_dir,
@@ -245,7 +247,6 @@ int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
if (!fsc->debugfs_caps)
goto out;
- dout("ea\n");
fsc->debugfs_dentry_lru = debugfs_create_file("dentry_lru",
0600,
fsc->client->debugfs_dir,
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index e0a2dc6fcaf..c29d6ae6887 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -40,13 +40,6 @@ int ceph_init_dentry(struct dentry *dentry)
if (dentry->d_fsdata)
return 0;
- if (ceph_snap(dentry->d_parent->d_inode) == CEPH_NOSNAP)
- dentry->d_op = &ceph_dentry_ops;
- else if (ceph_snap(dentry->d_parent->d_inode) == CEPH_SNAPDIR)
- dentry->d_op = &ceph_snapdir_dentry_ops;
- else
- dentry->d_op = &ceph_snap_dentry_ops;
-
di = kmem_cache_alloc(ceph_dentry_cachep, GFP_NOFS | __GFP_ZERO);
if (!di)
return -ENOMEM; /* oh well */
@@ -57,16 +50,41 @@ int ceph_init_dentry(struct dentry *dentry)
kmem_cache_free(ceph_dentry_cachep, di);
goto out_unlock;
}
+
+ if (ceph_snap(dentry->d_parent->d_inode) == CEPH_NOSNAP)
+ d_set_d_op(dentry, &ceph_dentry_ops);
+ else if (ceph_snap(dentry->d_parent->d_inode) == CEPH_SNAPDIR)
+ d_set_d_op(dentry, &ceph_snapdir_dentry_ops);
+ else
+ d_set_d_op(dentry, &ceph_snap_dentry_ops);
+
di->dentry = dentry;
di->lease_session = NULL;
- dentry->d_fsdata = di;
dentry->d_time = jiffies;
+ /* avoid reordering d_fsdata setup so that the check above is safe */
+ smp_mb();
+ dentry->d_fsdata = di;
ceph_dentry_lru_add(dentry);
out_unlock:
spin_unlock(&dentry->d_lock);
return 0;
}
+struct inode *ceph_get_dentry_parent_inode(struct dentry *dentry)
+{
+ struct inode *inode = NULL;
+
+ if (!dentry)
+ return NULL;
+
+ spin_lock(&dentry->d_lock);
+ if (!IS_ROOT(dentry)) {
+ inode = dentry->d_parent->d_inode;
+ ihold(inode);
+ }
+ spin_unlock(&dentry->d_lock);
+ return inode;
+}
/*
@@ -82,6 +100,14 @@ static unsigned fpos_off(loff_t p)
return p & 0xffffffff;
}
+static int fpos_cmp(loff_t l, loff_t r)
+{
+ int v = ceph_frag_compare(fpos_frag(l), fpos_frag(r));
+ if (v)
+ return v;
+ return (int)(fpos_off(l) - fpos_off(r));
+}
+
/*
* When possible, we try to satisfy a readdir by peeking at the
* dcache. We make this work by carefully ordering dentries on
@@ -89,15 +115,15 @@ static unsigned fpos_off(loff_t p)
* falling back to a "normal" sync readdir if any dentries in the dir
* are dropped.
*
- * I_COMPLETE tells indicates we have all dentries in the dir. It is
+ * Complete dir indicates that we have all dentries in the dir. It is
* defined IFF we hold CEPH_CAP_FILE_SHARED (which will be revoked by
* the MDS if/when the directory is modified).
*/
-static int __dcache_readdir(struct file *filp,
- void *dirent, filldir_t filldir)
+static int __dcache_readdir(struct file *file, struct dir_context *ctx,
+ u32 shared_gen)
{
- struct ceph_file_info *fi = filp->private_data;
- struct dentry *parent = filp->f_dentry;
+ struct ceph_file_info *fi = file->private_data;
+ struct dentry *parent = file->f_dentry;
struct inode *dir = parent->d_inode;
struct list_head *p;
struct dentry *dentry, *last;
@@ -108,14 +134,14 @@ static int __dcache_readdir(struct file *filp,
last = fi->dentry;
fi->dentry = NULL;
- dout("__dcache_readdir %p at %llu (last %p)\n", dir, filp->f_pos,
- last);
+ dout("__dcache_readdir %p v%u at %llu (last %p)\n",
+ dir, shared_gen, ctx->pos, last);
- spin_lock(&dcache_lock);
+ spin_lock(&parent->d_lock);
/* start at beginning? */
- if (filp->f_pos == 2 || (last &&
- filp->f_pos < ceph_dentry(last)->offset)) {
+ if (ctx->pos == 2 || last == NULL ||
+ fpos_cmp(ctx->pos, ceph_dentry(last)->offset) < 0) {
if (list_empty(&parent->d_subdirs))
goto out_unlock;
p = parent->d_subdirs.prev;
@@ -132,63 +158,65 @@ more:
d_unhashed(dentry) ? "!hashed" : "hashed",
parent->d_subdirs.prev, parent->d_subdirs.next);
if (p == &parent->d_subdirs) {
- fi->at_end = 1;
+ fi->flags |= CEPH_F_ATEND;
goto out_unlock;
}
- if (!d_unhashed(dentry) && dentry->d_inode &&
+ spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED);
+ if (di->lease_shared_gen == shared_gen &&
+ !d_unhashed(dentry) && dentry->d_inode &&
ceph_snap(dentry->d_inode) != CEPH_SNAPDIR &&
ceph_ino(dentry->d_inode) != CEPH_INO_CEPH &&
- filp->f_pos <= di->offset)
+ fpos_cmp(ctx->pos, di->offset) <= 0)
break;
dout(" skipping %p %.*s at %llu (%llu)%s%s\n", dentry,
dentry->d_name.len, dentry->d_name.name, di->offset,
- filp->f_pos, d_unhashed(dentry) ? " unhashed" : "",
+ ctx->pos, d_unhashed(dentry) ? " unhashed" : "",
!dentry->d_inode ? " null" : "");
+ spin_unlock(&dentry->d_lock);
p = p->prev;
dentry = list_entry(p, struct dentry, d_u.d_child);
di = ceph_dentry(dentry);
}
- atomic_inc(&dentry->d_count);
- spin_unlock(&dcache_lock);
+ dget_dlock(dentry);
+ spin_unlock(&dentry->d_lock);
+ spin_unlock(&parent->d_lock);
- dout(" %llu (%llu) dentry %p %.*s %p\n", di->offset, filp->f_pos,
- dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode);
- filp->f_pos = di->offset;
- err = filldir(dirent, dentry->d_name.name,
- dentry->d_name.len, di->offset,
- dentry->d_inode->i_ino,
- dentry->d_inode->i_mode >> 12);
+ /* make sure a dentry wasn't dropped while we didn't have parent lock */
+ if (!ceph_dir_is_complete(dir)) {
+ dout(" lost dir complete on %p; falling back to mds\n", dir);
+ dput(dentry);
+ err = -EAGAIN;
+ goto out;
+ }
- if (last) {
- if (err < 0) {
+ dout(" %llu (%llu) dentry %p %.*s %p\n", di->offset, ctx->pos,
+ dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode);
+ if (!dir_emit(ctx, dentry->d_name.name,
+ dentry->d_name.len,
+ ceph_translate_ino(dentry->d_sb, dentry->d_inode->i_ino),
+ dentry->d_inode->i_mode >> 12)) {
+ if (last) {
/* remember our position */
fi->dentry = last;
- fi->next_offset = di->offset;
- } else {
- dput(last);
+ fi->next_offset = fpos_off(di->offset);
}
+ dput(dentry);
+ return 0;
}
- last = dentry;
- if (err < 0)
- goto out;
-
- filp->f_pos++;
+ ctx->pos = di->offset + 1;
- /* make sure a dentry wasn't dropped while we didn't have dcache_lock */
- if (!ceph_i_test(dir, CEPH_I_COMPLETE)) {
- dout(" lost I_COMPLETE on %p; falling back to mds\n", dir);
- err = -EAGAIN;
- goto out;
- }
+ if (last)
+ dput(last);
+ last = dentry;
- spin_lock(&dcache_lock);
+ spin_lock(&parent->d_lock);
p = p->prev; /* advance to next dentry */
goto more;
out_unlock:
- spin_unlock(&dcache_lock);
+ spin_unlock(&parent->d_lock);
out:
if (last)
dput(last);
@@ -214,61 +242,64 @@ static int note_last_dentry(struct ceph_file_info *fi, const char *name,
return 0;
}
-static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir)
+static int ceph_readdir(struct file *file, struct dir_context *ctx)
{
- struct ceph_file_info *fi = filp->private_data;
- struct inode *inode = filp->f_dentry->d_inode;
+ struct ceph_file_info *fi = 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);
struct ceph_mds_client *mdsc = fsc->mdsc;
- unsigned frag = fpos_frag(filp->f_pos);
- int off = fpos_off(filp->f_pos);
+ unsigned frag = fpos_frag(ctx->pos);
+ int off = fpos_off(ctx->pos);
int err;
u32 ftype;
struct ceph_mds_reply_info_parsed *rinfo;
- const int max_entries = fsc->mount_options->max_readdir;
- const int max_bytes = fsc->mount_options->max_readdir_bytes;
- dout("readdir %p filp %p frag %u off %u\n", inode, filp, frag, off);
- if (fi->at_end)
+ dout("readdir %p file %p frag %u off %u\n", inode, file, frag, off);
+ if (fi->flags & CEPH_F_ATEND)
return 0;
/* always start with . and .. */
- if (filp->f_pos == 0) {
+ if (ctx->pos == 0) {
/* note dir version at start of readdir so we can tell
* if any dentries get dropped */
- fi->dir_release_count = ci->i_release_count;
+ fi->dir_release_count = atomic_read(&ci->i_release_count);
dout("readdir off 0 -> '.'\n");
- if (filldir(dirent, ".", 1, ceph_make_fpos(0, 0),
- inode->i_ino, inode->i_mode >> 12) < 0)
+ if (!dir_emit(ctx, ".", 1,
+ ceph_translate_ino(inode->i_sb, inode->i_ino),
+ inode->i_mode >> 12))
return 0;
- filp->f_pos = 1;
+ ctx->pos = 1;
off = 1;
}
- if (filp->f_pos == 1) {
+ if (ctx->pos == 1) {
+ ino_t ino = parent_ino(file->f_dentry);
dout("readdir off 1 -> '..'\n");
- if (filldir(dirent, "..", 2, ceph_make_fpos(0, 1),
- filp->f_dentry->d_parent->d_inode->i_ino,
- inode->i_mode >> 12) < 0)
+ if (!dir_emit(ctx, "..", 2,
+ ceph_translate_ino(inode->i_sb, ino),
+ inode->i_mode >> 12))
return 0;
- filp->f_pos = 2;
+ ctx->pos = 2;
off = 2;
}
/* can we use the dcache? */
- spin_lock(&inode->i_lock);
- if ((filp->f_pos == 2 || fi->dentry) &&
+ spin_lock(&ci->i_ceph_lock);
+ if ((ctx->pos == 2 || fi->dentry) &&
!ceph_test_mount_opt(fsc, NOASYNCREADDIR) &&
ceph_snap(inode) != CEPH_SNAPDIR &&
- (ci->i_ceph_flags & CEPH_I_COMPLETE) &&
+ __ceph_dir_is_complete(ci) &&
__ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1)) {
- spin_unlock(&inode->i_lock);
- err = __dcache_readdir(filp, dirent, filldir);
+ u32 shared_gen = ci->i_shared_gen;
+ spin_unlock(&ci->i_ceph_lock);
+ err = __dcache_readdir(file, ctx, shared_gen);
if (err != -EAGAIN)
return err;
+ frag = fpos_frag(ctx->pos);
+ off = fpos_off(ctx->pos);
} else {
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
}
if (fi->dentry) {
err = note_last_dentry(fi, fi->dentry->d_name.name,
@@ -294,16 +325,19 @@ more:
fi->last_readdir = NULL;
}
- /* requery frag tree, as the frag topology may have changed */
- frag = ceph_choose_frag(ceph_inode(inode), frag, NULL, NULL);
-
dout("readdir fetching %llx.%llx frag %x offset '%s'\n",
ceph_vinop(inode), frag, fi->last_name);
req = ceph_mdsc_create_request(mdsc, op, USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
- req->r_dentry = dget(filp->f_dentry);
+ err = ceph_alloc_readdir_reply_buffer(req, inode);
+ if (err) {
+ ceph_mdsc_put_request(req);
+ return err;
+ }
+ req->r_inode = inode;
+ ihold(inode);
+ req->r_dentry = dget(file->f_dentry);
/* hints to request -> mds selection code */
req->r_direct_mode = USE_AUTH_MDS;
req->r_direct_hash = ceph_frag_value(frag);
@@ -311,9 +345,6 @@ more:
req->r_path2 = kstrdup(fi->last_name, GFP_NOFS);
req->r_readdir_offset = fi->next_offset;
req->r_args.readdir.frag = cpu_to_le32(frag);
- req->r_args.readdir.max_entries = cpu_to_le32(max_entries);
- req->r_args.readdir.max_bytes = cpu_to_le32(max_bytes);
- req->r_num_caps = max_entries + 1;
err = ceph_mdsc_do_request(mdsc, NULL, req);
if (err < 0) {
ceph_mdsc_put_request(req);
@@ -326,19 +357,32 @@ more:
if (!req->r_did_prepopulate) {
dout("readdir !did_prepopulate");
- fi->dir_release_count--; /* preclude I_COMPLETE */
+ /* preclude from marking dir complete */
+ fi->dir_release_count--;
}
/* note next offset and last dentry name */
+ rinfo = &req->r_reply_info;
+ if (le32_to_cpu(rinfo->dir_dir->frag) != frag) {
+ frag = le32_to_cpu(rinfo->dir_dir->frag);
+ if (ceph_frag_is_leftmost(frag))
+ fi->next_offset = 2;
+ else
+ fi->next_offset = 0;
+ off = fi->next_offset;
+ }
+ fi->frag = frag;
fi->offset = fi->next_offset;
fi->last_readdir = req;
if (req->r_reply_info.dir_end) {
kfree(fi->last_name);
fi->last_name = NULL;
- fi->next_offset = 2;
+ if (ceph_frag_is_rightmost(frag))
+ fi->next_offset = 2;
+ else
+ fi->next_offset = 0;
} else {
- rinfo = &req->r_reply_info;
err = note_last_dentry(fi,
rinfo->dir_dname[rinfo->dir_nr-1],
rinfo->dir_dname_len[rinfo->dir_nr-1]);
@@ -351,27 +395,32 @@ more:
rinfo = &fi->last_readdir->r_reply_info;
dout("readdir frag %x num %d off %d chunkoff %d\n", frag,
rinfo->dir_nr, off, fi->offset);
- while (off - fi->offset >= 0 && off - fi->offset < rinfo->dir_nr) {
- u64 pos = ceph_make_fpos(frag, off);
+
+ ctx->pos = ceph_make_fpos(frag, off);
+ while (off >= fi->offset && off - fi->offset < rinfo->dir_nr) {
struct ceph_mds_reply_inode *in =
rinfo->dir_in[off - fi->offset].in;
+ struct ceph_vino vino;
+ ino_t ino;
+
dout("readdir off %d (%d/%d) -> %lld '%.*s' %p\n",
- off, off - fi->offset, rinfo->dir_nr, pos,
+ off, off - fi->offset, rinfo->dir_nr, ctx->pos,
rinfo->dir_dname_len[off - fi->offset],
rinfo->dir_dname[off - fi->offset], in);
BUG_ON(!in);
ftype = le32_to_cpu(in->mode) >> 12;
- if (filldir(dirent,
+ vino.ino = le64_to_cpu(in->ino);
+ vino.snap = le64_to_cpu(in->snapid);
+ ino = ceph_vino_to_ino(vino);
+ if (!dir_emit(ctx,
rinfo->dir_dname[off - fi->offset],
rinfo->dir_dname_len[off - fi->offset],
- pos,
- le64_to_cpu(in->ino),
- ftype) < 0) {
+ ceph_translate_ino(inode->i_sb, ino), ftype)) {
dout("filldir stopping us...\n");
return 0;
}
off++;
- filp->f_pos = pos + 1;
+ ctx->pos++;
}
if (fi->last_name) {
@@ -384,65 +433,73 @@ more:
if (!ceph_frag_is_rightmost(frag)) {
frag = ceph_frag_next(frag);
off = 0;
- filp->f_pos = ceph_make_fpos(frag, off);
+ ctx->pos = ceph_make_fpos(frag, off);
dout("readdir next frag is %x\n", frag);
goto more;
}
- fi->at_end = 1;
+ fi->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.
*/
- spin_lock(&inode->i_lock);
- if (ci->i_release_count == fi->dir_release_count) {
+ spin_lock(&ci->i_ceph_lock);
+ if (atomic_read(&ci->i_release_count) == fi->dir_release_count) {
dout(" marking %p complete\n", inode);
- ci->i_ceph_flags |= CEPH_I_COMPLETE;
- ci->i_max_offset = filp->f_pos;
+ __ceph_dir_set_complete(ci, fi->dir_release_count);
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
- dout("readdir %p filp %p done.\n", inode, filp);
+ dout("readdir %p file %p done.\n", inode, file);
return 0;
}
-static void reset_readdir(struct ceph_file_info *fi)
+static void reset_readdir(struct ceph_file_info *fi, unsigned frag)
{
if (fi->last_readdir) {
ceph_mdsc_put_request(fi->last_readdir);
fi->last_readdir = NULL;
}
kfree(fi->last_name);
- fi->next_offset = 2; /* compensate for . and .. */
+ fi->last_name = NULL;
+ if (ceph_frag_is_leftmost(frag))
+ fi->next_offset = 2; /* compensate for . and .. */
+ else
+ fi->next_offset = 0;
if (fi->dentry) {
dput(fi->dentry);
fi->dentry = NULL;
}
- fi->at_end = 0;
+ fi->flags &= ~CEPH_F_ATEND;
}
-static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int origin)
+static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int whence)
{
struct ceph_file_info *fi = file->private_data;
struct inode *inode = file->f_mapping->host;
- loff_t old_offset = offset;
+ loff_t old_offset = ceph_make_fpos(fi->frag, fi->next_offset);
loff_t retval;
mutex_lock(&inode->i_mutex);
- switch (origin) {
+ retval = -EINVAL;
+ switch (whence) {
case SEEK_END:
offset += inode->i_size + 2; /* FIXME */
break;
case SEEK_CUR:
offset += file->f_pos;
+ case SEEK_SET:
+ break;
+ default:
+ goto out;
}
- retval = -EINVAL;
- if (offset >= 0 && offset <= inode->i_sb->s_maxbytes) {
+
+ if (offset >= 0) {
if (offset != file->f_pos) {
file->f_pos = offset;
file->f_version = 0;
- fi->at_end = 0;
+ fi->flags &= ~CEPH_F_ATEND;
}
retval = offset;
@@ -451,39 +508,33 @@ static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int origin)
* seek to new frag, or seek prior to current chunk.
*/
if (offset == 0 ||
- fpos_frag(offset) != fpos_frag(old_offset) ||
+ fpos_frag(offset) != fi->frag ||
fpos_off(offset) < fi->offset) {
dout("dir_llseek dropping %p content\n", file);
- reset_readdir(fi);
+ reset_readdir(fi, fpos_frag(offset));
}
/* bump dir_release_count if we did a forward seek */
- if (offset > old_offset)
+ if (fpos_cmp(offset, old_offset) > 0)
fi->dir_release_count--;
}
+out:
mutex_unlock(&inode->i_mutex);
return retval;
}
/*
- * Process result of a lookup/open request.
- *
- * Mainly, make sure we return the final req->r_dentry (if it already
- * existed) in place of the original VFS-provided dentry when they
- * differ.
- *
- * Gracefully handle the case where the MDS replies with -ENOENT and
- * no trace (which it may do, at its discretion, e.g., if it doesn't
- * care to issue a lease on the negative dentry).
+ * Handle lookups for the hidden .snap directory.
*/
-struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
- struct dentry *dentry, int err)
+int ceph_handle_snapdir(struct ceph_mds_request *req,
+ struct dentry *dentry, int err)
{
struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
- struct inode *parent = dentry->d_parent->d_inode;
+ struct inode *parent = dentry->d_parent->d_inode; /* we hold i_mutex */
/* .snap dir? */
if (err == -ENOENT &&
+ ceph_snap(parent) == CEPH_NOSNAP &&
strcmp(dentry->d_name.name,
fsc->mount_options->snapdir_name) == 0) {
struct inode *inode = ceph_get_snapdir(parent);
@@ -493,7 +544,23 @@ struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
d_add(dentry, inode);
err = 0;
}
+ return err;
+}
+/*
+ * Figure out final result of a lookup/open request.
+ *
+ * Mainly, make sure we return the final req->r_dentry (if it already
+ * existed) in place of the original VFS-provided dentry when they
+ * differ.
+ *
+ * Gracefully handle the case where the MDS replies with -ENOENT and
+ * no trace (which it may do, at its discretion, e.g., if it doesn't
+ * care to issue a lease on the negative dentry).
+ */
+struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
+ struct dentry *dentry, int err)
+{
if (err == -ENOENT) {
/* no trace? */
err = 0;
@@ -528,7 +595,7 @@ static int is_root_ceph_dentry(struct inode *inode, struct dentry *dentry)
* the MDS so that it gets our 'caps wanted' value in a single op.
*/
static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
- struct nameidata *nd)
+ unsigned int flags)
{
struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
struct ceph_mds_client *mdsc = fsc->mdsc;
@@ -546,35 +613,26 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
if (err < 0)
return ERR_PTR(err);
- /* open (but not create!) intent? */
- if (nd &&
- (nd->flags & LOOKUP_OPEN) &&
- (nd->flags & LOOKUP_CONTINUE) == 0 && /* only open last component */
- !(nd->intent.open.flags & O_CREAT)) {
- int mode = nd->intent.open.create_mode & ~current->fs->umask;
- return ceph_lookup_open(dir, dentry, nd, mode, 1);
- }
-
/* can we conclude ENOENT locally? */
if (dentry->d_inode == NULL) {
struct ceph_inode_info *ci = ceph_inode(dir);
struct ceph_dentry_info *di = ceph_dentry(dentry);
- spin_lock(&dir->i_lock);
+ spin_lock(&ci->i_ceph_lock);
dout(" dir %p flags are %d\n", dir, ci->i_ceph_flags);
if (strncmp(dentry->d_name.name,
fsc->mount_options->snapdir_name,
dentry->d_name.len) &&
!is_root_ceph_dentry(dir, dentry) &&
- (ci->i_ceph_flags & CEPH_I_COMPLETE) &&
+ __ceph_dir_is_complete(ci) &&
(__ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1))) {
- spin_unlock(&dir->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
dout(" dir %p complete, -ENOENT\n", dir);
d_add(dentry, NULL);
di->lease_shared_gen = ci->i_shared_gen;
return NULL;
}
- spin_unlock(&dir->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
}
op = ceph_snap(dir) == CEPH_SNAPDIR ?
@@ -588,6 +646,7 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
req->r_args.getattr.mask = cpu_to_le32(CEPH_STAT_CAP_INODE);
req->r_locked_dir = dir;
err = ceph_mdsc_do_request(mdsc, NULL, req);
+ err = ceph_handle_snapdir(req, dentry, err);
dentry = ceph_finish_lookup(req, dentry, err);
ceph_mdsc_put_request(req); /* will dput(dentry) */
dout("lookup result=%p\n", dentry);
@@ -600,7 +659,7 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
*/
int ceph_handle_notrace_create(struct inode *dir, struct dentry *dentry)
{
- struct dentry *result = ceph_lookup(dir, dentry, NULL);
+ struct dentry *result = ceph_lookup(dir, dentry, 0);
if (result && !IS_ERR(result)) {
/*
@@ -618,7 +677,7 @@ int ceph_handle_notrace_create(struct inode *dir, struct dentry *dentry)
}
static int ceph_mknod(struct inode *dir, struct dentry *dentry,
- int mode, dev_t rdev)
+ umode_t mode, dev_t rdev)
{
struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
struct ceph_mds_client *mdsc = fsc->mdsc;
@@ -628,7 +687,7 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry,
if (ceph_snap(dir) != CEPH_NOSNAP)
return -EROFS;
- dout("mknod in dir %p dentry %p mode 0%o rdev %d\n",
+ dout("mknod in dir %p dentry %p mode 0%ho rdev %d\n",
dir, dentry, mode, rdev);
req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_MKNOD, USE_AUTH_MDS);
if (IS_ERR(req)) {
@@ -646,31 +705,18 @@ static int ceph_mknod(struct inode *dir, struct dentry *dentry,
if (!err && !req->r_reply_info.head->is_dentry)
err = ceph_handle_notrace_create(dir, dentry);
ceph_mdsc_put_request(req);
- if (err)
+
+ if (!err)
+ ceph_init_acl(dentry, dentry->d_inode, dir);
+ else
d_drop(dentry);
return err;
}
-static int ceph_create(struct inode *dir, struct dentry *dentry, int mode,
- struct nameidata *nd)
+static int ceph_create(struct inode *dir, struct dentry *dentry, umode_t mode,
+ bool excl)
{
- dout("create in dir %p dentry %p name '%.*s'\n",
- dir, dentry, dentry->d_name.len, dentry->d_name.name);
-
- if (ceph_snap(dir) != CEPH_NOSNAP)
- return -EROFS;
-
- if (nd) {
- BUG_ON((nd->flags & LOOKUP_OPEN) == 0);
- dentry = ceph_lookup_open(dir, dentry, nd, mode, 0);
- /* hrm, what should i do here if we get aliased? */
- if (IS_ERR(dentry))
- return PTR_ERR(dentry);
- return 0;
- }
-
- /* fall back to mknod */
- return ceph_mknod(dir, dentry, (mode & ~S_IFMT) | S_IFREG, 0);
+ return ceph_mknod(dir, dentry, mode, 0);
}
static int ceph_symlink(struct inode *dir, struct dentry *dentry,
@@ -700,12 +746,14 @@ static int ceph_symlink(struct inode *dir, struct dentry *dentry,
if (!err && !req->r_reply_info.head->is_dentry)
err = ceph_handle_notrace_create(dir, dentry);
ceph_mdsc_put_request(req);
- if (err)
+ if (!err)
+ ceph_init_acl(dentry, dentry->d_inode, dir);
+ else
d_drop(dentry);
return err;
}
-static int ceph_mkdir(struct inode *dir, struct dentry *dentry, int mode)
+static int ceph_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
{
struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
struct ceph_mds_client *mdsc = fsc->mdsc;
@@ -719,7 +767,7 @@ static int ceph_mkdir(struct inode *dir, struct dentry *dentry, int mode)
dout("mksnap dir %p snap '%.*s' dn %p\n", dir,
dentry->d_name.len, dentry->d_name.name, dentry);
} else if (ceph_snap(dir) == CEPH_NOSNAP) {
- dout("mkdir dir %p dn %p mode 0%o\n", dir, dentry, mode);
+ dout("mkdir dir %p dn %p mode 0%ho\n", dir, dentry, mode);
op = CEPH_MDS_OP_MKDIR;
} else {
goto out;
@@ -741,7 +789,9 @@ static int ceph_mkdir(struct inode *dir, struct dentry *dentry, int mode)
err = ceph_handle_notrace_create(dir, dentry);
ceph_mdsc_put_request(req);
out:
- if (err < 0)
+ if (!err)
+ ceph_init_acl(dentry, dentry->d_inode, dir);
+ else
d_drop(dentry);
return err;
}
@@ -766,15 +816,19 @@ static int ceph_link(struct dentry *old_dentry, struct inode *dir,
}
req->r_dentry = dget(dentry);
req->r_num_caps = 2;
- req->r_old_dentry = dget(old_dentry); /* or inode? hrm. */
+ req->r_old_dentry = dget(old_dentry);
req->r_locked_dir = dir;
req->r_dentry_drop = CEPH_CAP_FILE_SHARED;
req->r_dentry_unless = CEPH_CAP_FILE_EXCL;
+ /* release LINK_SHARED on source inode (mds will lock it) */
+ req->r_old_inode_drop = CEPH_CAP_LINK_SHARED;
err = ceph_mdsc_do_request(mdsc, dir, req);
- if (err)
+ if (err) {
d_drop(dentry);
- else if (!req->r_reply_info.head->is_dentry)
- d_instantiate(dentry, igrab(old_dentry->d_inode));
+ } else if (!req->r_reply_info.head->is_dentry) {
+ ihold(old_dentry->d_inode);
+ d_instantiate(dentry, old_dentry->d_inode);
+ }
ceph_mdsc_put_request(req);
return err;
}
@@ -790,12 +844,12 @@ static int drop_caps_for_unlink(struct inode *inode)
struct ceph_inode_info *ci = ceph_inode(inode);
int drop = CEPH_CAP_LINK_SHARED | CEPH_CAP_LINK_EXCL;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (inode->i_nlink == 1) {
drop |= ~(__ceph_caps_wanted(ci) | CEPH_CAP_PIN);
ci->i_ceph_flags |= CEPH_I_NODELAY;
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return drop;
}
@@ -819,7 +873,7 @@ static int ceph_unlink(struct inode *dir, struct dentry *dentry)
} else if (ceph_snap(dir) == CEPH_NOSNAP) {
dout("unlink/rmdir dir %p dn %p inode %p\n",
dir, dentry, inode);
- op = ((dentry->d_inode->i_mode & S_IFMT) == S_IFDIR) ?
+ op = S_ISDIR(dentry->d_inode->i_mode) ?
CEPH_MDS_OP_RMDIR : CEPH_MDS_OP_UNLINK;
} else
goto out;
@@ -860,9 +914,11 @@ static int ceph_rename(struct inode *old_dir, struct dentry *old_dentry,
req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_RENAME, USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
+ ihold(old_dir);
req->r_dentry = dget(new_dentry);
req->r_num_caps = 2;
req->r_old_dentry = dget(old_dentry);
+ req->r_old_dentry_dir = old_dir;
req->r_locked_dir = new_dir;
req->r_old_dentry_drop = CEPH_CAP_FILE_SHARED;
req->r_old_dentry_unless = CEPH_CAP_FILE_EXCL;
@@ -880,14 +936,16 @@ static int ceph_rename(struct inode *old_dir, struct dentry *old_dentry,
* to do it here.
*/
- /* d_move screws up d_subdirs order */
- ceph_i_clear(new_dir, CEPH_I_COMPLETE);
-
d_move(old_dentry, new_dentry);
/* ensure target dentry is invalidated, despite
rehashing bug in vfs_rename_dir */
ceph_invalidate_dentry_lease(new_dentry);
+
+ /* d_move screws up sibling dentries' offsets */
+ ceph_dir_clear_complete(old_dir);
+ ceph_dir_clear_complete(new_dir);
+
}
ceph_mdsc_put_request(req);
return err;
@@ -921,12 +979,12 @@ static int dentry_lease_is_valid(struct dentry *dentry)
spin_lock(&dentry->d_lock);
di = ceph_dentry(dentry);
- if (di && di->lease_session) {
+ if (di->lease_session) {
s = di->lease_session;
- spin_lock(&s->s_cap_lock);
+ spin_lock(&s->s_gen_ttl_lock);
gen = s->s_cap_gen;
ttl = s->s_cap_ttl;
- spin_unlock(&s->s_cap_lock);
+ spin_unlock(&s->s_gen_ttl_lock);
if (di->lease_gen == gen &&
time_before(jiffies, dentry->d_time) &&
@@ -963,10 +1021,10 @@ static int dir_lease_is_valid(struct inode *dir, struct dentry *dentry)
struct ceph_dentry_info *di = ceph_dentry(dentry);
int valid = 0;
- spin_lock(&dir->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (ci->i_shared_gen == di->lease_shared_gen)
valid = __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1);
- spin_unlock(&dir->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
dout("dir_lease_is_valid dir %p v%u dentry %p v%u = %d\n",
dir, (unsigned)ci->i_shared_gen, dentry,
(unsigned)di->lease_shared_gen, valid);
@@ -976,75 +1034,64 @@ static int dir_lease_is_valid(struct inode *dir, struct dentry *dentry)
/*
* Check if cached dentry can be trusted.
*/
-static int ceph_d_revalidate(struct dentry *dentry, struct nameidata *nd)
+static int ceph_d_revalidate(struct dentry *dentry, unsigned int flags)
{
- struct inode *dir = dentry->d_parent->d_inode;
+ int valid = 0;
+ struct inode *dir;
+
+ if (flags & LOOKUP_RCU)
+ return -ECHILD;
dout("d_revalidate %p '%.*s' inode %p offset %lld\n", dentry,
dentry->d_name.len, dentry->d_name.name, dentry->d_inode,
ceph_dentry(dentry)->offset);
+ dir = ceph_get_dentry_parent_inode(dentry);
+
/* always trust cached snapped dentries, snapdir dentry */
if (ceph_snap(dir) != CEPH_NOSNAP) {
dout("d_revalidate %p '%.*s' inode %p is SNAPPED\n", dentry,
dentry->d_name.len, dentry->d_name.name, dentry->d_inode);
- goto out_touch;
+ valid = 1;
+ } else if (dentry->d_inode &&
+ ceph_snap(dentry->d_inode) == CEPH_SNAPDIR) {
+ valid = 1;
+ } else if (dentry_lease_is_valid(dentry) ||
+ dir_lease_is_valid(dir, dentry)) {
+ if (dentry->d_inode)
+ valid = ceph_is_any_caps(dentry->d_inode);
+ else
+ valid = 1;
}
- if (dentry->d_inode && ceph_snap(dentry->d_inode) == CEPH_SNAPDIR)
- goto out_touch;
-
- if (dentry_lease_is_valid(dentry) ||
- dir_lease_is_valid(dir, dentry))
- goto out_touch;
- dout("d_revalidate %p invalid\n", dentry);
- d_drop(dentry);
- return 0;
-out_touch:
- ceph_dentry_lru_touch(dentry);
- return 1;
+ dout("d_revalidate %p %s\n", dentry, valid ? "valid" : "invalid");
+ if (valid) {
+ ceph_dentry_lru_touch(dentry);
+ } else {
+ ceph_dir_clear_complete(dir);
+ d_drop(dentry);
+ }
+ iput(dir);
+ return valid;
}
/*
- * When a dentry is released, clear the dir I_COMPLETE if it was part
- * of the current dir gen or if this is in the snapshot namespace.
+ * Release our ceph_dentry_info.
*/
-static void ceph_dentry_release(struct dentry *dentry)
+static void ceph_d_release(struct dentry *dentry)
{
struct ceph_dentry_info *di = ceph_dentry(dentry);
- struct inode *parent_inode = NULL;
- u64 snapid = CEPH_NOSNAP;
- if (!IS_ROOT(dentry)) {
- parent_inode = dentry->d_parent->d_inode;
- if (parent_inode)
- snapid = ceph_snap(parent_inode);
- }
- dout("dentry_release %p parent %p\n", dentry, parent_inode);
- if (parent_inode && snapid != CEPH_SNAPDIR) {
- struct ceph_inode_info *ci = ceph_inode(parent_inode);
-
- spin_lock(&parent_inode->i_lock);
- if (ci->i_shared_gen == di->lease_shared_gen ||
- snapid <= CEPH_MAXSNAP) {
- dout(" clearing %p complete (d_release)\n",
- parent_inode);
- ci->i_ceph_flags &= ~CEPH_I_COMPLETE;
- ci->i_release_count++;
- }
- spin_unlock(&parent_inode->i_lock);
- }
- if (di) {
- ceph_dentry_lru_del(dentry);
- if (di->lease_session)
- ceph_put_mds_session(di->lease_session);
- kmem_cache_free(ceph_dentry_cachep, di);
- dentry->d_fsdata = NULL;
- }
+ dout("d_release %p\n", dentry);
+ ceph_dentry_lru_del(dentry);
+ if (di->lease_session)
+ ceph_put_mds_session(di->lease_session);
+ kmem_cache_free(ceph_dentry_cachep, di);
+ dentry->d_fsdata = NULL;
}
static int ceph_snapdir_d_revalidate(struct dentry *dentry,
- struct nameidata *nd)
+ unsigned int flags)
{
/*
* Eventually, we'll want to revalidate snapped metadata
@@ -1053,7 +1100,30 @@ static int ceph_snapdir_d_revalidate(struct dentry *dentry,
return 1;
}
+/*
+ * When the VFS prunes a dentry from the cache, we need to clear the
+ * complete flag on the parent directory.
+ *
+ * Called under dentry->d_lock.
+ */
+static void ceph_d_prune(struct dentry *dentry)
+{
+ dout("ceph_d_prune %p\n", dentry);
+
+ /* do we have a valid parent? */
+ if (IS_ROOT(dentry))
+ return;
+ /* if we are not hashed, we don't affect dir's completeness */
+ if (d_unhashed(dentry))
+ return;
+
+ /*
+ * we hold d_lock, so d_parent is stable, and d_fsdata is never
+ * cleared until d_release
+ */
+ ceph_dir_clear_complete(dentry->d_parent->d_inode);
+}
/*
* read() on a dir. This weird interface hack only works if mounted
@@ -1063,19 +1133,20 @@ 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 inode *inode = file->f_dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct ceph_inode_info *ci = ceph_inode(inode);
int left;
+ const int bufsize = 1024;
if (!ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), DIRSTAT))
return -EISDIR;
if (!cf->dir_info) {
- cf->dir_info = kmalloc(1024, GFP_NOFS);
+ cf->dir_info = kmalloc(bufsize, GFP_NOFS);
if (!cf->dir_info)
return -ENOMEM;
cf->dir_info_len =
- sprintf(cf->dir_info,
+ snprintf(cf->dir_info, bufsize,
"entries: %20lld\n"
" files: %20lld\n"
" subdirs: %20lld\n"
@@ -1109,9 +1180,10 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
* an fsync() on a dir will wait for any uncommitted directory
* operations to commit.
*/
-static int ceph_dir_fsync(struct file *file, int datasync)
+static int ceph_dir_fsync(struct file *file, loff_t start, loff_t end,
+ int datasync)
{
- struct inode *inode = file->f_path.dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct ceph_inode_info *ci = ceph_inode(inode);
struct list_head *head = &ci->i_unsafe_dirops;
struct ceph_mds_request *req;
@@ -1119,6 +1191,11 @@ static int ceph_dir_fsync(struct file *file, int datasync)
int ret = 0;
dout("dir_fsync %p\n", inode);
+ ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
+ if (ret)
+ return ret;
+ mutex_lock(&inode->i_mutex);
+
spin_lock(&ci->i_unsafe_lock);
if (list_empty(head))
goto out;
@@ -1130,6 +1207,7 @@ static int ceph_dir_fsync(struct file *file, int datasync)
do {
ceph_mdsc_get_request(req);
spin_unlock(&ci->i_unsafe_lock);
+
dout("dir_fsync %p wait on tid %llu (until %llu)\n",
inode, req->r_tid, last_tid);
if (req->r_timeout) {
@@ -1142,9 +1220,9 @@ static int ceph_dir_fsync(struct file *file, int datasync)
} else {
wait_for_completion(&req->r_safe_completion);
}
- spin_lock(&ci->i_unsafe_lock);
ceph_mdsc_put_request(req);
+ spin_lock(&ci->i_unsafe_lock);
if (ret || list_empty(head))
break;
req = list_entry(head->next,
@@ -1152,6 +1230,8 @@ static int ceph_dir_fsync(struct file *file, int datasync)
} while (req->r_tid < last_tid);
out:
spin_unlock(&ci->i_unsafe_lock);
+ mutex_unlock(&inode->i_mutex);
+
return ret;
}
@@ -1167,13 +1247,11 @@ void ceph_dentry_lru_add(struct dentry *dn)
dout("dentry_lru_add %p %p '%.*s'\n", di, dn,
dn->d_name.len, dn->d_name.name);
- if (di) {
- mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
- spin_lock(&mdsc->dentry_lru_lock);
- list_add_tail(&di->lru, &mdsc->dentry_lru);
- mdsc->num_dentry++;
- spin_unlock(&mdsc->dentry_lru_lock);
- }
+ mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
+ spin_lock(&mdsc->dentry_lru_lock);
+ list_add_tail(&di->lru, &mdsc->dentry_lru);
+ mdsc->num_dentry++;
+ spin_unlock(&mdsc->dentry_lru_lock);
}
void ceph_dentry_lru_touch(struct dentry *dn)
@@ -1183,12 +1261,10 @@ void ceph_dentry_lru_touch(struct dentry *dn)
dout("dentry_lru_touch %p %p '%.*s' (offset %lld)\n", di, dn,
dn->d_name.len, dn->d_name.name, di->offset);
- if (di) {
- mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
- spin_lock(&mdsc->dentry_lru_lock);
- list_move_tail(&di->lru, &mdsc->dentry_lru);
- spin_unlock(&mdsc->dentry_lru_lock);
- }
+ mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
+ spin_lock(&mdsc->dentry_lru_lock);
+ list_move_tail(&di->lru, &mdsc->dentry_lru);
+ spin_unlock(&mdsc->dentry_lru_lock);
}
void ceph_dentry_lru_del(struct dentry *dn)
@@ -1198,18 +1274,35 @@ void ceph_dentry_lru_del(struct dentry *dn)
dout("dentry_lru_del %p %p '%.*s'\n", di, dn,
dn->d_name.len, dn->d_name.name);
- if (di) {
- mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
- spin_lock(&mdsc->dentry_lru_lock);
- list_del_init(&di->lru);
- mdsc->num_dentry--;
- spin_unlock(&mdsc->dentry_lru_lock);
+ mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
+ spin_lock(&mdsc->dentry_lru_lock);
+ list_del_init(&di->lru);
+ mdsc->num_dentry--;
+ spin_unlock(&mdsc->dentry_lru_lock);
+}
+
+/*
+ * Return name hash for a given dentry. This is dependent on
+ * the parent directory's hash function.
+ */
+unsigned ceph_dentry_hash(struct inode *dir, struct dentry *dn)
+{
+ struct ceph_inode_info *dci = ceph_inode(dir);
+
+ switch (dci->i_dir_layout.dl_dir_hash) {
+ case 0: /* for backward compat */
+ case CEPH_STR_HASH_LINUX:
+ return dn->d_name.hash;
+
+ default:
+ return ceph_str_hash(dci->i_dir_layout.dl_dir_hash,
+ dn->d_name.name, dn->d_name.len);
}
}
const struct file_operations ceph_dir_fops = {
.read = ceph_read_dir,
- .readdir = ceph_readdir,
+ .iterate = ceph_readdir,
.llseek = ceph_dir_llseek,
.open = ceph_open,
.release = ceph_release,
@@ -1226,6 +1319,8 @@ const struct inode_operations ceph_dir_iops = {
.getxattr = ceph_getxattr,
.listxattr = ceph_listxattr,
.removexattr = ceph_removexattr,
+ .get_acl = ceph_get_acl,
+ .set_acl = ceph_set_acl,
.mknod = ceph_mknod,
.symlink = ceph_symlink,
.mkdir = ceph_mkdir,
@@ -1234,18 +1329,21 @@ const struct inode_operations ceph_dir_iops = {
.rmdir = ceph_unlink,
.rename = ceph_rename,
.create = ceph_create,
+ .atomic_open = ceph_atomic_open,
};
const struct dentry_operations ceph_dentry_ops = {
.d_revalidate = ceph_d_revalidate,
- .d_release = ceph_dentry_release,
+ .d_release = ceph_d_release,
+ .d_prune = ceph_d_prune,
};
const struct dentry_operations ceph_snapdir_dentry_ops = {
.d_revalidate = ceph_snapdir_d_revalidate,
- .d_release = ceph_dentry_release,
+ .d_release = ceph_d_release,
};
const struct dentry_operations ceph_snap_dentry_ops = {
- .d_release = ceph_dentry_release,
+ .d_release = ceph_d_release,
+ .d_prune = ceph_d_prune,
};
diff --git a/fs/ceph/export.c b/fs/ceph/export.c
index 2297d942699..8d7d782f438 100644
--- a/fs/ceph/export.c
+++ b/fs/ceph/export.c
@@ -8,23 +8,6 @@
#include "mds_client.h"
/*
- * NFS export support
- *
- * NFS re-export of a ceph mount is, at present, only semireliable.
