Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client: (22 commits)
  ceph: do not carry i_lock for readdir from dcache
  fs/ceph/xattr.c: Use kmemdup
  rbd: passing wrong variable to bvec_kunmap_irq()
  rbd: null vs ERR_PTR
  ceph: fix num_pages_free accounting in pagelist
  ceph: add CEPH_MDS_OP_SETDIRLAYOUT and associated ioctl.
  ceph: don't crash when passed bad mount options
  ceph: fix debugfs warnings
  block: rbd: removing unnecessary test
  block: rbd: fixed may leaks
  ceph: switch from BKL to lock_flocks()
  ceph: preallocate flock state without locks held
  ceph: add pagelist_reserve, pagelist_truncate, pagelist_set_cursor
  ceph: use mapping->nrpages to determine if mapping is empty
  ceph: only invalidate on check_caps if we actually have pages
  ceph: do not hide .snap in root directory
  rbd: introduce rados block device (rbd), based on libceph
  ceph: factor out libceph from Ceph file system
  ceph-rbd: osdc support for osd call and rollback operations
  ceph: messenger and osdc changes for rbd
  ...

MAINTAINERS

@@ -1527,6 +1527,8 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client.git
 S:	Supported
 F:	Documentation/filesystems/ceph.txt
 F:	fs/ceph
+F:	net/ceph
+F:	include/linux/ceph
 
 CERTIFIED WIRELESS USB (WUSB) SUBSYSTEM:
 M:	David Vrabel <david.vrabel@csr.com>
@@ -4805,6 +4807,15 @@ F:	fs/qnx4/
 F:	include/linux/qnx4_fs.h
 F:	include/linux/qnxtypes.h
 
+RADOS BLOCK DEVICE (RBD)
+F:	include/linux/qnxtypes.h
+M:	Yehuda Sadeh <yehuda@hq.newdream.net>
+M:	Sage Weil <sage@newdream.net>
+M:	ceph-devel@vger.kernel.org
+S:	Supported
+F:	drivers/block/rbd.c
+F:	drivers/block/rbd_types.h
+
 RADEON FRAMEBUFFER DISPLAY DRIVER
 M:	Benjamin Herrenschmidt <benh@kernel.crashing.org>
 L:	linux-fbdev@vger.kernel.org

drivers/block/Kconfig

@@ -488,4 +488,21 @@ config BLK_DEV_HD
 
 	  If unsure, say N.
 
+config BLK_DEV_RBD
+	tristate "Rados block device (RBD)"
+	depends on INET && EXPERIMENTAL && BLOCK
+	select CEPH_LIB
+	select LIBCRC32C
+	select CRYPTO_AES
+	select CRYPTO
+	default n
+	help
+	  Say Y here if you want include the Rados block device, which stripes
+	  a block device over objects stored in the Ceph distributed object
+	  store.
+
+	  More information at http://ceph.newdream.net/.
+
+	  If unsure, say N.
+
 endif # BLK_DEV

drivers/block/Makefile

@@ -37,5 +37,6 @@ obj-$(CONFIG_BLK_DEV_HD)	+= hd.o
 
 obj-$(CONFIG_XEN_BLKDEV_FRONTEND)	+= xen-blkfront.o
 obj-$(CONFIG_BLK_DEV_DRBD)     += drbd/
+obj-$(CONFIG_BLK_DEV_RBD)     += rbd.o
 
 swim_mod-objs	:= swim.o swim_asm.o

drivers/block/rbd_types.h

@@ -0,0 +1,73 @@
+/*
+ * Ceph - scalable distributed file system
+ *
+ * Copyright (C) 2004-2010 Sage Weil <sage@newdream.net>
+ *
+ * This is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License version 2.1, as published by the Free Software
+ * Foundation.  See file COPYING.
+ *
+ */
+
+#ifndef CEPH_RBD_TYPES_H
+#define CEPH_RBD_TYPES_H
+
+#include <linux/types.h>
+
+/*
+ * rbd image 'foo' consists of objects
+ *   foo.rbd      - image metadata
+ *   foo.00000000
+ *   foo.00000001
+ *   ...          - data
+ */
+
+#define RBD_SUFFIX		".rbd"
+#define RBD_DIRECTORY           "rbd_directory"
+#define RBD_INFO                "rbd_info"
+
+#define RBD_DEFAULT_OBJ_ORDER	22   /* 4MB */
+#define RBD_MIN_OBJ_ORDER       16
+#define RBD_MAX_OBJ_ORDER       30
+
+#define RBD_MAX_OBJ_NAME_LEN	96
+#define RBD_MAX_SEG_NAME_LEN	128
+
+#define RBD_COMP_NONE		0
+#define RBD_CRYPT_NONE		0
+
+#define RBD_HEADER_TEXT		"<<< Rados Block Device Image >>>\n"
+#define RBD_HEADER_SIGNATURE	"RBD"
+#define RBD_HEADER_VERSION	"001.005"
+
+struct rbd_info {
+	__le64 max_id;
+} __attribute__ ((packed));
+
+struct rbd_image_snap_ondisk {
+	__le64 id;
+	__le64 image_size;
+} __attribute__((packed));
+
+struct rbd_image_header_ondisk {
+	char text[40];
+	char block_name[24];
+	char signature[4];
+	char version[8];
+	struct {
+		__u8 order;
+		__u8 crypt_type;
+		__u8 comp_type;
+		__u8 unused;
+	} __attribute__((packed)) options;
+	__le64 image_size;
+	__le64 snap_seq;
+	__le32 snap_count;
+	__le32 reserved;
+	__le64 snap_names_len;
+	struct rbd_image_snap_ondisk snaps[0];
+} __attribute__((packed));
+
+
+#endif

fs/ceph/Kconfig

@@ -1,9 +1,11 @@
 config CEPH_FS
         tristate "Ceph distributed file system (EXPERIMENTAL)"
 	depends on INET && EXPERIMENTAL
+	select CEPH_LIB
 	select LIBCRC32C
 	select CRYPTO_AES
 	select CRYPTO
+	default n
 	help
 	  Choose Y or M here to include support for mounting the
 	  experimental Ceph distributed file system.  Ceph is an extremely
@@ -14,15 +16,3 @@ config CEPH_FS
 
 	  If unsure, say N.
 
-config CEPH_FS_PRETTYDEBUG
-	bool "Include file:line in ceph debug output"
-	depends on CEPH_FS
-	default n
-	help
-	  If you say Y here, debug output will include a filename and
-	  line to aid debugging.  This icnreases kernel size and slows
-	  execution slightly when debug call sites are enabled (e.g.,
-	  via CONFIG_DYNAMIC_DEBUG).
-
-	  If unsure, say N.
-

fs/ceph/Makefile

@@ -8,15 +8,8 @@ obj-$(CONFIG_CEPH_FS) += ceph.o
 
 ceph-objs := super.o inode.o dir.o file.o locks.o addr.o ioctl.o \
 	export.o caps.o snap.o xattr.o \
-	messenger.o msgpool.o buffer.o pagelist.o \
+	mds_client.o mdsmap.o strings.o ceph_frag.o \
-	mds_client.o mdsmap.o \
+	debugfs.o
-	mon_client.o \
-	osd_client.o osdmap.o crush/crush.o crush/mapper.o crush/hash.o \
-	debugfs.o \
-	auth.o auth_none.o \
-	crypto.o armor.o \
-	auth_x.o \
-	ceph_fs.o ceph_strings.o ceph_hash.o ceph_frag.o
 
 else
 #Otherwise we were called directly from the command

fs/ceph/README

@@ -1,20 +0,0 @@
-#
-# The following files are shared by (and manually synchronized
-# between) the Ceph userland and kernel client.
-#
-# userland                  kernel
-src/include/ceph_fs.h	    fs/ceph/ceph_fs.h
-src/include/ceph_fs.cc	    fs/ceph/ceph_fs.c
-src/include/msgr.h	    fs/ceph/msgr.h
-src/include/rados.h	    fs/ceph/rados.h
-src/include/ceph_strings.cc fs/ceph/ceph_strings.c
-src/include/ceph_frag.h	    fs/ceph/ceph_frag.h
-src/include/ceph_frag.cc    fs/ceph/ceph_frag.c
-src/include/ceph_hash.h	    fs/ceph/ceph_hash.h
-src/include/ceph_hash.cc    fs/ceph/ceph_hash.c
-src/crush/crush.c	    fs/ceph/crush/crush.c
-src/crush/crush.h	    fs/ceph/crush/crush.h
-src/crush/mapper.c	    fs/ceph/crush/mapper.c
-src/crush/mapper.h	    fs/ceph/crush/mapper.h
-src/crush/hash.h	    fs/ceph/crush/hash.h
-src/crush/hash.c	    fs/ceph/crush/hash.c

fs/ceph/addr.c

@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/backing-dev.h>
 #include <linux/fs.h>
@@ -10,7 +10,8 @@
 #include <linux/task_io_accounting_ops.h>
 
 #include "super.h"
-#include "osd_client.h"
+#include "mds_client.h"
+#include <linux/ceph/osd_client.h>
 
 /*
  * Ceph address space ops.
@@ -193,7 +194,8 @@ static int readpage_nounlock(struct file *filp, struct page *page)
 {
 	struct inode *inode = filp->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_inode_to_client(inode)->osdc;
+	struct ceph_osd_client *osdc = 
+		&ceph_inode_to_client(inode)->client->osdc;
 	int err = 0;
 	u64 len = PAGE_CACHE_SIZE;
 
@@ -265,7 +267,8 @@ static int ceph_readpages(struct file *file, struct address_space *mapping,
 {
 	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)->osdc;
+	struct ceph_osd_client *osdc =
+		&ceph_inode_to_client(inode)->client->osdc;
 	int rc = 0;
 	struct page **pages;
 	loff_t offset;
@@ -365,7 +368,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 {
 	struct inode *inode;
 	struct ceph_inode_info *ci;
-	struct ceph_client *client;
+	struct ceph_fs_client *fsc;
 	struct ceph_osd_client *osdc;
 	loff_t page_off = page->index << PAGE_CACHE_SHIFT;
 	int len = PAGE_CACHE_SIZE;
@@ -383,8 +386,8 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 	}
 	inode = page->mapping->host;
 	ci = ceph_inode(inode);
-	client = ceph_inode_to_client(inode);
+	fsc = ceph_inode_to_client(inode);
-	osdc = &client->osdc;
+	osdc = &fsc->client->osdc;
 
 	/* verify this is a writeable snap context */
 	snapc = (void *)page->private;
@@ -414,10 +417,10 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
 	dout("writepage %p page %p index %lu on %llu~%u snapc %p\n",
 	     inode, page, page->index, page_off, len, snapc);
 
-	writeback_stat = atomic_long_inc_return(&client->writeback_count);
+	writeback_stat = atomic_long_inc_return(&fsc->writeback_count);
 	if (writeback_stat >
-	    CONGESTION_ON_THRESH(client->mount_args->congestion_kb))
+	    CONGESTION_ON_THRESH(fsc->mount_options->congestion_kb))
-		set_bdi_congested(&client->backing_dev_info, BLK_RW_ASYNC);
+		set_bdi_congested(&fsc->backing_dev_info, BLK_RW_ASYNC);
 
 	set_page_writeback(page);
 	err = ceph_osdc_writepages(osdc, ceph_vino(inode),
@@ -496,7 +499,7 @@ static void writepages_finish(struct ceph_osd_request *req,
 	struct address_space *mapping = inode->i_mapping;
 	__s32 rc = -EIO;
 	u64 bytes = 0;
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	long writeback_stat;
 	unsigned issued = ceph_caps_issued(ci);
 
@@ -529,10 +532,10 @@ static void writepages_finish(struct ceph_osd_request *req,
 		WARN_ON(!PageUptodate(page));
 
 		writeback_stat =
-			atomic_long_dec_return(&client->writeback_count);
+			atomic_long_dec_return(&fsc->writeback_count);
 		if (writeback_stat <
-		    CONGESTION_OFF_THRESH(client->mount_args->congestion_kb))
+		    CONGESTION_OFF_THRESH(fsc->mount_options->congestion_kb))
-			clear_bdi_congested(&client->backing_dev_info,
+			clear_bdi_congested(&fsc->backing_dev_info,
 					    BLK_RW_ASYNC);
 
 		ceph_put_snap_context((void *)page->private);
@@ -569,13 +572,13 @@ static void writepages_finish(struct ceph_osd_request *req,
  * 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_client *client,
+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(client->wb_pagevec_pool, GFP_NOFS);
+		req->r_pages = mempool_alloc(fsc->wb_pagevec_pool, GFP_NOFS);
 		req->r_pages_from_pool = 1;
 		WARN_ON(!req->r_pages);
 	}
@@ -590,7 +593,7 @@ static int ceph_writepages_start(struct address_space *mapping,
 	struct inode *inode = mapping->host;
 	struct backing_dev_info *bdi = mapping->backing_dev_info;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client;
+	struct ceph_fs_client *fsc;
 	pgoff_t index, start, end;
 	int range_whole = 0;
 	int should_loop = 1;
@@ -617,13 +620,13 @@ static int ceph_writepages_start(struct address_space *mapping,
 	     wbc->sync_mode == WB_SYNC_NONE ? "NONE" :
 	     (wbc->sync_mode == WB_SYNC_ALL ? "ALL" : "HOLD"));
 
-	client = ceph_inode_to_client(inode);
+	fsc = ceph_inode_to_client(inode);
-	if (client->mount_state == CEPH_MOUNT_SHUTDOWN) {
+	if (fsc->mount_state == CEPH_MOUNT_SHUTDOWN) {
 		pr_warning("writepage_start %p on forced umount\n", inode);
 		return -EIO; /* we're in a forced umount, don't write! */
 	}
-	if (client->mount_args->wsize && client->mount_args->wsize < wsize)
+	if (fsc->mount_options->wsize && fsc->mount_options->wsize < wsize)
-		wsize = client->mount_args->wsize;
+		wsize = fsc->mount_options->wsize;
 	if (wsize < PAGE_CACHE_SIZE)
 		wsize = PAGE_CACHE_SIZE;
 	max_pages_ever = wsize >> PAGE_CACHE_SHIFT;
@@ -769,7 +772,7 @@ get_more_pages:
 				offset = (unsigned long long)page->index
 					<< PAGE_CACHE_SHIFT;
 				len = wsize;
-				req = ceph_osdc_new_request(&client->osdc,
+				req = ceph_osdc_new_request(&fsc->client->osdc,
 					    &ci->i_layout,
 					    ceph_vino(inode),
 					    offset, &len,
@@ -782,7 +785,7 @@ get_more_pages:
 					    &inode->i_mtime, true, 1);
 				max_pages = req->r_num_pages;
 
-				alloc_page_vec(client, req);
+				alloc_page_vec(fsc, req);
 				req->r_callback = writepages_finish;
 				req->r_inode = inode;
 			}
@@ -794,10 +797,10 @@ get_more_pages:
 			     inode, page, page->index);
 
 			writeback_stat =
-			       atomic_long_inc_return(&client->writeback_count);
+			       atomic_long_inc_return(&fsc->writeback_count);
 			if (writeback_stat > CONGESTION_ON_THRESH(
-				    client->mount_args->congestion_kb)) {
+				    fsc->mount_options->congestion_kb)) {
-				set_bdi_congested(&client->backing_dev_info,
+				set_bdi_congested(&fsc->backing_dev_info,
 						  BLK_RW_ASYNC);
 			}
 
@@ -846,7 +849,7 @@ get_more_pages:
 		op->payload_len = cpu_to_le32(len);
 		req->r_request->hdr.data_len = cpu_to_le32(len);
 
-		ceph_osdc_start_request(&client->osdc, req, true);
+		ceph_osdc_start_request(&fsc->client->osdc, req, true);
 		req = NULL;
 
 		/* continue? */
@@ -915,7 +918,7 @@ static int ceph_update_writeable_page(struct file *file,
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	loff_t page_off = pos & PAGE_CACHE_MASK;
 	int pos_in_page = pos & ~PAGE_CACHE_MASK;
 	int end_in_page = pos_in_page + len;
@@ -1053,8 +1056,8 @@ static int ceph_write_end(struct file *file, struct address_space *mapping,
 			  struct page *page, void *fsdata)
 {
 	struct inode *inode = file->f_dentry->d_inode;
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	unsigned from = pos & (PAGE_CACHE_SIZE - 1);
 	int check_cap = 0;
 
@@ -1123,7 +1126,7 @@ 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 ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	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;

fs/ceph/caps.c

@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/fs.h>
 #include <linux/kernel.h>
@@ -9,8 +9,9 @@
 #include <linux/writeback.h>
 
 #include "super.h"
-#include "decode.h"
+#include "mds_client.h"
-#include "messenger.h"
+#include <linux/ceph/decode.h>
+#include <linux/ceph/messenger.h>
 
 /*
  * Capability management
@@ -287,11 +288,11 @@ void ceph_put_cap(struct ceph_mds_client *mdsc, struct ceph_cap *cap)
 	spin_unlock(&mdsc->caps_list_lock);
 }
 
-void ceph_reservation_status(struct ceph_client *client,
+void ceph_reservation_status(struct ceph_fs_client *fsc,
 			     int *total, int *avail, int *used, int *reserved,
 			     int *min)
 {
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 
 	if (total)
 		*total = mdsc->caps_total_count;
@@ -399,7 +400,7 @@ static void __insert_cap_node(struct ceph_inode_info *ci,
 static void __cap_set_timeouts(struct ceph_mds_client *mdsc,
 			       struct ceph_inode_info *ci)
 {
-	struct ceph_mount_args *ma = mdsc->client->mount_args;
+	struct ceph_mount_options *ma = mdsc->fsc->mount_options;
 
 	ci->i_hold_caps_min = round_jiffies(jiffies +
 					    ma->caps_wanted_delay_min * HZ);
@@ -515,7 +516,7 @@ int ceph_add_cap(struct inode *inode,
 		 unsigned seq, unsigned mseq, u64 realmino, int flags,
 		 struct ceph_cap_reservation *caps_reservation)
 {
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	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;
@@ -873,7 +874,7 @@ void __ceph_remove_cap(struct ceph_cap *cap)
 	struct ceph_mds_session *session = cap->session;
 	struct ceph_inode_info *ci = cap->ci;
 	struct ceph_mds_client *mdsc =
-		&ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
+		ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
 	int removed = 0;
 
 	dout("__ceph_remove_cap %p from %p\n", cap, &ci->vfs_inode);
@@ -1210,7 +1211,7 @@ void __ceph_flush_snaps(struct ceph_inode_info *ci,
 	int mds;
 	struct ceph_cap_snap *capsnap;
 	u32 mseq;
-	struct ceph_mds_client *mdsc = &ceph_inode_to_client(inode)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc;
 	struct ceph_mds_session *session = NULL; /* if session != NULL, we hold
 						    session->s_mutex */
 	u64 next_follows = 0;  /* keep track of how far we've gotten through the
@@ -1336,7 +1337,7 @@ static void ceph_flush_snaps(struct ceph_inode_info *ci)
 void __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;
+		ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
 	struct inode *inode = &ci->vfs_inode;
 	int was = ci->i_dirty_caps;
 	int dirty = 0;
@@ -1378,7 +1379,7 @@ void __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask)
 static int __mark_caps_flushing(struct inode *inode,
 				 struct ceph_mds_session *session)
 {
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int flushing;
 
@@ -1416,17 +1417,6 @@ static int __mark_caps_flushing(struct inode *inode,
 /*
  * try to invalidate mapping pages without blocking.
  */
-static int mapping_is_empty(struct address_space *mapping)
-{
-	struct page *page = find_get_page(mapping, 0);
-
-	if (!page)
-		return 1;
-
-	put_page(page);
-	return 0;
-}
-
 static int try_nonblocking_invalidate(struct inode *inode)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
@@ -1436,7 +1426,7 @@ static int try_nonblocking_invalidate(struct inode *inode)
 	invalidate_mapping_pages(&inode->i_data, 0, -1);
 	spin_lock(&inode->i_lock);
 
-	if (mapping_is_empty(&inode->i_data) &&
+	if (inode->i_data.nrpages == 0 &&
 	    invalidating_gen == ci->i_rdcache_gen) {
 		/* success. */
 		dout("try_nonblocking_invalidate %p success\n", inode);
@@ -1462,8 +1452,8 @@ static int try_nonblocking_invalidate(struct inode *inode)
 void ceph_check_caps(struct ceph_inode_info *ci, int flags,
 		     struct ceph_mds_session *session)
 {
-	struct ceph_client *client = ceph_inode_to_client(&ci->vfs_inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(&ci->vfs_inode);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct inode *inode = &ci->vfs_inode;
 	struct ceph_cap *cap;
 	int file_wanted, used;
@@ -1533,7 +1523,7 @@ retry_locked:
 	 */
 	if ((!is_delayed || mdsc->stopping) &&
 	    ci->i_wrbuffer_ref == 0 &&               /* no dirty pages... */
-	    ci->i_rdcache_gen &&                     /* may have cached pages */
+	    inode->i_data.nrpages &&                 /* have cached pages */
 	    (file_wanted == 0 ||                     /* no open files */
 	     (revoking & (CEPH_CAP_FILE_CACHE|
 			  CEPH_CAP_FILE_LAZYIO))) && /*  or revoking cache */
@@ -1706,7 +1696,7 @@ ack:
 static int try_flush_caps(struct inode *inode, struct ceph_mds_session *session,
 			  unsigned *flush_tid)
 {
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	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;
@@ -1872,7 +1862,7 @@ int ceph_write_inode(struct inode *inode, struct writeback_control *wbc)
 				       caps_are_flushed(inode, flush_tid));
 	} else {
 		struct ceph_mds_client *mdsc =
-			&ceph_sb_to_client(inode->i_sb)->mdsc;
+			ceph_sb_to_client(inode->i_sb)->mdsc;
 
 		spin_lock(&inode->i_lock);
 		if (__ceph_caps_dirty(ci))
@@ -2465,7 +2455,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid,
 	__releases(inode->i_lock)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 	unsigned seq = le32_to_cpu(m->seq);
 	int dirty = le32_to_cpu(m->dirty);
 	int cleaned = 0;
@@ -2713,7 +2703,7 @@ void ceph_handle_caps(struct ceph_mds_session *session,
 		      struct ceph_msg *msg)
 {
 	struct ceph_mds_client *mdsc = session->s_mdsc;
-	struct super_block *sb = mdsc->client->sb;
+	struct super_block *sb = mdsc->fsc->sb;
 	struct inode *inode;
 	struct ceph_cap *cap;
 	struct ceph_mds_caps *h;

fs/ceph/ceph_frag.c

@@ -1,7 +1,8 @@
 /*
  * Ceph 'frag' type
  */
-#include "types.h"
+#include <linux/module.h>
+#include <linux/ceph/types.h>
 
 int ceph_frag_compare(__u32 a, __u32 b)
 {

fs/ceph/debugfs.c

@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/device.h>
 #include <linux/slab.h>
@@ -7,143 +7,49 @@
 #include <linux/debugfs.h>
 #include <linux/seq_file.h>
 
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
+
 #include "super.h"
-#include "mds_client.h"
-#include "mon_client.h"
-#include "auth.h"
 
 #ifdef CONFIG_DEBUG_FS
 
-/*
+#include "mds_client.h"
- * Implement /sys/kernel/debug/ceph fun
- *
- * /sys/kernel/debug/ceph/client*  - an instance of the ceph client
- *      .../osdmap      - current osdmap
- *      .../mdsmap      - current mdsmap
- *      .../monmap      - current monmap
- *      .../osdc        - active osd requests
- *      .../mdsc        - active mds requests
- *      .../monc        - mon client state
- *      .../dentry_lru  - dump contents of dentry lru
- *      .../caps        - expose cap (reservation) stats
- *      .../bdi         - symlink to ../../bdi/something
- */
-
-static struct dentry *ceph_debugfs_dir;
-
-static int monmap_show(struct seq_file *s, void *p)
-{
-	int i;
-	struct ceph_client *client = s->private;
-
-	if (client->monc.monmap == NULL)
-		return 0;
-
-	seq_printf(s, "epoch %d\n", client->monc.monmap->epoch);
-	for (i = 0; i < client->monc.monmap->num_mon; i++) {
-		struct ceph_entity_inst *inst =
-			&client->monc.monmap->mon_inst[i];
-
-		seq_printf(s, "\t%s%lld\t%s\n",
-			   ENTITY_NAME(inst->name),
-			   pr_addr(&inst->addr.in_addr));
-	}
-	return 0;
-}
 
 static int mdsmap_show(struct seq_file *s, void *p)
 {
 	int i;
-	struct ceph_client *client = s->private;
+	struct ceph_fs_client *fsc = s->private;
 
-	if (client->mdsc.mdsmap == NULL)
+	if (fsc->mdsc == NULL || fsc->mdsc->mdsmap == NULL)
 		return 0;
-	seq_printf(s, "epoch %d\n", client->mdsc.mdsmap->m_epoch);
+	seq_printf(s, "epoch %d\n", fsc->mdsc->mdsmap->m_epoch);
-	seq_printf(s, "root %d\n", client->mdsc.mdsmap->m_root);
+	seq_printf(s, "root %d\n", fsc->mdsc->mdsmap->m_root);
 	seq_printf(s, "session_timeout %d\n",
-		       client->mdsc.mdsmap->m_session_timeout);
+		       fsc->mdsc->mdsmap->m_session_timeout);
 	seq_printf(s, "session_autoclose %d\n",
-		       client->mdsc.mdsmap->m_session_autoclose);
+		       fsc->mdsc->mdsmap->m_session_autoclose);
-	for (i = 0; i < client->mdsc.mdsmap->m_max_mds; i++) {
+	for (i = 0; i < fsc->mdsc->mdsmap->m_max_mds; i++) {
 		struct ceph_entity_addr *addr =
-			&client->mdsc.mdsmap->m_info[i].addr;
+			&fsc->mdsc->mdsmap->m_info[i].addr;
-		int state = client->mdsc.mdsmap->m_info[i].state;
+		int state = fsc->mdsc->mdsmap->m_info[i].state;
 
-		seq_printf(s, "\tmds%d\t%s\t(%s)\n", i, pr_addr(&addr->in_addr),
+		seq_printf(s, "\tmds%d\t%s\t(%s)\n", i,
+			       ceph_pr_addr(&addr->in_addr),
 			       ceph_mds_state_name(state));
 	}
 	return 0;
 }
 
-static int osdmap_show(struct seq_file *s, void *p)
+/*
-{
+ * mdsc debugfs
-	int i;
+ */
-	struct ceph_client *client = s->private;
-	struct rb_node *n;
-
-	if (client->osdc.osdmap == NULL)
-		return 0;
-	seq_printf(s, "epoch %d\n", client->osdc.osdmap->epoch);
-	seq_printf(s, "flags%s%s\n",
-		   (client->osdc.osdmap->flags & CEPH_OSDMAP_NEARFULL) ?
-		   " NEARFULL" : "",
-		   (client->osdc.osdmap->flags & CEPH_OSDMAP_FULL) ?
-		   " FULL" : "");
-	for (n = rb_first(&client->osdc.osdmap->pg_pools); n; n = rb_next(n)) {
-		struct ceph_pg_pool_info *pool =
-			rb_entry(n, struct ceph_pg_pool_info, node);
-		seq_printf(s, "pg_pool %d pg_num %d / %d, lpg_num %d / %d\n",
-			   pool->id, pool->v.pg_num, pool->pg_num_mask,
-			   pool->v.lpg_num, pool->lpg_num_mask);
-	}
-	for (i = 0; i < client->osdc.osdmap->max_osd; i++) {
-		struct ceph_entity_addr *addr =
-			&client->osdc.osdmap->osd_addr[i];
-		int state = client->osdc.osdmap->osd_state[i];
-		char sb[64];
-
-		seq_printf(s, "\tosd%d\t%s\t%3d%%\t(%s)\n",
-			   i, pr_addr(&addr->in_addr),
-			   ((client->osdc.osdmap->osd_weight[i]*100) >> 16),
-			   ceph_osdmap_state_str(sb, sizeof(sb), state));
-	}
-	return 0;
-}
-
-static int monc_show(struct seq_file *s, void *p)
-{
-	struct ceph_client *client = s->private;
-	struct ceph_mon_generic_request *req;
-	struct ceph_mon_client *monc = &client->monc;
-	struct rb_node *rp;
-
-	mutex_lock(&monc->mutex);
-
-	if (monc->have_mdsmap)
-		seq_printf(s, "have mdsmap %u\n", (unsigned)monc->have_mdsmap);
-	if (monc->have_osdmap)
-		seq_printf(s, "have osdmap %u\n", (unsigned)monc->have_osdmap);
-	if (monc->want_next_osdmap)
-		seq_printf(s, "want next osdmap\n");
-
-	for (rp = rb_first(&monc->generic_request_tree); rp; rp = rb_next(rp)) {
-		__u16 op;
-		req = rb_entry(rp, struct ceph_mon_generic_request, node);
-		op = le16_to_cpu(req->request->hdr.type);
-		if (op == CEPH_MSG_STATFS)
-			seq_printf(s, "%lld statfs\n", req->tid);
-		else
-			seq_printf(s, "%lld unknown\n", req->tid);
-	}
-
-	mutex_unlock(&monc->mutex);
-	return 0;
-}
-
 static int mdsc_show(struct seq_file *s, void *p)
 {
-	struct ceph_client *client = s->private;
+	struct ceph_fs_client *fsc = s->private;
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	struct rb_node *rp;
 	int pathlen;
@@ -214,61 +120,12 @@ static int mdsc_show(struct seq_file *s, void *p)
 	return 0;
 }
 
