linux/fs/nfs/blocklayout/blocklayout.h

205 lines
5.8 KiB
C
Raw Permalink Normal View History

/*
* linux/fs/nfs/blocklayout/blocklayout.h
*
* Module for the NFSv4.1 pNFS block layout driver.
*
* Copyright (c) 2006 The Regents of the University of Michigan.
* All rights reserved.
*
* Andy Adamson <andros@citi.umich.edu>
* Fred Isaman <iisaman@umich.edu>
*
* permission is granted to use, copy, create derivative works and
* redistribute this software and such derivative works for any purpose,
* so long as the name of the university of michigan is not used in
* any advertising or publicity pertaining to the use or distribution
* of this software without specific, written prior authorization. if
* the above copyright notice or any other identification of the
* university of michigan is included in any copy of any portion of
* this software, then the disclaimer below must also be included.
*
* this software is provided as is, without representation from the
* university of michigan as to its fitness for any purpose, and without
* warranty by the university of michigan of any kind, either express
* or implied, including without limitation the implied warranties of
* merchantability and fitness for a particular purpose. the regents
* of the university of michigan shall not be liable for any damages,
* including special, indirect, incidental, or consequential damages,
* with respect to any claim arising out or in connection with the use
* of the software, even if it has been or is hereafter advised of the
* possibility of such damages.
*/
#ifndef FS_NFS_NFS4BLOCKLAYOUT_H
#define FS_NFS_NFS4BLOCKLAYOUT_H
#include <linux/device-mapper.h>
#include <linux/nfs_fs.h>
#include <linux/sunrpc/rpc_pipe_fs.h>
#include "../nfs4_fs.h"
#include "../pnfs.h"
#include "../netns.h"
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
#define PAGE_CACHE_SECTORS (PAGE_SIZE >> SECTOR_SHIFT)
#define PAGE_CACHE_SECTOR_SHIFT (PAGE_SHIFT - SECTOR_SHIFT)
#define SECTOR_SIZE (1 << SECTOR_SHIFT)
struct pnfs_block_dev;
#define PNFS_BLOCK_MAX_UUIDS 4
#define PNFS_BLOCK_MAX_DEVICES 64
/*
* Random upper cap for the uuid length to avoid unbounded allocation.
* Not actually limited by the protocol.
*/
#define PNFS_BLOCK_UUID_LEN 128
struct pnfs_block_volume {
enum pnfs_block_volume_type type;
union {
struct {
int len;
int nr_sigs;
struct {
u64 offset;
u32 sig_len;
u8 sig[PNFS_BLOCK_UUID_LEN];
} sigs[PNFS_BLOCK_MAX_UUIDS];
} simple;
struct {
u64 start;
u64 len;
u32 volume;
} slice;
struct {
u32 volumes_count;
u32 volumes[PNFS_BLOCK_MAX_DEVICES];
} concat;
struct {
u64 chunk_size;
u32 volumes_count;
u32 volumes[PNFS_BLOCK_MAX_DEVICES];
} stripe;
struct {
enum scsi_code_set code_set;
enum scsi_designator_type designator_type;
int designator_len;
u8 designator[256];
u64 pr_key;
} scsi;
};
};
struct pnfs_block_dev_map {
u64 start;
u64 len;
u64 disk_offset;
struct block_device *bdev;
};
struct pnfs_block_dev {
struct nfs4_deviceid_node node;
u64 start;
u64 len;
nfs/blocklayout: Fix premature PR key unregistration During generic/069 runs with pNFS SCSI layouts, the NFS client emits the following in the system journal: kernel: pNFS: failed to open device /dev/disk/by-id/dm-uuid-mpath-0x6001405e3366f045b7949eb8e4540b51 (-2) kernel: pNFS: using block device sdb (reservation key 0x666b60901e7b26b3) kernel: pNFS: failed to open device /dev/disk/by-id/dm-uuid-mpath-0x6001405e3366f045b7949eb8e4540b51 (-2) kernel: pNFS: using block device sdb (reservation key 0x666b60901e7b26b3) kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: [sdb] tag#16 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#16 CDB: Write(10) 2a 00 00 00 00 50 00 00 08 00 kernel: reservation conflict error, dev sdb, sector 80 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 2 kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: [sdb] tag#18 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#17 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#18 CDB: Write(10) 2a 00 00 00 00 60 00 00 08 00 kernel: sd 6:0:0:1: [sdb] tag#17 CDB: Write(10) 2a 00 00 00 00 58 00 00 08 00 kernel: reservation conflict error, dev sdb, sector 96 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 0 kernel: reservation conflict error, dev sdb, sector 88 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 0 systemd[1]: fstests-generic-069.scope: Deactivated successfully. systemd[1]: fstests-generic-069.scope: Consumed 5.092s