- * The basic issue is that the Ceph architectures doesn't lend itself
- * well to generating filehandles that will remain valid forever.
- *
- * So, we do our best. If you're lucky, your inode will be in the
- * client's cache. If it's not, and you have a connectable fh, then
- * the MDS server may be able to find it for you. Otherwise, you get
- * ESTALE.
- *
- * There are ways to this more reliable, but in the non-connectable fh
- * case, we won't every work perfectly, and in the connectable case,
- * some changes are needed on the MDS side to work better.
- */
-
-/*
* Basic fh
*/
struct ceph_nfs_fh {
@@ -32,22 +15,18 @@ struct ceph_nfs_fh {
} __attribute__ ((packed));
/*
- * Larger 'connectable' fh that includes parent ino and name hash.
- * Use this whenever possible, as it works more reliably.
+ * Larger fh that includes parent ino.
*/
struct ceph_nfs_confh {
u64 ino, parent_ino;
- u32 parent_name_hash;
} __attribute__ ((packed));
-static int ceph_encode_fh(struct dentry *dentry, u32 *rawfh, int *max_len,
- int connectable)
+static int ceph_encode_fh(struct inode *inode, u32 *rawfh, int *max_len,
+ struct inode *parent_inode)
{
int type;
struct ceph_nfs_fh *fh = (void *)rawfh;
struct ceph_nfs_confh *cfh = (void *)rawfh;
- struct dentry *parent = dentry->d_parent;
- struct inode *inode = dentry->d_inode;
int connected_handle_length = sizeof(*cfh)/4;
int handle_length = sizeof(*fh)/4;
@@ -55,176 +34,217 @@ static int ceph_encode_fh(struct dentry *dentry, u32 *rawfh, int *max_len,
if (ceph_snap(inode) != CEPH_NOSNAP)
return -EINVAL;
- if (*max_len >= connected_handle_length) {
- dout("encode_fh %p connectable\n", dentry);
- cfh->ino = ceph_ino(dentry->d_inode);
- cfh->parent_ino = ceph_ino(parent->d_inode);
- cfh->parent_name_hash = parent->d_name.hash;
+ if (parent_inode && (*max_len < connected_handle_length)) {
*max_len = connected_handle_length;
- type = 2;
- } else if (*max_len >= handle_length) {
- if (connectable) {
- *max_len = connected_handle_length;
- return 255;
- }
- dout("encode_fh %p\n", dentry);
- fh->ino = ceph_ino(dentry->d_inode);
+ return FILEID_INVALID;
+ } else if (*max_len < handle_length) {
*max_len = handle_length;
- type = 1;
+ return FILEID_INVALID;
+ }
+
+ if (parent_inode) {
+ dout("encode_fh %llx with parent %llx\n",
+ ceph_ino(inode), ceph_ino(parent_inode));
+ cfh->ino = ceph_ino(inode);
+ cfh->parent_ino = ceph_ino(parent_inode);
+ *max_len = connected_handle_length;
+ type = FILEID_INO32_GEN_PARENT;
} else {
+ dout("encode_fh %llx\n", ceph_ino(inode));
+ fh->ino = ceph_ino(inode);
*max_len = handle_length;
- return 255;
+ type = FILEID_INO32_GEN;
}
return type;
}
-/*
- * convert regular fh to dentry
- *
- * FIXME: we should try harder by querying the mds for the ino.
- */
-static struct dentry *__fh_to_dentry(struct super_block *sb,
- struct ceph_nfs_fh *fh)
+static struct dentry *__fh_to_dentry(struct super_block *sb, u64 ino)
{
+ struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc;
struct inode *inode;
struct dentry *dentry;
struct ceph_vino vino;
int err;
- dout("__fh_to_dentry %llx\n", fh->ino);
- vino.ino = fh->ino;
+ vino.ino = ino;
vino.snap = CEPH_NOSNAP;
inode = ceph_find_inode(sb, vino);
- if (!inode)
- return ERR_PTR(-ESTALE);
+ if (!inode) {
+ struct ceph_mds_request *req;
+
+ req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LOOKUPINO,
+ USE_ANY_MDS);
+ if (IS_ERR(req))
+ return ERR_CAST(req);
+
+ req->r_ino1 = vino;
+ req->r_num_caps = 1;
+ err = ceph_mdsc_do_request(mdsc, NULL, req);
+ inode = req->r_target_inode;
+ if (inode)
+ ihold(inode);
+ ceph_mdsc_put_request(req);
+ if (!inode)
+ return ERR_PTR(-ESTALE);
+ }
dentry = d_obtain_alias(inode);
if (IS_ERR(dentry)) {
- pr_err("fh_to_dentry %llx -- inode %p but ENOMEM\n",
- fh->ino, inode);
iput(inode);
return dentry;
}
err = ceph_init_dentry(dentry);
-
if (err < 0) {
- iput(inode);
+ dput(dentry);
return ERR_PTR(err);
}
- dout("__fh_to_dentry %llx %p dentry %p\n", fh->ino, inode, dentry);
+ dout("__fh_to_dentry %llx %p dentry %p\n", ino, inode, dentry);
return dentry;
}
/*
- * convert connectable fh to dentry
+ * convert regular fh to dentry
*/
-static struct dentry *__cfh_to_dentry(struct super_block *sb,
- struct ceph_nfs_confh *cfh)
+static struct dentry *ceph_fh_to_dentry(struct super_block *sb,
+ struct fid *fid,
+ int fh_len, int fh_type)
+{
+ struct ceph_nfs_fh *fh = (void *)fid->raw;
+
+ if (fh_type != FILEID_INO32_GEN &&
+ fh_type != FILEID_INO32_GEN_PARENT)
+ return NULL;
+ if (fh_len < sizeof(*fh) / 4)
+ return NULL;
+
+ dout("fh_to_dentry %llx\n", fh->ino);
+ return __fh_to_dentry(sb, fh->ino);
+}
+
+static struct dentry *__get_parent(struct super_block *sb,
+ struct dentry *child, u64 ino)
{
struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc;
+ struct ceph_mds_request *req;
struct inode *inode;
struct dentry *dentry;
- struct ceph_vino vino;
int err;
- dout("__cfh_to_dentry %llx (%llx/%x)\n",
- cfh->ino, cfh->parent_ino, cfh->parent_name_hash);
-
- vino.ino = cfh->ino;
- vino.snap = CEPH_NOSNAP;
- inode = ceph_find_inode(sb, vino);
- if (!inode) {
- struct ceph_mds_request *req;
-
- req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LOOKUPHASH,
- USE_ANY_MDS);
- if (IS_ERR(req))
- return ERR_CAST(req);
+ req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LOOKUPPARENT,
+ USE_ANY_MDS);
+ if (IS_ERR(req))
+ return ERR_CAST(req);
- req->r_ino1 = vino;
- req->r_ino2.ino = cfh->parent_ino;
- req->r_ino2.snap = CEPH_NOSNAP;
- req->r_path2 = kmalloc(16, GFP_NOFS);
- snprintf(req->r_path2, 16, "%d", cfh->parent_name_hash);
- req->r_num_caps = 1;
- err = ceph_mdsc_do_request(mdsc, NULL, req);
- ceph_mdsc_put_request(req);
- inode = ceph_find_inode(sb, vino);
- if (!inode)
- return ERR_PTR(err ? err : -ESTALE);
+ if (child) {
+ req->r_inode = child->d_inode;
+ ihold(child->d_inode);
+ } else {
+ req->r_ino1 = (struct ceph_vino) {
+ .ino = ino,
+ .snap = CEPH_NOSNAP,
+ };
}
+ req->r_num_caps = 1;
+ err = ceph_mdsc_do_request(mdsc, NULL, req);
+ inode = req->r_target_inode;
+ if (inode)
+ ihold(inode);
+ ceph_mdsc_put_request(req);
+ if (!inode)
+ return ERR_PTR(-ENOENT);
dentry = d_obtain_alias(inode);
if (IS_ERR(dentry)) {
- pr_err("cfh_to_dentry %llx -- inode %p but ENOMEM\n",
- cfh->ino, inode);
iput(inode);
return dentry;
}
err = ceph_init_dentry(dentry);
if (err < 0) {
- iput(inode);
+ dput(dentry);
return ERR_PTR(err);
}
- dout("__cfh_to_dentry %llx %p dentry %p\n", cfh->ino, inode, dentry);
+ dout("__get_parent ino %llx parent %p ino %llx.%llx\n",
+ child ? ceph_ino(child->d_inode) : ino,
+ dentry, ceph_vinop(inode));
return dentry;
}
-static struct dentry *ceph_fh_to_dentry(struct super_block *sb, struct fid *fid,
- int fh_len, int fh_type)
+static struct dentry *ceph_get_parent(struct dentry *child)
{
- if (fh_type == 1)
- return __fh_to_dentry(sb, (struct ceph_nfs_fh *)fid->raw);
- else
- return __cfh_to_dentry(sb, (struct ceph_nfs_confh *)fid->raw);
+ /* don't re-export snaps */
+ if (ceph_snap(child->d_inode) != CEPH_NOSNAP)
+ return ERR_PTR(-EINVAL);
+
+ dout("get_parent %p ino %llx.%llx\n",
+ child, ceph_vinop(child->d_inode));
+ return __get_parent(child->d_sb, child, 0);
}
/*
- * get parent, if possible.
- *
- * FIXME: we could do better by querying the mds to discover the
- * parent.
+ * convert regular fh to parent
*/
static struct dentry *ceph_fh_to_parent(struct super_block *sb,
- struct fid *fid,
+ struct fid *fid,
int fh_len, int fh_type)
{
struct ceph_nfs_confh *cfh = (void *)fid->raw;
- struct ceph_vino vino;
- struct inode *inode;
struct dentry *dentry;
- int err;
- if (fh_type == 1)
- return ERR_PTR(-ESTALE);
+ if (fh_type != FILEID_INO32_GEN_PARENT)
+ return NULL;
+ if (fh_len < sizeof(*cfh) / 4)
+ return NULL;
- pr_debug("fh_to_parent %llx/%d\n", cfh->parent_ino,
- cfh->parent_name_hash);
+ dout("fh_to_parent %llx\n", cfh->parent_ino);
+ dentry = __get_parent(sb, NULL, cfh->ino);
+ if (IS_ERR(dentry) && PTR_ERR(dentry) == -ENOENT)
+ dentry = __fh_to_dentry(sb, cfh->parent_ino);
+ return dentry;
+}
- vino.ino = cfh->ino;
- vino.snap = CEPH_NOSNAP;
- inode = ceph_find_inode(sb, vino);
- if (!inode)
- return ERR_PTR(-ESTALE);
+static int ceph_get_name(struct dentry *parent, char *name,
+ struct dentry *child)
+{
+ struct ceph_mds_client *mdsc;
+ struct ceph_mds_request *req;
+ int err;
- dentry = d_obtain_alias(inode);
- if (IS_ERR(dentry)) {
- pr_err("fh_to_parent %llx -- inode %p but ENOMEM\n",
- cfh->ino, inode);
- iput(inode);
- return dentry;
- }
- err = ceph_init_dentry(dentry);
- if (err < 0) {
- iput(inode);
- return ERR_PTR(err);
+ mdsc = ceph_inode_to_client(child->d_inode)->mdsc;
+ req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_LOOKUPNAME,
+ USE_ANY_MDS);
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+
+ mutex_lock(&parent->d_inode->i_mutex);
+
+ req->r_inode = child->d_inode;
+ ihold(child->d_inode);
+ req->r_ino2 = ceph_vino(parent->d_inode);
+ req->r_locked_dir = parent->d_inode;
+ req->r_num_caps = 2;
+ err = ceph_mdsc_do_request(mdsc, NULL, req);
+
+ mutex_unlock(&parent->d_inode->i_mutex);
+
+ if (!err) {
+ struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
+ memcpy(name, rinfo->dname, rinfo->dname_len);
+ name[rinfo->dname_len] = 0;
+ dout("get_name %p ino %llx.%llx name %s\n",
+ child, ceph_vinop(child->d_inode), name);
+ } else {
+ dout("get_name %p ino %llx.%llx err %d\n",
+ child, ceph_vinop(child->d_inode), err);
}
- dout("fh_to_parent %llx %p dentry %p\n", cfh->ino, inode, dentry);
- return dentry;
+
+ ceph_mdsc_put_request(req);
+ return err;
}
const struct export_operations ceph_export_ops = {
.encode_fh = ceph_encode_fh,
.fh_to_dentry = ceph_fh_to_dentry,
.fh_to_parent = ceph_fh_to_parent,
+ .get_parent = ceph_get_parent,
+ .get_name = ceph_get_name,
};
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index e77c28cf369..302085100c2 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -4,11 +4,15 @@
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/file.h>
+#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/writeback.h>
+#include <linux/aio.h>
+#include <linux/falloc.h>
#include "super.h"
#include "mds_client.h"
+#include "cache.h"
/*
* Ceph file operations
@@ -54,7 +58,6 @@ prepare_open_request(struct super_block *sb, int flags, int create_mode)
req->r_fmode = ceph_flags_to_mode(flags);
req->r_args.open.flags = cpu_to_le32(flags);
req->r_args.open.mode = cpu_to_le32(create_mode);
- req->r_args.open.preferred = cpu_to_le32(-1);
out:
return req;
}
@@ -67,9 +70,23 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
{
struct ceph_file_info *cf;
int ret = 0;
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
+ struct ceph_mds_client *mdsc = fsc->mdsc;
switch (inode->i_mode & S_IFMT) {
case S_IFREG:
+ /* First file open request creates the cookie, we want to keep
+ * this cookie around for the filetime of the inode as not to
+ * have to worry about fscache register / revoke / operation
+ * races.
+ *
+ * Also, if we know the operation is going to invalidate data
+ * (non readonly) just nuke the cache right away.
+ */
+ ceph_fscache_register_inode_cookie(mdsc->fsc, ci);
+ if ((fmode & CEPH_FILE_MODE_WR))
+ ceph_fscache_invalidate(inode);
case S_IFDIR:
dout("init_file %p %p 0%o (regular)\n", inode, file,
inode->i_mode);
@@ -107,9 +124,6 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
}
/*
- * If the filp already has private_data, that means the file was
- * already opened by intent during lookup, and we do nothing.
- *
* If we already have the requisite capabilities, we can satisfy
* the open request locally (no need to request new caps from the
* MDS). We do, however, need to inform the MDS (asynchronously)
@@ -122,7 +136,7 @@ int ceph_open(struct inode *inode, struct file *file)
struct ceph_mds_client *mdsc = fsc->mdsc;
struct ceph_mds_request *req;
struct ceph_file_info *cf = file->private_data;
- struct inode *parent_inode = file->f_dentry->d_parent->d_inode;
+ struct inode *parent_inode = NULL;
int err;
int flags, fmode, wanted;
@@ -147,18 +161,20 @@ int ceph_open(struct inode *inode, struct file *file)
/* trivially open snapdir */
if (ceph_snap(inode) == CEPH_SNAPDIR) {
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
__ceph_get_fmode(ci, fmode);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return ceph_init_file(inode, file, fmode);
}
/*
- * No need to block if we have any caps. Update wanted set
+ * No need to block if we have caps on the auth MDS (for
+ * write) or any MDS (for read). Update wanted set
* asynchronously.
*/
- spin_lock(&inode->i_lock);
- if (__ceph_is_any_real_caps(ci)) {
+ spin_lock(&ci->i_ceph_lock);
+ if (__ceph_is_any_real_caps(ci) &&
+ (((fmode & CEPH_FILE_MODE_WR) == 0) || ci->i_auth_cap)) {
int mds_wanted = __ceph_caps_mds_wanted(ci);
int issued = __ceph_caps_issued(ci, NULL);
@@ -166,7 +182,7 @@ int ceph_open(struct inode *inode, struct file *file)
inode, fmode, ceph_cap_string(wanted),
ceph_cap_string(issued));
__ceph_get_fmode(ci, fmode);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
/* adjust wanted? */
if ((issued & wanted) != wanted &&
@@ -178,10 +194,11 @@ int ceph_open(struct inode *inode, struct file *file)
} else if (ceph_snap(inode) != CEPH_NOSNAP &&
(ci->i_snap_caps & wanted) == wanted) {
__ceph_get_fmode(ci, fmode);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return ceph_init_file(inode, file, fmode);
}
- spin_unlock(&inode->i_lock);
+
+ spin_unlock(&ci->i_ceph_lock);
dout("open fmode %d wants %s\n", fmode, ceph_cap_string(wanted));
req = prepare_open_request(inode->i_sb, flags, 0);
@@ -189,9 +206,14 @@ int ceph_open(struct inode *inode, struct file *file)
err = PTR_ERR(req);
goto out;
}
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
+
req->r_num_caps = 1;
+ if (flags & O_CREAT)
+ parent_inode = ceph_get_dentry_parent_inode(file->f_dentry);
err = ceph_mdsc_do_request(mdsc, parent_inode, req);
+ iput(parent_inode);
if (!err)
err = ceph_init_file(inode, file, req->r_fmode);
ceph_mdsc_put_request(req);
@@ -202,36 +224,34 @@ out:
/*
- * Do a lookup + open with a single request.
- *
- * If this succeeds, but some subsequent check in the vfs
- * may_open() fails, the struct *file gets cleaned up (i.e.
- * ceph_release gets called). So fear not!
- */
-/*
- * flags
- * path_lookup_open -> LOOKUP_OPEN
- * path_lookup_create -> LOOKUP_OPEN|LOOKUP_CREATE
+ * Do a lookup + open with a single request. If we get a non-existent
+ * file or symlink, return 1 so the VFS can retry.
*/
-struct dentry *ceph_lookup_open(struct inode *dir, struct dentry *dentry,
- struct nameidata *nd, int mode,
- int locked_dir)
+int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
+ struct file *file, unsigned flags, umode_t mode,
+ int *opened)
{
struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
struct ceph_mds_client *mdsc = fsc->mdsc;
- struct file *file = nd->intent.open.file;
- struct inode *parent_inode = get_dentry_parent_inode(file->f_dentry);
struct ceph_mds_request *req;
+ struct dentry *dn;
int err;
- int flags = nd->intent.open.flags - 1; /* silly vfs! */
- dout("ceph_lookup_open dentry %p '%.*s' flags %d mode 0%o\n",
- dentry, dentry->d_name.len, dentry->d_name.name, flags, mode);
+ dout("atomic_open %p dentry %p '%.*s' %s flags %d mode 0%o\n",
+ dir, dentry, dentry->d_name.len, dentry->d_name.name,
+ d_unhashed(dentry) ? "unhashed" : "hashed", flags, mode);
+
+ if (dentry->d_name.len > NAME_MAX)
+ return -ENAMETOOLONG;
+
+ err = ceph_init_dentry(dentry);
+ if (err < 0)
+ return err;
/* do the open */
req = prepare_open_request(dir->i_sb, flags, mode);
if (IS_ERR(req))
- return ERR_CAST(req);
+ return PTR_ERR(req);
req->r_dentry = dget(dentry);
req->r_num_caps = 2;
if (flags & O_CREAT) {
@@ -239,16 +259,44 @@ struct dentry *ceph_lookup_open(struct inode *dir, struct dentry *dentry,
req->r_dentry_unless = CEPH_CAP_FILE_EXCL;
}
req->r_locked_dir = dir; /* caller holds dir->i_mutex */
- err = ceph_mdsc_do_request(mdsc, parent_inode, req);
- dentry = ceph_finish_lookup(req, dentry, err);
- if (!err && (flags & O_CREAT) && !req->r_reply_info.head->is_dentry)
+ err = ceph_mdsc_do_request(mdsc,
+ (flags & (O_CREAT|O_TRUNC)) ? dir : NULL,
+ req);
+ if (err)
+ goto out_err;
+
+ err = ceph_handle_snapdir(req, dentry, err);
+ if (err == 0 && (flags & O_CREAT) && !req->r_reply_info.head->is_dentry)
err = ceph_handle_notrace_create(dir, dentry);
- if (!err)
- err = ceph_init_file(req->r_dentry->d_inode, file,
- req->r_fmode);
+
+ if (d_unhashed(dentry)) {
+ dn = ceph_finish_lookup(req, dentry, err);
+ if (IS_ERR(dn))
+ err = PTR_ERR(dn);
+ } else {
+ /* we were given a hashed negative dentry */
+ dn = NULL;
+ }
+ if (err)
+ goto out_err;
+ if (dn || dentry->d_inode == NULL || S_ISLNK(dentry->d_inode->i_mode)) {
+ /* make vfs retry on splice, ENOENT, or symlink */
+ dout("atomic_open finish_no_open on dn %p\n", dn);
+ err = finish_no_open(file, dn);
+ } else {
+ dout("atomic_open finish_open on dn %p\n", dn);
+ if (req->r_op == CEPH_MDS_OP_CREATE && req->r_reply_info.has_create_ino) {
+ ceph_init_acl(dentry, dentry->d_inode, dir);
+ *opened |= FILE_CREATED;
+ }
+ err = finish_open(file, dentry, ceph_open, opened);
+ }
+out_err:
+ if (!req->r_err && req->r_target_inode)
+ ceph_put_fmode(ceph_inode(req->r_target_inode), req->r_fmode);
ceph_mdsc_put_request(req);
- dout("ceph_lookup_open result=%p\n", dentry);
- return dentry;
+ dout("atomic_open result=%d\n", err);
+ return err;
}
int ceph_release(struct inode *inode, struct file *file)
@@ -280,13 +328,14 @@ int ceph_release(struct inode *inode, struct file *file)
static int striped_read(struct inode *inode,
u64 off, u64 len,
struct page **pages, int num_pages,
- int *checkeof)
+ int *checkeof, bool o_direct,
+ unsigned long buf_align)
{
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
struct ceph_inode_info *ci = ceph_inode(inode);
- u64 pos, this_len;
- int page_off = off & ~PAGE_CACHE_MASK; /* first byte's offset in page */
- int left, pages_left;
+ u64 pos, this_len, left;
+ int io_align, page_align;
+ int pages_left;
int read;
struct page **page_pos;
int ret;
@@ -300,58 +349,57 @@ static int striped_read(struct inode *inode,
page_pos = pages;
pages_left = num_pages;
read = 0;
+ io_align = off & ~PAGE_MASK;
more:
+ if (o_direct)
+ page_align = (pos - io_align + buf_align) & ~PAGE_MASK;
+ else
+ page_align = pos & ~PAGE_MASK;
this_len = left;
ret = ceph_osdc_readpages(&fsc->client->osdc, ceph_vino(inode),
&ci->i_layout, pos, &this_len,
ci->i_truncate_seq,
ci->i_truncate_size,
- page_pos, pages_left);
- hit_stripe = this_len < left;
- was_short = ret >= 0 && ret < this_len;
+ page_pos, pages_left, page_align);
if (ret == -ENOENT)
ret = 0;
- dout("striped_read %llu~%u (read %u) got %d%s%s\n", pos, left, read,
+ hit_stripe = this_len < left;
+ was_short = ret >= 0 && ret < this_len;
+ dout("striped_read %llu~%llu (read %u) got %d%s%s\n", pos, left, read,
ret, hit_stripe ? " HITSTRIPE" : "", was_short ? " SHORT" : "");
- if (ret > 0) {
- int didpages =
- ((pos & ~PAGE_CACHE_MASK) + ret) >> PAGE_CACHE_SHIFT;
-
- if (read < pos - off) {
- dout(" zero gap %llu to %llu\n", off + read, pos);
- ceph_zero_page_vector_range(page_off + read,
- pos - off - read, pages);
+ if (ret >= 0) {
+ int didpages;
+ if (was_short && (pos + ret < inode->i_size)) {
+ u64 tmp = min(this_len - ret,
+ inode->i_size - pos - ret);
+ dout(" zero gap %llu to %llu\n",
+ pos + ret, pos + ret + tmp);
+ ceph_zero_page_vector_range(page_align + read + ret,
+ tmp, pages);
+ ret += tmp;
}
+
+ didpages = (page_align + ret) >> PAGE_CACHE_SHIFT;
pos += ret;
read = pos - off;
left -= ret;
page_pos += didpages;
pages_left -= didpages;
- /* hit stripe? */
- if (left && hit_stripe)
+ /* hit stripe and need continue*/
+ if (left && hit_stripe && pos < inode->i_size)
goto more;
}
- if (was_short) {
- /* was original extent fully inside i_size? */
- if (pos + left <= inode->i_size) {
- dout("zero tail\n");
- ceph_zero_page_vector_range(page_off + read, len - read,
- pages);
- read = len;
- goto out;
- }
-
- /* check i_size */
- *checkeof = 1;
+ if (read > 0) {
+ ret = read;
+ /* did we bounce off eof? */
+ if (pos + left > inode->i_size)
+ *checkeof = 1;
}
-out:
- if (ret >= 0)
- ret = read;
dout("striped_read returns %d\n", ret);
return ret;
}
@@ -362,213 +410,361 @@ out:
*
* If the read spans object boundary, just do multiple reads.
*/
-static ssize_t ceph_sync_read(struct file *file, char __user *data,
- unsigned len, loff_t *poff, int *checkeof)
+static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *i,
+ int *checkeof)
{
- struct inode *inode = file->f_dentry->d_inode;
+ struct file *file = iocb->ki_filp;
+ struct inode *inode = file_inode(file);
struct page **pages;
- u64 off = *poff;
- int num_pages = calc_pages_for(off, len);
- int ret;
+ u64 off = iocb->ki_pos;
+ int num_pages, ret;
+ size_t len = iov_iter_count(i);
- dout("sync_read on file %p %llu~%u %s\n", file, off, len,
+ dout("sync_read on file %p %llu~%u %s\n", file, off,
+ (unsigned)len,
(file->f_flags & O_DIRECT) ? "O_DIRECT" : "");
+ /*
+ * flush any page cache pages in this range. this
+ * will make concurrent normal and sync io slow,
+ * but it will at least behave sensibly when they are
+ * in sequence.
+ */
+ ret = filemap_write_and_wait_range(inode->i_mapping, off,
+ off + len);
+ if (ret < 0)
+ return ret;
if (file->f_flags & O_DIRECT) {
- pages = ceph_get_direct_page_vector(data, num_pages, off, len);
+ while (iov_iter_count(i)) {
+ size_t start;
+ ssize_t n;
- /*
- * flush any page cache pages in this range. this
- * will make concurrent normal and O_DIRECT io slow,
- * but it will at least behave sensibly when they are
- * in sequence.
- */
- } else {
- pages = ceph_alloc_page_vector(num_pages, GFP_NOFS);
- }
- if (IS_ERR(pages))
- return PTR_ERR(pages);
+ n = iov_iter_get_pages_alloc(i, &pages, INT_MAX, &start);
+ if (n < 0)
+ return n;
- ret = filemap_write_and_wait(inode->i_mapping);
- if (ret < 0)
- goto done;
+ num_pages = (n + start + PAGE_SIZE - 1) / PAGE_SIZE;
- ret = striped_read(inode, off, len, pages, num_pages, checkeof);
+ ret = striped_read(inode, off, n,
+ pages, num_pages, checkeof,
+ 1, start);
- if (ret >= 0 && (file->f_flags & O_DIRECT) == 0)
- ret = ceph_copy_page_vector_to_user(pages, data, off, ret);
- if (ret >= 0)
- *poff = off + ret;
+ ceph_put_page_vector(pages, num_pages, true);
-done:
- if (file->f_flags & O_DIRECT)
- ceph_put_page_vector(pages, num_pages);
- else
+ if (ret <= 0)
+ break;
+ off += ret;
+ iov_iter_advance(i, ret);
+ if (ret < n)
+ break;
+ }
+ } else {
+ num_pages = calc_pages_for(off, len);
+ pages = ceph_alloc_page_vector(num_pages, GFP_NOFS);
+ if (IS_ERR(pages))
+ return PTR_ERR(pages);
+ ret = striped_read(inode, off, len, pages,
+ num_pages, checkeof, 0, 0);
+ if (ret > 0) {
+ int l, k = 0;
+ size_t left = ret;
+
+ while (left) {
+ int copy = min_t(size_t, PAGE_SIZE, left);
+ l = copy_page_to_iter(pages[k++], 0, copy, i);
+ off += l;
+ left -= l;
+ if (l < copy)
+ break;
+ }
+ }
ceph_release_page_vector(pages, num_pages);
+ }
+
+ if (off > iocb->ki_pos) {
+ ret = off - iocb->ki_pos;
+ iocb->ki_pos = off;
+ }
+
dout("sync_read result %d\n", ret);
return ret;
}
/*
- * Write commit callback, called if we requested both an ACK and
- * ONDISK commit reply from the OSD.
+ * Write commit request unsafe callback, called to tell us when a
+ * request is unsafe (that is, in flight--has been handed to the
+ * messenger to send to its target osd). It is called again when
+ * we've received a response message indicating the request is
+ * "safe" (its CEPH_OSD_FLAG_ONDISK flag is set), or when a request
+ * is completed early (and unsuccessfully) due to a timeout or
+ * interrupt.
+ *
+ * This is used if we requested both an ACK and ONDISK commit reply
+ * from the OSD.
*/
-static void sync_write_commit(struct ceph_osd_request *req,
- struct ceph_msg *msg)
+static void ceph_sync_write_unsafe(struct ceph_osd_request *req, bool unsafe)
{
struct ceph_inode_info *ci = ceph_inode(req->r_inode);
- dout("sync_write_commit %p tid %llu\n", req, req->r_tid);
- spin_lock(&ci->i_unsafe_lock);
- list_del_init(&req->r_unsafe_item);
- spin_unlock(&ci->i_unsafe_lock);
- ceph_put_cap_refs(ci, CEPH_CAP_FILE_WR);
+ dout("%s %p tid %llu %ssafe\n", __func__, req, req->r_tid,
+ unsafe ? "un" : "");
+ if (unsafe) {
+ ceph_get_cap_refs(ci, CEPH_CAP_FILE_WR);
+ spin_lock(&ci->i_unsafe_lock);
+ list_add_tail(&req->r_unsafe_item,
+ &ci->i_unsafe_writes);
+ spin_unlock(&ci->i_unsafe_lock);
+ } else {
+ spin_lock(&ci->i_unsafe_lock);
+ list_del_init(&req->r_unsafe_item);
+ spin_unlock(&ci->i_unsafe_lock);
+ ceph_put_cap_refs(ci, CEPH_CAP_FILE_WR);
+ }
}
+
/*
- * Synchronous write, straight from __user pointer or user pages (if
- * O_DIRECT).
+ * Synchronous write, straight from __user pointer or user pages.
*
* If write spans object boundary, just do multiple writes. (For a
* correct atomic write, we should e.g. take write locks on all
* objects, rollback on failure, etc.)
*/
-static ssize_t ceph_sync_write(struct file *file, const char __user *data,
- size_t left, loff_t *offset)
+static ssize_t
+ceph_sync_direct_write(struct kiocb *iocb, struct iov_iter *from)
{
- struct inode *inode = file->f_dentry->d_inode;
+ struct file *file = iocb->ki_filp;
+ struct inode *inode = file_inode(file);
struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+ struct ceph_snap_context *snapc;
+ struct ceph_vino vino;
struct ceph_osd_request *req;
struct page **pages;
int num_pages;
- long long unsigned pos;
- u64 len;
int written = 0;
int flags;
- int do_sync = 0;
int check_caps = 0;
int ret;
struct timespec mtime = CURRENT_TIME;
+ loff_t pos = iocb->ki_pos;
+ size_t count = iov_iter_count(from);
- if (ceph_snap(file->f_dentry->d_inode) != CEPH_NOSNAP)
+ if (ceph_snap(file_inode(file)) != CEPH_NOSNAP)
return -EROFS;
- dout("sync_write on file %p %lld~%u %s\n", file, *offset,
- (unsigned)left, (file->f_flags & O_DIRECT) ? "O_DIRECT" : "");
-
- if (file->f_flags & O_APPEND)
- pos = i_size_read(inode);
- else
- pos = *offset;
+ dout("sync_direct_write on file %p %lld~%u\n", file, pos,
+ (unsigned)count);
- ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + left);
+ ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + count);
if (ret < 0)
return ret;
ret = invalidate_inode_pages2_range(inode->i_mapping,
pos >> PAGE_CACHE_SHIFT,
- (pos + left) >> PAGE_CACHE_SHIFT);
+ (pos + count) >> PAGE_CACHE_SHIFT);
if (ret < 0)
dout("invalidate_inode_pages2_range returned %d\n", ret);
flags = CEPH_OSD_FLAG_ORDERSNAP |
CEPH_OSD_FLAG_ONDISK |
CEPH_OSD_FLAG_WRITE;
- if ((file->f_flags & (O_SYNC|O_DIRECT)) == 0)
- flags |= CEPH_OSD_FLAG_ACK;
- else
- do_sync = 1;
- /*
- * we may need to do multiple writes here if we span an object
- * boundary. this isn't atomic, unfortunately. :(
- */
-more:
- len = left;
- req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
- ceph_vino(inode), pos, &len,
- CEPH_OSD_OP_WRITE, flags,
- ci->i_snap_realm->cached_context,
- do_sync,
- ci->i_truncate_seq, ci->i_truncate_size,
- &mtime, false, 2);
- if (!req)
- return -ENOMEM;
-
- num_pages = calc_pages_for(pos, len);
+ while (iov_iter_count(from) > 0) {
+ u64 len = iov_iter_single_seg_count(from);
+ size_t start;
+ ssize_t n;
+
+ snapc = ci->i_snap_realm->cached_context;
+ vino = ceph_vino(inode);
+ req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
+ vino, pos, &len,
+ 2,/*include a 'startsync' command*/
+ CEPH_OSD_OP_WRITE, flags, snapc,
+ ci->i_truncate_seq,
+ ci->i_truncate_size,
+ false);
+ if (IS_ERR(req)) {
+ ret = PTR_ERR(req);
+ break;
+ }
- if (file->f_flags & O_DIRECT) {
- pages = ceph_get_direct_page_vector(data, num_pages, pos, len);
- if (IS_ERR(pages)) {
- ret = PTR_ERR(pages);
- goto out;
+ n = iov_iter_get_pages_alloc(from, &pages, len, &start);
+ if (unlikely(n < 0)) {
+ ret = n;
+ ceph_osdc_put_request(req);
+ break;
}
+ num_pages = (n + start + PAGE_SIZE - 1) / PAGE_SIZE;
/*
* throw out any page cache pages in this range. this
* may block.
*/
truncate_inode_pages_range(inode->i_mapping, pos,
- (pos+len) | (PAGE_CACHE_SIZE-1));
- } else {
+ (pos+n) | (PAGE_CACHE_SIZE-1));
+ osd_req_op_extent_osd_data_pages(req, 0, pages, n, start,
+ false, false);
+
+ /* BUG_ON(vino.snap != CEPH_NOSNAP); */
+ ceph_osdc_build_request(req, pos, snapc, vino.snap, &mtime);
+
+ ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
+ if (!ret)
+ ret = ceph_osdc_wait_request(&fsc->client->osdc, req);
+
+ ceph_put_page_vector(pages, num_pages, false);
+
+ ceph_osdc_put_request(req);
+ if (ret)
+ break;
+ pos += n;
+ written += n;
+ iov_iter_advance(from, n);
+
+ if (pos > i_size_read(inode)) {
+ check_caps = ceph_inode_set_size(inode, pos);
+ if (check_caps)
+ ceph_check_caps(ceph_inode(inode),
+ CHECK_CAPS_AUTHONLY,
+ NULL);
+ }
+ }
+
+ if (ret != -EOLDSNAPC && written > 0) {
+ iocb->ki_pos = pos;
+ ret = written;
+ }
+ return ret;
+}
+
+
+/*
+ * Synchronous write, straight from __user pointer or user pages.
+ *
+ * If write spans object boundary, just do multiple writes. (For a
+ * correct atomic write, we should e.g. take write locks on all
+ * objects, rollback on failure, etc.)
+ */
+static ssize_t ceph_sync_write(struct kiocb *iocb, struct iov_iter *from)
+{
+ struct file *file = iocb->ki_filp;
+ struct inode *inode = file_inode(file);
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+ struct ceph_snap_context *snapc;
+ struct ceph_vino vino;
+ struct ceph_osd_request *req;
+ struct page **pages;
+ u64 len;
+ int num_pages;
+ int written = 0;
+ int flags;
+ int check_caps = 0;
+ int ret;
+ struct timespec mtime = CURRENT_TIME;
+ loff_t pos = iocb->ki_pos;
+ size_t count = iov_iter_count(from);
+
+ if (ceph_snap(file_inode(file)) != CEPH_NOSNAP)
+ return -EROFS;
+
+ dout("sync_write on file %p %lld~%u\n", file, pos, (unsigned)count);
+
+ ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + count);
+ if (ret < 0)
+ return ret;
+
+ ret = invalidate_inode_pages2_range(inode->i_mapping,
+ pos >> PAGE_CACHE_SHIFT,
+ (pos + count) >> PAGE_CACHE_SHIFT);
+ if (ret < 0)
+ dout("invalidate_inode_pages2_range returned %d\n", ret);
+
+ flags = CEPH_OSD_FLAG_ORDERSNAP |
+ CEPH_OSD_FLAG_ONDISK |
+ CEPH_OSD_FLAG_WRITE |
+ CEPH_OSD_FLAG_ACK;
+
+ while ((len = iov_iter_count(from)) > 0) {
+ size_t left;
+ int n;
+
+ snapc = ci->i_snap_realm->cached_context;
+ vino = ceph_vino(inode);
+ req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
+ vino, pos, &len, 1,
+ CEPH_OSD_OP_WRITE, flags, snapc,
+ ci->i_truncate_seq,
+ ci->i_truncate_size,
+ false);
+ if (IS_ERR(req)) {
+ ret = PTR_ERR(req);
+ break;
+ }
+
+ /*
+ * write from beginning of first page,
+ * regardless of io alignment
+ */
+ num_pages = (len + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
+
pages = ceph_alloc_page_vector(num_pages, GFP_NOFS);
if (IS_ERR(pages)) {
ret = PTR_ERR(pages);
goto out;
}
- ret = ceph_copy_user_to_page_vector(pages, data, pos, len);
+
+ left = len;
+ for (n = 0; n < num_pages; n++) {
+ size_t plen = min_t(size_t, left, PAGE_SIZE);
+ ret = copy_page_from_iter(pages[n], 0, plen, from);
+ if (ret != plen) {
+ ret = -EFAULT;
+ break;
+ }
+ left -= ret;
+ }
+
if (ret < 0) {
ceph_release_page_vector(pages, num_pages);
goto out;
}
- if ((file->f_flags & O_SYNC) == 0) {
- /* get a second commit callback */
- req->r_safe_callback = sync_write_commit;
- req->r_own_pages = 1;
- }
- }
- req->r_pages = pages;
- req->r_num_pages = num_pages;
- req->r_inode = inode;
+ /* get a second commit callback */
+ req->r_unsafe_callback = ceph_sync_write_unsafe;
+ req->r_inode = inode;
- ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
- if (!ret) {
- if (req->r_safe_callback) {
- /*
- * Add to inode unsafe list only after we
- * start_request so that a tid has been assigned.
- */
- spin_lock(&ci->i_unsafe_lock);
- list_add(&req->r_unsafe_item, &ci->i_unsafe_writes);
- spin_unlock(&ci->i_unsafe_lock);
- ceph_get_cap_refs(ci, CEPH_CAP_FILE_WR);
- }
- ret = ceph_osdc_wait_request(&fsc->client->osdc, req);
- }
+ osd_req_op_extent_osd_data_pages(req, 0, pages, len, 0,
+ false, true);
- if (file->f_flags & O_DIRECT)
- ceph_put_page_vector(pages, num_pages);
- else if (file->f_flags & O_SYNC)
- ceph_release_page_vector(pages, num_pages);
+ /* BUG_ON(vino.snap != CEPH_NOSNAP); */
+ ceph_osdc_build_request(req, pos, snapc, vino.snap, &mtime);
+
+ ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
+ if (!ret)
+ ret = ceph_osdc_wait_request(&fsc->client->osdc, req);
out:
- ceph_osdc_put_request(req);
- if (ret == 0) {
- pos += len;
- written += len;
- left -= len;
- if (left)
- goto more;
+ ceph_osdc_put_request(req);
+ if (ret == 0) {
+ pos += len;
+ written += len;
+
+ if (pos > i_size_read(inode)) {
+ check_caps = ceph_inode_set_size(inode, pos);
+ if (check_caps)
+ ceph_check_caps(ceph_inode(inode),
+ CHECK_CAPS_AUTHONLY,
+ NULL);
+ }
+ } else
+ break;
+ }
+ if (ret != -EOLDSNAPC && written > 0) {
ret = written;
- *offset = pos;
- if (pos > i_size_read(inode))
- check_caps = ceph_inode_set_size(inode, pos);
- if (check_caps)
- ceph_check_caps(ceph_inode(inode), CHECK_CAPS_AUTHONLY,
- NULL);
+ iocb->ki_pos = pos;
}
return ret;
}
@@ -580,61 +776,69 @@ out:
*
* Hmm, the sync read case isn't actually async... should it be?
*/
-static ssize_t ceph_aio_read(struct kiocb *iocb, const struct iovec *iov,
- unsigned long nr_segs, loff_t pos)
+static ssize_t ceph_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
struct file *filp = iocb->ki_filp;
struct ceph_file_info *fi = filp->private_data;
- loff_t *ppos = &iocb->ki_pos;
- size_t len = iov->iov_len;
- struct inode *inode = filp->f_dentry->d_inode;
+ size_t len = iocb->ki_nbytes;
+ struct inode *inode = file_inode(filp);
struct ceph_inode_info *ci = ceph_inode(inode);
- void __user *base = iov->iov_base;
ssize_t ret;
int want, got = 0;
int checkeof = 0, read = 0;
- dout("aio_read %p %llx.%llx %llu~%u trying to get caps on %p\n",
- inode, ceph_vinop(inode), pos, (unsigned)len, inode);
again:
- __ceph_do_pending_vmtruncate(inode);
+ dout("aio_read %p %llx.%llx %llu~%u trying to get caps on %p\n",
+ inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len, inode);
+
if (fi->fmode & CEPH_FILE_MODE_LAZY)
want = CEPH_CAP_FILE_CACHE | CEPH_CAP_FILE_LAZYIO;
else
want = CEPH_CAP_FILE_CACHE;
ret = ceph_get_caps(ci, CEPH_CAP_FILE_RD, want, &got, -1);
if (ret < 0)
- goto out;
- dout("aio_read %p %llx.%llx %llu~%u got cap refs on %s\n",
- inode, ceph_vinop(inode), pos, (unsigned)len,
- ceph_cap_string(got));
+ return ret;
if ((got & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0 ||
(iocb->ki_filp->f_flags & O_DIRECT) ||
- (inode->i_sb->s_flags & MS_SYNCHRONOUS))
+ (fi->flags & CEPH_F_SYNC)) {
+
+ dout("aio_sync_read %p %llx.%llx %llu~%u got cap refs on %s\n",
+ inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len,
+ ceph_cap_string(got));
+
/* hmm, this isn't really async... */
- ret = ceph_sync_read(filp, base, len, ppos, &checkeof);
- else
- ret = generic_file_aio_read(iocb, iov, nr_segs, pos);
+ ret = ceph_sync_read(iocb, to, &checkeof);
+ } else {
+ dout("aio_read %p %llx.%llx %llu~%u got cap refs on %s\n",
+ inode, ceph_vinop(inode), iocb->ki_pos, (unsigned)len,
+ ceph_cap_string(got));
-out:
+ ret = generic_file_read_iter(iocb, to);
+ }
dout("aio_read %p %llx.%llx dropping cap refs on %s = %d\n",
inode, ceph_vinop(inode), ceph_cap_string(got), (int)ret);
ceph_put_cap_refs(ci, got);
if (checkeof && ret >= 0) {
- int statret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE);
+ int statret = ceph_do_getattr(inode,
+ CEPH_STAT_CAP_SIZE);
/* hit EOF or hole? */
- if (statret == 0 && *ppos < inode->i_size) {
- dout("aio_read sync_read hit hole, reading more\n");
+ if (statret == 0 && iocb->ki_pos < inode->i_size &&
+ ret < len) {
+ dout("sync_read hit hole, ppos %lld < size %lld"
+ ", reading more\n", iocb->ki_pos,
+ inode->i_size);
+
+ iov_iter_advance(to, ret);
read += ret;
- base += ret;
len -= ret;
checkeof = 0;
goto again;
}
}
+
if (ret >= 0)
ret += read;
@@ -651,95 +855,150 @@ out:
*
* If we are near ENOSPC, write synchronously.