-static int osdc_show(struct seq_file *s, void *pp)
-{
-	struct ceph_client *client = s->private;
-	struct ceph_osd_client *osdc = &client->osdc;
-	struct rb_node *p;
-
-	mutex_lock(&osdc->request_mutex);
-	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
-		struct ceph_osd_request *req;
-		struct ceph_osd_request_head *head;
-		struct ceph_osd_op *op;
-		int num_ops;
-		int opcode, olen;
-		int i;
-
-		req = rb_entry(p, struct ceph_osd_request, r_node);
-
-		seq_printf(s, "%lld\tosd%d\t%d.%x\t", req->r_tid,
-			   req->r_osd ? req->r_osd->o_osd : -1,
-			   le32_to_cpu(req->r_pgid.pool),
-			   le16_to_cpu(req->r_pgid.ps));
-
-		head = req->r_request->front.iov_base;
-		op = (void *)(head + 1);
-
-		num_ops = le16_to_cpu(head->num_ops);
-		olen = le32_to_cpu(head->object_len);
-		seq_printf(s, "%.*s", olen,
-			   (const char *)(head->ops + num_ops));
-
-		if (req->r_reassert_version.epoch)
-			seq_printf(s, "\t%u'%llu",
-			   (unsigned)le32_to_cpu(req->r_reassert_version.epoch),
-			   le64_to_cpu(req->r_reassert_version.version));
-		else
-			seq_printf(s, "\t");
-
-		for (i = 0; i < num_ops; i++) {
-			opcode = le16_to_cpu(op->op);
-			seq_printf(s, "\t%s", ceph_osd_op_name(opcode));
-			op++;
-		}
-
-		seq_printf(s, "\n");
-	}
-	mutex_unlock(&osdc->request_mutex);
-	return 0;
-}
-
 static int caps_show(struct seq_file *s, void *p)
 {
-	struct ceph_client *client = s->private;
+	struct ceph_fs_client *fsc = s->private;
 	int total, avail, used, reserved, min;
 
-	ceph_reservation_status(client, &total, &avail, &used, &reserved, &min);
+	ceph_reservation_status(fsc, &total, &avail, &used, &reserved, &min);
 	seq_printf(s, "total\t\t%d\n"
 		   "avail\t\t%d\n"
 		   "used\t\t%d\n"
@@ -280,8 +137,8 @@ static int caps_show(struct seq_file *s, void *p)
 
 static int dentry_lru_show(struct seq_file *s, void *ptr)
 {
-	struct ceph_client *client = s->private;
+	struct ceph_fs_client *fsc = s->private;
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_dentry_info *di;
 
 	spin_lock(&mdsc->dentry_lru_lock);
@@ -295,199 +152,124 @@ static int dentry_lru_show(struct seq_file *s, void *ptr)
 	return 0;
 }
 
-#define DEFINE_SHOW_FUNC(name)						\
+CEPH_DEFINE_SHOW_FUNC(mdsmap_show)
-static int name##_open(struct inode *inode, struct file *file)		\
+CEPH_DEFINE_SHOW_FUNC(mdsc_show)
-{									\
+CEPH_DEFINE_SHOW_FUNC(caps_show)
-	struct seq_file *sf;						\
+CEPH_DEFINE_SHOW_FUNC(dentry_lru_show)
-	int ret;							\
-									\
-	ret = single_open(file, name, NULL);				\
-	sf = file->private_data;					\
-	sf->private = inode->i_private;					\
-	return ret;							\
-}									\
-									\
-static const struct file_operations name##_fops = {			\
-	.open		= name##_open,					\
-	.read		= seq_read,					\
-	.llseek		= seq_lseek,					\
-	.release	= single_release,				\
-};
 
-DEFINE_SHOW_FUNC(monmap_show)
-DEFINE_SHOW_FUNC(mdsmap_show)
-DEFINE_SHOW_FUNC(osdmap_show)
-DEFINE_SHOW_FUNC(monc_show)
-DEFINE_SHOW_FUNC(mdsc_show)
-DEFINE_SHOW_FUNC(osdc_show)
-DEFINE_SHOW_FUNC(dentry_lru_show)
-DEFINE_SHOW_FUNC(caps_show)
 
+/*
+ * debugfs
+ */
 static int congestion_kb_set(void *data, u64 val)
 {
-	struct ceph_client *client = (struct ceph_client *)data;
+	struct ceph_fs_client *fsc = (struct ceph_fs_client *)data;
-
-	if (client)
-		client->mount_args->congestion_kb = (int)val;
 
+	fsc->mount_options->congestion_kb = (int)val;
 	return 0;
 }
 
 static int congestion_kb_get(void *data, u64 *val)
 {
-	struct ceph_client *client = (struct ceph_client *)data;
+	struct ceph_fs_client *fsc = (struct ceph_fs_client *)data;
-
-	if (client)
-		*val = (u64)client->mount_args->congestion_kb;
 
+	*val = (u64)fsc->mount_options->congestion_kb;
 	return 0;
 }
 
-
 DEFINE_SIMPLE_ATTRIBUTE(congestion_kb_fops, congestion_kb_get,
 			congestion_kb_set, "%llu\n");
 
-int __init ceph_debugfs_init(void)
+
+void ceph_fs_debugfs_cleanup(struct ceph_fs_client *fsc)
 {
-	ceph_debugfs_dir = debugfs_create_dir("ceph", NULL);
+	dout("ceph_fs_debugfs_cleanup\n");
-	if (!ceph_debugfs_dir)
+	debugfs_remove(fsc->debugfs_bdi);
-		return -ENOMEM;
+	debugfs_remove(fsc->debugfs_congestion_kb);
-	return 0;
+	debugfs_remove(fsc->debugfs_mdsmap);
+	debugfs_remove(fsc->debugfs_caps);
+	debugfs_remove(fsc->debugfs_mdsc);
+	debugfs_remove(fsc->debugfs_dentry_lru);
 }
 
-void ceph_debugfs_cleanup(void)
+int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
 {
-	debugfs_remove(ceph_debugfs_dir);
+	char name[100];
-}
+	int err = -ENOMEM;
 
-int ceph_debugfs_client_init(struct ceph_client *client)
+	dout("ceph_fs_debugfs_init\n");
-{
+	fsc->debugfs_congestion_kb =
-	int ret = 0;
-	char name[80];
-
-	snprintf(name, sizeof(name), "%pU.client%lld", &client->fsid,
-		 client->monc.auth->global_id);
-
-	client->debugfs_dir = debugfs_create_dir(name, ceph_debugfs_dir);
-	if (!client->debugfs_dir)
-		goto out;
-
-	client->monc.debugfs_file = debugfs_create_file("monc",
-						      0600,
-						      client->debugfs_dir,
-						      client,
-						      &monc_show_fops);
-	if (!client->monc.debugfs_file)
-		goto out;
-
-	client->mdsc.debugfs_file = debugfs_create_file("mdsc",
-						      0600,
-						      client->debugfs_dir,
-						      client,
-						      &mdsc_show_fops);
-	if (!client->mdsc.debugfs_file)
-		goto out;
-
-	client->osdc.debugfs_file = debugfs_create_file("osdc",
-						      0600,
-						      client->debugfs_dir,
-						      client,
-						      &osdc_show_fops);
-	if (!client->osdc.debugfs_file)
-		goto out;
-
-	client->debugfs_monmap = debugfs_create_file("monmap",
-					0600,
-					client->debugfs_dir,
-					client,
-					&monmap_show_fops);
-	if (!client->debugfs_monmap)
-		goto out;
-
-	client->debugfs_mdsmap = debugfs_create_file("mdsmap",
-					0600,
-					client->debugfs_dir,
-					client,
-					&mdsmap_show_fops);
-	if (!client->debugfs_mdsmap)
-		goto out;
-
-	client->debugfs_osdmap = debugfs_create_file("osdmap",
-					0600,
-					client->debugfs_dir,
-					client,
-					&osdmap_show_fops);
-	if (!client->debugfs_osdmap)
-		goto out;
-
-	client->debugfs_dentry_lru = debugfs_create_file("dentry_lru",
-					0600,
-					client->debugfs_dir,
-					client,
-					&dentry_lru_show_fops);
-	if (!client->debugfs_dentry_lru)
-		goto out;
-
-	client->debugfs_caps = debugfs_create_file("caps",
-						   0400,
-						   client->debugfs_dir,
-						   client,
-						   &caps_show_fops);
-	if (!client->debugfs_caps)
-		goto out;
-
-	client->debugfs_congestion_kb =
 		debugfs_create_file("writeback_congestion_kb",
 				    0600,
-				    client->debugfs_dir,
+				    fsc->client->debugfs_dir,
-				    client,
+				    fsc,
 				    &congestion_kb_fops);
-	if (!client->debugfs_congestion_kb)
+	if (!fsc->debugfs_congestion_kb)
 		goto out;
 
-	sprintf(name, "../../bdi/%s", dev_name(client->sb->s_bdi->dev));
+	dout("a\n");
-	client->debugfs_bdi = debugfs_create_symlink("bdi", client->debugfs_dir,
+
-						     name);
+	snprintf(name, sizeof(name), "../../bdi/%s",
+		 dev_name(fsc->backing_dev_info.dev));
+	fsc->debugfs_bdi =
+		debugfs_create_symlink("bdi",
+				       fsc->client->debugfs_dir,
+				       name);
+	if (!fsc->debugfs_bdi)
+		goto out;
+
+	dout("b\n");
+	fsc->debugfs_mdsmap = debugfs_create_file("mdsmap",
+					0600,
+					fsc->client->debugfs_dir,
+					fsc,
+					&mdsmap_show_fops);
+	if (!fsc->debugfs_mdsmap)
+		goto out;
+
+	dout("ca\n");
+	fsc->debugfs_mdsc = debugfs_create_file("mdsc",
+						0600,
+						fsc->client->debugfs_dir,
+						fsc,
+						&mdsc_show_fops);
+	if (!fsc->debugfs_mdsc)
+		goto out;
+
+	dout("da\n");
+	fsc->debugfs_caps = debugfs_create_file("caps",
+						   0400,
+						   fsc->client->debugfs_dir,
+						   fsc,
+						   &caps_show_fops);
+	if (!fsc->debugfs_caps)
+		goto out;
+
+	dout("ea\n");
+	fsc->debugfs_dentry_lru = debugfs_create_file("dentry_lru",
+					0600,
+					fsc->client->debugfs_dir,
+					fsc,
+					&dentry_lru_show_fops);
+	if (!fsc->debugfs_dentry_lru)
+		goto out;
 
 	return 0;
 
 out:
-	ceph_debugfs_client_cleanup(client);
+	ceph_fs_debugfs_cleanup(fsc);
-	return ret;
+	return err;
 }
 
-void ceph_debugfs_client_cleanup(struct ceph_client *client)
-{
-	debugfs_remove(client->debugfs_bdi);
-	debugfs_remove(client->debugfs_caps);
-	debugfs_remove(client->debugfs_dentry_lru);
-	debugfs_remove(client->debugfs_osdmap);
-	debugfs_remove(client->debugfs_mdsmap);
-	debugfs_remove(client->debugfs_monmap);
-	debugfs_remove(client->osdc.debugfs_file);
-	debugfs_remove(client->mdsc.debugfs_file);
-	debugfs_remove(client->monc.debugfs_file);
-	debugfs_remove(client->debugfs_congestion_kb);
-	debugfs_remove(client->debugfs_dir);
-}
 
 #else  /* CONFIG_DEBUG_FS */
 
-int __init ceph_debugfs_init(void)
+int ceph_fs_debugfs_init(struct ceph_fs_client *fsc)
 {
 	return 0;
 }
 
-void ceph_debugfs_cleanup(void)
+void ceph_fs_debugfs_cleanup(struct ceph_fs_client *fsc)
-{
-}
-
-int ceph_debugfs_client_init(struct ceph_client *client)
-{
-	return 0;
-}
-
-void ceph_debugfs_client_cleanup(struct ceph_client *client)
 {
 }
 

fs/ceph/dir.c

@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/spinlock.h>
 #include <linux/fs_struct.h>
@@ -7,6 +7,7 @@
 #include <linux/sched.h>
 
 #include "super.h"
+#include "mds_client.h"
 
 /*
  * Directory operations: readdir, lookup, create, link, unlink,
@@ -94,10 +95,7 @@ static unsigned fpos_off(loff_t p)
  */
 static int __dcache_readdir(struct file *filp,
 			    void *dirent, filldir_t filldir)
-		__releases(inode->i_lock)
-		__acquires(inode->i_lock)
 {
-	struct inode *inode = filp->f_dentry->d_inode;
 	struct ceph_file_info *fi = filp->private_data;
 	struct dentry *parent = filp->f_dentry;
 	struct inode *dir = parent->d_inode;
@@ -153,7 +151,6 @@ more:
 
 	atomic_inc(&dentry->d_count);
 	spin_unlock(&dcache_lock);
-	spin_unlock(&inode->i_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);
@@ -171,35 +168,30 @@ more:
 		} else {
 			dput(last);
 		}
-		last = NULL;
 	}
-
-	spin_lock(&inode->i_lock);
-	spin_lock(&dcache_lock);
-
 	last = dentry;
 
 	if (err < 0)
-		goto out_unlock;
+		goto out;
 
-	p = p->prev;
 	filp->f_pos++;
 
 	/* make sure a dentry wasn't dropped while we didn't have dcache_lock */
-	if ((ceph_inode(dir)->i_ceph_flags & CEPH_I_COMPLETE))
+	if (!ceph_i_test(dir, CEPH_I_COMPLETE)) {
-		goto more;
+		dout(" lost I_COMPLETE on %p; falling back to mds\n", dir);
-	dout(" lost I_COMPLETE on %p; falling back to mds\n", dir);
+		err = -EAGAIN;
-	err = -EAGAIN;
+		goto out;
+	}
+
+	spin_lock(&dcache_lock);
+	p = p->prev;	/* advance to next dentry */
+	goto more;
 
 out_unlock:
 	spin_unlock(&dcache_lock);
-
+out:
-	if (last) {
+	if (last)
-		spin_unlock(&inode->i_lock);
 		dput(last);
-		spin_lock(&inode->i_lock);
-	}
-
 	return err;
 }
 
@@ -227,15 +219,15 @@ static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	struct ceph_file_info *fi = filp->private_data;
 	struct inode *inode = filp->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	unsigned frag = fpos_frag(filp->f_pos);
 	int off = fpos_off(filp->f_pos);
 	int err;
 	u32 ftype;
 	struct ceph_mds_reply_info_parsed *rinfo;
-	const int max_entries = client->mount_args->max_readdir;
+	const int max_entries = fsc->mount_options->max_readdir;
-	const int max_bytes = client->mount_args->max_readdir_bytes;
+	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)
@@ -267,17 +259,17 @@ static int ceph_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	/* can we use the dcache? */
 	spin_lock(&inode->i_lock);
 	if ((filp->f_pos == 2 || fi->dentry) &&
-	    !ceph_test_opt(client, NOASYNCREADDIR) &&
+	    !ceph_test_mount_opt(fsc, NOASYNCREADDIR) &&
 	    ceph_snap(inode) != CEPH_SNAPDIR &&
 	    (ci->i_ceph_flags & CEPH_I_COMPLETE) &&
 	    __ceph_caps_issued_mask(ci, CEPH_CAP_FILE_SHARED, 1)) {
+		spin_unlock(&inode->i_lock);
 		err = __dcache_readdir(filp, dirent, filldir);
-		if (err != -EAGAIN) {
+		if (err != -EAGAIN)
-			spin_unlock(&inode->i_lock);
 			return err;
-		}
+	} else {
+		spin_unlock(&inode->i_lock);
 	}
-	spin_unlock(&inode->i_lock);
 	if (fi->dentry) {
 		err = note_last_dentry(fi, fi->dentry->d_name.name,
 				       fi->dentry->d_name.len);
@@ -487,14 +479,13 @@ static loff_t ceph_dir_llseek(struct file *file, loff_t offset, int origin)
 struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
 				  struct dentry *dentry, int err)
 {
-	struct ceph_client *client = ceph_sb_to_client(dentry->d_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
 	struct inode *parent = dentry->d_parent->d_inode;
 
 	/* .snap dir? */
 	if (err == -ENOENT &&
-	    ceph_vino(parent).ino != CEPH_INO_ROOT && /* no .snap in root dir */
 	    strcmp(dentry->d_name.name,
-		   client->mount_args->snapdir_name) == 0) {
+		   fsc->mount_options->snapdir_name) == 0) {
 		struct inode *inode = ceph_get_snapdir(parent);
 		dout("ENOENT on snapdir %p '%.*s', linking to snapdir %p\n",
 		     dentry, dentry->d_name.len, dentry->d_name.name, inode);
@@ -539,8 +530,8 @@ static int is_root_ceph_dentry(struct inode *inode, struct dentry *dentry)
 static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
 				  struct nameidata *nd)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int op;
 	int err;
@@ -572,7 +563,7 @@ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry,
 		spin_lock(&dir->i_lock);
 		dout(" dir %p flags are %d\n", dir, ci->i_ceph_flags);
 		if (strncmp(dentry->d_name.name,
-			    client->mount_args->snapdir_name,
+			    fsc->mount_options->snapdir_name,
 			    dentry->d_name.len) &&
 		    !is_root_ceph_dentry(dir, dentry) &&
 		    (ci->i_ceph_flags & CEPH_I_COMPLETE) &&
@@ -629,8 +620,8 @@ 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)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -685,8 +676,8 @@ static int ceph_create(struct inode *dir, struct dentry *dentry, int mode,
 static int ceph_symlink(struct inode *dir, struct dentry *dentry,
 			    const char *dest)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -716,8 +707,8 @@ static int ceph_symlink(struct inode *dir, struct dentry *dentry,
 
 static int ceph_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err = -EROFS;
 	int op;
@@ -758,8 +749,8 @@ out:
 static int ceph_link(struct dentry *old_dentry, struct inode *dir,
 		     struct dentry *dentry)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -813,8 +804,8 @@ static int drop_caps_for_unlink(struct inode *inode)
  */
 static int ceph_unlink(struct inode *dir, struct dentry *dentry)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct inode *inode = dentry->d_inode;
 	struct ceph_mds_request *req;
 	int err = -EROFS;
@@ -854,8 +845,8 @@ out:
 static int ceph_rename(struct inode *old_dir, struct dentry *old_dentry,
 		       struct inode *new_dir, struct dentry *new_dentry)
 {
-	struct ceph_client *client = ceph_sb_to_client(old_dir->i_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(old_dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -1076,7 +1067,7 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
 	struct ceph_inode_info *ci = ceph_inode(inode);
 	int left;
 
-	if (!ceph_test_opt(ceph_sb_to_client(inode->i_sb), DIRSTAT))
+	if (!ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), DIRSTAT))
 		return -EISDIR;
 
 	if (!cf->dir_info) {
@@ -1177,7 +1168,7 @@ 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;
+		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++;
@@ -1193,7 +1184,7 @@ 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;
+		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);
@@ -1208,7 +1199,7 @@ 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;
+		mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
 		spin_lock(&mdsc->dentry_lru_lock);
 		list_del_init(&di->lru);
 		mdsc->num_dentry--;

fs/ceph/export.c

@@ -1,10 +1,11 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/exportfs.h>
 #include <linux/slab.h>
 #include <asm/unaligned.h>
 
 #include "super.h"
+#include "mds_client.h"
 
 /*
  * NFS export support
@@ -120,7 +121,7 @@ static struct dentry *__fh_to_dentry(struct super_block *sb,
 static struct dentry *__cfh_to_dentry(struct super_block *sb,
 				      struct ceph_nfs_confh *cfh)
 {
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(sb)->mdsc;
 	struct inode *inode;
 	struct dentry *dentry;
 	struct ceph_vino vino;

fs/ceph/file.c

@@ -1,5 +1,6 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/file.h>
@@ -38,8 +39,8 @@
 static struct ceph_mds_request *
 prepare_open_request(struct super_block *sb, int flags, int create_mode)
 {
-	struct ceph_client *client = ceph_sb_to_client(sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int want_auth = USE_ANY_MDS;
 	int op = (flags & O_CREAT) ? CEPH_MDS_OP_CREATE : CEPH_MDS_OP_OPEN;
@@ -117,8 +118,8 @@ static int ceph_init_file(struct inode *inode, struct file *file, int fmode)
 int ceph_open(struct inode *inode, struct file *file)
 {
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client = ceph_sb_to_client(inode->i_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	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;
@@ -216,8 +217,8 @@ struct dentry *ceph_lookup_open(struct inode *dir, struct dentry *dentry,
 				struct nameidata *nd, int mode,
 				int locked_dir)
 {
-	struct ceph_client *client = ceph_sb_to_client(dir->i_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	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;
@@ -269,163 +270,6 @@ int ceph_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-/*
- * build a vector of user pages
- */
-static struct page **get_direct_page_vector(const char __user *data,
-					    int num_pages,
-					    loff_t off, size_t len)
-{
-	struct page **pages;
-	int rc;
-
-	pages = kmalloc(sizeof(*pages) * num_pages, GFP_NOFS);
-	if (!pages)
-		return ERR_PTR(-ENOMEM);
-
-	down_read(&current->mm->mmap_sem);
-	rc = get_user_pages(current, current->mm, (unsigned long)data,
-			    num_pages, 0, 0, pages, NULL);
-	up_read(&current->mm->mmap_sem);
-	if (rc < 0)
-		goto fail;
-	return pages;
-
-fail:
-	kfree(pages);
-	return ERR_PTR(rc);
-}
-
-static void put_page_vector(struct page **pages, int num_pages)
-{
-	int i;
-
-	for (i = 0; i < num_pages; i++)
-		put_page(pages[i]);
-	kfree(pages);
-}
-
-void ceph_release_page_vector(struct page **pages, int num_pages)
-{
-	int i;
-
-	for (i = 0; i < num_pages; i++)
-		__free_pages(pages[i], 0);
-	kfree(pages);
-}
-
-/*
- * allocate a vector new pages
- */
-static struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags)
-{
-	struct page **pages;
-	int i;
-
-	pages = kmalloc(sizeof(*pages) * num_pages, flags);
-	if (!pages)
-		return ERR_PTR(-ENOMEM);
-	for (i = 0; i < num_pages; i++) {
-		pages[i] = __page_cache_alloc(flags);
-		if (pages[i] == NULL) {
-			ceph_release_page_vector(pages, i);
-			return ERR_PTR(-ENOMEM);
-		}
-	}
-	return pages;
-}
-
-/*
- * copy user data into a page vector
- */
-static int copy_user_to_page_vector(struct page **pages,
-				    const char __user *data,
-				    loff_t off, size_t len)
-{
-	int i = 0;
-	int po = off & ~PAGE_CACHE_MASK;
-	int left = len;
-	int l, bad;
-
-	while (left > 0) {
-		l = min_t(int, PAGE_CACHE_SIZE-po, left);
-		bad = copy_from_user(page_address(pages[i]) + po, data, l);
-		if (bad == l)
-			return -EFAULT;
-		data += l - bad;
-		left -= l - bad;
-		po += l - bad;
-		if (po == PAGE_CACHE_SIZE) {
-			po = 0;
-			i++;
-		}
-	}
-	return len;
-}
-
-/*
- * copy user data from a page vector into a user pointer
- */
-static int copy_page_vector_to_user(struct page **pages, char __user *data,
-				    loff_t off, size_t len)
-{
-	int i = 0;
-	int po = off & ~PAGE_CACHE_MASK;
-	int left = len;
-	int l, bad;
-
-	while (left > 0) {
-		l = min_t(int, left, PAGE_CACHE_SIZE-po);
-		bad = copy_to_user(data, page_address(pages[i]) + po, l);
-		if (bad == l)
-			return -EFAULT;
-		data += l - bad;
-		left -= l - bad;
-		if (po) {
-			po += l - bad;
-			if (po == PAGE_CACHE_SIZE)
-				po = 0;
-		}
-		i++;
-	}
-	return len;
-}
-
-/*
- * Zero an extent within a page vector.  Offset is relative to the
- * start of the first page.
- */
-static void zero_page_vector_range(int off, int len, struct page **pages)
-{
-	int i = off >> PAGE_CACHE_SHIFT;
-
-	off &= ~PAGE_CACHE_MASK;
-
-	dout("zero_page_vector_page %u~%u\n", off, len);
-
-	/* leading partial page? */
-	if (off) {
-		int end = min((int)PAGE_CACHE_SIZE, off + len);
-		dout("zeroing %d %p head from %d\n", i, pages[i],
-		     (int)off);
-		zero_user_segment(pages[i], off, end);
-		len -= (end - off);
-		i++;
-	}
-	while (len >= PAGE_CACHE_SIZE) {
-		dout("zeroing %d %p len=%d\n", i, pages[i], len);
-		zero_user_segment(pages[i], 0, PAGE_CACHE_SIZE);
-		len -= PAGE_CACHE_SIZE;
-		i++;
-	}
-	/* trailing partial page? */
-	if (len) {
-		dout("zeroing %d %p tail to %d\n", i, pages[i], (int)len);
-		zero_user_segment(pages[i], 0, len);
-	}
-}
-
-
 /*
  * Read a range of bytes striped over one or more objects.  Iterate over
  * objects we stripe over.  (That's not atomic, but good enough for now.)
@@ -438,7 +282,7 @@ static int striped_read(struct inode *inode,
 			struct page **pages, int num_pages,
 			int *checkeof)
 {
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	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 */
@@ -459,7 +303,7 @@ static int striped_read(struct inode *inode,
 
 more:
 	this_len = left;
-	ret = ceph_osdc_readpages(&client->osdc, ceph_vino(inode),
+	ret = ceph_osdc_readpages(&fsc->client->osdc, ceph_vino(inode),
 				  &ci->i_layout, pos, &this_len,
 				  ci->i_truncate_seq,
 				  ci->i_truncate_size,
@@ -477,8 +321,8 @@ more:
 
 		if (read < pos - off) {
 			dout(" zero gap %llu to %llu\n", off + read, pos);
-			zero_page_vector_range(page_off + read,
+			ceph_zero_page_vector_range(page_off + read,
-					       pos - off - read, pages);
+						    pos - off - read, pages);
 		}
 		pos += ret;
 		read = pos - off;
@@ -495,8 +339,8 @@ more:
 		/* was original extent fully inside i_size? */
 		if (pos + left <= inode->i_size) {
 			dout("zero tail\n");
-			zero_page_vector_range(page_off + read, len - read,
+			ceph_zero_page_vector_range(page_off + read, len - read,
-					       pages);
+						    pages);
 			read = len;
 			goto out;
 		}
@@ -531,7 +375,7 @@ static ssize_t ceph_sync_read(struct file *file, char __user *data,
 	     (file->f_flags & O_DIRECT) ? "O_DIRECT" : "");
 
 	if (file->f_flags & O_DIRECT) {
-		pages = get_direct_page_vector(data, num_pages, off, len);
+		pages = ceph_get_direct_page_vector(data, num_pages, off, len);
 
 		/*
 		 * flush any page cache pages in this range.  this
@@ -552,13 +396,13 @@ static ssize_t ceph_sync_read(struct file *file, char __user *data,
 	ret = striped_read(inode, off, len, pages, num_pages, checkeof);
 
 	if (ret >= 0 && (file->f_flags & O_DIRECT) == 0)
-		ret = copy_page_vector_to_user(pages, data, off, ret);
+		ret = ceph_copy_page_vector_to_user(pages, data, off, ret);
 	if (ret >= 0)
 		*poff = off + ret;
 
 done:
 	if (file->f_flags & O_DIRECT)
-		put_page_vector(pages, num_pages);
+		ceph_put_page_vector(pages, num_pages);
 	else
 		ceph_release_page_vector(pages, num_pages);
 	dout("sync_read result %d\n", ret);
@@ -594,7 +438,7 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data,
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_client *client = ceph_inode_to_client(inode);
+	struct ceph_fs_client *fsc = ceph_inode_to_client(inode);
 	struct ceph_osd_request *req;
 	struct page **pages;
 	int num_pages;
@@ -642,7 +486,7 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data,
 	 */
 more:
 	len = left;
-	req = ceph_osdc_new_request(&client->osdc, &ci->i_layout,
+	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,
@@ -655,7 +499,7 @@ more:
 	num_pages = calc_pages_for(pos, len);
 
 	if (file->f_flags & O_DIRECT) {
-		pages = get_direct_page_vector(data, num_pages, pos, len);
+		pages = ceph_get_direct_page_vector(data, num_pages, pos, len);
 		if (IS_ERR(pages)) {
 			ret = PTR_ERR(pages);
 			goto out;
@@ -673,7 +517,7 @@ more:
 			ret = PTR_ERR(pages);
 			goto out;
 		}
-		ret = copy_user_to_page_vector(pages, data, pos, len);
+		ret = ceph_copy_user_to_page_vector(pages, data, pos, len);
 		if (ret < 0) {
 			ceph_release_page_vector(pages, num_pages);
 			goto out;
@@ -689,7 +533,7 @@ more:
 	req->r_num_pages = num_pages;
 	req->r_inode = inode;
 
-	ret = ceph_osdc_start_request(&client->osdc, req, false);
+	ret = ceph_osdc_start_request(&fsc->client->osdc, req, false);
 	if (!ret) {
 		if (req->r_safe_callback) {
 			/*
@@ -701,11 +545,11 @@ more:
 			spin_unlock(&ci->i_unsafe_lock);
 			ceph_get_cap_refs(ci, CEPH_CAP_FILE_WR);
 		}
-		ret = ceph_osdc_wait_request(&client->osdc, req);
+		ret = ceph_osdc_wait_request(&fsc->client->osdc, req);
 	}
 
 	if (file->f_flags & O_DIRECT)
-		put_page_vector(pages, num_pages);
+		ceph_put_page_vector(pages, num_pages);
 	else if (file->f_flags & O_SYNC)
 		ceph_release_page_vector(pages, num_pages);
 
@@ -814,7 +658,8 @@ static ssize_t ceph_aio_write(struct kiocb *iocb, const struct iovec *iov,
 	struct ceph_file_info *fi = file->private_data;
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_sb_to_client(inode->i_sb)->osdc;
+	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;

fs/ceph/inode.c

@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/module.h>
 #include <linux/fs.h>
@@ -13,7 +13,8 @@
 #include <linux/pagevec.h>
 
 #include "super.h"
-#include "decode.h"
+#include "mds_client.h"
+#include <linux/ceph/decode.h>
 
