diff options
Diffstat (limited to 'fs/afs')
| -rw-r--r-- | fs/afs/Kconfig | 29 | ||||
| -rw-r--r-- | fs/afs/Makefile | 7 | ||||
| -rw-r--r-- | fs/afs/afs.h | 42 | ||||
| -rw-r--r-- | fs/afs/afs_cm.h | 3 | ||||
| -rw-r--r-- | fs/afs/afs_fs.h | 10 | ||||
| -rw-r--r-- | fs/afs/afs_vl.h | 2 | ||||
| -rw-r--r-- | fs/afs/cache.c | 502 | ||||
| -rw-r--r-- | fs/afs/cache.h | 23 | ||||
| -rw-r--r-- | fs/afs/callback.c | 21 | ||||
| -rw-r--r-- | fs/afs/cell.c | 150 | ||||
| -rw-r--r-- | fs/afs/cmservice.c | 165 | ||||
| -rw-r--r-- | fs/afs/dir.c | 272 | ||||
| -rw-r--r-- | fs/afs/file.c | 316 | ||||
| -rw-r--r-- | fs/afs/flock.c | 585 | ||||
| -rw-r--r-- | fs/afs/fsclient.c | 1051 | ||||
| -rw-r--r-- | fs/afs/inode.c | 219 | ||||
| -rw-r--r-- | fs/afs/internal.h | 234 | ||||
| -rw-r--r-- | fs/afs/main.c | 59 | ||||
| -rw-r--r-- | fs/afs/misc.c | 18 | ||||
| -rw-r--r-- | fs/afs/mntpt.c | 189 | ||||
| -rw-r--r-- | fs/afs/netdevices.c | 8 | ||||
| -rw-r--r-- | fs/afs/proc.c | 234 | ||||
| -rw-r--r-- | fs/afs/rxrpc.c | 183 | ||||
| -rw-r--r-- | fs/afs/security.c | 29 | ||||
| -rw-r--r-- | fs/afs/server.c | 32 | ||||
| -rw-r--r-- | fs/afs/super.c | 233 | ||||
| -rw-r--r-- | fs/afs/vlclient.c | 1 | ||||
| -rw-r--r-- | fs/afs/vlocation.c | 56 | ||||
| -rw-r--r-- | fs/afs/vnode.c | 312 | ||||
| -rw-r--r-- | fs/afs/volume.c | 22 | ||||
| -rw-r--r-- | fs/afs/write.c | 763 |
31 files changed, 4551 insertions, 1219 deletions
diff --git a/fs/afs/Kconfig b/fs/afs/Kconfig new file mode 100644 index 00000000000..ebba3b18e5d --- /dev/null +++ b/fs/afs/Kconfig @@ -0,0 +1,29 @@ +config AFS_FS + tristate "Andrew File System support (AFS)" + depends on INET + select AF_RXRPC + select DNS_RESOLVER + help + If you say Y here, you will get an experimental Andrew File System + driver. It currently only supports unsecured read-only AFS access. + + See <file:Documentation/filesystems/afs.txt> for more information. + + If unsure, say N. + +config AFS_DEBUG + bool "AFS dynamic debugging" + depends on AFS_FS + help + Say Y here to make runtime controllable debugging messages appear. + + See <file:Documentation/filesystems/afs.txt> for more information. + + If unsure, say N. + +config AFS_FSCACHE + bool "Provide AFS client caching support" + depends on AFS_FS=m && FSCACHE || AFS_FS=y && FSCACHE=y + help + Say Y here if you want AFS data to be cached locally on disk through + the generic filesystem cache manager diff --git a/fs/afs/Makefile b/fs/afs/Makefile index cf83e5d6351..4f64b95d57b 100644 --- a/fs/afs/Makefile +++ b/fs/afs/Makefile @@ -2,12 +2,16 @@ # Makefile for Red Hat Linux AFS client. # +afs-cache-$(CONFIG_AFS_FSCACHE) := cache.o + kafs-objs := \ + $(afs-cache-y) \ callback.o \ cell.o \ cmservice.o \ dir.o \ file.o \ + flock.o \ fsclient.o \ inode.o \ main.o \ @@ -22,6 +26,7 @@ kafs-objs := \ vlclient.o \ vlocation.o \ vnode.o \ - volume.o + volume.o \ + write.o obj-$(CONFIG_AFS_FS) := kafs.o diff --git a/fs/afs/afs.h b/fs/afs/afs.h index 52d0752265b..3c462ff6db6 100644 --- a/fs/afs/afs.h +++ b/fs/afs/afs.h @@ -16,6 +16,9 @@ #define AFS_MAXCELLNAME 64 /* maximum length of a cell name */ #define AFS_MAXVOLNAME 64 /* maximum length of a volume name */ +#define AFSNAMEMAX 256 /* maximum length of a filename plus NUL */ +#define AFSPATHMAX 1024 /* maximum length of a pathname plus NUL */ +#define AFSOPAQUEMAX 1024 /* maximum length of an opaque field */ typedef unsigned afs_volid_t; typedef unsigned afs_vnodeid_t; @@ -34,6 +37,13 @@ typedef enum { AFS_FTYPE_SYMLINK = 3, } afs_file_type_t; +typedef enum { + AFS_LOCK_READ = 0, /* read lock request */ + AFS_LOCK_WRITE = 1, /* write lock request */ +} afs_lock_type_t; + +#define AFS_LOCKWAIT (5 * 60) /* time until a lock times out (seconds) */ + /* * AFS file identifier */ @@ -109,26 +119,20 @@ struct afs_file_status { u64 size; /* file size */ afs_dataversion_t data_version; /* current data version */ u32 author; /* author ID */ - u32 owner; /* owner ID */ - u32 group; /* group ID */ + kuid_t owner; /* owner ID */ + kgid_t group; /* group ID */ afs_access_t caller_access; /* access rights for authenticated caller */ afs_access_t anon_access; /* access rights for unauthenticated caller */ umode_t mode; /* UNIX mode */ struct afs_fid parent; /* parent dir ID for non-dirs only */ time_t mtime_client; /* last time client changed data */ time_t mtime_server; /* last time server changed data */ + s32 lock_count; /* file lock count (0=UNLK -1=WRLCK +ve=#RDLCK */ }; /* * AFS file status change request */ -struct afs_store_status { - u32 mask; /* which bits of the struct are set */ - u32 mtime_client; /* last time client changed data */ - u32 owner; /* owner ID */ - u32 group; /* group ID */ - umode_t mode; /* UNIX mode */ -}; #define AFS_SET_MTIME 0x01 /* set the mtime */ #define AFS_SET_OWNER 0x02 /* set the owner ID */ @@ -143,4 +147,24 @@ struct afs_volsync { time_t creation; /* volume creation time */ }; +/* + * AFS volume status record + */ +struct afs_volume_status { + u32 vid; /* volume ID */ + u32 parent_id; /* parent volume ID */ + u8 online; /* true if volume currently online and available */ + u8 in_service; /* true if volume currently in service */ + u8 blessed; /* same as in_service */ + u8 needs_salvage; /* true if consistency checking required */ + u32 type; /* volume type (afs_voltype_t) */ + u32 min_quota; /* minimum space set aside (blocks) */ + u32 max_quota; /* maximum space this volume may occupy (blocks) */ + u32 blocks_in_use; /* space this volume currently occupies (blocks) */ + u32 part_blocks_avail; /* space available in volume's partition */ + u32 part_max_blocks; /* size of volume's partition */ +}; + +#define AFS_BLOCK_SIZE 1024 + #endif /* AFS_H */ diff --git a/fs/afs/afs_cm.h b/fs/afs/afs_cm.h index 7b4d4fab4c8..255f5dd6040 100644 --- a/fs/afs/afs_cm.h +++ b/fs/afs/afs_cm.h @@ -24,7 +24,8 @@ enum AFS_CM_Operations { CBGetXStatsVersion = 209, /* get version of extended statistics */ CBGetXStats = 210, /* get contents of extended statistics data */ CBInitCallBackState3 = 213, /* initialise callback state, version 3 */ - CBGetCapabilities = 65538, /* get client capabilities */ + CBProbeUuid = 214, /* check the client hasn't rebooted */ + CBTellMeAboutYourself = 65538, /* get client capabilities */ }; #define AFS_CAP_ERROR_TRANSLATION 0x1 diff --git a/fs/afs/afs_fs.h b/fs/afs/afs_fs.h index 89e0d1650a7..eb647323d8f 100644 --- a/fs/afs/afs_fs.h +++ b/fs/afs/afs_fs.h @@ -18,6 +18,8 @@ enum AFS_FS_Operations { FSFETCHDATA = 130, /* AFS Fetch file data */ FSFETCHSTATUS = 132, /* AFS Fetch file status */ + FSSTOREDATA = 133, /* AFS Store file data */ + FSSTORESTATUS = 135, /* AFS Store file status */ FSREMOVEFILE = 136, /* AFS Remove a file */ FSCREATEFILE = 137, /* AFS Create a file */ FSRENAME = 138, /* AFS Rename or move a file or directory */ @@ -26,9 +28,15 @@ enum AFS_FS_Operations { FSMAKEDIR = 141, /* AFS Create a directory */ FSREMOVEDIR = 142, /* AFS Remove a directory */ FSGIVEUPCALLBACKS = 147, /* AFS Discard callback promises */ - FSGETVOLUMEINFO = 148, /* AFS Get root volume information */ + FSGETVOLUMEINFO = 148, /* AFS Get information about a volume */ + FSGETVOLUMESTATUS = 149, /* AFS Get volume status information */ FSGETROOTVOLUME = 151, /* AFS Get root volume name */ + FSSETLOCK = 156, /* AFS Request a file lock */ + FSEXTENDLOCK = 157, /* AFS Extend a file lock */ + FSRELEASELOCK = 158, /* AFS Release a file lock */ FSLOOKUP = 161, /* AFS lookup file in directory */ + FSFETCHDATA64 = 65537, /* AFS Fetch file data */ + FSSTOREDATA64 = 65538, /* AFS Store file data */ }; enum AFS_FS_Errors { diff --git a/fs/afs/afs_vl.h b/fs/afs/afs_vl.h index 8bbefe009ed..800f607ffaf 100644 --- a/fs/afs/afs_vl.h +++ b/fs/afs/afs_vl.h @@ -49,7 +49,7 @@ enum AFSVL_Errors { AFSVL_BADVOLOPER = 363542, /* Bad volume operation code */ AFSVL_BADRELLOCKTYPE = 363543, /* Bad release lock type */ AFSVL_RERELEASE = 363544, /* Status report: last release was aborted */ - AFSVL_BADSERVERFLAG = 363545, /* Invalid replication site server °ag */ + AFSVL_BADSERVERFLAG = 363545, /* Invalid replication site server flag */ AFSVL_PERM = 363546, /* No permission access */ AFSVL_NOMEM = 363547, /* malloc/realloc failed to alloc enough memory */ }; diff --git a/fs/afs/cache.c b/fs/afs/cache.c index de0d7de69ed..577763c3d88 100644 --- a/fs/afs/cache.c +++ b/fs/afs/cache.c @@ -1,6 +1,6 @@ /* AFS caching stuff * - * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2008 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -9,248 +9,394 @@ * 2 of the License, or (at your option) any later version. */ -#ifdef AFS_CACHING_SUPPORT -static cachefs_match_val_t afs_cell_cache_match(void *target, - const void *entry); -static void afs_cell_cache_update(void *source, void *entry); - -struct cachefs_index_def afs_cache_cell_index_def = { - .name = "cell_ix", - .data_size = sizeof(struct afs_cache_cell), - .keys[0] = { CACHEFS_INDEX_KEYS_ASCIIZ, 64 }, - .match = afs_cell_cache_match, - .update = afs_cell_cache_update, +#include <linux/sched.h> +#include "internal.h" + +static uint16_t afs_cell_cache_get_key(const void *cookie_netfs_data, + void *buffer, uint16_t buflen); +static uint16_t afs_cell_cache_get_aux(const void *cookie_netfs_data, + void *buffer, uint16_t buflen); +static enum fscache_checkaux afs_cell_cache_check_aux(void *cookie_netfs_data, + const void *buffer, + uint16_t buflen); + +static uint16_t afs_vlocation_cache_get_key(const void *cookie_netfs_data, + void *buffer, uint16_t buflen); +static uint16_t afs_vlocation_cache_get_aux(const void *cookie_netfs_data, + void *buffer, uint16_t buflen); +static enum fscache_checkaux afs_vlocation_cache_check_aux( + void *cookie_netfs_data, const void *buffer, uint16_t buflen); + +static uint16_t afs_volume_cache_get_key(const void *cookie_netfs_data, + void *buffer, uint16_t buflen); + +static uint16_t afs_vnode_cache_get_key(const void *cookie_netfs_data, + void *buffer, uint16_t buflen); +static void afs_vnode_cache_get_attr(const void *cookie_netfs_data, + uint64_t *size); +static uint16_t afs_vnode_cache_get_aux(const void *cookie_netfs_data, + void *buffer, uint16_t buflen); +static enum fscache_checkaux afs_vnode_cache_check_aux(void *cookie_netfs_data, + const void *buffer, + uint16_t buflen); +static void afs_vnode_cache_now_uncached(void *cookie_netfs_data); + +struct fscache_netfs afs_cache_netfs = { + .name = "afs", + .version = 0, +}; + +struct fscache_cookie_def afs_cell_cache_index_def = { + .name = "AFS.cell", + .type = FSCACHE_COOKIE_TYPE_INDEX, + .get_key = afs_cell_cache_get_key, + .get_aux = afs_cell_cache_get_aux, + .check_aux = afs_cell_cache_check_aux, +}; + +struct fscache_cookie_def afs_vlocation_cache_index_def = { + .name = "AFS.vldb", + .type = FSCACHE_COOKIE_TYPE_INDEX, + .get_key = afs_vlocation_cache_get_key, + .get_aux = afs_vlocation_cache_get_aux, + .check_aux = afs_vlocation_cache_check_aux, +}; + +struct fscache_cookie_def afs_volume_cache_index_def = { + .name = "AFS.volume", + .type = FSCACHE_COOKIE_TYPE_INDEX, + .get_key = afs_volume_cache_get_key, +}; + +struct fscache_cookie_def afs_vnode_cache_index_def = { + .name = "AFS.vnode", + .type = FSCACHE_COOKIE_TYPE_DATAFILE, + .get_key = afs_vnode_cache_get_key, + .get_attr = afs_vnode_cache_get_attr, + .get_aux = afs_vnode_cache_get_aux, + .check_aux = afs_vnode_cache_check_aux, + .now_uncached = afs_vnode_cache_now_uncached, }; -#endif /* - * match a cell record obtained from the cache + * set the key for the index entry */ -#ifdef AFS_CACHING_SUPPORT -static cachefs_match_val_t afs_cell_cache_match(void *target, - const void *entry) +static uint16_t afs_cell_cache_get_key(const void *cookie_netfs_data, + void *buffer, uint16_t bufmax) { - const struct afs_cache_cell *ccell = entry; - struct afs_cell *cell = target; + const struct afs_cell *cell = cookie_netfs_data; + uint16_t klen; - _enter("{%s},{%s}", ccell->name, cell->name); + _enter("%p,%p,%u", cell, buffer, bufmax); - if (strncmp(ccell->name, cell->name, sizeof(ccell->name)) == 0) { - _leave(" = SUCCESS"); - return CACHEFS_MATCH_SUCCESS; - } + klen = strlen(cell->name); + if (klen > bufmax) + return 0; - _leave(" = FAILED"); - return CACHEFS_MATCH_FAILED; + memcpy(buffer, cell->name, klen); + return klen; } -#endif /* - * update a cell record in the cache + * provide new auxiliary cache data */ -#ifdef AFS_CACHING_SUPPORT -static void afs_cell_cache_update(void *source, void *entry) +static uint16_t afs_cell_cache_get_aux(const void *cookie_netfs_data, + void *buffer, uint16_t bufmax) { - struct afs_cache_cell *ccell = entry; - struct afs_cell *cell = source; + const struct afs_cell *cell = cookie_netfs_data; + uint16_t dlen; - _enter("%p,%p", source, entry); + _enter("%p,%p,%u", cell, buffer, bufmax); - strncpy(ccell->name, cell->name, sizeof(ccell->name)); + dlen = cell->vl_naddrs * sizeof(cell->vl_addrs[0]); + dlen = min(dlen, bufmax); + dlen &= ~(sizeof(cell->vl_addrs[0]) - 1); - memcpy(ccell->vl_servers, - cell->vl_addrs, - min(sizeof(ccell->vl_servers), sizeof(cell->vl_addrs))); + memcpy(buffer, cell->vl_addrs, dlen); + return dlen; +} +/* + * check that the auxiliary data indicates that the entry is still valid + */ +static enum fscache_checkaux afs_cell_cache_check_aux(void *cookie_netfs_data, + const void *buffer, + uint16_t buflen) +{ + _leave(" = OKAY"); + return FSCACHE_CHECKAUX_OKAY; } -#endif - -#ifdef AFS_CACHING_SUPPORT -static cachefs_match_val_t afs_vlocation_cache_match(void *target, - const void *entry); -static void afs_vlocation_cache_update(void *source, void *entry); - -struct cachefs_index_def afs_vlocation_cache_index_def = { - .name = "vldb", - .data_size = sizeof(struct afs_cache_vlocation), - .keys[0] = { CACHEFS_INDEX_KEYS_ASCIIZ, 64 }, - .match = afs_vlocation_cache_match, - .update = afs_vlocation_cache_update, -}; -#endif +/*****************************************************************************/ /* - * match a VLDB record stored in the cache - * - may also load target from entry + * set the key for the index entry */ -#ifdef AFS_CACHING_SUPPORT -static cachefs_match_val_t afs_vlocation_cache_match(void *target, - const void *entry) +static uint16_t afs_vlocation_cache_get_key(const void *cookie_netfs_data, + void *buffer, uint16_t bufmax) { - const struct afs_cache_vlocation *vldb = entry; - struct afs_vlocation *vlocation = target; + const struct afs_vlocation *vlocation = cookie_netfs_data; + uint16_t klen; + + _enter("{%s},%p,%u", vlocation->vldb.name, buffer, bufmax); + + klen = strnlen(vlocation->vldb.name, sizeof(vlocation->vldb.name)); + if (klen > bufmax) + return 0; - _enter("{%s},{%s}", vlocation->vldb.name, vldb->name); + memcpy(buffer, vlocation->vldb.name, klen); - if (strncmp(vlocation->vldb.name, vldb->name, sizeof(vldb->name)) == 0 - ) { - if (!vlocation->valid || - vlocation->vldb.rtime == vldb->rtime + _leave(" = %u", klen); + return klen; +} + +/* + * provide new auxiliary cache data + */ +static uint16_t afs_vlocation_cache_get_aux(const void *cookie_netfs_data, + void *buffer, uint16_t bufmax) +{ + const struct afs_vlocation *vlocation = cookie_netfs_data; + uint16_t dlen; + + _enter("{%s},%p,%u", vlocation->vldb.name, buffer, bufmax); + + dlen = sizeof(struct afs_cache_vlocation); + dlen -= offsetof(struct afs_cache_vlocation, nservers); + if (dlen > bufmax) + return 0; + + memcpy(buffer, (uint8_t *)&vlocation->vldb.nservers, dlen); + + _leave(" = %u", dlen); + return dlen; +} + +/* + * check that the auxiliary data indicates that the entry is still valid + */ +static +enum fscache_checkaux afs_vlocation_cache_check_aux(void *cookie_netfs_data, + const void *buffer, + uint16_t buflen) +{ + const struct afs_cache_vlocation *cvldb; + struct afs_vlocation *vlocation = cookie_netfs_data; + uint16_t dlen; + + _enter("{%s},%p,%u", vlocation->vldb.name, buffer, buflen); + + /* check the size of the data is what we're expecting */ + dlen = sizeof(struct afs_cache_vlocation); + dlen -= offsetof(struct afs_cache_vlocation, nservers); + if (dlen != buflen) + return FSCACHE_CHECKAUX_OBSOLETE; + + cvldb = container_of(buffer, struct afs_cache_vlocation, nservers); + + /* if what's on disk is more valid than what's in memory, then use the + * VL record from the cache */ + if (!vlocation->valid || vlocation->vldb.rtime == cvldb->rtime) { + memcpy((uint8_t *)&vlocation->vldb.nservers, buffer, dlen); + vlocation->valid = 1; + _leave(" = SUCCESS [c->m]"); + return FSCACHE_CHECKAUX_OKAY; + } + + /* need to update the cache if the cached info differs */ + if (memcmp(&vlocation->vldb, buffer, dlen) != 0) { + /* delete if the volume IDs for this name differ */ + if (memcmp(&vlocation->vldb.vid, &cvldb->vid, + sizeof(cvldb->vid)) != 0 ) { - vlocation->vldb = *vldb; - vlocation->valid = 1; - _leave(" = SUCCESS [c->m]"); - return CACHEFS_MATCH_SUCCESS; - } else if (memcmp(&vlocation->vldb, vldb, sizeof(*vldb)) != 0) { - /* delete if VIDs for this name differ */ - if (memcmp(&vlocation->vldb.vid, - &vldb->vid, - sizeof(vldb->vid)) != 0) { - _leave(" = DELETE"); - return CACHEFS_MATCH_SUCCESS_DELETE; - } - - _leave(" = UPDATE"); - return CACHEFS_MATCH_SUCCESS_UPDATE; - } else { - _leave(" = SUCCESS"); - return CACHEFS_MATCH_SUCCESS; + _leave(" = OBSOLETE"); + return FSCACHE_CHECKAUX_OBSOLETE; } + + _leave(" = UPDATE"); + return FSCACHE_CHECKAUX_NEEDS_UPDATE; } - _leave(" = FAILED"); - return CACHEFS_MATCH_FAILED; + _leave(" = OKAY"); + return FSCACHE_CHECKAUX_OKAY; } -#endif +/*****************************************************************************/ /* - * update a VLDB record stored in the cache + * set the key for the volume index entry */ -#ifdef AFS_CACHING_SUPPORT -static void afs_vlocation_cache_update(void *source, void *entry) +static uint16_t afs_volume_cache_get_key(const void *cookie_netfs_data, + void *buffer, uint16_t bufmax) { - struct afs_cache_vlocation *vldb = entry; - struct afs_vlocation *vlocation = source; + const struct afs_volume *volume = cookie_netfs_data; + uint16_t klen; + + _enter("{%u},%p,%u", volume->type, buffer, bufmax); + + klen = sizeof(volume->type); + if (klen > bufmax) + return 0; - _enter(""); + memcpy(buffer, &volume->type, sizeof(volume->type)); + + _leave(" = %u", klen); + return klen; - *vldb = vlocation->vldb; } -#endif - -#ifdef AFS_CACHING_SUPPORT -static cachefs_match_val_t afs_volume_cache_match(void *target, - const void *entry); -static void afs_volume_cache_update(void *source, void *entry); - -struct cachefs_index_def afs_volume_cache_index_def = { - .name = "volume", - .data_size = sizeof(struct afs_cache_vhash), - .keys[0] = { CACHEFS_INDEX_KEYS_BIN, 1 }, - .keys[1] = { CACHEFS_INDEX_KEYS_BIN, 1 }, - .match = afs_volume_cache_match, - .update = afs_volume_cache_update, -}; -#endif +/*****************************************************************************/ /* - * match a volume hash record stored in the cache + * set the key for the index entry */ -#ifdef AFS_CACHING_SUPPORT -static cachefs_match_val_t afs_volume_cache_match(void *target, - const void *entry) +static uint16_t afs_vnode_cache_get_key(const void *cookie_netfs_data, + void *buffer, uint16_t bufmax) { - const struct afs_cache_vhash *vhash = entry; - struct afs_volume *volume = target; + const struct afs_vnode *vnode = cookie_netfs_data; + uint16_t klen; - _enter("{%u},{%u}", volume->type, vhash->vtype); + _enter("{%x,%x,%llx},%p,%u", + vnode->fid.vnode, vnode->fid.unique, vnode->status.data_version, + buffer, bufmax); - if (volume->type == vhash->vtype) { - _leave(" = SUCCESS"); - return CACHEFS_MATCH_SUCCESS; - } + klen = sizeof(vnode->fid.vnode); + if (klen > bufmax) + return 0; + + memcpy(buffer, &vnode->fid.vnode, sizeof(vnode->fid.vnode)); - _leave(" = FAILED"); - return CACHEFS_MATCH_FAILED; + _leave(" = %u", klen); + return klen; } -#endif /* - * update a volume hash record stored in the cache + * provide updated file attributes */ -#ifdef AFS_CACHING_SUPPORT -static void afs_volume_cache_update(void *source, void *entry) +static void afs_vnode_cache_get_attr(const void *cookie_netfs_data, + uint64_t *size) { - struct afs_cache_vhash *vhash = entry; - struct afs_volume *volume = source; + const struct afs_vnode *vnode = cookie_netfs_data; - _enter(""); + _enter("{%x,%x,%llx},", + vnode->fid.vnode, vnode->fid.unique, + vnode->status.data_version); - vhash->vtype = volume->type; + *size = vnode->status.size; } -#endif - -#ifdef AFS_CACHING_SUPPORT -static cachefs_match_val_t afs_vnode_cache_match(void *target, - const void *entry); -static void afs_vnode_cache_update(void *source, void *entry); - -struct cachefs_index_def afs_vnode_cache_index_def = { - .name = "vnode", - .data_size = sizeof(struct afs_cache_vnode), - .keys[0] = { CACHEFS_INDEX_KEYS_BIN, 4 }, - .match = afs_vnode_cache_match, - .update = afs_vnode_cache_update, -}; -#endif /* - * match a vnode record stored in the cache + * provide new auxiliary cache data + */ +static uint16_t afs_vnode_cache_get_aux(const void *cookie_netfs_data, + void *buffer, uint16_t bufmax) +{ + const struct afs_vnode *vnode = cookie_netfs_data; + uint16_t dlen; + + _enter("{%x,%x,%Lx},%p,%u", + vnode->fid.vnode, vnode->fid.unique, vnode->status.data_version, + buffer, bufmax); + + dlen = sizeof(vnode->fid.unique) + sizeof(vnode->status.data_version); + if (dlen > bufmax) + return 0; + + memcpy(buffer, &vnode->fid.unique, sizeof(vnode->fid.unique)); + buffer += sizeof(vnode->fid.unique); + memcpy(buffer, &vnode->status.data_version, + sizeof(vnode->status.data_version)); + + _leave(" = %u", dlen); + return dlen; +} + +/* + * check that the auxiliary data indicates that the entry is still valid */ -#ifdef AFS_CACHING_SUPPORT -static cachefs_match_val_t afs_vnode_cache_match(void *target, - const void *entry) +static enum fscache_checkaux afs_vnode_cache_check_aux(void *cookie_netfs_data, + const void *buffer, + uint16_t buflen) { - const struct afs_cache_vnode *cvnode = entry; - struct afs_vnode *vnode = target; - - _enter("{%x,%x,%Lx},{%x,%x,%Lx}", - vnode->fid.vnode, - vnode->fid.unique, - vnode->status.version, - cvnode->vnode_id, - cvnode->vnode_unique, - cvnode->data_version); - - if (vnode->fid.vnode != cvnode->vnode_id) { - _leave(" = FAILED"); - return CACHEFS_MATCH_FAILED; + struct afs_vnode *vnode = cookie_netfs_data; + uint16_t dlen; + + _enter("{%x,%x,%llx},%p,%u", + vnode->fid.vnode, vnode->fid.unique, vnode->status.data_version, + buffer, buflen); + + /* check the size of the data is what we're expecting */ + dlen = sizeof(vnode->fid.unique) + sizeof(vnode->status.data_version); + if (dlen != buflen) { + _leave(" = OBSOLETE [len %hx != %hx]", dlen, buflen); + return FSCACHE_CHECKAUX_OBSOLETE; } - if (vnode->fid.unique != cvnode->vnode_unique || - vnode->status.version != cvnode->data_version) { - _leave(" = DELETE"); - return CACHEFS_MATCH_SUCCESS_DELETE; + if (memcmp(buffer, + &vnode->fid.unique, + sizeof(vnode->fid.unique) + ) != 0) { + unsigned unique; + + memcpy(&unique, buffer, sizeof(unique)); + + _leave(" = OBSOLETE [uniq %x != %x]", + unique, vnode->fid.unique); + return FSCACHE_CHECKAUX_OBSOLETE; + } + + if (memcmp(buffer + sizeof(vnode->fid.unique), + &vnode->status.data_version, + sizeof(vnode->status.data_version) + ) != 0) { + afs_dataversion_t version; + + memcpy(&version, buffer + sizeof(vnode->fid.unique), + sizeof(version)); + + _leave(" = OBSOLETE [vers %llx != %llx]", + version, vnode->status.data_version); + return FSCACHE_CHECKAUX_OBSOLETE; } _leave(" = SUCCESS"); - return CACHEFS_MATCH_SUCCESS; + return FSCACHE_CHECKAUX_OKAY; } -#endif /* - * update a vnode record stored in the cache + * indication the cookie is no longer uncached + * - this function is called when the backing store currently caching a cookie + * is removed + * - the netfs should use this to clean up any markers indicating cached pages + * - this is mandatory for any object that may have data */ -#ifdef AFS_CACHING_SUPPORT -static void afs_vnode_cache_update(void *source, void *entry) +static void afs_vnode_cache_now_uncached(void *cookie_netfs_data) { - struct afs_cache_vnode *cvnode = entry; - struct afs_vnode *vnode = source; + struct afs_vnode *vnode = cookie_netfs_data; + struct pagevec pvec; + pgoff_t first; + int loop, nr_pages; + + _enter("{%x,%x,%Lx}", + vnode->fid.vnode, vnode->fid.unique, vnode->status.data_version); + + pagevec_init(&pvec, 0); + first = 0; + + for (;;) { + /* grab a bunch of pages to clean */ + nr_pages = pagevec_lookup(&pvec, vnode->vfs_inode.i_mapping, + first, + PAGEVEC_SIZE - pagevec_count(&pvec)); + if (!nr_pages) + break; - _enter(""); + for (loop = 0; loop < nr_pages; loop++) + ClearPageFsCache(pvec.pages[loop]); + + first = pvec.pages[nr_pages - 1]->index + 1; + + pvec.nr = nr_pages; + pagevec_release(&pvec); + cond_resched(); + } - cvnode->vnode_id = vnode->fid.vnode; - cvnode->vnode_unique = vnode->fid.unique; - cvnode->data_version = vnode->status.version; + _leave(""); } -#endif diff --git a/fs/afs/cache.h b/fs/afs/cache.h deleted file mode 100644 index 36a3642cf90..00000000000 --- a/fs/afs/cache.h +++ /dev/null @@ -1,23 +0,0 @@ -/* AFS local cache management interface - * - * Copyright (C) 2002 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - */ - -#ifndef AFS_CACHE_H -#define AFS_CACHE_H - -#undef AFS_CACHING_SUPPORT - -#include <linux/mm.h> -#ifdef AFS_CACHING_SUPPORT -#include <linux/cachefs.h> -#endif -#include "types.h" - -#endif /* AFS_CACHE_H */ diff --git a/fs/afs/callback.c b/fs/afs/callback.c index 9bdbf36a9aa..7ef637d7f3a 100644 --- a/fs/afs/callback.c +++ b/fs/afs/callback.c @@ -8,7 +8,7 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. * - * Authors: David Woodhouse <dwmw2@cambridge.redhat.com> + * Authors: David Woodhouse <dwmw2@infradead.org> * David Howells <dhowells@redhat.com> * */ @@ -17,9 +17,12 @@ #include <linux/module.h> #include <linux/init.h> #include <linux/circ_buf.h> +#include <linux/sched.h> #include "internal.h" +#if 0 unsigned afs_vnode_update_timeout = 10; +#endif /* 0 */ #define afs_breakring_space(server) \ CIRC_SPACE((server)->cb_break_head, (server)->cb_break_tail, \ @@ -44,7 +47,7 @@ void afs_init_callback_state(struct afs_server *server) while (!RB_EMPTY_ROOT(&server->cb_promises)) { vnode = rb_entry(server->cb_promises.rb_node, struct afs_vnode, cb_promise); - _debug("UNPROMISE { vid=%x vn=%u uq=%u}", + _debug("UNPROMISE { vid=%x:%u uq=%u}", vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique); rb_erase(&vnode->cb_promise, &server->cb_promises); vnode->cb_promised = false; @@ -84,11 +87,8 @@ void afs_broken_callback_work(struct work_struct *work) /* if the vnode's data version number changed then its contents * are different */ - if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) { - _debug("zap data {%x:%u}", - vnode->fid.vid, vnode->fid.vnode); - invalidate_remote_inode(&vnode->vfs_inode); - } + if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) + afs_zap_data(vnode); } out: @@ -127,6 +127,9 @@ static void afs_break_callback(struct afs_server *server, spin_unlock(&server->cb_lock); queue_work(afs_callback_update_worker, &vnode->cb_broken_work); + if (list_empty(&vnode->granted_locks) && + !list_empty(&vnode->pending_locks)) + afs_lock_may_be_available(vnode); spin_unlock(&vnode->lock); } } @@ -348,9 +351,7 @@ void afs_dispatch_give_up_callbacks(struct work_struct *work) */ void afs_flush_callback_breaks(struct afs_server *server) { - cancel_delayed_work(&server->cb_break_work); - queue_delayed_work(afs_callback_update_worker, - &server->cb_break_work, 0); + mod_delayed_work(afs_callback_update_worker, &server->cb_break_work, 0); } #if 0 diff --git a/fs/afs/cell.c b/fs/afs/cell.c index 9b1311a1df5..ca0a3cf9379 100644 --- a/fs/afs/cell.c +++ b/fs/afs/cell.c @@ -13,13 +13,15 @@ #include <linux/slab.h> #include <linux/key.h> #include <linux/ctype.h> +#include <linux/dns_resolver.h> +#include <linux/sched.h> #include <keys/rxrpc-type.h> #include "internal.h" DECLARE_RWSEM(afs_proc_cells_sem); LIST_HEAD(afs_proc_cells); -static struct list_head afs_cells = LIST_HEAD_INIT(afs_cells); +static LIST_HEAD(afs_cells); static DEFINE_RWLOCK(afs_cells_lock); static DECLARE_RWSEM(afs_cells_sem); /* add/remove serialisation */ static DECLARE_WAIT_QUEUE_HEAD(afs_cells_freeable_wq); @@ -29,20 +31,24 @@ static struct afs_cell *afs_cell_root; * allocate a cell record and fill in its name, VL server address list and * allocate an anonymous key */ -static struct afs_cell *afs_cell_alloc(const char *name, char *vllist) +static struct afs_cell *afs_cell_alloc(const char *name, unsigned namelen, + char *vllist) { struct afs_cell *cell; - size_t namelen; + struct key *key; char keyname[4 + AFS_MAXCELLNAME + 1], *cp, *dp, *next; + char *dvllist = NULL, *_vllist = NULL; + char delimiter = ':'; int ret; - _enter("%s,%s", name, vllist); + _enter("%*.*s,%s", namelen, namelen, name ?: "", vllist); BUG_ON(!name); /* TODO: want to look up "this cell" in the cache */ - namelen = strlen(name); - if (namelen > AFS_MAXCELLNAME) + if (namelen > AFS_MAXCELLNAME) { + _leave(" = -ENAMETOOLONG"); return ERR_PTR(-ENAMETOOLONG); + } /* allocate and initialise a cell record */ cell = kzalloc(sizeof(struct afs_cell) + namelen + 1, GFP_KERNEL); @@ -62,15 +68,35 @@ static struct afs_cell *afs_cell_alloc(const char *name, char *vllist) INIT_LIST_HEAD(&cell->vl_list); spin_lock_init(&cell->vl_lock); + /* if the ip address is invalid, try dns query */ + if (!vllist || strlen(vllist) < 7) { + ret = dns_query("afsdb", name, namelen, "ipv4", &dvllist, NULL); + if (ret < 0) { + if (ret == -ENODATA || ret == -EAGAIN || ret == -ENOKEY) + /* translate these errors into something + * userspace might understand */ + ret = -EDESTADDRREQ; + _leave(" = %d", ret); + return ERR_PTR(ret); + } + _vllist = dvllist; + + /* change the delimiter for user-space reply */ + delimiter = ','; + + } else { + _vllist = vllist; + } + /* fill in the VL server list from the rest of the string */ do { unsigned a, b, c, d; - next = strchr(vllist, ':'); + next = strchr(_vllist, delimiter); if (next) *next++ = 0; - if (sscanf(vllist, "%u.%u.%u.%u", &a, &b, &c, &d) != 4) + if (sscanf(_vllist, "%u.%u.%u.%u", &a, &b, &c, &d) != 4) goto bad_address; if (a > 255 || b > 255 || c > 255 || d > 255) @@ -79,7 +105,7 @@ static struct afs_cell *afs_cell_alloc(const char *name, char *vllist) cell->vl_addrs[cell->vl_naddrs++].s_addr = htonl((a << 24) | (b << 16) | (c << 8) | d); - } while (cell->vl_naddrs < AFS_CELL_MAX_ADDRS && (vllist = next)); + } while (cell->vl_naddrs < AFS_CELL_MAX_ADDRS && (_vllist = next)); /* create a key to represent an anonymous user */ memcpy(keyname, "afs@", 4); @@ -88,20 +114,14 @@ static struct afs_cell *afs_cell_alloc(const char *name, char *vllist) do { *dp++ = toupper(*cp); } while (*cp++); - cell->anonymous_key = key_alloc(&key_type_rxrpc, keyname, 0, 0, current, - KEY_POS_SEARCH, KEY_ALLOC_NOT_IN_QUOTA); - if (IS_ERR(cell->anonymous_key)) { - _debug("no key"); - ret = PTR_ERR(cell->anonymous_key); - goto error; - } - ret = key_instantiate_and_link(cell->anonymous_key, NULL, 0, - NULL, NULL); - if (ret < 0) { - _debug("instantiate failed"); + key = rxrpc_get_null_key(keyname); + if (IS_ERR(key)) { + _debug("no key"); + ret = PTR_ERR(key); goto error; } + cell->anonymous_key = key; _debug("anon key %p{%x}", cell->anonymous_key, key_serial(cell->anonymous_key)); @@ -114,42 +134,52 @@ bad_address: ret = -EINVAL; error: key_put(cell->anonymous_key); + kfree(dvllist); kfree(cell); _leave(" = %d", ret); return ERR_PTR(ret); } /* - * create a cell record - * - "name" is the name of the cell - * - "vllist" is a colon separated list of IP addresses in "a.b.c.d" format + * afs_cell_crate() - create a cell record + * @name: is the name of the cell. + * @namsesz: is the strlen of the cell name. + * @vllist: is a colon separated list of IP addresses in "a.b.c.d" format. + * @retref: is T to return the cell reference when the cell exists. */ -struct afs_cell *afs_cell_create(const char *name, char *vllist) +struct afs_cell *afs_cell_create(const char *name, unsigned namesz, + char *vllist, bool retref) { struct afs_cell *cell; int ret; - _enter("%s,%s", name, vllist); + _enter("%*.*s,%s", namesz, namesz, name ?: "", vllist); - cell = afs_cell_alloc(name, vllist); + down_write(&afs_cells_sem); + read_lock(&afs_cells_lock); + list_for_each_entry(cell, &afs_cells, link) { + if (strncasecmp(cell->name, name, namesz) == 0) + goto duplicate_name; + } + read_unlock(&afs_cells_lock); + + cell = afs_cell_alloc(name, namesz, vllist); if (IS_ERR(cell)) { _leave(" = %ld", PTR_ERR(cell)); + up_write(&afs_cells_sem); return cell; } - down_write(&afs_cells_sem); - /* add a proc directory for this cell */ ret = afs_proc_cell_setup(cell); if (ret < 0) goto error; -#ifdef AFS_CACHING_SUPPORT - /* put it up for caching */ - cachefs_acquire_cookie(afs_cache_netfs.primary_index, - &afs_vlocation_cache_index_def, - cell, - &cell->cache); +#ifdef CONFIG_AFS_FSCACHE + /* put it up for caching (this never returns an error) */ + cell->cache = fscache_acquire_cookie(afs_cache_netfs.primary_index, + &afs_cell_cache_index_def, + cell, true); #endif /* add to the cell lists */ @@ -171,6 +201,21 @@ error: kfree(cell); _leave(" = %d", ret); return ERR_PTR(ret); + +duplicate_name: + if (retref && !IS_ERR(cell)) + afs_get_cell(cell); + + read_unlock(&afs_cells_lock); + up_write(&afs_cells_sem); + + if (retref) { + _leave(" = %p", cell); + return cell; + } + + _leave(" = -EEXIST"); + return ERR_PTR(-EEXIST); } /* @@ -194,15 +239,13 @@ int afs_cell_init(char *rootcell) } cp = strchr(rootcell, ':'); - if (!cp) { - printk(KERN_ERR "kAFS: no VL server IP addresses specified\n"); - _leave(" = -EINVAL"); - return -EINVAL; - } + if (!cp) + _debug("kAFS: no VL server IP addresses specified"); + else + *cp++ = 0; /* allocate a cell record for the root cell */ - *cp++ = 0; - new_root = afs_cell_create(rootcell, cp); + new_root = afs_cell_create(rootcell, strlen(rootcell), cp, false); if (IS_ERR(new_root)) { _leave(" = %ld", PTR_ERR(new_root)); return PTR_ERR(new_root); @@ -222,11 +265,12 @@ int afs_cell_init(char *rootcell) /* * lookup a cell record */ -struct afs_cell *afs_cell_lookup(const char *name, unsigned namesz) +struct afs_cell *afs_cell_lookup(const char *name, unsigned namesz, + bool dns_cell) { struct afs_cell *cell; - _enter("\"%*.*s\",", namesz, namesz, name ? name : ""); + _enter("\"%*.*s\",", namesz, namesz, name ?: ""); down_read(&afs_cells_sem); read_lock(&afs_cells_lock); @@ -240,6 +284,8 @@ struct afs_cell *afs_cell_lookup(const char *name, unsigned namesz) } } cell = ERR_PTR(-ENOENT); + if (dns_cell) + goto create_cell; found: ; } else { @@ -247,7 +293,7 @@ struct afs_cell *afs_cell_lookup(const char *name, unsigned namesz) if (!cell) { /* this should not happen unless user tries to mount * when root cell is not set. Return an impossibly - * bizzare errno to alert the user. Things like + * bizarre errno to alert the user. Things like * ENOENT might be "more appropriate" but they happen * for other reasons. */ @@ -262,8 +308,18 @@ struct afs_cell *afs_cell_lookup(const char *name, unsigned namesz) up_read(&afs_cells_sem); _leave(" = %p", cell); return cell; + +create_cell: + read_unlock(&afs_cells_lock); + up_read(&afs_cells_sem); + + cell = afs_cell_create(name, namesz, NULL, true); + + _leave(" = %p", cell); + return cell; } +#if 0 /* * try and get a cell record */ @@ -279,6 +335,7 @@ struct afs_cell *afs_get_cell_maybe(struct afs_cell *cell) write_unlock(&afs_cells_lock); return cell; } +#endif /* 0 */ /* * destroy a cell record @@ -352,10 +409,9 @@ static void afs_cell_destroy(struct afs_cell *cell) list_del_init(&cell->proc_link); up_write(&afs_proc_cells_sem); -#ifdef AFS_CACHING_SUPPORT - cachefs_relinquish_cookie(cell->cache, 0); +#ifdef CONFIG_AFS_FSCACHE + fscache_relinquish_cookie(cell->cache, 0); #endif - key_put(cell->anonymous_key); kfree(cell); diff --git a/fs/afs/cmservice.c b/fs/afs/cmservice.c index d5b2ad6575b..4b0eff6da67 100644 --- a/fs/afs/cmservice.c +++ b/fs/afs/cmservice.c @@ -11,12 +11,15 @@ #include <linux/module.h> #include <linux/init.h> +#include <linux/slab.h> #include <linux/sched.h> #include <linux/ip.h> #include "internal.h" #include "afs_cm.h" +#if 0 struct workqueue_struct *afs_cm_workqueue; +#endif /* 0 */ static int afs_deliver_cb_init_call_back_state(struct afs_call *, struct sk_buff *, bool); @@ -24,8 +27,9 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *, struct sk_buff *, bool); static int afs_deliver_cb_probe(struct afs_call *, struct sk_buff *, bool); static int afs_deliver_cb_callback(struct afs_call *, struct sk_buff *, bool); -static int afs_deliver_cb_get_capabilities(struct afs_call *, struct sk_buff *, - bool); +static int afs_deliver_cb_probe_uuid(struct afs_call *, struct sk_buff *, bool); +static int afs_deliver_cb_tell_me_about_yourself(struct afs_call *, + struct sk_buff *, bool); static void afs_cm_destructor(struct afs_call *); /* @@ -69,11 +73,21 @@ static const struct afs_call_type afs_SRXCBProbe = { }; /* - * CB.GetCapabilities operation type + * CB.ProbeUuid operation type */ -static const struct afs_call_type afs_SRXCBGetCapabilites = { - .name = "CB.GetCapabilities", - .deliver = afs_deliver_cb_get_capabilities, +static const struct afs_call_type afs_SRXCBProbeUuid = { + .name = "CB.ProbeUuid", + .deliver = afs_deliver_cb_probe_uuid, + .abort_to_error = afs_abort_to_error, + .destructor = afs_cm_destructor, +}; + +/* + * CB.TellMeAboutYourself operation type + */ +static const struct afs_call_type afs_SRXCBTellMeAboutYourself = { + .name = "CB.TellMeAboutYourself", + .deliver = afs_deliver_cb_tell_me_about_yourself, .abort_to_error = afs_abort_to_error, .destructor = afs_cm_destructor, }; @@ -101,8 +115,8 @@ bool afs_cm_incoming_call(struct afs_call *call) case CBProbe: call->type = &afs_SRXCBProbe; return true; - case CBGetCapabilities: - call->type = &afs_SRXCBGetCapabilites; + case CBTellMeAboutYourself: + call->type = &afs_SRXCBTellMeAboutYourself; return true; default: return false; @@ -116,6 +130,15 @@ static void afs_cm_destructor(struct afs_call *call) { _enter(""); + /* Break the callbacks here so that we do it after the final ACK is + * received. The step number here must match the final number in + * afs_deliver_cb_callback(). + */ + if (call->unmarshall == 6) { + ASSERT(call->server && call->count && call->request); + afs_break_callbacks(call->server, call->count, call->request); + } + afs_put_server(call->server); call->server = NULL; kfree(call->buffer); @@ -258,6 +281,16 @@ static int afs_deliver_cb_callback(struct afs_call *call, struct sk_buff *skb, _debug("trailer"); if (skb->len != 0) return -EBADMSG; + + /* Record that the message was unmarshalled successfully so + * that the call destructor can know do the callback breaking + * work, even if the final ACK isn't received. + * + * If the step number changes, then afs_cm_destructor() must be + * updated also. + */ + call->unmarshall++; + case 6: break; } @@ -275,7 +308,7 @@ static int afs_deliver_cb_callback(struct afs_call *call, struct sk_buff *skb, call->server = server; INIT_WORK(&call->work, SRXAFSCB_CallBack); - schedule_work(&call->work); + queue_work(afs_wq, &call->work); return 0; } @@ -322,7 +355,7 @@ static int afs_deliver_cb_init_call_back_state(struct afs_call *call, call->server = server; INIT_WORK(&call->work, SRXAFSCB_InitCallBackState); - schedule_work(&call->work); + queue_work(afs_wq, &call->work); return 0; } @@ -353,7 +386,7 @@ static int afs_deliver_cb_init_call_back_state3(struct afs_call *call, call->server = server; INIT_WORK(&call->work, SRXAFSCB_InitCallBackState); - schedule_work(&call->work); + queue_work(afs_wq, &call->work); return 0; } @@ -386,14 +419,110 @@ static int afs_deliver_cb_probe(struct afs_call *call, struct sk_buff *skb, call->state = AFS_CALL_REPLYING; INIT_WORK(&call->work, SRXAFSCB_Probe); - schedule_work(&call->work); + queue_work(afs_wq, &call->work); + return 0; +} + +/* + * allow the fileserver to quickly find out if the fileserver has been rebooted + */ +static void SRXAFSCB_ProbeUuid(struct work_struct *work) +{ + struct afs_call *call = container_of(work, struct afs_call, work); + struct afs_uuid *r = call->request; + + struct { + __be32 match; + } reply; + + _enter(""); + + + if (memcmp(r, &afs_uuid, sizeof(afs_uuid)) == 0) + reply.match = htonl(0); + else + reply.match = htonl(1); + + afs_send_simple_reply(call, &reply, sizeof(reply)); + _leave(""); +} + +/* + * deliver request data to a CB.ProbeUuid call + */ +static int afs_deliver_cb_probe_uuid(struct afs_call *call, struct sk_buff *skb, + bool last) +{ + struct afs_uuid *r; + unsigned loop; + __be32 *b; + int ret; + + _enter("{%u},{%u},%d", call->unmarshall, skb->len, last); + + if (skb->len > 0) + return -EBADMSG; + if (!last) + return 0; + + switch (call->unmarshall) { + case 0: + call->offset = 0; + call->buffer = kmalloc(11 * sizeof(__be32), GFP_KERNEL); + if (!call->buffer) + return -ENOMEM; + call->unmarshall++; + + case 1: + _debug("extract UUID"); + ret = afs_extract_data(call, skb, last, call->buffer, + 11 * sizeof(__be32)); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + + _debug("unmarshall UUID"); + call->request = kmalloc(sizeof(struct afs_uuid), GFP_KERNEL); + if (!call->request) + return -ENOMEM; + + b = call->buffer; + r = call->request; + r->time_low = ntohl(b[0]); + r->time_mid = ntohl(b[1]); + r->time_hi_and_version = ntohl(b[2]); + r->clock_seq_hi_and_reserved = ntohl(b[3]); + r->clock_seq_low = ntohl(b[4]); + + for (loop = 0; loop < 6; loop++) + r->node[loop] = ntohl(b[loop + 5]); + + call->offset = 0; + call->unmarshall++; + + case 2: + _debug("trailer"); + if (skb->len != 0) + return -EBADMSG; + break; + } + + if (!last) + return 0; + + call->state = AFS_CALL_REPLYING; + + INIT_WORK(&call->work, SRXAFSCB_ProbeUuid); + queue_work(afs_wq, &call->work); return 0; } /* * allow the fileserver to ask about the cache manager's capabilities */ -static void SRXAFSCB_GetCapabilities(struct work_struct *work) +static void SRXAFSCB_TellMeAboutYourself(struct work_struct *work) { struct afs_interface *ifs; struct afs_call *call = container_of(work, struct afs_call, work); @@ -454,10 +583,10 @@ static void SRXAFSCB_GetCapabilities(struct work_struct *work) } /* - * deliver request data to a CB.GetCapabilities call + * deliver request data to a CB.TellMeAboutYourself call */ -static int afs_deliver_cb_get_capabilities(struct afs_call *call, - struct sk_buff *skb, bool last) +static int afs_deliver_cb_tell_me_about_yourself(struct afs_call *call, + struct sk_buff *skb, bool last) { _enter(",{%u},%d", skb->len, last); @@ -469,7 +598,7 @@ static int afs_deliver_cb_get_capabilities(struct afs_call *call, /* no unmarshalling required */ call->state = AFS_CALL_REPLYING; - INIT_WORK(&call->work, SRXAFSCB_GetCapabilities); - schedule_work(&call->work); + INIT_WORK(&call->work, SRXAFSCB_TellMeAboutYourself); + queue_work(afs_wq, &call->work); return 0; } diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 0c1e902f17a..529300327f4 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -12,24 +12,25 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> -#include <linux/slab.h> #include <linux/fs.h> +#include <linux/namei.h> #include <linux/pagemap.h> #include <linux/ctype.h> +#include <linux/sched.h> #include "internal.h" static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd); + unsigned int flags); static int afs_dir_open(struct inode *inode, struct file *file); -static int afs_readdir(struct file *file, void *dirent, filldir_t filldir); -static int afs_d_revalidate(struct dentry *dentry, struct nameidata *nd); -static int afs_d_delete(struct dentry *dentry); +static int afs_readdir(struct file *file, struct dir_context *ctx); +static int afs_d_revalidate(struct dentry *dentry, unsigned int flags); +static int afs_d_delete(const struct dentry *dentry); static void afs_d_release(struct dentry *dentry); static int afs_lookup_filldir(void *_cookie, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype); -static int afs_create(struct inode *dir, struct dentry *dentry, int mode, - struct nameidata *nd); -static int afs_mkdir(struct inode *dir, struct dentry *dentry, int mode); +static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, + bool excl); +static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); static int afs_rmdir(struct inode *dir, struct dentry *dentry); static int afs_unlink(struct inode *dir, struct dentry *dentry); static int afs_link(struct dentry *from, struct inode *dir, @@ -42,7 +43,9 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, const struct file_operations afs_dir_file_operations = { .open = afs_dir_open, .release = afs_release, - .readdir = afs_readdir, + .iterate = afs_readdir, + .lock = afs_lock, + .llseek = generic_file_llseek, }; const struct inode_operations afs_dir_inode_operations = { @@ -55,13 +58,15 @@ const struct inode_operations afs_dir_inode_operations = { .rmdir = afs_rmdir, .rename = afs_rename, .permission = afs_permission, - .getattr = afs_inode_getattr, + .getattr = afs_getattr, + .setattr = afs_setattr, }; -static struct dentry_operations afs_fs_dentry_operations = { +const struct dentry_operations afs_fs_dentry_operations = { .d_revalidate = afs_d_revalidate, .d_delete = afs_d_delete, .d_release = afs_d_release, + .d_automount = afs_d_automount, }; #define AFS_DIR_HASHTBL_SIZE 128 @@ -114,9 +119,9 @@ struct afs_dir_page { }; struct afs_lookup_cookie { + struct dir_context ctx; struct afs_fid fid; - const char *name; - size_t nlen; + struct qstr name; int found; }; @@ -137,7 +142,7 @@ static inline void afs_dir_check_page(struct inode *dir, struct page *page) if (page->index == 0 && qty != ntohs(dbuf->blocks[0].pagehdr.npages)) { printk("kAFS: %s(%lu): wrong number of dir blocks %d!