*/
-static ssize_t ceph_aio_write(struct kiocb *iocb, const struct iovec *iov,
- unsigned long nr_segs, loff_t pos)
+static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
struct file *file = iocb->ki_filp;
struct ceph_file_info *fi = file->private_data;
- struct inode *inode = file->f_dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_osd_client *osdc =
&ceph_sb_to_client(inode->i_sb)->client->osdc;
- loff_t endoff = pos + iov->iov_len;
- int want, got = 0;
- int ret, err;
+ ssize_t count = iov_iter_count(from), written = 0;
+ int err, want, got;
+ loff_t pos = iocb->ki_pos;
if (ceph_snap(inode) != CEPH_NOSNAP)
return -EROFS;
+ mutex_lock(&inode->i_mutex);
+
+ /* We can write back this queue in page reclaim */
+ current->backing_dev_info = file->f_mapping->backing_dev_info;
+
+ err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode));
+ if (err)
+ goto out;
+
+ if (count == 0)
+ goto out;
+ iov_iter_truncate(from, count);
+
+ err = file_remove_suid(file);
+ if (err)
+ goto out;
+
+ err = file_update_time(file);
+ if (err)
+ goto out;
+
retry_snap:
- if (ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_FULL))
- return -ENOSPC;
- __ceph_do_pending_vmtruncate(inode);
- dout("aio_write %p %llx.%llx %llu~%u getting caps. i_size %llu\n",
- inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len,
- inode->i_size);
+ if (ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_FULL)) {
+ err = -ENOSPC;
+ goto out;
+ }
+
+ dout("aio_write %p %llx.%llx %llu~%zd getting caps. i_size %llu\n",
+ inode, ceph_vinop(inode), pos, count, inode->i_size);
if (fi->fmode & CEPH_FILE_MODE_LAZY)
want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO;
else
want = CEPH_CAP_FILE_BUFFER;
- ret = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, &got, endoff);
- if (ret < 0)
+ got = 0;
+ err = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, &got, pos + count);
+ if (err < 0)
goto out;
- dout("aio_write %p %llx.%llx %llu~%u got cap refs on %s\n",
- inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len,
- ceph_cap_string(got));
+ dout("aio_write %p %llx.%llx %llu~%zd got cap refs on %s\n",
+ inode, ceph_vinop(inode), pos, count, ceph_cap_string(got));
if ((got & (CEPH_CAP_FILE_BUFFER|CEPH_CAP_FILE_LAZYIO)) == 0 ||
- (iocb->ki_filp->f_flags & O_DIRECT) ||
- (inode->i_sb->s_flags & MS_SYNCHRONOUS)) {
- ret = ceph_sync_write(file, iov->iov_base, iov->iov_len,
- &iocb->ki_pos);
- } else {
- ret = generic_file_aio_write(iocb, iov, nr_segs, pos);
-
- if ((ret >= 0 || ret == -EIOCBQUEUED) &&
- ((file->f_flags & O_SYNC) || IS_SYNC(file->f_mapping->host)
- || ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_NEARFULL))) {
- err = vfs_fsync_range(file, pos, pos + ret - 1, 1);
- if (err < 0)
- ret = err;
+ (file->f_flags & O_DIRECT) || (fi->flags & CEPH_F_SYNC)) {
+ struct iov_iter data;
+ mutex_unlock(&inode->i_mutex);
+ /* we might need to revert back to that point */
+ data = *from;
+ if (file->f_flags & O_DIRECT)
+ written = ceph_sync_direct_write(iocb, &data);
+ else
+ written = ceph_sync_write(iocb, &data);
+ if (written == -EOLDSNAPC) {
+ dout("aio_write %p %llx.%llx %llu~%u"
+ "got EOLDSNAPC, retrying\n",
+ inode, ceph_vinop(inode),
+ pos, (unsigned)count);
+ mutex_lock(&inode->i_mutex);
+ goto retry_snap;
}
+ if (written > 0)
+ iov_iter_advance(from, written);
+ } else {
+ loff_t old_size = inode->i_size;
+ /*
+ * No need to acquire the i_truncate_mutex. Because
+ * the MDS revokes Fwb caps before sending truncate
+ * message to us. We can't get Fwb cap while there
+ * are pending vmtruncate. So write and vmtruncate
+ * can not run at the same time
+ */
+ written = generic_perform_write(file, from, pos);
+ if (likely(written >= 0))
+ iocb->ki_pos = pos + written;
+ if (inode->i_size > old_size)
+ ceph_fscache_update_objectsize(inode);
+ mutex_unlock(&inode->i_mutex);
}
- if (ret >= 0) {
- spin_lock(&inode->i_lock);
- __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
- spin_unlock(&inode->i_lock);
+
+ if (written >= 0) {
+ int dirty;
+ spin_lock(&ci->i_ceph_lock);
+ dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
+ spin_unlock(&ci->i_ceph_lock);
+ if (dirty)
+ __mark_inode_dirty(inode, dirty);
}
-out:
dout("aio_write %p %llx.%llx %llu~%u dropping cap refs on %s\n",
- inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len,
+ inode, ceph_vinop(inode), pos, (unsigned)count,
ceph_cap_string(got));
ceph_put_cap_refs(ci, got);
- if (ret == -EOLDSNAPC) {
- dout("aio_write %p %llx.%llx %llu~%u got EOLDSNAPC, retrying\n",
- inode, ceph_vinop(inode), pos, (unsigned)iov->iov_len);
- goto retry_snap;
+ if (written >= 0 &&
+ ((file->f_flags & O_SYNC) || IS_SYNC(file->f_mapping->host) ||
+ ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_NEARFULL))) {
+ err = vfs_fsync_range(file, pos, pos + written - 1, 1);
+ if (err < 0)
+ written = err;
}
- return ret;
+ goto out_unlocked;
+
+out:
+ mutex_unlock(&inode->i_mutex);
+out_unlocked:
+ current->backing_dev_info = NULL;
+ return written ? written : err;
}
/*
* llseek. be sure to verify file size on SEEK_END.
*/
-static loff_t ceph_llseek(struct file *file, loff_t offset, int origin)
+static loff_t ceph_llseek(struct file *file, loff_t offset, int whence)
{
struct inode *inode = file->f_mapping->host;
int ret;
mutex_lock(&inode->i_mutex);
- __ceph_do_pending_vmtruncate(inode);
- switch (origin) {
- case SEEK_END:
+
+ if (whence == SEEK_END || whence == SEEK_DATA || whence == SEEK_HOLE) {
ret = ceph_do_getattr(inode, CEPH_STAT_CAP_SIZE);
if (ret < 0) {
offset = ret;
goto out;
}
+ }
+
+ switch (whence) {
+ case SEEK_END:
offset += inode->i_size;
break;
case SEEK_CUR:
@@ -755,39 +1014,239 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int origin)
}
offset += file->f_pos;
break;
+ case SEEK_DATA:
+ if (offset >= inode->i_size) {
+ ret = -ENXIO;
+ goto out;
+ }
+ break;
+ case SEEK_HOLE:
+ if (offset >= inode->i_size) {
+ ret = -ENXIO;
+ goto out;
+ }
+ offset = inode->i_size;
+ break;
}
- if (offset < 0 || offset > inode->i_sb->s_maxbytes) {
- offset = -EINVAL;
+ offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes);
+
+out:
+ mutex_unlock(&inode->i_mutex);
+ return offset;
+}
+
+static inline void ceph_zero_partial_page(
+ struct inode *inode, loff_t offset, unsigned size)
+{
+ struct page *page;
+ pgoff_t index = offset >> PAGE_CACHE_SHIFT;
+
+ page = find_lock_page(inode->i_mapping, index);
+ if (page) {
+ wait_on_page_writeback(page);
+ zero_user(page, offset & (PAGE_CACHE_SIZE - 1), size);
+ unlock_page(page);
+ page_cache_release(page);
+ }
+}
+
+static void ceph_zero_pagecache_range(struct inode *inode, loff_t offset,
+ loff_t length)
+{
+ loff_t nearly = round_up(offset, PAGE_CACHE_SIZE);
+ if (offset < nearly) {
+ loff_t size = nearly - offset;
+ if (length < size)
+ size = length;
+ ceph_zero_partial_page(inode, offset, size);
+ offset += size;
+ length -= size;
+ }
+ if (length >= PAGE_CACHE_SIZE) {
+ loff_t size = round_down(length, PAGE_CACHE_SIZE);
+ truncate_pagecache_range(inode, offset, offset + size - 1);
+ offset += size;
+ length -= size;
+ }
+ if (length)
+ ceph_zero_partial_page(inode, offset, length);
+}
+
+static int ceph_zero_partial_object(struct inode *inode,
+ loff_t offset, loff_t *length)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
+ struct ceph_osd_request *req;
+ int ret = 0;
+ loff_t zero = 0;
+ int op;
+
+ if (!length) {
+ op = offset ? CEPH_OSD_OP_DELETE : CEPH_OSD_OP_TRUNCATE;
+ length = &zero;
+ } else {
+ op = CEPH_OSD_OP_ZERO;
+ }
+
+ req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout,
+ ceph_vino(inode),
+ offset, length,
+ 1, op,
+ CEPH_OSD_FLAG_WRITE |
+ CEPH_OSD_FLAG_ONDISK,
+ NULL, 0, 0, false);
+ if (IS_ERR(req)) {
+ ret = PTR_ERR(req);
goto out;
}
- /* Special lock needed here? */
- if (offset != file->f_pos) {
- file->f_pos = offset;
- file->f_version = 0;
+ ceph_osdc_build_request(req, offset, NULL, ceph_vino(inode).snap,
+ &inode->i_mtime);
+
+ ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
+ if (!ret) {
+ ret = ceph_osdc_wait_request(&fsc->client->osdc, req);
+ if (ret == -ENOENT)
+ ret = 0;
}
+ ceph_osdc_put_request(req);
out:
+ return ret;
+}
+
+static int ceph_zero_objects(struct inode *inode, loff_t offset, loff_t length)
+{
+ int ret = 0;
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ s32 stripe_unit = ceph_file_layout_su(ci->i_layout);
+ s32 stripe_count = ceph_file_layout_stripe_count(ci->i_layout);
+ s32 object_size = ceph_file_layout_object_size(ci->i_layout);
+ u64 object_set_size = object_size * stripe_count;
+ u64 nearly, t;
+
+ /* round offset up to next period boundary */
+ nearly = offset + object_set_size - 1;
+ t = nearly;
+ nearly -= do_div(t, object_set_size);
+
+ while (length && offset < nearly) {
+ loff_t size = length;
+ ret = ceph_zero_partial_object(inode, offset, &size);
+ if (ret < 0)
+ return ret;
+ offset += size;
+ length -= size;
+ }
+ while (length >= object_set_size) {
+ int i;
+ loff_t pos = offset;
+ for (i = 0; i < stripe_count; ++i) {
+ ret = ceph_zero_partial_object(inode, pos, NULL);
+ if (ret < 0)
+ return ret;
+ pos += stripe_unit;
+ }
+ offset += object_set_size;
+ length -= object_set_size;
+ }
+ while (length) {
+ loff_t size = length;
+ ret = ceph_zero_partial_object(inode, offset, &size);
+ if (ret < 0)
+ return ret;
+ offset += size;
+ length -= size;
+ }
+ return ret;
+}
+
+static long ceph_fallocate(struct file *file, int mode,
+ loff_t offset, loff_t length)
+{
+ struct ceph_file_info *fi = file->private_data;
+ struct inode *inode = file_inode(file);
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_osd_client *osdc =
+ &ceph_inode_to_client(inode)->client->osdc;
+ int want, got = 0;
+ int dirty;
+ int ret = 0;
+ loff_t endoff = 0;
+ loff_t size;
+
+ if (!S_ISREG(inode->i_mode))
+ return -EOPNOTSUPP;
+
+ mutex_lock(&inode->i_mutex);
+
+ if (ceph_snap(inode) != CEPH_NOSNAP) {
+ ret = -EROFS;
+ goto unlock;
+ }
+
+ if (ceph_osdmap_flag(osdc->osdmap, CEPH_OSDMAP_FULL) &&
+ !(mode & FALLOC_FL_PUNCH_HOLE)) {
+ ret = -ENOSPC;
+ goto unlock;
+ }
+
+ size = i_size_read(inode);
+ if (!(mode & FALLOC_FL_KEEP_SIZE))
+ endoff = offset + length;
+
+ if (fi->fmode & CEPH_FILE_MODE_LAZY)
+ want = CEPH_CAP_FILE_BUFFER | CEPH_CAP_FILE_LAZYIO;
+ else
+ want = CEPH_CAP_FILE_BUFFER;
+
+ ret = ceph_get_caps(ci, CEPH_CAP_FILE_WR, want, &got, endoff);
+ if (ret < 0)
+ goto unlock;
+
+ if (mode & FALLOC_FL_PUNCH_HOLE) {
+ if (offset < size)
+ ceph_zero_pagecache_range(inode, offset, length);
+ ret = ceph_zero_objects(inode, offset, length);
+ } else if (endoff > size) {
+ truncate_pagecache_range(inode, size, -1);
+ if (ceph_inode_set_size(inode, endoff))
+ ceph_check_caps(ceph_inode(inode),
+ CHECK_CAPS_AUTHONLY, NULL);
+ }
+
+ if (!ret) {
+ spin_lock(&ci->i_ceph_lock);
+ dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_FILE_WR);
+ spin_unlock(&ci->i_ceph_lock);
+ if (dirty)
+ __mark_inode_dirty(inode, dirty);
+ }
+
+ ceph_put_cap_refs(ci, got);
+unlock:
mutex_unlock(&inode->i_mutex);
- return offset;
+ return ret;
}
const struct file_operations ceph_file_fops = {
.open = ceph_open,
.release = ceph_release,
.llseek = ceph_llseek,
- .read = do_sync_read,
- .write = do_sync_write,
- .aio_read = ceph_aio_read,
- .aio_write = ceph_aio_write,
+ .read = new_sync_read,
+ .write = new_sync_write,
+ .read_iter = ceph_read_iter,
+ .write_iter = ceph_write_iter,
.mmap = ceph_mmap,
.fsync = ceph_fsync,
.lock = ceph_lock,
.flock = ceph_flock,
.splice_read = generic_file_splice_read,
- .splice_write = generic_file_splice_write,
+ .splice_write = iter_file_splice_write,
.unlocked_ioctl = ceph_ioctl,
.compat_ioctl = ceph_ioctl,
+ .fallocate = ceph_fallocate,
};
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 1d6a45b5a04..04c89c266ce 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -2,7 +2,6 @@
#include <linux/module.h>
#include <linux/fs.h>
-#include <linux/smp_lock.h>
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/uaccess.h>
@@ -10,10 +9,12 @@
#include <linux/namei.h>
#include <linux/writeback.h>
#include <linux/vmalloc.h>
-#include <linux/pagevec.h>
+#include <linux/posix_acl.h>
+#include <linux/random.h>
#include "super.h"
#include "mds_client.h"
+#include "cache.h"
#include <linux/ceph/decode.h>
/*
@@ -37,6 +38,13 @@ static void ceph_vmtruncate_work(struct work_struct *work);
/*
* find or create an inode, given the ceph ino number
*/
+static int ceph_set_ino_cb(struct inode *inode, void *data)
+{
+ ceph_inode(inode)->i_vino = *(struct ceph_vino *)data;
+ inode->i_ino = ceph_vino_to_ino(*(struct ceph_vino *)data);
+ return 0;
+}
+
struct inode *ceph_get_inode(struct super_block *sb, struct ceph_vino vino)
{
struct inode *inode;
@@ -89,6 +97,8 @@ const struct inode_operations ceph_file_iops = {
.getxattr = ceph_getxattr,
.listxattr = ceph_listxattr,
.removexattr = ceph_removexattr,
+ .get_acl = ceph_get_acl,
+ .set_acl = ceph_set_acl,
};
@@ -170,9 +180,8 @@ struct ceph_inode_frag *__ceph_find_frag(struct ceph_inode_info *ci, u32 f)
* specified, copy the frag delegation info to the caller if
* it is present.
*/
-u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
- struct ceph_inode_frag *pfrag,
- int *found)
+static u32 __ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
+ struct ceph_inode_frag *pfrag, int *found)
{
u32 t = ceph_frag_make(0, 0);
struct ceph_inode_frag *frag;
@@ -182,7 +191,6 @@ u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
if (found)
*found = 0;
- mutex_lock(&ci->i_fragtree_mutex);
while (1) {
WARN_ON(!ceph_frag_contains_value(t, v));
frag = __ceph_find_frag(ci, t);
@@ -211,10 +219,19 @@ u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
}
dout("choose_frag(%x) = %x\n", v, t);
- mutex_unlock(&ci->i_fragtree_mutex);
return t;
}
+u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
+ struct ceph_inode_frag *pfrag, int *found)
+{
+ u32 ret;
+ mutex_lock(&ci->i_fragtree_mutex);
+ ret = __ceph_choose_frag(ci, v, pfrag, found);
+ mutex_unlock(&ci->i_fragtree_mutex);
+ return ret;
+}
+
/*
* Process dirfrag (delegation) info from the mds. Include leaf
* fragment in tree ONLY if ndist > 0. Otherwise, only
@@ -228,11 +245,17 @@ static int ceph_fill_dirfrag(struct inode *inode,
u32 id = le32_to_cpu(dirinfo->frag);
int mds = le32_to_cpu(dirinfo->auth);
int ndist = le32_to_cpu(dirinfo->ndist);
+ int diri_auth = -1;
int i;
int err = 0;
+ spin_lock(&ci->i_ceph_lock);
+ if (ci->i_auth_cap)
+ diri_auth = ci->i_auth_cap->mds;
+ spin_unlock(&ci->i_ceph_lock);
+
mutex_lock(&ci->i_fragtree_mutex);
- if (ndist == 0) {
+ if (ndist == 0 && mds == diri_auth) {
/* no delegation info needed. */
frag = __ceph_find_frag(ci, id);
if (!frag)
@@ -277,6 +300,75 @@ out:
return err;
}
+static int ceph_fill_fragtree(struct inode *inode,
+ struct ceph_frag_tree_head *fragtree,
+ struct ceph_mds_reply_dirfrag *dirinfo)
+{
+ struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_inode_frag *frag;
+ struct rb_node *rb_node;
+ int i;
+ u32 id, nsplits;
+ bool update = false;
+
+ mutex_lock(&ci->i_fragtree_mutex);
+ nsplits = le32_to_cpu(fragtree->nsplits);
+ if (nsplits) {
+ i = prandom_u32() % nsplits;
+ id = le32_to_cpu(fragtree->splits[i].frag);
+ if (!__ceph_find_frag(ci, id))
+ update = true;
+ } else if (!RB_EMPTY_ROOT(&ci->i_fragtree)) {
+ rb_node = rb_first(&ci->i_fragtree);
+ frag = rb_entry(rb_node, struct ceph_inode_frag, node);
+ if (frag->frag != ceph_frag_make(0, 0) || rb_next(rb_node))
+ update = true;
+ }
+ if (!update && dirinfo) {
+ id = le32_to_cpu(dirinfo->frag);
+ if (id != __ceph_choose_frag(ci, id, NULL, NULL))
+ update = true;
+ }
+ if (!update)
+ goto out_unlock;
+
+ dout("fill_fragtree %llx.%llx\n", ceph_vinop(inode));
+ rb_node = rb_first(&ci->i_fragtree);
+ for (i = 0; i < nsplits; i++) {
+ id = le32_to_cpu(fragtree->splits[i].frag);
+ frag = NULL;
+ while (rb_node) {
+ frag = rb_entry(rb_node, struct ceph_inode_frag, node);
+ if (ceph_frag_compare(frag->frag, id) >= 0) {
+ if (frag->frag != id)
+ frag = NULL;
+ else
+ rb_node = rb_next(rb_node);
+ break;
+ }
+ rb_node = rb_next(rb_node);
+ rb_erase(&frag->node, &ci->i_fragtree);
+ kfree(frag);
+ frag = NULL;
+ }
+ if (!frag) {
+ frag = __get_or_create_frag(ci, id);
+ if (IS_ERR(frag))
+ continue;
+ }
+ frag->split_by = le32_to_cpu(fragtree->splits[i].by);
+ dout(" frag %x split by %d\n", frag->frag, frag->split_by);
+ }
+ while (rb_node) {
+ frag = rb_entry(rb_node, struct ceph_inode_frag, node);
+ rb_node = rb_next(rb_node);
+ rb_erase(&frag->node, &ci->i_fragtree);
+ kfree(frag);
+ }
+out_unlock:
+ mutex_unlock(&ci->i_fragtree_mutex);
+ return 0;
+}
/*
* initialize a newly allocated inode.
@@ -292,12 +384,17 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
dout("alloc_inode %p\n", &ci->vfs_inode);
+ spin_lock_init(&ci->i_ceph_lock);
+
ci->i_version = 0;
ci->i_time_warp_seq = 0;
ci->i_ceph_flags = 0;
- ci->i_release_count = 0;
+ atomic_set(&ci->i_release_count, 1);
+ atomic_set(&ci->i_complete_count, 0);
ci->i_symlink = NULL;
+ memset(&ci->i_dir_layout, 0, sizeof(ci->i_dir_layout));
+
ci->i_fragtree = RB_ROOT;
mutex_init(&ci->i_fragtree_mutex);
@@ -324,9 +421,6 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
ci->i_hold_caps_min = 0;
ci->i_hold_caps_max = 0;
INIT_LIST_HEAD(&ci->i_cap_delay_list);
- ci->i_cap_exporting_mds = 0;
- ci->i_cap_exporting_mseq = 0;
- ci->i_cap_exporting_issued = 0;
INIT_LIST_HEAD(&ci->i_cap_snaps);
ci->i_head_snapc = NULL;
ci->i_snap_caps = 0;
@@ -334,6 +428,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
for (i = 0; i < CEPH_FILE_MODE_NUM; i++)
ci->i_nr_by_mode[i] = 0;
+ mutex_init(&ci->i_truncate_mutex);
ci->i_truncate_seq = 0;
ci->i_truncate_size = 0;
ci->i_truncate_pending = 0;
@@ -347,6 +442,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
ci->i_rd_ref = 0;
ci->i_rdcache_ref = 0;
ci->i_wr_ref = 0;
+ ci->i_wb_ref = 0;
ci->i_wrbuffer_ref = 0;
ci->i_wrbuffer_ref_head = 0;
ci->i_shared_gen = 0;
@@ -366,9 +462,19 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
INIT_WORK(&ci->i_vmtruncate_work, ceph_vmtruncate_work);
+ ceph_fscache_inode_init(ci);
+
return &ci->vfs_inode;
}
+static void ceph_i_callback(struct rcu_head *head)
+{
+ struct inode *inode = container_of(head, struct inode, i_rcu);
+ struct ceph_inode_info *ci = ceph_inode(inode);
+
+ kmem_cache_free(ceph_inode_cachep, ci);
+}
+
void ceph_destroy_inode(struct inode *inode)
{
struct ceph_inode_info *ci = ceph_inode(inode);
@@ -377,11 +483,13 @@ void ceph_destroy_inode(struct inode *inode)
dout("destroy_inode %p ino %llx.%llx\n", inode, ceph_vinop(inode));
+ ceph_fscache_unregister_inode_cookie(ci);
+
ceph_queue_caps_release(inode);
/*
* we may still have a snap_realm reference if there are stray
- * caps in i_cap_exporting_issued or i_snap_caps.
+ * caps in i_snap_caps.
*/
if (ci->i_snap_realm) {
struct ceph_mds_client *mdsc =
@@ -408,9 +516,18 @@ void ceph_destroy_inode(struct inode *inode)
if (ci->i_xattrs.prealloc_blob)
ceph_buffer_put(ci->i_xattrs.prealloc_blob);
- kmem_cache_free(ceph_inode_cachep, ci);
+ call_rcu(&inode->i_rcu, ceph_i_callback);
}
+int ceph_drop_inode(struct inode *inode)
+{
+ /*
+ * Positve dentry and corresponding inode are always accompanied
+ * in MDS reply. So no need to keep inode in the cache after
+ * dropping all its aliases.
+ */
+ return 1;
+}
/*
* Helpers to fill in size, ctime, mtime, and atime. We have to be
@@ -436,16 +553,20 @@ int ceph_fill_file_size(struct inode *inode, int issued,
dout("truncate_seq %u -> %u\n",
ci->i_truncate_seq, truncate_seq);
ci->i_truncate_seq = truncate_seq;
+
+ /* the MDS should have revoked these caps */
+ WARN_ON_ONCE(issued & (CEPH_CAP_FILE_EXCL |
+ CEPH_CAP_FILE_RD |
+ CEPH_CAP_FILE_WR |
+ CEPH_CAP_FILE_LAZYIO));
/*
* If we hold relevant caps, or in the case where we're
* not the only client referencing this file and we
* don't hold those caps, then we need to check whether
* the file is either opened or mmaped
*/
- if ((issued & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_RD|
- CEPH_CAP_FILE_WR|CEPH_CAP_FILE_BUFFER|
- CEPH_CAP_FILE_EXCL|
- CEPH_CAP_FILE_LAZYIO)) ||
+ if ((issued & (CEPH_CAP_FILE_CACHE|
+ CEPH_CAP_FILE_BUFFER)) ||
mapping_mapped(inode->i_mapping) ||
__ceph_caps_file_wanted(ci)) {
ci->i_truncate_pending++;
@@ -459,6 +580,10 @@ int ceph_fill_file_size(struct inode *inode, int issued,
truncate_size);
ci->i_truncate_size = truncate_size;
}
+
+ if (queue_trunc)
+ ceph_fscache_invalidate(inode);
+
return queue_trunc;
}
@@ -471,7 +596,9 @@ void ceph_fill_file_time(struct inode *inode, int issued,
if (issued & (CEPH_CAP_FILE_EXCL|
CEPH_CAP_FILE_WR|
- CEPH_CAP_FILE_BUFFER)) {
+ CEPH_CAP_FILE_BUFFER|
+ CEPH_CAP_AUTH_EXCL|
+ CEPH_CAP_XATTR_EXCL)) {
if (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,
@@ -511,7 +638,7 @@ void ceph_fill_file_time(struct inode *inode, int issued,
warn = 1;
}
} else {
- /* we have no write caps; whatever the MDS says is true */
+ /* we have no write|excl caps; whatever the MDS says is true */
if (ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) >= 0) {
inode->i_ctime = *ctime;
inode->i_mtime = *mtime;
@@ -537,20 +664,26 @@ static int fill_inode(struct inode *inode,
unsigned long ttl_from, int cap_fmode,
struct ceph_cap_reservation *caps_reservation)
{
+ struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
struct ceph_mds_reply_inode *info = iinfo->in;
struct ceph_inode_info *ci = ceph_inode(inode);
- int i;
- int issued, implemented;
+ int issued = 0, implemented, new_issued;
struct timespec mtime, atime, ctime;
- u32 nsplits;
struct ceph_buffer *xattr_blob = NULL;
+ struct ceph_cap *new_cap = NULL;
int err = 0;
- int queue_trunc = 0;
+ bool wake = false;
+ bool queue_trunc = false;
+ bool new_version = false;
dout("fill_inode %p ino %llx.%llx v %llu had %llu\n",
inode, ceph_vinop(inode), le64_to_cpu(info->version),
ci->i_version);
+ /* prealloc new cap struct */
+ if (info->cap.caps && ceph_snap(inode) == CEPH_NOSNAP)
+ new_cap = ceph_get_cap(mdsc, caps_reservation);
+
/*
* prealloc xattr data, if it looks like we'll need it. only
* if len > 4 (meaning there are actually xattrs; the first 4
@@ -563,52 +696,73 @@ static int fill_inode(struct inode *inode,
iinfo->xattr_len);
}
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
/*
* provided version will be odd if inode value is projected,
- * even if stable. skip the update if we have a newer info
- * (e.g., due to inode info racing form multiple MDSs), or if
- * we are getting projected (unstable) inode info.
+ * even if stable. skip the update if we have newer stable
+ * info (ours>=theirs, e.g. due to racing mds replies), unless
+ * we are getting projected (unstable) info (in which case the
+ * version is odd, and we want ours>theirs).
+ * us them
+ * 2 2 skip
+ * 3 2 skip
+ * 3 3 update
*/
- if (le64_to_cpu(info->version) > 0 &&
- (ci->i_version & ~1) > le64_to_cpu(info->version))
- goto no_change;
+ if (ci->i_version == 0 ||
+ ((info->cap.flags & CEPH_CAP_FLAG_AUTH) &&
+ le64_to_cpu(info->version) > (ci->i_version & ~1)))
+ new_version = true;
issued = __ceph_caps_issued(ci, &implemented);
issued |= implemented | __ceph_caps_dirty(ci);
+ new_issued = ~issued & le32_to_cpu(info->cap.caps);
/* update inode */
ci->i_version = le64_to_cpu(info->version);
inode->i_version++;
inode->i_rdev = le32_to_cpu(info->rdev);
+ inode->i_blkbits = fls(le32_to_cpu(info->layout.fl_stripe_unit)) - 1;
- if ((issued & CEPH_CAP_AUTH_EXCL) == 0) {
+ if ((new_version || (new_issued & CEPH_CAP_AUTH_SHARED)) &&
+ (issued & CEPH_CAP_AUTH_EXCL) == 0) {
inode->i_mode = le32_to_cpu(info->mode);
- inode->i_uid = le32_to_cpu(info->uid);
- inode->i_gid = le32_to_cpu(info->gid);
+ inode->i_uid = make_kuid(&init_user_ns, le32_to_cpu(info->uid));
+ inode->i_gid = make_kgid(&init_user_ns, le32_to_cpu(info->gid));
dout("%p mode 0%o uid.gid %d.%d\n", inode, inode->i_mode,
- inode->i_uid, inode->i_gid);
+ from_kuid(&init_user_ns, inode->i_uid),
+ from_kgid(&init_user_ns, inode->i_gid));
}
- if ((issued & CEPH_CAP_LINK_EXCL) == 0)
- inode->i_nlink = le32_to_cpu(info->nlink);
-
- /* be careful with mtime, atime, size */
- ceph_decode_timespec(&atime, &info->atime);
- ceph_decode_timespec(&mtime, &info->mtime);
- ceph_decode_timespec(&ctime, &info->ctime);
- queue_trunc = ceph_fill_file_size(inode, issued,
- le32_to_cpu(info->truncate_seq),
- le64_to_cpu(info->truncate_size),
- le64_to_cpu(info->size));
- ceph_fill_file_time(inode, issued,
- le32_to_cpu(info->time_warp_seq),
- &ctime, &mtime, &atime);
-
- ci->i_max_size = le64_to_cpu(info->max_size);
- ci->i_layout = info->layout;
- inode->i_blkbits = fls(le32_to_cpu(info->layout.fl_stripe_unit)) - 1;
+ if ((new_version || (new_issued & CEPH_CAP_LINK_SHARED)) &&
+ (issued & CEPH_CAP_LINK_EXCL) == 0)
+ set_nlink(inode, le32_to_cpu(info->nlink));
+
+ if (new_version || (new_issued & CEPH_CAP_ANY_RD)) {
+ /* be careful with mtime, atime, size */
+ ceph_decode_timespec(&atime, &info->atime);
+ ceph_decode_timespec(&mtime, &info->mtime);
+ ceph_decode_timespec(&ctime, &info->ctime);
+ ceph_fill_file_time(inode, issued,
+ le32_to_cpu(info->time_warp_seq),
+ &ctime, &mtime, &atime);
+ }
+
+ if (new_version ||
+ (new_issued & (CEPH_CAP_ANY_FILE_RD | CEPH_CAP_ANY_FILE_WR))) {
+ ci->i_layout = info->layout;
+ queue_trunc = ceph_fill_file_size(inode, issued,
+ le32_to_cpu(info->truncate_seq),
+ le64_to_cpu(info->truncate_size),
+ le64_to_cpu(info->size));
+ /* only update max_size on auth cap */
+ if ((info->cap.flags & CEPH_CAP_FLAG_AUTH) &&
+ ci->i_max_size != le64_to_cpu(info->max_size)) {
+ dout("max_size %lld -> %llu\n", ci->i_max_size,
+ le64_to_cpu(info->max_size));
+ ci->i_max_size = le64_to_cpu(info->max_size);
+ }
+ }
/* xattrs */
/* note that if i_xattrs.len <= 4, i_xattrs.data will still be NULL. */
@@ -621,6 +775,7 @@ static int fill_inode(struct inode *inode,
memcpy(ci->i_xattrs.blob->vec.iov_base,
iinfo->xattr_data, iinfo->xattr_len);
ci->i_xattrs.version = le64_to_cpu(info->xattr_version);
+ ceph_forget_all_cached_acls(inode);
xattr_blob = NULL;
}
@@ -643,20 +798,21 @@ static int fill_inode(struct inode *inode,
case S_IFLNK:
inode->i_op = &ceph_symlink_iops;
if (!ci->i_symlink) {
- int symlen = iinfo->symlink_len;
+ u32 symlen = iinfo->symlink_len;
char *sym;
- BUG_ON(symlen != inode->i_size);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
+
+ err = -EINVAL;
+ if (WARN_ON(symlen != inode->i_size))
+ goto out;
err = -ENOMEM;
- sym = kmalloc(symlen+1, GFP_NOFS);
+ sym = kstrndup(iinfo->symlink, symlen, GFP_NOFS);
if (!sym)
goto out;
- memcpy(sym, iinfo->symlink, symlen);
- sym[symlen] = 0;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (!ci->i_symlink)
ci->i_symlink = sym;
else
@@ -667,54 +823,30 @@ static int fill_inode(struct inode *inode,
inode->i_op = &ceph_dir_iops;
inode->i_fop = &ceph_dir_fops;
+ ci->i_dir_layout = iinfo->dir_layout;
+
ci->i_files = le64_to_cpu(info->files);
ci->i_subdirs = le64_to_cpu(info->subdirs);
ci->i_rbytes = le64_to_cpu(info->rbytes);
ci->i_rfiles = le64_to_cpu(info->rfiles);
ci->i_rsubdirs = le64_to_cpu(info->rsubdirs);
ceph_decode_timespec(&ci->i_rctime, &info->rctime);
-
- /* set dir completion flag? */
- if (ci->i_files == 0 && ci->i_subdirs == 0 &&
- ceph_snap(inode) == CEPH_NOSNAP &&
- (le32_to_cpu(info->cap.caps) & CEPH_CAP_FILE_SHARED) &&
- (issued & CEPH_CAP_FILE_EXCL) == 0 &&
- (ci->i_ceph_flags & CEPH_I_COMPLETE) == 0) {
- dout(" marking %p complete (empty)\n", inode);
- ci->i_ceph_flags |= CEPH_I_COMPLETE;
- ci->i_max_offset = 2;
- }
-
- /* it may be better to set st_size in getattr instead? */
- if (ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), RBYTES))
- inode->i_size = ci->i_rbytes;
break;
default:
pr_err("fill_inode %llx.%llx BAD mode 0%o\n",
ceph_vinop(inode), inode->i_mode);
}
-no_change:
- spin_unlock(&inode->i_lock);
-
- /* queue truncate if we saw i_size decrease */
- if (queue_trunc)
- ceph_queue_vmtruncate(inode);
-
- /* populate frag tree */
- /* FIXME: move me up, if/when version reflects fragtree changes */
- nsplits = le32_to_cpu(info->fragtree.nsplits);
- mutex_lock(&ci->i_fragtree_mutex);
- for (i = 0; i < nsplits; i++) {
- u32 id = le32_to_cpu(info->fragtree.splits[i].frag);
- struct ceph_inode_frag *frag = __get_or_create_frag(ci, id);
-
- if (IS_ERR(frag))
- continue;
- frag->split_by = le32_to_cpu(info->fragtree.splits[i].by);
- dout(" frag %x split by %d\n", frag->frag, frag->split_by);
+ /* set dir completion flag? */
+ if (S_ISDIR(inode->i_mode) &&
+ ci->i_files == 0 && ci->i_subdirs == 0 &&
+ ceph_snap(inode) == CEPH_NOSNAP &&
+ (le32_to_cpu(info->cap.caps) & CEPH_CAP_FILE_SHARED) &&
+ (issued & CEPH_CAP_FILE_EXCL) == 0 &&
+ !__ceph_dir_is_complete(ci)) {
+ dout(" marking %p complete (empty)\n", inode);
+ __ceph_dir_set_complete(ci, atomic_read(&ci->i_release_count));
}
- mutex_unlock(&ci->i_fragtree_mutex);
/* were we issued a capability? */
if (info->cap.caps) {
@@ -727,30 +859,41 @@ no_change:
le32_to_cpu(info->cap.seq),
le32_to_cpu(info->cap.mseq),
le64_to_cpu(info->cap.realm),
- info->cap.flags,
- caps_reservation);
+ info->cap.flags, &new_cap);
+ wake = true;
} else {
- spin_lock(&inode->i_lock);
dout(" %p got snap_caps %s\n", inode,
ceph_cap_string(le32_to_cpu(info->cap.caps)));
ci->i_snap_caps |= le32_to_cpu(info->cap.caps);
if (cap_fmode >= 0)
__ceph_get_fmode(ci, cap_fmode);
- spin_unlock(&inode->i_lock);
}
} else if (cap_fmode >= 0) {
- pr_warning("mds issued no caps on %llx.%llx\n",
+ pr_warn("mds issued no caps on %llx.%llx\n",
ceph_vinop(inode));
__ceph_get_fmode(ci, cap_fmode);
}
+ spin_unlock(&ci->i_ceph_lock);
+
+ if (wake)
+ wake_up_all(&ci->i_cap_wq);
+
+ /* queue truncate if we saw i_size decrease */
+ if (queue_trunc)
+ ceph_queue_vmtruncate(inode);
+
+ /* populate frag tree */
+ if (S_ISDIR(inode->i_mode))
+ ceph_fill_fragtree(inode, &info->fragtree, dirinfo);
/* update delegation info? */
if (dirinfo)
ceph_fill_dirfrag(inode, dirinfo);
err = 0;
-
out:
+ if (new_cap)
+ ceph_put_cap(mdsc, new_cap);
if (xattr_blob)
ceph_buffer_put(xattr_blob);
return err;
@@ -775,14 +918,14 @@ static void update_dentry_lease(struct dentry *dentry,
return;
spin_lock(&dentry->d_lock);
- dout("update_dentry_lease %p mask %d duration %lu ms ttl %lu\n",
- dentry, le16_to_cpu(lease->mask), duration, ttl);
+ dout("update_dentry_lease %p duration %lu ms ttl %lu\n",
+ dentry, duration, ttl);
/* make lease_rdcache_gen match directory */
dir = dentry->d_parent->d_inode;
di->lease_shared_gen = ceph_inode(dir)->i_shared_gen;
- if (lease->mask == 0)
+ if (duration == 0)
goto out_unlock;
if (di->lease_gen == session->s_cap_gen &&
@@ -807,37 +950,6 @@ out_unlock:
}
/*
- * Set dentry's directory position based on the current dir's max, and
- * order it in d_subdirs, so that dcache_readdir behaves.
- */
-static void ceph_set_dentry_offset(struct dentry *dn)
-{
- struct dentry *dir = dn->d_parent;
- struct inode *inode = dn->d_parent->d_inode;
- struct ceph_dentry_info *di;
-
- BUG_ON(!inode);
-
- di = ceph_dentry(dn);
-
- spin_lock(&inode->i_lock);
- if ((ceph_inode(inode)->i_ceph_flags & CEPH_I_COMPLETE) == 0) {
- spin_unlock(&inode->i_lock);
- return;
- }
- di->offset = ceph_inode(inode)->i_max_offset++;
- spin_unlock(&inode->i_lock);
-
- spin_lock(&dcache_lock);
- spin_lock(&dn->d_lock);
- list_move(&dn->d_u.d_child, &dir->d_subdirs);
- dout("set_dentry_offset %p %lld (%p %p)\n", dn, di->offset,
- dn->d_u.d_child.prev, dn->d_u.d_child.next);
- spin_unlock(&dn->d_lock);
- spin_unlock(&dcache_lock);
-}
-
-/*
* splice a dentry to an inode.
* caller must hold directory i_mutex for this to be safe.
*
@@ -846,7 +958,7 @@ static void ceph_set_dentry_offset(struct dentry *dn)
* the caller) if we fail.
*/
static struct dentry *splice_dentry(struct dentry *dn, struct inode *in,
- bool *prehash, bool set_offset)
+ bool *prehash)
{
struct dentry *realdn;
@@ -866,8 +978,8 @@ static struct dentry *splice_dentry(struct dentry *dn, struct inode *in,
} else if (realdn) {
dout("dn %p (%d) spliced with %p (%d) "
"inode %p ino %llx.%llx\n",
- dn, atomic_read(&dn->d_count),
- realdn, atomic_read(&realdn->d_count),
+ dn, d_count(dn),
+ realdn, d_count(realdn),
realdn->d_inode, ceph_vinop(realdn->d_inode));
dput(dn);
dn = realdn;
@@ -878,8 +990,6 @@ static struct dentry *splice_dentry(struct dentry *dn, struct inode *in,
}
if ((!prehash || *prehash) && d_unhashed(dn))
d_rehash(dn);
- if (set_offset)
- ceph_set_dentry_offset(dn);
out:
return dn;
}
@@ -900,10 +1010,8 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
{
struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
struct inode *in = NULL;
- struct ceph_mds_reply_inode *ininfo;
struct ceph_vino vino;
struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
- int i = 0;
int err = 0;
dout("fill_trace %p is_dentry %d is_target %d\n", req,
@@ -953,11 +1061,87 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
if (rinfo->head->is_dentry) {
struct inode *dir = req->r_locked_dir;
- err = fill_inode(dir, &rinfo->diri, rinfo->dirfrag,
- session, req->r_request_started, -1,
- &req->r_caps_reservation);
- if (err < 0)
- return err;
+ if (dir) {
+ err = fill_inode(dir, &rinfo->diri, rinfo->dirfrag,
+ session, req->r_request_started, -1,
+ &req->r_caps_reservation);
+ if (err < 0)
+ goto done;
+ } else {
+ WARN_ON_ONCE(1);
+ }
+
+ if (dir && req->r_op == CEPH_MDS_OP_LOOKUPNAME) {
+ struct qstr dname;
+ struct dentry *dn, *parent;
+
+ BUG_ON(!rinfo->head->is_target);
+ BUG_ON(req->r_dentry);
+
+ parent = d_find_any_alias(dir);
+ BUG_ON(!parent);
+
+ dname.name = rinfo->dname;
+ dname.len = rinfo->dname_len;
+ dname.hash = full_name_hash(dname.name, dname.len);
+ vino.ino = le64_to_cpu(rinfo->targeti.in->ino);
+ vino.snap = le64_to_cpu(rinfo->targeti.in->snapid);
+retry_lookup:
+ dn = d_lookup(parent, &dname);
+ dout("d_lookup on parent=%p name=%.*s got %p\n",
+ parent, dname.len, dname.name, dn);
+
+ if (!dn) {
+ dn = d_alloc(parent, &dname);
+ dout("d_alloc %p '%.*s' = %p\n", parent,
+ dname.len, dname.name, dn);
+ if (dn == NULL) {
+ dput(parent);
+ err = -ENOMEM;
+ goto done;
+ }
+ err = ceph_init_dentry(dn);
+ if (err < 0) {
+ dput(dn);
+ dput(parent);
+ goto done;
+ }
+ } else if (dn->d_inode &&
+ (ceph_ino(dn->d_inode) != vino.ino ||
+ ceph_snap(dn->d_inode) != vino.snap)) {
+ dout(" dn %p points to wrong inode %p\n",
+ dn, dn->d_inode);
+ d_delete(dn);
+ dput(dn);
+ goto retry_lookup;
+ }
+
+ req->r_dentry = dn;
+ dput(parent);
+ }
+ }
+
+ if (rinfo->head->is_target) {
+ vino.ino = le64_to_cpu(rinfo->targeti.in->ino);
+ vino.snap = le64_to_cpu(rinfo->targeti.in->snapid);
+
+ in = ceph_get_inode(sb, vino);
+ if (IS_ERR(in)) {
+ err = PTR_ERR(in);
+ goto done;
+ }
+ req->r_target_inode = in;
+
+ err = fill_inode(in, &rinfo->targeti, NULL,
+ session, req->r_request_started,
+ (!req->r_aborted && rinfo->head->result == 0) ?