 /*
  * Ceph inode operations
@@ -384,7 +385,7 @@ void ceph_destroy_inode(struct inode *inode)
 	 */
 	if (ci->i_snap_realm) {
 		struct ceph_mds_client *mdsc =
-			&ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
+			ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc;
 		struct ceph_snap_realm *realm = ci->i_snap_realm;
 
 		dout(" dropping residual ref to snap realm %p\n", realm);
@@ -685,7 +686,7 @@ static int fill_inode(struct inode *inode,
 		}
 
 		/* it may be better to set st_size in getattr instead? */
-		if (ceph_test_opt(ceph_sb_to_client(inode->i_sb), RBYTES))
+		if (ceph_test_mount_opt(ceph_sb_to_client(inode->i_sb), RBYTES))
 			inode->i_size = ci->i_rbytes;
 		break;
 	default:
@@ -901,7 +902,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 	struct inode *in = NULL;
 	struct ceph_mds_reply_inode *ininfo;
 	struct ceph_vino vino;
-	struct ceph_client *client = ceph_sb_to_client(sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(sb);
 	int i = 0;
 	int err = 0;
 
@@ -965,7 +966,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
 	 */
 	if (rinfo->head->is_dentry && !req->r_aborted &&
 	    (rinfo->head->is_target || strncmp(req->r_dentry->d_name.name,
-					       client->mount_args->snapdir_name,
+					       fsc->mount_options->snapdir_name,
 					       req->r_dentry->d_name.len))) {
 		/*
 		 * lookup link rename   : null -> possibly existing inode
@@ -1533,7 +1534,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr)
 	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;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(dentry->d_sb)->mdsc;
 	int issued;
 	int release = 0, dirtied = 0;
 	int mask = 0;
@@ -1728,8 +1729,8 @@ out:
  */
 int ceph_do_getattr(struct inode *inode, int mask)
 {
-	struct ceph_client *client = ceph_sb_to_client(inode->i_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(inode->i_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 

fs/ceph/ioctl.c

@@ -1,8 +1,10 @@
 #include <linux/in.h>
 
-#include "ioctl.h"
 #include "super.h"
-#include "ceph_debug.h"
+#include "mds_client.h"
+#include <linux/ceph/ceph_debug.h>
+
+#include "ioctl.h"
 
 
 /*
@@ -37,7 +39,7 @@ 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 ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	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;
@@ -89,6 +91,68 @@ static long ceph_ioctl_set_layout(struct file *file, void __user *arg)
 	return err;
 }
 
+/*
+ * Set a layout policy on a directory inode. All items in the tree
+ * rooted at this inode will inherit this layout on creation,
+ * (It doesn't apply retroactively )
+ * unless a subdirectory has its own layout policy.
+ */
+static long ceph_ioctl_set_layout_policy (struct file *file, void __user *arg)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	struct ceph_mds_request *req;
+	struct ceph_ioctl_layout l;
+	int err, i;
+	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;
+	}
+
+	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_args.setlayout.layout.fl_stripe_unit =
+			cpu_to_le32(l.stripe_unit);
+	req->r_args.setlayout.layout.fl_stripe_count =
+			cpu_to_le32(l.stripe_count);
+	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, inode, req);
+	ceph_mdsc_put_request(req);
+	return err;
+}
+
 /*
  * Return object name, size/offset information, and location (OSD
  * number, network address) for a given file offset.
@@ -98,7 +162,8 @@ 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 ceph_inode_info *ci = ceph_inode(inode);
-	struct ceph_osd_client *osdc = &ceph_sb_to_client(inode->i_sb)->osdc;
+	struct ceph_osd_client *osdc =
+		&ceph_sb_to_client(inode->i_sb)->client->osdc;
 	u64 len = 1, olen;
 	u64 tmp;
 	struct ceph_object_layout ol;
@@ -174,11 +239,15 @@ long ceph_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case CEPH_IOC_SET_LAYOUT:
 		return ceph_ioctl_set_layout(file, (void __user *)arg);
 
+	case CEPH_IOC_SET_LAYOUT_POLICY:
+		return ceph_ioctl_set_layout_policy(file, (void __user *)arg);
+
 	case CEPH_IOC_GET_DATALOC:
 		return ceph_ioctl_get_dataloc(file, (void __user *)arg);
 
 	case CEPH_IOC_LAZYIO:
 		return ceph_ioctl_lazyio(file);
 	}
+
 	return -ENOTTY;
 }

fs/ceph/ioctl.h

@@ -4,7 +4,7 @@
 #include <linux/ioctl.h>
 #include <linux/types.h>
 
-#define CEPH_IOCTL_MAGIC 0x97
+#define CEPH_IOCTL_MAGIC 0x98
 
 /* just use u64 to align sanely on all archs */
 struct ceph_ioctl_layout {
@@ -17,6 +17,8 @@ struct ceph_ioctl_layout {
 				   struct ceph_ioctl_layout)
 #define CEPH_IOC_SET_LAYOUT _IOW(CEPH_IOCTL_MAGIC, 2,		\
 				   struct ceph_ioctl_layout)
+#define CEPH_IOC_SET_LAYOUT_POLICY _IOW(CEPH_IOCTL_MAGIC, 5,	\
+				   struct ceph_ioctl_layout)
 
 /*
  * Extract identity, address of the OSD and object storing a given

fs/ceph/locks.c

@@ -1,11 +1,11 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/file.h>
 #include <linux/namei.h>
 
 #include "super.h"
 #include "mds_client.h"
-#include "pagelist.h"
+#include <linux/ceph/pagelist.h>
 
 /**
  * Implement fcntl and flock locking functions.
@@ -16,7 +16,7 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file,
 {
 	struct inode *inode = file->f_dentry->d_inode;
 	struct ceph_mds_client *mdsc =
-		&ceph_sb_to_client(inode->i_sb)->mdsc;
+		ceph_sb_to_client(inode->i_sb)->mdsc;
 	struct ceph_mds_request *req;
 	int err;
 
@@ -181,8 +181,9 @@ 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 BLK already held, and the lock numbers should have
+ * Must be called with lock_flocks() already held.
- * been gathered under the same lock holding window.
+ * If we encounter more of a specific lock type than expected,
+ * we return the value 1.
  */
 int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist,
 		      int num_fcntl_locks, int num_flock_locks)
@@ -190,6 +191,8 @@ int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist,
 	struct file_lock *lock;
 	struct ceph_filelock cephlock;
 	int err = 0;
+	int seen_fcntl = 0;
+	int seen_flock = 0;
 
 	dout("encoding %d flock and %d fcntl locks", num_flock_locks,
 	     num_fcntl_locks);
@@ -198,6 +201,11 @@ int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist,
 		goto fail;
 	for (lock = inode->i_flock; lock != NULL; lock = lock->fl_next) {
 		if (lock->fl_flags & FL_POSIX) {
+			++seen_fcntl;
+			if (seen_fcntl > num_fcntl_locks) {
+				err = -ENOSPC;
+				goto fail;
+			}
 			err = lock_to_ceph_filelock(lock, &cephlock);
 			if (err)
 				goto fail;
@@ -213,6 +221,11 @@ int ceph_encode_locks(struct inode *inode, struct ceph_pagelist *pagelist,
 		goto fail;
 	for (lock = inode->i_flock; lock != NULL; lock = lock->fl_next) {
 		if (lock->fl_flags & FL_FLOCK) {
+			++seen_flock;
+			if (seen_flock > num_flock_locks) {
+				err = -ENOSPC;
+				goto fail;
+			}
 			err = lock_to_ceph_filelock(lock, &cephlock);
 			if (err)
 				goto fail;

fs/ceph/mds_client.c

@@ -1,17 +1,21 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/fs.h>
 #include <linux/wait.h>
 #include <linux/slab.h>
 #include <linux/sched.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
 #include <linux/smp_lock.h>
 
-#include "mds_client.h"
-#include "mon_client.h"
 #include "super.h"
-#include "messenger.h"
+#include "mds_client.h"
-#include "decode.h"
+
-#include "auth.h"
+#include <linux/ceph/messenger.h>
-#include "pagelist.h"
+#include <linux/ceph/decode.h>
+#include <linux/ceph/pagelist.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
 
 /*
  * A cluster of MDS (metadata server) daemons is responsible for
@@ -286,8 +290,9 @@ void ceph_put_mds_session(struct ceph_mds_session *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->client->monc.auth->ops->destroy_authorizer(
+		     s->s_mdsc->fsc->client->monc.auth->ops->destroy_authorizer(
-				s->s_mdsc->client->monc.auth, s->s_authorizer);
+			     s->s_mdsc->fsc->client->monc.auth,
+			     s->s_authorizer);
 		kfree(s);
 	}
 }
@@ -344,7 +349,7 @@ 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->client->msgr, &s->s_con);
+	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;
@@ -599,7 +604,7 @@ static int __choose_mds(struct ceph_mds_client *mdsc,
 	} else if (req->r_dentry) {
 		struct inode *dir = req->r_dentry->d_parent->d_inode;
 
-		if (dir->i_sb != mdsc->client->sb) {
+		if (dir->i_sb != mdsc->fsc->sb) {
 			/* not this fs! */
 			inode = req->r_dentry->d_inode;
 		} else if (ceph_snap(dir) != CEPH_NOSNAP) {
@@ -884,7 +889,7 @@ static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap,
 	__ceph_remove_cap(cap);
 	if (!__ceph_is_any_real_caps(ci)) {
 		struct ceph_mds_client *mdsc =
-			&ceph_sb_to_client(inode->i_sb)->mdsc;
+			ceph_sb_to_client(inode->i_sb)->mdsc;
 
 		spin_lock(&mdsc->cap_dirty_lock);
 		if (!list_empty(&ci->i_dirty_item)) {
@@ -1146,7 +1151,7 @@ int ceph_add_cap_releases(struct ceph_mds_client *mdsc,
 	struct ceph_msg *msg, *partial = NULL;
 	struct ceph_mds_cap_release *head;
 	int err = -ENOMEM;
-	int extra = mdsc->client->mount_args->cap_release_safety;
+	int extra = mdsc->fsc->mount_options->cap_release_safety;
 	int num;
 
 	dout("add_cap_releases %p mds%d extra %d\n", session, session->s_mds,
@@ -2085,7 +2090,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 
 	/* insert trace into our cache */
 	mutex_lock(&req->r_fill_mutex);
-	err = ceph_fill_trace(mdsc->client->sb, req, req->r_session);
+	err = ceph_fill_trace(mdsc->fsc->sb, req, req->r_session);
 	if (err == 0) {
 		if (result == 0 && rinfo->dir_nr)
 			ceph_readdir_prepopulate(req, req->r_session);
@@ -2361,19 +2366,35 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap,
 
 	if (recon_state->flock) {
 		int num_fcntl_locks, num_flock_locks;
+		struct ceph_pagelist_cursor trunc_point;
 
-		lock_kernel();
+		ceph_pagelist_set_cursor(pagelist, &trunc_point);
-		ceph_count_locks(inode, &num_fcntl_locks, &num_flock_locks);
+		do {
-		rec.v2.flock_len = (2*sizeof(u32) +
+			lock_flocks();
-				    (num_fcntl_locks+num_flock_locks) *
+			ceph_count_locks(inode, &num_fcntl_locks,
-				    sizeof(struct ceph_filelock));
+					 &num_flock_locks);
+			rec.v2.flock_len = (2*sizeof(u32) +
+					    (num_fcntl_locks+num_flock_locks) *
+					    sizeof(struct ceph_filelock));
+			unlock_flocks();
 
-		err = ceph_pagelist_append(pagelist, &rec, reclen);
+			/* pre-alloc pagelist */
-		if (!err)
+			ceph_pagelist_truncate(pagelist, &trunc_point);
-			err = ceph_encode_locks(inode, pagelist,
+			err = ceph_pagelist_append(pagelist, &rec, reclen);
-						num_fcntl_locks,
+			if (!err)
-						num_flock_locks);
+				err = ceph_pagelist_reserve(pagelist,
-		unlock_kernel();
+							    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);
 	} else {
 		err = ceph_pagelist_append(pagelist, &rec, reclen);
 	}
@@ -2613,7 +2634,7 @@ static void handle_lease(struct ceph_mds_client *mdsc,
 			 struct ceph_mds_session *session,
 			 struct ceph_msg *msg)
 {
-	struct super_block *sb = mdsc->client->sb;
+	struct super_block *sb = mdsc->fsc->sb;
 	struct inode *inode;
 	struct ceph_inode_info *ci;
 	struct dentry *parent, *dentry;
@@ -2891,10 +2912,16 @@ static void delayed_work(struct work_struct *work)
 	schedule_delayed(mdsc);
 }
 
+int ceph_mdsc_init(struct ceph_fs_client *fsc)
 
-int ceph_mdsc_init(struct ceph_mds_client *mdsc, struct ceph_client *client)
 {
-	mdsc->client = client;
+	struct ceph_mds_client *mdsc;
+
+	mdsc = kzalloc(sizeof(struct ceph_mds_client), GFP_NOFS);
+	if (!mdsc)
+		return -ENOMEM;
+	mdsc->fsc = fsc;
+	fsc->mdsc = mdsc;
 	mutex_init(&mdsc->mutex);
 	mdsc->mdsmap = kzalloc(sizeof(*mdsc->mdsmap), GFP_NOFS);
 	if (mdsc->mdsmap == NULL)
@@ -2927,7 +2954,7 @@ int ceph_mdsc_init(struct ceph_mds_client *mdsc, struct ceph_client *client)
 	INIT_LIST_HEAD(&mdsc->dentry_lru);
 
 	ceph_caps_init(mdsc);
-	ceph_adjust_min_caps(mdsc, client->min_caps);
+	ceph_adjust_min_caps(mdsc, fsc->min_caps);
 
 	return 0;
 }
@@ -2939,7 +2966,7 @@ int ceph_mdsc_init(struct ceph_mds_client *mdsc, struct ceph_client *client)
 static void wait_requests(struct ceph_mds_client *mdsc)
 {
 	struct ceph_mds_request *req;
-	struct ceph_client *client = mdsc->client;
+	struct ceph_fs_client *fsc = mdsc->fsc;
 
 	mutex_lock(&mdsc->mutex);
 	if (__get_oldest_req(mdsc)) {
@@ -2947,7 +2974,7 @@ static void wait_requests(struct ceph_mds_client *mdsc)
 
 		dout("wait_requests waiting for requests\n");
 		wait_for_completion_timeout(&mdsc->safe_umount_waiters,
-				    client->mount_args->mount_timeout * HZ);
+				    fsc->client->options->mount_timeout * HZ);
 
 		/* tear down remaining requests */
 		mutex_lock(&mdsc->mutex);
@@ -3030,7 +3057,7 @@ void ceph_mdsc_sync(struct ceph_mds_client *mdsc)
 {
 	u64 want_tid, want_flush;
 
-	if (mdsc->client->mount_state == CEPH_MOUNT_SHUTDOWN)
+	if (mdsc->fsc->mount_state == CEPH_MOUNT_SHUTDOWN)
 		return;
 
 	dout("sync\n");
@@ -3053,7 +3080,7 @@ bool done_closing_sessions(struct ceph_mds_client *mdsc)
 {
 	int i, n = 0;
 
-	if (mdsc->client->mount_state == CEPH_MOUNT_SHUTDOWN)
+	if (mdsc->fsc->mount_state == CEPH_MOUNT_SHUTDOWN)
 		return true;
 
 	mutex_lock(&mdsc->mutex);
@@ -3071,8 +3098,8 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc)
 {
 	struct ceph_mds_session *session;
 	int i;
-	struct ceph_client *client = mdsc->client;
+	struct ceph_fs_client *fsc = mdsc->fsc;
-	unsigned long timeout = client->mount_args->mount_timeout * HZ;
+	unsigned long timeout = fsc->client->options->mount_timeout * HZ;
 
 	dout("close_sessions\n");
 
@@ -3119,7 +3146,7 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc)
 	dout("stopped\n");
 }
 
-void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
+static void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
 {
 	dout("stop\n");
 	cancel_delayed_work_sync(&mdsc->delayed_work); /* cancel timer */
@@ -3129,6 +3156,15 @@ void ceph_mdsc_stop(struct ceph_mds_client *mdsc)
 	ceph_caps_finalize(mdsc);
 }
 
+void ceph_mdsc_destroy(struct ceph_fs_client *fsc)
+{
+	struct ceph_mds_client *mdsc = fsc->mdsc;
+
+	ceph_mdsc_stop(mdsc);
+	fsc->mdsc = NULL;
+	kfree(mdsc);
+}
+
 
 /*
  * handle mds map update.
@@ -3145,14 +3181,14 @@ void ceph_mdsc_handle_map(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
 
 	ceph_decode_need(&p, end, sizeof(fsid)+2*sizeof(u32), bad);
 	ceph_decode_copy(&p, &fsid, sizeof(fsid));
-	if (ceph_check_fsid(mdsc->client, &fsid) < 0)
+	if (ceph_check_fsid(mdsc->fsc->client, &fsid) < 0)
 		return;
 	epoch = ceph_decode_32(&p);
 	maplen = ceph_decode_32(&p);
 	dout("handle_map epoch %u len %d\n", epoch, (int)maplen);
 
 	/* do we need it? */
-	ceph_monc_got_mdsmap(&mdsc->client->monc, epoch);
+	ceph_monc_got_mdsmap(&mdsc->fsc->client->monc, epoch);
 	mutex_lock(&mdsc->mutex);
 	if (mdsc->mdsmap && epoch <= mdsc->mdsmap->m_epoch) {
 		dout("handle_map epoch %u <= our %u\n",
@@ -3176,7 +3212,7 @@ void ceph_mdsc_handle_map(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
 	} else {
 		mdsc->mdsmap = newmap;  /* first mds map */
 	}
-	mdsc->client->sb->s_maxbytes = mdsc->mdsmap->m_max_file_size;
+	mdsc->fsc->sb->s_maxbytes = mdsc->mdsmap->m_max_file_size;
 
 	__wake_requests(mdsc, &mdsc->waiting_for_map);
 
@@ -3277,7 +3313,7 @@ static int get_authorizer(struct ceph_connection *con,
 {
 	struct ceph_mds_session *s = con->private;
 	struct ceph_mds_client *mdsc = s->s_mdsc;
-	struct ceph_auth_client *ac = mdsc->client->monc.auth;
+	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
 	int ret = 0;
 
 	if (force_new && s->s_authorizer) {
@@ -3311,7 +3347,7 @@ static int verify_authorizer_reply(struct ceph_connection *con, int len)
 {
 	struct ceph_mds_session *s = con->private;
 	struct ceph_mds_client *mdsc = s->s_mdsc;
-	struct ceph_auth_client *ac = mdsc->client->monc.auth;
+	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
 
 	return ac->ops->verify_authorizer_reply(ac, s->s_authorizer, len);
 }
@@ -3320,12 +3356,12 @@ static int invalidate_authorizer(struct ceph_connection *con)
 {
 	struct ceph_mds_session *s = con->private;
 	struct ceph_mds_client *mdsc = s->s_mdsc;
-	struct ceph_auth_client *ac = mdsc->client->monc.auth;
+	struct ceph_auth_client *ac = mdsc->fsc->client->monc.auth;
 
 	if (ac->ops->invalidate_authorizer)
 		ac->ops->invalidate_authorizer(ac, CEPH_ENTITY_TYPE_MDS);
 
-	return ceph_monc_validate_auth(&mdsc->client->monc);
+	return ceph_monc_validate_auth(&mdsc->fsc->client->monc);
 }
 
 static const struct ceph_connection_operations mds_con_ops = {
@@ -3338,7 +3374,4 @@ static const struct ceph_connection_operations mds_con_ops = {
 	.peer_reset = peer_reset,
 };
 
-
-
-
 /* eof */

fs/ceph/mds_client.h

@@ -8,9 +8,9 @@
 #include <linux/rbtree.h>
 #include <linux/spinlock.h>
 
-#include "types.h"
+#include <linux/ceph/types.h>
-#include "messenger.h"
+#include <linux/ceph/messenger.h>
-#include "mdsmap.h"
+#include <linux/ceph/mdsmap.h>
 
 /*
  * Some lock dependencies:
@@ -26,7 +26,7 @@
  *
  */
 
-struct ceph_client;
+struct ceph_fs_client;
 struct ceph_cap;
 
 /*
@@ -230,7 +230,7 @@ struct ceph_mds_request {
  * mds client state
  */
 struct ceph_mds_client {
-	struct ceph_client      *client;
+	struct ceph_fs_client  *fsc;
 	struct mutex            mutex;         /* all nested structures */
 
 	struct ceph_mdsmap      *mdsmap;
@@ -289,11 +289,6 @@ struct ceph_mds_client {
 	int		caps_avail_count;    /* unused, unreserved */
 	int		caps_min_count;      /* keep at least this many
 						(unreserved) */
-
-#ifdef CONFIG_DEBUG_FS
-	struct dentry 	  *debugfs_file;
-#endif
-
 	spinlock_t	  dentry_lru_lock;
 	struct list_head  dentry_lru;
 	int		  num_dentry;
@@ -316,10 +311,9 @@ extern void ceph_put_mds_session(struct ceph_mds_session *s);
 extern int ceph_send_msg_mds(struct ceph_mds_client *mdsc,
 			     struct ceph_msg *msg, int mds);
 
-extern int ceph_mdsc_init(struct ceph_mds_client *mdsc,
+extern int ceph_mdsc_init(struct ceph_fs_client *fsc);
-			   struct ceph_client *client);
 extern void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc);
-extern void ceph_mdsc_stop(struct ceph_mds_client *mdsc);
+extern void ceph_mdsc_destroy(struct ceph_fs_client *fsc);
 
 extern void ceph_mdsc_sync(struct ceph_mds_client *mdsc);
 

fs/ceph/mdsmap.c

@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/bug.h>
 #include <linux/err.h>
@@ -6,9 +6,9 @@
 #include <linux/slab.h>
 #include <linux/types.h>
 
-#include "mdsmap.h"
+#include <linux/ceph/mdsmap.h>
-#include "messenger.h"
+#include <linux/ceph/messenger.h>
-#include "decode.h"
+#include <linux/ceph/decode.h>
 
 #include "super.h"
 
@@ -117,7 +117,8 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 		}
 
 		dout("mdsmap_decode %d/%d %lld mds%d.%d %s %s\n",
-		     i+1, n, global_id, mds, inc, pr_addr(&addr.in_addr),
+		     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;

fs/ceph/pagelist.c

@@ -1,63 +0,0 @@
-
-#include <linux/gfp.h>
-#include <linux/pagemap.h>
-#include <linux/highmem.h>
-
-#include "pagelist.h"
-
-static void ceph_pagelist_unmap_tail(struct ceph_pagelist *pl)
-{
-	struct page *page = list_entry(pl->head.prev, struct page,
-				       lru);
-	kunmap(page);
-}
-
-int ceph_pagelist_release(struct ceph_pagelist *pl)
-{
-	if (pl->mapped_tail)
-		ceph_pagelist_unmap_tail(pl);
-
-	while (!list_empty(&pl->head)) {
-		struct page *page = list_first_entry(&pl->head, struct page,
-						     lru);
-		list_del(&page->lru);
-		__free_page(page);
-	}
-	return 0;
-}
-
-static int ceph_pagelist_addpage(struct ceph_pagelist *pl)
-{
-	struct page *page = __page_cache_alloc(GFP_NOFS);
-	if (!page)
-		return -ENOMEM;
-	pl->room += PAGE_SIZE;
-	list_add_tail(&page->lru, &pl->head);
-	if (pl->mapped_tail)
-		ceph_pagelist_unmap_tail(pl);
-	pl->mapped_tail = kmap(page);
-	return 0;
-}
-
-int ceph_pagelist_append(struct ceph_pagelist *pl, void *buf, size_t len)
-{
-	while (pl->room < len) {
-		size_t bit = pl->room;
-		int ret;
-
-		memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK),
-		       buf, bit);
-		pl->length += bit;
-		pl->room -= bit;
-		buf += bit;
-		len -= bit;
-		ret = ceph_pagelist_addpage(pl);
-		if (ret)
-			return ret;
-	}
-
-	memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK), buf, len);
-	pl->length += len;
-	pl->room -= len;
-	return 0;
-}

fs/ceph/snap.c

@@ -1,10 +1,12 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/sort.h>
 #include <linux/slab.h>
 
 #include "super.h"
-#include "decode.h"
+#include "mds_client.h"
+
+#include <linux/ceph/decode.h>
 
 /*
  * Snapshots in ceph are driven in large part by cooperation from the
@@ -526,7 +528,7 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
 			    struct ceph_cap_snap *capsnap)
 {
 	struct inode *inode = &ci->vfs_inode;
-	struct ceph_mds_client *mdsc = &ceph_sb_to_client(inode->i_sb)->mdsc;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
 
 	BUG_ON(capsnap->writing);
 	capsnap->size = inode->i_size;
@@ -747,7 +749,7 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc,
 		      struct ceph_mds_session *session,
 		      struct ceph_msg *msg)
 {
-	struct super_block *sb = mdsc->client->sb;
+	struct super_block *sb = mdsc->fsc->sb;
 	int mds = session->s_mds;
 	u64 split;
 	int op;

fs/ceph/strings.c

@@ -1,71 +1,9 @@
 /*
- * Ceph string constants
+ * Ceph fs string constants
  */
-#include "types.h"
+#include <linux/module.h>
+#include <linux/ceph/types.h>
 
-const char *ceph_entity_type_name(int type)
-{
-	switch (type) {
-	case CEPH_ENTITY_TYPE_MDS: return "mds";
-	case CEPH_ENTITY_TYPE_OSD: return "osd";
-	case CEPH_ENTITY_TYPE_MON: return "mon";
-	case CEPH_ENTITY_TYPE_CLIENT: return "client";
-	case CEPH_ENTITY_TYPE_AUTH: return "auth";
-	default: return "unknown";
-	}
-}
-
-const char *ceph_osd_op_name(int op)
-{
-	switch (op) {
-	case CEPH_OSD_OP_READ: return "read";
-	case CEPH_OSD_OP_STAT: return "stat";
-
-	case CEPH_OSD_OP_MASKTRUNC: return "masktrunc";
-
-	case CEPH_OSD_OP_WRITE: return "write";
-	case CEPH_OSD_OP_DELETE: return "delete";
-	case CEPH_OSD_OP_TRUNCATE: return "truncate";
-	case CEPH_OSD_OP_ZERO: return "zero";
-	case CEPH_OSD_OP_WRITEFULL: return "writefull";
-	case CEPH_OSD_OP_ROLLBACK: return "rollback";
-
-	case CEPH_OSD_OP_APPEND: return "append";
-	case CEPH_OSD_OP_STARTSYNC: return "startsync";
-	case CEPH_OSD_OP_SETTRUNC: return "settrunc";
-	case CEPH_OSD_OP_TRIMTRUNC: return "trimtrunc";
-
-	case CEPH_OSD_OP_TMAPUP: return "tmapup";
-	case CEPH_OSD_OP_TMAPGET: return "tmapget";
-	case CEPH_OSD_OP_TMAPPUT: return "tmapput";
-
-	case CEPH_OSD_OP_GETXATTR: return "getxattr";
-	case CEPH_OSD_OP_GETXATTRS: return "getxattrs";
-	case CEPH_OSD_OP_SETXATTR: return "setxattr";
-	case CEPH_OSD_OP_SETXATTRS: return "setxattrs";
-	case CEPH_OSD_OP_RESETXATTRS: return "resetxattrs";
-	case CEPH_OSD_OP_RMXATTR: return "rmxattr";
-	case CEPH_OSD_OP_CMPXATTR: return "cmpxattr";
-
-	case CEPH_OSD_OP_PULL: return "pull";
-	case CEPH_OSD_OP_PUSH: return "push";
-	case CEPH_OSD_OP_BALANCEREADS: return "balance-reads";
-	case CEPH_OSD_OP_UNBALANCEREADS: return "unbalance-reads";
-	case CEPH_OSD_OP_SCRUB: return "scrub";
-
-	case CEPH_OSD_OP_WRLOCK: return "wrlock";
-	case CEPH_OSD_OP_WRUNLOCK: return "wrunlock";
-	case CEPH_OSD_OP_RDLOCK: return "rdlock";
-	case CEPH_OSD_OP_RDUNLOCK: return "rdunlock";
-	case CEPH_OSD_OP_UPLOCK: return "uplock";
-	case CEPH_OSD_OP_DNLOCK: return "dnlock";
-
-	case CEPH_OSD_OP_CALL: return "call";
-
-	case CEPH_OSD_OP_PGLS: return "pgls";
-	}
-	return "???";
-}
 
 const char *ceph_mds_state_name(int s)
 {
@@ -177,17 +115,3 @@ const char *ceph_snap_op_name(int o)
 	}
 	return "???";
 }
-
-const char *ceph_pool_op_name(int op)
-{
-	switch (op) {
-	case POOL_OP_CREATE: return "create";
-	case POOL_OP_DELETE: return "delete";
-	case POOL_OP_AUID_CHANGE: return "auid change";
-	case POOL_OP_CREATE_SNAP: return "create snap";
-	case POOL_OP_DELETE_SNAP: return "delete snap";
-	case POOL_OP_CREATE_UNMANAGED_SNAP: return "create unmanaged snap";
-	case POOL_OP_DELETE_UNMANAGED_SNAP: return "delete unmanaged snap";
-	}
-	return "???";
-}

fs/ceph/super.h

@@ -1,7 +1,7 @@
 #ifndef _FS_CEPH_SUPER_H
 #define _FS_CEPH_SUPER_H
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <asm/unaligned.h>
 #include <linux/backing-dev.h>
@@ -14,13 +14,7 @@
 #include <linux/writeback.h>
 #include <linux/slab.h>
 
-#include "types.h"
+#include <linux/ceph/libceph.h>
-#include "messenger.h"
-#include "msgpool.h"
-#include "mon_client.h"
-#include "mds_client.h"
-#include "osd_client.h"
-#include "ceph_fs.h"
 