=%hu\n", - __FUNCTION__, dir->i_ino, qty, + __func__, dir->i_ino, qty, ntohs(dbuf->blocks[0].pagehdr.npages)); goto error; } @@ -156,7 +161,7 @@ static inline void afs_dir_check_page(struct inode *dir, struct page *page) for (tmp = 0; tmp < qty; tmp++) { if (dbuf->blocks[tmp].pagehdr.magic != AFS_DIR_MAGIC) { printk("kAFS: %s(%lu): bad magic %d/%d is %04hx\n", - __FUNCTION__, dir->i_ino, tmp, qty, + __func__, dir->i_ino, tmp, qty, ntohs(dbuf->blocks[tmp].pagehdr.magic)); goto error; } @@ -186,13 +191,9 @@ static struct page *afs_dir_get_page(struct inode *dir, unsigned long index, struct key *key) { struct page *page; - struct file file = { - .private_data = key, - }; - _enter("{%lu},%lu", dir->i_ino, index); - page = read_mapping_page(dir->i_mapping, index, &file); + page = read_cache_page(dir->i_mapping, index, afs_page_filler, key); if (!IS_ERR(page)) { kmap(page); if (!PageChecked(page)) @@ -227,20 +228,18 @@ static int afs_dir_open(struct inode *inode, struct file *file) /* * deal with one block in an AFS directory */ -static int afs_dir_iterate_block(unsigned *fpos, +static int afs_dir_iterate_block(struct dir_context *ctx, union afs_dir_block *block, - unsigned blkoff, - void *cookie, - filldir_t filldir) + unsigned blkoff) { union afs_dirent *dire; unsigned offset, next, curr; size_t nlen; - int tmp, ret; + int tmp; - _enter("%u,%x,%p,,",*fpos,blkoff,block); + _enter("%u,%x,%p,,",(unsigned)ctx->pos,blkoff,block); - curr = (*fpos - blkoff) / sizeof(union afs_dirent); + curr = (ctx->pos - blkoff) / sizeof(union afs_dirent); /* walk through the block, an entry at a time */ for (offset = AFS_DIRENT_PER_BLOCK - block->pagehdr.nentries; @@ -255,7 +254,7 @@ static int afs_dir_iterate_block(unsigned *fpos, _debug("ENT[%Zu.%u]: unused", blkoff / sizeof(union afs_dir_block), offset); if (offset >= curr) - *fpos = blkoff + + ctx->pos = blkoff + next * sizeof(union afs_dirent); continue; } @@ -301,19 +300,15 @@ static int afs_dir_iterate_block(unsigned *fpos, continue; /* found the next entry */ - ret = filldir(cookie, - dire->u.name, - nlen, - blkoff + offset * sizeof(union afs_dirent), + if (!dir_emit(ctx, dire->u.name, nlen, ntohl(dire->u.vnode), - filldir == afs_lookup_filldir ? - ntohl(dire->u.unique) : DT_UNKNOWN); - if (ret < 0) { + ctx->actor == afs_lookup_filldir ? + ntohl(dire->u.unique) : DT_UNKNOWN)) { _leave(" = 0 [full]"); return 0; } - *fpos = blkoff + next * sizeof(union afs_dirent); + ctx->pos = blkoff + next * sizeof(union afs_dirent); } _leave(" = 1 [more]"); @@ -323,8 +318,8 @@ static int afs_dir_iterate_block(unsigned *fpos, /* * iterate through the data blob that lists the contents of an AFS directory */ -static int afs_dir_iterate(struct inode *dir, unsigned *fpos, void *cookie, - filldir_t filldir, struct key *key) +static int afs_dir_iterate(struct inode *dir, struct dir_context *ctx, + struct key *key) { union afs_dir_block *dblock; struct afs_dir_page *dbuf; @@ -332,7 +327,7 @@ static int afs_dir_iterate(struct inode *dir, unsigned *fpos, void *cookie, unsigned blkoff, limit; int ret; - _enter("{%lu},%u,,", dir->i_ino, *fpos); + _enter("{%lu},%u,,", dir->i_ino, (unsigned)ctx->pos); if (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(dir)->flags)) { _leave(" = -ESTALE"); @@ -340,13 +335,13 @@ static int afs_dir_iterate(struct inode *dir, unsigned *fpos, void *cookie, } /* round the file position up to the next entry boundary */ - *fpos += sizeof(union afs_dirent) - 1; - *fpos &= ~(sizeof(union afs_dirent) - 1); + ctx->pos += sizeof(union afs_dirent) - 1; + ctx->pos &= ~(sizeof(union afs_dirent) - 1); /* walk through the blocks in sequence */ ret = 0; - while (*fpos < dir->i_size) { - blkoff = *fpos & ~(sizeof(union afs_dir_block) - 1); + while (ctx->pos < dir->i_size) { + blkoff = ctx->pos & ~(sizeof(union afs_dir_block) - 1); /* fetch the appropriate page from the directory */ page = afs_dir_get_page(dir, blkoff / PAGE_SIZE, key); @@ -363,8 +358,7 @@ static int afs_dir_iterate(struct inode *dir, unsigned *fpos, void *cookie, do { dblock = &dbuf->blocks[(blkoff % PAGE_SIZE) / sizeof(union afs_dir_block)]; - ret = afs_dir_iterate_block(fpos, dblock, blkoff, - cookie, filldir); + ret = afs_dir_iterate_block(ctx, dblock, blkoff); if (ret != 1) { afs_dir_put_page(page); goto out; @@ -372,7 +366,7 @@ static int afs_dir_iterate(struct inode *dir, unsigned *fpos, void *cookie, blkoff += sizeof(union afs_dir_block); - } while (*fpos < dir->i_size && blkoff < limit); + } while (ctx->pos < dir->i_size && blkoff < limit); afs_dir_put_page(page); ret = 0; @@ -386,23 +380,10 @@ out: /* * read an AFS directory */ -static int afs_readdir(struct file *file, void *cookie, filldir_t filldir) +static int afs_readdir(struct file *file, struct dir_context *ctx) { - unsigned fpos; - int ret; - - _enter("{%Ld,{%lu}}", - file->f_pos, file->f_path.dentry->d_inode->i_ino); - - ASSERT(file->private_data != NULL); - - fpos = file->f_pos; - ret = afs_dir_iterate(file->f_path.dentry->d_inode, &fpos, - cookie, filldir, file->private_data); - file->f_pos = fpos; - - _leave(" = %d", ret); - return ret; + return afs_dir_iterate(file_inode(file), + ctx, file->private_data); } /* @@ -415,15 +396,16 @@ static int afs_lookup_filldir(void *_cookie, const char *name, int nlen, { struct afs_lookup_cookie *cookie = _cookie; - _enter("{%s,%Zu},%s,%u,,%llu,%u", - cookie->name, cookie->nlen, name, nlen, + _enter("{%s,%u},%s,%u,,%llu,%u", + cookie->name.name, cookie->name.len, name, nlen, (unsigned long long) ino, dtype); /* insanity checks first */ BUILD_BUG_ON(sizeof(union afs_dir_block) != 2048); BUILD_BUG_ON(sizeof(union afs_dirent) != 32); - if (cookie->nlen != nlen || memcmp(cookie->name, name, nlen) != 0) { + if (cookie->name.len != nlen || + memcmp(cookie->name.name, name, nlen) != 0) { _leave(" = 0 [no]"); return 0; } @@ -443,24 +425,18 @@ static int afs_lookup_filldir(void *_cookie, const char *name, int nlen, static int afs_do_lookup(struct inode *dir, struct dentry *dentry, struct afs_fid *fid, struct key *key) { - struct afs_lookup_cookie cookie; - struct afs_super_info *as; - unsigned fpos; + struct afs_super_info *as = dir->i_sb->s_fs_info; + struct afs_lookup_cookie cookie = { + .ctx.actor = afs_lookup_filldir, + .name = dentry->d_name, + .fid.vid = as->volume->vid + }; int ret; _enter("{%lu},%p{%s},", dir->i_ino, dentry, dentry->d_name.name); - as = dir->i_sb->s_fs_info; - /* search the directory */ - cookie.name = dentry->d_name.name; - cookie.nlen = dentry->d_name.len; - cookie.fid.vid = as->volume->vid; - cookie.found = 0; - - fpos = 0; - ret = afs_dir_iterate(dir, &fpos, &cookie, afs_lookup_filldir, - key); + ret = afs_dir_iterate(dir, &cookie.ctx, key); if (ret < 0) { _leave(" = %d [iter]", ret); return ret; @@ -478,10 +454,44 @@ static int afs_do_lookup(struct inode *dir, struct dentry *dentry, } /* + * Try to auto mount the mountpoint with pseudo directory, if the autocell + * operation is setted. + */ +static struct inode *afs_try_auto_mntpt( + int ret, struct dentry *dentry, struct inode *dir, struct key *key, + struct afs_fid *fid) +{ + const char *devname = dentry->d_name.name; + struct afs_vnode *vnode = AFS_FS_I(dir); + struct inode *inode; + + _enter("%d, %p{%s}, {%x:%u}, %p", + ret, dentry, devname, vnode->fid.vid, vnode->fid.vnode, key); + + if (ret != -ENOENT || + !test_bit(AFS_VNODE_AUTOCELL, &vnode->flags)) + goto out; + + inode = afs_iget_autocell(dir, devname, strlen(devname), key); + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); + goto out; + } + + *fid = AFS_FS_I(inode)->fid; + _leave("= %p", inode); + return inode; + +out: + _leave("= %d", ret); + return ERR_PTR(ret); +} + +/* * look up an entry in a directory */ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct afs_vnode *vnode; struct afs_fid fid; @@ -491,12 +501,12 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, vnode = AFS_FS_I(dir); - _enter("{%x:%d},%p{%s},", + _enter("{%x:%u},%p{%s},", vnode->fid.vid, vnode->fid.vnode, dentry, dentry->d_name.name); ASSERTCMP(dentry->d_inode, ==, NULL); - if (dentry->d_name.len > 255) { + if (dentry->d_name.len >= AFSNAMEMAX) { _leave(" = -ENAMETOOLONG"); return ERR_PTR(-ENAMETOOLONG); } @@ -509,7 +519,7 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, key = afs_request_key(vnode->volume->cell); if (IS_ERR(key)) { _leave(" = %ld [key]", PTR_ERR(key)); - return ERR_PTR(PTR_ERR(key)); + return ERR_CAST(key); } ret = afs_validate(vnode, key); @@ -521,6 +531,13 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, ret = afs_do_lookup(dir, dentry, &fid, key); if (ret < 0) { + inode = afs_try_auto_mntpt(ret, dentry, dir, key, &fid); + if (!IS_ERR(inode)) { + key_put(key); + goto success; + } + + ret = PTR_ERR(inode); key_put(key); if (ret == -ENOENT) { d_add(dentry, NULL); @@ -537,17 +554,16 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, key_put(key); if (IS_ERR(inode)) { _leave(" = %ld", PTR_ERR(inode)); - return ERR_PTR(PTR_ERR(inode)); + return ERR_CAST(inode); } - dentry->d_op = &afs_fs_dentry_operations; - +success: d_add(dentry, inode); - _leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%lu }", + _leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%u }", fid.vnode, fid.unique, dentry->d_inode->i_ino, - dentry->d_inode->i_version); + dentry->d_inode->i_generation); return NULL; } @@ -557,15 +573,18 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, * - NOTE! the hit can be a negative hit too, so we can't assume we have an * inode */ -static int afs_d_revalidate(struct dentry *dentry, struct nameidata *nd) +static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) { struct afs_vnode *vnode, *dir; - struct afs_fid fid; + struct afs_fid uninitialized_var(fid); struct dentry *parent; struct key *key; void *dir_version; int ret; + if (flags & LOOKUP_RCU) + return -ECHILD; + vnode = AFS_FS_I(dentry->d_inode); if (dentry->d_inode) @@ -581,9 +600,6 @@ static int afs_d_revalidate(struct dentry *dentry, struct nameidata *nd) /* lock down the parent dentry so we can peer at it */ parent = dget_parent(dentry); - if (!parent->d_inode) - goto out_bad; - dir = AFS_FS_I(parent->d_inode); /* validate the parent directory */ @@ -627,9 +643,10 @@ static int afs_d_revalidate(struct dentry *dentry, struct nameidata *nd) * been deleted and replaced, and the original vnode ID has * been reused */ if (fid.unique != vnode->fid.unique) { - _debug("%s: file deleted (uq %u -> %u I:%lu)", + _debug("%s: file deleted (uq %u -> %u I:%u)", dentry->d_name.name, fid.unique, - vnode->fid.unique, dentry->d_inode->i_version); + vnode->fid.unique, + dentry->d_inode->i_generation); spin_lock(&vnode->lock); set_bit(AFS_VNODE_DELETED, &vnode->flags); spin_unlock(&vnode->lock); @@ -665,16 +682,12 @@ not_found: spin_unlock(&dentry->d_lock); out_bad: - if (dentry->d_inode) { - /* don't unhash if we have submounts */ - if (have_submounts(dentry)) - goto out_skip; - } + /* don't unhash if we have submounts */ + if (check_submounts_and_drop(dentry) != 0) + goto out_skip; _debug("dropping dentry %s/%s", parent->d_name.name, dentry->d_name.name); - shrink_dcache_parent(dentry); - d_drop(dentry); dput(parent); key_put(key); @@ -688,7 +701,7 @@ out_bad: * - called from dput() when d_count is going to 0. * - return 1 to request dentry be unhashed, 0 otherwise */ -static int afs_d_delete(struct dentry *dentry) +static int afs_d_delete(const struct dentry *dentry) { _enter("%s", dentry->d_name.name); @@ -696,8 +709,9 @@ static int afs_d_delete(struct dentry *dentry) goto zap; if (dentry->d_inode && - test_bit(AFS_VNODE_DELETED, &AFS_FS_I(dentry->d_inode)->flags)) - goto zap; + (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(dentry->d_inode)->flags) || + test_bit(AFS_VNODE_PSEUDODIR, &AFS_FS_I(dentry->d_inode)->flags))) + goto zap; _leave(" = 0 [keep]"); return 0; @@ -718,7 +732,7 @@ static void afs_d_release(struct dentry *dentry) /* * create a directory on an AFS filesystem */ -static int afs_mkdir(struct inode *dir, struct dentry *dentry, int mode) +static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct afs_file_status status; struct afs_callback cb; @@ -731,13 +745,9 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, int mode) dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%s},%o", + _enter("{%x:%u},{%s},%ho", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, mode); - ret = -ENAMETOOLONG; - if (dentry->d_name.len > 255) - goto error; - key = afs_request_key(dvnode->volume->cell); if (IS_ERR(key)) { ret = PTR_ERR(key); @@ -796,13 +806,9 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry) dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%s}", + _enter("{%x:%u},{%s}", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name); - ret = -ENAMETOOLONG; - if (dentry->d_name.len > 255) - goto error; - key = afs_request_key(dvnode->volume->cell); if (IS_ERR(key)) { ret = PTR_ERR(key); @@ -842,11 +848,11 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry) dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%s}", + _enter("{%x:%u},{%s}", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name); ret = -ENAMETOOLONG; - if (dentry->d_name.len > 255) + if (dentry->d_name.len >= AFSNAMEMAX) goto error; key = afs_request_key(dvnode->volume->cell); @@ -902,8 +908,8 @@ error: /* * create a regular file on an AFS filesystem */ -static int afs_create(struct inode *dir, struct dentry *dentry, int mode, - struct nameidata *nd) +static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, + bool excl) { struct afs_file_status status; struct afs_callback cb; @@ -916,13 +922,9 @@ static int afs_create(struct inode *dir, struct dentry *dentry, int mode, dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%s},%o,", + _enter("{%x:%u},{%s},%ho,", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, mode); - ret = -ENAMETOOLONG; - if (dentry->d_name.len > 255) - goto error; - key = afs_request_key(dvnode->volume->cell); if (IS_ERR(key)) { ret = PTR_ERR(key); @@ -983,15 +985,11 @@ static int afs_link(struct dentry *from, struct inode *dir, vnode = AFS_FS_I(from->d_inode); dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%x:%d},{%s}", + _enter("{%x:%u},{%x:%u},{%s}", vnode->fid.vid, vnode->fid.vnode, dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name); - ret = -ENAMETOOLONG; - if (dentry->d_name.len > 255) - goto error; - key = afs_request_key(dvnode->volume->cell); if (IS_ERR(key)) { ret = PTR_ERR(key); @@ -1002,7 +1000,7 @@ static int afs_link(struct dentry *from, struct inode *dir, if (ret < 0) goto link_error; - atomic_inc(&vnode->vfs_inode.i_count); + ihold(&vnode->vfs_inode); d_instantiate(dentry, &vnode->vfs_inode); key_put(key); _leave(" = 0"); @@ -1032,16 +1030,12 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry, dvnode = AFS_FS_I(dir); - _enter("{%x:%d},{%s},%s", + _enter("{%x:%u},{%s},%s", dvnode->fid.vid, dvnode->fid.vnode, dentry->d_name.name, content); - ret = -ENAMETOOLONG; - if (dentry->d_name.len > 255) - goto error; - ret = -EINVAL; - if (strlen(content) > 1023) + if (strlen(content) >= AFSPATHMAX) goto error; key = afs_request_key(dvnode->volume->cell); @@ -1104,16 +1098,12 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, orig_dvnode = AFS_FS_I(old_dir); new_dvnode = AFS_FS_I(new_dir); - _enter("{%x:%d},{%x:%d},{%x:%d},{%s}", + _enter("{%x:%u},{%x:%u},{%x:%u},{%s}", orig_dvnode->fid.vid, orig_dvnode->fid.vnode, vnode->fid.vid, vnode->fid.vnode, new_dvnode->fid.vid, new_dvnode->fid.vnode, new_dentry->d_name.name); - ret = -ENAMETOOLONG; - if (new_dentry->d_name.len > 255) - goto error; - key = afs_request_key(orig_dvnode->volume->cell); if (IS_ERR(key)) { ret = PTR_ERR(key); diff --git a/fs/afs/file.c b/fs/afs/file.c index ae256498f4f..932ce07948b 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -12,35 +12,53 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> -#include <linux/slab.h> #include <linux/fs.h> #include <linux/pagemap.h> +#include <linux/writeback.h> +#include <linux/gfp.h> #include "internal.h" -static int afs_file_readpage(struct file *file, struct page *page); -static void afs_file_invalidatepage(struct page *page, unsigned long offset); -static int afs_file_releasepage(struct page *page, gfp_t gfp_flags); +static int afs_readpage(struct file *file, struct page *page); +static void afs_invalidatepage(struct page *page, unsigned int offset, + unsigned int length); +static int afs_releasepage(struct page *page, gfp_t gfp_flags); +static int afs_launder_page(struct page *page); + +static int afs_readpages(struct file *filp, struct address_space *mapping, + struct list_head *pages, unsigned nr_pages); const struct file_operations afs_file_operations = { .open = afs_open, .release = afs_release, .llseek = generic_file_llseek, - .read = do_sync_read, - .aio_read = generic_file_aio_read, + .read = new_sync_read, + .write = new_sync_write, + .read_iter = generic_file_read_iter, + .write_iter = afs_file_write, .mmap = generic_file_readonly_mmap, - .sendfile = generic_file_sendfile, + .splice_read = generic_file_splice_read, + .fsync = afs_fsync, + .lock = afs_lock, + .flock = afs_flock, }; const struct inode_operations afs_file_inode_operations = { - .getattr = afs_inode_getattr, + .getattr = afs_getattr, + .setattr = afs_setattr, .permission = afs_permission, }; const struct address_space_operations afs_fs_aops = { - .readpage = afs_file_readpage, - .set_page_dirty = __set_page_dirty_nobuffers, - .releasepage = afs_file_releasepage, - .invalidatepage = afs_file_invalidatepage, + .readpage = afs_readpage, + .readpages = afs_readpages, + .set_page_dirty = afs_set_page_dirty, + .launder_page = afs_launder_page, + .releasepage = afs_releasepage, + .invalidatepage = afs_invalidatepage, + .write_begin = afs_write_begin, + .write_end = afs_write_end, + .writepage = afs_writepage, + .writepages = afs_writepages, }; /* @@ -52,7 +70,7 @@ int afs_open(struct inode *inode, struct file *file) struct key *key; int ret; - _enter("{%x:%x},", vnode->fid.vid, vnode->fid.vnode); + _enter("{%x:%u},", vnode->fid.vid, vnode->fid.vnode); key = afs_request_key(vnode->volume->cell); if (IS_ERR(key)) { @@ -78,79 +96,54 @@ int afs_release(struct inode *inode, struct file *file) { struct afs_vnode *vnode = AFS_FS_I(inode); - _enter("{%x:%x},", vnode->fid.vid, vnode->fid.vnode); + _enter("{%x:%u},", vnode->fid.vid, vnode->fid.vnode); key_put(file->private_data); _leave(" = 0"); return 0; } +#ifdef CONFIG_AFS_FSCACHE /* * deal with notification that a page was read from the cache */ -#ifdef AFS_CACHING_SUPPORT -static void afs_file_readpage_read_complete(void *cookie_data, - struct page *page, +static void afs_file_readpage_read_complete(struct page *page, void *data, int error) { - _enter("%p,%p,%p,%d", cookie_data, page, data, error); + _enter("%p,%p,%d", page, data, error); - if (error) - SetPageError(page); - else + /* if the read completes with an error, we just unlock the page and let + * the VM reissue the readpage */ + if (!error) SetPageUptodate(page); unlock_page(page); - -} -#endif - -/* - * deal with notification that a page was written to the cache - */ -#ifdef AFS_CACHING_SUPPORT -static void