+ req->r_fmode : -1,
+ &req->r_caps_reservation);
+ if (err < 0) {
+ pr_err("fill_inode badness %p %llx.%llx\n",
+ in, ceph_vinop(in));
+ goto done;
+ }
}
/*
@@ -965,6 +1149,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
* will have trouble splicing in the virtual snapdir later
*/
if (rinfo->head->is_dentry && !req->r_aborted &&
+ req->r_locked_dir &&
(rinfo->head->is_target || strncmp(req->r_dentry->d_name.name,
fsc->mount_options->snapdir_name,
req->r_dentry->d_name.len))) {
@@ -992,14 +1177,15 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
/* do we have a dn lease? */
have_lease = have_dir_cap ||
- (le16_to_cpu(rinfo->dlease->mask) &
- CEPH_LOCK_DN);
-
+ le32_to_cpu(rinfo->dlease->duration_ms);
if (!have_lease)
dout("fill_trace no dentry lease or dir cap\n");
/* rename? */
if (req->r_old_dentry && req->r_op == CEPH_MDS_OP_RENAME) {
+ struct inode *olddir = req->r_old_dentry_dir;
+ BUG_ON(!olddir);
+
dout(" src %p '%.*s' dst %p '%.*s'\n",
req->r_old_dentry,
req->r_old_dentry->d_name.len,
@@ -1008,9 +1194,6 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
dout("fill_trace doing d_move %p -> %p\n",
req->r_old_dentry, dn);
- /* d_move screws up d_subdirs order */
- ceph_i_clear(dir, CEPH_I_COMPLETE);
-
d_move(req->r_old_dentry, dn);
dout(" src %p '%.*s' dst %p '%.*s'\n",
req->r_old_dentry,
@@ -1022,15 +1205,14 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
rehashing bug in vfs_rename_dir */
ceph_invalidate_dentry_lease(dn);
- /* take overwritten dentry's readdir offset */
- dout("dn %p gets %p offset %lld (old offset %lld)\n",
- req->r_old_dentry, dn, ceph_dentry(dn)->offset,
+ /* d_move screws up sibling dentries' offsets */
+ ceph_dir_clear_complete(dir);
+ ceph_dir_clear_complete(olddir);
+
+ dout("dn %p gets new offset %lld\n", req->r_old_dentry,
ceph_dentry(req->r_old_dentry)->offset);
- ceph_dentry(req->r_old_dentry)->offset =
- ceph_dentry(dn)->offset;
dn = req->r_old_dentry; /* use old_dentry */
- in = dn->d_inode;
}
/* null dentry? */
@@ -1052,106 +1234,87 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
}
/* attach proper inode */
- ininfo = rinfo->targeti.in;
- vino.ino = le64_to_cpu(ininfo->ino);
- vino.snap = le64_to_cpu(ininfo->snapid);
if (!dn->d_inode) {
- in = ceph_get_inode(sb, vino);
- if (IS_ERR(in)) {
- pr_err("fill_trace bad get_inode "
- "%llx.%llx\n", vino.ino, vino.snap);
- err = PTR_ERR(in);
- d_delete(dn);
- goto done;
- }
- dn = splice_dentry(dn, in, &have_lease, true);
+ ceph_dir_clear_complete(dir);
+ ihold(in);
+ dn = splice_dentry(dn, in, &have_lease);
if (IS_ERR(dn)) {
err = PTR_ERR(dn);
goto done;
}
req->r_dentry = dn; /* may have spliced */
- igrab(in);
- } else if (ceph_ino(in) == vino.ino &&
- ceph_snap(in) == vino.snap) {
- igrab(in);
- } else {
+ } else if (dn->d_inode && dn->d_inode != in) {
dout(" %p links to %p %llx.%llx, not %llx.%llx\n",
- dn, in, ceph_ino(in), ceph_snap(in),
- vino.ino, vino.snap);
+ dn, dn->d_inode, ceph_vinop(dn->d_inode),
+ ceph_vinop(in));
have_lease = false;
- in = NULL;
}
if (have_lease)
update_dentry_lease(dn, rinfo->dlease, session,
req->r_request_started);
dout(" final dn %p\n", dn);
- i++;
- } else if (req->r_op == CEPH_MDS_OP_LOOKUPSNAP ||
- req->r_op == CEPH_MDS_OP_MKSNAP) {
+ } else if (!req->r_aborted &&
+ (req->r_op == CEPH_MDS_OP_LOOKUPSNAP ||
+ req->r_op == CEPH_MDS_OP_MKSNAP)) {
struct dentry *dn = req->r_dentry;
+ struct inode *dir = req->r_locked_dir;
/* fill out a snapdir LOOKUPSNAP dentry */
BUG_ON(!dn);
- BUG_ON(!req->r_locked_dir);
- BUG_ON(ceph_snap(req->r_locked_dir) != CEPH_SNAPDIR);
- ininfo = rinfo->targeti.in;
- vino.ino = le64_to_cpu(ininfo->ino);
- vino.snap = le64_to_cpu(ininfo->snapid);
- in = ceph_get_inode(sb, vino);
- if (IS_ERR(in)) {
- pr_err("fill_inode get_inode badness %llx.%llx\n",
- vino.ino, vino.snap);
- err = PTR_ERR(in);
- d_delete(dn);
- goto done;
- }
+ BUG_ON(!dir);
+ BUG_ON(ceph_snap(dir) != CEPH_SNAPDIR);
dout(" linking snapped dir %p to dn %p\n", in, dn);
- dn = splice_dentry(dn, in, NULL, true);
+ ceph_dir_clear_complete(dir);
+ ihold(in);
+ dn = splice_dentry(dn, in, NULL);
if (IS_ERR(dn)) {
err = PTR_ERR(dn);
goto done;
}
req->r_dentry = dn; /* may have spliced */
- igrab(in);
- rinfo->head->is_dentry = 1; /* fool notrace handlers */
}
+done:
+ dout("fill_trace done err=%d\n", err);
+ return err;
+}
- if (rinfo->head->is_target) {
- vino.ino = le64_to_cpu(rinfo->targeti.in->ino);
- vino.snap = le64_to_cpu(rinfo->targeti.in->snapid);
+/*
+ * Prepopulate our cache with readdir results, leases, etc.
+ */
+static int readdir_prepopulate_inodes_only(struct ceph_mds_request *req,
+ struct ceph_mds_session *session)
+{
+ struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
+ int i, err = 0;
- if (in == NULL || ceph_ino(in) != vino.ino ||
- ceph_snap(in) != vino.snap) {
- in = ceph_get_inode(sb, vino);
- if (IS_ERR(in)) {
- err = PTR_ERR(in);
- goto done;
- }
- }
- req->r_target_inode = in;
+ for (i = 0; i < rinfo->dir_nr; i++) {
+ struct ceph_vino vino;
+ struct inode *in;
+ int rc;
- err = fill_inode(in,
- &rinfo->targeti, NULL,
- session, req->r_request_started,
- (le32_to_cpu(rinfo->head->result) == 0) ?
- req->r_fmode : -1,
- &req->r_caps_reservation);
- if (err < 0) {
- pr_err("fill_inode badness %p %llx.%llx\n",
- in, ceph_vinop(in));
- goto done;
+ vino.ino = le64_to_cpu(rinfo->dir_in[i].in->ino);
+ vino.snap = le64_to_cpu(rinfo->dir_in[i].in->snapid);
+
+ in = ceph_get_inode(req->r_dentry->d_sb, vino);
+ if (IS_ERR(in)) {
+ err = PTR_ERR(in);
+ dout("new_inode badness got %d\n", err);
+ continue;
+ }
+ rc = fill_inode(in, &rinfo->dir_in[i], NULL, session,
+ req->r_request_started, -1,
+ &req->r_caps_reservation);
+ if (rc < 0) {
+ pr_err("fill_inode badness on %p got %d\n", in, rc);
+ err = rc;
+ continue;
}
}
-done:
- dout("fill_trace done err=%d\n", err);
return err;
}
-/*
- * Prepopulate our cache with readdir results, leases, etc.
- */
int ceph_readdir_prepopulate(struct ceph_mds_request *req,
struct ceph_mds_session *session)
{
@@ -1160,11 +1323,26 @@ int ceph_readdir_prepopulate(struct ceph_mds_request *req,
struct qstr dname;
struct dentry *dn;
struct inode *in;
- int err = 0, i;
+ int err = 0, ret, i;
struct inode *snapdir = NULL;
struct ceph_mds_request_head *rhead = req->r_request->front.iov_base;
- u64 frag = le32_to_cpu(rhead->args.readdir.frag);
struct ceph_dentry_info *di;
+ u64 r_readdir_offset = req->r_readdir_offset;
+ u32 frag = le32_to_cpu(rhead->args.readdir.frag);
+
+ if (rinfo->dir_dir &&
+ le32_to_cpu(rinfo->dir_dir->frag) != frag) {
+ dout("readdir_prepopulate got new frag %x -> %x\n",
+ frag, le32_to_cpu(rinfo->dir_dir->frag));
+ frag = le32_to_cpu(rinfo->dir_dir->frag);
+ if (ceph_frag_is_leftmost(frag))
+ r_readdir_offset = 2;
+ else
+ r_readdir_offset = 0;
+ }
+
+ if (req->r_aborted)
+ return readdir_prepopulate_inodes_only(req, session);
if (le32_to_cpu(rinfo->head->op) == CEPH_MDS_OP_LSSNAP) {
snapdir = ceph_get_snapdir(parent->d_inode);
@@ -1178,6 +1356,7 @@ int ceph_readdir_prepopulate(struct ceph_mds_request *req,
ceph_fill_dirfrag(parent->d_inode, rinfo->dir_dir);
}
+ /* FIXME: release caps/leases if error occurs */
for (i = 0; i < rinfo->dir_nr; i++) {
struct ceph_vino vino;
@@ -1202,9 +1381,10 @@ retry_lookup:
err = -ENOMEM;
goto out;
}
- err = ceph_init_dentry(dn);
- if (err < 0) {
+ ret = ceph_init_dentry(dn);
+ if (ret < 0) {
dput(dn);
+ err = ret;
goto out;
}
} else if (dn->d_inode &&
@@ -1217,16 +1397,13 @@ retry_lookup:
goto retry_lookup;
} else {
/* reorder parent's d_subdirs */
- spin_lock(&dcache_lock);
- spin_lock(&dn->d_lock);
+ spin_lock(&parent->d_lock);
+ spin_lock_nested(&dn->d_lock, DENTRY_D_LOCK_NESTED);
list_move(&dn->d_u.d_child, &parent->d_subdirs);
spin_unlock(&dn->d_lock);
- spin_unlock(&dcache_lock);
+ spin_unlock(&parent->d_lock);
}
- di = dn->d_fsdata;
- di->offset = ceph_make_fpos(frag, i + req->r_readdir_offset);
-
/* inode */
if (dn->d_inode) {
in = dn->d_inode;
@@ -1234,31 +1411,44 @@ retry_lookup:
in = ceph_get_inode(parent->d_sb, vino);
if (IS_ERR(in)) {
dout("new_inode badness\n");
- d_delete(dn);
+ d_drop(dn);
dput(dn);
err = PTR_ERR(in);
goto out;
}
- dn = splice_dentry(dn, in, NULL, false);
- if (IS_ERR(dn))
- dn = NULL;
}
if (fill_inode(in, &rinfo->dir_in[i], NULL, session,
req->r_request_started, -1,
&req->r_caps_reservation) < 0) {
pr_err("fill_inode badness on %p\n", in);
+ if (!dn->d_inode)
+ iput(in);
+ d_drop(dn);
goto next_item;
}
- if (dn)
- update_dentry_lease(dn, rinfo->dir_dlease[i],
- req->r_session,
- req->r_request_started);
+
+ if (!dn->d_inode) {
+ dn = splice_dentry(dn, in, NULL);
+ if (IS_ERR(dn)) {
+ err = PTR_ERR(dn);
+ dn = NULL;
+ goto next_item;
+ }
+ }
+
+ di = dn->d_fsdata;
+ di->offset = ceph_make_fpos(frag, i + r_readdir_offset);
+
+ update_dentry_lease(dn, rinfo->dir_dlease[i],
+ req->r_session,
+ req->r_request_started);
next_item:
if (dn)
dput(dn);
}
- req->r_did_prepopulate = true;
+ if (err == 0)
+ req->r_did_prepopulate = true;
out:
if (snapdir) {
@@ -1274,7 +1464,7 @@ int ceph_inode_set_size(struct inode *inode, loff_t size)
struct ceph_inode_info *ci = ceph_inode(inode);
int ret = 0;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
dout("set_size %p %llu -> %llu\n", inode, inode->i_size, size);
inode->i_size = size;
inode->i_blocks = (size + (1 << 9) - 1) >> 9;
@@ -1284,7 +1474,7 @@ int ceph_inode_set_size(struct inode *inode, loff_t size)
(ci->i_reported_size << 1) < ci->i_max_size)
ret = 1;
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return ret;
}
@@ -1294,12 +1484,13 @@ int ceph_inode_set_size(struct inode *inode, loff_t size)
*/
void ceph_queue_writeback(struct inode *inode)
{
+ ihold(inode);
if (queue_work(ceph_inode_to_client(inode)->wb_wq,
&ceph_inode(inode)->i_wb_work)) {
dout("ceph_queue_writeback %p\n", inode);
- igrab(inode);
} else {
dout("ceph_queue_writeback %p failed\n", inode);
+ iput(inode);
}
}
@@ -1319,55 +1510,13 @@ static void ceph_writeback_work(struct work_struct *work)
*/
void ceph_queue_invalidate(struct inode *inode)
{
+ ihold(inode);
if (queue_work(ceph_inode_to_client(inode)->pg_inv_wq,
&ceph_inode(inode)->i_pg_inv_work)) {
dout("ceph_queue_invalidate %p\n", inode);
- igrab(inode);
} else {
dout("ceph_queue_invalidate %p failed\n", inode);
- }
-}
-
-/*
- * invalidate any pages that are not dirty or under writeback. this
- * includes pages that are clean and mapped.
- */
-static void ceph_invalidate_nondirty_pages(struct address_space *mapping)
-{
- struct pagevec pvec;
- pgoff_t next = 0;
- int i;
-
- pagevec_init(&pvec, 0);
- while (pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) {
- for (i = 0; i < pagevec_count(&pvec); i++) {
- struct page *page = pvec.pages[i];
- pgoff_t index;
- int skip_page =
- (PageDirty(page) || PageWriteback(page));
-
- if (!skip_page)
- skip_page = !trylock_page(page);
-
- /*
- * We really shouldn't be looking at the ->index of an
- * unlocked page. But we're not allowed to lock these
- * pages. So we rely upon nobody altering the ->index
- * of this (pinned-by-us) page.
- */
- index = page->index;
- if (index > next)
- next = index;
- next++;
-
- if (skip_page)
- continue;
-
- generic_error_remove_page(mapping, page);
- unlock_page(page);
- }
- pagevec_release(&pvec);
- cond_resched();
+ iput(inode);
}
}
@@ -1383,44 +1532,47 @@ static void ceph_invalidate_work(struct work_struct *work)
u32 orig_gen;
int check = 0;
- spin_lock(&inode->i_lock);
+ mutex_lock(&ci->i_truncate_mutex);
+ spin_lock(&ci->i_ceph_lock);
dout("invalidate_pages %p gen %d revoking %d\n", inode,
ci->i_rdcache_gen, ci->i_rdcache_revoking);
- if (ci->i_rdcache_gen == 0 ||
- ci->i_rdcache_revoking != ci->i_rdcache_gen) {
- BUG_ON(ci->i_rdcache_revoking > ci->i_rdcache_gen);
- /* nevermind! */
- ci->i_rdcache_revoking = 0;
- spin_unlock(&inode->i_lock);
+ if (ci->i_rdcache_revoking != ci->i_rdcache_gen) {
+ if (__ceph_caps_revoking_other(ci, NULL, CEPH_CAP_FILE_CACHE))
+ check = 1;
+ spin_unlock(&ci->i_ceph_lock);
+ mutex_unlock(&ci->i_truncate_mutex);
goto out;
}
orig_gen = ci->i_rdcache_gen;
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
- ceph_invalidate_nondirty_pages(inode->i_mapping);
+ truncate_pagecache(inode, 0);
- spin_lock(&inode->i_lock);
- if (orig_gen == ci->i_rdcache_gen) {
+ spin_lock(&ci->i_ceph_lock);
+ if (orig_gen == ci->i_rdcache_gen &&
+ orig_gen == ci->i_rdcache_revoking) {
dout("invalidate_pages %p gen %d successful\n", inode,
ci->i_rdcache_gen);
- ci->i_rdcache_gen = 0;
- ci->i_rdcache_revoking = 0;
+ ci->i_rdcache_revoking--;
check = 1;
} else {
- dout("invalidate_pages %p gen %d raced, gen now %d\n",
- inode, orig_gen, ci->i_rdcache_gen);
+ dout("invalidate_pages %p gen %d raced, now %d revoking %d\n",
+ inode, orig_gen, ci->i_rdcache_gen,
+ ci->i_rdcache_revoking);
+ if (__ceph_caps_revoking_other(ci, NULL, CEPH_CAP_FILE_CACHE))
+ check = 1;
}
- spin_unlock(&inode->i_lock);
-
+ spin_unlock(&ci->i_ceph_lock);
+ mutex_unlock(&ci->i_truncate_mutex);
+out:
if (check)
ceph_check_caps(ci, 0, NULL);
-out:
iput(inode);
}
/*
- * called by trunc_wq; take i_mutex ourselves
+ * called by trunc_wq;
*
* We also truncate in a separate thread as well.
*/
@@ -1431,9 +1583,7 @@ static void ceph_vmtruncate_work(struct work_struct *work)
struct inode *inode = &ci->vfs_inode;
dout("vmtruncate_work %p\n", inode);
- mutex_lock(&inode->i_mutex);
__ceph_do_pending_vmtruncate(inode);
- mutex_unlock(&inode->i_mutex);
iput(inode);
}
@@ -1445,19 +1595,19 @@ void ceph_queue_vmtruncate(struct inode *inode)
{
struct ceph_inode_info *ci = ceph_inode(inode);
+ ihold(inode);
+
if (queue_work(ceph_sb_to_client(inode->i_sb)->trunc_wq,
&ci->i_vmtruncate_work)) {
dout("ceph_queue_vmtruncate %p\n", inode);
- igrab(inode);
} else {
dout("ceph_queue_vmtruncate %p failed, pending=%d\n",
inode, ci->i_truncate_pending);
+ iput(inode);
}
}
/*
- * called with i_mutex held.
- *
* Make sure any pending truncation is applied before doing anything
* that may depend on it.
*/
@@ -1465,13 +1615,15 @@ void __ceph_do_pending_vmtruncate(struct inode *inode)
{
struct ceph_inode_info *ci = ceph_inode(inode);
u64 to;
- int wrbuffer_refs, wake = 0;
+ int wrbuffer_refs, finish = 0;
+ mutex_lock(&ci->i_truncate_mutex);
retry:
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (ci->i_truncate_pending == 0) {
dout("__do_pending_vmtruncate %p none pending\n", inode);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
+ mutex_unlock(&ci->i_truncate_mutex);
return;
}
@@ -1482,32 +1634,39 @@ retry:
if (ci->i_wrbuffer_ref_head < ci->i_wrbuffer_ref) {
dout("__do_pending_vmtruncate %p flushing snaps first\n",
inode);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
filemap_write_and_wait_range(&inode->i_data, 0,
inode->i_sb->s_maxbytes);
goto retry;
}
+ /* there should be no reader or writer */
+ WARN_ON_ONCE(ci->i_rd_ref || ci->i_wr_ref);
+
to = ci->i_truncate_size;
wrbuffer_refs = ci->i_wrbuffer_ref;
dout("__do_pending_vmtruncate %p (%d) to %lld\n", inode,
ci->i_truncate_pending, to);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
+
+ truncate_pagecache(inode, to);
- truncate_inode_pages(inode->i_mapping, to);
+ spin_lock(&ci->i_ceph_lock);
+ if (to == ci->i_truncate_size) {
+ ci->i_truncate_pending = 0;
+ finish = 1;
+ }
+ spin_unlock(&ci->i_ceph_lock);
+ if (!finish)
+ goto retry;
- spin_lock(&inode->i_lock);
- ci->i_truncate_pending--;
- if (ci->i_truncate_pending == 0)
- wake = 1;
- spin_unlock(&inode->i_lock);
+ mutex_unlock(&ci->i_truncate_mutex);
if (wrbuffer_refs == 0)
ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
- if (wake)
- wake_up_all(&ci->i_cap_wq);
-}
+ wake_up_all(&ci->i_cap_wq);
+}
/*
* symlinks
@@ -1522,6 +1681,12 @@ static void *ceph_sym_follow_link(struct dentry *dentry, struct nameidata *nd)
static const struct inode_operations ceph_symlink_iops = {
.readlink = generic_readlink,
.follow_link = ceph_sym_follow_link,
+ .setattr = ceph_setattr,
+ .getattr = ceph_getattr,
+ .setxattr = ceph_setxattr,
+ .getxattr = ceph_getxattr,
+ .listxattr = ceph_listxattr,
+ .removexattr = ceph_removexattr,
};
/*
@@ -1531,7 +1696,6 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
{
struct inode *inode = dentry->d_inode;
struct ceph_inode_info *ci = ceph_inode(inode);
- struct inode *parent_inode = dentry->d_parent->d_inode;
const unsigned int ia_valid = attr->ia_valid;
struct ceph_mds_request *req;
struct ceph_mds_client *mdsc = ceph_sb_to_client(dentry->d_sb)->mdsc;
@@ -1539,12 +1703,11 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
int release = 0, dirtied = 0;
int mask = 0;
int err = 0;
+ int inode_dirty_flags = 0;
if (ceph_snap(inode) != CEPH_NOSNAP)
return -EROFS;
- __ceph_do_pending_vmtruncate(inode);
-
err = inode_change_ok(inode, attr);
if (err != 0)
return err;
@@ -1554,32 +1717,36 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
if (IS_ERR(req))
return PTR_ERR(req);
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
issued = __ceph_caps_issued(ci, NULL);
dout("setattr %p issued %s\n", inode, ceph_cap_string(issued));
if (ia_valid & ATTR_UID) {
dout("setattr %p uid %d -> %d\n", inode,
- inode->i_uid, attr->ia_uid);
+ from_kuid(&init_user_ns, inode->i_uid),
+ from_kuid(&init_user_ns, attr->ia_uid));
if (issued & CEPH_CAP_AUTH_EXCL) {
inode->i_uid = attr->ia_uid;
dirtied |= CEPH_CAP_AUTH_EXCL;
} else if ((issued & CEPH_CAP_AUTH_SHARED) == 0 ||
- attr->ia_uid != inode->i_uid) {
- req->r_args.setattr.uid = cpu_to_le32(attr->ia_uid);
+ !uid_eq(attr->ia_uid, inode->i_uid)) {
+ req->r_args.setattr.uid = cpu_to_le32(
+ from_kuid(&init_user_ns, attr->ia_uid));
mask |= CEPH_SETATTR_UID;
release |= CEPH_CAP_AUTH_SHARED;
}
}
if (ia_valid & ATTR_GID) {
dout("setattr %p gid %d -> %d\n", inode,
- inode->i_gid, attr->ia_gid);
+ from_kgid(&init_user_ns, inode->i_gid),
+ from_kgid(&init_user_ns, attr->ia_gid));
if (issued & CEPH_CAP_AUTH_EXCL) {
inode->i_gid = attr->ia_gid;
dirtied |= CEPH_CAP_AUTH_EXCL;
} else if ((issued & CEPH_CAP_AUTH_SHARED) == 0 ||
- attr->ia_gid != inode->i_gid) {
- req->r_args.setattr.gid = cpu_to_le32(attr->ia_gid);
+ !gid_eq(attr->ia_gid, inode->i_gid)) {
+ req->r_args.setattr.gid = cpu_to_le32(
+ from_kgid(&init_user_ns, attr->ia_gid));
mask |= CEPH_SETATTR_GID;
release |= CEPH_CAP_AUTH_SHARED;
}
@@ -1592,6 +1759,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
dirtied |= CEPH_CAP_AUTH_EXCL;
} else if ((issued & CEPH_CAP_AUTH_SHARED) == 0 ||
attr->ia_mode != inode->i_mode) {
+ inode->i_mode = attr->ia_mode;
req->r_args.setattr.mode = cpu_to_le32(attr->ia_mode);
mask |= CEPH_SETATTR_MODE;
release |= CEPH_CAP_AUTH_SHARED;
@@ -1697,28 +1865,40 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
dout("setattr %p ATTR_FILE ... hrm!\n", inode);
if (dirtied) {
- __ceph_mark_dirty_caps(ci, dirtied);
+ inode_dirty_flags = __ceph_mark_dirty_caps(ci, dirtied);
inode->i_ctime = CURRENT_TIME;
}
release &= issued;
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
+
+ if (inode_dirty_flags)
+ __mark_inode_dirty(inode, inode_dirty_flags);
+
+ if (ia_valid & ATTR_MODE) {
+ err = posix_acl_chmod(inode, attr->ia_mode);
+ if (err)
+ goto out_put;
+ }
if (mask) {
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_inode_drop = release;
req->r_args.setattr.mask = cpu_to_le32(mask);
req->r_num_caps = 1;
- err = ceph_mdsc_do_request(mdsc, parent_inode, req);
+ err = ceph_mdsc_do_request(mdsc, NULL, req);
}
dout("setattr %p result=%d (%s locally, %d remote)\n", inode, err,
ceph_cap_string(dirtied), mask);
ceph_mdsc_put_request(req);
- __ceph_do_pending_vmtruncate(inode);
+ if (mask & CEPH_SETATTR_SIZE)
+ __ceph_do_pending_vmtruncate(inode);
return err;
out:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
+out_put:
ceph_mdsc_put_request(req);
return err;
}
@@ -1739,14 +1919,15 @@ int ceph_do_getattr(struct inode *inode, int mask)
return 0;
}
- dout("do_getattr inode %p mask %s\n", inode, ceph_cap_string(mask));
+ dout("do_getattr inode %p mask %s mode 0%o\n", inode, ceph_cap_string(mask), inode->i_mode);
if (ceph_caps_issued_mask(ceph_inode(inode), mask, 1))
return 0;
req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_GETATTR, USE_ANY_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_num_caps = 1;
req->r_args.getattr.mask = cpu_to_le32(mask);
err = ceph_mdsc_do_request(mdsc, NULL, req);
@@ -1762,10 +1943,15 @@ int ceph_do_getattr(struct inode *inode, int mask)
*/
int ceph_permission(struct inode *inode, int mask)
{
- int err = ceph_do_getattr(inode, CEPH_CAP_AUTH_SHARED);
+ int err;
+
+ if (mask & MAY_NOT_BLOCK)
+ return -ECHILD;
+
+ err = ceph_do_getattr(inode, CEPH_CAP_AUTH_SHARED);
if (!err)
- err = generic_permission(inode, mask, NULL);
+ err = generic_permission(inode, mask);
return err;
}
@@ -1783,13 +1969,17 @@ int ceph_getattr(struct vfsmount *mnt, struct dentry *dentry,
err = ceph_do_getattr(inode, CEPH_STAT_CAP_INODE_ALL);
if (!err) {
generic_fillattr(inode, stat);
- stat->ino = inode->i_ino;
+ stat->ino = ceph_translate_ino(inode->i_sb, inode->i_ino);
if (ceph_snap(inode) != CEPH_NOSNAP)
stat->dev = ceph_snap(inode);
else
stat->dev = 0;
if (S_ISDIR(inode->i_mode)) {
- stat->size = ci->i_rbytes;
+ if (ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb),
+ RBYTES))
+ stat->size = ci->i_rbytes;
+ else
+ stat->size = ci->i_files + ci->i_subdirs;
stat->blocks = 0;
stat->blksize = 65536;
}
diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c
index 8888c9ba68d..a822a6e5829 100644
--- a/fs/ceph/ioctl.c
+++ b/fs/ceph/ioctl.c
@@ -1,9 +1,8 @@
+#include <linux/ceph/ceph_debug.h>
#include <linux/in.h>
#include "super.h"
#include "mds_client.h"
-#include <linux/ceph/ceph_debug.h>
-
#include "ioctl.h"
@@ -16,18 +15,17 @@
*/
static long ceph_ioctl_get_layout(struct file *file, void __user *arg)
{
- struct ceph_inode_info *ci = ceph_inode(file->f_dentry->d_inode);
+ struct ceph_inode_info *ci = ceph_inode(file_inode(file));
struct ceph_ioctl_layout l;
int err;
- err = ceph_do_getattr(file->f_dentry->d_inode, CEPH_STAT_CAP_LAYOUT);
+ err = ceph_do_getattr(file_inode(file), CEPH_STAT_CAP_LAYOUT);
if (!err) {
l.stripe_unit = ceph_file_layout_su(ci->i_layout);
l.stripe_count = ceph_file_layout_stripe_count(ci->i_layout);
l.object_size = ceph_file_layout_object_size(ci->i_layout);
l.data_pool = le32_to_cpu(ci->i_layout.fl_pg_pool);
- l.preferred_osd =
- (s32)le32_to_cpu(ci->i_layout.fl_pg_preferred);
+ l.preferred_osd = (s32)-1;
if (copy_to_user(arg, &l, sizeof(l)))
return -EFAULT;
}
@@ -35,45 +33,84 @@ static long ceph_ioctl_get_layout(struct file *file, void __user *arg)
return err;
}
+static long __validate_layout(struct ceph_mds_client *mdsc,
+ struct ceph_ioctl_layout *l)
+{
+ int i, err;
+
+ /* validate striping parameters */
+ if ((l->object_size & ~PAGE_MASK) ||
+ (l->stripe_unit & ~PAGE_MASK) ||
+ (l->stripe_unit != 0 &&
+ ((unsigned)l->object_size % (unsigned)l->stripe_unit)))
+ return -EINVAL;
+
+ /* make sure it's a valid data pool */
+ mutex_lock(&mdsc->mutex);
+ err = -EINVAL;
+ for (i = 0; i < mdsc->mdsmap->m_num_data_pg_pools; i++)
+ if (mdsc->mdsmap->m_data_pg_pools[i] == l->data_pool) {
+ err = 0;
+ break;
+ }
+ mutex_unlock(&mdsc->mutex);
+ if (err)
+ return err;
+
+ return 0;
+}
+
static long ceph_ioctl_set_layout(struct file *file, void __user *arg)
{
- struct inode *inode = file->f_dentry->d_inode;
- struct inode *parent_inode = file->f_dentry->d_parent->d_inode;
+ struct inode *inode = file_inode(file);
struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
struct ceph_mds_request *req;
struct ceph_ioctl_layout l;
- int err, i;
+ struct ceph_inode_info *ci = ceph_inode(file_inode(file));
+ struct ceph_ioctl_layout nl;
+ int err;
- /* copy and validate */
if (copy_from_user(&l, arg, sizeof(l)))
return -EFAULT;
- if ((l.object_size & ~PAGE_MASK) ||
- (l.stripe_unit & ~PAGE_MASK) ||
- !l.stripe_unit ||
- (l.object_size &&
- (unsigned)l.object_size % (unsigned)l.stripe_unit))
- return -EINVAL;
-
- /* make sure it's a valid data pool */
- if (l.data_pool > 0) {
- mutex_lock(&mdsc->mutex);
- err = -EINVAL;
- for (i = 0; i < mdsc->mdsmap->m_num_data_pg_pools; i++)
- if (mdsc->mdsmap->m_data_pg_pools[i] == l.data_pool) {
- err = 0;
- break;
- }
- mutex_unlock(&mdsc->mutex);
- if (err)
- return err;
- }
+ /* validate changed params against current layout */
+ err = ceph_do_getattr(file_inode(file), CEPH_STAT_CAP_LAYOUT);
+ if (err)
+ return err;
+
+ memset(&nl, 0, sizeof(nl));
+ if (l.stripe_count)
+ nl.stripe_count = l.stripe_count;
+ else
+ nl.stripe_count = ceph_file_layout_stripe_count(ci->i_layout);
+ if (l.stripe_unit)
+ nl.stripe_unit = l.stripe_unit;
+ else
+ nl.stripe_unit = ceph_file_layout_su(ci->i_layout);
+ if (l.object_size)
+ nl.object_size = l.object_size;
+ else
+ nl.object_size = ceph_file_layout_object_size(ci->i_layout);
+ if (l.data_pool)
+ nl.data_pool = l.data_pool;
+ else
+ nl.data_pool = ceph_file_layout_pg_pool(ci->i_layout);
+
+ /* this is obsolete, and always -1 */
+ nl.preferred_osd = le64_to_cpu(-1);
+
+ err = __validate_layout(mdsc, &nl);
+ if (err)
+ return err;
req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SETLAYOUT,
USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
+ req->r_num_caps = 1;
+
req->r_inode_drop = CEPH_CAP_FILE_SHARED | CEPH_CAP_FILE_EXCL;
req->r_args.setlayout.layout.fl_stripe_unit =
@@ -83,10 +120,8 @@ static long ceph_ioctl_set_layout(struct file *file, void __user *arg)
req->r_args.setlayout.layout.fl_object_size =
cpu_to_le32(l.object_size);
req->r_args.setlayout.layout.fl_pg_pool = cpu_to_le32(l.data_pool);
- req->r_args.setlayout.layout.fl_pg_preferred =
- cpu_to_le32(l.preferred_osd);
- err = ceph_mdsc_do_request(mdsc, parent_inode, req);
+ err = ceph_mdsc_do_request(mdsc, NULL, req);
ceph_mdsc_put_request(req);
return err;
}
@@ -99,43 +134,28 @@ static long ceph_ioctl_set_layout(struct file *file, void __user *arg)
*/
static long ceph_ioctl_set_layout_policy (struct file *file, void __user *arg)
{
- struct inode *inode = file->f_dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct ceph_mds_request *req;
struct ceph_ioctl_layout l;
- int err, i;
+ int err;
struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
/* copy and validate */
if (copy_from_user(&l, arg, sizeof(l)))
return -EFAULT;
- if ((l.object_size & ~PAGE_MASK) ||
- (l.stripe_unit & ~PAGE_MASK) ||
- !l.stripe_unit ||
- (l.object_size &&
- (unsigned)l.object_size % (unsigned)l.stripe_unit))
- return -EINVAL;
-
- /* make sure it's a valid data pool */
- if (l.data_pool > 0) {
- mutex_lock(&mdsc->mutex);
- err = -EINVAL;
- for (i = 0; i < mdsc->mdsmap->m_num_data_pg_pools; i++)
- if (mdsc->mdsmap->m_data_pg_pools[i] == l.data_pool) {
- err = 0;
- break;
- }
- mutex_unlock(&mdsc->mutex);
- if (err)
- return err;
- }
+ err = __validate_layout(mdsc, &l);
+ if (err)
+ return err;
req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SETDIRLAYOUT,
USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
+ req->r_num_caps = 1;
req->r_args.setlayout.layout.fl_stripe_unit =
cpu_to_le32(l.stripe_unit);
@@ -145,8 +165,6 @@ static long ceph_ioctl_set_layout_policy (struct file *file, void __user *arg)
cpu_to_le32(l.object_size);
req->r_args.setlayout.layout.fl_pg_pool =
cpu_to_le32(l.data_pool);
- req->r_args.setlayout.layout.fl_pg_preferred =
- cpu_to_le32(l.preferred_osd);
err = ceph_mdsc_do_request(mdsc, inode, req);
ceph_mdsc_put_request(req);
@@ -160,22 +178,29 @@ static long ceph_ioctl_set_layout_policy (struct file *file, void __user *arg)
static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg)
{
struct ceph_ioctl_dataloc dl;
- struct inode *inode = file->f_dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_osd_client *osdc =
&ceph_sb_to_client(inode->i_sb)->client->osdc;
+ struct ceph_object_locator oloc;
+ struct ceph_object_id oid;
u64 len = 1, olen;
u64 tmp;
- struct ceph_object_layout ol;
struct ceph_pg pgid;
+ int r;
/* copy and validate */
if (copy_from_user(&dl, arg, sizeof(dl)))
return -EFAULT;
down_read(&osdc->map_sem);
- ceph_calc_file_object_mapping(&ci->i_layout, dl.file_offset, &len,
- &dl.object_no, &dl.object_offset, &olen);
+ 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->map_sem);
+ return -EIO;
+ }
dl.file_offset -= dl.object_offset;
dl.object_size = ceph_file_layout_object_size(ci->i_layout);
dl.block_size = ceph_file_layout_su(ci->i_layout);
@@ -186,10 +211,16 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg)
snprintf(dl.object_name, sizeof(dl.object_name), "%llx.%08llx",
ceph_ino(inode), dl.object_no);
- ceph_calc_object_layout(&ol, dl.object_name, &ci->i_layout,
- osdc->osdmap);
- pgid = ol.ol_pgid;
+ oloc.pool = ceph_file_layout_pg_pool(ci->i_layout);
+ ceph_oid_set_name(&oid, dl.object_name);
+
+ r = ceph_oloc_oid_to_pg(osdc->osdmap, &oloc, &oid, &pgid);
+ if (r < 0) {
+ up_read(&osdc->map_sem);
+ return r;
+ }
+
dl.osd = ceph_calc_pg_primary(osdc->osdmap, pgid);
if (dl.osd >= 0) {
struct ceph_entity_addr *a =
@@ -211,15 +242,15 @@ static long ceph_ioctl_get_dataloc(struct file *file, void __user *arg)
static long ceph_ioctl_lazyio(struct file *file)
{
struct ceph_file_info *fi = file->private_data;
- struct inode *inode = file->f_dentry->d_inode;
+ struct inode *inode = file_inode(file);
struct ceph_inode_info *ci = ceph_inode(inode);
if ((fi->fmode & CEPH_FILE_MODE_LAZY) == 0) {
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
ci->i_nr_by_mode[fi->fmode]--;
fi->fmode |= CEPH_FILE_MODE_LAZY;
ci->i_nr_by_mode[fi->fmode]++;
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
dout("ioctl_layzio: file %p marked lazy\n", file);
ceph_check_caps(ci, 0, NULL);
@@ -229,6 +260,14 @@ static long ceph_ioctl_lazyio(struct file *file)
return 0;
}
+static long ceph_ioctl_syncio(struct file *file)
+{
+ struct ceph_file_info *fi = file->private_data;
+
+ fi->flags |= CEPH_F_SYNC;
+ return 0;
+}
+
long ceph_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
dout("ioctl file %p cmd %u arg %lu\n", file, cmd, arg);
@@ -247,6 +286,9 @@ long ceph_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
case CEPH_IOC_LAZYIO:
return ceph_ioctl_lazyio(file);
+
+ case CEPH_IOC_SYNCIO:
+ return ceph_ioctl_syncio(file);
}
return -ENOTTY;
diff --git a/fs/ceph/ioctl.h b/fs/ceph/ioctl.h
index a6ce54e94eb..c77028afb1e 100644
--- a/fs/ceph/ioctl.h
+++ b/fs/ceph/ioctl.h
@@ -4,12 +4,38 @@
#include <linux/ioctl.h>
#include <linux/types.h>
-#define CEPH_IOCTL_MAGIC 0x98
+#define CEPH_IOCTL_MAGIC 0x97
-/* just use u64 to align sanely on all archs */
+/*
+ * CEPH_IOC_GET_LAYOUT - get file layout or dir layout policy
+ * CEPH_IOC_SET_LAYOUT - set file layout
+ * CEPH_IOC_SET_LAYOUT_POLICY - set dir layout policy
+ *
+ * The file layout specifies how file data is striped over objects in
+ * the distributed object store, which object pool they belong to (if
+ * it differs from the default), and an optional 'preferred osd' to
+ * store them on.
+ *
+ * Files get a new layout based on the policy set on the containing
+ * directory or one of its ancestors. The GET_LAYOUT ioctl will let
+ * you examine the layout for a file or the policy on a directory.
+ *
+ * SET_LAYOUT will let you set a layout on a newly created file. This
+ * only works immediately after the file is created and before any
+ * data is written to it.
+ *
+ * SET_LAYOUT_POLICY will let you set a layout policy (default layout)
+ * on a directory that will apply to any new files created in that
+ * directory (or any child directory that doesn't specify a layout of
+ * its own).
+ */
+
+/* use u64 to align sanely on all archs */
struct ceph_ioctl_layout {
__u64 stripe_unit, stripe_count, object_size;
__u64 data_pool;
+
+ /* obsolete. new values ignored, always return -1 */
__s64 preferred_osd;
};
@@ -21,6 +47,8 @@ struct ceph_ioctl_layout {
struct ceph_ioctl_layout)
/*
+ * CEPH_IOC_GET_DATALOC - get location of file data in the cluster
+ *
* Extract identity, address of the OSD and object storing a given
* file offset.
*/
@@ -39,6 +67,34 @@ struct ceph_ioctl_dataloc {
#define CEPH_IOC_GET_DATALOC _IOWR(CEPH_IOCTL_MAGIC, 3, \
struct ceph_ioctl_dataloc)
+/*
+ * CEPH_IOC_LAZYIO - relax consistency
+ *
+ * Normally Ceph switches to synchronous IO when multiple clients have
+ * the file open (and or more for write). Reads and writes bypass the
+ * page cache and go directly to the OSD. Setting this flag on a file
+ * descriptor will allow buffered IO for this file in cases where the
+ * application knows it won't interfere with other nodes (or doesn't
+ * care).
+ */
#define CEPH_IOC_LAZYIO _IO(CEPH_IOCTL_MAGIC, 4)
+/*
+ * CEPH_IOC_SYNCIO - force synchronous IO
+ *
+ * This ioctl sets a file flag that forces the synchronous IO that
+ * bypasses the page cache, even if it is not necessary. This is
+ * essentially the opposite behavior of IOC_LAZYIO. This forces the
+ * same read/write path as a file opened by multiple clients when one
+ * or more of those clients is opened for write.