 /* f_type in struct statfs */
 #define CEPH_SUPER_MAGIC 0x00c36400
@@ -30,42 +24,25 @@
 #define CEPH_BLOCK_SHIFT   20  /* 1 MB */
 #define CEPH_BLOCK         (1 << CEPH_BLOCK_SHIFT)
 
-/*
+#define CEPH_MOUNT_OPT_DIRSTAT         (1<<4) /* `cat dirname` for stats */
- * Supported features
+#define CEPH_MOUNT_OPT_RBYTES          (1<<5) /* dir st_bytes = rbytes */
- */
+#define CEPH_MOUNT_OPT_NOASYNCREADDIR  (1<<7) /* no dcache readdir */
-#define CEPH_FEATURE_SUPPORTED CEPH_FEATURE_NOSRCADDR | CEPH_FEATURE_FLOCK
-#define CEPH_FEATURE_REQUIRED  CEPH_FEATURE_NOSRCADDR
 
-/*
+#define CEPH_MOUNT_OPT_DEFAULT    (CEPH_MOUNT_OPT_RBYTES)
- * mount options
- */
-#define CEPH_OPT_FSID             (1<<0)
-#define CEPH_OPT_NOSHARE          (1<<1) /* don't share client with other sbs */
-#define CEPH_OPT_MYIP             (1<<2) /* specified my ip */
-#define CEPH_OPT_DIRSTAT          (1<<4) /* funky `cat dirname` for stats */
-#define CEPH_OPT_RBYTES           (1<<5) /* dir st_bytes = rbytes */
-#define CEPH_OPT_NOCRC            (1<<6) /* no data crc on writes */
-#define CEPH_OPT_NOASYNCREADDIR   (1<<7) /* no dcache readdir */
 
-#define CEPH_OPT_DEFAULT   (CEPH_OPT_RBYTES)
+#define ceph_set_mount_opt(fsc, opt) \
+	(fsc)->mount_options->flags |= CEPH_MOUNT_OPT_##opt;
+#define ceph_test_mount_opt(fsc, opt) \
+	(!!((fsc)->mount_options->flags & CEPH_MOUNT_OPT_##opt))
 
-#define ceph_set_opt(client, opt) \
+#define CEPH_MAX_READDIR_DEFAULT        1024
-	(client)->mount_args->flags |= CEPH_OPT_##opt;
+#define CEPH_MAX_READDIR_BYTES_DEFAULT  (512*1024)
-#define ceph_test_opt(client, opt) \
+#define CEPH_SNAPDIRNAME_DEFAULT        ".snap"
-	(!!((client)->mount_args->flags & CEPH_OPT_##opt))
 
-
+struct ceph_mount_options {
-struct ceph_mount_args {
-	int sb_flags;
 	int flags;
-	struct ceph_fsid fsid;
+	int sb_flags;
-	struct ceph_entity_addr my_addr;
+
-	int num_mon;
-	struct ceph_entity_addr *mon_addr;
-	int mount_timeout;
-	int osd_idle_ttl;
-	int osd_timeout;
-	int osd_keepalive_timeout;
 	int wsize;
 	int rsize;            /* max readahead */
 	int congestion_kb;    /* max writeback in flight */
@@ -73,82 +50,25 @@ struct ceph_mount_args {
 	int cap_release_safety;
 	int max_readdir;       /* max readdir result (entires) */
 	int max_readdir_bytes; /* max readdir result (bytes) */
+
+	/*
+	 * everything above this point can be memcmp'd; everything below
+	 * is handled in compare_mount_options()
+	 */
+
 	char *snapdir_name;   /* default ".snap" */
-	char *name;
-	char *secret;
 };
 
-/*
+struct ceph_fs_client {
- * defaults
- */
-#define CEPH_MOUNT_TIMEOUT_DEFAULT  60
-#define CEPH_OSD_TIMEOUT_DEFAULT    60  /* seconds */
-#define CEPH_OSD_KEEPALIVE_DEFAULT  5
-#define CEPH_OSD_IDLE_TTL_DEFAULT    60
-#define CEPH_MOUNT_RSIZE_DEFAULT    (512*1024) /* readahead */
-#define CEPH_MAX_READDIR_DEFAULT    1024
-#define CEPH_MAX_READDIR_BYTES_DEFAULT    (512*1024)
-
-#define CEPH_MSG_MAX_FRONT_LEN	(16*1024*1024)
-#define CEPH_MSG_MAX_DATA_LEN	(16*1024*1024)
-
-#define CEPH_SNAPDIRNAME_DEFAULT ".snap"
-#define CEPH_AUTH_NAME_DEFAULT   "guest"
-/*
- * Delay telling the MDS we no longer want caps, in case we reopen
- * the file.  Delay a minimum amount of time, even if we send a cap
- * message for some other reason.  Otherwise, take the oppotunity to
- * update the mds to avoid sending another message later.
- */
-#define CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT      5  /* cap release delay */
-#define CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT     60  /* cap release delay */
-
-#define CEPH_CAP_RELEASE_SAFETY_DEFAULT        (CEPH_CAPS_PER_RELEASE * 4)
-
-/* mount state */
-enum {
-	CEPH_MOUNT_MOUNTING,
-	CEPH_MOUNT_MOUNTED,
-	CEPH_MOUNT_UNMOUNTING,
-	CEPH_MOUNT_UNMOUNTED,
-	CEPH_MOUNT_SHUTDOWN,
-};
-
-/*
- * subtract jiffies
- */
-static inline unsigned long time_sub(unsigned long a, unsigned long b)
-{
-	BUG_ON(time_after(b, a));
-	return (long)a - (long)b;
-}
-
-/*
- * per-filesystem client state
- *
- * possibly shared by multiple mount points, if they are
- * mounting the same ceph filesystem/cluster.
- */
-struct ceph_client {
-	struct ceph_fsid fsid;
-	bool have_fsid;
-
-	struct mutex mount_mutex;       /* serialize mount attempts */
-	struct ceph_mount_args *mount_args;
-
 	struct super_block *sb;
 
+	struct ceph_mount_options *mount_options;
+	struct ceph_client *client;
+
 	unsigned long mount_state;
-	wait_queue_head_t auth_wq;
-
-	int auth_err;
-
 	int min_caps;                  /* min caps i added */
 
-	struct ceph_messenger *msgr;   /* messenger instance */
+	struct ceph_mds_client *mdsc;
-	struct ceph_mon_client monc;
-	struct ceph_mds_client mdsc;
-	struct ceph_osd_client osdc;
 
 	/* writeback */
 	mempool_t *wb_pagevec_pool;
@@ -160,14 +80,14 @@ struct ceph_client {
 	struct backing_dev_info backing_dev_info;
 
 #ifdef CONFIG_DEBUG_FS
-	struct dentry *debugfs_monmap;
+	struct dentry *debugfs_dentry_lru, *debugfs_caps;
-	struct dentry *debugfs_mdsmap, *debugfs_osdmap;
-	struct dentry *debugfs_dir, *debugfs_dentry_lru, *debugfs_caps;
 	struct dentry *debugfs_congestion_kb;
 	struct dentry *debugfs_bdi;
+	struct dentry *debugfs_mdsc, *debugfs_mdsmap;
 #endif
 };
 
+
 /*
  * File i/o capability.  This tracks shared state with the metadata
  * server that allows us to cache or writeback attributes or to read
@@ -275,6 +195,20 @@ struct ceph_inode_xattr {
 	int should_free_val;
 };
 
+/*
+ * Ceph dentry state
+ */
+struct ceph_dentry_info {
+	struct ceph_mds_session *lease_session;
+	u32 lease_gen, lease_shared_gen;
+	u32 lease_seq;
+	unsigned long lease_renew_after, lease_renew_from;
+	struct list_head lru;
+	struct dentry *dentry;
+	u64 time;
+	u64 offset;
+};
+
 struct ceph_inode_xattrs_info {
 	/*
 	 * (still encoded) xattr blob. we avoid the overhead of parsing
@@ -296,11 +230,6 @@ struct ceph_inode_xattrs_info {
 /*
  * 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 */
-
 struct ceph_inode_info {
 	struct ceph_vino i_vino;   /* ceph ino + snap */
 
@@ -391,69 +320,9 @@ static inline struct ceph_inode_info *ceph_inode(struct inode *inode)
 	return container_of(inode, struct ceph_inode_info, vfs_inode);
 }
 
-static inline void ceph_i_clear(struct inode *inode, unsigned mask)
+static inline struct ceph_vino ceph_vino(struct inode *inode)
 {
-	struct ceph_inode_info *ci = ceph_inode(inode);
+	return ceph_inode(inode)->i_vino;
-
-	spin_lock(&inode->i_lock);
-	ci->i_ceph_flags &= ~mask;
-	spin_unlock(&inode->i_lock);
-}
-
-static inline void ceph_i_set(struct inode *inode, unsigned mask)
-{
-	struct ceph_inode_info *ci = ceph_inode(inode);
-
-	spin_lock(&inode->i_lock);
-	ci->i_ceph_flags |= mask;
-	spin_unlock(&inode->i_lock);
-}
-
-static inline bool ceph_i_test(struct inode *inode, unsigned mask)
-{
-	struct ceph_inode_info *ci = ceph_inode(inode);
-	bool r;
-
-	smp_mb();
-	r = (ci->i_ceph_flags & mask) == mask;
-	return r;
-}
-
-
-/* find a specific frag @f */
-extern struct ceph_inode_frag *__ceph_find_frag(struct ceph_inode_info *ci,
-						u32 f);
-
-/*
- * choose fragment for value @v.  copy frag content to pfrag, if leaf
- * exists
- */
-extern u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
-			    struct ceph_inode_frag *pfrag,
-			    int *found);
-
-/*
- * Ceph dentry state
- */
-struct ceph_dentry_info {
-	struct ceph_mds_session *lease_session;
-	u32 lease_gen, lease_shared_gen;
-	u32 lease_seq;
-	unsigned long lease_renew_after, lease_renew_from;
-	struct list_head lru;
-	struct dentry *dentry;
-	u64 time;
-	u64 offset;
-};
-
-static inline struct ceph_dentry_info *ceph_dentry(struct dentry *dentry)
-{
-	return (struct ceph_dentry_info *)dentry->d_fsdata;
-}
-
-static inline loff_t ceph_make_fpos(unsigned frag, unsigned off)
-{
-	return ((loff_t)frag << 32) | (loff_t)off;
 }
 
 /*
@@ -472,18 +341,6 @@ static inline ino_t ceph_vino_to_ino(struct ceph_vino vino)
 	return ino;
 }
 
-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;
-}
-
-static inline struct ceph_vino ceph_vino(struct inode *inode)
-{
-	return ceph_inode(inode)->i_vino;
-}
-
 /* for printf-style formatting */
 #define ceph_vinop(i) ceph_inode(i)->i_vino.ino, ceph_inode(i)->i_vino.snap
 
@@ -512,6 +369,73 @@ 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)
+{
+	struct ceph_inode_info *ci = ceph_inode(inode);
+
+	spin_lock(&inode->i_lock);
+	ci->i_ceph_flags &= ~mask;
+	spin_unlock(&inode->i_lock);
+}
+
+static inline void ceph_i_set(struct inode *inode, unsigned mask)
+{
+	struct ceph_inode_info *ci = ceph_inode(inode);
+
+	spin_lock(&inode->i_lock);
+	ci->i_ceph_flags |= mask;
+	spin_unlock(&inode->i_lock);
+}
+
+static inline bool ceph_i_test(struct inode *inode, unsigned mask)
+{
+	struct ceph_inode_info *ci = ceph_inode(inode);
+	bool r;
+
+	spin_lock(&inode->i_lock);
+	r = (ci->i_ceph_flags & mask) == mask;
+	spin_unlock(&inode->i_lock);
+	return r;
+}
+
+
+/* find a specific frag @f */
+extern struct ceph_inode_frag *__ceph_find_frag(struct ceph_inode_info *ci,
+						u32 f);
+
+/*
+ * choose fragment for value @v.  copy frag content to pfrag, if leaf
+ * exists
+ */
+extern u32 ceph_choose_frag(struct ceph_inode_info *ci, u32 v,
+			    struct ceph_inode_frag *pfrag,
+			    int *found);
+
+static inline struct ceph_dentry_info *ceph_dentry(struct dentry *dentry)
+{
+	return (struct ceph_dentry_info *)dentry->d_fsdata;
+}
+
+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
  */
@@ -576,18 +500,18 @@ extern int 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);
-extern void ceph_reservation_status(struct ceph_client *client,
+extern void ceph_reservation_status(struct ceph_fs_client *client,
 				    int *total, int *avail, int *used,
 				    int *reserved, int *min);
 
-static inline struct ceph_client *ceph_inode_to_client(struct inode *inode)
+static inline struct ceph_fs_client *ceph_inode_to_client(struct inode *inode)
 {
-	return (struct ceph_client *)inode->i_sb->s_fs_info;
+	return (struct ceph_fs_client *)inode->i_sb->s_fs_info;
 }
 
-static inline struct ceph_client *ceph_sb_to_client(struct super_block *sb)
+static inline struct ceph_fs_client *ceph_sb_to_client(struct super_block *sb)
 {
-	return (struct ceph_client *)sb->s_fs_info;
+	return (struct ceph_fs_client *)sb->s_fs_info;
 }
 
 
@@ -616,51 +540,6 @@ struct ceph_file_info {
 
 
 
-/*
- * snapshots
- */
-
-/*
- * A "snap context" is the set of existing snapshots when we
- * write data.  It is used by the OSD to guide its COW behavior.
- *
- * The ceph_snap_context is refcounted, and attached to each dirty
- * page, indicating which context the dirty data belonged when it was
- * dirtied.
- */
-struct ceph_snap_context {
-	atomic_t nref;
-	u64 seq;
-	int num_snaps;
-	u64 snaps[];
-};
-
-static inline struct ceph_snap_context *
-ceph_get_snap_context(struct ceph_snap_context *sc)
-{
-	/*
-	printk("get_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
-	       atomic_read(&sc->nref)+1);
-	*/
-	if (sc)
-		atomic_inc(&sc->nref);
-	return sc;
-}
-
-static inline void ceph_put_snap_context(struct ceph_snap_context *sc)
-{
-	if (!sc)
-		return;
-	/*
-	printk("put_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
-	       atomic_read(&sc->nref)-1);
-	*/
-	if (atomic_dec_and_test(&sc->nref)) {
-		/*printk(" deleting snap_context %p\n", sc);*/
-		kfree(sc);
-	}
-}
-
 /*
  * A "snap realm" describes a subset of the file hierarchy sharing
  * the same set of snapshots that apply to it.  The realms themselves
@@ -699,16 +578,33 @@ struct ceph_snap_realm {
 	spinlock_t inodes_with_caps_lock;
 };
 
-
+static inline int default_congestion_kb(void)
-
-/*
- * calculate the number of pages a given length and offset map onto,
- * if we align the data.
- */
-static inline int calc_pages_for(u64 off, u64 len)
 {
-	return ((off+len+PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT) -
+	int congestion_kb;
-		(off >> PAGE_CACHE_SHIFT);
+
+	/*
+	 * Copied from NFS
+	 *
+	 * congestion size, scale with available memory.
+	 *
+	 *  64MB:    8192k
+	 * 128MB:   11585k
+	 * 256MB:   16384k
+	 * 512MB:   23170k
+	 *   1GB:   32768k
+	 *   2GB:   46340k
+	 *   4GB:   65536k
+	 *   8GB:   92681k
+	 *  16GB:  131072k
+	 *
+	 * This allows larger machines to have larger/more transfers.
+	 * Limit the default to 256M
+	 */
+	congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10);
+	if (congestion_kb > 256*1024)
+		congestion_kb = 256*1024;
+
+	return congestion_kb;
 }
 
 
@@ -741,16 +637,6 @@ static inline bool __ceph_have_pending_cap_snap(struct ceph_inode_info *ci)
 			   ci_item)->writing;
 }
 
-
-/* super.c */
-extern struct kmem_cache *ceph_inode_cachep;
-extern struct kmem_cache *ceph_cap_cachep;
-extern struct kmem_cache *ceph_dentry_cachep;
-extern struct kmem_cache *ceph_file_cachep;
-
-extern const char *ceph_msg_type_name(int type);
-extern int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid);
-
 /* inode.c */
 extern const struct inode_operations ceph_file_iops;
 
@@ -857,12 +743,18 @@ 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_release(struct inode *inode, struct file *filp);
-extern void ceph_release_page_vector(struct page **pages, int num_pages);
 
 /* dir.c */
 extern const struct file_operations ceph_dir_fops;
@@ -892,12 +784,6 @@ extern long ceph_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 /* export.c */
 extern const struct export_operations ceph_export_ops;
 
-/* debugfs.c */
-extern int ceph_debugfs_init(void);
-extern void ceph_debugfs_cleanup(void);
-extern int ceph_debugfs_client_init(struct ceph_client *client);
-extern void ceph_debugfs_client_cleanup(struct ceph_client *client);
-
 /* locks.c */
 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);
@@ -914,4 +800,8 @@ static inline struct inode *get_dentry_parent_inode(struct dentry *dentry)
 	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);
+
 #endif /* _FS_CEPH_SUPER_H */

fs/ceph/xattr.c

@@ -1,6 +1,9 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
+
 #include "super.h"
-#include "decode.h"
+#include "mds_client.h"
+
+#include <linux/ceph/decode.h>
 
 #include <linux/xattr.h>
 #include <linux/slab.h>
@@ -620,12 +623,12 @@ out:
 static int ceph_sync_setxattr(struct dentry *dentry, const char *name,
 			      const char *value, size_t size, int flags)
 {
-	struct ceph_client *client = ceph_sb_to_client(dentry->d_sb);
+	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 = &client->mdsc;
+	struct ceph_mds_client *mdsc = fsc->mdsc;
 	int err;
 	int i, nr_pages;
 	struct page **pages = NULL;
@@ -713,10 +716,9 @@ int ceph_setxattr(struct dentry *dentry, const char *name,
 
 	/* preallocate memory for xattr name, value, index node */
 	err = -ENOMEM;
-	newname = kmalloc(name_len + 1, GFP_NOFS);
+	newname = kmemdup(name, name_len + 1, GFP_NOFS);
 	if (!newname)
 		goto out;
-	memcpy(newname, name, name_len + 1);
 
 	if (val_len) {
 		newval = kmalloc(val_len + 1, GFP_NOFS);
@@ -777,8 +779,8 @@ out:
 
 static int ceph_send_removexattr(struct dentry *dentry, const char *name)
 {
-	struct ceph_client *client = ceph_sb_to_client(dentry->d_sb);
+	struct ceph_fs_client *fsc = ceph_sb_to_client(dentry->d_sb);
-	struct ceph_mds_client *mdsc = &client->mdsc;
+	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;

include/linux/ceph/auth.h

@@ -1,8 +1,8 @@
 #ifndef _FS_CEPH_AUTH_H
 #define _FS_CEPH_AUTH_H
 
-#include "types.h"
+#include <linux/ceph/types.h>
-#include "buffer.h"
+#include <linux/ceph/buffer.h>
 
 /*
  * Abstract interface for communicating with the authenticate module.

include/linux/ceph/ceph_debug.h

@@ -3,7 +3,7 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#ifdef CONFIG_CEPH_FS_PRETTYDEBUG
+#ifdef CONFIG_CEPH_LIB_PRETTYDEBUG
 
 /*
  * wrap pr_debug to include a filename:lineno prefix on each line.
@@ -14,7 +14,8 @@
 # if defined(DEBUG) || defined(CONFIG_DYNAMIC_DEBUG)
 extern const char *ceph_file_part(const char *s, int len);
 #  define dout(fmt, ...)						\
-	pr_debug(" %12.12s:%-4d : " fmt,				\
+	pr_debug("%.*s %12.12s:%-4d : " fmt,				\
+		 8 - (int)sizeof(KBUILD_MODNAME), "    ",		\
 		 ceph_file_part(__FILE__, sizeof(__FILE__)),		\
 		 __LINE__, ##__VA_ARGS__)
 # else

include/linux/ceph/ceph_fs.h

@@ -299,6 +299,7 @@ enum {
 	CEPH_MDS_OP_SETATTR    = 0x01108,
 	CEPH_MDS_OP_SETFILELOCK= 0x01109,
 	CEPH_MDS_OP_GETFILELOCK= 0x00110,
+	CEPH_MDS_OP_SETDIRLAYOUT=0x0110a,
 
 	CEPH_MDS_OP_MKNOD      = 0x01201,
 	CEPH_MDS_OP_LINK       = 0x01202,

include/linux/ceph/debugfs.h

@@ -0,0 +1,33 @@
+#ifndef _FS_CEPH_DEBUGFS_H
+#define _FS_CEPH_DEBUGFS_H
+
+#include "ceph_debug.h"
+#include "types.h"
+
+#define CEPH_DEFINE_SHOW_FUNC(name)					\
+static int name##_open(struct inode *inode, struct file *file)		\
+{									\
+	struct seq_file *sf;						\
+	int ret;							\
+									\
+	ret = single_open(file, name, NULL);				\
+	sf = file->private_data;					\
+	sf->private = inode->i_private;					\
+	return ret;							\
+}									\
+									\
+static const struct file_operations name##_fops = {			\
+	.open		= name##_open,					\
+	.read		= seq_read,					\
+	.llseek		= seq_lseek,					\
+	.release	= single_release,				\
+};
+
+/* debugfs.c */
+extern int ceph_debugfs_init(void);
+extern void ceph_debugfs_cleanup(void);
+extern int ceph_debugfs_client_init(struct ceph_client *client);
+extern void ceph_debugfs_client_cleanup(struct ceph_client *client);
+
+#endif
+

include/linux/ceph/decode.h

@@ -191,6 +191,11 @@ static inline void ceph_encode_string(void **p, void *end,
 		ceph_encode_need(p, end, n, bad);		\
 		ceph_encode_copy(p, pv, n);			\
 	} while (0)
+#define ceph_encode_string_safe(p, end, s, n, bad)		\
+	do {							\
+		ceph_encode_need(p, end, n, bad);		\
+		ceph_encode_string(p, end, s, n);		\
+	} while (0)
 
 
 #endif

include/linux/ceph/libceph.h

@@ -0,0 +1,249 @@
+#ifndef _FS_CEPH_LIBCEPH_H
+#define _FS_CEPH_LIBCEPH_H
+
+#include "ceph_debug.h"
+
+#include <asm/unaligned.h>
+#include <linux/backing-dev.h>
+#include <linux/completion.h>
+#include <linux/exportfs.h>
+#include <linux/fs.h>
+#include <linux/mempool.h>
+#include <linux/pagemap.h>
+#include <linux/wait.h>
+#include <linux/writeback.h>
+#include <linux/slab.h>
+
+#include "types.h"
+#include "messenger.h"
+#include "msgpool.h"
+#include "mon_client.h"
+#include "osd_client.h"
+#include "ceph_fs.h"
+
+/*
+ * Supported features
+ */
+#define CEPH_FEATURE_SUPPORTED_DEFAULT CEPH_FEATURE_NOSRCADDR
+#define CEPH_FEATURE_REQUIRED_DEFAULT  CEPH_FEATURE_NOSRCADDR
+
+/*
+ * mount options
+ */
+#define CEPH_OPT_FSID             (1<<0)
+#define CEPH_OPT_NOSHARE          (1<<1) /* don't share client with other sbs */
+#define CEPH_OPT_MYIP             (1<<2) /* specified my ip */
+#define CEPH_OPT_NOCRC            (1<<3) /* no data crc on writes */
+
+#define CEPH_OPT_DEFAULT   (0);
+
+#define ceph_set_opt(client, opt) \
+	(client)->options->flags |= CEPH_OPT_##opt;
+#define ceph_test_opt(client, opt) \
+	(!!((client)->options->flags & CEPH_OPT_##opt))
+
+struct ceph_options {
+	int flags;
+	struct ceph_fsid fsid;
+	struct ceph_entity_addr my_addr;
+	int mount_timeout;
+	int osd_idle_ttl;
+	int osd_timeout;
+	int osd_keepalive_timeout;
+
+	/*
+	 * any type that can't be simply compared or doesn't need need
+	 * to be compared should go beyond this point,
+	 * ceph_compare_options() should be updated accordingly
+	 */
+
+	struct ceph_entity_addr *mon_addr; /* should be the first
+					      pointer type of args */
+	int num_mon;
+	char *name;
+	char *secret;
+};
+
+/*
+ * defaults
+ */
+#define CEPH_MOUNT_TIMEOUT_DEFAULT  60
+#define CEPH_OSD_TIMEOUT_DEFAULT    60  /* seconds */
+#define CEPH_OSD_KEEPALIVE_DEFAULT  5
+#define CEPH_OSD_IDLE_TTL_DEFAULT    60
+#define CEPH_MOUNT_RSIZE_DEFAULT    (512*1024) /* readahead */
+
+#define CEPH_MSG_MAX_FRONT_LEN	(16*1024*1024)
+#define CEPH_MSG_MAX_DATA_LEN	(16*1024*1024)
+
+#define CEPH_AUTH_NAME_DEFAULT   "guest"
+
+/*
+ * Delay telling the MDS we no longer want caps, in case we reopen
+ * the file.  Delay a minimum amount of time, even if we send a cap
+ * message for some other reason.  Otherwise, take the oppotunity to
+ * update the mds to avoid sending another message later.
+ */
+#define CEPH_CAPS_WANTED_DELAY_MIN_DEFAULT      5  /* cap release delay */
+#define CEPH_CAPS_WANTED_DELAY_MAX_DEFAULT     60  /* cap release delay */
+
+#define CEPH_CAP_RELEASE_SAFETY_DEFAULT        (CEPH_CAPS_PER_RELEASE * 4)
+
+/* mount state */
+enum {
+	CEPH_MOUNT_MOUNTING,
+	CEPH_MOUNT_MOUNTED,
+	CEPH_MOUNT_UNMOUNTING,
+	CEPH_MOUNT_UNMOUNTED,
+	CEPH_MOUNT_SHUTDOWN,
+};
+
+/*
+ * subtract jiffies
+ */
+static inline unsigned long time_sub(unsigned long a, unsigned long b)
+{
+	BUG_ON(time_after(b, a));
+	return (long)a - (long)b;
+}
+
+struct ceph_mds_client;
+
+/*
+ * per client state
+ *
+ * possibly shared by multiple mount points, if they are
+ * mounting the same ceph filesystem/cluster.
+ */
+struct ceph_client {
+	struct ceph_fsid fsid;
+	bool have_fsid;
+
+	void *private;
+
+	struct ceph_options *options;
+
+	struct mutex mount_mutex;      /* serialize mount attempts */
+	wait_queue_head_t auth_wq;
+	int auth_err;
+
+	int (*extra_mon_dispatch)(struct ceph_client *, struct ceph_msg *);
+
+	u32 supported_features;
+	u32 required_features;
+
+	struct ceph_messenger *msgr;   /* messenger instance */
+	struct ceph_mon_client monc;
+	struct ceph_osd_client osdc;
+
+#ifdef CONFIG_DEBUG_FS
+	struct dentry *debugfs_dir;
+	struct dentry *debugfs_monmap;
+	struct dentry *debugfs_osdmap;
+#endif
+};
+
+
+
+/*
+ * snapshots
+ */
+
+/*
+ * A "snap context" is the set of existing snapshots when we
+ * write data.  It is used by the OSD to guide its COW behavior.
+ *
+ * The ceph_snap_context is refcounted, and attached to each dirty
+ * page, indicating which context the dirty data belonged when it was
+ * dirtied.
+ */
+struct ceph_snap_context {
+	atomic_t nref;
+	u64 seq;
+	int num_snaps;
+	u64 snaps[];
+};
+
+static inline struct ceph_snap_context *
+ceph_get_snap_context(struct ceph_snap_context *sc)
+{
+	/*
+	printk("get_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
+	       atomic_read(&sc->nref)+1);
+	*/
+	if (sc)
+		atomic_inc(&sc->nref);
+	return sc;
+}
+
+static inline void ceph_put_snap_context(struct ceph_snap_context *sc)
+{
+	if (!sc)
+		return;
+	/*
+	printk("put_snap_context %p %d -> %d\n", sc, atomic_read(&sc->nref),
+	       atomic_read(&sc->nref)-1);
+	*/
+	if (atomic_dec_and_test(&sc->nref)) {
+		/*printk(" deleting snap_context %p\n", sc);*/
+		kfree(sc);
+	}
+}
+
+/*
+ * calculate the number of pages a given length and offset map onto,
+ * if we align the data.
+ */
+static inline int calc_pages_for(u64 off, u64 len)
+{
+	return ((off+len+PAGE_CACHE_SIZE-1) >> PAGE_CACHE_SHIFT) -
+		(off >> PAGE_CACHE_SHIFT);
+}
+
+/* ceph_common.c */
+extern const char *ceph_msg_type_name(int type);
+extern int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid);
+extern struct kmem_cache *ceph_inode_cachep;
+extern struct kmem_cache *ceph_cap_cachep;
+extern struct kmem_cache *ceph_dentry_cachep;
+extern struct kmem_cache *ceph_file_cachep;
+
+extern int ceph_parse_options(struct ceph_options **popt, char *options,
+			      const char *dev_name, const char *dev_name_end,
+			      int (*parse_extra_token)(char *c, void *private),
+			      void *private);
+extern void ceph_destroy_options(struct ceph_options *opt);
+extern int ceph_compare_options(struct ceph_options *new_opt,
+				struct ceph_client *client);
+extern struct ceph_client *ceph_create_client(struct ceph_options *opt,
+					      void *private);
+extern u64 ceph_client_id(struct ceph_client *client);
+extern void ceph_destroy_client(struct ceph_client *client);
+extern int __ceph_open_session(struct ceph_client *client,
+			       unsigned long started);
+extern int ceph_open_session(struct ceph_client *client);
+
+/* pagevec.c */
+extern void ceph_release_page_vector(struct page **pages, int num_pages);
+
+extern struct page **ceph_get_direct_page_vector(const char __user *data,
+					    int num_pages,
+					    loff_t off, size_t len);
+extern void ceph_put_page_vector(struct page **pages, int num_pages);
+extern void ceph_release_page_vector(struct page **pages, int num_pages);
+extern struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags);
+extern int ceph_copy_user_to_page_vector(struct page **pages,
+					 const char __user *data,
+					 loff_t off, size_t len);
+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 int ceph_copy_page_vector_to_user(struct page **pages, char __user *data,
+				    loff_t off, size_t len);
+extern void ceph_zero_page_vector_range(int off, int len, struct page **pages);
+
+
+#endif /* _FS_CEPH_SUPER_H */

include/linux/ceph/messenger.h

@@ -65,6 +65,9 @@ struct ceph_messenger {
 	 */
 	u32 global_seq;
 	spinlock_t global_seq_lock;
+
+	u32 supported_features;
+	u32 required_features;
 };
 