CPU time. systemd[1]: media-test.mount: Deactivated successfully. systemd[1]: media-scratch.mount: Deactivated successfully. kernel: sd 6:0:0:1: reservation conflict kernel: failed to unregister PR key. This appears to be due to a race. bl_alloc_lseg() calls this: 561 static struct nfs4_deviceid_node * 562 bl_find_get_deviceid(struct nfs_server *server, 563 const struct nfs4_deviceid *id, const struct cred *cred, 564 gfp_t gfp_mask) 565 { 566 struct nfs4_deviceid_node *node; 567 unsigned long start, end; 568 569 retry: 570 node = nfs4_find_get_deviceid(server, id, cred, gfp_mask); 571 if (!node) 572 return ERR_PTR(-ENODEV); nfs4_find_get_deviceid() does a lookup without the spin lock first. If it can't find a matching deviceid, it creates a new device_info (which calls bl_alloc_deviceid_node, and that registers the device's PR key). Then it takes the nfs4_deviceid_lock and looks up the deviceid again. If it finds it this time, bl_find_get_deviceid() frees the spare (new) device_info, which unregisters the PR key for the same device. Any subsequent I/O from this client on that device gets EBADE. The umount later unregisters the device's PR key again. To prevent this problem, register the PR key after the deviceid_node lookup. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2024-06-25 20:02:06 +00:00
enum pnfs_block_volume_type type;
u32 nr_children;
struct pnfs_block_dev *children;
u64 chunk_size;
struct file *bdev_file;
u64 disk_offset;
nfs/blocklayout: Fix premature PR key unregistration During generic/069 runs with pNFS SCSI layouts, the NFS client emits the following in the system journal: kernel: pNFS: failed to open device /dev/disk/by-id/dm-uuid-mpath-0x6001405e3366f045b7949eb8e4540b51 (-2) kernel: pNFS: using block device sdb (reservation key 0x666b60901e7b26b3) kernel: pNFS: failed to open device /dev/disk/by-id/dm-uuid-mpath-0x6001405e3366f045b7949eb8e4540b51 (-2) kernel: pNFS: using block device sdb (reservation key 0x666b60901e7b26b3) kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: [sdb] tag#16 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#16 CDB: Write(10) 2a 00 00 00 00 50 00 00 08 00 kernel: reservation conflict error, dev sdb, sector 80 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 2 kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: [sdb] tag#18 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#17 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#18 CDB: Write(10) 2a 00 00 00 00 60 00 00 08 00 kernel: sd 6:0:0:1: [sdb] tag#17 CDB: Write(10) 2a 00 00 00 00 58 00 00 08 00 kernel: reservation conflict error, dev sdb, sector 96 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 0 kernel: reservation conflict error, dev sdb, sector 88 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 0 systemd[1]: fstests-generic-069.scope: Deactivated successfully. systemd[1]: fstests-generic-069.scope: Consumed 5.092s CPU time. systemd[1]: media-test.mount: Deactivated successfully. systemd[1]: media-scratch.mount: Deactivated successfully. kernel: sd 6:0:0:1: reservation conflict kernel: failed to unregister PR key. This appears to be due to a race. bl_alloc_lseg() calls this: 561 static struct nfs4_deviceid_node * 562 bl_find_get_deviceid(struct nfs_server *server, 563 const struct nfs4_deviceid *id, const struct cred *cred, 564 gfp_t gfp_mask) 565 { 566 struct nfs4_deviceid_node *node; 567 unsigned long start, end; 568 569 retry: 570 node = nfs4_find_get_deviceid(server, id, cred, gfp_mask); 571 if (!node) 572 return ERR_PTR(-ENODEV); nfs4_find_get_deviceid() does a lookup without the spin lock first. If it can't find a matching deviceid, it creates a new device_info (which calls bl_alloc_deviceid_node, and that registers the device's PR key). Then it takes the nfs4_deviceid_lock and looks up the deviceid again. If it finds it this time, bl_find_get_deviceid() frees the spare (new) device_info, which unregisters the PR key for the same device. Any subsequent I/O from this client on that device gets EBADE. The umount later unregisters the device's PR key again. To prevent this problem, register the PR key after the deviceid_node lookup. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2024-06-25 20:02:06 +00:00
unsigned long flags;
u64 pr_key;
bool (*map)(struct pnfs_block_dev *dev, u64 offset,
struct pnfs_block_dev_map *map);
};