+ *
+ * Note that this type of sync IO takes a different path than a file
+ * opened with O_SYNC/D_SYNC (writes hit the page cache and are
+ * immediately flushed on page boundaries). It is very similar to
+ * O_DIRECT (writes bypass the page cache) excep that O_DIRECT writes
+ * are not copied (user page must remain stable) and O_DIRECT writes
+ * have alignment restrictions (on the buffer and file offset).
+ */
+#define CEPH_IOC_SYNCIO _IO(CEPH_IOCTL_MAGIC, 5)
+
#endif
diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c
index 40abde93c34..fbc39c47bac 100644
--- a/fs/ceph/locks.c
+++ b/fs/ceph/locks.c
@@ -2,49 +2,97 @@
#include <linux/file.h>
#include <linux/namei.h>
+#include <linux/random.h>
#include "super.h"
#include "mds_client.h"
#include <linux/ceph/pagelist.h>
+static u64 lock_secret;
+
+static inline u64 secure_addr(void *addr)
+{
+ u64 v = lock_secret ^ (u64)(unsigned long)addr;
+ /*
+ * Set the most significant bit, so that MDS knows the 'owner'
+ * is sufficient to identify the owner of lock. (old code uses
+ * both 'owner' and 'pid')
+ */
+ v |= (1ULL << 63);
+ return v;
+}
+
+void __init ceph_flock_init(void)
+{
+ get_random_bytes(&lock_secret, sizeof(lock_secret));
+}
+
/**
* Implement fcntl and flock locking functions.
*/
static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
- u64 pid, u64 pid_ns,
- int cmd, u64 start, u64 length, u8 wait)
+ int cmd, u8 wait, struct file_lock *fl)
{
- struct inode *inode = file->f_dentry->d_inode;
- struct ceph_mds_client *mdsc =
- ceph_sb_to_client(inode->i_sb)->mdsc;
+ struct inode *inode = file_inode(file);
+ struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
struct ceph_mds_request *req;
int err;
+ u64 length = 0;
+ u64 owner;
req = ceph_mdsc_create_request(mdsc, operation, USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
+ req->r_num_caps = 1;
- dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, "
- "length: %llu, wait: %d, type`: %d", (int)lock_type,
- (int)operation, pid, start, length, wait, cmd);
+ /* mds requires start and length rather than start and end */
+ if (LLONG_MAX == fl->fl_end)
+ length = 0;
+ else
+ length = fl->fl_end - fl->fl_start + 1;
+
+ 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,
+ (int)operation, owner, (u64)fl->fl_pid, fl->fl_start, length,
+ wait, fl->fl_type);
req->r_args.filelock_change.rule = lock_type;
req->r_args.filelock_change.type = cmd;
- req->r_args.filelock_change.pid = cpu_to_le64(pid);
- /* This should be adjusted, but I'm not sure if
- namespaces actually get id numbers*/
- req->r_args.filelock_change.pid_namespace =
- cpu_to_le64((u64)pid_ns);
- req->r_args.filelock_change.start = cpu_to_le64(start);
+ req->r_args.filelock_change.owner = cpu_to_le64(owner);
+ req->r_args.filelock_change.pid = cpu_to_le64((u64)fl->fl_pid);
+ req->r_args.filelock_change.start = cpu_to_le64(fl->fl_start);
req->r_args.filelock_change.length = cpu_to_le64(length);
req->r_args.filelock_change.wait = wait;
err = ceph_mdsc_do_request(mdsc, inode, req);
+
+ if (operation == CEPH_MDS_OP_GETFILELOCK) {
+ fl->fl_pid = le64_to_cpu(req->r_reply_info.filelock_reply->pid);
+ if (CEPH_LOCK_SHARED == req->r_reply_info.filelock_reply->type)
+ fl->fl_type = F_RDLCK;
+ else if (CEPH_LOCK_EXCL == req->r_reply_info.filelock_reply->type)
+ fl->fl_type = F_WRLCK;
+ else
+ fl->fl_type = F_UNLCK;
+
+ fl->fl_start = le64_to_cpu(req->r_reply_info.filelock_reply->start);
+ length = le64_to_cpu(req->r_reply_info.filelock_reply->start) +
+ le64_to_cpu(req->r_reply_info.filelock_reply->length);
+ if (length >= 1)
+ fl->fl_end = length -1;
+ else
+ fl->fl_end = 0;
+
+ }
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,
- (int)operation, pid, start, length, wait, cmd, err);
+ "length: %llu, wait: %d, type: %d, err code %d", (int)lock_type,
+ (int)operation, (u64)fl->fl_pid, fl->fl_start,
+ length, wait, fl->fl_type, err);
return err;
}
@@ -54,20 +102,24 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
*/
int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
{
- u64 length;
u8 lock_cmd;
int err;
u8 wait = 0;
u16 op = CEPH_MDS_OP_SETFILELOCK;
- fl->fl_nspid = get_pid(task_tgid(current));
- dout("ceph_lock, fl_pid:%d", fl->fl_pid);
+ if (!(fl->fl_flags & FL_POSIX))
+ return -ENOLCK;
+ /* No mandatory locks */
+ if (__mandatory_lock(file->f_mapping->host) && fl->fl_type != F_UNLCK)
+ return -ENOLCK;
+
+ dout("ceph_lock, fl_owner: %p", fl->fl_owner);
/* set wait bit as appropriate, then make command as Ceph expects it*/
- if (F_SETLKW == cmd)
- wait = 1;
- if (F_GETLK == cmd)
+ if (IS_GETLK(cmd))
op = CEPH_MDS_OP_GETFILELOCK;
+ else if (IS_SETLKW(cmd))
+ wait = 1;
if (F_RDLCK == fl->fl_type)
lock_cmd = CEPH_LOCK_SHARED;
@@ -76,87 +128,75 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl)
else
lock_cmd = CEPH_LOCK_UNLOCK;
- if (LLONG_MAX == fl->fl_end)
- length = 0;
- else
- length = fl->fl_end - fl->fl_start + 1;
-
- err = ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
- (u64)fl->fl_pid,
- (u64)(unsigned long)fl->fl_nspid,
- lock_cmd, fl->fl_start,
- length, wait);
+ err = ceph_lock_message(CEPH_LOCK_FCNTL, op, file, lock_cmd, wait, fl);
if (!err) {
- dout("mds locked, locking locally");
- err = posix_lock_file(file, fl, NULL);
- if (err && (CEPH_MDS_OP_SETFILELOCK == op)) {
- /* undo! This should only happen if the kernel detects
- * local deadlock. */
- ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
- (u64)fl->fl_pid,
- (u64)(unsigned long)fl->fl_nspid,
- CEPH_LOCK_UNLOCK, fl->fl_start,
- length, 0);
- dout("got %d on posix_lock_file, undid lock", err);
+ if (op != CEPH_MDS_OP_GETFILELOCK) {
+ dout("mds locked, locking locally");
+ err = posix_lock_file(file, fl, NULL);
+ if (err && (CEPH_MDS_OP_SETFILELOCK == op)) {
+ /* undo! This should only happen if
+ * the kernel detects local
+ * deadlock. */
+ ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
+ CEPH_LOCK_UNLOCK, 0, fl);
+ dout("got %d on posix_lock_file, undid lock",
+ err);
+ }
}
- } else {
- dout("mds returned error code %d", err);
+
+ } else if (err == -ERESTARTSYS) {
+ dout("undoing lock\n");
+ ceph_lock_message(CEPH_LOCK_FCNTL, op, file,
+ CEPH_LOCK_UNLOCK, 0, fl);
}
return err;
}
int ceph_flock(struct file *file, int cmd, struct file_lock *fl)
{
- u64 length;
u8 lock_cmd;
int err;
- u8 wait = 1;
-
- fl->fl_nspid = get_pid(task_tgid(current));
- dout("ceph_flock, fl_pid:%d", fl->fl_pid);
-
- /* set wait bit, then clear it out of cmd*/
- if (cmd & LOCK_NB)
- wait = 0;
- cmd = cmd & (LOCK_SH | LOCK_EX | LOCK_UN);
- /* set command sequence that Ceph wants to see:
- shared lock, exclusive lock, or unlock */
- if (LOCK_SH == cmd)
+ u8 wait = 0;
+
+ if (!(fl->fl_flags & FL_FLOCK))
+ return -ENOLCK;
+ /* No mandatory locks */
+ if (__mandatory_lock(file->f_mapping->host) && fl->fl_type != F_UNLCK)
+ return -ENOLCK;
+
+ dout("ceph_flock, fl_file: %p", fl->fl_file);
+
+ if (IS_SETLKW(cmd))
+ wait = 1;
+
+ if (F_RDLCK == fl->fl_type)
lock_cmd = CEPH_LOCK_SHARED;
- else if (LOCK_EX == cmd)
+ else if (F_WRLCK == fl->fl_type)
lock_cmd = CEPH_LOCK_EXCL;
else
lock_cmd = CEPH_LOCK_UNLOCK;
- /* mds requires start and length rather than start and end */
- if (LLONG_MAX == fl->fl_end)
- length = 0;
- else
- length = fl->fl_end - fl->fl_start + 1;
err = ceph_lock_message(CEPH_LOCK_FLOCK, CEPH_MDS_OP_SETFILELOCK,
- file, (u64)fl->fl_pid,
- (u64)(unsigned long)fl->fl_nspid,
- lock_cmd, fl->fl_start,
- length, wait);
+ file, lock_cmd, wait, fl);
if (!err) {
err = flock_lock_file_wait(file, fl);
if (err) {
ceph_lock_message(CEPH_LOCK_FLOCK,
CEPH_MDS_OP_SETFILELOCK,
- file, (u64)fl->fl_pid,
- (u64)(unsigned long)fl->fl_nspid,
- CEPH_LOCK_UNLOCK, fl->fl_start,
- length, 0);
+ file, CEPH_LOCK_UNLOCK, 0, fl);
dout("got %d on flock_lock_file_wait, undid lock", err);
}
- } else {
- dout("mds error code %d", err);
+ } else if (err == -ERESTARTSYS) {
+ dout("undoing lock\n");
+ ceph_lock_message(CEPH_LOCK_FLOCK,
+ CEPH_MDS_OP_SETFILELOCK,
+ file, CEPH_LOCK_UNLOCK, 0, fl);
}
return err;
}
/**
- * Must be called with BKL already held. Fills in the passed
+ * Must be called with lock_flocks() already held. Fills in the passed
* counter variables, so you can prepare pagelist metadata before calling
* ceph_encode_locks.
*/
@@ -178,27 +218,23 @@ void ceph_count_locks(struct inode *inode, int *fcntl_count, int *flock_count)
}
/**
- * Encode the flock and fcntl locks for the given inode into the pagelist.
- * Format is: #fcntl locks, sequential fcntl locks, #flock locks,
- * sequential flock locks.
- * Must be called with lock_flocks() already held.
- * If we encounter more of a specific lock type than expected,
- * we return the value 1.
+ * Encode the flock and fcntl locks for the given inode into the ceph_filelock
+ * array. Must be called with inode->i_lock already held.
+ * If we encounter more of a specific lock type than expected, return -ENOSPC.
*/
-int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist,
- int num_fcntl_locks, int num_flock_locks)
+int ceph_encode_locks_to_buffer(struct inode *inode,
+ struct ceph_filelock *flocks,
+ int num_fcntl_locks, int num_flock_locks)
{
struct file_lock *lock;
- struct ceph_filelock cephlock;
int err = 0;
int seen_fcntl = 0;
int seen_flock = 0;
+ int l = 0;
dout("encoding %d flock and %d fcntl locks", num_flock_locks,
num_fcntl_locks);
- err = ceph_pagelist_append(pagelist, &num_fcntl_locks, sizeof(u32));
- if (err)
- goto fail;
+
for (lock = inode->i_flock; lock != NULL; lock = lock->fl_next) {
if (lock->fl_flags & FL_POSIX) {
++seen_fcntl;
@@ -206,19 +242,12 @@ int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist,
err = -ENOSPC;
goto fail;
}
- err = lock_to_ceph_filelock(lock, &cephlock);
+ err = lock_to_ceph_filelock(lock, &flocks[l]);
if (err)
goto fail;
- err = ceph_pagelist_append(pagelist, &cephlock,
- sizeof(struct ceph_filelock));
+ ++l;
}
- if (err)
- goto fail;
}
-
- err = ceph_pagelist_append(pagelist, &num_flock_locks, sizeof(u32));
- if (err)
- goto fail;
for (lock = inode->i_flock; lock != NULL; lock = lock->fl_next) {
if (lock->fl_flags & FL_FLOCK) {
++seen_flock;
@@ -226,19 +255,51 @@ int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist,
err = -ENOSPC;
goto fail;
}
- err = lock_to_ceph_filelock(lock, &cephlock);
+ err = lock_to_ceph_filelock(lock, &flocks[l]);
if (err)
goto fail;
- err = ceph_pagelist_append(pagelist, &cephlock,
- sizeof(struct ceph_filelock));
+ ++l;
}
- if (err)
- goto fail;
}
fail:
return err;
}
+/**
+ * Copy the encoded flock and fcntl locks into the pagelist.
+ * Format is: #fcntl locks, sequential fcntl locks, #flock locks,
+ * sequential flock locks.
+ * Returns zero on success.
+ */
+int ceph_locks_to_pagelist(struct ceph_filelock *flocks,
+ struct ceph_pagelist *pagelist,
+ int num_fcntl_locks, int num_flock_locks)
+{
+ int err = 0;
+ __le32 nlocks;
+
+ nlocks = cpu_to_le32(num_fcntl_locks);
+ err = ceph_pagelist_append(pagelist, &nlocks, sizeof(nlocks));
+ if (err)
+ goto out_fail;
+
+ err = ceph_pagelist_append(pagelist, flocks,
+ num_fcntl_locks * sizeof(*flocks));
+ if (err)
+ goto out_fail;
+
+ nlocks = cpu_to_le32(num_flock_locks);
+ err = ceph_pagelist_append(pagelist, &nlocks, sizeof(nlocks));
+ if (err)
+ goto out_fail;
+
+ err = ceph_pagelist_append(pagelist,
+ &flocks[num_fcntl_locks],
+ num_flock_locks * sizeof(*flocks));
+out_fail:
+ return err;
+}
+
/*
* Given a pointer to a lock, convert it to a ceph filelock
*/
@@ -246,13 +307,11 @@ int lock_to_ceph_filelock(struct file_lock *lock,
struct ceph_filelock *cephlock)
{
int err = 0;
-
cephlock->start = cpu_to_le64(lock->fl_start);
cephlock->length = cpu_to_le64(lock->fl_end - lock->fl_start + 1);
cephlock->client = cpu_to_le64(0);
- cephlock->pid = cpu_to_le64(lock->fl_pid);
- cephlock->pid_namespace =
- cpu_to_le64((u64)(unsigned long)lock->fl_nspid);
+ cephlock->pid = cpu_to_le64((u64)lock->fl_pid);
+ cephlock->owner = cpu_to_le64(secure_addr(lock->fl_owner));
switch (lock->fl_type) {
case F_RDLCK:
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 3142b15940c..92a2548278f 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -3,14 +3,15 @@
#include <linux/fs.h>
#include <linux/wait.h>
#include <linux/slab.h>
+#include <linux/gfp.h>
#include <linux/sched.h>
#include <linux/debugfs.h>
#include <linux/seq_file.h>
-#include <linux/smp_lock.h>
#include "super.h"
#include "mds_client.h"
+#include <linux/ceph/ceph_features.h>
#include <linux/ceph/messenger.h>
#include <linux/ceph/decode.h>
#include <linux/ceph/pagelist.h>
@@ -43,6 +44,7 @@
*/
struct ceph_reconnect_state {
+ int nr_caps;
struct ceph_pagelist *pagelist;
bool flock;
};
@@ -61,7 +63,8 @@ static const struct ceph_connection_operations mds_con_ops;
* parse individual inode info
*/
static int parse_reply_info_in(void **p, void *end,
- struct ceph_mds_reply_info_in *info)
+ struct ceph_mds_reply_info_in *info,
+ u64 features)
{
int err = -EIO;
@@ -75,6 +78,12 @@ static int parse_reply_info_in(void **p, void *end,
info->symlink = *p;
*p += info->symlink_len;
+ if (features & CEPH_FEATURE_DIRLAYOUTHASH)
+ ceph_decode_copy_safe(p, end, &info->dir_layout,
+ sizeof(info->dir_layout), bad);
+ else
+ memset(&info->dir_layout, 0, sizeof(info->dir_layout));
+
ceph_decode_32_safe(p, end, info->xattr_len, bad);
ceph_decode_need(p, end, info->xattr_len, bad);
info->xattr_data = *p;
@@ -89,12 +98,13 @@ bad:
* target inode.
*/
static int parse_reply_info_trace(void **p, void *end,
- struct ceph_mds_reply_info_parsed *info)
+ struct ceph_mds_reply_info_parsed *info,
+ u64 features)
{
int err;
if (info->head->is_dentry) {
- err = parse_reply_info_in(p, end, &info->diri);
+ err = parse_reply_info_in(p, end, &info->diri, features);
if (err < 0)
goto out_bad;
@@ -115,7 +125,7 @@ static int parse_reply_info_trace(void **p, void *end,
}
if (info->head->is_target) {
- err = parse_reply_info_in(p, end, &info->targeti);
+ err = parse_reply_info_in(p, end, &info->targeti, features);
if (err < 0)
goto out_bad;
}
@@ -135,7 +145,8 @@ out_bad:
* parse readdir results
*/
static int parse_reply_info_dir(void **p, void *end,
- struct ceph_mds_reply_info_parsed *info)
+ struct ceph_mds_reply_info_parsed *info,
+ u64 features)
{
u32 num, i = 0;
int err;
@@ -155,21 +166,18 @@ static int parse_reply_info_dir(void **p, void *end,
if (num == 0)
goto done;
- /* alloc large array */
- info->dir_nr = num;
- info->dir_in = kcalloc(num, sizeof(*info->dir_in) +
- sizeof(*info->dir_dname) +
- sizeof(*info->dir_dname_len) +
- sizeof(*info->dir_dlease),
- GFP_NOFS);
- if (info->dir_in == NULL) {
- err = -ENOMEM;
- goto out_bad;
- }
+ BUG_ON(!info->dir_in);
info->dir_dname = (void *)(info->dir_in + num);
info->dir_dname_len = (void *)(info->dir_dname + num);
info->dir_dlease = (void *)(info->dir_dname_len + num);
+ if ((unsigned long)(info->dir_dlease + num) >
+ (unsigned long)info->dir_in + info->dir_buf_size) {
+ pr_err("dir contents are larger than expected\n");
+ WARN_ON(1);
+ goto bad;
+ }
+ info->dir_nr = num;
while (num) {
/* dentry */
ceph_decode_need(p, end, sizeof(u32)*2, bad);
@@ -183,7 +191,7 @@ static int parse_reply_info_dir(void **p, void *end,
*p += sizeof(struct ceph_mds_reply_lease);
/* inode */
- err = parse_reply_info_in(p, end, &info->dir_in[i]);
+ err = parse_reply_info_in(p, end, &info->dir_in[i], features);
if (err < 0)
goto out_bad;
i++;
@@ -203,10 +211,74 @@ out_bad:
}
/*
+ * parse fcntl F_GETLK results
+ */
+static int parse_reply_info_filelock(void **p, void *end,
+ struct ceph_mds_reply_info_parsed *info,
+ u64 features)
+{
+ if (*p + sizeof(*info->filelock_reply) > end)
+ goto bad;
+
+ info->filelock_reply = *p;
+ *p += sizeof(*info->filelock_reply);
+
+ if (unlikely(*p != end))
+ goto bad;
+ return 0;
+
+bad:
+ return -EIO;
+}
+
+/*
+ * parse create results
+ */
+static int parse_reply_info_create(void **p, void *end,
+ struct ceph_mds_reply_info_parsed *info,
+ u64 features)
+{
+ if (features & CEPH_FEATURE_REPLY_CREATE_INODE) {
+ if (*p == end) {
+ info->has_create_ino = false;
+ } else {
+ info->has_create_ino = true;
+ info->ino = ceph_decode_64(p);
+ }
+ }
+
+ if (unlikely(*p != end))
+ goto bad;
+ return 0;
+
+bad:
+ return -EIO;
+}
+
+/*
+ * parse extra results
+ */
+static int parse_reply_info_extra(void **p, void *end,
+ struct ceph_mds_reply_info_parsed *info,
+ u64 features)
+{
+ if (info->head->op == CEPH_MDS_OP_GETFILELOCK)
+ return parse_reply_info_filelock(p, end, info, features);
+ else if (info->head->op == CEPH_MDS_OP_READDIR ||
+ info->head->op == CEPH_MDS_OP_LSSNAP)
+ return parse_reply_info_dir(p, end, info, features);
+ else if (info->head->op == CEPH_MDS_OP_CREATE)
+ return parse_reply_info_create(p, end, info, features);
+ else
+ return -EIO;
+}
+
+/*
* parse entire mds reply
*/
static int parse_reply_info(struct ceph_msg *msg,
- struct ceph_mds_reply_info_parsed *info)
+ struct ceph_mds_reply_info_parsed *info,
+ u64 features)
{
void *p, *end;
u32 len;
@@ -219,15 +291,17 @@ static int parse_reply_info(struct ceph_msg *msg,
/* trace */
ceph_decode_32_safe(&p, end, len, bad);
if (len > 0) {
- err = parse_reply_info_trace(&p, p+len, info);
+ ceph_decode_need(&p, end, len, bad);
+ err = parse_reply_info_trace(&p, p+len, info, features);
if (err < 0)
goto out_bad;
}
- /* dir content */
+ /* extra */
ceph_decode_32_safe(&p, end, len, bad);
if (len > 0) {
- err = parse_reply_info_dir(&p, p+len, info);
+ ceph_decode_need(&p, end, len, bad);
+ err = parse_reply_info_extra(&p, p+len, info, features);
if (err < 0)
goto out_bad;
}
@@ -251,7 +325,9 @@ out_bad:
static void destroy_reply_info(struct ceph_mds_reply_info_parsed *info)
{
- kfree(info->dir_in);
+ if (!info->dir_in)
+ return;
+ free_pages((unsigned long)info->dir_in, get_order(info->dir_buf_size));
}
@@ -289,10 +365,10 @@ void ceph_put_mds_session(struct ceph_mds_session *s)
dout("mdsc put_session %p %d -> %d\n", s,
atomic_read(&s->s_ref), atomic_read(&s->s_ref)-1);
if (atomic_dec_and_test(&s->s_ref)) {
- if (s->s_authorizer)
- s->s_mdsc->fsc->client->monc.auth->ops->destroy_authorizer(
- s->s_mdsc->fsc->client->monc.auth,
- s->s_authorizer);
+ if (s->s_auth.authorizer)
+ ceph_auth_destroy_authorizer(
+ s->s_mdsc->fsc->client->monc.auth,
+ s->s_auth.authorizer);
kfree(s);
}
}
@@ -339,6 +415,9 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
{
struct ceph_mds_session *s;
+ if (mds >= mdsc->mdsmap->m_max_mds)
+ return ERR_PTR(-EINVAL);
+
s = kzalloc(sizeof(*s), GFP_NOFS);
if (!s)
return ERR_PTR(-ENOMEM);
@@ -349,15 +428,13 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
s->s_seq = 0;
mutex_init(&s->s_mutex);
- ceph_con_init(mdsc->fsc->client->msgr, &s->s_con);
- s->s_con.private = s;
- s->s_con.ops = &mds_con_ops;
- s->s_con.peer_name.type = CEPH_ENTITY_TYPE_MDS;
- s->s_con.peer_name.num = cpu_to_le64(mds);
+ ceph_con_init(&s->s_con, s, &mds_con_ops, &mdsc->fsc->client->msgr);
- spin_lock_init(&s->s_cap_lock);
+ spin_lock_init(&s->s_gen_ttl_lock);
s->s_cap_gen = 0;
- s->s_cap_ttl = 0;
+ s->s_cap_ttl = jiffies - 1;
+
+ spin_lock_init(&s->s_cap_lock);
s->s_renew_requested = 0;
s->s_renew_seq = 0;
INIT_LIST_HEAD(&s->s_caps);
@@ -367,6 +444,7 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
INIT_LIST_HEAD(&s->s_waiting);
INIT_LIST_HEAD(&s->s_unsafe);
s->s_num_cap_releases = 0;
+ s->s_cap_reconnect = 0;
s->s_cap_iterator = NULL;
INIT_LIST_HEAD(&s->s_cap_releases);
INIT_LIST_HEAD(&s->s_cap_releases_done);
@@ -393,7 +471,8 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
mdsc->sessions[mds] = s;
atomic_inc(&s->s_ref); /* one ref to sessions[], one to caller */
- ceph_con_open(&s->s_con, ceph_mdsmap_get_addr(mdsc->mdsmap, mds));
+ ceph_con_open(&s->s_con, CEPH_ENTITY_TYPE_MDS, mds,
+ ceph_mdsmap_get_addr(mdsc->mdsmap, mds));
return s;
@@ -433,29 +512,33 @@ void ceph_mdsc_release_request(struct kref *kref)
struct ceph_mds_request *req = container_of(kref,
struct ceph_mds_request,
r_kref);
+ destroy_reply_info(&req->r_reply_info);
if (req->r_request)
ceph_msg_put(req->r_request);
- if (req->r_reply) {
+ if (req->r_reply)
ceph_msg_put(req->r_reply);
- destroy_reply_info(&req->r_reply_info);
- }
if (req->r_inode) {
- ceph_put_cap_refs(ceph_inode(req->r_inode),
- CEPH_CAP_PIN);
+ ceph_put_cap_refs(ceph_inode(req->r_inode), CEPH_CAP_PIN);
iput(req->r_inode);
}
if (req->r_locked_dir)
- ceph_put_cap_refs(ceph_inode(req->r_locked_dir),
- CEPH_CAP_PIN);
+ ceph_put_cap_refs(ceph_inode(req->r_locked_dir), CEPH_CAP_PIN);
if (req->r_target_inode)
iput(req->r_target_inode);
if (req->r_dentry)
dput(req->r_dentry);
- if (req->r_old_dentry) {
- ceph_put_cap_refs(
- ceph_inode(req->r_old_dentry->d_parent->d_inode),
- CEPH_CAP_PIN);
+ if (req->r_old_dentry)
dput(req->r_old_dentry);
+ if (req->r_old_dentry_dir) {
+ /*
+ * track (and drop pins for) r_old_dentry_dir
+ * separately, since r_old_dentry's d_parent may have
+ * changed between the dir mutex being dropped and
+ * this request being freed.
+ */
+ ceph_put_cap_refs(ceph_inode(req->r_old_dentry_dir),
+ CEPH_CAP_PIN);
+ iput(req->r_old_dentry_dir);
}
kfree(req->r_path1);
kfree(req->r_path2);
@@ -529,9 +612,13 @@ static void __register_request(struct ceph_mds_client *mdsc,
ceph_mdsc_get_request(req);
__insert_request(mdsc, req);
+ req->r_uid = current_fsuid();
+ req->r_gid = current_fsgid();
+
if (dir) {
struct ceph_inode_info *ci = ceph_inode(dir);
+ ihold(dir);
spin_lock(&ci->i_unsafe_lock);
req->r_unsafe_dir = dir;
list_add_tail(&req->r_unsafe_dir_item, &ci->i_unsafe_dirops);
@@ -552,8 +639,13 @@ static void __unregister_request(struct ceph_mds_client *mdsc,
spin_lock(&ci->i_unsafe_lock);
list_del_init(&req->r_unsafe_dir_item);
spin_unlock(&ci->i_unsafe_lock);
+
+ iput(req->r_unsafe_dir);
+ req->r_unsafe_dir = NULL;
}
+ complete_all(&req->r_safe_completion);
+
ceph_mdsc_put_request(req);
}
@@ -565,8 +657,14 @@ static void __unregister_request(struct ceph_mds_client *mdsc,
*
* Called under mdsc->mutex.
*/
-struct dentry *get_nonsnap_parent(struct dentry *dentry)
+static struct dentry *get_nonsnap_parent(struct dentry *dentry)
{
+ /*
+ * we don't need to worry about protecting the d_parent access
+ * here because we never renaming inside the snapped namespace
+ * except to resplice to another snapdir, and either the old or new
+ * result is a valid result.
+ */
while (!IS_ROOT(dentry) && ceph_snap(dentry->d_inode) != CEPH_NOSNAP)
dentry = dentry->d_parent;
return dentry;
@@ -602,7 +700,9 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
if (req->r_inode) {
inode = req->r_inode;
} else if (req->r_dentry) {
- struct inode *dir = req->r_dentry->d_parent->d_inode;
+ /* ignore race with rename; old or new d_parent is okay */
+ struct dentry *parent = req->r_dentry->d_parent;
+ struct inode *dir = parent->d_inode;
if (dir->i_sb != mdsc->fsc->sb) {
/* not this fs! */
@@ -610,18 +710,18 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
} else if (ceph_snap(dir) != CEPH_NOSNAP) {
/* direct snapped/virtual snapdir requests
* based on parent dir inode */
- struct dentry *dn =
- get_nonsnap_parent(req->r_dentry->d_parent);
+ struct dentry *dn = get_nonsnap_parent(parent);
inode = dn->d_inode;
dout("__choose_mds using nonsnap parent %p\n", inode);
- } else if (req->r_dentry->d_inode) {
+ } else {
/* dentry target */
inode = req->r_dentry->d_inode;
- } else {
- /* dir + name */
- inode = dir;
- hash = req->r_dentry->d_name.hash;
- is_hash = true;
+ if (!inode || mode == USE_AUTH_MDS) {
+ /* dir + name */
+ inode = dir;
+ hash = ceph_dentry_hash(dir, req->r_dentry);
+ is_hash = true;
+ }
}
}
@@ -647,9 +747,11 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
dout("choose_mds %p %llx.%llx "
"frag %u mds%d (%d/%d)\n",
inode, ceph_vinop(inode),
- frag.frag, frag.mds,
+ frag.frag, mds,
(int)r, frag.ndist);
- return mds;
+ if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
+ CEPH_MDS_STATE_ACTIVE)
+ return mds;
}
/* since this file/dir wasn't known to be
@@ -662,26 +764,28 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
dout("choose_mds %p %llx.%llx "
"frag %u mds%d (auth)\n",
inode, ceph_vinop(inode), frag.frag, mds);
- return mds;
+ if (ceph_mdsmap_get_state(mdsc->mdsmap, mds) >=
+ CEPH_MDS_STATE_ACTIVE)
+ return mds;
}
}
}
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
cap = NULL;
if (mode == USE_AUTH_MDS)
cap = ci->i_auth_cap;
if (!cap && !RB_EMPTY_ROOT(&ci->i_caps))
cap = rb_entry(rb_first(&ci->i_caps), struct ceph_cap, ci_node);
if (!cap) {
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
goto random;
}
mds = cap->session->s_mds;
dout("choose_mds %p %llx.%llx mds%d (%scap %p)\n",
inode, ceph_vinop(inode), mds,
cap == ci->i_auth_cap ? "auth " : "", cap);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return mds;
random:
@@ -699,7 +803,8 @@ static struct ceph_msg *create_session_msg(u32 op, u64 seq)
struct ceph_msg *msg;
struct ceph_mds_session_head *h;
- msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h), GFP_NOFS);
+ msg = ceph_msg_new(CEPH_MSG_CLIENT_SESSION, sizeof(*h), GFP_NOFS,
+ false);
if (!msg) {
pr_err("create_session_msg ENOMEM creating msg\n");
return NULL;
@@ -742,35 +847,56 @@ static int __open_session(struct ceph_mds_client *mdsc,
*
* called under mdsc->mutex
*/
+static struct ceph_mds_session *
+__open_export_target_session(struct ceph_mds_client *mdsc, int target)
+{
+ struct ceph_mds_session *session;
+
+ session = __ceph_lookup_mds_session(mdsc, target);
+ if (!session) {
+ session = register_session(mdsc, target);
+ if (IS_ERR(session))
+ return session;
+ }
+ if (session->s_state == CEPH_MDS_SESSION_NEW ||
+ session->s_state == CEPH_MDS_SESSION_CLOSING)
+ __open_session(mdsc, session);
+
+ return session;
+}
+
+struct ceph_mds_session *
+ceph_mdsc_open_export_target_session(struct ceph_mds_client *mdsc, int target)
+{
+ struct ceph_mds_session *session;
+
+ dout("open_export_target_session to mds%d\n", target);
+
+ mutex_lock(&mdsc->mutex);
+ session = __open_export_target_session(mdsc, target);
+ mutex_unlock(&mdsc->mutex);
+
+ return session;
+}
+
static void __open_export_target_sessions(struct ceph_mds_client *mdsc,
struct ceph_mds_session *session)
{
struct ceph_mds_info *mi;
struct ceph_mds_session *ts;
int i, mds = session->s_mds;
- int target;
if (mds >= mdsc->mdsmap->m_max_mds)
return;
+
mi = &mdsc->mdsmap->m_info[mds];
dout("open_export_target_sessions for mds%d (%d targets)\n",
session->s_mds, mi->num_export_targets);
for (i = 0; i < mi->num_export_targets; i++) {
- target = mi->export_targets[i];
- ts = __ceph_lookup_mds_session(mdsc, target);
- if (!ts) {
- ts = register_session(mdsc, target);
- if (IS_ERR(ts))
- return;
- }
- if (session->s_state == CEPH_MDS_SESSION_NEW ||
- session->s_state == CEPH_MDS_SESSION_CLOSING)
- __open_session(mdsc, session);
- else
- dout(" mds%d target mds%d %p is %s\n", session->s_mds,
- i, ts, session_state_name(ts->s_state));
- ceph_put_mds_session(ts);
+ ts = __open_export_target_session(mdsc, mi->export_targets[i]);
+ if (!IS_ERR(ts))
+ ceph_put_mds_session(ts);
}
}
@@ -885,8 +1011,8 @@ static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap,
dout("removing cap %p, ci is %p, inode is %p\n",
cap, ci, &ci->vfs_inode);
- spin_lock(&inode->i_lock);
- __ceph_remove_cap(cap);
+ spin_lock(&ci->i_ceph_lock);
+ __ceph_remove_cap(cap, false);
if (!__ceph_is_any_real_caps(ci)) {
struct ceph_mds_client *mdsc =
ceph_sb_to_client(inode->i_sb)->mdsc;
@@ -918,7 +1044,7 @@ static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap,
}
spin_unlock(&mdsc->cap_dirty_lock);
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
while (drop--)
iput(inode);
return 0;
@@ -931,6 +1057,37 @@ static void remove_session_caps(struct ceph_mds_session *session)
{
dout("remove_session_caps on %p\n", session);
iterate_session_caps(session, remove_session_caps_cb, NULL);
+
+ spin_lock(&session->s_cap_lock);
+ if (session->s_nr_caps > 0) {
+ struct super_block *sb = session->s_mdsc->fsc->sb;
+ struct inode *inode;
+ struct ceph_cap *cap, *prev = NULL;
+ struct ceph_vino vino;
+ /*
+ * iterate_session_caps() skips inodes that are being
+ * deleted, we need to wait until deletions are complete.
+ * __wait_on_freeing_inode() is designed for the job,
+ * but it is not exported, so use lookup inode function
+ * to access it.
+ */
+ while (!list_empty(&session->s_caps)) {
+ cap = list_entry(session->s_caps.next,
+ struct ceph_cap, session_caps);
+ if (cap == prev)
+ break;
+ prev = cap;
+ vino = cap->ci->i_vino;
+ spin_unlock(&session->s_cap_lock);
+
+ inode = ceph_find_inode(sb, vino);
+ iput(inode);
+
+ spin_lock(&session->s_cap_lock);
+ }
+ }
+ spin_unlock(&session->s_cap_lock);
+
BUG_ON(session->s_nr_caps > 0);
BUG_ON(!list_empty(&session->s_cap_flushing));
cleanup_cap_releases(session);
@@ -949,10 +1106,10 @@ static int wake_up_session_cb(struct inode *inode, struct ceph_cap *cap,
wake_up_all(&ci->i_cap_wq);
if (arg) {
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
ci->i_wanted_max_size = 0;
ci->i_requested_max_size = 0;
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
}
return 0;
}
@@ -1001,6 +1158,21 @@ static int send_renew_caps(struct ceph_mds_client *mdsc,
return 0;
}
+static int send_flushmsg_ack(struct ceph_mds_client *mdsc,
+ struct ceph_mds_session *session, u64 seq)
+{
+ struct ceph_msg *msg;
+
+ dout("send_flushmsg_ack to mds%d (%s)s seq %lld\n",
+ session->s_mds, session_state_name(session->s_state), seq);
+ msg = create_session_msg(CEPH_SESSION_FLUSHMSG_ACK, seq);
+ if (!msg)
+ return -ENOMEM;
+ ceph_con_send(&session->s_con, msg);
+ return 0;
+}
+
+
/*
* Note new cap ttl, and any transition from stale -> not stale (fresh?).
*
@@ -1013,8 +1185,7 @@ static void renewed_caps(struct ceph_mds_client *mdsc,
int wake = 0;
spin_lock(&session->s_cap_lock);
- was_stale = is_renew && (session->s_cap_ttl == 0 ||
- time_after_eq(jiffies, session->s_cap_ttl));
+ was_stale = is_renew && time_after_eq(jiffies, session->s_cap_ttl);
session->s_cap_ttl = session->s_renew_requested +
mdsc->mdsmap->m_session_timeout*HZ;
@@ -1080,31 +1251,36 @@ static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg)
{
struct ceph_mds_session *session = arg;
struct ceph_inode_info *ci = ceph_inode(inode);
- int used, oissued, mine;
+ int used, wanted, oissued, mine;
if (session->s_trim_caps <= 0)
return -1;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
mine = cap->issued | cap->implemented;
used = __ceph_caps_used(ci);
+ wanted = __ceph_caps_file_wanted(ci);
oissued = __ceph_caps_issued_other(ci, cap);
- dout("trim_caps_cb %p cap %p mine %s oissued %s used %s\n",
+ dout("trim_caps_cb %p cap %p mine %s oissued %s used %s wanted %s\n",
inode, cap, ceph_cap_string(mine), ceph_cap_string(oissued),
- ceph_cap_string(used));
- if (ci->i_dirty_caps)
- goto out; /* dirty caps */
- if ((used & ~oissued) & mine)
+ ceph_cap_string(used), ceph_cap_string(wanted));
+ if (cap == ci->i_auth_cap) {
+ if (ci->i_dirty_caps | ci->i_flushing_caps)
+ goto out;
+ if ((used | wanted) & CEPH_CAP_ANY_WR)
+ goto out;
+ }
+ if ((used | wanted) & ~oissued & mine)
goto out; /* we need these caps */
session->s_trim_caps--;
if (oissued) {
/* we aren't the only cap.. just remove us */
- __ceph_remove_cap(cap);
+ __ceph_remove_cap(cap, true);
} else {
/* try to drop referring dentries */
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
d_prune_aliases(inode);
dout("trim_caps_cb %p cap %p pruned, count now %d\n",
inode, cap, atomic_read(&inode->i_count));
@@ -1112,7 +1288,7 @@ static int trim_caps_cb(struct inode *inode, struct ceph_cap *cap, void *arg)
}
out:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return 0;
}
@@ -1135,6 +1311,9 @@ static int trim_caps(struct ceph_mds_client *mdsc,
trim_caps - session->s_trim_caps);
session->s_trim_caps = 0;
}
+
+ ceph_add_cap_releases(mdsc, session);
+ ceph_send_cap_releases(mdsc, session);
return 0;
}
@@ -1175,7 +1354,7 @@ int ceph_add_cap_releases(struct ceph_mds_client *mdsc,
while (session->s_num_cap_releases < session->s_nr_caps + extra) {
spin_unlock(&session->s_cap_lock);
msg = ceph_msg_new(CEPH_MSG_CLIENT_CAPRELEASE, PAGE_CACHE_SIZE,
- GFP_NOFS);
+ GFP_NOFS, false);
if (!msg)
goto out_unlocked;
dout("add_cap_releases %p msg %p now %d\n", session, msg,
@@ -1230,7 +1409,7 @@ static int check_cap_flush(struct ceph_mds_client *mdsc, u64 want_flush_seq)
i_flushing_item);
struct inode *inode = &ci->vfs_inode;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (ci->i_cap_flush_seq <= want_flush_seq) {
dout("check_cap_flush still flushing %p "
"seq %lld <= %lld to mds%d\n", inode,
@@ -1238,7 +1417,7 @@ static int check_cap_flush(struct ceph_mds_client *mdsc, u64 want_flush_seq)
session->s_mds);
ret = 0;
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
}
mutex_unlock(&session->s_mutex);
ceph_put_mds_session(session);
@@ -1284,16 +1463,19 @@ static void discard_cap_releases(struct ceph_mds_client *mdsc,
unsigned num;
dout("discard_cap_releases mds%d\n", session->s_mds);
- spin_lock(&session->s_cap_lock);
- /* zero out the in-progress message */
- msg = list_first_entry(&session->s_cap_releases,
- struct ceph_msg, list_head);
- head = msg->front.iov_base;
- num = le32_to_cpu(head->num);
- dout("discard_cap_releases mds%d %p %u\n", session->s_mds, msg, num);
- head->num = cpu_to_le32(0);
- session->s_num_cap_releases += num;
+ if (!list_empty(&session->s_cap_releases)) {
+ /* zero out the in-progress message */
+ msg = list_first_entry(&session->s_cap_releases,
+ struct ceph_msg, list_head);
+ head = msg->front.iov_base;
+ num = le32_to_cpu(head->num);
+ dout("discard_cap_releases mds%d %p %u\n",
+ session->s_mds, msg, num);
+ head->num = cpu_to_le32(0);
+ msg->front.iov_len = sizeof(*head);
+ session->s_num_cap_releases += num;
+ }
/* requeue completed messages */
while (!list_empty(&session->s_cap_releases_done)) {
@@ -1310,14 +1492,49 @@ static void discard_cap_releases(struct ceph_mds_client *mdsc,
msg->front.iov_len = sizeof(*head);
list_add(&msg->list_head, &session->s_cap_releases);
}
-
- spin_unlock(&session->s_cap_lock);
}
/*
* requests
*/
+int ceph_alloc_readdir_reply_buffer(struct ceph_mds_request *req,
+ struct inode *dir)
+{
+ struct ceph_inode_info *ci = ceph_inode(dir);
+ struct ceph_mds_reply_info_parsed *rinfo = &req->r_reply_info;
+ struct ceph_mount_options *opt = req->r_mdsc->fsc->mount_options;
+ size_t size = sizeof(*rinfo->dir_in) + sizeof(*rinfo->dir_dname_len) +
+ sizeof(*rinfo->dir_dname) + sizeof(*rinfo->dir_dlease);
+ int order, num_entries;
+
+ spin_lock(&ci->i_ceph_lock);
+ num_entries = ci->i_files + ci->i_subdirs;
+ spin_unlock(&ci->i_ceph_lock);
+ num_entries = max(num_entries, 1);
+ num_entries = min(num_entries, opt->max_readdir);
+
+ order = get_order(size * num_entries);
+ while (order >= 0) {
+ rinfo->dir_in = (void*)__get_free_pages(GFP_NOFS | __GFP_NOWARN,
+ order);
+ if (rinfo->dir_in)
+ break;
+ order--;
+ }
+ if (!rinfo->dir_in)
+ return -ENOMEM;
+
+ num_entries = (PAGE_SIZE << order) / size;
+ num_entries = min(num_entries, opt->max_readdir);
+
+ rinfo->dir_buf_size = PAGE_SIZE << order;
+ req->r_num_caps = num_entries + 1;
+ req->r_args.readdir.max_entries = cpu_to_le32(num_entries);
+ req->r_args.readdir.max_bytes = cpu_to_le32(opt->max_readdir_bytes);
+ return 0;
+}
+
/*
* Create an mds request.