 /*
@@ -82,6 +85,10 @@ struct ceph_msg {
 	struct ceph_pagelist *pagelist; /* instead of pages */
 	struct list_head list_head;
 	struct kref kref;
+	struct bio  *bio;		/* instead of pages/pagelist */
+	struct bio  *bio_iter;		/* bio iterator */
+	int bio_seg;			/* current bio segment */
+	struct ceph_pagelist *trail;	/* the trailing part of the data */
 	bool front_is_vmalloc;
 	bool more_to_follow;
 	bool needs_out_seq;
@@ -205,7 +212,7 @@ struct ceph_connection {
 };
 
 
-extern const char *pr_addr(const struct sockaddr_storage *ss);
+extern const char *ceph_pr_addr(const struct sockaddr_storage *ss);
 extern int ceph_parse_ips(const char *c, const char *end,
 			  struct ceph_entity_addr *addr,
 			  int max_count, int *count);
@@ -216,7 +223,8 @@ extern void ceph_msgr_exit(void);
 extern void ceph_msgr_flush(void);
 
 extern struct ceph_messenger *ceph_messenger_create(
-	struct ceph_entity_addr *myaddr);
+	struct ceph_entity_addr *myaddr,
+	u32 features, u32 required);
 extern void ceph_messenger_destroy(struct ceph_messenger *);
 
 extern void ceph_con_init(struct ceph_messenger *msgr,

include/linux/ceph/mon_client.h

@@ -79,6 +79,7 @@ struct ceph_mon_client {
 	u64 last_tid;
 
 	/* mds/osd map */
+	int want_mdsmap;
 	int want_next_osdmap; /* 1 = want, 2 = want+asked */
 	u32 have_osdmap, have_mdsmap;
 

include/linux/ceph/osd_client.h

@@ -15,6 +15,7 @@ struct ceph_snap_context;
 struct ceph_osd_request;
 struct ceph_osd_client;
 struct ceph_authorizer;
+struct ceph_pagelist;
 
 /*
  * completion callback for async writepages
@@ -68,6 +69,7 @@ struct ceph_osd_request {
 	struct list_head  r_unsafe_item;
 
 	struct inode *r_inode;         	      /* for use by callbacks */
+	void *r_priv;			      /* ditto */
 
 	char              r_oid[40];          /* object name */
 	int               r_oid_len;
@@ -80,6 +82,11 @@ struct ceph_osd_request {
 	struct page     **r_pages;            /* pages for data payload */
 	int               r_pages_from_pool;
 	int               r_own_pages;        /* if true, i own page list */
+#ifdef CONFIG_BLOCK
+	struct bio       *r_bio;	      /* instead of pages */
+#endif
+
+	struct ceph_pagelist *r_trail;	      /* trailing part of the data */
 };
 
 struct ceph_osd_client {
@@ -110,6 +117,42 @@ struct ceph_osd_client {
 	struct ceph_msgpool	msgpool_op_reply;
 };
 
+struct ceph_osd_req_op {
+	u16 op;           /* CEPH_OSD_OP_* */
+	u32 flags;        /* CEPH_OSD_FLAG_* */
+	union {
+		struct {
+			u64 offset, length;
+			u64 truncate_size;
+			u32 truncate_seq;
+		} extent;
+		struct {
+			const char *name;
+			u32 name_len;
+			const char  *val;
+			u32 value_len;
+			__u8 cmp_op;       /* CEPH_OSD_CMPXATTR_OP_* */
+			__u8 cmp_mode;     /* CEPH_OSD_CMPXATTR_MODE_* */
+		} xattr;
+		struct {
+			const char *class_name;
+			__u8 class_len;
+			const char *method_name;
+			__u8 method_len;
+			__u8 argc;
+			const char *indata;
+			u32 indata_len;
+		} cls;
+		struct {
+			u64 cookie, count;
+		} pgls;
+	        struct {
+		        u64 snapid;
+	        } snap;
+	};
+	u32 payload_len;
+};
+
 extern int ceph_osdc_init(struct ceph_osd_client *osdc,
 			  struct ceph_client *client);
 extern void ceph_osdc_stop(struct ceph_osd_client *osdc);
@@ -119,6 +162,30 @@ extern void ceph_osdc_handle_reply(struct ceph_osd_client *osdc,
 extern void ceph_osdc_handle_map(struct ceph_osd_client *osdc,
 				 struct ceph_msg *msg);
 
+extern void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
+			struct ceph_file_layout *layout,
+			u64 snapid,
+			u64 off, u64 *plen, u64 *bno,
+			struct ceph_osd_request *req,
+			struct ceph_osd_req_op *op);
+
+extern struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
+					       int flags,
+					       struct ceph_snap_context *snapc,
+					       struct ceph_osd_req_op *ops,
+					       bool use_mempool,
+					       gfp_t gfp_flags,
+					       struct page **pages,
+					       struct bio *bio);
+
+extern void ceph_osdc_build_request(struct ceph_osd_request *req,
+				    u64 off, u64 *plen,
+				    struct ceph_osd_req_op *src_ops,
+				    struct ceph_snap_context *snapc,
+				    struct timespec *mtime,
+				    const char *oid,
+				    int oid_len);
+
 extern struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *,
 				      struct ceph_file_layout *layout,
 				      struct ceph_vino vino,

include/linux/ceph/osdmap.h

@@ -4,7 +4,7 @@
 #include <linux/rbtree.h>
 #include "types.h"
 #include "ceph_fs.h"
-#include "crush/crush.h"
+#include <linux/crush/crush.h>
 
 /*
  * The osd map describes the current membership of the osd cluster and
@@ -125,4 +125,6 @@ extern int ceph_calc_pg_acting(struct ceph_osdmap *osdmap, struct ceph_pg pgid,
 extern int ceph_calc_pg_primary(struct ceph_osdmap *osdmap,
 				struct ceph_pg pgid);
 
+extern int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name);
+
 #endif

include/linux/ceph/pagelist.h

@@ -8,6 +8,14 @@ struct ceph_pagelist {
 	void *mapped_tail;
 	size_t length;
 	size_t room;
+	struct list_head free_list;
+	size_t num_pages_free;
+};
+
+struct ceph_pagelist_cursor {
+	struct ceph_pagelist *pl;   /* pagelist, for error checking */
+	struct list_head *page_lru; /* page in list */
+	size_t room;		    /* room remaining to reset to */
 };
 
 static inline void ceph_pagelist_init(struct ceph_pagelist *pl)
@@ -16,10 +24,23 @@ static inline void ceph_pagelist_init(struct ceph_pagelist *pl)
 	pl->mapped_tail = NULL;
 	pl->length = 0;
 	pl->room = 0;
+	INIT_LIST_HEAD(&pl->free_list);
+	pl->num_pages_free = 0;
 }
+
 extern int ceph_pagelist_release(struct ceph_pagelist *pl);
 
-extern int ceph_pagelist_append(struct ceph_pagelist *pl, void *d, size_t l);
+extern int ceph_pagelist_append(struct ceph_pagelist *pl, const void *d, size_t l);
+
+extern int ceph_pagelist_reserve(struct ceph_pagelist *pl, size_t space);
+
+extern int ceph_pagelist_free_reserve(struct ceph_pagelist *pl);
+
+extern void ceph_pagelist_set_cursor(struct ceph_pagelist *pl,
+				     struct ceph_pagelist_cursor *c);
+
+extern int ceph_pagelist_truncate(struct ceph_pagelist *pl,
+				  struct ceph_pagelist_cursor *c);
 
 static inline int ceph_pagelist_encode_64(struct ceph_pagelist *pl, u64 v)
 {

net/Kconfig

@@ -293,6 +293,7 @@ source "net/wimax/Kconfig"
 source "net/rfkill/Kconfig"
 source "net/9p/Kconfig"
 source "net/caif/Kconfig"
+source "net/ceph/Kconfig"
 
 
 endif   # if NET

net/Makefile

@@ -68,3 +68,4 @@ obj-$(CONFIG_SYSCTL)		+= sysctl_net.o
 endif
 obj-$(CONFIG_WIMAX)		+= wimax/
 obj-$(CONFIG_DNS_RESOLVER)	+= dns_resolver/
+obj-$(CONFIG_CEPH_LIB)		+= ceph/

net/ceph/Kconfig

@@ -0,0 +1,28 @@
+config CEPH_LIB
+        tristate "Ceph core library (EXPERIMENTAL)"
+	depends on INET && EXPERIMENTAL
+	select LIBCRC32C
+	select CRYPTO_AES
+	select CRYPTO
+	default n
+	help
+	  Choose Y or M here to include cephlib, which provides the
+	  common functionality to both the Ceph filesystem and
+	  to the rados block device (rbd).
+
+	  More information at http://ceph.newdream.net/.
+
+	  If unsure, say N.
+
+config CEPH_LIB_PRETTYDEBUG
+	bool "Include file:line in ceph debug output"
+	depends on CEPH_LIB
+	default n
+	help
+	  If you say Y here, debug output will include a filename and
+	  line to aid debugging.  This increases kernel size and slows
+	  execution slightly when debug call sites are enabled (e.g.,
+	  via CONFIG_DYNAMIC_DEBUG).
+
+	  If unsure, say N.
+

net/ceph/Makefile

@@ -0,0 +1,37 @@
+#
+# Makefile for CEPH filesystem.
+#
+
+ifneq ($(KERNELRELEASE),)
+
+obj-$(CONFIG_CEPH_LIB) += libceph.o
+
+libceph-objs := ceph_common.o messenger.o msgpool.o buffer.o pagelist.o \
+	mon_client.o \
+	osd_client.o osdmap.o crush/crush.o crush/mapper.o crush/hash.o \
+	debugfs.o \
+	auth.o auth_none.o \
+	crypto.o armor.o \
+	auth_x.o \
+	ceph_fs.o ceph_strings.o ceph_hash.o \
+	pagevec.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_LIB=m modules
+
+modules_install:
+	$(MAKE) -C $(KERNELDIR) M=$(PWD) CONFIG_CEPH_LIB=m modules_install
+
+clean:
+	$(MAKE) -C $(KERNELDIR) M=$(PWD) clean
+
+endif

net/ceph/auth.c

@@ -1,16 +1,16 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/module.h>
 #include <linux/err.h>
 #include <linux/slab.h>
 
-#include "types.h"
+#include <linux/ceph/types.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/messenger.h>
 #include "auth_none.h"
 #include "auth_x.h"
-#include "decode.h"
-#include "super.h"
 
-#include "messenger.h"
 
 /*
  * get protocol handler

net/ceph/auth_none.c

@@ -1,14 +1,15 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/err.h>
 #include <linux/module.h>
 #include <linux/random.h>
 #include <linux/slab.h>
 
+#include <linux/ceph/decode.h>
+#include <linux/ceph/auth.h>
+
 #include "auth_none.h"
-#include "auth.h"
-#include "decode.h"
 
 static void reset(struct ceph_auth_client *ac)
 {

net/ceph/auth_none.h

@@ -2,8 +2,7 @@
 #define _FS_CEPH_AUTH_NONE_H
 
 #include <linux/slab.h>
-
+#include <linux/ceph/auth.h>
-#include "auth.h"
 
 /*
  * null security mode.

net/ceph/auth_x.c

@@ -1,16 +1,17 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/err.h>
 #include <linux/module.h>
 #include <linux/random.h>
 #include <linux/slab.h>
 
+#include <linux/ceph/decode.h>
+#include <linux/ceph/auth.h>
+
+#include "crypto.h"
 #include "auth_x.h"
 #include "auth_x_protocol.h"
-#include "crypto.h"
-#include "auth.h"
-#include "decode.h"
 
 #define TEMP_TICKET_BUF_LEN	256
 

net/ceph/auth_x.h

@@ -3,8 +3,9 @@
 
 #include <linux/rbtree.h>
 
+#include <linux/ceph/auth.h>
+
 #include "crypto.h"
-#include "auth.h"
 #include "auth_x_protocol.h"
 
 /*

net/ceph/buffer.c

@@ -1,10 +1,11 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/slab.h>
 
-#include "buffer.h"
+#include <linux/ceph/buffer.h>
-#include "decode.h"
+#include <linux/ceph/decode.h>
 
 struct ceph_buffer *ceph_buffer_new(size_t len, gfp_t gfp)
 {
@@ -32,6 +33,7 @@ struct ceph_buffer *ceph_buffer_new(size_t len, gfp_t gfp)
 	dout("buffer_new %p\n", b);
 	return b;
 }
+EXPORT_SYMBOL(ceph_buffer_new);
 
 void ceph_buffer_release(struct kref *kref)
 {
@@ -46,6 +48,7 @@ void ceph_buffer_release(struct kref *kref)
 	}
 	kfree(b);
 }
+EXPORT_SYMBOL(ceph_buffer_release);
 
 int ceph_decode_buffer(struct ceph_buffer **b, void **p, void *end)
 {

net/ceph/ceph_common.c

@@ -0,0 +1,529 @@
+
+#include <linux/ceph/ceph_debug.h>
+#include <linux/backing-dev.h>
+#include <linux/ctype.h>
+#include <linux/fs.h>
+#include <linux/inet.h>
+#include <linux/in6.h>
+#include <linux/module.h>
+#include <linux/mount.h>
+#include <linux/parser.h>
+#include <linux/sched.h>
+#include <linux/seq_file.h>
+#include <linux/slab.h>
+#include <linux/statfs.h>
+#include <linux/string.h>
+
+
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/debugfs.h>
+#include <linux/ceph/decode.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+
+
+
+/*
+ * find filename portion of a path (/foo/bar/baz -> baz)
+ */
+const char *ceph_file_part(const char *s, int len)
+{
+	const char *e = s + len;
+
+	while (e != s && *(e-1) != '/')
+		e--;
+	return e;
+}
+EXPORT_SYMBOL(ceph_file_part);
+
+const char *ceph_msg_type_name(int type)
+{
+	switch (type) {
+	case CEPH_MSG_SHUTDOWN: return "shutdown";
+	case CEPH_MSG_PING: return "ping";
+	case CEPH_MSG_AUTH: return "auth";
+	case CEPH_MSG_AUTH_REPLY: return "auth_reply";
+	case CEPH_MSG_MON_MAP: return "mon_map";
+	case CEPH_MSG_MON_GET_MAP: return "mon_get_map";
+	case CEPH_MSG_MON_SUBSCRIBE: return "mon_subscribe";
+	case CEPH_MSG_MON_SUBSCRIBE_ACK: return "mon_subscribe_ack";
+	case CEPH_MSG_STATFS: return "statfs";
+	case CEPH_MSG_STATFS_REPLY: return "statfs_reply";
+	case CEPH_MSG_MDS_MAP: return "mds_map";
+	case CEPH_MSG_CLIENT_SESSION: return "client_session";
+	case CEPH_MSG_CLIENT_RECONNECT: return "client_reconnect";
+	case CEPH_MSG_CLIENT_REQUEST: return "client_request";
+	case CEPH_MSG_CLIENT_REQUEST_FORWARD: return "client_request_forward";
+	case CEPH_MSG_CLIENT_REPLY: return "client_reply";
+	case CEPH_MSG_CLIENT_CAPS: return "client_caps";
+	case CEPH_MSG_CLIENT_CAPRELEASE: return "client_cap_release";
+	case CEPH_MSG_CLIENT_SNAP: return "client_snap";
+	case CEPH_MSG_CLIENT_LEASE: return "client_lease";
+	case CEPH_MSG_OSD_MAP: return "osd_map";
+	case CEPH_MSG_OSD_OP: return "osd_op";
+	case CEPH_MSG_OSD_OPREPLY: return "osd_opreply";
+	default: return "unknown";
+	}
+}
+EXPORT_SYMBOL(ceph_msg_type_name);
+
+/*
+ * Initially learn our fsid, or verify an fsid matches.
+ */
+int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid)
+{
+	if (client->have_fsid) {
+		if (ceph_fsid_compare(&client->fsid, fsid)) {
+			pr_err("bad fsid, had %pU got %pU",
+			       &client->fsid, fsid);
+			return -1;
+		}
+	} else {
+		pr_info("client%lld fsid %pU\n", ceph_client_id(client), fsid);
+		memcpy(&client->fsid, fsid, sizeof(*fsid));
+		ceph_debugfs_client_init(client);
+		client->have_fsid = true;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_check_fsid);
+
+static int strcmp_null(const char *s1, const char *s2)
+{
+	if (!s1 && !s2)
+		return 0;
+	if (s1 && !s2)
+		return -1;
+	if (!s1 && s2)
+		return 1;
+	return strcmp(s1, s2);
+}
+
+int ceph_compare_options(struct ceph_options *new_opt,
+			 struct ceph_client *client)
+{
+	struct ceph_options *opt1 = new_opt;
+	struct ceph_options *opt2 = client->options;
+	int ofs = offsetof(struct ceph_options, mon_addr);
+	int i;
+	int ret;
+
+	ret = memcmp(opt1, opt2, ofs);
+	if (ret)
+		return ret;
+
+	ret = strcmp_null(opt1->name, opt2->name);
+	if (ret)
+		return ret;
+
+	ret = strcmp_null(opt1->secret, opt2->secret);
+	if (ret)
+		return ret;
+
+	/* any matching mon ip implies a match */
+	for (i = 0; i < opt1->num_mon; i++) {
+		if (ceph_monmap_contains(client->monc.monmap,
+				 &opt1->mon_addr[i]))
+			return 0;
+	}
+	return -1;
+}
+EXPORT_SYMBOL(ceph_compare_options);
+
+
+static int parse_fsid(const char *str, struct ceph_fsid *fsid)
+{
+	int i = 0;
+	char tmp[3];
+	int err = -EINVAL;
+	int d;
+
+	dout("parse_fsid '%s'\n", str);
+	tmp[2] = 0;
+	while (*str && i < 16) {
+		if (ispunct(*str)) {
+			str++;
+			continue;
+		}
+		if (!isxdigit(str[0]) || !isxdigit(str[1]))
+			break;
+		tmp[0] = str[0];
+		tmp[1] = str[1];
+		if (sscanf(tmp, "%x", &d) < 1)
+			break;
+		fsid->fsid[i] = d & 0xff;
+		i++;
+		str += 2;
+	}
+
+	if (i == 16)
+		err = 0;
+	dout("parse_fsid ret %d got fsid %pU", err, fsid);
+	return err;
+}
+
+/*
+ * ceph options
+ */
+enum {
+	Opt_osdtimeout,
+	Opt_osdkeepalivetimeout,
+	Opt_mount_timeout,
+	Opt_osd_idle_ttl,
+	Opt_last_int,
+	/* int args above */
+	Opt_fsid,
+	Opt_name,
+	Opt_secret,
+	Opt_ip,
+	Opt_last_string,
+	/* string args above */
+	Opt_noshare,
+	Opt_nocrc,
+};
+
+static match_table_t opt_tokens = {
+	{Opt_osdtimeout, "osdtimeout=%d"},
+	{Opt_osdkeepalivetimeout, "osdkeepalive=%d"},
+	{Opt_mount_timeout, "mount_timeout=%d"},
+	{Opt_osd_idle_ttl, "osd_idle_ttl=%d"},
+	/* int args above */
+	{Opt_fsid, "fsid=%s"},
+	{Opt_name, "name=%s"},
+	{Opt_secret, "secret=%s"},
+	{Opt_ip, "ip=%s"},
+	/* string args above */
+	{Opt_noshare, "noshare"},
+	{Opt_nocrc, "nocrc"},
+	{-1, NULL}
+};
+
+void ceph_destroy_options(struct ceph_options *opt)
+{
+	dout("destroy_options %p\n", opt);
+	kfree(opt->name);
+	kfree(opt->secret);
+	kfree(opt);
+}
+EXPORT_SYMBOL(ceph_destroy_options);
+
+int ceph_parse_options(struct ceph_options **popt, char *options,
+		       const char *dev_name, const char *dev_name_end,
+		       int (*parse_extra_token)(char *c, void *private),
+		       void *private)
+{
+	struct ceph_options *opt;
+	const char *c;
+	int err = -ENOMEM;
+	substring_t argstr[MAX_OPT_ARGS];
+
+	opt = kzalloc(sizeof(*opt), GFP_KERNEL);
+	if (!opt)
+		return err;
+	opt->mon_addr = kcalloc(CEPH_MAX_MON, sizeof(*opt->mon_addr),
+				GFP_KERNEL);
+	if (!opt->mon_addr)
+		goto out;
+
+	dout("parse_options %p options '%s' dev_name '%s'\n", opt, options,
+	     dev_name);
+
+	/* start with defaults */
+	opt->flags = CEPH_OPT_DEFAULT;
+	opt->osd_timeout = CEPH_OSD_TIMEOUT_DEFAULT;
+	opt->osd_keepalive_timeout = CEPH_OSD_KEEPALIVE_DEFAULT;
+	opt->mount_timeout = CEPH_MOUNT_TIMEOUT_DEFAULT; /* seconds */
+	opt->osd_idle_ttl = CEPH_OSD_IDLE_TTL_DEFAULT;   /* seconds */
+
+	/* get mon ip(s) */
+	/* ip1[:port1][,ip2[:port2]...] */
+	err = ceph_parse_ips(dev_name, dev_name_end, opt->mon_addr,
+			     CEPH_MAX_MON, &opt->num_mon);
+	if (err < 0)
+		goto out;
+
+	/* parse mount options */
+	while ((c = strsep(&options, ",")) != NULL) {
+		int token, intval, ret;
+		if (!*c)
+			continue;
+		err = -EINVAL;
+		token = match_token((char *)c, opt_tokens, argstr);
+		if (token < 0 && parse_extra_token) {
+			/* extra? */
+			err = parse_extra_token((char *)c, private);
+			if (err < 0) {
+				pr_err("bad option at '%s'\n", c);
+				goto out;
+			}
+			continue;
+		}
+		if (token < Opt_last_int) {
+			ret = match_int(&argstr[0], &intval);
+			if (ret < 0) {
+				pr_err("bad mount option arg (not int) "
+				       "at '%s'\n", c);
+				continue;
+			}
+			dout("got int token %d val %d\n", token, intval);
+		} else if (token > Opt_last_int && token < Opt_last_string) {
+			dout("got string token %d val %s\n", token,
+			     argstr[0].from);
+		} else {
+			dout("got token %d\n", token);
+		}
+		switch (token) {
+		case Opt_ip:
+			err = ceph_parse_ips(argstr[0].from,
+					     argstr[0].to,
+					     &opt->my_addr,
+					     1, NULL);
+			if (err < 0)
+				goto out;
+			opt->flags |= CEPH_OPT_MYIP;
+			break;
+
+		case Opt_fsid:
+			err = parse_fsid(argstr[0].from, &opt->fsid);
+			if (err == 0)
+				opt->flags |= CEPH_OPT_FSID;
+			break;
+		case Opt_name:
+			opt->name = kstrndup(argstr[0].from,
+					      argstr[0].to-argstr[0].from,
+					      GFP_KERNEL);
+			break;
+		case Opt_secret:
+			opt->secret = kstrndup(argstr[0].from,
+						argstr[0].to-argstr[0].from,
+						GFP_KERNEL);
+			break;
+
+			/* misc */
+		case Opt_osdtimeout:
+			opt->osd_timeout = intval;
+			break;
+		case Opt_osdkeepalivetimeout:
+			opt->osd_keepalive_timeout = intval;
+			break;
+		case Opt_osd_idle_ttl:
+			opt->osd_idle_ttl = intval;
+			break;
+		case Opt_mount_timeout:
+			opt->mount_timeout = intval;
+			break;
+
+		case Opt_noshare:
+			opt->flags |= CEPH_OPT_NOSHARE;
+			break;
+
+		case Opt_nocrc:
+			opt->flags |= CEPH_OPT_NOCRC;
+			break;
+
+		default:
+			BUG_ON(token);
+		}
+	}
+
+	/* success */
+	*popt = opt;
+	return 0;
+
+out:
+	ceph_destroy_options(opt);
+	return err;
+}
+EXPORT_SYMBOL(ceph_parse_options);
+
+u64 ceph_client_id(struct ceph_client *client)
+{
+	return client->monc.auth->global_id;
+}
+EXPORT_SYMBOL(ceph_client_id);
+
+/*
+ * create a fresh client instance
+ */
+struct ceph_client *ceph_create_client(struct ceph_options *opt, void *private)
+{
+	struct ceph_client *client;
+	int err = -ENOMEM;
+
+	client = kzalloc(sizeof(*client), GFP_KERNEL);
+	if (client == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	client->private = private;
+	client->options = opt;
+
+	mutex_init(&client->mount_mutex);
+	init_waitqueue_head(&client->auth_wq);
+	client->auth_err = 0;
+
+	client->extra_mon_dispatch = NULL;
+	client->supported_features = CEPH_FEATURE_SUPPORTED_DEFAULT;
+	client->required_features = CEPH_FEATURE_REQUIRED_DEFAULT;
+
+	client->msgr = NULL;
+
+	/* subsystems */
+	err = ceph_monc_init(&client->monc, client);
+	if (err < 0)
+		goto fail;
+	err = ceph_osdc_init(&client->osdc, client);
+	if (err < 0)
+		goto fail_monc;
+
+	return client;
+
+fail_monc:
+	ceph_monc_stop(&client->monc);
+fail:
+	kfree(client);
+	return ERR_PTR(err);
+}
+EXPORT_SYMBOL(ceph_create_client);
+
+void ceph_destroy_client(struct ceph_client *client)
+{
+	dout("destroy_client %p\n", client);
+
+	/* unmount */
+	ceph_osdc_stop(&client->osdc);
+
+	/*
+	 * make sure mds and osd connections close out before destroying
+	 * the auth module, which is needed to free those connections'
+	 * ceph_authorizers.
+	 */
+	ceph_msgr_flush();
+
+	ceph_monc_stop(&client->monc);
+
+	ceph_debugfs_client_cleanup(client);
+
+	if (client->msgr)
+		ceph_messenger_destroy(client->msgr);
+
+	ceph_destroy_options(client->options);
+
+	kfree(client);
+	dout("destroy_client %p done\n", client);
+}
+EXPORT_SYMBOL(ceph_destroy_client);
+
+/*
+ * true if we have the mon map (and have thus joined the cluster)
+ */
+static int have_mon_and_osd_map(struct ceph_client *client)
+{
+	return client->monc.monmap && client->monc.monmap->epoch &&
+	       client->osdc.osdmap && client->osdc.osdmap->epoch;
+}
+
+/*
+ * mount: join the ceph cluster, and open root directory.
+ */
+int __ceph_open_session(struct ceph_client *client, unsigned long started)
+{
+	struct ceph_entity_addr *myaddr = NULL;
+	int err;
+	unsigned long timeout = client->options->mount_timeout * HZ;
+
+	/* initialize the messenger */
+	if (client->msgr == NULL) {
+		if (ceph_test_opt(client, MYIP))
+			myaddr = &client->options->my_addr;
+		client->msgr = ceph_messenger_create(myaddr,
+					client->supported_features,
+					client->required_features);
+		if (IS_ERR(client->msgr)) {
+			client->msgr = NULL;
+			return PTR_ERR(client->msgr);
+		}
+		client->msgr->nocrc = ceph_test_opt(client, NOCRC);
+	}
+
+	/* open session, and wait for mon and osd maps */
+	err = ceph_monc_open_session(&client->monc);
+	if (err < 0)
+		return err;
+
+	while (!have_mon_and_osd_map(client)) {
+		err = -EIO;
+		if (timeout && time_after_eq(jiffies, started + timeout))
+			return err;
+
+		/* wait */
+		dout("mount waiting for mon_map\n");
+		err = wait_event_interruptible_timeout(client->auth_wq,
+			have_mon_and_osd_map(client) || (client->auth_err < 0),
+			timeout);
+		if (err == -EINTR || err == -ERESTARTSYS)
+			return err;
+		if (client->auth_err < 0)
+			return client->auth_err;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(__ceph_open_session);
+
+
+int ceph_open_session(struct ceph_client *client)
+{
+	int ret;
+	unsigned long started = jiffies;  /* note the start time */
+
+	dout("open_session start\n");
+	mutex_lock(&client->mount_mutex);
+
+	ret = __ceph_open_session(client, started);
+
+	mutex_unlock(&client->mount_mutex);
+	return ret;
+}
+EXPORT_SYMBOL(ceph_open_session);
+
+
+static int __init init_ceph_lib(void)
+{
+	int ret = 0;
+
+	ret = ceph_debugfs_init();
+	if (ret < 0)
+		goto out;
+
+	ret = ceph_msgr_init();
+	if (ret < 0)
+		goto out_debugfs;
+
+	pr_info("loaded (mon/osd proto %d/%d, osdmap %d/%d %d/%d)\n",
+		CEPH_MONC_PROTOCOL, CEPH_OSDC_PROTOCOL,
+		CEPH_OSDMAP_VERSION, CEPH_OSDMAP_VERSION_EXT,
+		CEPH_OSDMAP_INC_VERSION, CEPH_OSDMAP_INC_VERSION_EXT);
+
+	return 0;
+
+out_debugfs:
+	ceph_debugfs_cleanup();
+out:
+	return ret;
+}
+
+static void __exit exit_ceph_lib(void)
+{
+	dout("exit_ceph_lib\n");
+	ceph_msgr_exit();
+	ceph_debugfs_cleanup();
+}
+
+module_init(init_ceph_lib);
+module_exit(exit_ceph_lib);
+
+MODULE_AUTHOR("Sage Weil <sage@newdream.net>");
+MODULE_AUTHOR("Yehuda Sadeh <yehuda@hq.newdream.net>");
+MODULE_AUTHOR("Patience Warnick <patience@newdream.net>");
+MODULE_DESCRIPTION("Ceph filesystem for Linux");
+MODULE_LICENSE("GPL");

net/ceph/ceph_fs.c

@@ -1,7 +1,8 @@
 /*
  * Some non-inline ceph helpers
  */
-#include "types.h"
+#include <linux/module.h>
+#include <linux/ceph/types.h>
 