nfs/blocklayout: Fix premature PR key unregistration During generic/069 runs with pNFS SCSI layouts, the NFS client emits the following in the system journal: kernel: pNFS: failed to open device /dev/disk/by-id/dm-uuid-mpath-0x6001405e3366f045b7949eb8e4540b51 (-2) kernel: pNFS: using block device sdb (reservation key 0x666b60901e7b26b3) kernel: pNFS: failed to open device /dev/disk/by-id/dm-uuid-mpath-0x6001405e3366f045b7949eb8e4540b51 (-2) kernel: pNFS: using block device sdb (reservation key 0x666b60901e7b26b3) kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: [sdb] tag#16 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#16 CDB: Write(10) 2a 00 00 00 00 50 00 00 08 00 kernel: reservation conflict error, dev sdb, sector 80 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 2 kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: [sdb] tag#18 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#17 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#18 CDB: Write(10) 2a 00 00 00 00 60 00 00 08 00 kernel: sd 6:0:0:1: [sdb] tag#17 CDB: Write(10) 2a 00 00 00 00 58 00 00 08 00 kernel: reservation conflict error, dev sdb, sector 96 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 0 kernel: reservation conflict error, dev sdb, sector 88 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 0 systemd[1]: fstests-generic-069.scope: Deactivated successfully. systemd[1]: fstests-generic-069.scope: Consumed 5.092s CPU time. systemd[1]: media-test.mount: Deactivated successfully. systemd[1]: media-scratch.mount: Deactivated successfully. kernel: sd 6:0:0:1: reservation conflict kernel: failed to unregister PR key. This appears to be due to a race. bl_alloc_lseg() calls this: 561 static struct nfs4_deviceid_node * 562 bl_find_get_deviceid(struct nfs_server *server, 563 const struct nfs4_deviceid *id, const struct cred *cred, 564 gfp_t gfp_mask) 565 { 566 struct nfs4_deviceid_node *node; 567 unsigned long start, end; 568 569 retry: 570 node = nfs4_find_get_deviceid(server, id, cred, gfp_mask); 571 if (!node) 572 return ERR_PTR(-ENODEV); nfs4_find_get_deviceid() does a lookup without the spin lock first. If it can't find a matching deviceid, it creates a new device_info (which calls bl_alloc_deviceid_node, and that registers the device's PR key). Then it takes the nfs4_deviceid_lock and looks up the deviceid again. If it finds it this time, bl_find_get_deviceid() frees the spare (new) device_info, which unregisters the PR key for the same device. Any subsequent I/O from this client on that device gets EBADE. The umount later unregisters the device's PR key again. To prevent this problem, register the PR key after the deviceid_node lookup. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2024-06-25 20:02:06 +00:00
/* pnfs_block_dev flag bits */
enum {
PNFS_BDEV_REGISTERED = 0,
};
/* sector_t fields are all in 512-byte sectors */
struct pnfs_block_extent {
pnfs/blocklayout: rewrite extent tracking Currently the block layout driver tracks extents in three separate data structures: - the two list of pnfs_block_extent structures returned by the server - the list of sectors that were in invalid state but have been written to - a list of pnfs_block_short_extent structures for LAYOUTCOMMIT All of these share the property that they are not only highly inefficient data structures, but also that operations on them are even more inefficient than nessecary. In addition there are various implementation defects like: - using an int to track sectors, causing corruption for large offsets - incorrect normalization of page or block granularity ranges - insufficient error handling - incorrect synchronization as extents can be modified while they are in use This patch replace all three data with a single unified rbtree structure tracking all extents, as well as their in-memory state, although we still need to instance for read-only and read-write extent due to the arcane client side COW feature in the block layouts spec. To fix the problem of extent possibly being modified while in use we make sure to return a copy of the extent for use in the write path - the extent can only be invalidated by a layout recall or return which has to wait until the I/O operations finished due to refcounts on the layout segment. The new extent tree work similar to the schemes used by block based filesystems like XFS or ext4. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
2014-09-10 15:23:34 +00:00
union {
struct rb_node be_node;