*/
@@ -1341,6 +1558,8 @@ ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode)
init_completion(&req->r_safe_completion);
INIT_LIST_HEAD(&req->r_unsafe_item);
+ req->r_stamp = CURRENT_TIME;
+
req->r_op = op;
req->r_direct_mode = mode;
return req;
@@ -1384,12 +1603,15 @@ char *ceph_mdsc_build_path(struct dentry *dentry, int *plen, u64 *base,
struct dentry *temp;
char *path;
int len, pos;
+ unsigned seq;
if (dentry == NULL)
return ERR_PTR(-EINVAL);
retry:
len = 0;
+ seq = read_seqbegin(&rename_lock);
+ rcu_read_lock();
for (temp = dentry; !IS_ROOT(temp);) {
struct inode *inode = temp->d_inode;
if (inode && ceph_snap(inode) == CEPH_SNAPDIR)
@@ -1400,11 +1622,8 @@ retry:
else
len += 1 + temp->d_name.len;
temp = temp->d_parent;
- if (temp == NULL) {
- pr_err("build_path corrupt dentry %p\n", dentry);
- return ERR_PTR(-EINVAL);
- }
}
+ rcu_read_unlock();
if (len)
len--; /* no leading '/' */
@@ -1413,32 +1632,35 @@ retry:
return ERR_PTR(-ENOMEM);
pos = len;
path[pos] = 0; /* trailing null */
+ rcu_read_lock();
for (temp = dentry; !IS_ROOT(temp) && pos != 0; ) {
- struct inode *inode = temp->d_inode;
+ struct inode *inode;
+ spin_lock(&temp->d_lock);
+ inode = temp->d_inode;
if (inode && ceph_snap(inode) == CEPH_SNAPDIR) {
dout("build_path path+%d: %p SNAPDIR\n",
pos, temp);
} else if (stop_on_nosnap && inode &&
ceph_snap(inode) == CEPH_NOSNAP) {
+ spin_unlock(&temp->d_lock);
break;
} else {
pos -= temp->d_name.len;
- if (pos < 0)
+ if (pos < 0) {
+ spin_unlock(&temp->d_lock);
break;
+ }
strncpy(path + pos, temp->d_name.name,
temp->d_name.len);
}
+ spin_unlock(&temp->d_lock);
if (pos)
path[--pos] = '/';
temp = temp->d_parent;
- if (temp == NULL) {
- pr_err("build_path corrupt dentry\n");
- kfree(path);
- return ERR_PTR(-EINVAL);
- }
}
- if (pos != 0) {
+ rcu_read_unlock();
+ if (pos != 0 || read_seqretry(&rename_lock, seq)) {
pr_err("build_path did not end path lookup where "
"expected, namelen is %d, pos is %d\n", len, pos);
/* presumably this is only possible if racing with a
@@ -1452,7 +1674,7 @@ retry:
*base = ceph_ino(temp->d_inode);
*plen = len;
dout("build_path on %p %d built %llx '%.*s'\n",
- dentry, atomic_read(&dentry->d_count), *base, len, path);
+ dentry, d_count(dentry), *base, len, path);
return path;
}
@@ -1517,10 +1739,10 @@ static int set_request_path_attr(struct inode *rinode, struct dentry *rdentry,
r = build_dentry_path(rdentry, ppath, pathlen, ino, freepath);
dout(" dentry %p %llx/%.*s\n", rdentry, *ino, *pathlen,
*ppath);
- } else if (rpath) {
+ } else if (rpath || rino) {
*ino = rino;
*ppath = rpath;
- *pathlen = strlen(rpath);
+ *pathlen = rpath ? strlen(rpath) : 0;
dout(" path %.*s\n", *pathlen, rpath);
}
@@ -1563,7 +1785,8 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
}
len = sizeof(*head) +
- pathlen1 + pathlen2 + 2*(1 + sizeof(u32) + sizeof(u64));
+ pathlen1 + pathlen2 + 2*(1 + sizeof(u32) + sizeof(u64)) +
+ sizeof(struct timespec);
/* calculate (max) length for cap releases */
len += sizeof(struct ceph_mds_request_release) *
@@ -1574,12 +1797,13 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
if (req->r_old_dentry_drop)
len += req->r_old_dentry->d_name.len;
- msg = ceph_msg_new(CEPH_MSG_CLIENT_REQUEST, len, GFP_NOFS);
+ msg = ceph_msg_new(CEPH_MSG_CLIENT_REQUEST, len, GFP_NOFS, false);
if (!msg) {
msg = ERR_PTR(-ENOMEM);
goto out_free2;
}
+ msg->hdr.version = 2;
msg->hdr.tid = cpu_to_le64(req->r_tid);
head = msg->front.iov_base;
@@ -1588,8 +1812,8 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
head->mdsmap_epoch = cpu_to_le32(mdsc->mdsmap->m_epoch);
head->op = cpu_to_le32(req->r_op);
- head->caller_uid = cpu_to_le32(current_fsuid());
- head->caller_gid = cpu_to_le32(current_fsgid());
+ head->caller_uid = cpu_to_le32(from_kuid(&init_user_ns, req->r_uid));
+ head->caller_gid = cpu_to_le32(from_kgid(&init_user_ns, req->r_gid));
head->args = req->r_args;
ceph_encode_filepath(&p, end, ino1, path1);
@@ -1616,12 +1840,19 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
mds, req->r_old_inode_drop, req->r_old_inode_unless, 0);
head->num_releases = cpu_to_le16(releases);
+ /* time stamp */
+ ceph_encode_copy(&p, &req->r_stamp, sizeof(req->r_stamp));
+
BUG_ON(p > end);
msg->front.iov_len = p - msg->front.iov_base;
msg->hdr.front_len = cpu_to_le32(msg->front.iov_len);
- msg->pages = req->r_pages;
- msg->nr_pages = req->r_num_pages;
+ if (req->r_data_len) {
+ /* outbound data set only by ceph_sync_setxattr() */
+ BUG_ON(!req->r_pages);
+ ceph_msg_data_add_pages(msg, req->r_pages, req->r_data_len, 0);
+ }
+
msg->hdr.data_len = cpu_to_le32(req->r_data_len);
msg->hdr.data_off = cpu_to_le16(0);
@@ -1659,7 +1890,6 @@ static int __prepare_send_request(struct ceph_mds_client *mdsc,
struct ceph_msg *msg;
int flags = 0;
- req->r_mds = mds;
req->r_attempts++;
if (req->r_inode) {
struct ceph_cap *cap =
@@ -1736,8 +1966,11 @@ static int __do_request(struct ceph_mds_client *mdsc,
int mds = -1;
int err = -EAGAIN;
- if (req->r_err || req->r_got_result)
+ if (req->r_err || req->r_got_result) {
+ if (req->r_aborted)
+ __unregister_request(mdsc, req);
goto out;
+ }
if (req->r_timeout &&
time_after_eq(jiffies, req->r_started + req->r_timeout)) {
@@ -1746,6 +1979,8 @@ static int __do_request(struct ceph_mds_client *mdsc,
goto finish;
}
+ put_request_session(req);
+
mds = __choose_mds(mdsc, req);
if (mds < 0 ||
ceph_mdsmap_get_state(mdsc->mdsmap, mds) < CEPH_MDS_STATE_ACTIVE) {
@@ -1763,6 +1998,8 @@ static int __do_request(struct ceph_mds_client *mdsc,
goto finish;
}
}
+ req->r_session = get_session(session);
+
dout("do_request mds%d session %p state %s\n", mds, session,
session_state_name(session->s_state));
if (session->s_state != CEPH_MDS_SESSION_OPEN &&
@@ -1775,7 +2012,6 @@ static int __do_request(struct ceph_mds_client *mdsc,
}
/* send request */
- req->r_session = get_session(session);
req->r_resend_mds = -1; /* forget any previous mds hint */
if (req->r_request_started == 0) /* note request start time */
@@ -1804,10 +2040,16 @@ finish:
static void __wake_requests(struct ceph_mds_client *mdsc,
struct list_head *head)
{
- struct ceph_mds_request *req, *nreq;
+ struct ceph_mds_request *req;
+ LIST_HEAD(tmp_list);
+
+ list_splice_init(head, &tmp_list);
- list_for_each_entry_safe(req, nreq, head, r_wait) {
+ while (!list_empty(&tmp_list)) {
+ req = list_entry(tmp_list.next,
+ struct ceph_mds_request, r_wait);
list_del_init(&req->r_wait);
+ dout(" wake request %p tid %llu\n", req, req->r_tid);
__do_request(mdsc, req);
}
}
@@ -1829,7 +2071,6 @@ static void kick_requests(struct ceph_mds_client *mdsc, int mds)
if (req->r_session &&
req->r_session->s_mds == mds) {
dout(" kicking tid %llu\n", req->r_tid);
- put_request_session(req);
__do_request(mdsc, req);
}
}
@@ -1862,10 +2103,9 @@ int ceph_mdsc_do_request(struct ceph_mds_client *mdsc,
ceph_get_cap_refs(ceph_inode(req->r_inode), CEPH_CAP_PIN);
if (req->r_locked_dir)
ceph_get_cap_refs(ceph_inode(req->r_locked_dir), CEPH_CAP_PIN);
- if (req->r_old_dentry)
- ceph_get_cap_refs(
- ceph_inode(req->r_old_dentry->d_parent->d_inode),
- CEPH_CAP_PIN);
+ if (req->r_old_dentry_dir)
+ ceph_get_cap_refs(ceph_inode(req->r_old_dentry_dir),
+ CEPH_CAP_PIN);
/* issue */
mutex_lock(&mdsc->mutex);
@@ -1923,20 +2163,16 @@ out:
}
/*
- * Invalidate dir I_COMPLETE, dentry lease state on an aborted MDS
+ * Invalidate dir's completeness, dentry lease state on an aborted MDS
* namespace request.
*/
void ceph_invalidate_dir_request(struct ceph_mds_request *req)
{
struct inode *inode = req->r_locked_dir;
- struct ceph_inode_info *ci = ceph_inode(inode);
- dout("invalidate_dir_request %p (I_COMPLETE, lease(s))\n", inode);
- spin_lock(&inode->i_lock);
- ci->i_ceph_flags &= ~CEPH_I_COMPLETE;
- ci->i_release_count++;
- spin_unlock(&inode->i_lock);
+ dout("invalidate_dir_request %p (complete, lease(s))\n", inode);
+ ceph_dir_clear_complete(inode);
if (req->r_dentry)
ceph_invalidate_dentry_lease(req->r_dentry);
if (req->r_old_dentry)
@@ -1989,13 +2225,13 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
/* dup? */
if ((req->r_got_unsafe && !head->safe) ||
(req->r_got_safe && head->safe)) {
- pr_warning("got a dup %s reply on %llu from mds%d\n",
+ pr_warn("got a dup %s reply on %llu from mds%d\n",
head->safe ? "safe" : "unsafe", tid, mds);
mutex_unlock(&mdsc->mutex);
goto out;
}
if (req->r_got_safe && !head->safe) {
- pr_warning("got unsafe after safe on %llu from mds%d\n",
+ pr_warn("got unsafe after safe on %llu from mds%d\n",
tid, mds);
mutex_unlock(&mdsc->mutex);
goto out;
@@ -2012,23 +2248,16 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
*/
if (result == -ESTALE) {
dout("got ESTALE on request %llu", req->r_tid);
- if (!req->r_inode) {
- /* do nothing; not an authority problem */
- } else if (req->r_direct_mode != USE_AUTH_MDS) {
+ if (req->r_direct_mode != USE_AUTH_MDS) {
dout("not using auth, setting for that now");
req->r_direct_mode = USE_AUTH_MDS;
__do_request(mdsc, req);
mutex_unlock(&mdsc->mutex);
goto out;
} else {
- struct ceph_inode_info *ci = ceph_inode(req->r_inode);
- struct ceph_cap *cap =
- ceph_get_cap_for_mds(ci, req->r_mds);;
-
- dout("already using auth");
- if ((!cap || cap != ci->i_auth_cap) ||
- (cap->mseq != req->r_sent_on_mseq)) {
- dout("but cap changed, so resending");
+ int mds = __choose_mds(mdsc, req);
+ if (mds >= 0 && mds != req->r_session->s_mds) {
+ dout("but auth changed, so resending");
__do_request(mdsc, req);
mutex_unlock(&mdsc->mutex);
goto out;
@@ -2041,7 +2270,6 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
if (head->safe) {
req->r_got_safe = true;
__unregister_request(mdsc, req);
- complete_all(&req->r_safe_completion);
if (req->r_got_unsafe) {
/*
@@ -2067,12 +2295,12 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
dout("handle_reply tid %lld result %d\n", tid, result);
rinfo = &req->r_reply_info;
- err = parse_reply_info(msg, rinfo);
+ err = parse_reply_info(msg, rinfo, session->s_con.peer_features);
mutex_unlock(&mdsc->mutex);
mutex_lock(&session->s_mutex);
if (err < 0) {
- pr_err("mdsc_handle_reply got corrupt reply mds%d\n", mds);
+ pr_err("mdsc_handle_reply got corrupt reply mds%d(tid:%lld)\n", mds, tid);
ceph_msg_dump(msg);
goto out_err;
}
@@ -2092,7 +2320,8 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
mutex_lock(&req->r_fill_mutex);
err = ceph_fill_trace(mdsc->fsc->sb, req, req->r_session);
if (err == 0) {
- if (result == 0 && rinfo->dir_nr)
+ if (result == 0 && (req->r_op == CEPH_MDS_OP_READDIR ||
+ req->r_op == CEPH_MDS_OP_LSSNAP))
ceph_readdir_prepopulate(req, req->r_session);
ceph_unreserve_caps(mdsc, &req->r_caps_reservation);
}
@@ -2242,10 +2471,10 @@ static void handle_session(struct ceph_mds_session *session,
case CEPH_SESSION_STALE:
pr_info("mds%d caps went stale, renewing\n",
session->s_mds);
- spin_lock(&session->s_cap_lock);
+ spin_lock(&session->s_gen_ttl_lock);
session->s_cap_gen++;
- session->s_cap_ttl = 0;
- spin_unlock(&session->s_cap_lock);
+ session->s_cap_ttl = jiffies - 1;
+ spin_unlock(&session->s_gen_ttl_lock);
send_renew_caps(mdsc, session);
break;
@@ -2253,6 +2482,10 @@ static void handle_session(struct ceph_mds_session *session,
trim_caps(mdsc, session, le32_to_cpu(h->max_caps));
break;
+ case CEPH_SESSION_FLUSHMSG:
+ send_flushmsg_ack(mdsc, session, seq);
+ break;
+
default:
pr_err("mdsc_handle_session bad op %d mds%d\n", op, mds);
WARN_ON(1);
@@ -2339,9 +2572,11 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
if (err)
goto out_free;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
cap->seq = 0; /* reset cap seq */
cap->issue_seq = 0; /* and issue_seq */
+ cap->mseq = 0; /* and migrate_seq */
+ cap->cap_gen = cap->session->s_cap_gen;
if (recon_state->flock) {
rec.v2.cap_id = cpu_to_le64(cap->cap_id);
@@ -2362,43 +2597,50 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
rec.v1.pathbase = cpu_to_le64(pathbase);
reclen = sizeof(rec.v1);
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
if (recon_state->flock) {
int num_fcntl_locks, num_flock_locks;
- struct ceph_pagelist_cursor trunc_point;
-
- ceph_pagelist_set_cursor(pagelist, &trunc_point);
- do {
- lock_flocks();
- ceph_count_locks(inode, &num_fcntl_locks,
- &num_flock_locks);
- rec.v2.flock_len = (2*sizeof(u32) +
- (num_fcntl_locks+num_flock_locks) *
- sizeof(struct ceph_filelock));
- unlock_flocks();
-
- /* pre-alloc pagelist */
- ceph_pagelist_truncate(pagelist, &trunc_point);
- err = ceph_pagelist_append(pagelist, &rec, reclen);
- if (!err)
- err = ceph_pagelist_reserve(pagelist,
- rec.v2.flock_len);
-
- /* encode locks */
- if (!err) {
- lock_flocks();
- err = ceph_encode_locks(inode,
- pagelist,
- num_fcntl_locks,
- num_flock_locks);
- unlock_flocks();
- }
- } while (err == -ENOSPC);
+ struct ceph_filelock *flocks;
+
+encode_again:
+ spin_lock(&inode->i_lock);
+ ceph_count_locks(inode, &num_fcntl_locks, &num_flock_locks);
+ spin_unlock(&inode->i_lock);
+ flocks = kmalloc((num_fcntl_locks+num_flock_locks) *
+ sizeof(struct ceph_filelock), GFP_NOFS);
+ if (!flocks) {
+ err = -ENOMEM;
+ goto out_free;
+ }
+ spin_lock(&inode->i_lock);
+ err = ceph_encode_locks_to_buffer(inode, flocks,
+ num_fcntl_locks,
+ num_flock_locks);
+ spin_unlock(&inode->i_lock);
+ if (err) {
+ kfree(flocks);
+ if (err == -ENOSPC)
+ goto encode_again;
+ goto out_free;
+ }
+ /*
+ * number of encoded locks is stable, so copy to pagelist
+ */
+ rec.v2.flock_len = cpu_to_le32(2*sizeof(u32) +
+ (num_fcntl_locks+num_flock_locks) *
+ sizeof(struct ceph_filelock));
+ err = ceph_pagelist_append(pagelist, &rec, reclen);
+ if (!err)
+ err = ceph_locks_to_pagelist(flocks, pagelist,
+ num_fcntl_locks,
+ num_flock_locks);
+ kfree(flocks);
} else {
err = ceph_pagelist_append(pagelist, &rec, reclen);
}
+ recon_state->nr_caps++;
out_free:
kfree(path);
out_dput:
@@ -2426,6 +2668,7 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
struct rb_node *p;
int mds = session->s_mds;
int err = -ENOMEM;
+ int s_nr_caps;
struct ceph_pagelist *pagelist;
struct ceph_reconnect_state recon_state;
@@ -2436,7 +2679,7 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
goto fail_nopagelist;
ceph_pagelist_init(pagelist);
- reply = ceph_msg_new(CEPH_MSG_CLIENT_RECONNECT, 0, GFP_NOFS);
+ reply = ceph_msg_new(CEPH_MSG_CLIENT_RECONNECT, 0, GFP_NOFS, false);
if (!reply)
goto fail_nomsg;
@@ -2444,7 +2687,9 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
session->s_state = CEPH_MDS_SESSION_RECONNECTING;
session->s_seq = 0;
+ ceph_con_close(&session->s_con);
ceph_con_open(&session->s_con,
+ CEPH_ENTITY_TYPE_MDS, mds,
ceph_mdsmap_get_addr(mdsc->mdsmap, mds));
/* replay unsafe requests */
@@ -2455,20 +2700,38 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
dout("session %p state %s\n", session,
session_state_name(session->s_state));
+ spin_lock(&session->s_gen_ttl_lock);
+ session->s_cap_gen++;
+ spin_unlock(&session->s_gen_ttl_lock);
+
+ spin_lock(&session->s_cap_lock);
+ /*
+ * notify __ceph_remove_cap() that we are composing cap reconnect.
+ * If a cap get released before being added to the cap reconnect,
+ * __ceph_remove_cap() should skip queuing cap release.
+ */
+ session->s_cap_reconnect = 1;
/* drop old cap expires; we're about to reestablish that state */
discard_cap_releases(mdsc, session);
+ spin_unlock(&session->s_cap_lock);
/* traverse this session's caps */
- err = ceph_pagelist_encode_32(pagelist, session->s_nr_caps);
+ s_nr_caps = session->s_nr_caps;
+ err = ceph_pagelist_encode_32(pagelist, s_nr_caps);
if (err)
goto fail;
+ recon_state.nr_caps = 0;
recon_state.pagelist = pagelist;
recon_state.flock = session->s_con.peer_features & CEPH_FEATURE_FLOCK;
err = iterate_session_caps(session, encode_caps_cb, &recon_state);
if (err < 0)
goto fail;
+ spin_lock(&session->s_cap_lock);
+ session->s_cap_reconnect = 0;
+ spin_unlock(&session->s_cap_lock);
+
/*
* snaprealms. we provide mds with the ino, seq (version), and
* parent for all of our realms. If the mds has any newer info,
@@ -2489,11 +2752,20 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
goto fail;
}
- reply->pagelist = pagelist;
if (recon_state.flock)
reply->hdr.version = cpu_to_le16(2);
+
+ /* raced with cap release? */
+ if (s_nr_caps != recon_state.nr_caps) {
+ struct page *page = list_first_entry(&pagelist->head,
+ struct page, lru);
+ __le32 *addr = kmap_atomic(page);
+ *addr = cpu_to_le32(recon_state.nr_caps);
+ kunmap_atomic(addr);
+ }
+
reply->hdr.data_len = cpu_to_le32(pagelist->length);
- reply->nr_pages = calc_pages_for(0, pagelist->length);
+ ceph_msg_data_add_pagelist(reply, pagelist);
ceph_con_send(&session->s_con, reply);
mutex_unlock(&session->s_mutex);
@@ -2549,7 +2821,8 @@ static void check_new_map(struct ceph_mds_client *mdsc,
ceph_mdsmap_is_laggy(newmap, i) ? " (laggy)" : "",
session_state_name(s->s_state));
- if (memcmp(ceph_mdsmap_get_addr(oldmap, i),
+ if (i >= newmap->m_max_mds ||
+ memcmp(ceph_mdsmap_get_addr(oldmap, i),
ceph_mdsmap_get_addr(newmap, i),
sizeof(struct ceph_entity_addr))) {
if (s->s_state == CEPH_MDS_SESSION_OPENING) {
@@ -2636,14 +2909,12 @@ static void handle_lease(struct ceph_mds_client *mdsc,
{
struct super_block *sb = mdsc->fsc->sb;
struct inode *inode;
- struct ceph_inode_info *ci;
struct dentry *parent, *dentry;
struct ceph_dentry_info *di;
int mds = session->s_mds;
struct ceph_mds_lease *h = msg->front.iov_base;
u32 seq;
struct ceph_vino vino;
- int mask;
struct qstr dname;
int release = 0;
@@ -2654,7 +2925,6 @@ static void handle_lease(struct ceph_mds_client *mdsc,
goto bad;
vino.ino = le64_to_cpu(h->ino);
vino.snap = CEPH_NOSNAP;
- mask = le16_to_cpu(h->mask);
seq = le32_to_cpu(h->seq);
dname.name = (void *)h + sizeof(*h) + sizeof(u32);
dname.len = msg->front.iov_len - sizeof(*h) - sizeof(u32);
@@ -2666,14 +2936,13 @@ static void handle_lease(struct ceph_mds_client *mdsc,
/* lookup inode */
inode = ceph_find_inode(sb, vino);
- dout("handle_lease %s, mask %d, ino %llx %p %.*s\n",
- ceph_lease_op_name(h->action), mask, vino.ino, inode,
+ dout("handle_lease %s, ino %llx %p %.*s\n",
+ ceph_lease_op_name(h->action), vino.ino, inode,
dname.len, dname.name);
if (inode == NULL) {
dout("handle_lease no inode %llx\n", vino.ino);
goto release;
}
- ci = ceph_inode(inode);
/* dentry */
parent = d_find_alias(inode);
@@ -2692,7 +2961,7 @@ static void handle_lease(struct ceph_mds_client *mdsc,
di = ceph_dentry(dentry);
switch (h->action) {
case CEPH_MDS_LEASE_REVOKE:
- if (di && di->lease_session == session) {
+ if (di->lease_session == session) {
if (ceph_seq_cmp(di->lease_seq, seq) > 0)
h->seq = cpu_to_le32(di->lease_seq);
__ceph_mdsc_drop_dentry_lease(dentry);
@@ -2701,7 +2970,7 @@ static void handle_lease(struct ceph_mds_client *mdsc,
break;
case CEPH_MDS_LEASE_RENEW:
- if (di && di->lease_session == session &&
+ if (di->lease_session == session &&
di->lease_gen == session->s_cap_gen &&
di->lease_renew_from &&
di->lease_renew_after == 0) {
@@ -2753,12 +3022,11 @@ void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session,
dnamelen = dentry->d_name.len;
len += dnamelen;
- msg = ceph_msg_new(CEPH_MSG_CLIENT_LEASE, len, GFP_NOFS);
+ msg = ceph_msg_new(CEPH_MSG_CLIENT_LEASE, len, GFP_NOFS, false);
if (!msg)
return;
lease = msg->front.iov_base;
lease->action = action;
- lease->mask = cpu_to_le16(1);
lease->ino = cpu_to_le64(ceph_vino(inode).ino);
lease->first = lease->last = cpu_to_le64(ceph_vino(inode).snap);
lease->seq = cpu_to_le32(seq);
@@ -2780,7 +3048,7 @@ void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session,
* Pass @inode always, @dentry is optional.
*/
void ceph_mdsc_lease_release(struct ceph_mds_client *mdsc, struct inode *inode,
- struct dentry *dentry, int mask)
+ struct dentry *dentry)
{
struct ceph_dentry_info *di;
struct ceph_mds_session *session;
@@ -2788,7 +3056,6 @@ void ceph_mdsc_lease_release(struct ceph_mds_client *mdsc, struct inode *inode,
BUG_ON(inode == NULL);
BUG_ON(dentry == NULL);
- BUG_ON(mask == 0);
/* is dentry lease valid? */
spin_lock(&dentry->d_lock);
@@ -2798,8 +3065,8 @@ void ceph_mdsc_lease_release(struct ceph_mds_client *mdsc, struct inode *inode,
di->lease_gen != di->lease_session->s_cap_gen ||
!time_before(jiffies, dentry->d_time)) {
dout("lease_release inode %p dentry %p -- "
- "no lease on %d\n",
- inode, dentry, mask);
+ "no lease\n",
+ inode, dentry);
spin_unlock(&dentry->d_lock);
return;
}
@@ -2810,8 +3077,8 @@ void ceph_mdsc_lease_release(struct ceph_mds_client *mdsc, struct inode *inode,
__ceph_mdsc_drop_dentry_lease(dentry);
spin_unlock(&dentry->d_lock);
- dout("lease_release inode %p dentry %p mask %d to mds%d\n",
- inode, dentry, mask, session->s_mds);
+ dout("lease_release inode %p dentry %p to mds%d\n",
+ inode, dentry, session->s_mds);
ceph_mdsc_lease_send_msg(session, inode, dentry,
CEPH_MDS_LEASE_RELEASE, seq);
ceph_put_mds_session(session);
@@ -2924,8 +3191,10 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc)
fsc->mdsc = mdsc;
mutex_init(&mdsc->mutex);
mdsc->mdsmap = kzalloc(sizeof(*mdsc->mdsmap), GFP_NOFS);
- if (mdsc->mdsmap == NULL)
+ if (mdsc->mdsmap == NULL) {
+ kfree(mdsc);
return -ENOMEM;
+ }
init_completion(&mdsc->safe_umount_waiters);
init_waitqueue_head(&mdsc->session_close_wq);
@@ -2947,6 +3216,7 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc)
spin_lock_init(&mdsc->snap_flush_lock);
mdsc->cap_flush_seq = 0;
INIT_LIST_HEAD(&mdsc->cap_dirty);
+ INIT_LIST_HEAD(&mdsc->cap_dirty_migrating);
mdsc->num_cap_flushing = 0;
spin_lock_init(&mdsc->cap_dirty_lock);
init_waitqueue_head(&mdsc->cap_flushing_wq);
@@ -3076,7 +3346,7 @@ void ceph_mdsc_sync(struct ceph_mds_client *mdsc)
/*
* true if all sessions are closed, or we force unmount
*/
-bool done_closing_sessions(struct ceph_mds_client *mdsc)
+static bool done_closing_sessions(struct ceph_mds_client *mdsc)
{
int i, n = 0;
@@ -3160,9 +3430,15 @@ void ceph_mdsc_destroy(struct ceph_fs_client *fsc)
{
struct ceph_mds_client *mdsc = fsc->mdsc;
+ dout("mdsc_destroy %p\n", mdsc);
ceph_mdsc_stop(mdsc);
+
+ /* flush out any connection work with references to us */
+ ceph_msgr_flush();
+
fsc->mdsc = NULL;
kfree(mdsc);
+ dout("mdsc_destroy %p done\n", mdsc);
}
@@ -3243,8 +3519,8 @@ static void con_put(struct ceph_connection *con)
{
struct ceph_mds_session *s = con->private;
+ dout("mdsc con_put %p (%d)\n", s, atomic_read(&s->s_ref) - 1);
ceph_put_mds_session(s);
- dout("mdsc con_put %p (%d)\n", s, atomic_read(&s->s_ref));
}
/*
@@ -3256,7 +3532,7 @@ static void peer_reset(struct ceph_connection *con)
struct ceph_mds_session *s = con->private;
struct ceph_mds_client *mdsc = s->s_mdsc;
- pr_warning("mds%d closed our session\n", s->s_mds);
+ pr_warn("mds%d closed our session\n", s->s_mds);
send_mds_reconnect(mdsc, s);
}
@@ -3307,39 +3583,37 @@ out:
/*
* authentication
*/
-static int get_authorizer(struct ceph_connection *con,
- void **buf, int *len, int *proto,
- void **reply_buf, int *reply_len, int force_new)
+
+/*
+ * Note: returned pointer is the address of a structure that's
+ * managed separately. Caller must *not* attempt to free it.
+ */
+static struct ceph_auth_handshake *get_authorizer(struct ceph_connection *con,
+ int *proto, int force_new)
{
struct ceph_mds_session *s = con->private;
struct ceph_mds_client *mdsc = s->s_mdsc;
struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
- int ret = 0;
-
- if (force_new && s->s_authorizer) {
- ac->ops->destroy_authorizer(ac, s->s_authorizer);
- s->s_authorizer = NULL;
- }
- if (s->s_authorizer == NULL) {
- if (ac->ops->create_authorizer) {
- ret = ac->ops->create_authorizer(
- ac, CEPH_ENTITY_TYPE_MDS,
- &s->s_authorizer,
- &s->s_authorizer_buf,
- &s->s_authorizer_buf_len,
- &s->s_authorizer_reply_buf,
- &s->s_authorizer_reply_buf_len);
- if (ret)
- return ret;
- }
- }
+ struct ceph_auth_handshake *auth = &s->s_auth;
+ if (force_new && auth->authorizer) {
+ ceph_auth_destroy_authorizer(ac, auth->authorizer);
+ auth->authorizer = NULL;
+ }
+ if (!auth->authorizer) {
+ int ret = ceph_auth_create_authorizer(ac, CEPH_ENTITY_TYPE_MDS,
+ auth);
+ if (ret)
+ return ERR_PTR(ret);
+ } else {
+ int ret = ceph_auth_update_authorizer(ac, CEPH_ENTITY_TYPE_MDS,
+ auth);
+ if (ret)
+ return ERR_PTR(ret);
+ }
*proto = ac->protocol;
- *buf = s->s_authorizer_buf;
- *len = s->s_authorizer_buf_len;
- *reply_buf = s->s_authorizer_reply_buf;
- *reply_len = s->s_authorizer_reply_buf_len;
- return 0;
+
+ return auth;
}
@@ -3349,7 +3623,7 @@ static int verify_authorizer_reply(struct ceph_connection *con, int len)
struct ceph_mds_client *mdsc = s->s_mdsc;
struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
- return ac->ops->verify_authorizer_reply(ac, s->s_authorizer, len);
+ return ceph_auth_verify_authorizer_reply(ac, s->s_auth.authorizer, len);
}
static int invalidate_authorizer(struct ceph_connection *con)
@@ -3358,12 +3632,32 @@ static int invalidate_authorizer(struct ceph_connection *con)
struct ceph_mds_client *mdsc = s->s_mdsc;
struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
- if (ac->ops->invalidate_authorizer)
- ac->ops->invalidate_authorizer(ac, CEPH_ENTITY_TYPE_MDS);
+ ceph_auth_invalidate_authorizer(ac, CEPH_ENTITY_TYPE_MDS);
return ceph_monc_validate_auth(&mdsc->fsc->client->monc);
}
+static struct ceph_msg *mds_alloc_msg(struct ceph_connection *con,
+ struct ceph_msg_header *hdr, int *skip)
+{
+ struct ceph_msg *msg;
+ int type = (int) le16_to_cpu(hdr->type);
+ int front_len = (int) le32_to_cpu(hdr->front_len);
+
+ if (con->in_msg)
+ return con->in_msg;
+
+ *skip = 0;
+ msg = ceph_msg_new(type, front_len, GFP_NOFS, false);
+ if (!msg) {
+ pr_err("unable to allocate msg type %d len %d\n",
+ type, front_len);
+ return NULL;
+ }
+
+ return msg;
+}
+
static const struct ceph_connection_operations mds_con_ops = {
.get = con_get,
.put = con_put,
@@ -3372,6 +3666,7 @@ static const struct ceph_connection_operations mds_con_ops = {
.verify_authorizer_reply = verify_authorizer_reply,
.invalidate_authorizer = invalidate_authorizer,
.peer_reset = peer_reset,
+ .alloc_msg = mds_alloc_msg,
};
/* eof */
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index d66d63c7235..e00737cf523 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -11,6 +11,7 @@
#include <linux/ceph/types.h>
#include <linux/ceph/messenger.h>
#include <linux/ceph/mdsmap.h>
+#include <linux/ceph/auth.h>
/*
* Some lock dependencies:
@@ -20,7 +21,7 @@
*
* mdsc->snap_rwsem
*
- * inode->i_lock
+ * ci->i_ceph_lock
* mdsc->snap_flush_lock
* mdsc->cap_delay_lock
*
@@ -35,6 +36,7 @@ struct ceph_cap;
*/
struct ceph_mds_reply_info_in {
struct ceph_mds_reply_inode *in;
+ struct ceph_dir_layout dir_layout;
u32 symlink_len;
char *symlink;
u32 xattr_len;
@@ -42,26 +44,44 @@ struct ceph_mds_reply_info_in {
};
/*
- * parsed info about an mds reply, including information about the
- * target inode and/or its parent directory and dentry, and directory
- * contents (for readdir results).
+ * parsed info about an mds reply, including information about
+ * either: 1) the target inode and/or its parent directory and dentry,
+ * and directory contents (for readdir results), or
+ * 2) the file range lock info (for fcntl F_GETLK results).
*/
struct ceph_mds_reply_info_parsed {
struct ceph_mds_reply_head *head;
+ /* trace */
struct ceph_mds_reply_info_in diri, targeti;
struct ceph_mds_reply_dirfrag *dirfrag;
char *dname;
u32 dname_len;
struct ceph_mds_reply_lease *dlease;
- struct ceph_mds_reply_dirfrag *dir_dir;
- int dir_nr;
- char **dir_dname;
- u32 *dir_dname_len;
- struct ceph_mds_reply_lease **dir_dlease;
- struct ceph_mds_reply_info_in *dir_in;
- u8 dir_complete, dir_end;
+ /* extra */
+ union {
+ /* for fcntl F_GETLK results */
+ struct ceph_filelock *filelock_reply;
+
+ /* for readdir results */
+ struct {
+ struct ceph_mds_reply_dirfrag *dir_dir;
+ size_t dir_buf_size;
+ int dir_nr;
+ char **dir_dname;
+ u32 *dir_dname_len;
+ struct ceph_mds_reply_lease **dir_dlease;
+ struct ceph_mds_reply_info_in *dir_in;
+ u8 dir_complete, dir_end;
+ };
+
+ /* for create results */
+ struct {
+ bool has_create_ino;
+ u64 ino;
+ };
+ };
/* encoded blob describing snapshot contexts for certain
operations (e.g., open) */
@@ -101,17 +121,19 @@ struct ceph_mds_session {
struct ceph_connection s_con;
- struct ceph_authorizer *s_authorizer;
- void *s_authorizer_buf, *s_authorizer_reply_buf;
- size_t s_authorizer_buf_len, s_authorizer_reply_buf_len;
+ struct ceph_auth_handshake s_auth;
- /* protected by s_cap_lock */
- spinlock_t s_cap_lock;
+ /* protected by s_gen_ttl_lock */
+ spinlock_t s_gen_ttl_lock;
u32 s_cap_gen; /* inc each time we get mds stale msg */
unsigned long s_cap_ttl; /* when session caps expire */
+
+ /* protected by s_cap_lock */
+ spinlock_t s_cap_lock;
struct list_head s_caps; /* all caps issued by this session */
int s_nr_caps, s_trim_caps;
int s_num_cap_releases;
+ int s_cap_reconnect;
struct list_head s_cap_releases; /* waiting cap_release messages */
struct list_head s_cap_releases_done; /* ready to send */
struct ceph_cap *s_cap_iterator;
@@ -154,12 +176,12 @@ struct ceph_mds_request {
struct ceph_mds_client *r_mdsc;
int r_op; /* mds op code */
- int r_mds;
/* operation on what? */
struct inode *r_inode; /* arg1 */
struct dentry *r_dentry; /* arg1 */
struct dentry *r_old_dentry; /* arg2: rename from or link from */
+ struct inode *r_old_dentry_dir; /* arg2: old dentry's parent dir */
char *r_path1, *r_path2;
struct ceph_vino r_ino1, r_ino2;
@@ -170,6 +192,9 @@ struct ceph_mds_request {
union ceph_mds_request_args r_args;
int r_fmode; /* file mode, if expecting cap */
+ kuid_t r_uid;
+ kgid_t r_gid;
+ struct timespec r_stamp;
/* for choosing which mds to send this request to */
int r_direct_mode;
@@ -265,6 +290,7 @@ struct ceph_mds_client {
u64 cap_flush_seq;
struct list_head cap_dirty; /* inodes with dirty caps */
+ struct list_head cap_dirty_migrating; /* ...that are migration... */
int num_cap_flushing; /* # caps we are flushing */
spinlock_t cap_dirty_lock; /* protects above items */
wait_queue_head_t cap_flushing_wq;
@@ -319,10 +345,11 @@ extern void ceph_mdsc_sync(struct ceph_mds_client *mdsc);
extern void ceph_mdsc_lease_release(struct ceph_mds_client *mdsc,
struct inode *inode,
- struct dentry *dn, int mask);
+ struct dentry *dn);
extern void ceph_invalidate_dir_request(struct ceph_mds_request *req);
-
+extern int ceph_alloc_readdir_reply_buffer(struct ceph_mds_request *req,
+ struct inode *dir);
extern struct ceph_mds_request *
ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op, int mode);
extern void ceph_mdsc_submit_request(struct ceph_mds_client *mdsc,
@@ -359,6 +386,8 @@ extern void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session,
extern void ceph_mdsc_handle_map(struct ceph_mds_client *mdsc,
struct ceph_msg *msg);
+extern struct ceph_mds_session *
+ceph_mdsc_open_export_target_session(struct ceph_mds_client *mdsc, int target);
extern void ceph_mdsc_open_export_target_sessions(struct ceph_mds_client *mdsc,
struct ceph_mds_session *session);
diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
index 73b7d44e8a3..261531e55e9 100644
--- a/fs/ceph/mdsmap.c
+++ b/fs/ceph/mdsmap.c
@@ -20,7 +20,10 @@ int ceph_mdsmap_get_random_mds(struct ceph_mdsmap *m)
{
int n = 0;
int i;
- char r;
+
+ /* special case for one mds */
+ if (1 == m->m_max_mds && m->m_info[0].state > 0)
+ return 0;
/* count */
for (i = 0; i < m->m_max_mds; i++)
@@ -30,8 +33,7 @@ int ceph_mdsmap_get_random_mds(struct ceph_mdsmap *m)
return -1;
/* pick */
- get_random_bytes(&r, 1);
- n = r % n;
+ n = prandom_u32() % n;
i = 0;
for (i = 0; n > 0; i++, n--)
while (m->m_info[i].state <= 0)
@@ -59,6 +61,10 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
return ERR_PTR(-ENOMEM);
ceph_decode_16_safe(p, end, version, bad);
+ if (version > 3) {
+ pr_warn("got mdsmap version %d > 3, failing", version);
+ goto bad;
+ }
ceph_decode_need(p, end, 8*sizeof(u32) + sizeof(u64), bad);
m->m_epoch = ceph_decode_32(p);
@@ -86,6 +92,7 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
u32 num_export_targets;
void *pexport_targets = NULL;
struct ceph_timespec laggy_since;
+ struct ceph_mds_info *info;
ceph_decode_need(p, end, sizeof(u64)*2 + 1 + sizeof(u32), bad);
global_id = ceph_decode_64(p);
@@ -120,37 +127,40 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
i+1, n, global_id, mds, inc,
ceph_pr_addr(&addr.in_addr),
ceph_mds_state_name(state));
- if (mds >= 0 && mds < m->m_max_mds && state > 0) {
- m->m_info[mds].global_id = global_id;
- m->m_info[mds].state = state;
- m->m_info[mds].addr = addr;
- m->m_info[mds].laggy =
- (laggy_since.tv_sec != 0 ||
- laggy_since.tv_nsec != 0);
- m->m_info[mds].num_export_targets = num_export_targets;
- if (num_export_targets) {
- m->m_info[mds].export_targets =
- kcalloc(num_export_targets, sizeof(u32),
- GFP_NOFS);
- for (j = 0; j < num_export_targets; j++)
- m->m_info[mds].export_targets[j] =
- ceph_decode_32(&pexport_targets);
- } else {
- m->m_info[mds].export_targets = NULL;
- }
+
+ if (mds < 0 || mds >= m->m_max_mds || state <= 0)
+ continue;
+
+ info = &m->m_info[mds];
+ info->global_id = global_id;
+ info->state = state;
+ info->addr = addr;
+ info->laggy = (laggy_since.tv_sec != 0 ||
+ laggy_since.tv_nsec != 0);
+ info->num_export_targets = num_export_targets;
+ if (num_export_targets) {
+ info->export_targets = kcalloc(num_export_targets,
+ sizeof(u32), GFP_NOFS);
+ if (info->export_targets == NULL)
+ goto badmem;
+ for (j = 0; j < num_export_targets; j++)
+ info->export_targets[j] =
+ ceph_decode_32(&pexport_targets);
+ } else {
+ info->export_targets = NULL;
}
}
/* pg_pools */
ceph_decode_32_safe(p, end, n, bad);
m->m_num_data_pg_pools = n;
- m->m_data_pg_pools = kcalloc(n, sizeof(u32), GFP_NOFS);
+ m->m_data_pg_pools = kcalloc(n, sizeof(u64), GFP_NOFS);
if (!m->m_data_pg_pools)
goto badmem;
- ceph_decode_need(p, end, sizeof(u32)*(n+1), bad);
+ ceph_decode_need(p, end, sizeof(u64)*(n+1), bad);
for (i = 0; i < n; i++)
- m->m_data_pg_pools[i] = ceph_decode_32(p);
- m->m_cas_pg_pool = ceph_decode_32(p);
+ m->m_data_pg_pools[i] = ceph_decode_64(p);
+ m->m_cas_pg_pool = ceph_decode_64(p);
/* ok, we don't care about the rest. */
dout("mdsmap_decode success epoch %u\n", m->m_epoch);
@@ -164,7 +174,7 @@ bad:
DUMP_PREFIX_OFFSET, 16, 1,
start, end - start, true);
ceph_mdsmap_destroy(m);
- return ERR_PTR(-EINVAL);
+ return ERR_PTR(err);
}
void ceph_mdsmap_destroy(struct ceph_mdsmap *m)
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index 39c243acd06..f01645a2775 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -206,7 +206,7 @@ void ceph_put_snap_realm(struct ceph_mds_client *mdsc,
up_write(&mdsc->snap_rwsem);
} else {
spin_lock(&mdsc->snap_empty_lock);
- list_add(&mdsc->snap_empty, &realm->empty_item);
+ list_add(&realm->empty_item, &mdsc->snap_empty);
spin_unlock(&mdsc->snap_empty_lock);
}
}
@@ -296,8 +296,7 @@ static int build_snap_context(struct ceph_snap_realm *realm)
struct ceph_snap_realm *parent = realm->parent;
struct ceph_snap_context *snapc;
int err = 0;
- int i;
- int num = realm->num_prior_parent_snaps + realm->num_snaps;
+ u32 num = realm->num_prior_parent_snaps + realm->num_snaps;
/*
* build parent context, if it hasn't been built.