 /*
  * return true if @layout appears to be valid
@@ -52,6 +53,7 @@ int ceph_flags_to_mode(int flags)
 
 	return mode;
 }
+EXPORT_SYMBOL(ceph_flags_to_mode);
 
 int ceph_caps_for_mode(int mode)
 {
@@ -70,3 +72,4 @@ int ceph_caps_for_mode(int mode)
 
 	return caps;
 }
+EXPORT_SYMBOL(ceph_caps_for_mode);

net/ceph/ceph_hash.c

@@ -1,5 +1,5 @@
 
-#include "types.h"
+#include <linux/ceph/types.h>
 
 /*
  * Robert Jenkin's hash function.

net/ceph/ceph_strings.c

@@ -0,0 +1,84 @@
+/*
+ * Ceph string constants
+ */
+#include <linux/module.h>
+#include <linux/ceph/types.h>
+
+const char *ceph_entity_type_name(int type)
+{
+	switch (type) {
+	case CEPH_ENTITY_TYPE_MDS: return "mds";
+	case CEPH_ENTITY_TYPE_OSD: return "osd";
+	case CEPH_ENTITY_TYPE_MON: return "mon";
+	case CEPH_ENTITY_TYPE_CLIENT: return "client";
+	case CEPH_ENTITY_TYPE_AUTH: return "auth";
+	default: return "unknown";
+	}
+}
+
+const char *ceph_osd_op_name(int op)
+{
+	switch (op) {
+	case CEPH_OSD_OP_READ: return "read";
+	case CEPH_OSD_OP_STAT: return "stat";
+
+	case CEPH_OSD_OP_MASKTRUNC: return "masktrunc";
+
+	case CEPH_OSD_OP_WRITE: return "write";
+	case CEPH_OSD_OP_DELETE: return "delete";
+	case CEPH_OSD_OP_TRUNCATE: return "truncate";
+	case CEPH_OSD_OP_ZERO: return "zero";
+	case CEPH_OSD_OP_WRITEFULL: return "writefull";
+	case CEPH_OSD_OP_ROLLBACK: return "rollback";
+
+	case CEPH_OSD_OP_APPEND: return "append";
+	case CEPH_OSD_OP_STARTSYNC: return "startsync";
+	case CEPH_OSD_OP_SETTRUNC: return "settrunc";
+	case CEPH_OSD_OP_TRIMTRUNC: return "trimtrunc";
+
+	case CEPH_OSD_OP_TMAPUP: return "tmapup";
+	case CEPH_OSD_OP_TMAPGET: return "tmapget";
+	case CEPH_OSD_OP_TMAPPUT: return "tmapput";
+
+	case CEPH_OSD_OP_GETXATTR: return "getxattr";
+	case CEPH_OSD_OP_GETXATTRS: return "getxattrs";
+	case CEPH_OSD_OP_SETXATTR: return "setxattr";
+	case CEPH_OSD_OP_SETXATTRS: return "setxattrs";
+	case CEPH_OSD_OP_RESETXATTRS: return "resetxattrs";
+	case CEPH_OSD_OP_RMXATTR: return "rmxattr";
+	case CEPH_OSD_OP_CMPXATTR: return "cmpxattr";
+
+	case CEPH_OSD_OP_PULL: return "pull";
+	case CEPH_OSD_OP_PUSH: return "push";
+	case CEPH_OSD_OP_BALANCEREADS: return "balance-reads";
+	case CEPH_OSD_OP_UNBALANCEREADS: return "unbalance-reads";
+	case CEPH_OSD_OP_SCRUB: return "scrub";
+
+	case CEPH_OSD_OP_WRLOCK: return "wrlock";
+	case CEPH_OSD_OP_WRUNLOCK: return "wrunlock";
+	case CEPH_OSD_OP_RDLOCK: return "rdlock";
+	case CEPH_OSD_OP_RDUNLOCK: return "rdunlock";
+	case CEPH_OSD_OP_UPLOCK: return "uplock";
+	case CEPH_OSD_OP_DNLOCK: return "dnlock";
+
+	case CEPH_OSD_OP_CALL: return "call";
+
+	case CEPH_OSD_OP_PGLS: return "pgls";
+	}
+	return "???";
+}
+
+
+const char *ceph_pool_op_name(int op)
+{
+	switch (op) {
+	case POOL_OP_CREATE: return "create";
+	case POOL_OP_DELETE: return "delete";
+	case POOL_OP_AUID_CHANGE: return "auid change";
+	case POOL_OP_CREATE_SNAP: return "create snap";
+	case POOL_OP_DELETE_SNAP: return "delete snap";
+	case POOL_OP_CREATE_UNMANAGED_SNAP: return "create unmanaged snap";
+	case POOL_OP_DELETE_UNMANAGED_SNAP: return "delete unmanaged snap";
+	}
+	return "???";
+}

net/ceph/crush/crush.c

@@ -8,7 +8,7 @@
 # define BUG_ON(x) assert(!(x))
 #endif
 
-#include "crush.h"
+#include <linux/crush/crush.h>
 
 const char *crush_bucket_alg_name(int alg)
 {

net/ceph/crush/hash.c

@@ -1,6 +1,6 @@
 
 #include <linux/types.h>
-#include "hash.h"
+#include <linux/crush/hash.h>
 
 /*
  * Robert Jenkins' function for mixing 32-bit values

net/ceph/crush/mapper.c

@@ -18,8 +18,8 @@
 # define kfree(x) free(x)
 #endif
 
-#include "crush.h"
+#include <linux/crush/crush.h>
-#include "hash.h"
+#include <linux/crush/hash.h>
 
 /*
  * Implement the core CRUSH mapping algorithm.

net/ceph/crypto.c

@@ -1,13 +1,13 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/err.h>
 #include <linux/scatterlist.h>
 #include <linux/slab.h>
 #include <crypto/hash.h>
 
+#include <linux/ceph/decode.h>
 #include "crypto.h"
-#include "decode.h"
 
 int ceph_crypto_key_encode(struct ceph_crypto_key *key, void **p, void *end)
 {

net/ceph/crypto.h

@@ -1,8 +1,8 @@
 #ifndef _FS_CEPH_CRYPTO_H
 #define _FS_CEPH_CRYPTO_H
 
-#include "types.h"
+#include <linux/ceph/types.h>
-#include "buffer.h"
+#include <linux/ceph/buffer.h>
 
 /*
  * cryptographic secret

net/ceph/debugfs.c

@@ -0,0 +1,267 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/ctype.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+
+#include <linux/ceph/libceph.h>
+#include <linux/ceph/mon_client.h>
+#include <linux/ceph/auth.h>
+#include <linux/ceph/debugfs.h>
+
+#ifdef CONFIG_DEBUG_FS
+
+/*
+ * Implement /sys/kernel/debug/ceph fun
+ *
+ * /sys/kernel/debug/ceph/client*  - an instance of the ceph client
+ *      .../osdmap      - current osdmap
+ *      .../monmap      - current monmap
+ *      .../osdc        - active osd requests
+ *      .../monc        - mon client state
+ *      .../dentry_lru  - dump contents of dentry lru
+ *      .../caps        - expose cap (reservation) stats
+ *      .../bdi         - symlink to ../../bdi/something
+ */
+
+static struct dentry *ceph_debugfs_dir;
+
+static int monmap_show(struct seq_file *s, void *p)
+{
+	int i;
+	struct ceph_client *client = s->private;
+
+	if (client->monc.monmap == NULL)
+		return 0;
+
+	seq_printf(s, "epoch %d\n", client->monc.monmap->epoch);
+	for (i = 0; i < client->monc.monmap->num_mon; i++) {
+		struct ceph_entity_inst *inst =
+			&client->monc.monmap->mon_inst[i];
+
+		seq_printf(s, "\t%s%lld\t%s\n",
+			   ENTITY_NAME(inst->name),
+			   ceph_pr_addr(&inst->addr.in_addr));
+	}
+	return 0;
+}
+
+static int osdmap_show(struct seq_file *s, void *p)
+{
+	int i;
+	struct ceph_client *client = s->private;
+	struct rb_node *n;
+
+	if (client->osdc.osdmap == NULL)
+		return 0;
+	seq_printf(s, "epoch %d\n", client->osdc.osdmap->epoch);
+	seq_printf(s, "flags%s%s\n",
+		   (client->osdc.osdmap->flags & CEPH_OSDMAP_NEARFULL) ?
+		   " NEARFULL" : "",
+		   (client->osdc.osdmap->flags & CEPH_OSDMAP_FULL) ?
+		   " FULL" : "");
+	for (n = rb_first(&client->osdc.osdmap->pg_pools); n; n = rb_next(n)) {
+		struct ceph_pg_pool_info *pool =
+			rb_entry(n, struct ceph_pg_pool_info, node);
+		seq_printf(s, "pg_pool %d pg_num %d / %d, lpg_num %d / %d\n",
+			   pool->id, pool->v.pg_num, pool->pg_num_mask,
+			   pool->v.lpg_num, pool->lpg_num_mask);
+	}
+	for (i = 0; i < client->osdc.osdmap->max_osd; i++) {
+		struct ceph_entity_addr *addr =
+			&client->osdc.osdmap->osd_addr[i];
+		int state = client->osdc.osdmap->osd_state[i];
+		char sb[64];
+
+		seq_printf(s, "\tosd%d\t%s\t%3d%%\t(%s)\n",
+			   i, ceph_pr_addr(&addr->in_addr),
+			   ((client->osdc.osdmap->osd_weight[i]*100) >> 16),
+			   ceph_osdmap_state_str(sb, sizeof(sb), state));
+	}
+	return 0;
+}
+
+static int monc_show(struct seq_file *s, void *p)
+{
+	struct ceph_client *client = s->private;
+	struct ceph_mon_generic_request *req;
+	struct ceph_mon_client *monc = &client->monc;
+	struct rb_node *rp;
+
+	mutex_lock(&monc->mutex);
+
+	if (monc->have_mdsmap)
+		seq_printf(s, "have mdsmap %u\n", (unsigned)monc->have_mdsmap);
+	if (monc->have_osdmap)
+		seq_printf(s, "have osdmap %u\n", (unsigned)monc->have_osdmap);
+	if (monc->want_next_osdmap)
+		seq_printf(s, "want next osdmap\n");
+
+	for (rp = rb_first(&monc->generic_request_tree); rp; rp = rb_next(rp)) {
+		__u16 op;
+		req = rb_entry(rp, struct ceph_mon_generic_request, node);
+		op = le16_to_cpu(req->request->hdr.type);
+		if (op == CEPH_MSG_STATFS)
+			seq_printf(s, "%lld statfs\n", req->tid);
+		else
+			seq_printf(s, "%lld unknown\n", req->tid);
+	}
+
+	mutex_unlock(&monc->mutex);
+	return 0;
+}
+
+static int osdc_show(struct seq_file *s, void *pp)
+{
+	struct ceph_client *client = s->private;
+	struct ceph_osd_client *osdc = &client->osdc;
+	struct rb_node *p;
+
+	mutex_lock(&osdc->request_mutex);
+	for (p = rb_first(&osdc->requests); p; p = rb_next(p)) {
+		struct ceph_osd_request *req;
+		struct ceph_osd_request_head *head;
+		struct ceph_osd_op *op;
+		int num_ops;
+		int opcode, olen;
+		int i;
+
+		req = rb_entry(p, struct ceph_osd_request, r_node);
+
+		seq_printf(s, "%lld\tosd%d\t%d.%x\t", req->r_tid,
+			   req->r_osd ? req->r_osd->o_osd : -1,
+			   le32_to_cpu(req->r_pgid.pool),
+			   le16_to_cpu(req->r_pgid.ps));
+
+		head = req->r_request->front.iov_base;
+		op = (void *)(head + 1);
+
+		num_ops = le16_to_cpu(head->num_ops);
+		olen = le32_to_cpu(head->object_len);
+		seq_printf(s, "%.*s", olen,
+			   (const char *)(head->ops + num_ops));
+
+		if (req->r_reassert_version.epoch)
+			seq_printf(s, "\t%u'%llu",
+			   (unsigned)le32_to_cpu(req->r_reassert_version.epoch),
+			   le64_to_cpu(req->r_reassert_version.version));
+		else
+			seq_printf(s, "\t");
+
+		for (i = 0; i < num_ops; i++) {
+			opcode = le16_to_cpu(op->op);
+			seq_printf(s, "\t%s", ceph_osd_op_name(opcode));
+			op++;
+		}
+
+		seq_printf(s, "\n");
+	}
+	mutex_unlock(&osdc->request_mutex);
+	return 0;
+}
+
+CEPH_DEFINE_SHOW_FUNC(monmap_show)
+CEPH_DEFINE_SHOW_FUNC(osdmap_show)
+CEPH_DEFINE_SHOW_FUNC(monc_show)
+CEPH_DEFINE_SHOW_FUNC(osdc_show)
+
+int ceph_debugfs_init(void)
+{
+	ceph_debugfs_dir = debugfs_create_dir("ceph", NULL);
+	if (!ceph_debugfs_dir)
+		return -ENOMEM;
+	return 0;
+}
+
+void ceph_debugfs_cleanup(void)
+{
+	debugfs_remove(ceph_debugfs_dir);
+}
+
+int ceph_debugfs_client_init(struct ceph_client *client)
+{
+	int ret = -ENOMEM;
+	char name[80];
+
+	snprintf(name, sizeof(name), "%pU.client%lld", &client->fsid,
+		 client->monc.auth->global_id);
+
+	client->debugfs_dir = debugfs_create_dir(name, ceph_debugfs_dir);
+	if (!client->debugfs_dir)
+		goto out;
+
+	client->monc.debugfs_file = debugfs_create_file("monc",
+						      0600,
+						      client->debugfs_dir,
+						      client,
+						      &monc_show_fops);
+	if (!client->monc.debugfs_file)
+		goto out;
+
+	client->osdc.debugfs_file = debugfs_create_file("osdc",
+						      0600,
+						      client->debugfs_dir,
+						      client,
+						      &osdc_show_fops);
+	if (!client->osdc.debugfs_file)
+		goto out;
+
+	client->debugfs_monmap = debugfs_create_file("monmap",
+					0600,
+					client->debugfs_dir,
+					client,
+					&monmap_show_fops);
+	if (!client->debugfs_monmap)
+		goto out;
+
+	client->debugfs_osdmap = debugfs_create_file("osdmap",
+					0600,
+					client->debugfs_dir,
+					client,
+					&osdmap_show_fops);
+	if (!client->debugfs_osdmap)
+		goto out;
+
+	return 0;
+
+out:
+	ceph_debugfs_client_cleanup(client);
+	return ret;
+}
+
+void ceph_debugfs_client_cleanup(struct ceph_client *client)
+{
+	debugfs_remove(client->debugfs_osdmap);
+	debugfs_remove(client->debugfs_monmap);
+	debugfs_remove(client->osdc.debugfs_file);
+	debugfs_remove(client->monc.debugfs_file);
+	debugfs_remove(client->debugfs_dir);
+}
+
+#else  /* CONFIG_DEBUG_FS */
+
+int ceph_debugfs_init(void)
+{
+	return 0;
+}
+
+void ceph_debugfs_cleanup(void)
+{
+}
+
+int ceph_debugfs_client_init(struct ceph_client *client)
+{
+	return 0;
+}
+
+void ceph_debugfs_client_cleanup(struct ceph_client *client)
+{
+}
+
+#endif  /* CONFIG_DEBUG_FS */
+
+EXPORT_SYMBOL(ceph_debugfs_init);
+EXPORT_SYMBOL(ceph_debugfs_cleanup);

net/ceph/messenger.c

@@ -1,4 +1,4 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/crc32c.h>
 #include <linux/ctype.h>
@@ -9,12 +9,14 @@
 #include <linux/slab.h>
 #include <linux/socket.h>
 #include <linux/string.h>
+#include <linux/bio.h>
+#include <linux/blkdev.h>
 #include <net/tcp.h>
 
-#include "super.h"
+#include <linux/ceph/libceph.h>
-#include "messenger.h"
+#include <linux/ceph/messenger.h>
-#include "decode.h"
+#include <linux/ceph/decode.h>
-#include "pagelist.h"
+#include <linux/ceph/pagelist.h>
 
 /*
  * Ceph uses the messenger to exchange ceph_msg messages with other
@@ -48,7 +50,7 @@ static char addr_str[MAX_ADDR_STR][MAX_ADDR_STR_LEN];
 static DEFINE_SPINLOCK(addr_str_lock);
 static int last_addr_str;
 
-const char *pr_addr(const struct sockaddr_storage *ss)
+const char *ceph_pr_addr(const struct sockaddr_storage *ss)
 {
 	int i;
 	char *s;
@@ -79,6 +81,7 @@ const char *pr_addr(const struct sockaddr_storage *ss)
 
 	return s;
 }
+EXPORT_SYMBOL(ceph_pr_addr);
 
 static void encode_my_addr(struct ceph_messenger *msgr)
 {
@@ -91,7 +94,7 @@ static void encode_my_addr(struct ceph_messenger *msgr)
  */
 struct workqueue_struct *ceph_msgr_wq;
 
-int __init ceph_msgr_init(void)
+int ceph_msgr_init(void)
 {
 	ceph_msgr_wq = create_workqueue("ceph-msgr");
 	if (IS_ERR(ceph_msgr_wq)) {
@@ -102,16 +105,19 @@ int __init ceph_msgr_init(void)
 	}
 	return 0;
 }
+EXPORT_SYMBOL(ceph_msgr_init);
 
 void ceph_msgr_exit(void)
 {
 	destroy_workqueue(ceph_msgr_wq);
 }
+EXPORT_SYMBOL(ceph_msgr_exit);
 
 void ceph_msgr_flush(void)
 {
 	flush_workqueue(ceph_msgr_wq);
 }
+EXPORT_SYMBOL(ceph_msgr_flush);
 
 
 /*
@@ -221,19 +227,19 @@ static struct socket *ceph_tcp_connect(struct ceph_connection *con)
 
 	set_sock_callbacks(sock, con);
 
-	dout("connect %s\n", pr_addr(&con->peer_addr.in_addr));
+	dout("connect %s\n", ceph_pr_addr(&con->peer_addr.in_addr));
 
 	ret = sock->ops->connect(sock, (struct sockaddr *)paddr, sizeof(*paddr),
 				 O_NONBLOCK);
 	if (ret == -EINPROGRESS) {
 		dout("connect %s EINPROGRESS sk_state = %u\n",
-		     pr_addr(&con->peer_addr.in_addr),
+		     ceph_pr_addr(&con->peer_addr.in_addr),
 		     sock->sk->sk_state);
 		ret = 0;
 	}
 	if (ret < 0) {
 		pr_err("connect %s error %d\n",
-		       pr_addr(&con->peer_addr.in_addr), ret);
+		       ceph_pr_addr(&con->peer_addr.in_addr), ret);
 		sock_release(sock);
 		con->sock = NULL;
 		con->error_msg = "connect error";
@@ -334,7 +340,8 @@ static void reset_connection(struct ceph_connection *con)
  */
 void ceph_con_close(struct ceph_connection *con)
 {
-	dout("con_close %p peer %s\n", con, pr_addr(&con->peer_addr.in_addr));
+	dout("con_close %p peer %s\n", con,
+	     ceph_pr_addr(&con->peer_addr.in_addr));
 	set_bit(CLOSED, &con->state);  /* in case there's queued work */
 	clear_bit(STANDBY, &con->state);  /* avoid connect_seq bump */
 	clear_bit(LOSSYTX, &con->state);  /* so we retry next connect */
@@ -347,19 +354,21 @@ void ceph_con_close(struct ceph_connection *con)
 	mutex_unlock(&con->mutex);
 	queue_con(con);
 }
+EXPORT_SYMBOL(ceph_con_close);
 
 /*
  * Reopen a closed connection, with a new peer address.
  */
 void ceph_con_open(struct ceph_connection *con, struct ceph_entity_addr *addr)
 {
-	dout("con_open %p %s\n", con, pr_addr(&addr->in_addr));
+	dout("con_open %p %s\n", con, ceph_pr_addr(&addr->in_addr));
 	set_bit(OPENING, &con->state);
 	clear_bit(CLOSED, &con->state);
 	memcpy(&con->peer_addr, addr, sizeof(*addr));
 	con->delay = 0;      /* reset backoff memory */
 	queue_con(con);
 }
+EXPORT_SYMBOL(ceph_con_open);
 
 /*
  * return true if this connection ever successfully opened
@@ -406,6 +415,7 @@ void ceph_con_init(struct ceph_messenger *msgr, struct ceph_connection *con)
 	INIT_LIST_HEAD(&con->out_sent);
 	INIT_DELAYED_WORK(&con->work, con_work);
 }
+EXPORT_SYMBOL(ceph_con_init);
 
 
 /*
@@ -529,8 +539,11 @@ static void prepare_write_message(struct ceph_connection *con)
 	if (le32_to_cpu(m->hdr.data_len) > 0) {
 		/* initialize page iterator */
 		con->out_msg_pos.page = 0;
-		con->out_msg_pos.page_pos =
+		if (m->pages)
-			le16_to_cpu(m->hdr.data_off) & ~PAGE_MASK;
+			con->out_msg_pos.page_pos =
+				le16_to_cpu(m->hdr.data_off) & ~PAGE_MASK;
+		else
+			con->out_msg_pos.page_pos = 0;
 		con->out_msg_pos.data_pos = 0;
 		con->out_msg_pos.did_page_crc = 0;
 		con->out_more = 1;  /* data + footer will follow */
@@ -647,7 +660,7 @@ static void prepare_write_connect(struct ceph_messenger *msgr,
 	dout("prepare_write_connect %p cseq=%d gseq=%d proto=%d\n", con,
 	     con->connect_seq, global_seq, proto);
 
-	con->out_connect.features = cpu_to_le64(CEPH_FEATURE_SUPPORTED);
+	con->out_connect.features = cpu_to_le64(msgr->supported_features);
 	con->out_connect.host_type = cpu_to_le32(CEPH_ENTITY_TYPE_CLIENT);
 	con->out_connect.connect_seq = cpu_to_le32(con->connect_seq);
 	con->out_connect.global_seq = cpu_to_le32(global_seq);
@@ -712,6 +725,31 @@ out:
 	return ret;  /* done! */
 }
 
+#ifdef CONFIG_BLOCK
+static void init_bio_iter(struct bio *bio, struct bio **iter, int *seg)
+{
+	if (!bio) {
+		*iter = NULL;
+		*seg = 0;
+		return;
+	}
+	*iter = bio;
+	*seg = bio->bi_idx;
+}
+
+static void iter_bio_next(struct bio **bio_iter, int *seg)
+{
+	if (*bio_iter == NULL)
+		return;
+
+	BUG_ON(*seg >= (*bio_iter)->bi_vcnt);
+
+	(*seg)++;
+	if (*seg == (*bio_iter)->bi_vcnt)
+		init_bio_iter((*bio_iter)->bi_next, bio_iter, seg);
+}
+#endif
+
 /*
  * Write as much message data payload as we can.  If we finish, queue
  * up the footer.
@@ -726,21 +764,46 @@ static int write_partial_msg_pages(struct ceph_connection *con)
 	size_t len;
 	int crc = con->msgr->nocrc;
 	int ret;
+	int total_max_write;
+	int in_trail = 0;
+	size_t trail_len = (msg->trail ? msg->trail->length : 0);
 
 	dout("write_partial_msg_pages %p msg %p page %d/%d offset %d\n",
 	     con, con->out_msg, con->out_msg_pos.page, con->out_msg->nr_pages,
 	     con->out_msg_pos.page_pos);
 
-	while (con->out_msg_pos.page < con->out_msg->nr_pages) {
+#ifdef CONFIG_BLOCK
+	if (msg->bio && !msg->bio_iter)
+		init_bio_iter(msg->bio, &msg->bio_iter, &msg->bio_seg);
+#endif
+
+	while (data_len > con->out_msg_pos.data_pos) {
 		struct page *page = NULL;
 		void *kaddr = NULL;
+		int max_write = PAGE_SIZE;
+		int page_shift = 0;
+
+		total_max_write = data_len - trail_len -
+			con->out_msg_pos.data_pos;
 
 		/*
 		 * if we are calculating the data crc (the default), we need
 		 * to map the page.  if our pages[] has been revoked, use the
 		 * zero page.
 		 */
-		if (msg->pages) {
+
+		/* have we reached the trail part of the data? */
+		if (con->out_msg_pos.data_pos >= data_len - trail_len) {
+			in_trail = 1;
+
+			total_max_write = data_len - con->out_msg_pos.data_pos;
+
+			page = list_first_entry(&msg->trail->head,
+						struct page, lru);
+			if (crc)
+				kaddr = kmap(page);
+			max_write = PAGE_SIZE;
+		} else if (msg->pages) {
 			page = msg->pages[con->out_msg_pos.page];
 			if (crc)
 				kaddr = kmap(page);
@@ -749,13 +812,25 @@ static int write_partial_msg_pages(struct ceph_connection *con)
 						struct page, lru);
 			if (crc)
 				kaddr = kmap(page);
+#ifdef CONFIG_BLOCK
+		} else if (msg->bio) {
+			struct bio_vec *bv;
+
+			bv = bio_iovec_idx(msg->bio_iter, msg->bio_seg);
+			page = bv->bv_page;
+			page_shift = bv->bv_offset;
+			if (crc)
+				kaddr = kmap(page) + page_shift;
+			max_write = bv->bv_len;
+#endif
 		} else {
 			page = con->msgr->zero_page;
 			if (crc)
 				kaddr = page_address(con->msgr->zero_page);
 		}
-		len = min((int)(PAGE_SIZE - con->out_msg_pos.page_pos),
+		len = min_t(int, max_write - con->out_msg_pos.page_pos,
-			  (int)(data_len - con->out_msg_pos.data_pos));
+			    total_max_write);
+
 		if (crc && !con->out_msg_pos.did_page_crc) {
 			void *base = kaddr + con->out_msg_pos.page_pos;
 			u32 tmpcrc = le32_to_cpu(con->out_msg->footer.data_crc);
@@ -765,13 +840,14 @@ static int write_partial_msg_pages(struct ceph_connection *con)
 				cpu_to_le32(crc32c(tmpcrc, base, len));
 			con->out_msg_pos.did_page_crc = 1;
 		}
-
 		ret = kernel_sendpage(con->sock, page,
-				      con->out_msg_pos.page_pos, len,
+				      con->out_msg_pos.page_pos + page_shift,
+				      len,
 				      MSG_DONTWAIT | MSG_NOSIGNAL |
 				      MSG_MORE);
 
-		if (crc && (msg->pages || msg->pagelist))
+		if (crc &&
+		    (msg->pages || msg->pagelist || msg->bio || in_trail))
 			kunmap(page);
 
 		if (ret <= 0)
@@ -783,9 +859,16 @@ static int write_partial_msg_pages(struct ceph_connection *con)
 			con->out_msg_pos.page_pos = 0;
 			con->out_msg_pos.page++;
 			con->out_msg_pos.did_page_crc = 0;
-			if (msg->pagelist)
+			if (in_trail)
+				list_move_tail(&page->lru,
+					       &msg->trail->head);
+			else if (msg->pagelist)
 				list_move_tail(&page->lru,
 					       &msg->pagelist->head);
+#ifdef CONFIG_BLOCK
+			else if (msg->bio)
+				iter_bio_next(&msg->bio_iter, &msg->bio_seg);
+#endif
 		}
 	}
 
@@ -938,7 +1021,7 @@ static int verify_hello(struct ceph_connection *con)
 {
 	if (memcmp(con->in_banner, CEPH_BANNER, strlen(CEPH_BANNER))) {
 		pr_err("connect to %s got bad banner\n",
-		       pr_addr(&con->peer_addr.in_addr));
+		       ceph_pr_addr(&con->peer_addr.in_addr));
 		con->error_msg = "protocol error, bad banner";
 		return -1;
 	}
@@ -1041,7 +1124,7 @@ int ceph_parse_ips(const char *c, const char *end,
 
 		addr_set_port(ss, port);
 