struct list_head be_list;
};
struct nfs4_deviceid_node *be_device;
sector_t be_f_offset; /* the starting offset in the file */
sector_t be_length; /* the size of the extent */
sector_t be_v_offset; /* the starting offset in the volume */
enum pnfs_block_extent_state be_state; /* the state of this extent */
pnfs/blocklayout: rewrite extent tracking Currently the block layout driver tracks extents in three separate data structures: - the two list of pnfs_block_extent structures returned by the server - the list of sectors that were in invalid state but have been written to - a list of pnfs_block_short_extent structures for LAYOUTCOMMIT All of these share the property that they are not only highly inefficient data structures, but also that operations on them are even more inefficient than nessecary. In addition there are various implementation defects like: - using an int to track sectors, causing corruption for large offsets - incorrect normalization of page or block granularity ranges - insufficient error handling - incorrect synchronization as extents can be modified while they are in use This patch replace all three data with a single unified rbtree structure tracking all extents, as well as their in-memory state, although we still need to instance for read-only and read-write extent due to the arcane client side COW feature in the block layouts spec. To fix the problem of extent possibly being modified while in use we make sure to return a copy of the extent for use in the write path - the extent can only be invalidated by a layout recall or return which has to wait until the I/O operations finished due to refcounts on the layout segment. The new extent tree work similar to the schemes used by block based filesystems like XFS or ext4. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
2014-09-10 15:23:34 +00:00
#define EXTENT_WRITTEN 1
#define EXTENT_COMMITTING 2
unsigned int be_tag;
pnfsblock: encode_layoutcommit In blocklayout driver. There are two things happening while layoutcommit/cleanup. 1. the modified extents are encoded. 2. On cleanup the extents are put back on the layout rw extents list, for reads. In the new system where actual xdr encoding is done in encode_layoutcommit() directly into xdr buffer, these are the new commit stages: 1. On setup_layoutcommit, the range is adjusted as before and a structure is allocated for communication with bl_encode_layoutcommit && bl_cleanup_layoutcommit (Generic layer provides a void-star to hang it on) 2. bl_encode_layoutcommit is called to do the actual encoding directly into xdr. The commit-extent-list is not freed and is stored on above structure. FIXME: The code is not yet converted to the new XDR cleanup 3. On cleanup the commit-extent-list is put back by a call to set_to_rw() as before, but with no need for XDR decoding of the list as before. And the commit-extent-list is freed. Finally allocated structure is freed. [rm inode and pnfs_layout_hdr args from cleanup_layoutcommit()] [pnfsblock: get rid of deprecated xdr macros] Signed-off-by: Jim Rees <rees@umich.edu> Signed-off-by: Peng Tao <peng_tao@emc.com> Signed-off-by: Fred Isaman <iisaman@citi.umich.edu> [blocklayout: encode_layoutcommit implementation] Signed-off-by: Boaz Harrosh <bharrosh@panasas.com> [pnfsblock: fix bug setting up layoutcommit.] Signed-off-by: Tao Guo <guotao@nrchpc.ac.cn> [pnfsblock: prevent commit list corruption] [pnfsblock: fix layoutcommit with an empty opaque] Signed-off-by: Fred Isaman <iisaman@citi.umich.edu> Signed-off-by: Benny Halevy <bhalevy@panasas.com> Signed-off-by: Benny Halevy <bhalevy@tonian.com> Signed-off-by: Jim Rees <rees@umich.edu> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2011-07-31 00:52:51 +00:00
};
struct pnfs_block_layout {