@@ -321,28 +320,29 @@ static int build_snap_context(struct ceph_snap_realm *realm)
realm->cached_context->seq == realm->seq &&
(!parent ||
realm->cached_context->seq >= parent->cached_context->seq)) {
- dout("build_snap_context %llx %p: %p seq %lld (%d snaps)"
+ dout("build_snap_context %llx %p: %p seq %lld (%u snaps)"
" (unchanged)\n",
realm->ino, realm, realm->cached_context,
realm->cached_context->seq,
- realm->cached_context->num_snaps);
+ (unsigned int) realm->cached_context->num_snaps);
return 0;
}
/* alloc new snap context */
err = -ENOMEM;
- if (num > ULONG_MAX / sizeof(u64) - sizeof(*snapc))
+ if (num > (SIZE_MAX - sizeof(*snapc)) / sizeof(u64))
goto fail;
- snapc = kzalloc(sizeof(*snapc) + num*sizeof(u64), GFP_NOFS);
+ snapc = ceph_create_snap_context(num, GFP_NOFS);
if (!snapc)
goto fail;
- atomic_set(&snapc->nref, 1);
/* build (reverse sorted) snap vector */
num = 0;
snapc->seq = realm->seq;
if (parent) {
- /* include any of parent's snaps occuring _after_ my
+ u32 i;
+
+ /* include any of parent's snaps occurring _after_ my
parent became my parent */
for (i = 0; i < parent->cached_context->num_snaps; i++)
if (parent->cached_context->snaps[i] >=
@@ -361,8 +361,9 @@ static int build_snap_context(struct ceph_snap_realm *realm)
sort(snapc->snaps, num, sizeof(u64), cmpu64_rev, NULL);
snapc->num_snaps = num;
- dout("build_snap_context %llx %p: %p seq %lld (%d snaps)\n",
- realm->ino, realm, snapc, snapc->seq, snapc->num_snaps);
+ dout("build_snap_context %llx %p: %p seq %lld (%u snaps)\n",
+ realm->ino, realm, snapc, snapc->seq,
+ (unsigned int) snapc->num_snaps);
if (realm->cached_context)
ceph_put_snap_context(realm->cached_context);
@@ -402,9 +403,9 @@ static void rebuild_snap_realms(struct ceph_snap_realm *realm)
* helper to allocate and decode an array of snapids. free prior
* instance, if any.
*/
-static int dup_array(u64 **dst, __le64 *src, int num)
+static int dup_array(u64 **dst, __le64 *src, u32 num)
{
- int i;
+ u32 i;
kfree(*dst);
if (num) {
@@ -446,9 +447,18 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci)
return;
}
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
used = __ceph_caps_used(ci);
dirty = __ceph_caps_dirty(ci);
+
+ /*
+ * If there is a write in progress, treat that as a dirty Fw,
+ * even though it hasn't completed yet; by the time we finish
+ * up this capsnap it will be.
+ */
+ if (used & CEPH_CAP_FILE_WR)
+ dirty |= CEPH_CAP_FILE_WR;
+
if (__ceph_have_pending_cap_snap(ci)) {
/* there is no point in queuing multiple "pending" cap_snaps,
as no new writes are allowed to start when pending, so any
@@ -456,15 +466,21 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci)
cap_snap. lucky us. */
dout("queue_cap_snap %p already pending\n", inode);
kfree(capsnap);
- } else if (ci->i_wrbuffer_ref_head || (used & CEPH_CAP_FILE_WR) ||
- (dirty & (CEPH_CAP_AUTH_EXCL|CEPH_CAP_XATTR_EXCL|
- CEPH_CAP_FILE_EXCL|CEPH_CAP_FILE_WR))) {
+ } else if (dirty & (CEPH_CAP_AUTH_EXCL|CEPH_CAP_XATTR_EXCL|
+ CEPH_CAP_FILE_EXCL|CEPH_CAP_FILE_WR)) {
struct ceph_snap_context *snapc = ci->i_head_snapc;
- dout("queue_cap_snap %p cap_snap %p queuing under %p\n", inode,
- capsnap, snapc);
- igrab(inode);
-
+ /*
+ * if we are a sync write, we may need to go to the snaprealm
+ * to get the current snapc.
+ */
+ if (!snapc)
+ snapc = ci->i_snap_realm->cached_context;
+
+ dout("queue_cap_snap %p cap_snap %p queuing under %p %s\n",
+ inode, capsnap, snapc, ceph_cap_string(dirty));
+ ihold(inode);
+
atomic_set(&capsnap->nref, 1);
capsnap->ci = ci;
INIT_LIST_HEAD(&capsnap->ci_item);
@@ -513,7 +529,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci)
kfree(capsnap);
}
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
}
/*
@@ -522,7 +538,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci)
*
* If capsnap can now be flushed, add to snap_flush list, and return 1.
*
- * Caller must hold i_lock.
+ * Caller must hold i_ceph_lock.
*/
int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
struct ceph_cap_snap *capsnap)
@@ -584,10 +600,14 @@ static void queue_realm_cap_snaps(struct ceph_snap_realm *realm)
if (lastinode)
iput(lastinode);
- dout("queue_realm_cap_snaps %p %llx children\n", realm, realm->ino);
- list_for_each_entry(child, &realm->children, child_item)
- queue_realm_cap_snaps(child);
+ list_for_each_entry(child, &realm->children, child_item) {
+ dout("queue_realm_cap_snaps %p %llx queue child %p %llx\n",
+ realm, realm->ino, child, child->ino);
+ list_del_init(&child->dirty_item);
+ list_add(&child->dirty_item, &realm->dirty_item);
+ }
+ list_del_init(&realm->dirty_item);
dout("queue_realm_cap_snaps %p %llx done\n", realm, realm->ino);
}
@@ -683,7 +703,9 @@ more:
* queue cap snaps _after_ we've built the new snap contexts,
* so that i_head_snapc can be set appropriately.
*/
- list_for_each_entry(realm, &dirty_realms, dirty_item) {
+ while (!list_empty(&dirty_realms)) {
+ realm = list_first_entry(&dirty_realms, struct ceph_snap_realm,
+ dirty_item);
queue_realm_cap_snaps(realm);
}
@@ -716,11 +738,11 @@ static void flush_snaps(struct ceph_mds_client *mdsc)
ci = list_first_entry(&mdsc->snap_flush_list,
struct ceph_inode_info, i_snap_flush_item);
inode = &ci->vfs_inode;
- igrab(inode);
+ ihold(inode);
spin_unlock(&mdsc->snap_flush_lock);
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
__ceph_flush_snaps(ci, &session, 0);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
iput(inode);
spin_lock(&mdsc->snap_flush_lock);
}
@@ -826,7 +848,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc,
continue;
ci = ceph_inode(inode);
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (!ci->i_snap_realm)
goto skip_inode;
/*
@@ -855,7 +877,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc,
oldrealm = ci->i_snap_realm;
ci->i_snap_realm = realm;
spin_unlock(&realm->inodes_with_caps_lock);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
ceph_get_snap_realm(mdsc, realm);
ceph_put_snap_realm(mdsc, oldrealm);
@@ -864,7 +886,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc,
continue;
skip_inode:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
iput(inode);
}
diff --git a/fs/ceph/strings.c b/fs/ceph/strings.c
index cd5097d7c80..51cc23e4811 100644
--- a/fs/ceph/strings.c
+++ b/fs/ceph/strings.c
@@ -15,6 +15,7 @@ const char *ceph_mds_state_name(int s)
case CEPH_MDS_STATE_BOOT: return "up:boot";
case CEPH_MDS_STATE_STANDBY: return "up:standby";
case CEPH_MDS_STATE_STANDBY_REPLAY: return "up:standby-replay";
+ case CEPH_MDS_STATE_REPLAYONCE: return "up:oneshot-replay";
case CEPH_MDS_STATE_CREATING: return "up:creating";
case CEPH_MDS_STATE_STARTING: return "up:starting";
/* up and in */
@@ -40,6 +41,8 @@ const char *ceph_session_op_name(int op)
case CEPH_SESSION_RENEWCAPS: return "renewcaps";
case CEPH_SESSION_STALE: return "stale";
case CEPH_SESSION_RECALL_STATE: return "recall_state";
+ case CEPH_SESSION_FLUSHMSG: return "flushmsg";
+ case CEPH_SESSION_FLUSHMSG_ACK: return "flushmsg_ack";
}
return "???";
}
@@ -50,10 +53,14 @@ const char *ceph_mds_op_name(int op)
case CEPH_MDS_OP_LOOKUP: return "lookup";
case CEPH_MDS_OP_LOOKUPHASH: return "lookuphash";
case CEPH_MDS_OP_LOOKUPPARENT: return "lookupparent";
+ case CEPH_MDS_OP_LOOKUPINO: return "lookupino";
+ case CEPH_MDS_OP_LOOKUPNAME: return "lookupname";
case CEPH_MDS_OP_GETATTR: return "getattr";
case CEPH_MDS_OP_SETXATTR: return "setxattr";
case CEPH_MDS_OP_SETATTR: return "setattr";
case CEPH_MDS_OP_RMXATTR: return "rmxattr";
+ case CEPH_MDS_OP_SETLAYOUT: return "setlayou";
+ case CEPH_MDS_OP_SETDIRLAYOUT: return "setdirlayout";
case CEPH_MDS_OP_READDIR: return "readdir";
case CEPH_MDS_OP_MKNOD: return "mknod";
case CEPH_MDS_OP_LINK: return "link";
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index 08b460ae053..06150fd745a 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -17,7 +17,9 @@
#include "super.h"
#include "mds_client.h"
+#include "cache.h"
+#include <linux/ceph/ceph_features.h>
#include <linux/ceph/decode.h>
#include <linux/ceph/mon_client.h>
#include <linux/ceph/auth.h>
@@ -70,17 +72,21 @@ static int ceph_statfs(struct dentry *dentry, struct kstatfs *buf)
/*
* express utilization in terms of large blocks to avoid
* overflow on 32-bit machines.
+ *
+ * NOTE: for the time being, we make bsize == frsize to humor
+ * not-yet-ancient versions of glibc that are broken.
+ * Someday, we will probably want to report a real block
+ * size... whatever that may mean for a network file system!
*/
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) - le64_to_cpu(st.kb_used)) >>
- (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;
buf->f_namelen = NAME_MAX;
- buf->f_frsize = PAGE_CACHE_SIZE;
/* leave fsid little-endian, regardless of host endianness */
fsid = *(u64 *)(&monmap->fsid) ^ *((u64 *)&monmap->fsid + 1);
@@ -115,6 +121,7 @@ static int ceph_sync_fs(struct super_block *sb, int wait)
enum {
Opt_wsize,
Opt_rsize,
+ Opt_rasize,
Opt_caps_wanted_delay_min,
Opt_caps_wanted_delay_max,
Opt_cap_release_safety,
@@ -130,12 +137,24 @@ enum {
Opt_nodirstat,
Opt_rbytes,
Opt_norbytes,
+ Opt_asyncreaddir,
Opt_noasyncreaddir,
+ Opt_dcache,
+ Opt_nodcache,
+ Opt_ino32,
+ Opt_noino32,
+ Opt_fscache,
+ Opt_nofscache,
+#ifdef CONFIG_CEPH_FS_POSIX_ACL
+ Opt_acl,
+#endif
+ Opt_noacl
};
static match_table_t fsopt_tokens = {
{Opt_wsize, "wsize=%d"},
{Opt_rsize, "rsize=%d"},
+ {Opt_rasize, "rasize=%d"},
{Opt_caps_wanted_delay_min, "caps_wanted_delay_min=%d"},
{Opt_caps_wanted_delay_max, "caps_wanted_delay_max=%d"},
{Opt_cap_release_safety, "cap_release_safety=%d"},
@@ -149,7 +168,18 @@ static match_table_t fsopt_tokens = {
{Opt_nodirstat, "nodirstat"},
{Opt_rbytes, "rbytes"},
{Opt_norbytes, "norbytes"},
+ {Opt_asyncreaddir, "asyncreaddir"},
{Opt_noasyncreaddir, "noasyncreaddir"},
+ {Opt_dcache, "dcache"},
+ {Opt_nodcache, "nodcache"},
+ {Opt_ino32, "ino32"},
+ {Opt_noino32, "noino32"},
+ {Opt_fscache, "fsc"},
+ {Opt_nofscache, "nofsc"},
+#ifdef CONFIG_CEPH_FS_POSIX_ACL
+ {Opt_acl, "acl"},
+#endif
+ {Opt_noacl, "noacl"},
{-1, NULL}
};
@@ -195,6 +225,9 @@ static int parse_fsopt_token(char *c, void *private)
case Opt_rsize:
fsopt->rsize = intval;
break;
+ case Opt_rasize:
+ fsopt->rasize = intval;
+ break;
case Opt_caps_wanted_delay_min:
fsopt->caps_wanted_delay_min = intval;
break;
@@ -222,9 +255,38 @@ static int parse_fsopt_token(char *c, void *private)
case Opt_norbytes:
fsopt->flags &= ~CEPH_MOUNT_OPT_RBYTES;
break;
+ case Opt_asyncreaddir:
+ fsopt->flags &= ~CEPH_MOUNT_OPT_NOASYNCREADDIR;
+ break;
case Opt_noasyncreaddir:
fsopt->flags |= CEPH_MOUNT_OPT_NOASYNCREADDIR;
break;
+ case Opt_dcache:
+ fsopt->flags |= CEPH_MOUNT_OPT_DCACHE;
+ break;
+ case Opt_nodcache:
+ fsopt->flags &= ~CEPH_MOUNT_OPT_DCACHE;
+ break;
+ case Opt_ino32:
+ fsopt->flags |= CEPH_MOUNT_OPT_INO32;
+ break;
+ case Opt_noino32:
+ fsopt->flags &= ~CEPH_MOUNT_OPT_INO32;
+ break;
+ case Opt_fscache:
+ fsopt->flags |= CEPH_MOUNT_OPT_FSCACHE;
+ break;
+ case Opt_nofscache:
+ fsopt->flags &= ~CEPH_MOUNT_OPT_FSCACHE;
+ break;
+#ifdef CONFIG_CEPH_FS_POSIX_ACL
+ case Opt_acl:
+ fsopt->sb_flags |= MS_POSIXACL;
+ break;
+#endif
+ case Opt_noacl:
+ fsopt->sb_flags &= ~MS_POSIXACL;
+ break;
default:
BUG_ON(token);
}
@@ -277,7 +339,10 @@ static int parse_mount_options(struct ceph_mount_options **pfsopt,
{
struct ceph_mount_options *fsopt;
const char *dev_name_end;
- int err = -ENOMEM;
+ int err;
+
+ if (!dev_name || !*dev_name)
+ return -EINVAL;
fsopt = kzalloc(sizeof(*fsopt), GFP_KERNEL);
if (!fsopt)
@@ -285,37 +350,54 @@ static int parse_mount_options(struct ceph_mount_options **pfsopt,
dout("parse_mount_options %p, dev_name '%s'\n", fsopt, dev_name);
- fsopt->sb_flags = flags;
- fsopt->flags = CEPH_MOUNT_OPT_DEFAULT;
-
- fsopt->rsize = CEPH_MOUNT_RSIZE_DEFAULT;
- fsopt->snapdir_name = kstrdup(CEPH_SNAPDIRNAME_DEFAULT, GFP_KERNEL);
- 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;
- }
- dev_name_end = *path;
- dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name);
+ fsopt->sb_flags = flags;
+ fsopt->flags = CEPH_MOUNT_OPT_DEFAULT;
+
+ fsopt->rsize = CEPH_RSIZE_DEFAULT;
+ fsopt->rasize = CEPH_RASIZE_DEFAULT;
+ fsopt->snapdir_name = kstrdup(CEPH_SNAPDIRNAME_DEFAULT, GFP_KERNEL);
+ fsopt->caps_wanted_delay_min = CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT;
+ fsopt->caps_wanted_delay_max = CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT;
+ 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();
- /* path on server */
- *path += 2;
+ /*
+ * Distinguish the server list from the path in "dev_name".
+ * Internally we do not include the leading '/' in the path.
+ *
+ * "dev_name" will look like:
+ * <server_spec>[,<server_spec>...]:[<path>]
+ * where
+ * <server_spec> is <ip>[:<port>]
+ * <path> is optional, but if present must begin with '/'
+ */
+ dev_name_end = strchr(dev_name, '/');
+ if (dev_name_end) {
+ /* skip over leading '/' for path */
+ *path = dev_name_end + 1;
+ } else {
+ /* path is empty */
+ dev_name_end = dev_name + strlen(dev_name);
+ *path = dev_name_end;
+ }
+ err = -EINVAL;
+ dev_name_end--; /* back up to ':' separator */
+ if (dev_name_end < dev_name || *dev_name_end != ':') {
+ pr_err("device name is missing path (no : separator in %s)\n",
+ dev_name);
+ goto out;
+ }
+ dout("device name '%.*s'\n", (int)(dev_name_end - dev_name), dev_name);
dout("server path '%s'\n", *path);
- err = ceph_parse_options(popt, options, dev_name, dev_name_end,
+ *popt = ceph_parse_options(options, dev_name, dev_name_end,
parse_fsopt_token, (void *)fsopt);
- if (err)
+ if (IS_ERR(*popt)) {
+ err = PTR_ERR(*popt);
goto out;
+ }
/* success */
*pfsopt = fsopt;
@@ -329,11 +411,11 @@ out:
/**
* ceph_show_options - Show mount options in /proc/mounts
* @m: seq_file to write to
- * @mnt: mount descriptor
+ * @root: root of that (sub)tree
*/
-static int ceph_show_options(struct seq_file *m, struct vfsmount *mnt)
+static int ceph_show_options(struct seq_file *m, struct dentry *root)
{
- struct ceph_fs_client *fsc = ceph_sb_to_client(mnt->mnt_sb);
+ struct ceph_fs_client *fsc = ceph_sb_to_client(root->d_sb);
struct ceph_mount_options *fsopt = fsc->mount_options;
struct ceph_options *opt = fsc->client->options;
@@ -346,15 +428,13 @@ static int ceph_show_options(struct seq_file *m, struct vfsmount *mnt)
if (opt->name)
seq_printf(m, ",name=%s", opt->name);
- if (opt->secret)
+ if (opt->key)
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);
@@ -365,11 +445,28 @@ static int ceph_show_options(struct seq_file *m, struct vfsmount *mnt)
seq_puts(m, ",norbytes");
if (fsopt->flags & CEPH_MOUNT_OPT_NOASYNCREADDIR)
seq_puts(m, ",noasyncreaddir");
+ if (fsopt->flags & CEPH_MOUNT_OPT_DCACHE)
+ seq_puts(m, ",dcache");
+ else
+ seq_puts(m, ",nodcache");
+ if (fsopt->flags & CEPH_MOUNT_OPT_FSCACHE)
+ seq_puts(m, ",fsc");
+ else
+ seq_puts(m, ",nofsc");
+
+#ifdef CONFIG_CEPH_FS_POSIX_ACL
+ if (fsopt->sb_flags & MS_POSIXACL)
+ seq_puts(m, ",acl");
+ else
+ seq_puts(m, ",noacl");
+#endif
if (fsopt->wsize)
seq_printf(m, ",wsize=%d", fsopt->wsize);
- if (fsopt->rsize != CEPH_MOUNT_RSIZE_DEFAULT)
+ if (fsopt->rsize != CEPH_RSIZE_DEFAULT)
seq_printf(m, ",rsize=%d", fsopt->rsize);
+ if (fsopt->rasize != CEPH_RASIZE_DEFAULT)
+ seq_printf(m, ",rasize=%d", fsopt->rasize);
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)
@@ -412,23 +509,29 @@ static int extra_mon_dispatch(struct ceph_client *client, struct ceph_msg *msg)
/*
* create a new fs client
*/
-struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt,
+static struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt,
struct ceph_options *opt)
{
struct ceph_fs_client *fsc;
+ const u64 supported_features =
+ CEPH_FEATURE_FLOCK |
+ CEPH_FEATURE_DIRLAYOUTHASH;
+ const u64 required_features = 0;
+ int page_count;
+ size_t size;
int err = -ENOMEM;
fsc = kzalloc(sizeof(*fsc), GFP_KERNEL);
if (!fsc)
return ERR_PTR(-ENOMEM);
- fsc->client = ceph_create_client(opt, fsc);
+ fsc->client = ceph_create_client(opt, fsc, supported_features,
+ required_features);
if (IS_ERR(fsc->client)) {
err = PTR_ERR(fsc->client);
goto fail;
}
fsc->client->extra_mon_dispatch = extra_mon_dispatch;
- fsc->client->supported_features |= CEPH_FEATURE_FLOCK;
fsc->client->monc.want_mdsmap = 1;
fsc->mount_options = fsopt;
@@ -443,28 +546,40 @@ struct ceph_fs_client *create_fs_client(struct ceph_mount_options *fsopt,
goto fail_client;
err = -ENOMEM;
- fsc->wb_wq = create_workqueue("ceph-writeback");
+ /*
+ * 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);
if (fsc->wb_wq == NULL)
goto fail_bdi;
- fsc->pg_inv_wq = create_singlethread_workqueue("ceph-pg-invalid");
+ fsc->pg_inv_wq = alloc_workqueue("ceph-pg-invalid", 0, 1);
if (fsc->pg_inv_wq == NULL)
goto fail_wb_wq;
- fsc->trunc_wq = create_singlethread_workqueue("ceph-trunc");
+ fsc->trunc_wq = alloc_workqueue("ceph-trunc", 0, 1);
if (fsc->trunc_wq == NULL)
goto fail_pg_inv_wq;
/* set up mempools */
err = -ENOMEM;
- fsc->wb_pagevec_pool = mempool_create_kmalloc_pool(10,
- fsc->mount_options->wsize >> PAGE_CACHE_SHIFT);
+ page_count = fsc->mount_options->wsize >> PAGE_CACHE_SHIFT;
+ size = sizeof (struct page *) * (page_count ? page_count : 1);
+ fsc->wb_pagevec_pool = mempool_create_kmalloc_pool(10, size);
if (!fsc->wb_pagevec_pool)
goto fail_trunc_wq;
+ /* setup fscache */
+ if ((fsopt->flags & CEPH_MOUNT_OPT_FSCACHE) &&
+ (ceph_fscache_register_fs(fsc) != 0))
+ goto fail_fscache;
+
/* caps */
fsc->min_caps = fsopt->max_readdir;
return fsc;
+fail_fscache:
+ ceph_fscache_unregister_fs(fsc);
fail_trunc_wq:
destroy_workqueue(fsc->trunc_wq);
fail_pg_inv_wq:
@@ -480,10 +595,12 @@ fail:
return ERR_PTR(err);
}
-void destroy_fs_client(struct ceph_fs_client *fsc)
+static void destroy_fs_client(struct ceph_fs_client *fsc)
{
dout("destroy_fs_client %p\n", fsc);
+ ceph_fscache_unregister_fs(fsc);
+
destroy_workqueue(fsc->wb_wq);
destroy_workqueue(fsc->pg_inv_wq);
destroy_workqueue(fsc->trunc_wq);
@@ -518,6 +635,8 @@ static void ceph_inode_init_once(void *foo)
static int __init init_caches(void)
{
+ int error = -ENOMEM;
+
ceph_inode_cachep = kmem_cache_create("ceph_inode_info",
sizeof(struct ceph_inode_info),
__alignof__(struct ceph_inode_info),
@@ -541,23 +660,33 @@ static int __init init_caches(void)
if (ceph_file_cachep == NULL)
goto bad_file;
- return 0;
+ if ((error = ceph_fscache_register()))
+ goto bad_file;
+ return 0;
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;
+ return error;
}
static void destroy_caches(void)
{
+ /*
+ * Make sure all delayed rcu free inodes are flushed before we
+ * destroy cache.
+ */
+ rcu_barrier();
+
kmem_cache_destroy(ceph_inode_cachep);
kmem_cache_destroy(ceph_cap_cachep);
kmem_cache_destroy(ceph_dentry_cachep);
kmem_cache_destroy(ceph_file_cachep);
+
+ ceph_fscache_unregister();
}
@@ -580,6 +709,7 @@ static const struct super_operations ceph_super_ops = {
.alloc_inode = ceph_alloc_inode,
.destroy_inode = ceph_destroy_inode,
.write_inode = ceph_write_inode,
+ .drop_inode = ceph_drop_inode,
.sync_fs = ceph_sync_fs,
.put_super = ceph_put_super,
.show_options = ceph_show_options,
@@ -614,17 +744,25 @@ static struct dentry *open_root_dentry(struct ceph_fs_client *fsc,
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 &&
- fsc->sb->s_root == NULL)
- root = d_alloc_root(req->r_target_inode);
- else
- root = d_obtain_alias(req->r_target_inode);
+ struct inode *inode = req->r_target_inode;
req->r_target_inode = NULL;
+ dout("open_root_inode success\n");
+ if (ceph_ino(inode) == CEPH_INO_ROOT &&
+ fsc->sb->s_root == NULL) {
+ root = d_make_root(inode);
+ if (!root) {
+ root = ERR_PTR(-ENOMEM);
+ goto out;
+ }
+ } else {
+ root = d_obtain_alias(inode);
+ }
+ ceph_init_dentry(root);
dout("open_root_inode success, root dentry is %p\n", root);
} else {
root = ERR_PTR(err);
}
+out:
ceph_mdsc_put_request(req);
return root;
}
@@ -705,6 +843,7 @@ static int ceph_set_super(struct super_block *s, void *data)
s->s_flags = fsc->mount_options->sb_flags;
s->s_maxbytes = 1ULL << 40; /* temp value until we get mdsmap */
+ s->s_xattr = ceph_xattr_handlers;
s->s_fs_info = fsc;
fsc->sb = s;
@@ -763,12 +902,16 @@ static int ceph_register_bdi(struct super_block *sb,
{
int err;
- /* set ra_pages based on rsize mount option? */
- if (fsc->mount_options->rsize >= PAGE_CACHE_SIZE)
+ /* set ra_pages based on rasize mount option? */
+ if (fsc->mount_options->rasize >= PAGE_CACHE_SIZE)
fsc->backing_dev_info.ra_pages =
- (fsc->mount_options->rsize + PAGE_CACHE_SIZE - 1)
+ (fsc->mount_options->rasize + PAGE_CACHE_SIZE - 1)
>> PAGE_SHIFT;
- err = bdi_register(&fsc->backing_dev_info, NULL, "ceph-%d",
+ else
+ fsc->backing_dev_info.ra_pages =
+ default_backing_dev_info.ra_pages;
+
+ err = bdi_register(&fsc->backing_dev_info, NULL, "ceph-%ld",
atomic_long_inc_return(&bdi_seq));
if (!err)
sb->s_bdi = &fsc->backing_dev_info;
@@ -788,6 +931,10 @@ static struct dentry *ceph_mount(struct file_system_type *fs_type,
struct ceph_options *opt = NULL;
dout("ceph_mount\n");
+
+#ifdef CONFIG_CEPH_FS_POSIX_ACL
+ flags |= MS_POSIXACL;
+#endif
err = parse_mount_options(&fsopt, &opt, flags, data, dev_name, &path);
if (err < 0) {
res = ERR_PTR(err);
@@ -798,8 +945,8 @@ static struct dentry *ceph_mount(struct file_system_type *fs_type,
fsc = create_fs_client(fsopt, opt);
if (IS_ERR(fsc)) {
res = ERR_CAST(fsc);
- kfree(fsopt);
- kfree(opt);
+ destroy_mount_options(fsopt);
+ ceph_destroy_options(opt);
goto out_final;
}
@@ -811,7 +958,7 @@ static struct dentry *ceph_mount(struct file_system_type *fs_type,
if (ceph_test_opt(fsc->client, NOSHARE))
compare_super = NULL;
- sb = sget(fs_type, compare_super, ceph_set_super, fsc);
+ sb = sget(fs_type, compare_super, ceph_set_super, flags, fsc);
if (IS_ERR(sb)) {
res = ERR_CAST(sb);
goto out;
@@ -868,6 +1015,7 @@ static struct file_system_type ceph_fs_type = {
.kill_sb = ceph_kill_sb,
.fs_flags = FS_RENAME_DOES_D_MOVE,
};
+MODULE_ALIAS_FS("ceph");
#define _STRINGIFY(x) #x
#define STRINGIFY(x) _STRINGIFY(x)
@@ -878,6 +1026,8 @@ static int __init init_ceph(void)
if (ret)
goto out;
+ ceph_flock_init();
+ ceph_xattr_init();
ret = register_filesystem(&ceph_fs_type);
if (ret)
goto out_icache;
@@ -887,6 +1037,7 @@ static int __init init_ceph(void)
return 0;
out_icache:
+ ceph_xattr_exit();
destroy_caches();
out:
return ret;
@@ -896,6 +1047,7 @@ static void __exit exit_ceph(void)
{
dout("exit_ceph\n");
unregister_filesystem(&ceph_fs_type);
+ ceph_xattr_exit();
destroy_caches();
}
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index 1886294e12f..12b20744e38 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -13,20 +13,28 @@
#include <linux/wait.h>
#include <linux/writeback.h>
#include <linux/slab.h>
+#include <linux/posix_acl.h>
#include <linux/ceph/libceph.h>
+#ifdef CONFIG_CEPH_FSCACHE
+#include <linux/fscache.h>
+#endif
+
/* f_type in struct statfs */
#define CEPH_SUPER_MAGIC 0x00c36400
/* large granularity for statfs utilization stats to facilitate
* large volume sizes on 32-bit machines. */
-#define CEPH_BLOCK_SHIFT 20 /* 1 MB */
+#define CEPH_BLOCK_SHIFT 22 /* 4 MB */
#define CEPH_BLOCK (1 << CEPH_BLOCK_SHIFT)
#define CEPH_MOUNT_OPT_DIRSTAT (1<<4) /* `cat dirname` for stats */
#define CEPH_MOUNT_OPT_RBYTES (1<<5) /* dir st_bytes = rbytes */
#define CEPH_MOUNT_OPT_NOASYNCREADDIR (1<<7) /* no dcache readdir */
+#define CEPH_MOUNT_OPT_INO32 (1<<8) /* 32 bit inos */
+#define CEPH_MOUNT_OPT_DCACHE (1<<9) /* use dcache for readdir etc */
+#define CEPH_MOUNT_OPT_FSCACHE (1<<10) /* use fscache */
#define CEPH_MOUNT_OPT_DEFAULT (CEPH_MOUNT_OPT_RBYTES)
@@ -35,6 +43,8 @@
#define ceph_test_mount_opt(fsc, opt) \
(!!((fsc)->mount_options->flags & CEPH_MOUNT_OPT_##opt))
+#define CEPH_RSIZE_DEFAULT 0 /* max read size */
+#define CEPH_RASIZE_DEFAULT (8192*1024) /* readahead */
#define CEPH_MAX_READDIR_DEFAULT 1024
#define CEPH_MAX_READDIR_BYTES_DEFAULT (512*1024)
#define CEPH_SNAPDIRNAME_DEFAULT ".snap"
@@ -43,8 +53,9 @@ struct ceph_mount_options {
int flags;
int sb_flags;
- int wsize;
- int rsize; /* max readahead */
+ int wsize; /* max write size */
+ int rsize; /* max read size */
+ int rasize; /* max readahead */
int congestion_kb; /* max writeback in flight */
int caps_wanted_delay_min, caps_wanted_delay_max;
int cap_release_safety;
@@ -85,6 +96,11 @@ struct ceph_fs_client {
struct dentry *debugfs_bdi;
struct dentry *debugfs_mdsc, *debugfs_mdsmap;
#endif
+
+#ifdef CONFIG_CEPH_FSCACHE
+ struct fscache_cookie *fscache;
+ struct workqueue_struct *revalidate_wq;
+#endif
};
@@ -132,9 +148,9 @@ struct ceph_cap_snap {
int issued, dirty;
struct ceph_snap_context *context;
- mode_t mode;
- uid_t uid;
- gid_t gid;
+ umode_t mode;
+ kuid_t uid;
+ kgid_t gid;
struct ceph_buffer *xattr_blob;
u64 xattr_version;
@@ -233,12 +249,16 @@ struct ceph_inode_xattrs_info {
struct ceph_inode_info {
struct ceph_vino i_vino; /* ceph ino + snap */
+ spinlock_t i_ceph_lock;
+
u64 i_version;
u32 i_time_warp_seq;
unsigned i_ceph_flags;
- unsigned long i_release_count;
+ atomic_t i_release_count;
+ atomic_t i_complete_count;
+ struct ceph_dir_layout i_dir_layout;
struct ceph_file_layout i_layout;
char *i_symlink;
@@ -246,14 +266,13 @@ struct ceph_inode_info {
struct timespec i_rctime;
u64 i_rbytes, i_rfiles, i_rsubdirs;
u64 i_files, i_subdirs;
- u64 i_max_offset; /* largest readdir offset, set with I_COMPLETE */
struct rb_root i_fragtree;
struct mutex i_fragtree_mutex;
struct ceph_inode_xattrs_info i_xattrs;
- /* capabilities. protected _both_ by i_lock and cap->session's
+ /* capabilities. protected _both_ by i_ceph_lock and cap->session's
* s_mutex. */
struct rb_root i_caps; /* cap list */
struct ceph_cap *i_auth_cap; /* authoritative cap, if any */
@@ -268,9 +287,6 @@ struct ceph_inode_info {
unsigned long i_hold_caps_min; /* jiffies */
unsigned long i_hold_caps_max; /* jiffies */
struct list_head i_cap_delay_list; /* for delayed cap release to mds */
- int i_cap_exporting_mds; /* to handle cap migration between */
- unsigned i_cap_exporting_mseq; /* mds's. */
- unsigned i_cap_exporting_issued;
struct ceph_cap_reservation i_cap_migration_resv;
struct list_head i_cap_snaps; /* snapped state pending flush to mds */
struct ceph_snap_context *i_head_snapc; /* set if wr_buffer_head > 0 or
@@ -279,6 +295,7 @@ struct ceph_inode_info {
int i_nr_by_mode[CEPH_FILE_MODE_NUM]; /* open file counts */
+ struct mutex i_truncate_mutex;
u32 i_truncate_seq; /* last truncate to smaller size */
u64 i_truncate_size; /* and the size we last truncated down to */
int i_truncate_pending; /* still need to call vmtruncate */
@@ -290,12 +307,10 @@ struct ceph_inode_info {
/* held references to caps */
int i_pin_ref;
- int i_rd_ref, i_rdcache_ref, i_wr_ref;
+ int i_rd_ref, i_rdcache_ref, i_wr_ref, i_wb_ref;
int i_wrbuffer_ref, i_wrbuffer_ref_head;
u32 i_shared_gen; /* increment each time we get FILE_SHARED */
- u32 i_rdcache_gen; /* we increment this each time we get
- FILE_CACHE. If it's non-zero, we
- _may_ have cached pages. */
+ u32 i_rdcache_gen; /* incremented each time we get FILE_CACHE. */
u32 i_rdcache_revoking; /* RDCACHE gen to async invalidate, if any */
struct list_head i_unsafe_writes; /* uncommitted sync writes */
@@ -312,6 +327,11 @@ struct ceph_inode_info {
struct work_struct i_vmtruncate_work;
+#ifdef CONFIG_CEPH_FSCACHE
+ struct fscache_cookie *fscache;
+ u32 i_fscache_gen; /* sequence, for delayed fscache validate */
+ struct work_struct i_revalidate_work;
+#endif
struct inode vfs_inode; /* at end */
};
@@ -320,6 +340,16 @@ static inline struct ceph_inode_info *ceph_inode(struct inode *inode)
return container_of(inode, struct ceph_inode_info, vfs_inode);
}
+static inline struct ceph_fs_client *ceph_inode_to_client(struct inode *inode)
+{
+ return (struct ceph_fs_client *)inode->i_sb->s_fs_info;
+}
+
+static inline struct ceph_fs_client *ceph_sb_to_client(struct super_block *sb)
+{
+ return (struct ceph_fs_client *)sb->s_fs_info;
+}
+
static inline struct ceph_vino ceph_vino(struct inode *inode)
{
return ceph_inode(inode)->i_vino;
@@ -328,18 +358,49 @@ static inline struct ceph_vino ceph_vino(struct inode *inode)
/*
* ino_t is <64 bits on many architectures, blech.
*
- * don't include snap in ino hash, at least for now.
+ * i_ino (kernel inode) st_ino (userspace)
+ * i386 32 32
+ * x86_64+ino32 64 32
+ * x86_64 64 64
+ */
+static inline u32 ceph_ino_to_ino32(__u64 vino)
+{
+ u32 ino = vino & 0xffffffff;
+ ino ^= vino >> 32;
+ if (!ino)
+ ino = 2;
+ return ino;
+}
+
+/*
+ * kernel i_ino value
*/
static inline ino_t ceph_vino_to_ino(struct ceph_vino vino)
{
- ino_t ino = (ino_t)vino.ino; /* ^ (vino.snap << 20); */
#if BITS_PER_LONG == 32
- ino ^= vino.ino >> (sizeof(u64)-sizeof(ino_t)) * 8;
- if (!ino)
- ino = 1;
+ return ceph_ino_to_ino32(vino.ino);
+#else
+ return (ino_t)vino.ino;
#endif
+}
+
+/*
+ * user-visible ino (stat, filldir)
+ */
+#if BITS_PER_LONG == 32
+static inline ino_t ceph_translate_ino(struct super_block *sb, ino_t ino)
+{
+ return ino;
+}
+#else
+static inline ino_t ceph_translate_ino(struct super_block *sb, ino_t ino)
+{
+ if (ceph_test_mount_opt(ceph_sb_to_client(sb), INO32))
+ ino = ceph_ino_to_ino32(ino);
return ino;
}
+#endif
+
/* for printf-style formatting */
#define ceph_vinop(i) ceph_inode(i)->i_vino.ino, ceph_inode(i)->i_vino.snap
@@ -372,38 +433,35 @@ static inline struct inode *ceph_find_inode(struct super_block *sb,
/*
* Ceph inode.