-		dout("parse_ips got %s\n", pr_addr(ss));
+		dout("parse_ips got %s\n", ceph_pr_addr(ss));
 
 		if (p == end)
 			break;
@@ -1061,6 +1144,7 @@ bad:
 	pr_err("parse_ips bad ip '%.*s'\n", (int)(end - c), c);
 	return -EINVAL;
 }
+EXPORT_SYMBOL(ceph_parse_ips);
 
 static int process_banner(struct ceph_connection *con)
 {
@@ -1082,9 +1166,9 @@ static int process_banner(struct ceph_connection *con)
 	    !(addr_is_blank(&con->actual_peer_addr.in_addr) &&
 	      con->actual_peer_addr.nonce == con->peer_addr.nonce)) {
 		pr_warning("wrong peer, want %s/%d, got %s/%d\n",
-			   pr_addr(&con->peer_addr.in_addr),
+			   ceph_pr_addr(&con->peer_addr.in_addr),
 			   (int)le32_to_cpu(con->peer_addr.nonce),
-			   pr_addr(&con->actual_peer_addr.in_addr),
+			   ceph_pr_addr(&con->actual_peer_addr.in_addr),
 			   (int)le32_to_cpu(con->actual_peer_addr.nonce));
 		con->error_msg = "wrong peer at address";
 		return -1;
@@ -1102,7 +1186,7 @@ static int process_banner(struct ceph_connection *con)
 		addr_set_port(&con->msgr->inst.addr.in_addr, port);
 		encode_my_addr(con->msgr);
 		dout("process_banner learned my addr is %s\n",
-		     pr_addr(&con->msgr->inst.addr.in_addr));
+		     ceph_pr_addr(&con->msgr->inst.addr.in_addr));
 	}
 
 	set_bit(NEGOTIATING, &con->state);
@@ -1123,8 +1207,8 @@ static void fail_protocol(struct ceph_connection *con)
 
 static int process_connect(struct ceph_connection *con)
 {
-	u64 sup_feat = CEPH_FEATURE_SUPPORTED;
+	u64 sup_feat = con->msgr->supported_features;
-	u64 req_feat = CEPH_FEATURE_REQUIRED;
+	u64 req_feat = con->msgr->required_features;
 	u64 server_feat = le64_to_cpu(con->in_reply.features);
 
 	dout("process_connect on %p tag %d\n", con, (int)con->in_tag);
@@ -1134,7 +1218,7 @@ static int process_connect(struct ceph_connection *con)
 		pr_err("%s%lld %s feature set mismatch,"
 		       " my %llx < server's %llx, missing %llx\n",
 		       ENTITY_NAME(con->peer_name),
-		       pr_addr(&con->peer_addr.in_addr),
+		       ceph_pr_addr(&con->peer_addr.in_addr),
 		       sup_feat, server_feat, server_feat & ~sup_feat);
 		con->error_msg = "missing required protocol features";
 		fail_protocol(con);
@@ -1144,7 +1228,7 @@ static int process_connect(struct ceph_connection *con)
 		pr_err("%s%lld %s protocol version mismatch,"
 		       " my %d != server's %d\n",
 		       ENTITY_NAME(con->peer_name),
-		       pr_addr(&con->peer_addr.in_addr),
+		       ceph_pr_addr(&con->peer_addr.in_addr),
 		       le32_to_cpu(con->out_connect.protocol_version),
 		       le32_to_cpu(con->in_reply.protocol_version));
 		con->error_msg = "protocol version mismatch";
@@ -1178,7 +1262,7 @@ static int process_connect(struct ceph_connection *con)
 		     le32_to_cpu(con->in_connect.connect_seq));
 		pr_err("%s%lld %s connection reset\n",
 		       ENTITY_NAME(con->peer_name),
-		       pr_addr(&con->peer_addr.in_addr));
+		       ceph_pr_addr(&con->peer_addr.in_addr));
 		reset_connection(con);
 		prepare_write_connect(con->msgr, con, 0);
 		prepare_read_connect(con);
@@ -1223,7 +1307,7 @@ static int process_connect(struct ceph_connection *con)
 			pr_err("%s%lld %s protocol feature mismatch,"
 			       " my required %llx > server's %llx, need %llx\n",
 			       ENTITY_NAME(con->peer_name),
-			       pr_addr(&con->peer_addr.in_addr),
+			       ceph_pr_addr(&con->peer_addr.in_addr),
 			       req_feat, server_feat, req_feat & ~server_feat);
 			con->error_msg = "missing required protocol features";
 			fail_protocol(con);
@@ -1305,8 +1389,7 @@ static int read_partial_message_section(struct ceph_connection *con,
 					struct kvec *section,
 					unsigned int sec_len, u32 *crc)
 {
-	int left;
+	int ret, left;
-	int ret;
 
 	BUG_ON(!section);
 
@@ -1329,13 +1412,83 @@ static int read_partial_message_section(struct ceph_connection *con,
 static struct ceph_msg *ceph_alloc_msg(struct ceph_connection *con,
 				struct ceph_msg_header *hdr,
 				int *skip);
+
+
+static int read_partial_message_pages(struct ceph_connection *con,
+				      struct page **pages,
+				      unsigned data_len, int datacrc)
+{
+	void *p;
+	int ret;
+	int left;
+
+	left = min((int)(data_len - con->in_msg_pos.data_pos),
+		   (int)(PAGE_SIZE - con->in_msg_pos.page_pos));
+	/* (page) data */
+	BUG_ON(pages == NULL);
+	p = kmap(pages[con->in_msg_pos.page]);
+	ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
+			       left);
+	if (ret > 0 && datacrc)
+		con->in_data_crc =
+			crc32c(con->in_data_crc,
+				  p + con->in_msg_pos.page_pos, ret);
+	kunmap(pages[con->in_msg_pos.page]);
+	if (ret <= 0)
+		return ret;
+	con->in_msg_pos.data_pos += ret;
+	con->in_msg_pos.page_pos += ret;
+	if (con->in_msg_pos.page_pos == PAGE_SIZE) {
+		con->in_msg_pos.page_pos = 0;
+		con->in_msg_pos.page++;
+	}
+
+	return ret;
+}
+
+#ifdef CONFIG_BLOCK
+static int read_partial_message_bio(struct ceph_connection *con,
+				    struct bio **bio_iter, int *bio_seg,
+				    unsigned data_len, int datacrc)
+{
+	struct bio_vec *bv = bio_iovec_idx(*bio_iter, *bio_seg);
+	void *p;
+	int ret, left;
+
+	if (IS_ERR(bv))
+		return PTR_ERR(bv);
+
+	left = min((int)(data_len - con->in_msg_pos.data_pos),
+		   (int)(bv->bv_len - con->in_msg_pos.page_pos));
+
+	p = kmap(bv->bv_page) + bv->bv_offset;
+
+	ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
+			       left);
+	if (ret > 0 && datacrc)
+		con->in_data_crc =
+			crc32c(con->in_data_crc,
+				  p + con->in_msg_pos.page_pos, ret);
+	kunmap(bv->bv_page);
+	if (ret <= 0)
+		return ret;
+	con->in_msg_pos.data_pos += ret;
+	con->in_msg_pos.page_pos += ret;
+	if (con->in_msg_pos.page_pos == bv->bv_len) {
+		con->in_msg_pos.page_pos = 0;
+		iter_bio_next(bio_iter, bio_seg);
+	}
+
+	return ret;
+}
+#endif
+
 /*
  * read (part of) a message.
  */
 static int read_partial_message(struct ceph_connection *con)
 {
 	struct ceph_msg *m = con->in_msg;
-	void *p;
 	int ret;
 	int to, left;
 	unsigned front_len, middle_len, data_len, data_off;
@@ -1381,7 +1534,7 @@ static int read_partial_message(struct ceph_connection *con)
 	if ((s64)seq - (s64)con->in_seq < 1) {
 		pr_info("skipping %s%lld %s seq %lld, expected %lld\n",
 			ENTITY_NAME(con->peer_name),
-			pr_addr(&con->peer_addr.in_addr),
+			ceph_pr_addr(&con->peer_addr.in_addr),
 			seq, con->in_seq + 1);
 		con->in_base_pos = -front_len - middle_len - data_len -
 			sizeof(m->footer);
@@ -1422,7 +1575,10 @@ static int read_partial_message(struct ceph_connection *con)
 			m->middle->vec.iov_len = 0;
 
 		con->in_msg_pos.page = 0;
-		con->in_msg_pos.page_pos = data_off & ~PAGE_MASK;
+		if (m->pages)
+			con->in_msg_pos.page_pos = data_off & ~PAGE_MASK;
+		else
+			con->in_msg_pos.page_pos = 0;
 		con->in_msg_pos.data_pos = 0;
 	}
 
@@ -1440,27 +1596,29 @@ static int read_partial_message(struct ceph_connection *con)
 		if (ret <= 0)
 			return ret;
 	}
+#ifdef CONFIG_BLOCK
+	if (m->bio && !m->bio_iter)
+		init_bio_iter(m->bio, &m->bio_iter, &m->bio_seg);
+#endif
 
 	/* (page) data */
 	while (con->in_msg_pos.data_pos < data_len) {
-		left = min((int)(data_len - con->in_msg_pos.data_pos),
+		if (m->pages) {
-			   (int)(PAGE_SIZE - con->in_msg_pos.page_pos));
+			ret = read_partial_message_pages(con, m->pages,
-		BUG_ON(m->pages == NULL);
+						 data_len, datacrc);
-		p = kmap(m->pages[con->in_msg_pos.page]);
+			if (ret <= 0)
-		ret = ceph_tcp_recvmsg(con->sock, p + con->in_msg_pos.page_pos,
+				return ret;
-				       left);
+#ifdef CONFIG_BLOCK
-		if (ret > 0 && datacrc)
+		} else if (m->bio) {
-			con->in_data_crc =
+
-				crc32c(con->in_data_crc,
+			ret = read_partial_message_bio(con,
-					  p + con->in_msg_pos.page_pos, ret);
+						 &m->bio_iter, &m->bio_seg,
-		kunmap(m->pages[con->in_msg_pos.page]);
+						 data_len, datacrc);
-		if (ret <= 0)
+			if (ret <= 0)
-			return ret;
+				return ret;
-		con->in_msg_pos.data_pos += ret;
+#endif
-		con->in_msg_pos.page_pos += ret;
+		} else {
-		if (con->in_msg_pos.page_pos == PAGE_SIZE) {
+			BUG_ON(1);
-			con->in_msg_pos.page_pos = 0;
-			con->in_msg_pos.page++;
 		}
 	}
 
@@ -1874,9 +2032,9 @@ out:
 static void ceph_fault(struct ceph_connection *con)
 {
 	pr_err("%s%lld %s %s\n", ENTITY_NAME(con->peer_name),
-	       pr_addr(&con->peer_addr.in_addr), con->error_msg);
+	       ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg);
 	dout("fault %p state %lu to peer %s\n",
-	     con, con->state, pr_addr(&con->peer_addr.in_addr));
+	     con, con->state, ceph_pr_addr(&con->peer_addr.in_addr));
 
 	if (test_bit(LOSSYTX, &con->state)) {
 		dout("fault on LOSSYTX channel\n");
@@ -1936,7 +2094,9 @@ out:
 /*
  * create a new messenger instance
  */
-struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr)
+struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr,
+					     u32 supported_features,
+					     u32 required_features)
 {
 	struct ceph_messenger *msgr;
 
@@ -1944,6 +2104,9 @@ struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr)
 	if (msgr == NULL)
 		return ERR_PTR(-ENOMEM);
 
+	msgr->supported_features = supported_features;
+	msgr->required_features = required_features;
+
 	spin_lock_init(&msgr->global_seq_lock);
 
 	/* the zero page is needed if a request is "canceled" while the message
@@ -1966,6 +2129,7 @@ struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr)
 	dout("messenger_create %p\n", msgr);
 	return msgr;
 }
+EXPORT_SYMBOL(ceph_messenger_create);
 
 void ceph_messenger_destroy(struct ceph_messenger *msgr)
 {
@@ -1975,6 +2139,7 @@ void ceph_messenger_destroy(struct ceph_messenger *msgr)
 	kfree(msgr);
 	dout("destroyed messenger %p\n", msgr);
 }
+EXPORT_SYMBOL(ceph_messenger_destroy);
 
 /*
  * Queue up an outgoing message on the given connection.
@@ -2011,6 +2176,7 @@ void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg)
 	if (test_and_set_bit(WRITE_PENDING, &con->state) == 0)
 		queue_con(con);
 }
+EXPORT_SYMBOL(ceph_con_send);
 
 /*
  * Revoke a message that was previously queued for send
@@ -2076,6 +2242,7 @@ void ceph_con_keepalive(struct ceph_connection *con)
 	    test_and_set_bit(WRITE_PENDING, &con->state) == 0)
 		queue_con(con);
 }
+EXPORT_SYMBOL(ceph_con_keepalive);
 
 
 /*
@@ -2136,6 +2303,10 @@ struct ceph_msg *ceph_msg_new(int type, int front_len, gfp_t flags)
 	m->nr_pages = 0;
 	m->pages = NULL;
 	m->pagelist = NULL;
+	m->bio = NULL;
+	m->bio_iter = NULL;
+	m->bio_seg = 0;
+	m->trail = NULL;
 
 	dout("ceph_msg_new %p front %d\n", m, front_len);
 	return m;
@@ -2146,6 +2317,7 @@ out:
 	pr_err("msg_new can't create type %d front %d\n", type, front_len);
 	return NULL;
 }
+EXPORT_SYMBOL(ceph_msg_new);
 
 /*
  * Allocate "middle" portion of a message, if it is needed and wasn't
@@ -2250,11 +2422,14 @@ void ceph_msg_last_put(struct kref *kref)
 		m->pagelist = NULL;
 	}
 
+	m->trail = NULL;
+
 	if (m->pool)
 		ceph_msgpool_put(m->pool, m);
 	else
 		ceph_msg_kfree(m);
 }
+EXPORT_SYMBOL(ceph_msg_last_put);
 
 void ceph_msg_dump(struct ceph_msg *msg)
 {
@@ -2275,3 +2450,4 @@ void ceph_msg_dump(struct ceph_msg *msg)
 		       DUMP_PREFIX_OFFSET, 16, 1,
 		       &msg->footer, sizeof(msg->footer), true);
 }
+EXPORT_SYMBOL(ceph_msg_dump);

net/ceph/mon_client.c

@@ -1,14 +1,16 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/types.h>
 #include <linux/slab.h>
 #include <linux/random.h>
 #include <linux/sched.h>
 
-#include "mon_client.h"
+#include <linux/ceph/mon_client.h>
-#include "super.h"
+#include <linux/ceph/libceph.h>
-#include "auth.h"
+#include <linux/ceph/decode.h>
-#include "decode.h"
+
+#include <linux/ceph/auth.h>
 
 /*
  * Interact with Ceph monitor cluster.  Handle requests for new map
@@ -74,7 +76,7 @@ struct ceph_monmap *ceph_monmap_decode(void *p, void *end)
 	     m->num_mon);
 	for (i = 0; i < m->num_mon; i++)
 		dout("monmap_decode  mon%d is %s\n", i,
-		     pr_addr(&m->mon_inst[i].addr.in_addr));
+		     ceph_pr_addr(&m->mon_inst[i].addr.in_addr));
 	return m;
 
 bad:
@@ -191,30 +193,33 @@ static void __send_subscribe(struct ceph_mon_client *monc)
 		struct ceph_msg *msg = monc->m_subscribe;
 		struct ceph_mon_subscribe_item *i;
 		void *p, *end;
+		int num;
 
 		p = msg->front.iov_base;
 		end = p + msg->front_max;
 
-		dout("__send_subscribe to 'mdsmap' %u+\n",
+		num = 1 + !!monc->want_next_osdmap + !!monc->want_mdsmap;
-		     (unsigned)monc->have_mdsmap);
+		ceph_encode_32(&p, num);
+
 		if (monc->want_next_osdmap) {
 			dout("__send_subscribe to 'osdmap' %u\n",
 			     (unsigned)monc->have_osdmap);
-			ceph_encode_32(&p, 3);
 			ceph_encode_string(&p, end, "osdmap", 6);
 			i = p;
 			i->have = cpu_to_le64(monc->have_osdmap);
 			i->onetime = 1;
 			p += sizeof(*i);
 			monc->want_next_osdmap = 2;  /* requested */
-		} else {
-			ceph_encode_32(&p, 2);
 		}
-		ceph_encode_string(&p, end, "mdsmap", 6);
+		if (monc->want_mdsmap) {
-		i = p;
+			dout("__send_subscribe to 'mdsmap' %u+\n",
-		i->have = cpu_to_le64(monc->have_mdsmap);
+			     (unsigned)monc->have_mdsmap);
-		i->onetime = 0;
+			ceph_encode_string(&p, end, "mdsmap", 6);
-		p += sizeof(*i);
+			i = p;
+			i->have = cpu_to_le64(monc->have_mdsmap);
+			i->onetime = 0;
+			p += sizeof(*i);
+		}
 		ceph_encode_string(&p, end, "monmap", 6);
 		i = p;
 		i->have = 0;
@@ -243,7 +248,8 @@ static void handle_subscribe_ack(struct ceph_mon_client *monc,
 	mutex_lock(&monc->mutex);
 	if (monc->hunting) {
 		pr_info("mon%d %s session established\n",
-			monc->cur_mon, pr_addr(&monc->con->peer_addr.in_addr));
+			monc->cur_mon,
+			ceph_pr_addr(&monc->con->peer_addr.in_addr));
 		monc->hunting = false;
 	}
 	dout("handle_subscribe_ack after %d seconds\n", seconds);
@@ -266,6 +272,7 @@ int ceph_monc_got_mdsmap(struct ceph_mon_client *monc, u32 got)
 	mutex_unlock(&monc->mutex);
 	return 0;
 }
+EXPORT_SYMBOL(ceph_monc_got_mdsmap);
 
 int ceph_monc_got_osdmap(struct ceph_mon_client *monc, u32 got)
 {
@@ -310,6 +317,7 @@ int ceph_monc_open_session(struct ceph_mon_client *monc)
 	mutex_unlock(&monc->mutex);
 	return 0;
 }
+EXPORT_SYMBOL(ceph_monc_open_session);
 
 /*
  * The monitor responds with mount ack indicate mount success.  The
@@ -540,6 +548,7 @@ out:
 	kref_put(&req->kref, release_generic_request);
 	return err;
 }
+EXPORT_SYMBOL(ceph_monc_do_statfs);
 
 /*
  * pool ops
@@ -651,6 +660,7 @@ int ceph_monc_create_snapid(struct ceph_mon_client *monc,
 				   pool, 0, (char *)snapid, sizeof(*snapid));
 
 }
+EXPORT_SYMBOL(ceph_monc_create_snapid);
 
 int ceph_monc_delete_snapid(struct ceph_mon_client *monc,
 			    u32 pool, u64 snapid)
@@ -708,9 +718,9 @@ static void delayed_work(struct work_struct *work)
  */
 static int build_initial_monmap(struct ceph_mon_client *monc)
 {
-	struct ceph_mount_args *args = monc->client->mount_args;
+	struct ceph_options *opt = monc->client->options;
-	struct ceph_entity_addr *mon_addr = args->mon_addr;
+	struct ceph_entity_addr *mon_addr = opt->mon_addr;
-	int num_mon = args->num_mon;
+	int num_mon = opt->num_mon;
 	int i;
 
 	/* build initial monmap */
@@ -728,11 +738,6 @@ static int build_initial_monmap(struct ceph_mon_client *monc)
 	}
 	monc->monmap->num_mon = num_mon;
 	monc->have_fsid = false;
-
-	/* release addr memory */
-	kfree(args->mon_addr);
-	args->mon_addr = NULL;
-	args->num_mon = 0;
 	return 0;
 }
 
@@ -753,8 +758,8 @@ int ceph_monc_init(struct ceph_mon_client *monc, struct ceph_client *cl)
 	monc->con = NULL;
 
 	/* authentication */
-	monc->auth = ceph_auth_init(cl->mount_args->name,
+	monc->auth = ceph_auth_init(cl->options->name,
-				    cl->mount_args->secret);
+				    cl->options->secret);
 	if (IS_ERR(monc->auth))
 		return PTR_ERR(monc->auth);
 	monc->auth->want_keys =
@@ -808,6 +813,7 @@ out_monmap:
 out:
 	return err;
 }
+EXPORT_SYMBOL(ceph_monc_init);
 
 void ceph_monc_stop(struct ceph_mon_client *monc)
 {
@@ -832,6 +838,7 @@ void ceph_monc_stop(struct ceph_mon_client *monc)
 
 	kfree(monc->monmap);
 }
+EXPORT_SYMBOL(ceph_monc_stop);
 
 static void handle_auth_reply(struct ceph_mon_client *monc,
 			      struct ceph_msg *msg)
@@ -889,6 +896,7 @@ int ceph_monc_validate_auth(struct ceph_mon_client *monc)
 	mutex_unlock(&monc->mutex);
 	return ret;
 }
+EXPORT_SYMBOL(ceph_monc_validate_auth);
 
 /*
  * handle incoming message
@@ -922,15 +930,16 @@ static void dispatch(struct ceph_connection *con, struct ceph_msg *msg)
 		ceph_monc_handle_map(monc, msg);
 		break;
 
-	case CEPH_MSG_MDS_MAP:
-		ceph_mdsc_handle_map(&monc->client->mdsc, msg);
-		break;
-
 	case CEPH_MSG_OSD_MAP:
 		ceph_osdc_handle_map(&monc->client->osdc, msg);
 		break;
 
 	default:
+		/* can the chained handler handle it? */
+		if (monc->client->extra_mon_dispatch &&
+		    monc->client->extra_mon_dispatch(monc->client, msg) == 0)
+			break;
+			
 		pr_err("received unknown message type %d %s\n", type,
 		       ceph_msg_type_name(type));
 	}
@@ -994,7 +1003,7 @@ static void mon_fault(struct ceph_connection *con)
 	if (monc->con && !monc->hunting)
 		pr_info("mon%d %s session lost, "
 			"hunting for new mon\n", monc->cur_mon,
-			pr_addr(&monc->con->peer_addr.in_addr));
+			ceph_pr_addr(&monc->con->peer_addr.in_addr));
 
 	__close_session(monc);
 	if (!monc->hunting) {

net/ceph/msgpool.c

@@ -1,11 +1,11 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
 #include <linux/err.h>
 #include <linux/sched.h>
 #include <linux/types.h>
 #include <linux/vmalloc.h>
 
-#include "msgpool.h"
+#include <linux/ceph/msgpool.h>
 
 static void *alloc_fn(gfp_t gfp_mask, void *arg)
 {

net/ceph/osd_client.c

@@ -1,17 +1,22 @@
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/err.h>
 #include <linux/highmem.h>
 #include <linux/mm.h>
 #include <linux/pagemap.h>
 #include <linux/slab.h>
 #include <linux/uaccess.h>
+#ifdef CONFIG_BLOCK
+#include <linux/bio.h>
+#endif
 
-#include "super.h"
+#include <linux/ceph/libceph.h>
-#include "osd_client.h"
+#include <linux/ceph/osd_client.h>
-#include "messenger.h"
+#include <linux/ceph/messenger.h>
-#include "decode.h"
+#include <linux/ceph/decode.h>
-#include "auth.h"
+#include <linux/ceph/auth.h>
+#include <linux/ceph/pagelist.h>
 
 #define OSD_OP_FRONT_LEN	4096
 #define OSD_OPREPLY_FRONT_LEN	512
@@ -22,6 +27,59 @@ static int __kick_requests(struct ceph_osd_client *osdc,
 
 static void kick_requests(struct ceph_osd_client *osdc, struct ceph_osd *osd);
 
+static int op_needs_trail(int op)
+{
+	switch (op) {
+	case CEPH_OSD_OP_GETXATTR:
+	case CEPH_OSD_OP_SETXATTR:
+	case CEPH_OSD_OP_CMPXATTR:
+	case CEPH_OSD_OP_CALL:
+		return 1;
+	default:
+		return 0;
+	}
+}
+
+static int op_has_extent(int op)
+{
+	return (op == CEPH_OSD_OP_READ ||
+		op == CEPH_OSD_OP_WRITE);
+}
+
+void ceph_calc_raw_layout(struct ceph_osd_client *osdc,
+			struct ceph_file_layout *layout,
+			u64 snapid,
+			u64 off, u64 *plen, u64 *bno,
+			struct ceph_osd_request *req,
+			struct ceph_osd_req_op *op)
+{
+	struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base;
+	u64 orig_len = *plen;
+	u64 objoff, objlen;    /* extent in object */
+
+	reqhead->snapid = cpu_to_le64(snapid);
+
+	/* object extent? */
+	ceph_calc_file_object_mapping(layout, off, plen, bno,
+				      &objoff, &objlen);
+	if (*plen < orig_len)
+		dout(" skipping last %llu, final file extent %llu~%llu\n",
+		     orig_len - *plen, off, *plen);
+
+	if (op_has_extent(op->op)) {
+		op->extent.offset = objoff;
+		op->extent.length = objlen;
+	}
+	req->r_num_pages = calc_pages_for(off, *plen);
+	if (op->op == CEPH_OSD_OP_WRITE)
+		op->payload_len = *plen;
+
+	dout("calc_layout bno=%llx %llu~%llu (%d pages)\n",
+	     *bno, objoff, objlen, req->r_num_pages);
+
+}
+EXPORT_SYMBOL(ceph_calc_raw_layout);
+
 /*
  * Implement client access to distributed object storage cluster.
  *
@@ -48,34 +106,19 @@ static void kick_requests(struct ceph_osd_client *osdc, struct ceph_osd *osd);
  * fill osd op in request message.
  */
 static void calc_layout(struct ceph_osd_client *osdc,
-			struct ceph_vino vino, struct ceph_file_layout *layout,
+			struct ceph_vino vino,
+			struct ceph_file_layout *layout,
 			u64 off, u64 *plen,
-			struct ceph_osd_request *req)
+			struct ceph_osd_request *req,
+			struct ceph_osd_req_op *op)
 {
-	struct ceph_osd_request_head *reqhead = req->r_request->front.iov_base;
-	struct ceph_osd_op *op = (void *)(reqhead + 1);
-	u64 orig_len = *plen;
-	u64 objoff, objlen;    /* extent in object */
 	u64 bno;
 
-	reqhead->snapid = cpu_to_le64(vino.snap);
+	ceph_calc_raw_layout(osdc, layout, vino.snap, off,
-
+			     plen, &bno, req, op);
-	/* object extent? */
-	ceph_calc_file_object_mapping(layout, off, plen, &bno,
-				      &objoff, &objlen);
-	if (*plen < orig_len)
-		dout(" skipping last %llu, final file extent %llu~%llu\n",
-		     orig_len - *plen, off, *plen);
 
 	sprintf(req->r_oid, "%llx.%08llx", vino.ino, bno);
 	req->r_oid_len = strlen(req->r_oid);
-
-	op->extent.offset = cpu_to_le64(objoff);
-	op->extent.length = cpu_to_le64(objlen);
-	req->r_num_pages = calc_pages_for(off, *plen);
-
-	dout("calc_layout %s (%d) %llu~%llu (%d pages)\n",
-	     req->r_oid, req->r_oid_len, objoff, objlen, req->r_num_pages);
 }
 