pnfs/blocklayout: rewrite extent tracking Currently the block layout driver tracks extents in three separate data structures: - the two list of pnfs_block_extent structures returned by the server - the list of sectors that were in invalid state but have been written to - a list of pnfs_block_short_extent structures for LAYOUTCOMMIT All of these share the property that they are not only highly inefficient data structures, but also that operations on them are even more inefficient than nessecary. In addition there are various implementation defects like: - using an int to track sectors, causing corruption for large offsets - incorrect normalization of page or block granularity ranges - insufficient error handling - incorrect synchronization as extents can be modified while they are in use This patch replace all three data with a single unified rbtree structure tracking all extents, as well as their in-memory state, although we still need to instance for read-only and read-write extent due to the arcane client side COW feature in the block layouts spec. To fix the problem of extent possibly being modified while in use we make sure to return a copy of the extent for use in the write path - the extent can only be invalidated by a layout recall or return which has to wait until the I/O operations finished due to refcounts on the layout segment. The new extent tree work similar to the schemes used by block based filesystems like XFS or ext4. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
2014-09-10 15:23:34 +00:00
struct pnfs_layout_hdr bl_layout;
struct rb_root bl_ext_rw;
struct rb_root bl_ext_ro;
spinlock_t bl_ext_lock; /* Protects list manipulation */
bool bl_scsi_layout;
u64 bl_lwb;
};
static inline struct pnfs_block_layout *
BLK_LO2EXT(struct pnfs_layout_hdr *lo)
{
return container_of(lo, struct pnfs_block_layout, bl_layout);
}
static inline struct pnfs_block_layout *
BLK_LSEG2EXT(struct pnfs_layout_segment *lseg)
{
return BLK_LO2EXT(lseg->pls_layout);
}
struct bl_pipe_msg {
struct rpc_pipe_msg msg;
wait_queue_head_t *bl_wq;
};
struct bl_msg_hdr {
u8 type;
u16 totallen; /* length of entire message, including hdr itself */
};
#define BL_DEVICE_UMOUNT 0x0 /* Umount--delete devices */
#define BL_DEVICE_MOUNT 0x1 /* Mount--create devices*/
#define BL_DEVICE_REQUEST_INIT 0x0 /* Start request */
#define BL_DEVICE_REQUEST_PROC 0x1 /* User level process succeeds */
#define BL_DEVICE_REQUEST_ERR 0x2 /* User level process fails */
/* dev.c */
nfs/blocklayout: Fix premature PR key unregistration During generic/069 runs with pNFS SCSI layouts, the NFS client emits the following in the system journal: kernel: pNFS: failed to open device /dev/disk/by-id/dm-uuid-mpath-0x6001405e3366f045b7949eb8e4540b51 (-2) kernel: pNFS: using block device sdb (reservation key 0x666b60901e7b26b3) kernel: pNFS: failed to open device /dev/disk/by-id/dm-uuid-mpath-0x6001405e3366f045b7949eb8e4540b51 (-2) kernel: pNFS: using block device sdb (reservation key 0x666b60901e7b26b3) kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: [sdb] tag#16 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#16 CDB: Write(10) 2a 00 00 00 00 50 00 00 08 00 kernel: reservation conflict error, dev sdb, sector 80 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 2 kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: reservation conflict kernel: sd 6:0:0:1: [sdb] tag#18 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#17 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_OK cmd_age=0s kernel: sd 6:0:0:1: [sdb] tag#18 CDB: Write(10) 2a 00 00 00 00 60 00 00 08 00 kernel: sd 6:0:0:1: [sdb] tag#17 CDB: Write(10) 2a 00 00 00 00 58 00 00 08 00 kernel: reservation conflict error, dev sdb, sector 96 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 0 kernel: reservation conflict error, dev sdb, sector 88 op 0x1:(WRITE) flags 0x0 phys_seg 1 prio class 0 systemd[1]: fstests-generic-069.scope: Deactivated successfully. systemd[1]: fstests-generic-069.scope: Consumed 5.092s CPU time. systemd[1]: media-test.mount: Deactivated successfully. systemd[1]: media-scratch.mount: Deactivated successfully. kernel: sd 6:0:0:1: reservation conflict kernel: failed to unregister PR key. This appears to be due to a race. bl_alloc_lseg() calls this: 561 static struct nfs4_deviceid_node * 562 bl_find_get_deviceid(struct nfs_server *server, 563 const struct nfs4_deviceid *id, const struct cred *cred, 564 gfp_t gfp_mask) 565 { 566 struct nfs4_deviceid_node *node; 567 unsigned long start, end; 568 569 retry: 570 node = nfs4_find_get_deviceid(server, id, cred, gfp_mask); 571 if (!node) 572 return ERR_PTR(-ENODEV); nfs4_find_get_deviceid() does a lookup without the spin lock first. If it can't find a matching deviceid, it creates a new device_info (which calls bl_alloc_deviceid_node, and that registers the device's PR key). Then it takes the nfs4_deviceid_lock and looks up the deviceid again. If it finds it this time, bl_find_get_deviceid() frees the spare (new) device_info, which unregisters the PR key for the same device. Any subsequent I/O from this client on that device gets EBADE. The umount later unregisters the device's PR key again. To prevent this problem, register the PR key after the deviceid_node lookup. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Benjamin Coddington <bcodding@redhat.com> Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