*/
-#define CEPH_I_COMPLETE 1 /* we have complete directory cached */
#define CEPH_I_NODELAY 4 /* do not delay cap release */
#define CEPH_I_FLUSH 8 /* do not delay flush of dirty metadata */
#define CEPH_I_NOFLUSH 16 /* do not flush dirty caps */
-static inline void ceph_i_clear(struct inode *inode, unsigned mask)
+static inline void __ceph_dir_set_complete(struct ceph_inode_info *ci,
+ int release_count)
{
- struct ceph_inode_info *ci = ceph_inode(inode);
-
- spin_lock(&inode->i_lock);
- ci->i_ceph_flags &= ~mask;
- spin_unlock(&inode->i_lock);
+ atomic_set(&ci->i_complete_count, release_count);
}
-static inline void ceph_i_set(struct inode *inode, unsigned mask)
+static inline void __ceph_dir_clear_complete(struct ceph_inode_info *ci)
{
- struct ceph_inode_info *ci = ceph_inode(inode);
+ atomic_inc(&ci->i_release_count);
+}
- spin_lock(&inode->i_lock);
- ci->i_ceph_flags |= mask;
- spin_unlock(&inode->i_lock);
+static inline bool __ceph_dir_is_complete(struct ceph_inode_info *ci)
+{
+ return atomic_read(&ci->i_complete_count) ==
+ atomic_read(&ci->i_release_count);
}
-static inline bool ceph_i_test(struct inode *inode, unsigned mask)
+static inline void ceph_dir_clear_complete(struct inode *inode)
{
- struct ceph_inode_info *ci = ceph_inode(inode);
- bool r;
+ __ceph_dir_clear_complete(ceph_inode(inode));
+}
- spin_lock(&inode->i_lock);
- r = (ci->i_ceph_flags & mask) == mask;
- spin_unlock(&inode->i_lock);
- return r;
+static inline bool ceph_dir_is_complete(struct inode *inode)
+{
+ return __ceph_dir_is_complete(ceph_inode(inode));
}
@@ -429,13 +487,6 @@ static inline loff_t ceph_make_fpos(unsigned frag, unsigned off)
return ((loff_t)frag << 32) | (loff_t)off;
}
-static inline int ceph_set_ino_cb(struct inode *inode, void *data)
-{
- ceph_inode(inode)->i_vino = *(struct ceph_vino *)data;
- inode->i_ino = ceph_vino_to_ino(*(struct ceph_vino *)data);
- return 0;
-}
-
/*
* caps helpers
*/
@@ -452,9 +503,9 @@ extern int __ceph_caps_issued_other(struct ceph_inode_info *ci,
static inline int ceph_caps_issued(struct ceph_inode_info *ci)
{
int issued;
- spin_lock(&ci->vfs_inode.i_lock);
+ spin_lock(&ci->i_ceph_lock);
issued = __ceph_caps_issued(ci, NULL);
- spin_unlock(&ci->vfs_inode.i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return issued;
}
@@ -462,9 +513,9 @@ static inline int ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask,
int touch)
{
int r;
- spin_lock(&ci->vfs_inode.i_lock);
+ spin_lock(&ci->i_ceph_lock);
r = __ceph_caps_issued_mask(ci, mask, touch);
- spin_unlock(&ci->vfs_inode.i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return r;
}
@@ -472,8 +523,10 @@ static inline int __ceph_caps_dirty(struct ceph_inode_info *ci)
{
return ci->i_dirty_caps | ci->i_flushing_caps;
}
-extern void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask);
+extern int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask);
+extern int __ceph_caps_revoking_other(struct ceph_inode_info *ci,
+ struct ceph_cap *ocap, int mask);
extern int ceph_caps_revoking(struct ceph_inode_info *ci, int mask);
extern int __ceph_caps_used(struct ceph_inode_info *ci);
@@ -496,7 +549,7 @@ extern int __ceph_caps_mds_wanted(struct ceph_inode_info *ci);
extern void ceph_caps_init(struct ceph_mds_client *mdsc);
extern void ceph_caps_finalize(struct ceph_mds_client *mdsc);
extern void ceph_adjust_min_caps(struct ceph_mds_client *mdsc, int delta);
-extern int ceph_reserve_caps(struct ceph_mds_client *mdsc,
+extern void ceph_reserve_caps(struct ceph_mds_client *mdsc,
struct ceph_cap_reservation *ctx, int need);
extern int ceph_unreserve_caps(struct ceph_mds_client *mdsc,
struct ceph_cap_reservation *ctx);
@@ -504,34 +557,28 @@ extern void ceph_reservation_status(struct ceph_fs_client *client,
int *total, int *avail, int *used,
int *reserved, int *min);
-static inline struct ceph_fs_client *ceph_inode_to_client(struct inode *inode)
-{
- return (struct ceph_fs_client *)inode->i_sb->s_fs_info;
-}
-
-static inline struct ceph_fs_client *ceph_sb_to_client(struct super_block *sb)
-{
- return (struct ceph_fs_client *)sb->s_fs_info;
-}
/*
* we keep buffered readdir results attached to file->private_data
*/
+#define CEPH_F_SYNC 1
+#define CEPH_F_ATEND 2
+
struct ceph_file_info {
- int fmode; /* initialized on open */
+ short fmode; /* initialized on open */
+ short flags; /* CEPH_F_* */
/* readdir: position within the dir */
u32 frag;
struct ceph_mds_request *last_readdir;
- int at_end;
/* readdir: position within a frag */
unsigned offset; /* offset of last chunk, adjusted for . and .. */
- u64 next_offset; /* offset of next chunk (last_name's + 1) */
+ unsigned next_offset; /* offset of next chunk (last_name's + 1) */
char *last_name; /* last entry in previous chunk */
struct dentry *dentry; /* next dentry (for dcache readdir) */
- unsigned long dir_release_count;
+ int dir_release_count;
/* used for -o dirstat read() on directory thing */
char *dir_info;
@@ -559,9 +606,9 @@ struct ceph_snap_realm {
u64 parent_since; /* snapid when our current parent became so */
u64 *prior_parent_snaps; /* snaps inherited from any parents we */
- int num_prior_parent_snaps; /* had prior to parent_since */
+ u32 num_prior_parent_snaps; /* had prior to parent_since */
u64 *snaps; /* snaps specific to this realm */
- int num_snaps;
+ u32 num_snaps;
struct ceph_snap_realm *parent;
struct list_head children; /* list of child realms */
@@ -642,6 +689,7 @@ extern const struct inode_operations ceph_file_iops;
extern struct inode *ceph_alloc_inode(struct super_block *sb);
extern void ceph_destroy_inode(struct inode *inode);
+extern int ceph_drop_inode(struct inode *inode);
extern struct inode *ceph_get_inode(struct super_block *sb,
struct ceph_vino vino);
@@ -675,35 +723,75 @@ extern int ceph_getattr(struct vfsmount *mnt, struct dentry *dentry,
/* xattr.c */
extern int ceph_setxattr(struct dentry *, const char *, const void *,
size_t, int);
+int __ceph_setxattr(struct dentry *, const char *, const void *, size_t, int);
+ssize_t __ceph_getxattr(struct inode *, const char *, void *, size_t);
+int __ceph_removexattr(struct dentry *, const char *);
extern ssize_t ceph_getxattr(struct dentry *, const char *, void *, size_t);
extern ssize_t ceph_listxattr(struct dentry *, char *, size_t);
extern int ceph_removexattr(struct dentry *, const char *);
extern void __ceph_build_xattrs_blob(struct ceph_inode_info *ci);
extern void __ceph_destroy_xattrs(struct ceph_inode_info *ci);
+extern void __init ceph_xattr_init(void);
+extern void ceph_xattr_exit(void);
+
+/* acl.c */
+extern const struct xattr_handler *ceph_xattr_handlers[];
+
+#ifdef CONFIG_CEPH_FS_POSIX_ACL
+
+struct posix_acl *ceph_get_acl(struct inode *, int);
+int ceph_set_acl(struct inode *inode, struct posix_acl *acl, int type);
+int ceph_init_acl(struct dentry *, struct inode *, struct inode *);
+
+static inline void ceph_forget_all_cached_acls(struct inode *inode)
+{
+ forget_all_cached_acls(inode);
+}
+
+#else
+
+#define ceph_get_acl NULL
+#define ceph_set_acl NULL
+
+static inline int ceph_init_acl(struct dentry *dentry, struct inode *inode,
+ struct inode *dir)
+{
+ return 0;
+}
+
+static inline int ceph_acl_chmod(struct dentry *dentry, struct inode *inode)
+{
+ return 0;
+}
+
+static inline void ceph_forget_all_cached_acls(struct inode *inode)
+{
+}
+
+#endif
/* caps.c */
extern const char *ceph_cap_string(int c);
extern void ceph_handle_caps(struct ceph_mds_session *session,
struct ceph_msg *msg);
-extern int ceph_add_cap(struct inode *inode,
- struct ceph_mds_session *session, u64 cap_id,
- int fmode, unsigned issued, unsigned wanted,
- unsigned cap, unsigned seq, u64 realmino, int flags,
- struct ceph_cap_reservation *caps_reservation);
-extern void __ceph_remove_cap(struct ceph_cap *cap);
-static inline void ceph_remove_cap(struct ceph_cap *cap)
-{
- struct inode *inode = &cap->ci->vfs_inode;
- spin_lock(&inode->i_lock);
- __ceph_remove_cap(cap);
- spin_unlock(&inode->i_lock);
-}
+extern struct ceph_cap *ceph_get_cap(struct ceph_mds_client *mdsc,
+ struct ceph_cap_reservation *ctx);
+extern void ceph_add_cap(struct inode *inode,
+ struct ceph_mds_session *session, u64 cap_id,
+ int fmode, unsigned issued, unsigned wanted,
+ unsigned cap, unsigned seq, u64 realmino, int flags,
+ struct ceph_cap **new_cap);
+extern void __ceph_remove_cap(struct ceph_cap *cap, bool queue_release);
extern void ceph_put_cap(struct ceph_mds_client *mdsc,
struct ceph_cap *cap);
+extern int ceph_is_any_caps(struct inode *inode);
+extern void __queue_cap_release(struct ceph_mds_session *session, u64 ino,
+ u64 cap_id, u32 migrate_seq, u32 issue_seq);
extern void ceph_queue_caps_release(struct inode *inode);
extern int ceph_write_inode(struct inode *inode, struct writeback_control *wbc);
-extern int ceph_fsync(struct file *file, int datasync);
+extern int ceph_fsync(struct file *file, loff_t start, loff_t end,
+ int datasync);
extern void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc,
struct ceph_mds_session *session);
extern struct ceph_cap *ceph_get_cap_for_mds(struct ceph_inode_info *ci,
@@ -743,17 +831,11 @@ extern int ceph_mmap(struct file *file, struct vm_area_struct *vma);
/* file.c */
extern const struct file_operations ceph_file_fops;
extern const struct address_space_operations ceph_aops;
-extern int ceph_copy_to_page_vector(struct page **pages,
- const char *data,
- loff_t off, size_t len);
-extern int ceph_copy_from_page_vector(struct page **pages,
- char *data,
- loff_t off, size_t len);
-extern struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags);
+
extern int ceph_open(struct inode *inode, struct file *file);
-extern struct dentry *ceph_lookup_open(struct inode *dir, struct dentry *dentry,
- struct nameidata *nd, int mode,
- int locked_dir);
+extern int ceph_atomic_open(struct inode *dir, struct dentry *dentry,
+ struct file *file, unsigned flags, umode_t mode,
+ int *opened);
extern int ceph_release(struct inode *inode, struct file *filp);
/* dir.c */
@@ -763,6 +845,8 @@ extern const struct dentry_operations ceph_dentry_ops, ceph_snap_dentry_ops,
ceph_snapdir_dentry_ops;
extern int ceph_handle_notrace_create(struct inode *dir, struct dentry *dentry);
+extern int ceph_handle_snapdir(struct ceph_mds_request *req,
+ struct dentry *dentry, int err);
extern struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
struct dentry *dentry, int err);
@@ -770,6 +854,8 @@ extern void ceph_dentry_lru_add(struct dentry *dn);
extern void ceph_dentry_lru_touch(struct dentry *dn);
extern void ceph_dentry_lru_del(struct dentry *dn);
extern void ceph_invalidate_dentry_lease(struct dentry *dentry);
+extern unsigned ceph_dentry_hash(struct inode *dir, struct dentry *dn);
+extern struct inode *ceph_get_dentry_parent_inode(struct dentry *dentry);
/*
* our d_ops vary depending on whether the inode is live,
@@ -785,21 +871,19 @@ extern long ceph_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
extern const struct export_operations ceph_export_ops;
/* locks.c */
+extern __init void ceph_flock_init(void);
extern int ceph_lock(struct file *file, int cmd, struct file_lock *fl);
extern int ceph_flock(struct file *file, int cmd, struct file_lock *fl);
extern void ceph_count_locks(struct inode *inode, int *p_num, int *f_num);
-extern int ceph_encode_locks(struct inode *i, struct ceph_pagelist *p,
- int p_locks, int f_locks);
+extern int ceph_encode_locks_to_buffer(struct inode *inode,
+ struct ceph_filelock *flocks,
+ int num_fcntl_locks,
+ int num_flock_locks);
+extern int ceph_locks_to_pagelist(struct ceph_filelock *flocks,
+ struct ceph_pagelist *pagelist,
+ int num_fcntl_locks, int num_flock_locks);
extern int lock_to_ceph_filelock(struct file_lock *fl, struct ceph_filelock *c);
-static inline struct inode *get_dentry_parent_inode(struct dentry *dentry)
-{
- if (dentry && dentry->d_parent)
- return dentry->d_parent->d_inode;
-
- return NULL;
-}
-
/* debugfs.c */
extern int ceph_fs_debugfs_init(struct ceph_fs_client *client);
extern void ceph_fs_debugfs_cleanup(struct ceph_fs_client *client);
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index 6e12a6ba5f7..c9c2b887381 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -6,13 +6,33 @@
#include <linux/ceph/decode.h>
#include <linux/xattr.h>
+#include <linux/posix_acl_xattr.h>
#include <linux/slab.h>
+#define XATTR_CEPH_PREFIX "ceph."
+#define XATTR_CEPH_PREFIX_LEN (sizeof (XATTR_CEPH_PREFIX) - 1)
+
+static int __remove_xattr(struct ceph_inode_info *ci,
+ struct ceph_inode_xattr *xattr);
+
+/*
+ * List of handlers for synthetic system.* attributes. Other
+ * attributes are handled directly.
+ */
+const struct xattr_handler *ceph_xattr_handlers[] = {
+#ifdef CONFIG_CEPH_FS_POSIX_ACL
+ &posix_acl_access_xattr_handler,
+ &posix_acl_default_xattr_handler,
+#endif
+ NULL,
+};
+
static bool ceph_is_valid_xattr(const char *name)
{
- return !strncmp(name, "ceph.", 5) ||
+ return !strncmp(name, XATTR_CEPH_PREFIX, XATTR_CEPH_PREFIX_LEN) ||
!strncmp(name, XATTR_SECURITY_PREFIX,
XATTR_SECURITY_PREFIX_LEN) ||
+ !strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN) ||
!strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN) ||
!strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN);
}
@@ -21,101 +41,235 @@ static bool ceph_is_valid_xattr(const char *name)
* These define virtual xattrs exposing the recursive directory
* statistics and layout metadata.
*/
-struct ceph_vxattr_cb {
- bool readonly;
+struct ceph_vxattr {
char *name;
+ size_t name_size; /* strlen(name) + 1 (for '\0') */
size_t (*getxattr_cb)(struct ceph_inode_info *ci, char *val,
size_t size);
+ bool readonly, hidden;
+ bool (*exists_cb)(struct ceph_inode_info *ci);
};
+/* layouts */
+
+static bool ceph_vxattrcb_layout_exists(struct ceph_inode_info *ci)
+{
+ size_t s;
+ char *p = (char *)&ci->i_layout;
+
+ for (s = 0; s < sizeof(ci->i_layout); s++, p++)
+ if (*p)
+ return true;
+ return false;
+}
+
+static size_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val,
+ size_t size)
+{
+ int ret;
+ struct ceph_fs_client *fsc = ceph_sb_to_client(ci->vfs_inode.i_sb);
+ struct ceph_osd_client *osdc = &fsc->client->osdc;
+ s64 pool = ceph_file_layout_pg_pool(ci->i_layout);
+ const char *pool_name;
+ char buf[128];
+
+ dout("ceph_vxattrcb_layout %p\n", &ci->vfs_inode);
+ down_read(&osdc->map_sem);
+ pool_name = ceph_pg_pool_name_by_id(osdc->osdmap, pool);
+ if (pool_name) {
+ size_t len = strlen(pool_name);
+ ret = snprintf(buf, sizeof(buf),
+ "stripe_unit=%lld stripe_count=%lld object_size=%lld pool=",
+ (unsigned long long)ceph_file_layout_su(ci->i_layout),
+ (unsigned long long)ceph_file_layout_stripe_count(ci->i_layout),
+ (unsigned long long)ceph_file_layout_object_size(ci->i_layout));
+ if (!size) {
+ ret += len;
+ } else if (ret + len > size) {
+ ret = -ERANGE;
+ } else {
+ memcpy(val, buf, ret);
+ memcpy(val + ret, pool_name, len);
+ ret += len;
+ }
+ } else {
+ ret = snprintf(buf, sizeof(buf),
+ "stripe_unit=%lld stripe_count=%lld object_size=%lld pool=%lld",
+ (unsigned long long)ceph_file_layout_su(ci->i_layout),
+ (unsigned long long)ceph_file_layout_stripe_count(ci->i_layout),
+ (unsigned long long)ceph_file_layout_object_size(ci->i_layout),
+ (unsigned long long)pool);
+ if (size) {
+ if (ret <= size)
+ memcpy(val, buf, ret);
+ else
+ ret = -ERANGE;
+ }
+ }
+ up_read(&osdc->map_sem);
+ return ret;
+}
+
+static size_t ceph_vxattrcb_layout_stripe_unit(struct ceph_inode_info *ci,
+ char *val, size_t size)
+{
+ return snprintf(val, size, "%lld",
+ (unsigned long long)ceph_file_layout_su(ci->i_layout));
+}
+
+static size_t ceph_vxattrcb_layout_stripe_count(struct ceph_inode_info *ci,
+ char *val, size_t size)
+{
+ return snprintf(val, size, "%lld",
+ (unsigned long long)ceph_file_layout_stripe_count(ci->i_layout));
+}
+
+static size_t ceph_vxattrcb_layout_object_size(struct ceph_inode_info *ci,
+ char *val, size_t size)
+{
+ return snprintf(val, size, "%lld",
+ (unsigned long long)ceph_file_layout_object_size(ci->i_layout));
+}
+
+static size_t ceph_vxattrcb_layout_pool(struct ceph_inode_info *ci,
+ char *val, size_t size)
+{
+ int ret;
+ struct ceph_fs_client *fsc = ceph_sb_to_client(ci->vfs_inode.i_sb);
+ struct ceph_osd_client *osdc = &fsc->client->osdc;
+ s64 pool = ceph_file_layout_pg_pool(ci->i_layout);
+ const char *pool_name;
+
+ down_read(&osdc->map_sem);
+ pool_name = ceph_pg_pool_name_by_id(osdc->osdmap, pool);
+ if (pool_name)
+ ret = snprintf(val, size, "%s", pool_name);
+ else
+ ret = snprintf(val, size, "%lld", (unsigned long long)pool);
+ up_read(&osdc->map_sem);
+ return ret;
+}
+
/* directories */
-static size_t ceph_vxattrcb_entries(struct ceph_inode_info *ci, char *val,
+static size_t ceph_vxattrcb_dir_entries(struct ceph_inode_info *ci, char *val,
size_t size)
{
return snprintf(val, size, "%lld", ci->i_files + ci->i_subdirs);
}
-static size_t ceph_vxattrcb_files(struct ceph_inode_info *ci, char *val,
+static size_t ceph_vxattrcb_dir_files(struct ceph_inode_info *ci, char *val,
size_t size)
{
return snprintf(val, size, "%lld", ci->i_files);
}
-static size_t ceph_vxattrcb_subdirs(struct ceph_inode_info *ci, char *val,
+static size_t ceph_vxattrcb_dir_subdirs(struct ceph_inode_info *ci, char *val,
size_t size)
{
return snprintf(val, size, "%lld", ci->i_subdirs);
}
-static size_t ceph_vxattrcb_rentries(struct ceph_inode_info *ci, char *val,
+static size_t ceph_vxattrcb_dir_rentries(struct ceph_inode_info *ci, char *val,
size_t size)
{
return snprintf(val, size, "%lld", ci->i_rfiles + ci->i_rsubdirs);
}
-static size_t ceph_vxattrcb_rfiles(struct ceph_inode_info *ci, char *val,
+static size_t ceph_vxattrcb_dir_rfiles(struct ceph_inode_info *ci, char *val,
size_t size)
{
return snprintf(val, size, "%lld", ci->i_rfiles);
}
-static size_t ceph_vxattrcb_rsubdirs(struct ceph_inode_info *ci, char *val,
+static size_t ceph_vxattrcb_dir_rsubdirs(struct ceph_inode_info *ci, char *val,
size_t size)
{
return snprintf(val, size, "%lld", ci->i_rsubdirs);
}
-static size_t ceph_vxattrcb_rbytes(struct ceph_inode_info *ci, char *val,
+static size_t ceph_vxattrcb_dir_rbytes(struct ceph_inode_info *ci, char *val,
size_t size)
{
return snprintf(val, size, "%lld", ci->i_rbytes);
}
-static size_t ceph_vxattrcb_rctime(struct ceph_inode_info *ci, char *val,
+static size_t ceph_vxattrcb_dir_rctime(struct ceph_inode_info *ci, char *val,
size_t size)
{
- return snprintf(val, size, "%ld.%ld", (long)ci->i_rctime.tv_sec,
+ return snprintf(val, size, "%ld.09%ld", (long)ci->i_rctime.tv_sec,
(long)ci->i_rctime.tv_nsec);
}
-static struct ceph_vxattr_cb ceph_dir_vxattrs[] = {
- { true, "ceph.dir.entries", ceph_vxattrcb_entries},
- { true, "ceph.dir.files", ceph_vxattrcb_files},
- { true, "ceph.dir.subdirs", ceph_vxattrcb_subdirs},
- { true, "ceph.dir.rentries", ceph_vxattrcb_rentries},
- { true, "ceph.dir.rfiles", ceph_vxattrcb_rfiles},
- { true, "ceph.dir.rsubdirs", ceph_vxattrcb_rsubdirs},
- { true, "ceph.dir.rbytes", ceph_vxattrcb_rbytes},
- { true, "ceph.dir.rctime", ceph_vxattrcb_rctime},
- { true, NULL, NULL }
-};
-/* files */
+#define CEPH_XATTR_NAME(_type, _name) XATTR_CEPH_PREFIX #_type "." #_name
+#define CEPH_XATTR_NAME2(_type, _name, _name2) \
+ XATTR_CEPH_PREFIX #_type "." #_name "." #_name2
-static size_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val,
- size_t size)
-{
- int ret;
+#define XATTR_NAME_CEPH(_type, _name) \
+ { \
+ .name = CEPH_XATTR_NAME(_type, _name), \
+ .name_size = sizeof (CEPH_XATTR_NAME(_type, _name)), \
+ .getxattr_cb = ceph_vxattrcb_ ## _type ## _ ## _name, \
+ .readonly = true, \
+ .hidden = false, \
+ .exists_cb = NULL, \
+ }
+#define XATTR_LAYOUT_FIELD(_type, _name, _field) \
+ { \
+ .name = CEPH_XATTR_NAME2(_type, _name, _field), \
+ .name_size = sizeof (CEPH_XATTR_NAME2(_type, _name, _field)), \
+ .getxattr_cb = ceph_vxattrcb_ ## _name ## _ ## _field, \
+ .readonly = false, \
+ .hidden = true, \
+ .exists_cb = ceph_vxattrcb_layout_exists, \
+ }
- ret = snprintf(val, size,
- "chunk_bytes=%lld\nstripe_count=%lld\nobject_size=%lld\n",
- (unsigned long long)ceph_file_layout_su(ci->i_layout),
- (unsigned long long)ceph_file_layout_stripe_count(ci->i_layout),
- (unsigned long long)ceph_file_layout_object_size(ci->i_layout));
- if (ceph_file_layout_pg_preferred(ci->i_layout))
- ret += snprintf(val + ret, size, "preferred_osd=%lld\n",
- (unsigned long long)ceph_file_layout_pg_preferred(
- ci->i_layout));
- return ret;
-}
+static struct ceph_vxattr ceph_dir_vxattrs[] = {
+ {
+ .name = "ceph.dir.layout",
+ .name_size = sizeof("ceph.dir.layout"),
+ .getxattr_cb = ceph_vxattrcb_layout,
+ .readonly = false,
+ .hidden = true,
+ .exists_cb = ceph_vxattrcb_layout_exists,
+ },
+ XATTR_LAYOUT_FIELD(dir, layout, stripe_unit),
+ XATTR_LAYOUT_FIELD(dir, layout, stripe_count),
+ XATTR_LAYOUT_FIELD(dir, layout, object_size),
+ XATTR_LAYOUT_FIELD(dir, layout, pool),
+ XATTR_NAME_CEPH(dir, entries),
+ XATTR_NAME_CEPH(dir, files),
+ XATTR_NAME_CEPH(dir, subdirs),
+ XATTR_NAME_CEPH(dir, rentries),
+ XATTR_NAME_CEPH(dir, rfiles),
+ XATTR_NAME_CEPH(dir, rsubdirs),
+ XATTR_NAME_CEPH(dir, rbytes),
+ XATTR_NAME_CEPH(dir, rctime),
+ { .name = NULL, 0 } /* Required table terminator */
+};
+static size_t ceph_dir_vxattrs_name_size; /* total size of all names */
+
+/* files */
-static struct ceph_vxattr_cb ceph_file_vxattrs[] = {
- { true, "ceph.layout", ceph_vxattrcb_layout},
- { NULL, NULL }
+static struct ceph_vxattr ceph_file_vxattrs[] = {
+ {
+ .name = "ceph.file.layout",
+ .name_size = sizeof("ceph.file.layout"),
+ .getxattr_cb = ceph_vxattrcb_layout,
+ .readonly = false,
+ .hidden = true,
+ .exists_cb = ceph_vxattrcb_layout_exists,
+ },
+ XATTR_LAYOUT_FIELD(file, layout, stripe_unit),
+ XATTR_LAYOUT_FIELD(file, layout, stripe_count),
+ XATTR_LAYOUT_FIELD(file, layout, object_size),
+ XATTR_LAYOUT_FIELD(file, layout, pool),
+ { .name = NULL, 0 } /* Required table terminator */
};
+static size_t ceph_file_vxattrs_name_size; /* total size of all names */
-static struct ceph_vxattr_cb *ceph_inode_vxattrs(struct inode *inode)
+static struct ceph_vxattr *ceph_inode_vxattrs(struct inode *inode)
{
if (S_ISDIR(inode->i_mode))
return ceph_dir_vxattrs;
@@ -124,22 +278,67 @@ static struct ceph_vxattr_cb *ceph_inode_vxattrs(struct inode *inode)
return NULL;
}
-static struct ceph_vxattr_cb *ceph_match_vxattr(struct ceph_vxattr_cb *vxattr,
+static size_t ceph_vxattrs_name_size(struct ceph_vxattr *vxattrs)
+{
+ if (vxattrs == ceph_dir_vxattrs)
+ return ceph_dir_vxattrs_name_size;
+ if (vxattrs == ceph_file_vxattrs)
+ return ceph_file_vxattrs_name_size;
+ BUG();
+
+ return 0;
+}
+
+/*
+ * Compute the aggregate size (including terminating '\0') of all
+ * virtual extended attribute names in the given vxattr table.
+ */
+static size_t __init vxattrs_name_size(struct ceph_vxattr *vxattrs)
+{
+ struct ceph_vxattr *vxattr;
+ size_t size = 0;
+
+ for (vxattr = vxattrs; vxattr->name; vxattr++)
+ if (!vxattr->hidden)
+ size += vxattr->name_size;
+
+ return size;
+}
+
+/* Routines called at initialization and exit time */
+
+void __init ceph_xattr_init(void)
+{
+ ceph_dir_vxattrs_name_size = vxattrs_name_size(ceph_dir_vxattrs);
+ ceph_file_vxattrs_name_size = vxattrs_name_size(ceph_file_vxattrs);
+}
+
+void ceph_xattr_exit(void)
+{
+ ceph_dir_vxattrs_name_size = 0;
+ ceph_file_vxattrs_name_size = 0;
+}
+
+static struct ceph_vxattr *ceph_match_vxattr(struct inode *inode,
const char *name)
{
- do {
- if (strcmp(vxattr->name, name) == 0)
- return vxattr;
- vxattr++;
- } while (vxattr->name);
+ struct ceph_vxattr *vxattr = ceph_inode_vxattrs(inode);
+
+ if (vxattr) {
+ while (vxattr->name) {
+ if (!strcmp(vxattr->name, name))
+ return vxattr;
+ vxattr++;
+ }
+ }
+
return NULL;
}
static int __set_xattr(struct ceph_inode_info *ci,
const char *name, int name_len,
const char *val, int val_len,
- int dirty,
- int should_free_name, int should_free_val,
+ int flags, int update_xattr,
struct ceph_inode_xattr **newxattr)
{
struct rb_node **p;
@@ -168,12 +367,31 @@ static int __set_xattr(struct ceph_inode_info *ci,
xattr = NULL;
}
+ if (update_xattr) {
+ int err = 0;
+ if (xattr && (flags & XATTR_CREATE))
+ err = -EEXIST;
+ else if (!xattr && (flags & XATTR_REPLACE))
+ err = -ENODATA;
+ if (err) {
+ kfree(name);
+ kfree(val);
+ return err;
+ }
+ if (update_xattr < 0) {
+ if (xattr)
+ __remove_xattr(ci, xattr);
+ kfree(name);
+ return 0;
+ }
+ }
+
if (!xattr) {
new = 1;
xattr = *newxattr;
xattr->name = name;
xattr->name_len = name_len;
- xattr->should_free_name = should_free_name;
+ xattr->should_free_name = update_xattr;
ci->i_xattrs.count++;
dout("__set_xattr count=%d\n", ci->i_xattrs.count);
@@ -183,7 +401,7 @@ static int __set_xattr(struct ceph_inode_info *ci,
if (xattr->should_free_val)
kfree((void *)xattr->val);
- if (should_free_name) {
+ if (update_xattr) {
kfree((void *)name);
name = xattr->name;
}
@@ -198,8 +416,8 @@ static int __set_xattr(struct ceph_inode_info *ci,
xattr->val = "";
xattr->val_len = val_len;
- xattr->dirty = dirty;
- xattr->should_free_val = (val && should_free_val);
+ xattr->dirty = update_xattr;
+ xattr->should_free_val = (val && update_xattr);
if (new) {
rb_link_node(&xattr->node, parent, p);
@@ -219,6 +437,7 @@ static struct ceph_inode_xattr *__get_xattr(struct ceph_inode_info *ci,
struct rb_node **p;
struct rb_node *parent = NULL;
struct ceph_inode_xattr *xattr = NULL;
+ int name_len = strlen(name);
int c;
p = &ci->i_xattrs.index.rb_node;
@@ -226,6 +445,8 @@ static struct ceph_inode_xattr *__get_xattr(struct ceph_inode_info *ci,
parent = *p;
xattr = rb_entry(parent, struct ceph_inode_xattr, node);
c = strncmp(name, xattr->name, xattr->name_len);
+ if (c == 0 && name_len > xattr->name_len)
+ c = 1;
if (c < 0)
p = &(*p)->rb_left;
else if (c > 0)
@@ -258,7 +479,7 @@ static int __remove_xattr(struct ceph_inode_info *ci,
struct ceph_inode_xattr *xattr)
{
if (!xattr)
- return -EOPNOTSUPP;
+ return -ENODATA;
rb_erase(&xattr->node, &ci->i_xattrs.index);
@@ -340,8 +561,8 @@ void __ceph_destroy_xattrs(struct ceph_inode_info *ci)
}
static int __build_xattrs(struct inode *inode)
- __releases(inode->i_lock)
- __acquires(inode->i_lock)
+ __releases(ci->i_ceph_lock)
+ __acquires(ci->i_ceph_lock)
{
u32 namelen;
u32 numattr = 0;
@@ -369,7 +590,7 @@ start:
end = p + ci->i_xattrs.blob->vec.iov_len;
ceph_decode_32_safe(&p, end, numattr, bad);
xattr_version = ci->i_xattrs.version;
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
xattrs = kcalloc(numattr, sizeof(struct ceph_xattr *),
GFP_NOFS);
@@ -384,12 +605,13 @@ start:
goto bad_lock;
}
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (ci->i_xattrs.version != xattr_version) {
/* lost a race, retry */
for (i = 0; i < numattr; i++)
kfree(xattrs[i]);
kfree(xattrs);
+ xattrs = NULL;
goto start;
}
err = -EIO;
@@ -403,7 +625,7 @@ start:
p += len;
err = __set_xattr(ci, name, namelen, val, len,
- 0, 0, 0, &xattrs[numattr]);
+ 0, 0, &xattrs[numattr]);
if (err < 0)
goto bad;
@@ -415,7 +637,7 @@ start:
return err;
bad_lock:
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
bad:
if (xattrs) {
for (i = 0; i < numattr; i++)
@@ -492,24 +714,25 @@ void __ceph_build_xattrs_blob(struct ceph_inode_info *ci)
}
}
-ssize_t ceph_getxattr(struct dentry *dentry, const char *name, void *value,
+ssize_t __ceph_getxattr(struct inode *inode, const char *name, void *value,
size_t size)
{
- struct inode *inode = dentry->d_inode;
struct ceph_inode_info *ci = ceph_inode(inode);
- struct ceph_vxattr_cb *vxattrs = ceph_inode_vxattrs(inode);
int err;
struct ceph_inode_xattr *xattr;
- struct ceph_vxattr_cb *vxattr = NULL;
+ struct ceph_vxattr *vxattr = NULL;
if (!ceph_is_valid_xattr(name))
return -ENODATA;
/* let's see if a virtual xattr was requested */
- if (vxattrs)
- vxattr = ceph_match_vxattr(vxattrs, name);
+ vxattr = ceph_match_vxattr(inode, name);
+ if (vxattr && !(vxattr->exists_cb && !vxattr->exists_cb(ci))) {
+ err = vxattr->getxattr_cb(ci, value, size);
+ return err;
+ }
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
dout("getxattr %p ver=%lld index_ver=%lld\n", inode,
ci->i_xattrs.version, ci->i_xattrs.index_version);
@@ -517,19 +740,14 @@ ssize_t ceph_getxattr(struct dentry *dentry, const char *name, void *value,
(ci->i_xattrs.index_version >= ci->i_xattrs.version)) {
goto get_xattr;
} else {
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
/* get xattrs from mds (if we don't already have them) */
err = ceph_do_getattr(inode, CEPH_STAT_CAP_XATTR);
if (err)
return err;
}
- spin_lock(&inode->i_lock);
-
- if (vxattr && vxattr->readonly) {
- err = vxattr->getxattr_cb(ci, value, size);
- goto out;
- }
+ spin_lock(&ci->i_ceph_lock);
err = __build_xattrs(inode);
if (err < 0)
@@ -538,11 +756,8 @@ ssize_t ceph_getxattr(struct dentry *dentry, const char *name, void *value,
get_xattr:
err = -ENODATA; /* == ENOATTR */
xattr = __get_xattr(ci, name);
- if (!xattr) {
- if (vxattr)
- err = vxattr->getxattr_cb(ci, value, size);
+ if (!xattr)
goto out;
- }
err = -ERANGE;
if (size && size < xattr->val_len)
@@ -555,22 +770,31 @@ get_xattr:
memcpy(value, xattr->val, xattr->val_len);
out:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return err;
}
+ssize_t ceph_getxattr(struct dentry *dentry, const char *name, void *value,
+ size_t size)
+{
+ if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN))
+ return generic_getxattr(dentry, name, value, size);
+
+ return __ceph_getxattr(dentry->d_inode, name, value, size);
+}
+
ssize_t ceph_listxattr(struct dentry *dentry, char *names, size_t size)
{
struct inode *inode = dentry->d_inode;
struct ceph_inode_info *ci = ceph_inode(inode);
- struct ceph_vxattr_cb *vxattrs = ceph_inode_vxattrs(inode);
+ struct ceph_vxattr *vxattrs = ceph_inode_vxattrs(inode);
u32 vir_namelen = 0;
u32 namelen;
int err;
u32 len;
int i;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
dout("listxattr %p ver=%lld index_ver=%lld\n", inode,
ci->i_xattrs.version, ci->i_xattrs.index_version);
@@ -578,45 +802,53 @@ ssize_t ceph_listxattr(struct dentry *dentry, char *names, size_t size)
(ci->i_xattrs.index_version >= ci->i_xattrs.version)) {
goto list_xattr;
} else {
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
err = ceph_do_getattr(inode, CEPH_STAT_CAP_XATTR);
if (err)
return err;
}
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
err = __build_xattrs(inode);
if (err < 0)
goto out;
list_xattr:
- vir_namelen = 0;
- /* include virtual dir xattrs */
- if (vxattrs)
- for (i = 0; vxattrs[i].name; i++)
- vir_namelen += strlen(vxattrs[i].name) + 1;
+ /*
+ * Start with virtual dir xattr names (if any) (including
+ * terminating '\0' characters for each).
+ */
+ vir_namelen = ceph_vxattrs_name_size(vxattrs);
+
/* adding 1 byte per each variable due to the null termination */
- namelen = vir_namelen + ci->i_xattrs.names_size + ci->i_xattrs.count;
+ namelen = ci->i_xattrs.names_size + ci->i_xattrs.count;
err = -ERANGE;
- if (size && namelen > size)
+ if (size && vir_namelen + namelen > size)
goto out;
- err = namelen;
+ err = namelen + vir_namelen;
if (size == 0)
goto out;
names = __copy_xattr_names(ci, names);
/* virtual xattr names, too */
- if (vxattrs)
+ err = namelen;
+ if (vxattrs) {
for (i = 0; vxattrs[i].name; i++) {
- len = sprintf(names, "%s", vxattrs[i].name);
- names += len + 1;
+ if (!vxattrs[i].hidden &&
+ !(vxattrs[i].exists_cb &&
+ !vxattrs[i].exists_cb(ci))) {
+ len = sprintf(names, "%s", vxattrs[i].name);
+ names += len + 1;
+ err += len + 1;
+ }
}
+ }
out:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
return err;
}
@@ -626,7 +858,6 @@ static int ceph_sync_setxattr(struct dentry *dentry, const char *name,
struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
struct inode *inode = dentry->d_inode;
struct ceph_inode_info *ci = ceph_inode(inode);
- struct inode *parent_inode = dentry->d_parent->d_inode;
struct ceph_mds_request *req;
struct ceph_mds_client *mdsc = fsc->mdsc;
int err;
@@ -655,6 +886,9 @@ static int ceph_sync_setxattr(struct dentry *dentry, const char *name,
dout("setxattr value=%.*s\n", (int)size, value);
+ if (!value)
+ flags |= CEPH_XATTR_REMOVE;
+
/* do request */
req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SETXATTR,
USE_AUTH_MDS);
@@ -662,7 +896,8 @@ static int ceph_sync_setxattr(struct dentry *dentry, const char *name,
err = PTR_ERR(req);
goto out;
}
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_inode_drop = CEPH_CAP_XATTR_SHARED;
req->r_num_caps = 1;
req->r_args.setxattr.flags = cpu_to_le32(flags);
@@ -673,7 +908,7 @@ static int ceph_sync_setxattr(struct dentry *dentry, const char *name,
req->r_data_len = size;
dout("xattr.ver (before): %lld\n", ci->i_xattrs.version);
- err = ceph_mdsc_do_request(mdsc, parent_inode, req);
+ err = ceph_mdsc_do_request(mdsc, NULL, req);
ceph_mdsc_put_request(req);
dout("xattr.ver (after): %lld\n", ci->i_xattrs.version);
@@ -686,33 +921,32 @@ out:
return err;
}
-int ceph_setxattr(struct dentry *dentry, const char *name,
- const void *value, size_t size, int flags)
+int __ceph_setxattr(struct dentry *dentry, const char *name,
+ const void *value, size_t size, int flags)
{
struct inode *inode = dentry->d_inode;
+ struct ceph_vxattr *vxattr;
struct ceph_inode_info *ci = ceph_inode(inode);
- struct ceph_vxattr_cb *vxattrs = ceph_inode_vxattrs(inode);
+ int issued;
int err;
+ int dirty = 0;
int name_len = strlen(name);
int val_len = size;
char *newname = NULL;
char *newval = NULL;
struct ceph_inode_xattr *xattr = NULL;
- int issued;
int required_blob_size;
- if (ceph_snap(inode) != CEPH_NOSNAP)
- return -EROFS;
-
if (!ceph_is_valid_xattr(name))
return -EOPNOTSUPP;
- if (vxattrs) {
- struct ceph_vxattr_cb *vxattr =
- ceph_match_vxattr(vxattrs, name);
- if (vxattr && vxattr->readonly)
- return -EOPNOTSUPP;
- }
+ vxattr = ceph_match_vxattr(inode, name);
+ if (vxattr && vxattr->readonly)
+ return -EOPNOTSUPP;
+
+ /* pass any unhandled ceph.* xattrs through to the MDS */
+ if (!strncmp(name, XATTR_CEPH_PREFIX, XATTR_CEPH_PREFIX_LEN))
+ goto do_sync_unlocked;
/* preallocate memory for xattr name, value, index node */
err = -ENOMEM;
@@ -721,20 +955,19 @@ int ceph_setxattr(struct dentry *dentry, const char *name,
goto out;
if (val_len) {
- newval = kmalloc(val_len + 1, GFP_NOFS);
+ newval = kmemdup(value, val_len, GFP_NOFS);
if (!newval)
goto out;
- memcpy(newval, value, val_len);
- newval[val_len] = '\0';
}
xattr = kmalloc(sizeof(struct ceph_inode_xattr), GFP_NOFS);
if (!xattr)
goto out;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
retry:
issued = __ceph_caps_issued(ci, NULL);
+ dout("setxattr %p issued %s\n", inode, ceph_cap_string(issued));
if (!(issued & CEPH_CAP_XATTR_EXCL))
goto do_sync;
__build_xattrs(inode);
@@ -743,32 +976,37 @@ retry:
if (!ci->i_xattrs.prealloc_blob ||
required_blob_size > ci->i_xattrs.prealloc_blob->alloc_len) {
- struct ceph_buffer *blob = NULL;
+ struct ceph_buffer *blob;
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
dout(" preaallocating new blob size=%d\n", required_blob_size);
blob = ceph_buffer_new(required_blob_size, GFP_NOFS);
if (!blob)
goto out;
- spin_lock(&inode->i_lock);
+ spin_lock(&ci->i_ceph_lock);
if (ci->i_xattrs.prealloc_blob)
ceph_buffer_put(ci->i_xattrs.prealloc_blob);
ci->i_xattrs.prealloc_blob = blob;
goto retry;
}
- dout("setxattr %p issued %s\n", inode, ceph_cap_string(issued));
- err = __set_xattr(ci, newname, name_len, newval,
- val_len, 1, 1, 1, &xattr);
- __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
- ci->i_xattrs.dirty = true;
- inode->i_ctime = CURRENT_TIME;
- spin_unlock(&inode->i_lock);
+ err = __set_xattr(ci, newname, name_len, newval, val_len,
+ flags, value ? 1 : -1, &xattr);
+
+ if (!err) {
+ dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
+ ci->i_xattrs.dirty = true;
+ inode->i_ctime = CURRENT_TIME;
+ }
+ spin_unlock(&ci->i_ceph_lock);
+ if (dirty)
+ __mark_inode_dirty(inode, dirty);
return err;
do_sync:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
+do_sync_unlocked:
err = ceph_sync_setxattr(dentry, name, value, size, flags);
out:
kfree(newname);
@@ -777,12 +1015,23 @@ out:
return err;
}
+int ceph_setxattr(struct dentry *dentry, const char *name,
+ const void *value, size_t size, int flags)
+{
+ if (ceph_snap(dentry->d_inode) != CEPH_NOSNAP)
+ return -EROFS;
+
+ if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN))
+ return generic_setxattr(dentry, name, value, size, flags);
+
+ return __ceph_setxattr(dentry, name, value, size, flags);
+}
+
static int ceph_send_removexattr(struct dentry *dentry, const char *name)
{
struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
struct ceph_mds_client *mdsc = fsc->mdsc;
struct inode *inode = dentry->d_inode;
- struct inode *parent_inode = dentry->d_parent->d_inode;
struct ceph_mds_request *req;
int err;
@@ -790,56 +1039,90 @@ static int ceph_send_removexattr(struct dentry *dentry, const char *name)
USE_AUTH_MDS);
if (IS_ERR(req))
return PTR_ERR(req);
- req->r_inode = igrab(inode);
+ req->r_inode = inode;
+ ihold(inode);
req->r_inode_drop = CEPH_CAP_XATTR_SHARED;
req->r_num_caps = 1;
req->r_path2 = kstrdup(name, GFP_NOFS);
- err = ceph_mdsc_do_request(mdsc, parent_inode, req);
+ err = ceph_mdsc_do_request(mdsc, NULL, req);
ceph_mdsc_put_request(req);
return err;
}
-int ceph_removexattr(struct dentry *dentry, const char *name)
+int __ceph_removexattr(struct dentry *dentry, const char *name)
{
struct inode *inode = dentry->d_inode;
+ struct ceph_vxattr *vxattr;
struct ceph_inode_info *ci = ceph_inode(inode);
- struct ceph_vxattr_cb *vxattrs = ceph_inode_vxattrs(inode);
int issued;
int err;
-
- if (ceph_snap(inode) != CEPH_NOSNAP)
- return -EROFS;
+ int required_blob_size;
+ int dirty;
if (!ceph_is_valid_xattr(name))
return -EOPNOTSUPP;
- if (vxattrs) {
- struct ceph_vxattr_cb *vxattr =
- ceph_match_vxattr(vxattrs, name);
- if (vxattr && vxattr->readonly)
- return -EOPNOTSUPP;
- }
+ vxattr = ceph_match_vxattr(inode, name);
+ if (vxattr && vxattr->readonly)
+ return -EOPNOTSUPP;
- spin_lock(&inode->i_lock);
- __build_xattrs(inode);
+ /* pass any unhandled ceph.* xattrs through to the MDS */
+ if (!strncmp(name, XATTR_CEPH_PREFIX, XATTR_CEPH_PREFIX_LEN))
+ goto do_sync_unlocked;
+
+ err = -ENOMEM;
+ spin_lock(&ci->i_ceph_lock);
+retry:
issued = __ceph_caps_issued(ci, NULL);
dout("removexattr %p issued %s\n", inode, ceph_cap_string(issued));
if (!(issued & CEPH_CAP_XATTR_EXCL))
goto do_sync;
+ __build_xattrs(inode);
+
+ required_blob_size = __get_required_blob_size(ci, 0, 0);
+
+ if (!ci->i_xattrs.prealloc_blob ||
+ required_blob_size > ci->i_xattrs.prealloc_blob->alloc_len) {
+ struct ceph_buffer *blob;
+
+ spin_unlock(&ci->i_ceph_lock);
+ dout(" preaallocating new blob size=%d\n", required_blob_size);
+ blob = ceph_buffer_new(required_blob_size, GFP_NOFS);
+ if (!blob)
+ goto out;
+ spin_lock(&ci->i_ceph_lock);
+ if (ci->i_xattrs.prealloc_blob)
+ ceph_buffer_put(ci->i_xattrs.prealloc_blob);
+ ci->i_xattrs.prealloc_blob = blob;
+ goto retry;
+ }
err = __remove_xattr_by_name(ceph_inode(inode), name);
- __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
+
+ dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
ci->i_xattrs.dirty = true;
inode->i_ctime = CURRENT_TIME;
-
- spin_unlock(&inode->i_lock);
-
+ spin_unlock(&ci->i_ceph_lock);
+ if (dirty)
+ __mark_inode_dirty(inode, dirty);
return err;
do_sync:
- spin_unlock(&inode->i_lock);
+ spin_unlock(&ci->i_ceph_lock);
+do_sync_unlocked:
err = ceph_send_removexattr(dentry, name);
+out:
return err;
}
+int ceph_removexattr(struct dentry *dentry, const char *name)
+{
+ if (ceph_snap(dentry->d_inode) != CEPH_NOSNAP)
+ return -EROFS;
+
+ if (!strncmp(name, XATTR_SYSTEM_PREFIX, XATTR_SYSTEM_PREFIX_LEN))
+ return generic_removexattr(dentry, name);
+
+ return __ceph_removexattr(dentry, name);
+}
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-04 08:22:42 +0000