afs_file_readpage_write_complete(void *cookie_data, - struct page *page, - void *data, - int error) -{ - _enter("%p,%p,%p,%d", cookie_data, page, data, error); - - unlock_page(page); } #endif /* - * AFS read page from file (or symlink) + * read page from file, directory or symlink, given a key to use */ -static int afs_file_readpage(struct file *file, struct page *page) +int afs_page_filler(void *data, struct page *page) { - struct afs_vnode *vnode; - struct inode *inode; - struct key *key; + struct inode *inode = page->mapping->host; + struct afs_vnode *vnode = AFS_FS_I(inode); + struct key *key = data; size_t len; off_t offset; int ret; - inode = page->mapping->host; - - ASSERT(file != NULL); - key = file->private_data; - ASSERT(key != NULL); - _enter("{%x},{%lu},{%lu}", key_serial(key), inode->i_ino, page->index); - vnode = AFS_FS_I(inode); - BUG_ON(!PageLocked(page)); ret = -ESTALE; if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) goto error; -#ifdef AFS_CACHING_SUPPORT /* is it cached? */ - ret = cachefs_read_or_alloc_page(vnode->cache, +#ifdef CONFIG_AFS_FSCACHE + ret = fscache_read_or_alloc_page(vnode->cache, page, afs_file_readpage_read_complete, NULL, @@ -158,20 +151,21 @@ static int afs_file_readpage(struct file *file, struct page *page) #else ret = -ENOBUFS; #endif - switch (ret) { - /* read BIO submitted and wb-journal entry found */ - case 1: - BUG(); // TODO - handle wb-journal match - /* read BIO submitted (page in cache) */ case 0: break; - /* no page available in cache */ - case -ENOBUFS: + /* page not yet cached */ case -ENODATA: + _debug("cache said ENODATA"); + goto go_on; + + /* page will not be cached */ + case -ENOBUFS: + _debug("cache said ENOBUFS"); default: + go_on: offset = page->index << PAGE_CACHE_SHIFT; len = min_t(size_t, i_size_read(inode) - offset, PAGE_SIZE); @@ -185,27 +179,25 @@ static int afs_file_readpage(struct file *file, struct page *page) set_bit(AFS_VNODE_DELETED, &vnode->flags); ret = -ESTALE; } -#ifdef AFS_CACHING_SUPPORT - cachefs_uncache_page(vnode->cache, page); + +#ifdef CONFIG_AFS_FSCACHE + fscache_uncache_page(vnode->cache, page); #endif + BUG_ON(PageFsCache(page)); goto error; } SetPageUptodate(page); -#ifdef AFS_CACHING_SUPPORT - if (cachefs_write_page(vnode->cache, - page, - afs_file_readpage_write_complete, - NULL, - GFP_KERNEL) != 0 - ) { - cachefs_uncache_page(vnode->cache, page); - unlock_page(page); + /* send the page to the cache */ +#ifdef CONFIG_AFS_FSCACHE + if (PageFsCache(page) && + fscache_write_page(vnode->cache, page, GFP_KERNEL) != 0) { + fscache_uncache_page(vnode->cache, page); + BUG_ON(PageFsCache(page)); } -#else - unlock_page(page); #endif + unlock_page(page); } _leave(" = 0"); @@ -219,80 +211,170 @@ error: } /* - * get a page cookie for the specified page + * read page from file, directory or symlink, given a file to nominate the key + * to be used */ -#ifdef AFS_CACHING_SUPPORT -int afs_cache_get_page_cookie(struct page *page, - struct cachefs_page **_page_cookie) +static int afs_readpage(struct file *file, struct page *page) { + struct key *key; int ret; - _enter(""); - ret = cachefs_page_get_private(page,_page_cookie, GFP_NOIO); - - _leave(" = %d", ret); + if (file) { + key = file->private_data; + ASSERT(key != NULL); + ret = afs_page_filler(key, page); + } else { + struct inode *inode = page->mapping->host; + key = afs_request_key(AFS_FS_S(inode->i_sb)->volume->cell); + if (IS_ERR(key)) { + ret = PTR_ERR(key); + } else { + ret = afs_page_filler(key, page); + key_put(key); + } + } return ret; } + +/* + * read a set of pages + */ +static int afs_readpages(struct file *file, struct address_space *mapping, + struct list_head *pages, unsigned nr_pages) +{ + struct key *key = file->private_data; + struct afs_vnode *vnode; + int ret = 0; + + _enter("{%d},{%lu},,%d", + key_serial(key), mapping->host->i_ino, nr_pages); + + ASSERT(key != NULL); + + vnode = AFS_FS_I(mapping->host); + if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { + _leave(" = -ESTALE"); + return -ESTALE; + } + + /* attempt to read as many of the pages as possible */ +#ifdef CONFIG_AFS_FSCACHE + ret = fscache_read_or_alloc_pages(vnode->cache, + mapping, + pages, + &nr_pages, + afs_file_readpage_read_complete, + NULL, + mapping_gfp_mask(mapping)); +#else + ret = -ENOBUFS; #endif + switch (ret) { + /* all pages are being read from the cache */ + case 0: + BUG_ON(!list_empty(pages)); + BUG_ON(nr_pages != 0); + _leave(" = 0 [reading all]"); + return 0; + + /* there were pages that couldn't be read from the cache */ + case -ENODATA: + case -ENOBUFS: + break; + + /* other error */ + default: + _leave(" = %d", ret); + return ret; + } + + /* load the missing pages from the network */ + ret = read_cache_pages(mapping, pages, afs_page_filler, key); + + _leave(" = %d [netting]", ret); + return ret; +} + +/* + * write back a dirty page + */ +static int afs_launder_page(struct page *page) +{ + _enter("{%lu}", page->index); + + return 0; +} + /* * invalidate part or all of a page + * - release a page and clean up its private data if offset is 0 (indicating + * the entire page) */ -static void afs_file_invalidatepage(struct page *page, unsigned long offset) +static void afs_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { - int ret = 1; + struct afs_writeback *wb = (struct afs_writeback *) page_private(page); - _enter("{%lu},%lu", page->index, offset); + _enter("{%lu},%u,%u", page->index, offset, length); BUG_ON(!PageLocked(page)); - if (PagePrivate(page)) { -#ifdef AFS_CACHING_SUPPORT - struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); - cachefs_uncache_page(vnode->cache,page); + /* we clean up only if the entire page is being invalidated */ + if (offset == 0 && length == PAGE_CACHE_SIZE) { +#ifdef CONFIG_AFS_FSCACHE + if (PageFsCache(page)) { + struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); + fscache_wait_on_page_write(vnode->cache, page); + fscache_uncache_page(vnode->cache, page); + } #endif - /* We release buffers only if the entire page is being - * invalidated. - * The get_block cached value has been unconditionally - * invalidated, so real IO is not possible anymore. - */ - if (offset == 0) { - BUG_ON(!PageLocked(page)); - - ret = 0; - if (!PageWriteback(page)) - ret = page->mapping->a_ops->releasepage(page, - 0); - /* possibly should BUG_ON(!ret); - neilb */ + if (PagePrivate(page)) { + if (wb && !PageWriteback(page)) { + set_page_private(page, 0); + afs_put_writeback(wb); + } + + if (!page_private(page)) + ClearPagePrivate(page); } } - _leave(" = %d", ret); + _leave(""); } /* - * release a page and cleanup its private data + * release a page and clean up its private state if it's not busy + * - return true if the page can now be released, false if not */ -static int afs_file_releasepage(struct page *page, gfp_t gfp_flags) +static int afs_releasepage(struct page *page, gfp_t gfp_flags) { - struct cachefs_page *pageio; - - _enter("{%lu},%x", page->index, gfp_flags); - - if (PagePrivate(page)) { -#ifdef AFS_CACHING_SUPPORT - struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); - cachefs_uncache_page(vnode->cache, page); + struct afs_writeback *wb = (struct afs_writeback *) page_private(page); + struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); + + _enter("{{%x:%u}[%lu],%lx},%x", + vnode->fid.vid, vnode->fid.vnode, page->index, page->flags, + gfp_flags); + + /* deny if page is being written to the cache and the caller hasn't + * elected to wait */ +#ifdef CONFIG_AFS_FSCACHE + if (!fscache_maybe_release_page(vnode->cache, page, gfp_flags)) { + _leave(" = F [cache busy]"); + return 0; + } #endif - pageio = (struct cachefs_page *) page_private(page); - set_page_private(page, 0); + if (PagePrivate(page)) { + if (wb) { + set_page_private(page, 0); + afs_put_writeback(wb); + } ClearPagePrivate(page); - - kfree(pageio); } - _leave(" = 0"); - return 0; + /* indicate that the page can be released */ + _leave(" = T"); + return 1; } diff --git a/fs/afs/flock.c b/fs/afs/flock.c new file mode 100644 index 00000000000..4baf1d2b39e --- /dev/null +++ b/fs/afs/flock.c @@ -0,0 +1,585 @@ +/* AFS file locking support + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include "internal.h" + +#define AFS_LOCK_GRANTED 0 +#define AFS_LOCK_PENDING 1 + +static void afs_fl_copy_lock(struct file_lock *new, struct file_lock *fl); +static void afs_fl_release_private(struct file_lock *fl); + +static struct workqueue_struct *afs_lock_manager; +static DEFINE_MUTEX(afs_lock_manager_mutex); + +static const struct file_lock_operations afs_lock_ops = { + .fl_copy_lock = afs_fl_copy_lock, + .fl_release_private = afs_fl_release_private, +}; + +/* + * initialise the lock manager thread if it isn't already running + */ +static int afs_init_lock_manager(void) +{ + int ret; + + ret = 0; + if (!afs_lock_manager) { + mutex_lock(&afs_lock_manager_mutex); + if (!afs_lock_manager) { + afs_lock_manager = + create_singlethread_workqueue("kafs_lockd"); + if (!afs_lock_manager) + ret = -ENOMEM; + } + mutex_unlock(&afs_lock_manager_mutex); + } + return ret; +} + +/* + * destroy the lock manager thread if it's running + */ +void __exit afs_kill_lock_manager(void) +{ + if (afs_lock_manager) + destroy_workqueue(afs_lock_manager); +} + +/* + * if the callback is broken on this vnode, then the lock may now be available + */ +void afs_lock_may_be_available(struct afs_vnode *vnode) +{ + _enter("{%x:%u}", vnode->fid.vid, vnode->fid.vnode); + + queue_delayed_work(afs_lock_manager, &vnode->lock_work, 0); +} + +/* + * the lock will time out in 5 minutes unless we extend it, so schedule + * extension in a bit less than that time + */ +static void afs_schedule_lock_extension(struct afs_vnode *vnode) +{ + queue_delayed_work(afs_lock_manager, &vnode->lock_work, + AFS_LOCKWAIT * HZ / 2); +} + +/* + * grant one or more locks (readlocks are allowed to jump the queue if the + * first lock in the queue is itself a readlock) + * - the caller must hold the vnode lock + */ +static void afs_grant_locks(struct afs_vnode *vnode, struct file_lock *fl) +{ + struct file_lock *p, *_p; + + list_move_tail(&fl->fl_u.afs.link, &vnode->granted_locks); + if (fl->fl_type == F_RDLCK) { + list_for_each_entry_safe(p, _p, &vnode->pending_locks, + fl_u.afs.link) { + if (p->fl_type == F_RDLCK) { + p->fl_u.afs.state = AFS_LOCK_GRANTED; + list_move_tail(&p->fl_u.afs.link, + &vnode->granted_locks); + wake_up(&p->fl_wait); + } + } + } +} + +/* + * do work for a lock, including: + * - probing for a lock we're waiting on but didn't get immediately + * - extending a lock that's close to timing out + */ +void afs_lock_work(struct work_struct *work) +{ + struct afs_vnode *vnode = + container_of(work, struct afs_vnode, lock_work.work); + struct file_lock *fl; + afs_lock_type_t type; + struct key *key; + int ret; + + _enter("{%x:%u}", vnode->fid.vid, vnode->fid.vnode); + + spin_lock(&vnode->lock); + + if (test_bit(AFS_VNODE_UNLOCKING, &vnode->flags)) { + _debug("unlock"); + spin_unlock(&vnode->lock); + + /* attempt to release the server lock; if it fails, we just + * wait 5 minutes and it'll time out anyway */ + ret = afs_vnode_release_lock(vnode, vnode->unlock_key); + if (ret < 0) + printk(KERN_WARNING "AFS:" + " Failed to release lock on {%x:%x} error %d\n", + vnode->fid.vid, vnode->fid.vnode, ret); + + spin_lock(&vnode->lock); + key_put(vnode->unlock_key); + vnode->unlock_key = NULL; + clear_bit(AFS_VNODE_UNLOCKING, &vnode->flags); + } + + /* if we've got a lock, then it must be time to extend that lock as AFS + * locks time out after 5 minutes */ + if (!list_empty(&vnode->granted_locks)) { + _debug("extend"); + + if (test_and_set_bit(AFS_VNODE_LOCKING, &vnode->flags)) + BUG(); + fl = list_entry(vnode->granted_locks.next, + struct file_lock, fl_u.afs.link); + key = key_get(fl->fl_file->private_data); + spin_unlock(&vnode->lock); + + ret = afs_vnode_extend_lock(vnode, key); + clear_bit(AFS_VNODE_LOCKING, &vnode->flags); + key_put(key); + switch (ret) { + case 0: + afs_schedule_lock_extension(vnode); + break; + default: + /* ummm... we failed to extend the lock - retry + * extension shortly */ + printk(KERN_WARNING "AFS:" + " Failed to extend lock on {%x:%x} error %d\n", + vnode->fid.vid, vnode->fid.vnode, ret); + queue_delayed_work(afs_lock_manager, &vnode->lock_work, + HZ * 10); + break; + } + _leave(" [extend]"); + return; + } + + /* if we don't have a granted lock, then we must've been called back by + * the server, and so if might be possible to get a lock we're + * currently waiting for */ + if (!list_empty(&vnode->pending_locks)) { + _debug("get"); + + if (test_and_set_bit(AFS_VNODE_LOCKING, &vnode->flags)) + BUG(); + fl = list_entry(vnode->pending_locks.next, + struct file_lock, fl_u.afs.link); + key = key_get(fl->fl_file->private_data); + type = (fl->fl_type == F_RDLCK) ? + AFS_LOCK_READ : AFS_LOCK_WRITE; + spin_unlock(&vnode->lock); + + ret = afs_vnode_set_lock(vnode, key, type); + clear_bit(AFS_VNODE_LOCKING, &vnode->flags); + switch (ret) { + case -EWOULDBLOCK: + _debug("blocked"); + break; + case 0: + _debug("acquired"); + if (type == AFS_LOCK_READ) + set_bit(AFS_VNODE_READLOCKED, &vnode->flags); + else + set_bit(AFS_VNODE_WRITELOCKED, &vnode->flags); + ret = AFS_LOCK_GRANTED; + default: + spin_lock(&vnode->lock); + /* the pending lock may have been withdrawn due to a + * signal */ + if (list_entry(vnode->pending_locks.next, + struct file_lock, fl_u.afs.link) == fl) { + fl->fl_u.afs.state = ret; + if (ret == AFS_LOCK_GRANTED) + afs_grant_locks(vnode, fl); + else + list_del_init(&fl->fl_u.afs.link); + wake_up(&fl->fl_wait); + spin_unlock(&vnode->lock); + } else { + _debug("withdrawn"); + clear_bit(AFS_VNODE_READLOCKED, &vnode->flags); + clear_bit(AFS_VNODE_WRITELOCKED, &vnode->flags); + spin_unlock(&vnode->lock); + afs_vnode_release_lock(vnode, key); + if (!list_empty(&vnode->pending_locks)) + afs_lock_may_be_available(vnode); + } + break; + } + key_put(key); + _leave(" [pend]"); + return; + } + + /* looks like the lock request was withdrawn on a signal */ + spin_unlock(&vnode->lock); + _leave(" [no locks]"); +} + +/* + * pass responsibility for the unlocking of a vnode on the server to the + * manager thread, lest a pending signal in the calling thread interrupt + * AF_RXRPC + * - the caller must hold the vnode lock + */ +static void afs_defer_unlock(struct afs_vnode *vnode, struct key *key) +{ + cancel_delayed_work(&vnode->lock_work); + if (!test_and_clear_bit(AFS_VNODE_READLOCKED, &vnode->flags) && + !test_and_clear_bit(AFS_VNODE_WRITELOCKED, &vnode->flags)) + BUG(); + if (test_and_set_bit(AFS_VNODE_UNLOCKING, &vnode->flags)) + BUG(); + vnode->unlock_key = key_get(key); + afs_lock_may_be_available(vnode); +} + +/* + * request a lock on a file on the server + */ +static int afs_do_setlk(struct file *file, struct file_lock *fl) +{ + struct inode *inode = file_inode(file); + struct afs_vnode *vnode = AFS_FS_I(inode); + afs_lock_type_t type; + struct key *key = file->private_data; + int ret; + + _enter("{%x:%u},%u", vnode->fid.vid, vnode->fid.vnode, fl->fl_type); + + /* only whole-file locks are supported */ + if (fl->fl_start != 0 || fl->fl_end != OFFSET_MAX) + return -EINVAL; + + ret = afs_init_lock_manager(); + if (ret < 0) + return ret; + + fl->fl_ops = &afs_lock_ops; + INIT_LIST_HEAD(&fl->fl_u.afs.link); + fl->fl_u.afs.state = AFS_LOCK_PENDING; + + type = (fl->fl_type == F_RDLCK) ? AFS_LOCK_READ : AFS_LOCK_WRITE; + + spin_lock(&inode->i_lock); + + /* make sure we've got a callback on this file and that our view of the + * data version is up to date */ + ret = afs_vnode_fetch_status(vnode, NULL, key); + if (ret < 0) + goto error; + + if (vnode->status.lock_count != 0 && !(fl->fl_flags & FL_SLEEP)) { + ret = -EAGAIN; + goto error; + } + + spin_lock(&vnode->lock); + + /* if we've already got a readlock on the server then we can instantly + * grant another readlock, irrespective of whether there are any + * pending writelocks */ + if (type == AFS_LOCK_READ && + vnode->flags & (1 << AFS_VNODE_READLOCKED)) { + _debug("instant readlock"); + ASSERTCMP(vnode->flags & + ((1 << AFS_VNODE_LOCKING) | + (1 << AFS_VNODE_WRITELOCKED)), ==, 0); + ASSERT(!list_empty(&vnode->granted_locks)); + goto sharing_existing_lock; + } + + /* if there's no-one else with a lock on this vnode, then we need to + * ask the server for a lock */ + if (list_empty(&vnode->pending_locks) && + list_empty(&vnode->granted_locks)) { + _debug("not locked"); + ASSERTCMP(vnode->flags & + ((1 << AFS_VNODE_LOCKING) | + (1 << AFS_VNODE_READLOCKED) | + (1 << AFS_VNODE_WRITELOCKED)), ==, 0); + list_add_tail(&fl->fl_u.afs.link, &vnode->pending_locks); + set_bit(AFS_VNODE_LOCKING, &vnode->flags); + spin_unlock(&vnode->lock); + + ret = afs_vnode_set_lock(vnode, key, type); + clear_bit(AFS_VNODE_LOCKING, &vnode->flags); + switch (ret) { + case 0: + _debug("acquired"); + goto acquired_server_lock; + case -EWOULDBLOCK: + _debug("would block"); + spin_lock(&vnode->lock); + ASSERT(list_empty(&vnode->granted_locks)); + ASSERTCMP(vnode->pending_locks.next, ==, + &fl->fl_u.afs.link); + goto wait; + default: + spin_lock(&vnode->lock); + list_del_init(&fl->fl_u.afs.link); + spin_unlock(&vnode->lock); + goto error; + } + } + + /* otherwise, we need to wait for a local lock to become available */ + _debug("wait local"); + list_add_tail(&fl->fl_u.afs.link, &vnode->pending_locks); +wait: + if (!(fl->fl_flags & FL_SLEEP)) { + _debug("noblock"); + ret = -EAGAIN; + goto abort_attempt; + } + spin_unlock(&vnode->lock); + + /* now we need to sleep and wait for the lock manager thread to get the + * lock from the server */ + _debug("sleep"); + ret = wait_event_interruptible(fl->fl_wait, + fl->fl_u.afs.state <= AFS_LOCK_GRANTED); + if (fl->fl_u.afs.state <= AFS_LOCK_GRANTED) { + ret = fl->fl_u.afs.state; + if (ret < 0) + goto error; + spin_lock(&vnode->lock); + goto given_lock; + } + + /* we were interrupted, but someone may still be in the throes of + * giving us the lock */ + _debug("intr"); + ASSERTCMP(ret, ==, -ERESTARTSYS); + + spin_lock(&vnode->lock); + if (fl->fl_u.afs.state <= AFS_LOCK_GRANTED) { + ret = fl->fl_u.afs.state; + if (ret < 0) { + spin_unlock(&vnode->lock); + goto error; + } + goto given_lock; + } + +abort_attempt: + /* we aren't going to get the lock, either because we're unwilling to + * wait, or because some signal happened */ + _debug("abort"); + if (list_empty(&vnode->granted_locks) && + vnode->pending_locks.next == &fl->fl_u.afs.link) { + if (vnode->pending_locks.prev != &fl->fl_u.afs.link) { + /* kick the next pending lock into having a go */ + list_del_init(&fl->fl_u.afs.link); + afs_lock_may_be_available(vnode); + } + } else { + list_del_init(&fl->fl_u.afs.link); + } + spin_unlock(&vnode->lock); + goto error; + +acquired_server_lock: + /* we've acquired a server lock, but it needs to be renewed after 5 + * mins */ + spin_lock(&vnode->lock); + afs_schedule_lock_extension(vnode); + if (type == AFS_LOCK_READ) + set_bit(AFS_VNODE_READLOCKED, &vnode->flags); + else + set_bit(AFS_VNODE_WRITELOCKED, &vnode->flags); +sharing_existing_lock: + /* the lock has been granted as far as we're concerned... */ + fl->fl_u.afs.state = AFS_LOCK_GRANTED; + list_move_tail(&fl->fl_u.afs.link, &vnode->granted_locks); +given_lock: + /* ... but we do still need to get the VFS's blessing */ + ASSERT(!(vnode->flags & (1 << AFS_VNODE_LOCKING))); + ASSERT((vnode->flags & ((1 << AFS_VNODE_READLOCKED) | + (1 << AFS_VNODE_WRITELOCKED))) != 0); + ret = posix_lock_file(file, fl, NULL); + if (ret < 0) + goto vfs_rejected_lock; + spin_unlock(&vnode->lock); + + /* again, make sure we've got a callback on this file and, again, make + * sure that our view of the data version is up to date (we ignore + * errors incurred here and deal with the consequences elsewhere) */ + afs_vnode_fetch_status(vnode, NULL, key); + +error: + spin_unlock(&inode->i_lock); + _leave(" = %d", ret); + return ret; + +vfs_rejected_lock: + /* the VFS rejected the lock we just obtained, so we have to discard + * what we just got */ + _debug("vfs refused %d", ret); + list_del_init(&fl->fl_u.afs.link); + if (list_empty(&vnode->granted_locks)) + afs_defer_unlock(vnode, key); + goto abort_attempt; +} + +/* + * unlock on a file on the server + */ +static int afs_do_unlk(struct file *file, struct file_lock *fl) +{ + struct afs_vnode *vnode = AFS_FS_I(file->f_mapping->host); + struct key *key = file->private_data; + int ret; + + _enter("{%x:%u},%u", vnode->fid.vid, vnode->fid.vnode, fl->fl_type); + + /* only whole-file unlocks are supported */ + if (fl->fl_start != 0 || fl->fl_end != OFFSET_MAX) + return -EINVAL; + + fl->fl_ops = &afs_lock_ops; + INIT_LIST_HEAD(&fl->fl_u.afs.link); + fl->fl_u.afs.state = AFS_LOCK_PENDING; + + spin_lock(&vnode->lock); + ret = posix_lock_file(file, fl, NULL); + if (ret < 0) { + spin_unlock(&vnode->lock); + _leave(" = %d [vfs]", ret); + return ret; + } + + /* discard the server lock only if all granted locks are gone */ + if (list_empty(&vnode->granted_locks)) + afs_defer_unlock(vnode, key); + spin_unlock(&vnode->lock); + _leave(" = 0"); + return 0; +} + +/* + * return information about a lock we currently hold, if indeed we hold one + */ +static int afs_do_getlk(struct file *file, struct file_lock *fl) +{ + struct afs_vnode *vnode = AFS_FS_I(file->f_mapping->host); + struct key *key = file->private_data; + int ret, lock_count; + + _enter(""); + + fl->fl_type = F_UNLCK; + + mutex_lock(&vnode->vfs_inode.i_mutex); + + /* check local lock records first */ + ret = 0; + posix_test_lock(file, fl); + if (fl->fl_type == F_UNLCK) { + /* no local locks; consult the server */ + ret = afs_vnode_fetch_status(vnode, NULL, key); + if (ret < 0) + goto error; + lock_count = vnode->status.lock_count; + if (lock_count) { + if (lock_count > 0) + fl->fl_type = F_RDLCK; + else + fl->fl_type = F_WRLCK; + fl->fl_start = 0; + fl->fl_end = OFFSET_MAX; + } + } + +error: + mutex_unlock(&vnode->vfs_inode.i_mutex); + _leave(" = %d [%hd]", ret, fl->fl_type); + return ret; +} + +/* + * manage POSIX locks on a file + */ +int afs_lock(struct file *file, int cmd, struct file_lock *fl) +{ + struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); + + _enter("{%x:%u},%d,{t=%x,fl=%x,r=%Ld:%Ld}", + vnode->fid.vid, vnode->fid.vnode, cmd, + fl->fl_type, fl->fl_flags, + (long long) fl->fl_start, (long long) fl->fl_end); + + /* AFS doesn't support mandatory locks */ + if (__mandatory_lock(&vnode->vfs_inode) && fl->fl_type != F_UNLCK) + return -ENOLCK; + + if (IS_GETLK(cmd)) + return afs_do_getlk(file, fl); + if (fl->fl_type == F_UNLCK) + return afs_do_unlk(file, fl); + return afs_do_setlk(file, fl); +} + +/* + * manage FLOCK locks on a file + */ +int afs_flock(struct file *file, int cmd, struct file_lock *fl) +{ + struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); + + _enter("{%x:%u},%d,{t=%x,fl=%x}", + vnode->fid.vid, vnode->fid.vnode, cmd, + fl->fl_type, fl->fl_flags); + + /* + * No BSD flocks over NFS allowed. + * Note: we could try to fake a POSIX lock request here by + * using ((u32) filp | 0x80000000) or some such as the pid. + * Not sure whether that would be unique, though, or whether + * that would break in other places. + */ + if (!