 /*
@@ -101,12 +144,259 @@ void ceph_osdc_release_request(struct kref *kref)
 	if (req->r_own_pages)
 		ceph_release_page_vector(req->r_pages,
 					 req->r_num_pages);
+#ifdef CONFIG_BLOCK
+	if (req->r_bio)
+		bio_put(req->r_bio);
+#endif
 	ceph_put_snap_context(req->r_snapc);
+	if (req->r_trail) {
+		ceph_pagelist_release(req->r_trail);
+		kfree(req->r_trail);
+	}
 	if (req->r_mempool)
 		mempool_free(req, req->r_osdc->req_mempool);
 	else
 		kfree(req);
 }
+EXPORT_SYMBOL(ceph_osdc_release_request);
+
+static int get_num_ops(struct ceph_osd_req_op *ops, int *needs_trail)
+{
+	int i = 0;
+
+	if (needs_trail)
+		*needs_trail = 0;
+	while (ops[i].op) {
+		if (needs_trail && op_needs_trail(ops[i].op))
+			*needs_trail = 1;
+		i++;
+	}
+
+	return i;
+}
+
+struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc,
+					       int flags,
+					       struct ceph_snap_context *snapc,
+					       struct ceph_osd_req_op *ops,
+					       bool use_mempool,
+					       gfp_t gfp_flags,
+					       struct page **pages,
+					       struct bio *bio)
+{
+	struct ceph_osd_request *req;
+	struct ceph_msg *msg;
+	int needs_trail;
+	int num_op = get_num_ops(ops, &needs_trail);
+	size_t msg_size = sizeof(struct ceph_osd_request_head);
+
+	msg_size += num_op*sizeof(struct ceph_osd_op);
+
+	if (use_mempool) {
+		req = mempool_alloc(osdc->req_mempool, gfp_flags);
+		memset(req, 0, sizeof(*req));
+	} else {
+		req = kzalloc(sizeof(*req), gfp_flags);
+	}
+	if (req == NULL)
+		return NULL;
+
+	req->r_osdc = osdc;
+	req->r_mempool = use_mempool;
+
+	kref_init(&req->r_kref);
+	init_completion(&req->r_completion);
+	init_completion(&req->r_safe_completion);
+	INIT_LIST_HEAD(&req->r_unsafe_item);
+	req->r_flags = flags;
+
+	WARN_ON((flags & (CEPH_OSD_FLAG_READ|CEPH_OSD_FLAG_WRITE)) == 0);
+
+	/* create reply message */
+	if (use_mempool)
+		msg = ceph_msgpool_get(&osdc->msgpool_op_reply, 0);
+	else
+		msg = ceph_msg_new(CEPH_MSG_OSD_OPREPLY,
+				   OSD_OPREPLY_FRONT_LEN, gfp_flags);
+	if (!msg) {
+		ceph_osdc_put_request(req);
+		return NULL;
+	}
+	req->r_reply = msg;
+
+	/* allocate space for the trailing data */
+	if (needs_trail) {
+		req->r_trail = kmalloc(sizeof(struct ceph_pagelist), gfp_flags);
+		if (!req->r_trail) {
+			ceph_osdc_put_request(req);
+			return NULL;
+		}
+		ceph_pagelist_init(req->r_trail);
+	}
+	/* create request message; allow space for oid */
+	msg_size += 40;
+	if (snapc)
+		msg_size += sizeof(u64) * snapc->num_snaps;
+	if (use_mempool)
+		msg = ceph_msgpool_get(&osdc->msgpool_op, 0);
+	else
+		msg = ceph_msg_new(CEPH_MSG_OSD_OP, msg_size, gfp_flags);
+	if (!msg) {
+		ceph_osdc_put_request(req);
+		return NULL;
+	}
+
+	msg->hdr.type = cpu_to_le16(CEPH_MSG_OSD_OP);
+	memset(msg->front.iov_base, 0, msg->front.iov_len);
+
+	req->r_request = msg;
+	req->r_pages = pages;
+#ifdef CONFIG_BLOCK
+	if (bio) {
+		req->r_bio = bio;
+		bio_get(req->r_bio);
+	}
+#endif
+
+	return req;
+}
+EXPORT_SYMBOL(ceph_osdc_alloc_request);
+
+static void osd_req_encode_op(struct ceph_osd_request *req,
+			      struct ceph_osd_op *dst,
+			      struct ceph_osd_req_op *src)
+{
+	dst->op = cpu_to_le16(src->op);
+
+	switch (dst->op) {
+	case CEPH_OSD_OP_READ:
+	case CEPH_OSD_OP_WRITE:
+		dst->extent.offset =
+			cpu_to_le64(src->extent.offset);
+		dst->extent.length =
+			cpu_to_le64(src->extent.length);
+		dst->extent.truncate_size =
+			cpu_to_le64(src->extent.truncate_size);
+		dst->extent.truncate_seq =
+			cpu_to_le32(src->extent.truncate_seq);
+		break;
+
+	case CEPH_OSD_OP_GETXATTR:
+	case CEPH_OSD_OP_SETXATTR:
+	case CEPH_OSD_OP_CMPXATTR:
+		BUG_ON(!req->r_trail);
+
+		dst->xattr.name_len = cpu_to_le32(src->xattr.name_len);
+		dst->xattr.value_len = cpu_to_le32(src->xattr.value_len);
+		dst->xattr.cmp_op = src->xattr.cmp_op;
+		dst->xattr.cmp_mode = src->xattr.cmp_mode;
+		ceph_pagelist_append(req->r_trail, src->xattr.name,
+				     src->xattr.name_len);
+		ceph_pagelist_append(req->r_trail, src->xattr.val,
+				     src->xattr.value_len);
+		break;
+	case CEPH_OSD_OP_CALL:
+		BUG_ON(!req->r_trail);
+
+		dst->cls.class_len = src->cls.class_len;
+		dst->cls.method_len = src->cls.method_len;
+		dst->cls.indata_len = cpu_to_le32(src->cls.indata_len);
+
+		ceph_pagelist_append(req->r_trail, src->cls.class_name,
+				     src->cls.class_len);
+		ceph_pagelist_append(req->r_trail, src->cls.method_name,
+				     src->cls.method_len);
+		ceph_pagelist_append(req->r_trail, src->cls.indata,
+				     src->cls.indata_len);
+		break;
+	case CEPH_OSD_OP_ROLLBACK:
+		dst->snap.snapid = cpu_to_le64(src->snap.snapid);
+		break;
+	case CEPH_OSD_OP_STARTSYNC:
+		break;
+	default:
+		pr_err("unrecognized osd opcode %d\n", dst->op);
+		WARN_ON(1);
+		break;
+	}
+	dst->payload_len = cpu_to_le32(src->payload_len);
+}
+
+/*
+ * build new request AND message
+ *
+ */
+void ceph_osdc_build_request(struct ceph_osd_request *req,
+			     u64 off, u64 *plen,
+			     struct ceph_osd_req_op *src_ops,
+			     struct ceph_snap_context *snapc,
+			     struct timespec *mtime,
+			     const char *oid,
+			     int oid_len)
+{
+	struct ceph_msg *msg = req->r_request;
+	struct ceph_osd_request_head *head;
+	struct ceph_osd_req_op *src_op;
+	struct ceph_osd_op *op;
+	void *p;
+	int num_op = get_num_ops(src_ops, NULL);
+	size_t msg_size = sizeof(*head) + num_op*sizeof(*op);
+	int flags = req->r_flags;
+	u64 data_len = 0;
+	int i;
+
+	head = msg->front.iov_base;
+	op = (void *)(head + 1);
+	p = (void *)(op + num_op);
+
+	req->r_snapc = ceph_get_snap_context(snapc);
+
+	head->client_inc = cpu_to_le32(1); /* always, for now. */
+	head->flags = cpu_to_le32(flags);
+	if (flags & CEPH_OSD_FLAG_WRITE)
+		ceph_encode_timespec(&head->mtime, mtime);
+	head->num_ops = cpu_to_le16(num_op);
+
+
+	/* fill in oid */
+	head->object_len = cpu_to_le32(oid_len);
+	memcpy(p, oid, oid_len);
+	p += oid_len;
+
+	src_op = src_ops;
+	while (src_op->op) {
+		osd_req_encode_op(req, op, src_op);
+		src_op++;
+		op++;
+	}
+
+	if (req->r_trail)
+		data_len += req->r_trail->length;
+
+	if (snapc) {
+		head->snap_seq = cpu_to_le64(snapc->seq);
+		head->num_snaps = cpu_to_le32(snapc->num_snaps);
+		for (i = 0; i < snapc->num_snaps; i++) {
+			put_unaligned_le64(snapc->snaps[i], p);
+			p += sizeof(u64);
+		}
+	}
+
+	if (flags & CEPH_OSD_FLAG_WRITE) {
+		req->r_request->hdr.data_off = cpu_to_le16(off);
+		req->r_request->hdr.data_len = cpu_to_le32(*plen + data_len);
+	} else if (data_len) {
+		req->r_request->hdr.data_off = 0;
+		req->r_request->hdr.data_len = cpu_to_le32(data_len);
+	}
+
+	BUG_ON(p > msg->front.iov_base + msg->front.iov_len);
+	msg_size = p - msg->front.iov_base;
+	msg->front.iov_len = msg_size;
+	msg->hdr.front_len = cpu_to_le32(msg_size);
+	return;
+}
+EXPORT_SYMBOL(ceph_osdc_build_request);
 
 /*
  * build new request AND message, calculate layout, and adjust file
@@ -131,110 +421,40 @@ struct ceph_osd_request *ceph_osdc_new_request(struct ceph_osd_client *osdc,
 					       struct timespec *mtime,
 					       bool use_mempool, int num_reply)
 {
+	struct ceph_osd_req_op ops[3];
 	struct ceph_osd_request *req;
-	struct ceph_msg *msg;
-	struct ceph_osd_request_head *head;
-	struct ceph_osd_op *op;
-	void *p;
-	int num_op = 1 + do_sync;
-	size_t msg_size = sizeof(*head) + num_op*sizeof(*op);
-	int i;
 
-	if (use_mempool) {
+	ops[0].op = opcode;
-		req = mempool_alloc(osdc->req_mempool, GFP_NOFS);
+	ops[0].extent.truncate_seq = truncate_seq;
-		memset(req, 0, sizeof(*req));
+	ops[0].extent.truncate_size = truncate_size;
-	} else {
+	ops[0].payload_len = 0;
-		req = kzalloc(sizeof(*req), GFP_NOFS);
-	}
-	if (req == NULL)
-		return NULL;
-
-	req->r_osdc = osdc;
-	req->r_mempool = use_mempool;
-	kref_init(&req->r_kref);
-	init_completion(&req->r_completion);
-	init_completion(&req->r_safe_completion);
-	INIT_LIST_HEAD(&req->r_unsafe_item);
-	req->r_flags = flags;
-
-	WARN_ON((flags & (CEPH_OSD_FLAG_READ|CEPH_OSD_FLAG_WRITE)) == 0);
-
-	/* create reply message */
-	if (use_mempool)
-		msg = ceph_msgpool_get(&osdc->msgpool_op_reply, 0);
-	else
-		msg = ceph_msg_new(CEPH_MSG_OSD_OPREPLY,
-				   OSD_OPREPLY_FRONT_LEN, GFP_NOFS);
-	if (!msg) {
-		ceph_osdc_put_request(req);
-		return NULL;
-	}
-	req->r_reply = msg;
-
-	/* create request message; allow space for oid */
-	msg_size += 40;
-	if (snapc)
-		msg_size += sizeof(u64) * snapc->num_snaps;
-	if (use_mempool)
-		msg = ceph_msgpool_get(&osdc->msgpool_op, 0);
-	else
-		msg = ceph_msg_new(CEPH_MSG_OSD_OP, msg_size, GFP_NOFS);
-	if (!msg) {
-		ceph_osdc_put_request(req);
-		return NULL;
-	}
-	msg->hdr.type = cpu_to_le16(CEPH_MSG_OSD_OP);
-	memset(msg->front.iov_base, 0, msg->front.iov_len);
-	head = msg->front.iov_base;
-	op = (void *)(head + 1);
-	p = (void *)(op + num_op);
-
-	req->r_request = msg;
-	req->r_snapc = ceph_get_snap_context(snapc);
-
-	head->client_inc = cpu_to_le32(1); /* always, for now. */
-	head->flags = cpu_to_le32(flags);
-	if (flags & CEPH_OSD_FLAG_WRITE)
-		ceph_encode_timespec(&head->mtime, mtime);
-	head->num_ops = cpu_to_le16(num_op);
-	op->op = cpu_to_le16(opcode);
-
-	/* calculate max write size */
-	calc_layout(osdc, vino, layout, off, plen, req);
-	req->r_file_layout = *layout;  /* keep a copy */
-
-	if (flags & CEPH_OSD_FLAG_WRITE) {
-		req->r_request->hdr.data_off = cpu_to_le16(off);
-		req->r_request->hdr.data_len = cpu_to_le32(*plen);
-		op->payload_len = cpu_to_le32(*plen);
-	}
-	op->extent.truncate_size = cpu_to_le64(truncate_size);
-	op->extent.truncate_seq = cpu_to_le32(truncate_seq);
-
-	/* fill in oid */
-	head->object_len = cpu_to_le32(req->r_oid_len);
-	memcpy(p, req->r_oid, req->r_oid_len);
-	p += req->r_oid_len;
 
 	if (do_sync) {
-		op++;
+		ops[1].op = CEPH_OSD_OP_STARTSYNC;
-		op->op = cpu_to_le16(CEPH_OSD_OP_STARTSYNC);
+		ops[1].payload_len = 0;
-	}
+		ops[2].op = 0;
-	if (snapc) {
+	} else
-		head->snap_seq = cpu_to_le64(snapc->seq);
+		ops[1].op = 0;
-		head->num_snaps = cpu_to_le32(snapc->num_snaps);
+
-		for (i = 0; i < snapc->num_snaps; i++) {
+	req = ceph_osdc_alloc_request(osdc, flags,
-			put_unaligned_le64(snapc->snaps[i], p);
+					 snapc, ops,
-			p += sizeof(u64);
+					 use_mempool,
-		}
+					 GFP_NOFS, NULL, NULL);
-	}
+	if (IS_ERR(req))
+		return req;
+
+	/* calculate max write size */
+	calc_layout(osdc, vino, layout, off, plen, req, ops);
+	req->r_file_layout = *layout;  /* keep a copy */
+
+	ceph_osdc_build_request(req, off, plen, ops,
+				snapc,
+				mtime,
+				req->r_oid, req->r_oid_len);
 
-	BUG_ON(p > msg->front.iov_base + msg->front.iov_len);
-	msg_size = p - msg->front.iov_base;
-	msg->front.iov_len = msg_size;
-	msg->hdr.front_len = cpu_to_le32(msg_size);
 	return req;
 }
+EXPORT_SYMBOL(ceph_osdc_new_request);
 
 /*
  * We keep osd requests in an rbtree, sorted by ->r_tid.
@@ -389,7 +609,7 @@ static void __move_osd_to_lru(struct ceph_osd_client *osdc,
 	dout("__move_osd_to_lru %p\n", osd);
 	BUG_ON(!list_empty(&osd->o_osd_lru));
 	list_add_tail(&osd->o_osd_lru, &osdc->osd_lru);
-	osd->lru_ttl = jiffies + osdc->client->mount_args->osd_idle_ttl * HZ;
+	osd->lru_ttl = jiffies + osdc->client->options->osd_idle_ttl * HZ;
 }
 
 static void __remove_osd_from_lru(struct ceph_osd *osd)
@@ -483,7 +703,7 @@ static struct ceph_osd *__lookup_osd(struct ceph_osd_client *osdc, int o)
 static void __schedule_osd_timeout(struct ceph_osd_client *osdc)
 {
 	schedule_delayed_work(&osdc->timeout_work,
-			osdc->client->mount_args->osd_keepalive_timeout * HZ);
+			osdc->client->options->osd_keepalive_timeout * HZ);
 }
 
 static void __cancel_osd_timeout(struct ceph_osd_client *osdc)
@@ -684,9 +904,9 @@ static void handle_timeout(struct work_struct *work)
 		container_of(work, struct ceph_osd_client, timeout_work.work);
 	struct ceph_osd_request *req, *last_req = NULL;
 	struct ceph_osd *osd;
-	unsigned long timeout = osdc->client->mount_args->osd_timeout * HZ;
+	unsigned long timeout = osdc->client->options->osd_timeout * HZ;
 	unsigned long keepalive =
-		osdc->client->mount_args->osd_keepalive_timeout * HZ;
+		osdc->client->options->osd_keepalive_timeout * HZ;
 	unsigned long last_stamp = 0;
 	struct rb_node *p;
 	struct list_head slow_osds;
@@ -773,7 +993,7 @@ static void handle_osds_timeout(struct work_struct *work)
 		container_of(work, struct ceph_osd_client,
 			     osds_timeout_work.work);
 	unsigned long delay =
-		osdc->client->mount_args->osd_idle_ttl * HZ >> 2;
+		osdc->client->options->osd_idle_ttl * HZ >> 2;
 
 	dout("osds timeout\n");
 	down_read(&osdc->map_sem);
@@ -1104,6 +1324,10 @@ int ceph_osdc_start_request(struct ceph_osd_client *osdc,
 
 	req->r_request->pages = req->r_pages;
 	req->r_request->nr_pages = req->r_num_pages;
+#ifdef CONFIG_BLOCK
+	req->r_request->bio = req->r_bio;
+#endif
+	req->r_request->trail = req->r_trail;
 
 	register_request(osdc, req);
 
@@ -1131,6 +1355,7 @@ int ceph_osdc_start_request(struct ceph_osd_client *osdc,
 	up_read(&osdc->map_sem);
 	return rc;
 }
+EXPORT_SYMBOL(ceph_osdc_start_request);
 
 /*
  * wait for a request to complete
@@ -1153,6 +1378,7 @@ int ceph_osdc_wait_request(struct ceph_osd_client *osdc,
 	dout("wait_request tid %llu result %d\n", req->r_tid, req->r_result);
 	return req->r_result;
 }
+EXPORT_SYMBOL(ceph_osdc_wait_request);
 
 /*
  * sync - wait for all in-flight requests to flush.  avoid starvation.
@@ -1186,6 +1412,7 @@ void ceph_osdc_sync(struct ceph_osd_client *osdc)
 	mutex_unlock(&osdc->request_mutex);
 	dout("sync done (thru tid %llu)\n", last_tid);
 }
+EXPORT_SYMBOL(ceph_osdc_sync);
 
 /*
  * init, shutdown
@@ -1211,7 +1438,7 @@ int ceph_osdc_init(struct ceph_osd_client *osdc, struct ceph_client *client)
 	INIT_DELAYED_WORK(&osdc->osds_timeout_work, handle_osds_timeout);
 
 	schedule_delayed_work(&osdc->osds_timeout_work,
-	   round_jiffies_relative(osdc->client->mount_args->osd_idle_ttl * HZ));
+	   round_jiffies_relative(osdc->client->options->osd_idle_ttl * HZ));
 
 	err = -ENOMEM;
 	osdc->req_mempool = mempool_create_kmalloc_pool(10,
@@ -1237,6 +1464,7 @@ out_mempool:
 out:
 	return err;
 }
+EXPORT_SYMBOL(ceph_osdc_init);
 
 void ceph_osdc_stop(struct ceph_osd_client *osdc)
 {
@@ -1251,6 +1479,7 @@ void ceph_osdc_stop(struct ceph_osd_client *osdc)
 	ceph_msgpool_destroy(&osdc->msgpool_op);
 	ceph_msgpool_destroy(&osdc->msgpool_op_reply);
 }
+EXPORT_SYMBOL(ceph_osdc_stop);
 
 /*
  * Read some contiguous pages.  If we cross a stripe boundary, shorten
@@ -1288,6 +1517,7 @@ int ceph_osdc_readpages(struct ceph_osd_client *osdc,
 	dout("readpages result %d\n", rc);
 	return rc;
 }
+EXPORT_SYMBOL(ceph_osdc_readpages);
 
 /*
  * do a synchronous write on N pages
@@ -1330,6 +1560,7 @@ int ceph_osdc_writepages(struct ceph_osd_client *osdc, struct ceph_vino vino,
 	dout("writepages result %d\n", rc);
 	return rc;
 }
+EXPORT_SYMBOL(ceph_osdc_writepages);
 
 /*
  * handle incoming message
@@ -1420,6 +1651,9 @@ static struct ceph_msg *get_reply(struct ceph_connection *con,
 		}
 		m->pages = req->r_pages;
 		m->nr_pages = req->r_num_pages;
+#ifdef CONFIG_BLOCK
+		m->bio = req->r_bio;
+#endif
 	}
 	*skip = 0;
 	req->r_con_filling_msg = ceph_con_get(con);

net/ceph/osdmap.c

@@ -1,14 +1,15 @@
 
-#include "ceph_debug.h"
+#include <linux/ceph/ceph_debug.h>
 
+#include <linux/module.h>
 #include <linux/slab.h>
 #include <asm/div64.h>
 
-#include "super.h"
+#include <linux/ceph/libceph.h>
-#include "osdmap.h"
+#include <linux/ceph/osdmap.h>
-#include "crush/hash.h"
+#include <linux/ceph/decode.h>
-#include "crush/mapper.h"
+#include <linux/crush/hash.h>
-#include "decode.h"
+#include <linux/crush/mapper.h>
 
 char *ceph_osdmap_state_str(char *str, int len, int state)
 {
@@ -417,6 +418,20 @@ static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, int id)
 	return NULL;
 }
 
+int ceph_pg_poolid_by_name(struct ceph_osdmap *map, const char *name)
+{
+	struct rb_node *rbp;
+
+	for (rbp = rb_first(&map->pg_pools); rbp; rbp = rb_next(rbp)) {
+		struct ceph_pg_pool_info *pi =
+			rb_entry(rbp, struct ceph_pg_pool_info, node);
+		if (pi->name && strcmp(pi->name, name) == 0)
+			return pi->id;
+	}
+	return -ENOENT;
+}
+EXPORT_SYMBOL(ceph_pg_poolid_by_name);
+
 static void __remove_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *pi)
 {
 	rb_erase(&pi->node, root);
@@ -966,6 +981,7 @@ void ceph_calc_file_object_mapping(struct ceph_file_layout *layout,
 
 	dout(" obj extent %llu~%llu\n", *oxoff, *oxlen);
 }
+EXPORT_SYMBOL(ceph_calc_file_object_mapping);
 
 /*
  * calculate an object layout (i.e. pgid) from an oid,
@@ -1011,6 +1027,7 @@ int ceph_calc_object_layout(struct ceph_object_layout *ol,
 	ol->ol_stripe_unit = fl->fl_object_stripe_unit;
 	return 0;
 }
+EXPORT_SYMBOL(ceph_calc_object_layout);
 
 /*
  * Calculate raw osd vector for the given pgid.  Return pointer to osd
@@ -1108,3 +1125,4 @@ int ceph_calc_pg_primary(struct ceph_osdmap *osdmap, struct ceph_pg pgid)
 			return osds[i];
 	return -1;
 }
+EXPORT_SYMBOL(ceph_calc_pg_primary);

net/ceph/pagelist.c

@@ -0,0 +1,154 @@
+
+#include <linux/module.h>
+#include <linux/gfp.h>
+#include <linux/pagemap.h>
+#include <linux/highmem.h>
+#include <linux/ceph/pagelist.h>
+
+static void ceph_pagelist_unmap_tail(struct ceph_pagelist *pl)
+{
+	if (pl->mapped_tail) {
+		struct page *page = list_entry(pl->head.prev, struct page, lru);
+		kunmap(page);
+		pl->mapped_tail = NULL;
+	}
+}
+
+int ceph_pagelist_release(struct ceph_pagelist *pl)
+{
+	ceph_pagelist_unmap_tail(pl);
+	while (!list_empty(&pl->head)) {
+		struct page *page = list_first_entry(&pl->head, struct page,
+						     lru);
+		list_del(&page->lru);
+		__free_page(page);
+	}
+	ceph_pagelist_free_reserve(pl);
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_release);
+
+static int ceph_pagelist_addpage(struct ceph_pagelist *pl)
+{
+	struct page *page;
+
+	if (!pl->num_pages_free) {
+		page = __page_cache_alloc(GFP_NOFS);
+	} else {
+		page = list_first_entry(&pl->free_list, struct page, lru);
+		list_del(&page->lru);
+		--pl->num_pages_free;
+	}
+	if (!page)
+		return -ENOMEM;
+	pl->room += PAGE_SIZE;
+	ceph_pagelist_unmap_tail(pl);
+	list_add_tail(&page->lru, &pl->head);
+	pl->mapped_tail = kmap(page);
+	return 0;
+}
+
+int ceph_pagelist_append(struct ceph_pagelist *pl, const void *buf, size_t len)
+{
+	while (pl->room < len) {
+		size_t bit = pl->room;
+		int ret;
+
+		memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK),
+		       buf, bit);
+		pl->length += bit;
+		pl->room -= bit;
+		buf += bit;
+		len -= bit;
+		ret = ceph_pagelist_addpage(pl);
+		if (ret)
+			return ret;
+	}
+
+	memcpy(pl->mapped_tail + (pl->length & ~PAGE_CACHE_MASK), buf, len);
+	pl->length += len;
+	pl->room -= len;
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_append);
+
+/**
+ * Allocate enough pages for a pagelist to append the given amount
+ * of data without without allocating.
+ * Returns: 0 on success, -ENOMEM on error.
+ */
+int ceph_pagelist_reserve(struct ceph_pagelist *pl, size_t space)
+{
+	if (space <= pl->room)
+		return 0;
+	space -= pl->room;
+	space = (space + PAGE_SIZE - 1) >> PAGE_SHIFT;   /* conv to num pages */
+
+	while (space > pl->num_pages_free) {
+		struct page *page = __page_cache_alloc(GFP_NOFS);
+		if (!page)
+			return -ENOMEM;
+		list_add_tail(&page->lru, &pl->free_list);
+		++pl->num_pages_free;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_reserve);
+
+/**
+ * Free any pages that have been preallocated.
+ */
+int ceph_pagelist_free_reserve(struct ceph_pagelist *pl)
+{
+	while (!list_empty(&pl->free_list)) {
+		struct page *page = list_first_entry(&pl->free_list,
+						     struct page, lru);
+		list_del(&page->lru);
+		__free_page(page);
+		--pl->num_pages_free;
+	}
+	BUG_ON(pl->num_pages_free);
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_free_reserve);
+
+/**
+ * Create a truncation point.
+ */
+void ceph_pagelist_set_cursor(struct ceph_pagelist *pl,
+			      struct ceph_pagelist_cursor *c)
+{
+	c->pl = pl;
+	c->page_lru = pl->head.prev;
+	c->room = pl->room;
+}
+EXPORT_SYMBOL(ceph_pagelist_set_cursor);
+
+/**
+ * Truncate a pagelist to the given point. Move extra pages to reserve.
+ * This won't sleep.
+ * Returns: 0 on success,
+ *          -EINVAL if the pagelist doesn't match the trunc point pagelist
+ */
+int ceph_pagelist_truncate(struct ceph_pagelist *pl,
+			   struct ceph_pagelist_cursor *c)
+{
+	struct page *page;
+
+	if (pl != c->pl)
+		return -EINVAL;
+	ceph_pagelist_unmap_tail(pl);
+	while (pl->head.prev != c->page_lru) {
+		page = list_entry(pl->head.prev, struct page, lru);
+		list_del(&page->lru);                /* remove from pagelist */
+		list_add_tail(&page->lru, &pl->free_list); /* add to reserve */
+		++pl->num_pages_free;
+	}
+	pl->room = c->room;
+	if (!list_empty(&pl->head)) {
+		page = list_entry(pl->head.prev, struct page, lru);
+		pl->mapped_tail = kmap(page);
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ceph_pagelist_truncate);

net/ceph/pagevec.c

@@ -0,0 +1,223 @@
+#include <linux/ceph/ceph_debug.h>
+
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/file.h>
+#include <linux/namei.h>
+#include <linux/writeback.h>
+
+#include <linux/ceph/libceph.h>
+
+/*
+ * build a vector of user pages
+ */
+struct page **ceph_get_direct_page_vector(const char __user *data,
+						 int num_pages,
+						 loff_t off, size_t len)
+{
+	struct page **pages;
+	int rc;
+
+	pages = kmalloc(sizeof(*pages) * num_pages, GFP_NOFS);
+	if (!pages)
+		return ERR_PTR(-ENOMEM);
+
+	down_read(&current->mm->mmap_sem);
+	rc = get_user_pages(current, current->mm, (unsigned long)data,
+			    num_pages, 0, 0, pages, NULL);
+	up_read(&current->mm->mmap_sem);
+	if (rc < 0)
+		goto fail;
+	return pages;
+
+fail:
+	kfree(pages);
+	return ERR_PTR(rc);
+}
+EXPORT_SYMBOL(ceph_get_direct_page_vector);
+
+void ceph_put_page_vector(struct page **pages, int num_pages)
+{
+	int i;
+
+	for (i = 0; i < num_pages; i++)
+		put_page(pages[i]);
+	kfree(pages);
+}
+EXPORT_SYMBOL(ceph_put_page_vector);
+
+void ceph_release_page_vector(struct page **pages, int num_pages)
+{
+	int i;
+
+	for (i = 0; i < num_pages; i++)
+		__free_pages(pages[i], 0);
+	kfree(pages);
+}
+EXPORT_SYMBOL(ceph_release_page_vector);
+
+/*
+ * allocate a vector new pages
+ */
+struct page **ceph_alloc_page_vector(int num_pages, gfp_t flags)
+{
+	struct page **pages;
+	int i;
+
+	pages = kmalloc(sizeof(*pages) * num_pages, flags);
+	if (!pages)
+		return ERR_PTR(-ENOMEM);
+	for (i = 0; i < num_pages; i++) {
+		pages[i] = __page_cache_alloc(flags);
+		if (pages[i] == NULL) {
+			ceph_release_page_vector(pages, i);
+			return ERR_PTR(-ENOMEM);
+		}
+	}
+	return pages;
+}
+EXPORT_SYMBOL(ceph_alloc_page_vector);
+
+/*
+ * copy user data into a page vector
+ */
+int ceph_copy_user_to_page_vector(struct page **pages,
+					 const char __user *data,
+					 loff_t off, size_t len)
+{
+	int i = 0;
+	int po = off & ~PAGE_CACHE_MASK;
+	int left = len;
+	int l, bad;
+
+	while (left > 0) {
+		l = min_t(int, PAGE_CACHE_SIZE-po, left);
+		bad = copy_from_user(page_address(pages[i]) + po, data, l);
+		if (bad == l)
+			return -EFAULT;
+		data += l - bad;
+		left -= l - bad;
+		po += l - bad;
+		if (po == PAGE_CACHE_SIZE) {
+			po = 0;
+			i++;
+		}
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_user_to_page_vector);
+
+int ceph_copy_to_page_vector(struct page **pages,
+				    const char *data,
+				    loff_t off, size_t len)
+{
+	int i = 0;
+	size_t po = off & ~PAGE_CACHE_MASK;
+	size_t left = len;
+	size_t l;
+
+	while (left > 0) {
+		l = min_t(size_t, PAGE_CACHE_SIZE-po, left);
+		memcpy(page_address(pages[i]) + po, data, l);
+		data += l;
+		left -= l;
+		po += l;
+		if (po == PAGE_CACHE_SIZE) {
+			po = 0;
+			i++;
+		}
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_to_page_vector);
+
+int ceph_copy_from_page_vector(struct page **pages,
+				    char *data,
+				    loff_t off, size_t len)
+{
+	int i = 0;
+	size_t po = off & ~PAGE_CACHE_MASK;
+	size_t left = len;
+	size_t l;
+
+	while (left > 0) {
+		l = min_t(size_t, PAGE_CACHE_SIZE-po, left);
+		memcpy(data, page_address(pages[i]) + po, l);
+		data += l;
+		left -= l;
+		po += l;
+		if (po == PAGE_CACHE_SIZE) {
+			po = 0;
+			i++;
+		}
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_from_page_vector);
+
+/*
+ * copy user data from a page vector into a user pointer
+ */
+int ceph_copy_page_vector_to_user(struct page **pages,
+					 char __user *data,
+					 loff_t off, size_t len)
+{
+	int i = 0;
+	int po = off & ~PAGE_CACHE_MASK;
+	int left = len;
+	int l, bad;
+
+	while (left > 0) {
+		l = min_t(int, left, PAGE_CACHE_SIZE-po);
+		bad = copy_to_user(data, page_address(pages[i]) + po, l);
+		if (bad == l)
+			return -EFAULT;
+		data += l - bad;
+		left -= l - bad;
+		if (po) {
+			po += l - bad;
+			if (po == PAGE_CACHE_SIZE)
+				po = 0;
+		}
+		i++;
+	}
+	return len;
+}
+EXPORT_SYMBOL(ceph_copy_page_vector_to_user);
+
+/*
+ * Zero an extent within a page vector.  Offset is relative to the
+ * start of the first page.
+ */
+void ceph_zero_page_vector_range(int off, int len, struct page **pages)
+{
+	int i = off >> PAGE_CACHE_SHIFT;
+
+	off &= ~PAGE_CACHE_MASK;
+
+	dout("zero_page_vector_page %u~%u\n", off, len);
+
+	/* leading partial page? */
+	if (off) {
+		int end = min((int)PAGE_CACHE_SIZE, off + len);
+		dout("zeroing %d %p head from %d\n", i, pages[i],
+		     (int)off);
+		zero_user_segment(pages[i], off, end);
+		len -= (end - off);
+		i++;
+	}
+	while (len >= PAGE_CACHE_SIZE) {
+		dout("zeroing %d %p len=%d\n", i, pages[i], len);
+		zero_user_segment(pages[i], 0, PAGE_CACHE_SIZE);
+		len -= PAGE_CACHE_SIZE;
+		i++;
+	}
+	/* trailing partial page? */
+	if (len) {
+		dout("zeroing %d %p tail to %d\n", i, pages[i], (int)len);
+		zero_user_segment(pages[i], 0, len);
+	}
+}
+EXPORT_SYMBOL(ceph_zero_page_vector_range);
+