2024-06-25 20:02:06 +00:00
bool bl_register_dev(struct pnfs_block_dev *d);
struct nfs4_deviceid_node *bl_alloc_deviceid_node(struct nfs_server *server,
struct pnfs_device *pdev, gfp_t gfp_mask);
void bl_free_deviceid_node(struct nfs4_deviceid_node *d);
pnfs/blocklayout: rewrite extent tracking Currently the block layout driver tracks extents in three separate data structures: - the two list of pnfs_block_extent structures returned by the server - the list of sectors that were in invalid state but have been written to - a list of pnfs_block_short_extent structures for LAYOUTCOMMIT All of these share the property that they are not only highly inefficient data structures, but also that operations on them are even more inefficient than nessecary. In addition there are various implementation defects like: - using an int to track sectors, causing corruption for large offsets - incorrect normalization of page or block granularity ranges - insufficient error handling - incorrect synchronization as extents can be modified while they are in use This patch replace all three data with a single unified rbtree structure tracking all extents, as well as their in-memory state, although we still need to instance for read-only and read-write extent due to the arcane client side COW feature in the block layouts spec. To fix the problem of extent possibly being modified while in use we make sure to return a copy of the extent for use in the write path - the extent can only be invalidated by a layout recall or return which has to wait until the I/O operations finished due to refcounts on the layout segment. The new extent tree work similar to the schemes used by block based filesystems like XFS or ext4. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
2014-09-10 15:23:34 +00:00
/* extent_tree.c */
int ext_tree_insert(struct pnfs_block_layout *bl,
struct pnfs_block_extent *new);
int ext_tree_remove(struct pnfs_block_layout *bl, bool rw, sector_t start,
sector_t end);
int ext_tree_mark_written(struct pnfs_block_layout *bl, sector_t start,
sector_t len, u64 lwb);
pnfs/blocklayout: rewrite extent tracking Currently the block layout driver tracks extents in three separate data structures: - the two list of pnfs_block_extent structures returned by the server - the list of sectors that were in invalid state but have been written to - a list of pnfs_block_short_extent structures for LAYOUTCOMMIT All of these share the property that they are not only highly inefficient data structures, but also that operations on them are even more inefficient than nessecary. In addition there are various implementation defects like: - using an int to track sectors, causing corruption for large offsets - incorrect normalization of page or block granularity ranges - insufficient error handling - incorrect synchronization as extents can be modified while they are in use This patch replace all three data with a single unified rbtree structure tracking all extents, as well as their in-memory state, although we still need to instance for read-only and read-write extent due to the arcane client side COW feature in the block layouts spec. To fix the problem of extent possibly being modified while in use we make sure to return a copy of the extent for use in the write path - the extent can only be invalidated by a layout recall or return which has to wait until the I/O operations finished due to refcounts on the layout segment. The new extent tree work similar to the schemes used by block based filesystems like XFS or ext4. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>
2014-09-10 15:23:34 +00:00
bool ext_tree_lookup(struct pnfs_block_layout *bl, sector_t isect,
struct pnfs_block_extent *ret, bool rw);
int ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg);
void ext_tree_mark_committed(struct nfs4_layoutcommit_args *arg, int status);
/* rpc_pipefs.c */
dev_t bl_resolve_deviceid(struct nfs_server *server,
struct pnfs_block_volume *b, gfp_t gfp_mask);
int __init bl_init_pipefs(void);
void bl_cleanup_pipefs(void);
#endif /* FS_NFS_NFS4BLOCKLAYOUT_H */