(fl->fl_flags & FL_FLOCK)) + return -ENOLCK; + + /* we're simulating flock() locks using posix locks on the server */ + if (fl->fl_type == F_UNLCK) + return afs_do_unlk(file, fl); + return afs_do_setlk(file, fl); +} + +/* + * the POSIX lock management core VFS code copies the lock record and adds the + * copy into its own list, so we need to add that copy to the vnode's lock + * queue in the same place as the original (which will be deleted shortly + * after) + */ +static void afs_fl_copy_lock(struct file_lock *new, struct file_lock *fl) +{ + _enter(""); + + list_add(&new->fl_u.afs.link, &fl->fl_u.afs.link); +} + +/* + * need to remove this lock from the vnode queue when it's removed from the + * VFS's list + */ +static void afs_fl_release_private(struct file_lock *fl) +{ + _enter(""); + + list_del_init(&fl->fl_u.afs.link); +} diff --git a/fs/afs/fsclient.c b/fs/afs/fsclient.c index e54e6c2ad34..c2e930ec288 100644 --- a/fs/afs/fsclient.c +++ b/fs/afs/fsclient.c @@ -10,6 +10,7 @@ */ #include <linux/init.h> +#include <linux/slab.h> #include <linux/sched.h> #include <linux/circ_buf.h> #include "internal.h" @@ -33,12 +34,16 @@ static void xdr_decode_AFSFid(const __be32 **_bp, struct afs_fid *fid) */ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, struct afs_file_status *status, - struct afs_vnode *vnode) + struct afs_vnode *vnode, + afs_dataversion_t *store_version) { + afs_dataversion_t expected_version; const __be32 *bp = *_bp; umode_t mode; u64 data_version, size; u32 changed = 0; /* becomes non-zero if ctime-type changes seen */ + kuid_t owner; + kgid_t group; #define EXTRACT(DST) \ do { \ @@ -53,7 +58,9 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, size = ntohl(*bp++); data_version = ntohl(*bp++); EXTRACT(status->author); - EXTRACT(status->owner); + owner = make_kuid(&init_user_ns, ntohl(*bp++)); + changed |= !uid_eq(owner, status->owner); + status->owner = owner; EXTRACT(status->caller_access); /* call ticket dependent */ EXTRACT(status->anon_access); EXTRACT(status->mode); @@ -62,10 +69,12 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, bp++; /* seg size */ status->mtime_client = ntohl(*bp++); status->mtime_server = ntohl(*bp++); - EXTRACT(status->group); + group = make_kgid(&init_user_ns, ntohl(*bp++)); + changed |= !gid_eq(group, status->group); + status->group = group; bp++; /* sync counter */ data_version |= (u64) ntohl(*bp++) << 32; - bp++; /* lock count */ + EXTRACT(status->lock_count); size |= (u64) ntohl(*bp++) << 32; bp++; /* spare 4 */ *_bp = bp; @@ -86,8 +95,8 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, i_size_write(&vnode->vfs_inode, size); vnode->vfs_inode.i_uid = status->owner; vnode->vfs_inode.i_gid = status->group; - vnode->vfs_inode.i_version = vnode->fid.unique; - vnode->vfs_inode.i_nlink = status->nlink; + vnode->vfs_inode.i_generation = vnode->fid.unique; + set_nlink(&vnode->vfs_inode, status->nlink); mode = vnode->vfs_inode.i_mode; mode &= ~S_IALLUGO; @@ -99,9 +108,14 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, vnode->vfs_inode.i_ctime.tv_sec = status->mtime_server; vnode->vfs_inode.i_mtime = vnode->vfs_inode.i_ctime; vnode->vfs_inode.i_atime = vnode->vfs_inode.i_ctime; + vnode->vfs_inode.i_version = data_version; } - if (status->data_version != data_version) { + expected_version = status->data_version; + if (store_version) + expected_version = *store_version; + + if (expected_version != data_version) { status->data_version = data_version; if (vnode && !test_bit(AFS_VNODE_UNSET, &vnode->flags)) { _debug("vnode modified %llx on {%x:%u}", @@ -110,6 +124,8 @@ static void xdr_decode_AFSFetchStatus(const __be32 **_bp, set_bit(AFS_VNODE_MODIFIED, &vnode->flags); set_bit(AFS_VNODE_ZAP_DATA, &vnode->flags); } + } else if (store_version) { + status->data_version = data_version; } } @@ -156,6 +172,67 @@ static void xdr_decode_AFSVolSync(const __be32 **_bp, } /* + * encode the requested attributes into an AFSStoreStatus block + */ +static void xdr_encode_AFS_StoreStatus(__be32 **_bp, struct iattr *attr) +{ + __be32 *bp = *_bp; + u32 mask = 0, mtime = 0, owner = 0, group = 0, mode = 0; + + mask = 0; + if (attr->ia_valid & ATTR_MTIME) { + mask |= AFS_SET_MTIME; + mtime = attr->ia_mtime.tv_sec; + } + + if (attr->ia_valid & ATTR_UID) { + mask |= AFS_SET_OWNER; + owner = from_kuid(&init_user_ns, attr->ia_uid); + } + + if (attr->ia_valid & ATTR_GID) { + mask |= AFS_SET_GROUP; + group = from_kgid(&init_user_ns, attr->ia_gid); + } + + if (attr->ia_valid & ATTR_MODE) { + mask |= AFS_SET_MODE; + mode = attr->ia_mode & S_IALLUGO; + } + + *bp++ = htonl(mask); + *bp++ = htonl(mtime); + *bp++ = htonl(owner); + *bp++ = htonl(group); + *bp++ = htonl(mode); + *bp++ = 0; /* segment size */ + *_bp = bp; +} + +/* + * decode an AFSFetchVolumeStatus block + */ +static void xdr_decode_AFSFetchVolumeStatus(const __be32 **_bp, + struct afs_volume_status *vs) +{ + const __be32 *bp = *_bp; + + vs->vid = ntohl(*bp++); + vs->parent_id = ntohl(*bp++); + vs->online = ntohl(*bp++); + vs->in_service = ntohl(*bp++); + vs->blessed = ntohl(*bp++); + vs->needs_salvage = ntohl(*bp++); + vs->type = ntohl(*bp++); + vs->min_quota = ntohl(*bp++); + vs->max_quota = ntohl(*bp++); + vs->blocks_in_use = ntohl(*bp++); + vs->part_blocks_avail = ntohl(*bp++); + vs->part_max_blocks = ntohl(*bp++); + *_bp = bp; +} + +/* * deliver reply data to an FS.FetchStatus */ static int afs_deliver_fs_fetch_status(struct afs_call *call, @@ -175,7 +252,7 @@ static int afs_deliver_fs_fetch_status(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); xdr_decode_AFSCallBack(&bp, vnode); if (call->reply2) xdr_decode_AFSVolSync(&bp, call->reply2); @@ -206,7 +283,7 @@ int afs_fs_fetch_file_status(struct afs_server *server, struct afs_call *call; __be32 *bp; - _enter(",%x,{%x:%d},,", + _enter(",%x,{%x:%u},,", key_serial(key), vnode->fid.vid, vnode->fid.vnode); call = afs_alloc_flat_call(&afs_RXFSFetchStatus, 16, (21 + 3 + 6) * 4); @@ -247,10 +324,16 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call, case 0: call->offset = 0; call->unmarshall++; + if (call->operation_ID != FSFETCHDATA64) { + call->unmarshall++; + goto no_msw; + } - /* extract the returned data length */ + /* extract the upper part of the returned data length of an + * FSFETCHDATA64 op (which should always be 0 using this + * client) */ case 1: - _debug("extract data length"); + _debug("extract data length (MSW)"); ret = afs_extract_data(call, skb, last, &call->tmp, 4); switch (ret) { case 0: break; @@ -259,38 +342,51 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call, } call->count = ntohl(call->tmp); - _debug("DATA length: %u", call->count); - if (call->count > PAGE_SIZE) + _debug("DATA length MSW: %u", call->count); + if (call->count > 0) return -EBADMSG; call->offset = 0; call->unmarshall++; - if (call->count < PAGE_SIZE) { - page = call->reply3; - buffer = kmap_atomic(page, KM_USER0); - memset(buffer + PAGE_SIZE - call->count, 0, - call->count); - kunmap_atomic(buffer, KM_USER0); - } - - /* extract the returned data */ + no_msw: + /* extract the returned data length */ case 2: - _debug("extract data"); - page = call->reply3; - buffer = kmap_atomic(page, KM_USER0); - ret = afs_extract_data(call, skb, last, buffer, call->count); - kunmap_atomic(buffer, KM_USER0); + _debug("extract data length"); + ret = afs_extract_data(call, skb, last, &call->tmp, 4); switch (ret) { case 0: break; case -EAGAIN: return 0; default: return ret; } + call->count = ntohl(call->tmp); + _debug("DATA length: %u", call->count); + if (call->count > PAGE_SIZE) + return -EBADMSG; call->offset = 0; call->unmarshall++; - /* extract the metadata */ + /* extract the returned data */ case 3: + _debug("extract data"); + if (call->count > 0) { + page = call->reply3; + buffer = kmap_atomic(page); + ret = afs_extract_data(call, skb, last, buffer, + call->count); + kunmap_atomic(buffer); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + } + + call->offset = 0; + call->unmarshall++; + + /* extract the metadata */ + case 4: ret = afs_extract_data(call, skb, last, call->buffer, (21 + 3 + 6) * 4); switch (ret) { @@ -300,7 +396,7 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call, } bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); xdr_decode_AFSCallBack(&bp, vnode); if (call->reply2) xdr_decode_AFSVolSync(&bp, call->reply2); @@ -308,7 +404,7 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call, call->offset = 0; call->unmarshall++; - case 4: + case 5: _debug("trailer"); if (skb->len != 0) return -EBADMSG; @@ -318,6 +414,14 @@ static int afs_deliver_fs_fetch_data(struct afs_call *call, if (!last) return 0; + if (call->count < PAGE_SIZE) { + _debug("clear"); + page = call->reply3; + buffer = kmap_atomic(page); + memset(buffer + call->count, 0, PAGE_SIZE - call->count); + kunmap_atomic(buffer); + } + _leave(" = 0 [done]"); return 0; } @@ -332,6 +436,56 @@ static const struct afs_call_type afs_RXFSFetchData = { .destructor = afs_flat_call_destructor, }; +static const struct afs_call_type afs_RXFSFetchData64 = { + .name = "FS.FetchData64", + .deliver = afs_deliver_fs_fetch_data, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +/* + * fetch data from a very large file + */ +static int afs_fs_fetch_data64(struct afs_server *server, + struct key *key, + struct afs_vnode *vnode, + off_t offset, size_t length, + struct page *buffer, + const struct afs_wait_mode *wait_mode) +{ + struct afs_call *call; + __be32 *bp; + + _enter(""); + + ASSERTCMP(length, <, ULONG_MAX); + + call = afs_alloc_flat_call(&afs_RXFSFetchData64, 32, (21 + 3 + 6) * 4); + if (!call) + return -ENOMEM; + + call->key = key; + call->reply = vnode; + call->reply2 = NULL; /* volsync */ + call->reply3 = buffer; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + call->operation_ID = FSFETCHDATA64; + + /* marshall the parameters */ + bp = call->request; + bp[0] = htonl(FSFETCHDATA64); + bp[1] = htonl(vnode->fid.vid); + bp[2] = htonl(vnode->fid.vnode); + bp[3] = htonl(vnode->fid.unique); + bp[4] = htonl(upper_32_bits(offset)); + bp[5] = htonl((u32) offset); + bp[6] = 0; + bp[7] = htonl((u32) length); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + /* * fetch data from a file */ @@ -345,6 +499,10 @@ int afs_fs_fetch_data(struct afs_server *server, struct afs_call *call; __be32 *bp; + if (upper_32_bits(offset) || upper_32_bits(offset + length)) + return afs_fs_fetch_data64(server, key, vnode, offset, length, + buffer, wait_mode); + _enter(""); call = afs_alloc_flat_call(&afs_RXFSFetchData, 24, (21 + 3 + 6) * 4); @@ -357,6 +515,7 @@ int afs_fs_fetch_data(struct afs_server *server, call->reply3 = buffer; call->service_id = FS_SERVICE; call->port = htons(AFS_FS_PORT); + call->operation_ID = FSFETCHDATA; /* marshall the parameters */ bp = call->request; @@ -476,8 +635,8 @@ static int afs_deliver_fs_create_vnode(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; xdr_decode_AFSFid(&bp, call->reply2); - xdr_decode_AFSFetchStatus(&bp, call->reply3, NULL); - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); + xdr_decode_AFSFetchStatus(&bp, call->reply3, NULL, NULL); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); xdr_decode_AFSCallBack_raw(&bp, call->reply4); /* xdr_decode_AFSVolSync(&bp, call->replyX); */ @@ -574,7 +733,7 @@ static int afs_deliver_fs_remove(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); /* xdr_decode_AFSVolSync(&bp, call->replyX); */ _leave(" = 0 [done]"); @@ -657,8 +816,8 @@ static int afs_deliver_fs_link(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); - xdr_decode_AFSFetchStatus(&bp, &dvnode->status, dvnode); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); + xdr_decode_AFSFetchStatus(&bp, &dvnode->status, dvnode, NULL); /* xdr_decode_AFSVolSync(&bp, call->replyX); */ _leave(" = 0 [done]"); @@ -746,8 +905,8 @@ static int afs_deliver_fs_symlink(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; xdr_decode_AFSFid(&bp, call->reply2); - xdr_decode_AFSFetchStatus(&bp, call->reply3, NULL); - xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode); + xdr_decode_AFSFetchStatus(&bp, call->reply3, NULL, NULL); + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, NULL); /* xdr_decode_AFSVolSync(&bp, call->replyX); */ _leave(" = 0 [done]"); @@ -852,9 +1011,10 @@ static int afs_deliver_fs_rename(struct afs_call *call, /* unmarshall the reply once we've received all of it */ bp = call->buffer; - xdr_decode_AFSFetchStatus(&bp, &orig_dvnode->status, orig_dvnode); + xdr_decode_AFSFetchStatus(&bp, &orig_dvnode->status, orig_dvnode, NULL); if (new_dvnode != orig_dvnode) - xdr_decode_AFSFetchStatus(&bp, &new_dvnode->status, new_dvnode); + xdr_decode_AFSFetchStatus(&bp, &new_dvnode->status, new_dvnode, + NULL); /* xdr_decode_AFSVolSync(&bp, call->replyX); */ _leave(" = 0 [done]"); @@ -936,3 +1096,816 @@ int afs_fs_rename(struct afs_server *server, return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); } + +/* + * deliver reply data to an FS.StoreData + */ +static int afs_deliver_fs_store_data(struct afs_call *call, + struct sk_buff *skb, bool last) +{ + struct afs_vnode *vnode = call->reply; + const __be32 *bp; + + _enter(",,%u", last); + + afs_transfer_reply(call, skb); + if (!last) { + _leave(" = 0 [more]"); + return 0; + } + + if (call->reply_size != call->reply_max) { + _leave(" = -EBADMSG [%u != %u]", + call->reply_size, call->reply_max); + return -EBADMSG; + } + + /* unmarshall the reply once we've received all of it */ + bp = call->buffer; + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, + &call->store_version); + /* xdr_decode_AFSVolSync(&bp, call->replyX); */ + + afs_pages_written_back(vnode, call); + + _leave(" = 0 [done]"); + return 0; +} + +/* + * FS.StoreData operation type + */ +static const struct afs_call_type afs_RXFSStoreData = { + .name = "FS.StoreData", + .deliver = afs_deliver_fs_store_data, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +static const struct afs_call_type afs_RXFSStoreData64 = { + .name = "FS.StoreData64", + .deliver = afs_deliver_fs_store_data, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +/* + * store a set of pages to a very large file + */ +static int afs_fs_store_data64(struct afs_server *server, + struct afs_writeback *wb, + pgoff_t first, pgoff_t last, + unsigned offset, unsigned to, + loff_t size, loff_t pos, loff_t i_size, + const struct afs_wait_mode *wait_mode) +{ + struct afs_vnode *vnode = wb->vnode; + struct afs_call *call; + __be32 *bp; + + _enter(",%x,{%x:%u},,", + key_serial(wb->key), vnode->fid.vid, vnode->fid.vnode); + + call = afs_alloc_flat_call(&afs_RXFSStoreData64, + (4 + 6 + 3 * 2) * 4, + (21 + 6) * 4); + if (!call) + return -ENOMEM; + + call->wb = wb; + call->key = wb->key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + call->mapping = vnode->vfs_inode.i_mapping; + call->first = first; + call->last = last; + call->first_offset = offset; + call->last_to = to; + call->send_pages = true; + call->store_version = vnode->status.data_version + 1; + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSSTOREDATA64); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + + *bp++ = 0; /* mask */ + *bp++ = 0; /* mtime */ + *bp++ = 0; /* owner */ + *bp++ = 0; /* group */ + *bp++ = 0; /* unix mode */ + *bp++ = 0; /* segment size */ + + *bp++ = htonl(pos >> 32); + *bp++ = htonl((u32) pos); + *bp++ = htonl(size >> 32); + *bp++ = htonl((u32) size); + *bp++ = htonl(i_size >> 32); + *bp++ = htonl((u32) i_size); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + +/* + * store a set of pages + */ +int afs_fs_store_data(struct afs_server *server, struct afs_writeback *wb, + pgoff_t first, pgoff_t last, + unsigned offset, unsigned to, + const struct afs_wait_mode *wait_mode) +{ + struct afs_vnode *vnode = wb->vnode; + struct afs_call *call; + loff_t size, pos, i_size; + __be32 *bp; + + _enter(",%x,{%x:%u},,", + key_serial(wb->key), vnode->fid.vid, vnode->fid.vnode); + + size = to - offset; + if (first != last) + size += (loff_t)(last - first) << PAGE_SHIFT; + pos = (loff_t)first << PAGE_SHIFT; + pos += offset; + + i_size = i_size_read(&vnode->vfs_inode); + if (pos + size > i_size) + i_size = size + pos; + + _debug("size %llx, at %llx, i_size %llx", + (unsigned long long) size, (unsigned long long) pos, + (unsigned long long) i_size); + + if (pos >> 32 || i_size >> 32 || size >> 32 || (pos + size) >> 32) + return afs_fs_store_data64(server, wb, first, last, offset, to, + size, pos, i_size, wait_mode); + + call = afs_alloc_flat_call(&afs_RXFSStoreData, + (4 + 6 + 3) * 4, + (21 + 6) * 4); + if (!call) + return -ENOMEM; + + call->wb = wb; + call->key = wb->key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + call->mapping = vnode->vfs_inode.i_mapping; + call->first = first; + call->last = last; + call->first_offset = offset; + call->last_to = to; + call->send_pages = true; + call->store_version = vnode->status.data_version + 1; + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSSTOREDATA); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + + *bp++ = 0; /* mask */ + *bp++ = 0; /* mtime */ + *bp++ = 0; /* owner */ + *bp++ = 0; /* group */ + *bp++ = 0; /* unix mode */ + *bp++ = 0; /* segment size */ + + *bp++ = htonl(pos); + *bp++ = htonl(size); + *bp++ = htonl(i_size); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + +/* + * deliver reply data to an FS.StoreStatus + */ +static int afs_deliver_fs_store_status(struct afs_call *call, + struct sk_buff *skb, bool last) +{ + afs_dataversion_t *store_version; + struct afs_vnode *vnode = call->reply; + const __be32 *bp; + + _enter(",,%u", last); + + afs_transfer_reply(call, skb); + if (!last) { + _leave(" = 0 [more]"); + return 0; + } + + if (call->reply_size != call->reply_max) { + _leave(" = -EBADMSG [%u != %u]", + call->reply_size, call->reply_max); + return -EBADMSG; + } + + /* unmarshall the reply once we've received all of it */ + store_version = NULL; + if (call->operation_ID == FSSTOREDATA) + store_version = &call->store_version; + + bp = call->buffer; + xdr_decode_AFSFetchStatus(&bp, &vnode->status, vnode, store_version); + /* xdr_decode_AFSVolSync(&bp, call->replyX); */ + + _leave(" = 0 [done]"); + return 0; +} + +/* + * FS.StoreStatus operation type + */ +static const struct afs_call_type afs_RXFSStoreStatus = { + .name = "FS.StoreStatus", + .deliver = afs_deliver_fs_store_status, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +static const struct afs_call_type afs_RXFSStoreData_as_Status = { + .name = "FS.StoreData", + .deliver = afs_deliver_fs_store_status, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +static const struct afs_call_type afs_RXFSStoreData64_as_Status = { + .name = "FS.StoreData64", + .deliver = afs_deliver_fs_store_status, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +/* + * set the attributes on a very large file, using FS.StoreData rather than + * FS.StoreStatus so as to alter the file size also + */ +static int afs_fs_setattr_size64(struct afs_server *server, struct key *key, + struct afs_vnode *vnode, struct iattr *attr, + const struct afs_wait_mode *wait_mode) +{ + struct afs_call *call; + __be32 *bp; + + _enter(",%x,{%x:%u},,", + key_serial(key), vnode->fid.vid, vnode->fid.vnode); + + ASSERT(attr->ia_valid & ATTR_SIZE); + + call = afs_alloc_flat_call(&afs_RXFSStoreData64_as_Status, + (4 + 6 + 3 * 2) * 4, + (21 + 6) * 4); + if (!call) + return -ENOMEM; + + call->key = key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + call->store_version = vnode->status.data_version + 1; + call->operation_ID = FSSTOREDATA; + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSSTOREDATA64); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + + xdr_encode_AFS_StoreStatus(&bp, attr); + + *bp++ = 0; /* position of start of write */ + *bp++ = 0; + *bp++ = 0; /* size of write */ + *bp++ = 0; + *bp++ = htonl(attr->ia_size >> 32); /* new file length */ + *bp++ = htonl((u32) attr->ia_size); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + +/* + * set the attributes on a file, using FS.StoreData rather than FS.StoreStatus + * so as to alter the file size also + */ +static int afs_fs_setattr_size(struct afs_server *server, struct key *key, + struct afs_vnode *vnode, struct iattr *attr, + const struct afs_wait_mode *wait_mode) +{ + struct afs_call *call; + __be32 *bp; + + _enter(",%x,{%x:%u},,", + key_serial(key), vnode->fid.vid, vnode->fid.vnode); + + ASSERT(attr->ia_valid & ATTR_SIZE); + if (attr->ia_size >> 32) + return afs_fs_setattr_size64(server, key, vnode, attr, + wait_mode); + + call = afs_alloc_flat_call(&afs_RXFSStoreData_as_Status, + (4 + 6 + 3) * 4, + (21 + 6) * 4); + if (!call) + return -ENOMEM; + + call->key = key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + call->store_version = vnode->status.data_version + 1; + call->operation_ID = FSSTOREDATA; + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSSTOREDATA); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + + xdr_encode_AFS_StoreStatus(&bp, attr); + + *bp++ = 0; /* position of start of write */ + *bp++ = 0; /* size of write */ + *bp++ = htonl(attr->ia_size); /* new file length */ + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + +/* + * set the attributes on a file, using FS.StoreData if there's a change in file + * size, and FS.StoreStatus otherwise + */ +int afs_fs_setattr(struct afs_server *server, struct key *key, + struct afs_vnode *vnode, struct iattr *attr, + const struct afs_wait_mode *wait_mode) +{ + struct afs_call *call; + __be32 *bp; + + if (attr->ia_valid & ATTR_SIZE) + return afs_fs_setattr_size(server, key, vnode, attr, + wait_mode); + + _enter(",%x,{%x:%u},,", + key_serial(key), vnode->fid.vid, vnode->fid.vnode); + + call = afs_alloc_flat_call(&afs_RXFSStoreStatus, + (4 + 6) * 4, + (21 + 6) * 4); + if (!call) + return -ENOMEM; + + call->key = key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + call->operation_ID = FSSTORESTATUS; + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSSTORESTATUS); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + + xdr_encode_AFS_StoreStatus(&bp, attr); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + +/* + * deliver reply data to an FS.GetVolumeStatus + */ +static int afs_deliver_fs_get_volume_status(struct afs_call *call, + struct sk_buff *skb, bool last) +{ + const __be32 *bp; + char *p; + int ret; + + _enter("{%u},{%u},%d", call->unmarshall, skb->len, last); + + switch (call->unmarshall) { + case 0: + call->offset = 0; + call->unmarshall++; + + /* extract the returned status record */ + case 1: + _debug("extract status"); + ret = afs_extract_data(call, skb, last, call->buffer, + 12 * 4); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + + bp = call->buffer; + xdr_decode_AFSFetchVolumeStatus(&bp, call->reply2); + call->offset = 0; + call->unmarshall++; + + /* extract the volume name length */ + case 2: + ret = afs_extract_data(call, skb, last, &call->tmp, 4); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + + call->count = ntohl(call->tmp); + _debug("volname length: %u", call->count); + if (call->count >= AFSNAMEMAX) + return -EBADMSG; + call->offset = 0; + call->unmarshall++; + + /* extract the volume name */ + case 3: + _debug("extract volname"); + if (call->count > 0) { + ret = afs_extract_data(call, skb, last, call->reply3, + call->count); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + } + + p = call->reply3; + p[call->count] = 0; + _debug("volname '%s'", p); + + call->offset = 0; + call->unmarshall++; + + /* extract the volume name padding */ + if ((call->count & 3) == 0) { + call->unmarshall++; + goto no_volname_padding; + } + call->count = 4 - (call->count & 3); + + case 4: + ret = afs_extract_data(call, skb, last, call->buffer, + call->count); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + + call->offset = 0; + call->unmarshall++; + no_volname_padding: + + /* extract the offline message length */ + case 5: + ret = afs_extract_data(call, skb, last, &call->tmp, 4); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + + call->count = ntohl(call->tmp); + _debug("offline msg length: %u", call->count); + if (call->count >= AFSNAMEMAX) + return -EBADMSG; + call->offset = 0; + call->unmarshall++; + + /* extract the offline message */ + case 6: + _debug("extract offline"); + if (call->count > 0) { + ret = afs_extract_data(call, skb, last, call->reply3, + call->count); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + } + + p = call->reply3; + p[call->count] = 0; + _debug("offline '%s'", p); + + call->offset = 0; + call->unmarshall++; + + /* extract the offline message padding */ + if ((call->count & 3) == 0) { + call->unmarshall++; + goto no_offline_padding; + } + call->count = 4 - (call->count & 3); + + case 7: + ret = afs_extract_data(call, skb, last, call->buffer, + call->count); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + + call->offset = 0; + call->unmarshall++; + no_offline_padding: + + /* extract the message of the day length */ + case 8: + ret = afs_extract_data(call, skb, last, &call->tmp, 4); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + + call->count = ntohl(call->tmp); + _debug("motd length: %u", call->count); + if (call->count >= AFSNAMEMAX) + return -EBADMSG; + call->offset = 0; + call->unmarshall++; + + /* extract the message of the day */ + case 9: + _debug("extract motd"); + if (call->count > 0) { + ret = afs_extract_data(call, skb, last, call->reply3, + call->count); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + } + + p = call->reply3; + p[call->count] = 0; + _debug("motd '%s'", p); + + call->offset = 0; + call->unmarshall++; + + /* extract the message of the day padding */ + if ((call->count & 3) == 0) { + call->unmarshall++; + goto no_motd_padding; + } + call->count = 4 - (call->count & 3); + + case 10: + ret = afs_extract_data(call, skb, last, call->buffer, + call->count); + switch (ret) { + case 0: break; + case -EAGAIN: return 0; + default: return ret; + } + + call->offset = 0; + call->unmarshall++; + no_motd_padding: + + case 11: + _debug("trailer %d", skb->len); + if (skb->len != 0) + return -EBADMSG; + break; + } + + if (!last) + return 0; + + _leave(" = 0 [done]"); + return 0; +} + +/* + * destroy an FS.GetVolumeStatus call + */ +static void afs_get_volume_status_call_destructor(struct afs_call *call) +{ + kfree(call->reply3); + call->reply3 = NULL; + afs_flat_call_destructor(call); +} + +/* + * FS.GetVolumeStatus operation type + */ +static const struct afs_call_type afs_RXFSGetVolumeStatus = { + .name = "FS.GetVolumeStatus", + .deliver = afs_deliver_fs_get_volume_status, + .abort_to_error = afs_abort_to_error, + .destructor = afs_get_volume_status_call_destructor, +}; + +/* + * fetch the status of a volume + */ +int afs_fs_get_volume_status(struct afs_server *server, + struct key *key, + struct afs_vnode *vnode, + struct afs_volume_status *vs, + const struct afs_wait_mode *wait_mode) +{ + struct afs_call *call; + __be32 *bp; + void *tmpbuf; + + _enter(""); + + tmpbuf = kmalloc(AFSOPAQUEMAX, GFP_KERNEL); + if (!tmpbuf) + return -ENOMEM; + + call = afs_alloc_flat_call(&afs_RXFSGetVolumeStatus, 2 * 4, 12 * 4); + if (!call) { + kfree(tmpbuf); + return -ENOMEM; + } + + call->key = key; + call->reply = vnode; + call->reply2 = vs; + call->reply3 = tmpbuf; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + + /* marshall the parameters */ + bp = call->request; + bp[0] = htonl(FSGETVOLUMESTATUS); + bp[1] = htonl(vnode->fid.vid); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + +/* + * deliver reply data to an FS.SetLock, FS.ExtendLock or FS.ReleaseLock + */ +static int afs_deliver_fs_xxxx_lock(struct afs_call *call, + struct sk_buff *skb, bool last) +{ + const __be32 *bp; + + _enter("{%u},{%u},%d", call->unmarshall, skb->len, last); + + afs_transfer_reply(call, skb); + if (!last) + return 0; + + if (call->reply_size != call->reply_max) + return -EBADMSG; + + /* unmarshall the reply once we've received all of it */ + bp = call->buffer; + /* xdr_decode_AFSVolSync(&bp, call->replyX); */ + + _leave(" = 0 [done]"); + return 0; +} + +/* + * FS.SetLock operation type + */ +static const struct afs_call_type afs_RXFSSetLock = { + .name = "FS.SetLock", + .deliver = afs_deliver_fs_xxxx_lock, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +/* + * FS.ExtendLock operation type + */ +static const struct afs_call_type afs_RXFSExtendLock = { + .name = "FS.ExtendLock", + .deliver = afs_deliver_fs_xxxx_lock, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +/* + * FS.ReleaseLock operation type + */ +static const struct afs_call_type afs_RXFSReleaseLock = { + .name = "FS.ReleaseLock", + .deliver = afs_deliver_fs_xxxx_lock, + .abort_to_error = afs_abort_to_error, + .destructor = afs_flat_call_destructor, +}; + +/* + * get a lock on a file + */ +int afs_fs_set_lock(struct afs_server *server, + struct key *key, + struct afs_vnode *vnode, + afs_lock_type_t type, + const struct afs_wait_mode *wait_mode) +{ + struct afs_call *call; + __be32 *bp; + + _enter(""); + + call = afs_alloc_flat_call(&afs_RXFSSetLock, 5 * 4, 6 * 4); + if (!call) + return -ENOMEM; + + call->key = key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSSETLOCK); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + *bp++ = htonl(type); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + +/* + * extend a lock on a file + */ +int afs_fs_extend_lock(struct afs_server *server, + struct key *key, + struct afs_vnode *vnode, + const struct afs_wait_mode *wait_mode) +{ + struct afs_call *call; + __be32 *bp; + + _enter(""); + + call = afs_alloc_flat_call(&afs_RXFSExtendLock, 4 * 4, 6 * 4); + if (!call) + return -ENOMEM; + + call->key = key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSEXTENDLOCK); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} + +/* + * release a lock on a file + */ +int afs_fs_release_lock(struct afs_server *server, + struct key *key, + struct afs_vnode *vnode, + const struct afs_wait_mode *wait_mode) +{ + struct afs_call *call; + __be32 *bp; + + _enter(""); + + call = afs_alloc_flat_call(&afs_RXFSReleaseLock, 4 * 4, 6 * 4); + if (!call) + return -ENOMEM; + + call->key = key; + call->reply = vnode; + call->service_id = FS_SERVICE; + call->port = htons(AFS_FS_PORT); + + /* marshall the parameters */ + bp = call->request; + *bp++ = htonl(FSRELEASELOCK); + *bp++ = htonl(vnode->fid.vid); + *bp++ = htonl(vnode->fid.vnode); + *bp++ = htonl(vnode->fid.unique); + + return afs_make_call(&server->addr, call, GFP_NOFS, wait_mode); +} diff --git a/fs/afs/inode.c b/fs/afs/inode.c index c184a4ee599..29467128844 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -8,7 +8,7 @@ * along with this program; if not, write to the Free Software * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. * - * Authors: David Woodhouse <dwmw2@cambridge.redhat.com> + * Authors: David Woodhouse <dwmw2@infradead.org> * David Howells <dhowells@redhat.com> * */ @@ -16,9 +16,11 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> -#include <linux/slab.h> #include <linux/fs.h> #include <linux/pagemap.h> +#include <linux/sched.h> +#include <linux/mount.h> +#include <linux/namei.h> #include "internal.h" struct afs_iget_data { @@ -60,15 +62,21 @@ static int afs_inode_map_status(struct afs_vnode *vnode, struct key *key) return -EBADMSG; } - inode->i_nlink = vnode->status.nlink; +#ifdef CONFIG_AFS_FSCACHE + if (vnode->status.size != inode->i_size) + fscache_attr_changed(vnode->cache); +#endif + + set_nlink(inode, vnode->status.nlink); inode->i_uid = vnode->status.owner; - inode->i_gid = 0; + inode->i_gid = GLOBAL_ROOT_GID; inode->i_size = vnode->status.size; inode->i_ctime.tv_sec = vnode->status.mtime_server; inode->i_ctime.tv_nsec = 0; inode->i_atime = inode->i_mtime = inode->i_ctime; inode->i_blocks = 0; - inode->i_version = vnode->fid.unique; + inode->i_generation = vnode->fid.unique; + inode->i_version = vnode->status.data_version; inode->i_mapping->a_ops = &afs_fs_aops; /* check to see whether a symbolic link is really a mountpoint */ @@ -93,7 +101,17 @@ static int afs_iget5_test(struct inode *inode, void *opaque) struct afs_iget_data *data = opaque; return inode->i_ino == data->fid.vnode && - inode->i_version == data->fid.unique; + inode->i_generation == data->fid.unique; +} + +/* + * iget5() comparator for inode created by autocell operations + * + * These pseudo inodes don't match anything. + */ +static int afs_iget5_autocell_test(struct inode *inode, void *opaque) +{ + return 0; } /* @@ -105,7 +123,7 @@ static int afs_iget5_set(struct inode *inode, void *opaque) struct afs_vnode *vnode = AFS_FS_I(inode); inode->i_ino = data->fid.vnode; - inode->i_version = data->fid.unique; + inode->i_generation = data->fid.unique; vnode->fid = data->fid; vnode->volume = data->volume; @@ -113,6 +131,68 @@ static int afs_iget5_set(struct inode *inode, void *opaque) } /* + * inode retrieval for autocell + */ +struct inode *afs_iget_autocell(struct inode *dir, const char *dev_name, + int namesz, struct key *key) +{ + struct afs_iget_data data; + struct afs_super_info *as; + struct afs_vnode *vnode; + struct super_block *sb; + struct inode *inode; + static atomic_t afs_autocell_ino; + + _enter("{%x:%u},%*.*s,", + AFS_FS_I(dir)->fid.vid, AFS_FS_I(dir)->fid.vnode, + namesz, namesz, dev_name ?: ""); + + sb = dir->i_sb; + as = sb->s_fs_info; + data.volume = as->volume; + data.fid.vid = as->volume->vid; + data.fid.unique = 0; + data.fid.vnode = 0; + + inode = iget5_locked(sb, atomic_inc_return(&afs_autocell_ino), + afs_iget5_autocell_test, afs_iget5_set, + &data); + if (!inode) { + _leave(" = -ENOMEM"); + return ERR_PTR(-ENOMEM); + } + + _debug("GOT INODE %p { ino=%lu, vl=%x, vn=%x, u=%x }", + inode, inode->i_ino, data.fid.vid, data.fid.vnode, + data.fid.unique); + + vnode = AFS_FS_I(inode); + + /* there shouldn't be an existing inode */ + BUG_ON(!(inode->i_state & I_NEW)); + + inode->i_size = 0; + inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO; + inode->i_op = &afs_autocell_inode_operations; + set_nlink(inode, 2); + inode->i_uid = GLOBAL_ROOT_UID; + inode->i_gid = GLOBAL_ROOT_GID; + inode->i_ctime.tv_sec = get_seconds(); + inode->i_ctime.tv_nsec = 0; + inode->i_atime = inode->i_mtime = inode->i_ctime; + inode->i_blocks = 0; + inode->i_version = 0; + inode->i_generation = 0; + + set_bit(AFS_VNODE_PSEUDODIR, &vnode->flags); + set_bit(AFS_VNODE_MOUNTPOINT, &vnode->flags); + inode->i_flags |= S_AUTOMOUNT | S_NOATIME; + unlock_new_inode(inode); + _leave(" = %p", inode); + return inode; +} + +/* * inode retrieval */ struct inode *afs_iget(struct super_block *sb, struct key *key, @@ -125,7 +205,7 @@ struct inode *afs_iget(struct super_block *sb, struct key *key, struct inode *inode; int ret; - _enter(",{%u,%u,%u},,", fid->vid, fid->vnode, fid->unique); + _enter(",{%x:%u.%u},,", fid->vid, fid->vnode, fid->unique); as = sb->s_fs_info; data.volume = as->volume; @@ -148,15 +228,6 @@ struct inode *afs_iget(struct super_block *sb, struct key *key, return inode; } -#ifdef AFS_CACHING_SUPPORT - /* set up caching before reading the status, as fetch-status reads the - * first page of symlinks to see if they're really mntpts */ - cachefs_acquire_cookie(vnode->volume->cache, - NULL, - vnode, - &vnode->cache); -#endif - if (!status) { /* it's a remotely extant inode */ set_bit(AFS_VNODE_CB_BROKEN, &vnode->flags); @@ -182,6 +253,15 @@ struct inode *afs_iget(struct super_block *sb, struct key *key, } } + /* set up caching before mapping the status, as map-status reads the + * first page of symlinks to see if they're really mountpoints */ + inode->i_size = vnode->status.size; +#ifdef CONFIG_AFS_FSCACHE + vnode->cache = fscache_acquire_cookie(vnode->volume->cache, + &afs_vnode_cache_index_def, + vnode, true); +#endif + ret = afs_inode_map_status(vnode, key); if (ret < 0) goto bad_inode; @@ -195,15 +275,33 @@ struct inode *afs_iget(struct super_block *sb, struct key *key, /* failure */ bad_inode: - make_bad_inode(inode); - unlock_new_inode(inode); - iput(inode); - +#ifdef CONFIG_AFS_FSCACHE + fscache_relinquish_cookie(vnode->cache, 0); + vnode->cache = NULL; +#endif + iget_failed(inode); _leave(" = %d [bad]", ret); return ERR_PTR(ret); } /* + * mark the data attached to an inode as obsolete due to a write on the server + * - might also want to ditch all the outstanding writes and dirty pages + */ +void afs_zap_data(struct afs_vnode *vnode) +{ + _enter("{%x:%u}", vnode->fid.vid, vnode->fid.vnode); + + /* nuke all the non-dirty pages that aren't locked, mapped or being + * written back in a regular file and completely discard the pages in a + * directory or symlink */ + if (S_ISREG(vnode->vfs_inode.i_mode)) + invalidate_remote_inode(&vnode->vfs_inode); + else + invalidate_inode_pages2(vnode->vfs_inode.i_mapping); +} + +/* * validate a vnode/inode * - there are several things we need to check * - parent dir data changes (rm, rmdir, rename, mkdir, create, link, @@ -258,10 +356,8 @@ int afs_validate(struct afs_vnode *vnode, struct key *key) /* if the vnode's data version number changed then its contents are * different */ - if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) { - _debug("zap data {%x:%d}", vnode->fid.vid, vnode->fid.vnode); - invalidate_remote_inode(&vnode->vfs_inode); - } + if (test_and_clear_bit(AFS_VNODE_ZAP_DATA, &vnode->flags)) + afs_zap_data(vnode); clear_bit(AFS_VNODE_MODIFIED, &vnode->flags); mutex_unlock(&vnode->validate_lock); @@ -278,30 +374,43 @@ error_unlock: /* * read the attributes of an inode */ -int afs_inode_getattr(struct vfsmount *mnt, struct dentry *dentry, +int afs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { struct inode *inode; inode = dentry->d_inode; - _enter("{ ino=%lu v=%lu }", inode->i_ino, inode->i_version); + _enter("{ ino=%lu v=%u }", inode->i_ino, inode->i_generation); generic_fillattr(inode, stat); return 0; } /* + * discard an AFS inode + */ +int afs_drop_inode(struct inode *inode) +{ + _enter(""); + + if (test_bit(AFS_VNODE_PSEUDODIR, &AFS_FS_I(inode)->flags)) + return generic_delete_inode(inode); + else + return generic_drop_inode(inode); +} + +/* * clear an AFS inode */ -void afs_clear_inode(struct inode *inode) +void afs_evict_inode(struct inode *inode) { struct afs_permits *permits; struct afs_vnode *vnode; vnode = AFS_FS_I(inode); - _enter("{%x:%d.%d} v=%u x=%u t=%u }", + _enter("{%x:%u.%d} v=%u x=%u t=%u }", vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique, @@ -313,6 +422,9 @@ void afs_clear_inode(struct inode *inode) ASSERTCMP(inode->i_ino, ==, vnode->fid.vnode); + truncate_inode_pages_final(&inode->i_data); + clear_inode(inode); + afs_give_up_callback(vnode); if (vnode->server) { @@ -323,10 +435,11 @@ void afs_clear_inode(struct inode *inode) vnode->server = NULL; } + ASSERT(list_empty(&vnode->writebacks)); ASSERT(!vnode->cb_promised); -#ifdef AFS_CACHING_SUPPORT - cachefs_relinquish_cookie(vnode->cache, 0); +#ifdef CONFIG_AFS_FSCACHE + fscache_relinquish_cookie(vnode->cache, 0); vnode->cache = NULL; #endif @@ -339,3 +452,47 @@ void afs_clear_inode(struct inode *inode) _leave(""); } + +/* + * set the attributes of an inode + */ +int afs_setattr(struct dentry *dentry, struct iattr *attr) +{ + struct afs_vnode *vnode = AFS_FS_I(dentry->d_inode); + struct key *key; + int ret; + + _enter("{%x:%u},{n=%s},%x", + vnode->fid.vid, vnode->fid.vnode, dentry->d_name.name, + attr->ia_valid); + + if (!(attr->ia_valid & (ATTR_SIZE | ATTR_MODE | ATTR_UID | ATTR_GID | + ATTR_MTIME))) { + _leave(" = 0 [unsupported]"); + return 0; + } + + /* flush any dirty data outstanding on a regular file */ + if (S_ISREG(vnode->vfs_inode.i_mode)) { + filemap_write_and_wait(vnode->vfs_inode.i_mapping); + afs_writeback_all(vnode); + } + + if (attr->ia_valid & ATTR_FILE) { + key = attr->ia_file->private_data; + } else { + key = afs_request_key(vnode->volume->cell); + if (IS_ERR(key)) { + ret = PTR_ERR(key); + goto error; + } + } + + ret = afs_vnode_setattr(vnode, key, attr); + if (!(attr->ia_valid & ATTR_FILE)) + key_put(key); + +error: + _leave(" = %d", ret); + return ret; +} diff --git a/fs/afs/internal.h b/fs/afs/internal.h index d90c158cd93..71d5982312f 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -16,11 +16,17 @@ #include <linux/skbuff.h> #include <linux/rxrpc.h> #include <linux/key.h> +#include <linux/workqueue.h> +#include <linux/sched.h> +#include <linux/fscache.h> +#include <linux/backing-dev.h> + #include "afs.h" #include "afs_vl.h" #define AFS_CELL_MAX_ADDRS 15 +struct pagevec; struct afs_call; typedef enum { @@ -36,6 +42,7 @@ typedef enum { struct afs_mount_params { bool rwpath; /* T if the parent should be considered R/W */ bool force; /* T to force cell type */ + bool autocell; /* T if set auto mount operation */ afs_voltype_t type; /* type of volume requested */ int volnamesz; /* size of volume name */ const char *volname; /* name of volume to mount */ @@ -68,6 +75,7 @@ struct afs_call { const struct afs_call_type *type; /* type of call */ const struct afs_wait_mode *wait_mode; /* completion wait mode */ wait_queue_head_t waitq; /* processes awaiting completion */ + void (*async_workfn)(struct afs_call *call); /* asynchronous work function */ struct work_struct async_work; /* asynchronous work processor */ struct work_struct work; /* actual work processor */ struct sk_buff_head rx_queue; /* received packets */ @@ -75,12 +83,15 @@ struct afs_call { struct key *key; /* security for this call */ struct afs_server *server; /* server affected by incoming CM call */ void *request; /* request data (first part) */ - void *request2; /* request data (second part) */ + struct address_space *mapping; /* page set */ + struct afs_writeback *wb; /* writeback being performed */ void *buffer; /* reply receive buffer */ void *reply; /* reply buffer (first part) */ void *reply2; /* reply buffer (second part) */ void *reply3; /* reply buffer (third part) */ void *reply4; /* reply buffer (fourth part) */ + pgoff_t first; /* first page in mapping to deal with */ + pgoff_t last; /* last page in mapping to deal with */ enum { /* call state */ AFS_CALL_REQUESTING, /* request is being sent for outgoing call */ AFS_CALL_AWAIT_REPLY, /* awaiting reply to outgoing call */ @@ -97,14 +108,18 @@ struct afs_call { unsigned request_size; /* size of request data */ unsigned reply_max; /* maximum size of reply */ unsigned reply_size; /* current size of reply */ - unsigned short offset; /* offset into received data store */ + unsigned first_offset; /* offset into mapping[first] */ + unsigned last_to; /* amount of mapping[last] */ + unsigned offset; /* offset into received data store */ unsigned char unmarshall; /* unmarshalling phase */ bool incoming; /* T if incoming call */ + bool send_pages; /* T if data from mapping should be sent */ u16 service_id; /* RxRPC service ID to call */ __be16 port; /* target UDP port */ __be32 operation_ID; /* operation ID for an incoming call */ u32 count; /* count for use in unmarshalling */ __be32 tmp; /* place to extract temporary data */ + afs_dataversion_t store_version; /* updated version expected from store */ }; struct afs_call_type { @@ -124,6 +139,32 @@ struct afs_call_type { }; /* + * record of an outstanding writeback on a vnode + */ +struct afs_writeback { + struct list_head link; /* link in vnode->writebacks */ + struct work_struct writer; /* work item to perform the writeback */ + struct afs_vnode *vnode; /* vnode to which this write applies */ + struct key *key; /* owner of this write */ + wait_queue_head_t waitq; /* completion and ready wait queue */ + pgoff_t first; /* first page in batch */ + pgoff_t point; /* last page in current store op */ + pgoff_t last; /* last page in batch (inclusive) */ + unsigned offset_first; /* offset into first page of start of write */ + unsigned to_last; /* offset into last page of end of write */ + int num_conflicts; /* count of conflicting writes in list */ + int usage; + bool conflicts; /* T if has dependent conflicts */ + enum { + AFS_WBACK_SYNCING, /* synchronisation being performed */ + AFS_WBACK_PENDING, /* write pending */ + AFS_WBACK_CONFLICTING, /* conflicting writes posted */ + AFS_WBACK_WRITING, /* writing back */ + AFS_WBACK_COMPLETE /* the writeback record has been unlinked */ + } state __attribute__((packed)); +}; + +/* * AFS superblock private data * - there's one superblock per volume */ @@ -155,9 +196,8 @@ struct afs_cell { struct list_head link; /* main cell list link */ struct key *anonymous_key; /* anonymous user key for this cell */ struct list_head proc_link; /* /proc cell list link */ - struct proc_dir_entry *proc_dir; /* /proc dir for this cell */ -#ifdef AFS_CACHING_SUPPORT - struct cachefs_cookie *cache; /* caching cookie */ +#ifdef CONFIG_AFS_FSCACHE + struct fscache_cookie *cache; /* caching cookie */ #endif /* server record management */ @@ -212,8 +252,8 @@ struct afs_vlocation { struct list_head grave; /* link in master graveyard list */ struct list_head update; /* link in master update list */ struct afs_cell *cell; /* cell to which volume belongs */ -#ifdef AFS_CACHING_SUPPORT - struct cachefs_cookie *cache; /* caching cookie */ +#ifdef CONFIG_AFS_FSCACHE + struct fscache_cookie *cache; /* caching cookie */ #endif struct afs_cache_vlocation vldb; /* volume information DB record */ struct afs_volume *vols[3]; /* volume access record pointer (index by type) */ @@ -265,8 +305,8 @@ struct afs_volume { atomic_t usage; struct afs_cell *cell; /* cell to which belongs (unrefd ptr) */ struct afs_vlocation *vlocation; /* volume location */ -#ifdef AFS_CACHING_SUPPORT - struct cachefs_cookie *cache; /* caching cookie */ +#ifdef CONFIG_AFS_FSCACHE + struct fscache_cookie *cache; /* caching cookie */ #endif afs_volid_t vid; /* volume ID */ afs_voltype_t type; /* type of volume */ @@ -275,6 +315,7 @@ struct afs_volume { unsigned short rjservers; /* number of servers discarded due to -ENOMEDIUM */ struct afs_server *servers[8]; /* servers on which volume resides (ordered) */ struct rw_semaphore server_sem; /* lock for accessing current server */ + struct backing_dev_info bdi; }; /* @@ -296,8 +337,8 @@ struct afs_vnode { struct afs_server *server; /* server currently supplying this file */ struct afs_fid fid; /* the file identifier for this inode */ struct afs_file_status status; /* AFS status info for this file */ -#ifdef AFS_CACHING_SUPPORT - struct cachefs_cookie *cache; /* caching cookie */ +#ifdef CONFIG_AFS_FSCACHE + struct fscache_cookie *cache; /* caching cookie */ #endif struct afs_permits *permits; /* cache of permits so far obtained */ struct mutex permits_lock; /* lock for altering permits list */ @@ -305,6 +346,7 @@ struct afs_vnode { wait_queue_head_t update_waitq; /* status fetch waitqueue */ int update_cnt; /* number of outstanding ops that will update the * status */ + spinlock_t writeback_lock; /* lock for writebacks */ spinlock_t lock; /* waitqueue/flags lock */ unsigned long flags; #define AFS_VNODE_CB_BROKEN 0 /* set if vnode's callback was broken */ @@ -313,9 +355,21 @@ struct afs_vnode { #define AFS_VNODE_ZAP_DATA 3 /* set if vnode's data should be invalidated */ #define AFS_VNODE_DELETED 4 /* set if vnode deleted on server */ #define AFS_VNODE_MOUNTPOINT 5 /* set if vnode is a mountpoint symlink */ +#define AFS_VNODE_LOCKING 6 /* set if waiting for lock on vnode */ +#define AFS_VNODE_READLOCKED 7 /* set if vnode is read-locked on the server */ +#define AFS_VNODE_WRITELOCKED 8 /* set if vnode is write-locked on the server */ +#define AFS_VNODE_UNLOCKING 9 /* set if vnode is being unlocked on the server */ +#define AFS_VNODE_AUTOCELL 10 /* set if Vnode is an auto mount point */ +#define AFS_VNODE_PSEUDODIR 11 /* set if Vnode is a pseudo directory */ long acl_order; /* ACL check count (callback break count) */ + struct list_head writebacks; /* alterations in pagecache that need writing */ + struct list_head pending_locks; /* locks waiting to be granted */ + struct list_head granted_locks; /* locks granted on this file */ + struct delayed_work lock_work; /* work to be done in locking */ + struct key *unlock_key; /* key to be used in unlocking */ + /* outstanding callback notification on this file */ struct rb_node server_rb; /* link in server->fs_vnodes */ struct rb_node cb_promise; /* link in server->cb_promises */ @@ -380,6 +434,22 @@ struct afs_uuid { /*****************************************************************************/ /* + * cache.c + */ +#ifdef CONFIG_AFS_FSCACHE +extern struct fscache_netfs afs_cache_netfs; +extern struct fscache_cookie_def afs_cell_cache_index_def; +extern struct fscache_cookie_def afs_vlocation_cache_index_def; +extern struct fscache_cookie_def afs_volume_cache_index_def; +extern struct fscache_cookie_def afs_vnode_cache_index_def; +#else +#define afs_cell_cache_index_def (*(struct fscache_cookie_def *) NULL) +#define afs_vlocation_cache_index_def (*(struct fscache_cookie_def *) NULL) +#define afs_volume_cache_index_def (*(struct fscache_cookie_def *) NULL) +#define afs_vnode_cache_index_def (*(struct fscache_cookie_def *) NULL) +#endif + +/* * callback.c */ extern void afs_init_callback_state(struct afs_server *); @@ -398,14 +468,11 @@ extern void afs_callback_update_kill(void); */ extern struct rw_semaphore afs_proc_cells_sem; extern struct list_head afs_proc_cells; -#ifdef AFS_CACHING_SUPPORT -extern struct cachefs_index_def afs_cache_cell_index_def; -#endif #define afs_get_cell(C) do { atomic_inc(&(C)->usage); } while(0) extern int afs_cell_init(char *); -extern struct afs_cell *afs_cell_create(const char *, char *); -extern struct afs_cell *afs_cell_lookup(const char *, unsigned); +extern struct afs_cell *afs_cell_create(const char *, unsigned, char *, bool); +extern struct afs_cell *afs_cell_lookup(const char *, unsigned, bool); extern struct afs_cell *afs_grab_cell(struct afs_cell *); extern void afs_put_cell(struct afs_cell *); extern void afs_cell_purge(void); @@ -419,10 +486,9 @@ extern bool afs_cm_incoming_call(struct afs_call *); * dir.c */ extern const struct inode_operations afs_dir_inode_operations; +extern const struct dentry_operations afs_fs_dentry_operations; extern const struct file_operations afs_dir_file_operations; -extern int afs_permission(struct inode *, int, struct nameidata *); - /* * file.c */ @@ -432,10 +498,16 @@ extern const struct file_operations afs_file_operations; extern int afs_open(struct inode *, struct file *); extern int afs_release(struct inode *, struct file *); +extern int afs_page_filler(void *, struct page *); -#ifdef AFS_CACHING_SUPPORT -extern int afs_cache_get_page_cookie(struct page *, struct cachefs_page **); -#endif +/* + * flock.c + */ +extern void __exit afs_kill_lock_manager(void); +extern void afs_lock_work(struct work_struct *); +extern void afs_lock_may_be_available(struct afs_vnode *); +extern int afs_lock(struct file *, int, struct file_lock *); +extern int afs_flock(struct file *, int, struct file_lock *); /* * fsclient.c @@ -467,26 +539,46 @@ extern int afs_fs_rename(struct afs_server *, struct key *, struct afs_vnode *, const char *, struct afs_vnode *, const char *, const struct afs_wait_mode *); +extern int afs_fs_store_data(struct afs_server *, struct afs_writeback *, + pgoff_t, pgoff_t, unsigned, unsigned, + const struct afs_wait_mode *); +extern int afs_fs_setattr(struct afs_server *, struct key *, + struct afs_vnode *, struct iattr *, + const struct afs_wait_mode *); +extern int afs_fs_get_volume_status(struct afs_server *, struct key *, + struct afs_vnode *, + struct afs_volume_status *, + const struct afs_wait_mode *); +extern int afs_fs_set_lock(struct afs_server *, struct key *, + struct afs_vnode *, afs_lock_type_t, + const struct afs_wait_mode *); +extern int afs_fs_extend_lock(struct afs_server *, struct key *, + struct afs_vnode *, + const struct afs_wait_mode *); +extern int afs_fs_release_lock(struct afs_server *, struct key *, + struct afs_vnode *, + const struct afs_wait_mode *); /* * inode.c */ +extern struct inode *afs_iget_autocell(struct inode *, const char *, int, + struct key *); extern struct inode *afs_iget(struct super_block *, struct key *, struct afs_fid *, struct afs_file_status *, struct afs_callback *); +extern void afs_zap_data(struct afs_vnode *); extern int afs_validate(struct afs_vnode *, struct key *); -extern int afs_inode_getattr(struct vfsmount *, struct dentry *, - struct kstat *); -extern void afs_zap_permits(struct rcu_head *); -extern void afs_clear_inode(struct inode *); +extern int afs_getattr(struct vfsmount *, struct dentry *, struct kstat *); +extern int afs_setattr(struct dentry *, struct iattr *); +extern void afs_evict_inode(struct inode *); +extern int afs_drop_inode(struct inode *); /* * main.c */ +extern struct workqueue_struct *afs_wq; extern struct afs_uuid afs_uuid; -#ifdef AFS_CACHING_SUPPORT -extern struct cachefs_netfs afs_cache_netfs; -#endif /* * misc.c @@ -497,12 +589,12 @@ extern int afs_abort_to_error(u32); * mntpt.c */ extern const struct inode_operations afs_mntpt_inode_operations; +extern const struct inode_operations afs_autocell_inode_operations; extern const struct file_operations afs_mntpt_file_operations; -extern unsigned long afs_mntpt_expiry_timeout; +extern struct vfsmount *afs_d_automount(struct path *); extern int afs_mntpt_check_symlink(struct afs_vnode *, struct key *); extern void afs_mntpt_kill_timer(void); -extern void afs_umount_begin(struct vfsmount *, int); /* * proc.c @@ -533,8 +625,9 @@ extern int afs_extract_data(struct afs_call *, struct sk_buff *, bool, void *, */ extern void afs_clear_permits(struct afs_vnode *); extern void afs_cache_permit(struct afs_vnode *, struct key *, long); +extern void afs_zap_permits(struct rcu_head *); extern struct key *afs_request_key(struct afs_cell *); -extern int afs_permission(struct inode *, int, struct nameidata *); +extern int afs_permission(struct inode *, int); /* * server.c @@ -568,10 +661,6 @@ extern int afs_get_MAC_address(u8 *, size_t); /* * vlclient.c */ -#ifdef AFS_CACHING_SUPPORT -extern struct cachefs_index_def afs_vlocation_cache_index_def; -#endif - extern int afs_vl_get_entry_by_name(struct in_addr *, struct key *, const char *, struct afs_cache_vlocation *, const struct afs_wait_mode *); @@ -595,12 +684,6 @@ extern void afs_vlocation_purge(void); /* * vnode.c */ -#ifdef AFS_CACHING_SUPPORT -extern struct cachefs_index_def afs_vnode_cache_index_def; -#endif - -extern struct afs_timer_ops afs_vnode_cb_timed_out_ops; - static inline struct afs_vnode *AFS_FS_I(struct inode *inode) { return container_of(inode, struct afs_vnode, vfs_inode); @@ -629,14 +712,19 @@ extern int afs_vnode_symlink(struct afs_vnode *, struct key *, const char *, struct afs_file_status *, struct afs_server **); extern int afs_vnode_rename(struct afs_vnode *, struct afs_vnode *, struct key *, const char *, const char *); +extern int afs_vnode_store_data(struct afs_writeback *, pgoff_t, pgoff_t, + unsigned, unsigned); +extern int afs_vnode_setattr(struct afs_vnode *, struct key *, struct iattr *); +extern int afs_vnode_get_volume_status(struct afs_vnode *, struct key *, + struct afs_volume_status *); +extern int afs_vnode_set_lock(struct afs_vnode *, struct key *, + afs_lock_type_t); +extern int afs_vnode_extend_lock(struct afs_vnode *, struct key *); +extern int afs_vnode_release_lock(struct afs_vnode *, struct key *); /* * volume.c */ -#ifdef AFS_CACHING_SUPPORT -extern struct cachefs_index_def afs_volume_cache_index_def; -#endif - #define afs_get_volume(V) do { atomic_inc(&(V)->usage); } while(0) extern void afs_put_volume(struct afs_volume *); @@ -645,6 +733,25 @@ extern struct afs_server *afs_volume_pick_fileserver(struct afs_vnode *); extern int afs_volume_release_fileserver(struct afs_vnode *, struct afs_server *, int); +/* + * write.c + */ +extern int afs_set_page_dirty(struct page *); +extern void afs_put_writeback(struct afs_writeback *); +extern int afs_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata); +extern int afs_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata); +extern int afs_writepage(struct page *, struct writeback_control *); +extern int afs_writepages(struct address_space *, struct writeback_control *); +extern void afs_pages_written_back(struct afs_vnode *, struct afs_call *); +extern ssize_t afs_file_write(struct kiocb *, struct iov_iter *); +extern int afs_writeback_all(struct afs_vnode *); +extern int afs_fsync(struct file *, loff_t, loff_t, int); + + /*****************************************************************************/ /* * debug tracing @@ -652,16 +759,10 @@ extern int afs_volume_release_fileserver(struct afs_vnode *, extern unsigned afs_debug; #define dbgprintk(FMT,...) \ - printk("[%x%-6.6s] "FMT"\n", smp_processor_id(), current->comm ,##__VA_ARGS__) - -/* make sure we maintain the format strings, even when debugging is disabled */ -static inline __attribute__((format(printf,1,2))) -void _dbprintk(const char *fmt, ...) -{ -} + printk("[%-6.6s] "FMT"\n", current->comm ,##__VA_ARGS__) -#define kenter(FMT,...) dbgprintk("==> %s("FMT")",__FUNCTION__ ,##__VA_ARGS__) -#define kleave(FMT,...) dbgprintk("<== %s()"FMT"",__FUNCTION__ ,##__VA_ARGS__) +#define kenter(FMT,...) dbgprintk("==> %s("FMT")",__func__ ,##__VA_ARGS__) +#define kleave(FMT,...) dbgprintk("<== %s()"FMT"",__func__ ,##__VA_ARGS__) #define kdebug(FMT,...) dbgprintk(" "FMT ,##__VA_ARGS__) @@ -694,9 +795,9 @@ do { \ } while (0) #else -#define _enter(FMT,...) _dbprintk("==> %s("FMT")",__FUNCTION__ ,##__VA_ARGS__) -#define _leave(FMT,...) _dbprintk("<== %s()"FMT"",__FUNCTION__ ,##__VA_ARGS__) -#define _debug(FMT,...) _dbprintk(" "FMT ,##__VA_ARGS__) +#define _enter(FMT,...) no_printk("==> %s("FMT")",__func__ ,##__VA_ARGS__) +#define _leave(FMT,...) no_printk("<== %s()"FMT"",__func__ ,##__VA_ARGS__) +#define _debug(FMT,...) no_printk(" "FMT ,##__VA_ARGS__) #endif /* @@ -726,6 +827,21 @@ do { \ } \ } while(0) +#define ASSERTRANGE(L, OP1, N, OP2, H) \ +do { \ + if (unlikely(!((L) OP1 (N)) || !((N) OP2 (H)))) { \ + printk(KERN_ERR "\n"); \ + printk(KERN_ERR "AFS: Assertion failed\n"); \ + printk(KERN_ERR "%lu "#OP1" %lu "#OP2" %lu is false\n", \ + (unsigned long)(L), (unsigned long)(N), \ + (unsigned long)(H)); \ + printk(KERN_ERR "0x%lx "#OP1" 0x%lx "#OP2" 0x%lx is false\n", \ + (unsigned long)(L), (unsigned long)(N), \ + (unsigned long)(H)); \ + BUG(); \ + } \ +} while(0) + #define ASSERTIF(C, X) \ do { \ if (unlikely((C) && !(X))) { \ @@ -758,6 +874,10 @@ do { \ do { \ } while(0) +#define ASSERTRANGE(L, OP1, N, OP2, H) \ +do { \ +} while(0) + #define ASSERTIF(C, X) \ do { \ } while(0) diff --git a/fs/afs/main.c b/fs/afs/main.c index 80ec6fd19a7..35de0c04729 100644 --- a/fs/afs/main.c +++ b/fs/afs/main.c @@ -1,6 +1,6 @@ /* AFS client file system * - * Copyright (C) 2002 Red Hat, Inc. All Rights Reserved. + * Copyright (C) 2002,5 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) * * This program is free software; you can redistribute it and/or @@ -13,6 +13,7 @@ #include <linux/moduleparam.h> #include <linux/init.h> #include <linux/completion.h> +#include <linux/sched.h> #include "internal.h" MODULE_DESCRIPTION("AFS Client File System"); @@ -21,26 +22,15 @@ MODULE_LICENSE("GPL"); unsigned afs_debug; module_param_named(debug, afs_debug, uint, S_IWUSR | S_IRUGO); -MODULE_PARM_DESC(afs_debug, "AFS debugging mask"); +MODULE_PARM_DESC(debug, "AFS debugging mask"); static char *rootcell; module_param(rootcell, charp, 0); MODULE_PARM_DESC(rootcell, "root AFS cell name and VL server IP addr list"); -#ifdef AFS_CACHING_SUPPORT -static struct cachefs_netfs_operations afs_cache_ops = { - .get_page_cookie = afs_cache_get_page_cookie, -}; - -struct cachefs_netfs afs_cache_netfs = { - .name = "afs", - .version = 0, - .ops = &afs_cache_ops, -}; -#endif - struct afs_uuid afs_uuid; +struct workqueue_struct *afs_wq; /* * get a client UUID @@ -65,13 +55,13 @@ static int __init afs_get_client_UUID(void) afs_uuid.time_low = uuidtime; afs_uuid.time_mid = uuidtime >> 32; afs_uuid.time_hi_and_version = (uuidtime >> 48) & AFS_UUID_TIMEHI_MASK; - afs_uuid.time_hi_and_version = AFS_UUID_VERSION_TIME; + afs_uuid.time_hi_and_version |= AFS_UUID_VERSION_TIME; get_random_bytes(&clockseq, 2); afs_uuid.clock_seq_low = clockseq; afs_uuid.clock_seq_hi_and_reserved = (clockseq >> 8) & AFS_UUID_CLOCKHI_MASK; - afs_uuid.clock_seq_hi_and_reserved = AFS_UUID_VARIANT_STD; + afs_uuid.clock_seq_hi_and_reserved |= AFS_UUID_VARIANT_STD; _debug("AFS UUID: %08x-%04x-%04x-%02x%02x-%02x%02x%02x%02x%02x%02x", afs_uuid.time_low, @@ -98,15 +88,20 @@ static int __init afs_init(void) if (ret < 0) return ret; + /* create workqueue */ + ret = -ENOMEM; + afs_wq = alloc_workqueue("afs", 0, 0); + if (!afs_wq) + return ret; + /* register the /proc stuff */ ret = afs_proc_init(); if (ret < 0) - return ret; + goto error_proc; -#ifdef AFS_CACHING_SUPPORT +#ifdef CONFIG_AFS_FSCACHE /* we want to be able to cache */ - ret = cachefs_register_netfs(&afs_cache_netfs, - &afs_cache_cell_index_def); + ret = fscache_register_netfs(&afs_cache_netfs); if (ret < 0) goto error_cache; #endif @@ -123,6 +118,8 @@ static int __init afs_init(void) /* initialise the callback update process */ ret = afs_callback_update_init(); + if (ret < 0) + goto error_callback_update_init; /* create the RxRPC transport */ ret = afs_open_socket(); @@ -139,16 +136,20 @@ static int __init afs_init(void) error_fs: afs_close_socket(); error_open_socket: + afs_callback_update_kill(); +error_callback_update_init: + afs_vlocation_purge(); error_vl_update_init: + afs_cell_purge(); error_cell_init: -#ifdef AFS_CACHING_SUPPORT - cachefs_unregister_netfs(&afs_cache_netfs); +#ifdef CONFIG_AFS_FSCACHE + fscache_unregister_netfs(&afs_cache_netfs); error_cache: #endif - afs_callback_update_kill(); - afs_vlocation_purge(); - afs_cell_purge(); afs_proc_cleanup(); +error_proc: + destroy_workqueue(afs_wq); + rcu_barrier(); printk(KERN_ERR "kAFS: failed to register: %d\n", ret); return ret; } @@ -166,16 +167,18 @@ static void __exit afs_exit(void) printk(KERN_INFO "kAFS: Red Hat AFS client v0.1 unregistering.\n"); afs_fs_exit(); + afs_kill_lock_manager(); afs_close_socket(); afs_purge_servers(); afs_callback_update_kill(); afs_vlocation_purge(); - flush_scheduled_work(); + destroy_workqueue(afs_wq); afs_cell_purge(); -#ifdef AFS_CACHING_SUPPORT - cachefs_unregister_netfs(&afs_cache_netfs); +#ifdef CONFIG_AFS_FSCACHE + fscache_unregister_netfs(&afs_cache_netfs); #endif afs_proc_cleanup(); + rcu_barrier(); } module_exit(afs_exit); diff --git a/fs/afs/misc.c b/fs/afs/misc.c index cdb9792d816..0dd4dafee10 100644 --- a/fs/afs/misc.c +++ b/fs/afs/misc.c @@ -12,6 +12,7 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/errno.h> +#include <rxrpc/packet.h> #include "internal.h" #include "afs_fs.h" @@ -22,6 +23,7 @@ int afs_abort_to_error(u32 abort_code) { switch (abort_code) { case 13: return -EACCES; + case 27: return -EFBIG; case 30: return -EROFS; case VSALVAGE: return -EIO; case VNOVNODE: return -ENOENT; @@ -34,6 +36,7 @@ int afs_abort_to_error(u32 abort_code) case VOVERQUOTA: return -EDQUOT; case VBUSY: return -EBUSY; case VMOVED: return -ENXIO; + case 0x2f6df0a: return -EWOULDBLOCK; case 0x2f6df0c: return -EACCES; case 0x2f6df0f: return -EBUSY; case 0x2f6df10: return -EEXIST; @@ -52,6 +55,21 @@ int afs_abort_to_error(u32 abort_code) case 0x2f6df24: return -ENOLCK; case 0x2f6df26: return -ENOTEMPTY; case 0x2f6df78: return -EDQUOT; + + case RXKADINCONSISTENCY: return -EPROTO; + case RXKADPACKETSHORT: return -EPROTO; + case RXKADLEVELFAIL: return -EKEYREJECTED; + case RXKADTICKETLEN: return -EKEYREJECTED; + case RXKADOUTOFSEQUENCE: return -EPROTO; + case RXKADNOAUTH: return -EKEYREJECTED; + case RXKADBADKEY: return -EKEYREJECTED; + case RXKADBADTICKET: return -EKEYREJECTED; + case RXKADUNKNOWNKEY: return -EKEYREJECTED; + case RXKADEXPIRED: return -EKEYEXPIRED; + case RXKADSEALEDINCON: return -EKEYREJECTED; + case RXKADDATALEN: return -EKEYREJECTED; + case RXKADILLEGALLEVEL: return -EKEYREJECTED; + default: return -EREMOTEIO; } } diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index 034fcfd4e33..9682c33d5da 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -12,37 +12,39 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> -#include <linux/slab.h> #include <linux/fs.h> #include <linux/pagemap.h> #include <linux/mount.h> #include <linux/namei.h> -#include <linux/mnt_namespace.h> +#include <linux/gfp.h> #include "internal.h" static struct dentry *afs_mntpt_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd); + unsigned int flags); static int afs_mntpt_open(struct inode *inode, struct file *file); -static void *afs_mntpt_follow_link(struct dentry *dentry, struct nameidata *nd); static void afs_mntpt_expiry_timed_out(struct work_struct *work); const struct file_operations afs_mntpt_file_operations = { .open = afs_mntpt_open, + .llseek = noop_llseek, }; const struct inode_operations afs_mntpt_inode_operations = { .lookup = afs_mntpt_lookup, - .follow_link = afs_mntpt_follow_link, .readlink = page_readlink, - .getattr = afs_inode_getattr, + .getattr = afs_getattr, +}; + +const struct inode_operations afs_autocell_inode_operations = { + .getattr = afs_getattr, }; static LIST_HEAD(afs_vfsmounts); static DECLARE_DELAYED_WORK(afs_mntpt_expiry_timer, afs_mntpt_expiry_timed_out); -unsigned long afs_mntpt_expiry_timeout = 10 * 60; +static unsigned long afs_mntpt_expiry_timeout = 10 * 60; /* * check a symbolic link to see whether it actually encodes a mountpoint @@ -50,18 +52,17 @@ unsigned long afs_mntpt_expiry_timeout = 10 * 60; */ int afs_mntpt_check_symlink(struct afs_vnode *vnode, struct key *key) { - struct file file = { - .private_data = key, - }; struct page *page; size_t size; char *buf; int ret; - _enter("{%u,%u}", vnode->fid.vnode, vnode->fid.unique); + _enter("{%x:%u,%u}", + vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique); /* read the contents of the symlink into the pagecache */ - page = read_mapping_page(AFS_VNODE_TO_I(vnode)->i_mapping, 0, &file); + page = read_cache_page(AFS_VNODE_TO_I(vnode)->i_mapping, 0, + afs_page_filler, key); if (IS_ERR(page)) { ret = PTR_ERR(page); goto out; @@ -84,6 +85,7 @@ int afs_mntpt_check_symlink(struct afs_vnode *vnode, struct key *key) _debug("symlink is a mountpoint"); spin_lock(&vnode->lock); set_bit(AFS_VNODE_MOUNTPOINT, &vnode->flags); + vnode->vfs_inode.i_flags |= S_AUTOMOUNT; spin_unlock(&vnode->lock); } @@ -102,7 +104,7 @@ out: */ static struct dentry *afs_mntpt_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { _enter("%p,%p{%p{%s},%s}", dir, @@ -138,51 +140,78 @@ static struct vfsmount *afs_mntpt_do_automount(struct dentry *mntpt) { struct afs_super_info *super; struct vfsmount *mnt; - struct page *page = NULL; - size_t size; - char *buf, *devname = NULL, *options = NULL; + struct afs_vnode *vnode; + struct page *page; + char *devname, *options; + bool rwpath = false; int ret; _enter("{%s}", mntpt->d_name.name); BUG_ON(!mntpt->d_inode); - ret = -EINVAL; - size = mntpt->d_inode->i_size; - if (size > PAGE_SIZE - 1) - goto error; - ret = -ENOMEM; devname = (char *) get_zeroed_page(GFP_KERNEL); if (!devname) - goto error; + goto error_no_devname; options = (char *) get_zeroed_page(GFP_KERNEL); if (!options) - goto error; - - /* read the contents of the AFS special symlink */ - page = read_mapping_page(mntpt->d_inode->i_mapping, 0, NULL); - if (IS_ERR(page)) { - ret = PTR_ERR(page); - goto error; + goto error_no_options; + + vnode = AFS_FS_I(mntpt->d_inode); + if (test_bit(AFS_VNODE_PSEUDODIR, &vnode->flags)) { + /* if the directory is a pseudo directory, use the d_name */ + static const char afs_root_cell[] = ":root.cell."; + unsigned size = mntpt->d_name.len; + + ret = -ENOENT; + if (size < 2 || size > AFS_MAXCELLNAME) + goto error_no_page; + + if (mntpt->d_name.name[0] == '.') { + devname[0] = '#'; + memcpy(devname + 1, mntpt->d_name.name, size - 1); + memcpy(devname + size, afs_root_cell, + sizeof(afs_root_cell)); + rwpath = true; + } else { + devname[0] = '%'; + memcpy(devname + 1, mntpt->d_name.name, size); + memcpy(devname + size + 1, afs_root_cell, + sizeof(afs_root_cell)); + } + } else { + /* read the contents of the AFS special symlink */ + loff_t size = i_size_read(mntpt->d_inode); + char *buf; + + ret = -EINVAL; + if (size > PAGE_SIZE - 1) + goto error_no_page; + + page = read_mapping_page(mntpt->d_inode->i_mapping, 0, NULL); + if (IS_ERR(page)) { + ret = PTR_ERR(page); + goto error_no_page; + } + + ret = -EIO; + if (PageError(page)) + goto error; + + buf = kmap_atomic(page); + memcpy(devname, buf, size); + kunmap_atomic(buf); + page_cache_release(page); + page = NULL; } - ret = -EIO; - if (PageError(page)) - goto error; - - buf = kmap(page); - memcpy(devname, buf, size); - kunmap(page); - page_cache_release(page); - page = NULL; - /* work out what options we want */ super = AFS_FS_S(mntpt->d_sb); memcpy(options, "cell=", 5); strcpy(options + 5, super->volume->cell->name); - if (super->volume->type == AFSVL_RWVOL) + if (super->volume->type == AFSVL_RWVOL || rwpath) strcat(options, ",rwpath"); /* try and do the mount */ @@ -196,64 +225,35 @@ static struct vfsmount *afs_mntpt_do_automount(struct dentry *mntpt) return mnt; error: - if (page) - page_cache_release(page); - if (devname) - free_page((unsigned long) devname); - if (options) - free_page((unsigned long) options); + page_cache_release(page); +error_no_page: + free_page((unsigned long) options); +error_no_options: + free_page((unsigned long) devname); +error_no_devname: _leave(" = %d", ret); return ERR_PTR(ret); } /* - * follow a link from a mountpoint directory, thus causing it to be mounted + * handle an automount point */ -static void *afs_mntpt_follow_link(struct dentry *dentry, struct nameidata *nd) +struct vfsmount *afs_d_automount(struct path *path) { struct vfsmount *newmnt; - int err; - _enter("%p{%s},{%s:%p{%s},}", - dentry, - dentry->d_name.name, - nd->mnt->mnt_devname, - dentry, - nd->dentry->d_name.name); + _enter("{%s}", path->dentry->d_name.name); - dput(nd->dentry); - nd->dentry = dget(dentry); - - newmnt = afs_mntpt_do_automount(nd->dentry); - if (IS_ERR(newmnt)) { - path_release(nd); - return (void *)newmnt; - } + newmnt = afs_mntpt_do_automount(path->dentry); + if (IS_ERR(newmnt)) + return newmnt; - mntget(newmnt); - err = do_add_mount(newmnt, nd, MNT_SHRINKABLE, &afs_vfsmounts); - switch (err) { - case 0: - mntput(nd->mnt); - dput(nd->dentry); - nd->mnt = newmnt; - nd->dentry = dget(newmnt->mnt_root); - schedule_delayed_work(&afs_mntpt_expiry_timer, - afs_mntpt_expiry_timeout * HZ); - break; - case -EBUSY: - /* someone else made a mount here whilst we were busy */ - while (d_mountpoint(nd->dentry) && - follow_down(&nd->mnt, &nd->dentry)) - ; - err = 0; - default: - mntput(newmnt); - break; - } - - _leave(" = %d", err); - return ERR_PTR(err); + mntget(newmnt); /* prevent immediate expiration */ + mnt_set_expiry(newmnt, &afs_vfsmounts); + queue_delayed_work(afs_wq, &afs_mntpt_expiry_timer, + afs_mntpt_expiry_timeout * HZ); + _leave(" = %p", newmnt); + return newmnt; } /* @@ -265,8 +265,8 @@ static void afs_mntpt_expiry_timed_out(struct work_struct *work) if (!list_empty(&afs_vfsmounts)) { mark_mounts_for_expiry(&afs_vfsmounts); - schedule_delayed_work(&afs_mntpt_expiry_timer, - afs_mntpt_expiry_timeout * HZ); + queue_delayed_work(afs_wq, &afs_mntpt_expiry_timer, + afs_mntpt_expiry_timeout * HZ); } _leave(""); @@ -280,14 +280,5 @@ void afs_mntpt_kill_timer(void) _enter(""); ASSERT(list_empty(&afs_vfsmounts)); - cancel_delayed_work(&afs_mntpt_expiry_timer); - flush_scheduled_work(); -} - -/* - * begin unmount by attempting to remove all automounted mountpoints we added - */ -void afs_umount_begin(struct vfsmount *vfsmnt, int flags) -{ - shrink_submounts(vfsmnt, &afs_vfsmounts); + cancel_delayed_work_sync(&afs_mntpt_expiry_timer); } diff --git a/fs/afs/netdevices.c b/fs/afs/netdevices.c index fc27d4b52e5..7ad36506c25 100644 --- a/fs/afs/netdevices.c +++ b/fs/afs/netdevices.c @@ -8,6 +8,7 @@ #include <linux/inetdevice.h> #include <linux/netdevice.h> #include <linux/if_arp.h> +#include <net/net_namespace.h> #include "internal.h" /* @@ -19,11 +20,10 @@ int afs_get_MAC_address(u8 *mac, size_t maclen) struct net_device *dev; int ret = -ENODEV; - if (maclen != ETH_ALEN) - BUG(); + BUG_ON(maclen != ETH_ALEN); rtnl_lock(); - dev = __dev_getfirstbyhwtype(ARPHRD_ETHER); + dev = __dev_getfirstbyhwtype(&init_net, ARPHRD_ETHER); if (dev) { memcpy(mac, dev->dev_addr, maclen); ret = 0; @@ -47,7 +47,7 @@ int afs_get_ipv4_interfaces(struct afs_interface *bufs, size_t maxbufs, ASSERT(maxbufs > 0); rtnl_lock(); - for_each_netdev(dev) { + for_each_netdev(&init_net, dev) { if (dev->type == ARPHRD_LOOPBACK && !wantloopback) continue; idev = __in_dev_get_rtnl(dev); diff --git a/fs/afs/proc.c b/fs/afs/proc.c index d5601f617cd..24a905b076f 100644 --- a/fs/afs/proc.c +++ b/fs/afs/proc.c @@ -13,6 +13,7 @@ #include <linux/module.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> +#include <linux/sched.h> #include <asm/uaccess.h> #include "internal.h" @@ -27,7 +28,7 @@ static int afs_proc_cells_show(struct seq_file *m, void *v); static ssize_t afs_proc_cells_write(struct file *file, const char __user *buf, size_t size, loff_t *_pos); -static struct seq_operations afs_proc_cells_ops = { +static const struct seq_operations afs_proc_cells_ops = { .start = afs_proc_cells_start, .next = afs_proc_cells_next, .stop = afs_proc_cells_stop, @@ -42,8 +43,6 @@ static const struct file_operations afs_proc_cells_fops = { .release = seq_release, }; -static int afs_proc_rootcell_open(struct inode *inode, struct file *file); -static int afs_proc_rootcell_release(struct inode *inode, struct file *file); static ssize_t afs_proc_rootcell_read(struct file *file, char __user *buf, size_t size, loff_t *_pos); static ssize_t afs_proc_rootcell_write(struct file *file, @@ -51,23 +50,19 @@ static ssize_t afs_proc_rootcell_write(struct file *file, size_t size, loff_t *_pos); static const struct file_operations afs_proc_rootcell_fops = { - .open = afs_proc_rootcell_open, .read = afs_proc_rootcell_read, .write = afs_proc_rootcell_write, .llseek = no_llseek, - .release = afs_proc_rootcell_release }; static int afs_proc_cell_volumes_open(struct inode *inode, struct file *file); -static int afs_proc_cell_volumes_release(struct inode *inode, - struct file *file); static void *afs_proc_cell_volumes_start(struct seq_file *p, loff_t *pos); static void *afs_proc_cell_volumes_next(struct seq_file *p, void *v, loff_t *pos); static void afs_proc_cell_volumes_stop(struct seq_file *p, void *v); static int afs_proc_cell_volumes_show(struct seq_file *m, void *v); -static struct seq_operations afs_proc_cell_volumes_ops = { +static const struct seq_operations afs_proc_cell_volumes_ops = { .start = afs_proc_cell_volumes_start, .next = afs_proc_cell_volumes_next, .stop = afs_proc_cell_volumes_stop, @@ -78,20 +73,18 @@ static const struct file_operations afs_proc_cell_volumes_fops = { .open = afs_proc_cell_volumes_open, .read = seq_read, .llseek = seq_lseek, - .release = afs_proc_cell_volumes_release, + .release = seq_release, }; static int afs_proc_cell_vlservers_open(struct inode *inode, struct file *file); -static int afs_proc_cell_vlservers_release(struct inode *inode, - struct file *file); static void *afs_proc_cell_vlservers_start(struct seq_file *p, loff_t *pos); static void *afs_proc_cell_vlservers_next(struct seq_file *p, void *v, loff_t *pos); static void afs_proc_cell_vlservers_stop(struct seq_file *p, void *v); static int afs_proc_cell_vlservers_show(struct seq_file *m, void *v); -static struct seq_operations afs_proc_cell_vlservers_ops = { +static const struct seq_operations afs_proc_cell_vlservers_ops = { .start = afs_proc_cell_vlservers_start, .next = afs_proc_cell_vlservers_next, .stop = afs_proc_cell_vlservers_stop, @@ -102,19 +95,17 @@ static const struct file_operations afs_proc_cell_vlservers_fops = { .open = afs_proc_cell_vlservers_open, .read = seq_read, .llseek = seq_lseek, - .release = afs_proc_cell_vlservers_release, + .release = seq_release, }; static int afs_proc_cell_servers_open(struct inode *inode, struct file *file); -static int afs_proc_cell_servers_release(struct inode *inode, - struct file *file); static void *afs_proc_cell_servers_start(struct seq_file *p, loff_t *pos); static void *afs_proc_cell_servers_next(struct seq_file *p, void *v, loff_t *pos); static void afs_proc_cell_servers_stop(struct seq_file *p, void *v); static int afs_proc_cell_servers_show(struct seq_file *m, void *v); -static struct seq_operations afs_proc_cell_servers_ops = { +static const struct seq_operations afs_proc_cell_servers_ops = { .start = afs_proc_cell_servers_start, .next = afs_proc_cell_servers_next, .stop = afs_proc_cell_servers_stop, @@ -125,7 +116,7 @@ static const struct file_operations afs_proc_cell_servers_fops = { .open = afs_proc_cell_servers_open, .read = seq_read, .llseek = seq_lseek, - .release = afs_proc_cell_servers_release, + .release = seq_release, }; /* @@ -133,34 +124,21 @@ static const struct file_operations afs_proc_cell_servers_fops = { */ int afs_proc_init(void) { - struct proc_dir_entry *p; - _enter(""); proc_afs = proc_mkdir("fs/afs", NULL); if (!proc_afs) goto error_dir; - proc_afs->owner = THIS_MODULE; - - p = create_proc_entry("cells", 0, proc_afs); - if (!p) - goto error_cells; - p->proc_fops = &afs_proc_cells_fops; - p->owner = THIS_MODULE; - p = create_proc_entry("rootcell", 0, proc_afs); - if (!p) - goto error_rootcell; - p->proc_fops = &afs_proc_rootcell_fops; - p->owner = THIS_MODULE; + if (!proc_create("cells", 0644, proc_afs, &afs_proc_cells_fops) || + !proc_create("rootcell", 0644, proc_afs, &afs_proc_rootcell_fops)) + goto error_tree; _leave(" = 0"); return 0; -error_rootcell: - remove_proc_entry("cells", proc_afs); -error_cells: - remove_proc_entry("fs/afs", NULL); +error_tree: + remove_proc_subtree("fs/afs", NULL); error_dir: _leave(" = -ENOMEM"); return -ENOMEM; @@ -171,9 +149,7 @@ error_dir: */ void afs_proc_cleanup(void) { - remove_proc_entry("rootcell", proc_afs); - remove_proc_entry("cells", proc_afs); - remove_proc_entry("fs/afs", NULL); + remove_proc_subtree("fs/afs", NULL); } /* @@ -189,7 +165,7 @@ static int afs_proc_cells_open(struct inode *inode, struct file *file) return ret; m = file->private_data; - m->private = PDE(inode)->data; + m->private = PDE_DATA(inode); return 0; } @@ -200,23 +176,9 @@ static int afs_proc_cells_open(struct inode *inode, struct file *file) */ static void *afs_proc_cells_start(struct seq_file *m, loff_t *_pos) { - struct list_head *_p; - loff_t pos = *_pos; - /* lock the list against modification */ down_read(&afs_proc_cells_sem); - - /* allow for the header line */ - if (!pos) - return (void *) 1; - pos--; - - /* find the n'th element in the list */ - list_for_each(_p, &afs_proc_cells) - if (!pos--) - break; - - return _p != &afs_proc_cells ? _p : NULL; + return seq_list_start_head(&afs_proc_cells, *_pos); } /* @@ -224,14 +186,7 @@ static void *afs_proc_cells_start(struct seq_file *m, loff_t *_pos) */ static void *afs_proc_cells_next(struct seq_file *p, void *v, loff_t *pos) { - struct list_head *_p; - - (*pos)++; - - _p = v; - _p = v == (void *) 1 ? afs_proc_cells.next : _p->next; - - return _p != &afs_proc_cells ? _p : NULL; + return seq_list_next(v, &afs_proc_cells, pos); } /* @@ -249,7 +204,7 @@ static int afs_proc_cells_show(struct seq_file *m, void *v) { struct afs_cell *cell = list_entry(v, struct afs_cell, proc_link); - if (v == (void *) 1) { + if (v == &afs_proc_cells) { /* display header on line 1 */ seq_puts(m, "USE NAME\n"); return 0; @@ -314,7 +269,7 @@ static ssize_t afs_proc_cells_write(struct file *file, const char __user *buf, if (strcmp(kbuf, "add") == 0) { struct afs_cell *cell; - cell = afs_cell_create(name, args); + cell = afs_cell_create(name, strlen(name), args, false); if (IS_ERR(cell)) { ret = PTR_ERR(cell); goto done; @@ -339,19 +294,6 @@ inval: goto done; } -/* - * Stubs for /proc/fs/afs/rootcell - */ -static int afs_proc_rootcell_open(struct inode *inode, struct file *file) -{ - return 0; -} - -static int afs_proc_rootcell_release(struct inode *inode, struct file *file) -{ - return 0; -} - static ssize_t afs_proc_rootcell_read(struct file *file, char __user *buf, size_t size, loff_t *_pos) { @@ -407,44 +349,27 @@ nomem: */ int afs_proc_cell_setup(struct afs_cell *cell) { - struct proc_dir_entry *p; + struct proc_dir_entry *dir; _enter("%p{%s}", cell, cell->name); - cell->proc_dir = proc_mkdir(cell->name, proc_afs); - if (!cell->proc_dir) + dir = proc_mkdir(cell->name, proc_afs); + if (!dir) goto error_dir; - p = create_proc_entry("servers", 0, cell->proc_dir); - if (!p) - goto error_servers; - p->proc_fops = &afs_proc_cell_servers_fops; - p->owner = THIS_MODULE; - p->data = cell; - - p = create_proc_entry("vlservers", 0, cell->proc_dir); - if (!p) - goto error_vlservers; - p->proc_fops = &afs_proc_cell_vlservers_fops; - p->owner = THIS_MODULE; - p->data = cell; - - p = create_proc_entry("volumes", 0, cell->proc_dir); - if (!p) - goto error_volumes; - p->proc_fops = &afs_proc_cell_volumes_fops; - p->owner = THIS_MODULE; - p->data = cell; + if (!proc_create_data("servers", 0, dir, + &afs_proc_cell_servers_fops, cell) || + !proc_create_data("vlservers", 0, dir, + &afs_proc_cell_vlservers_fops, cell) || + !proc_create_data("volumes", 0, dir, + &afs_proc_cell_volumes_fops, cell)) + goto error_tree; _leave(" = 0"); return 0; -error_volumes: - remove_proc_entry("vlservers", cell->proc_dir); -error_vlservers: - remove_proc_entry("servers", cell->proc_dir); -error_servers: - remove_proc_entry(cell->name, proc_afs); +error_tree: + remove_proc_subtree(cell->name, proc_afs); error_dir: _leave(" = -ENOMEM"); return -ENOMEM; @@ -457,10 +382,7 @@ void afs_proc_cell_remove(struct afs_cell *cell) { _enter(""); - remove_proc_entry("volumes", cell->proc_dir); - remove_proc_entry("vlservers", cell->proc_dir); - remove_proc_entry("servers", cell->proc_dir); - remove_proc_entry(cell->name, proc_afs); + remove_proc_subtree(cell->name, proc_afs); _leave(""); } @@ -474,7 +396,7 @@ static int afs_proc_cell_volumes_open(struct inode *inode, struct file *file) struct seq_file *m; int ret; - cell = PDE(inode)->data; + cell = PDE_DATA(inode); if (!cell) return -ENOENT; @@ -489,39 +411,18 @@ static int afs_proc_cell_volumes_open(struct inode *inode, struct file *file) } /* - * close the file and release the ref to the cell - */ -static int afs_proc_cell_volumes_release(struct inode *inode, struct file *file) -{ - return seq_release(inode, file); -} - -/* * set up the iterator to start reading from the cells list and return the * first item */ static void *afs_proc_cell_volumes_start(struct seq_file *m, loff_t *_pos) { - struct list_head *_p; struct afs_cell *cell = m->private; - loff_t pos = *_pos; _enter("cell=%p pos=%Ld", cell, *_pos); /* lock the list against modification */ down_read(&cell->vl_sem); - - /* allow for the header line */ - if (!pos) - return (void *) 1; - pos--; - - /* find the n'th element in the list */ - list_for_each(_p, &cell->vl_list) - if (!pos--) - break; - - return _p != &cell->vl_list ? _p : NULL; + return seq_list_start_head(&cell->vl_list, *_pos); } /* @@ -530,17 +431,10 @@ static void *afs_proc_cell_volumes_start(struct seq_file *m, loff_t *_pos) static void *afs_proc_cell_volumes_next(struct seq_file *p, void *v, loff_t *_pos) { - struct list_head *_p; struct afs_cell *cell = p->private; _enter("cell=%p pos=%Ld", cell, *_pos); - - (*_pos)++; - - _p = v; - _p = (v == (void *) 1) ? cell->vl_list.next : _p->next; - - return (_p != &cell->vl_list) ? _p : NULL; + return seq_list_next(v, &cell->vl_list, _pos); } /* @@ -553,7 +447,7 @@ static void afs_proc_cell_volumes_stop(struct seq_file *p, void *v) up_read(&cell->vl_sem); } -const char afs_vlocation_states[][4] = { +static const char afs_vlocation_states[][4] = { [AFS_VL_NEW] = "New", [AFS_VL_CREATING] = "Crt", [AFS_VL_VALID] = "Val", @@ -568,11 +462,12 @@ const char afs_vlocation_states[][4] = { */ static int afs_proc_cell_volumes_show(struct seq_file *m, void *v) { + struct afs_cell *cell = m->private; struct afs_vlocation *vlocation = list_entry(v, struct afs_vlocation, link); /* display header on line 1 */ - if (v == (void *) 1) { + if (v == &cell->vl_list) { seq_puts(m, "USE STT VLID[0] VLID[1] VLID[2] NAME\n"); return 0; } @@ -599,7 +494,7 @@ static int afs_proc_cell_vlservers_open(struct inode *inode, struct file *file) struct seq_file *m; int ret; - cell = PDE(inode)->data; + cell = PDE_DATA(inode); if (!cell) return -ENOENT; @@ -614,15 +509,6 @@ static int afs_proc_cell_vlservers_open(struct inode *inode, struct file *file) } /* - * close the file and release the ref to the cell - */ -static int afs_proc_cell_vlservers_release(struct inode *inode, - struct file *file) -{ - return seq_release(inode, file); -} - -/* * set up the iterator to start reading from the cells list and return the * first item */ @@ -690,7 +576,7 @@ static int afs_proc_cell_vlservers_show(struct seq_file *m, void *v) } /* display one cell per line on subsequent lines */ - seq_printf(m, "%u.%u.%u.%u\n", NIPQUAD(addr->s_addr)); + seq_printf(m, "%pI4\n", &addr->s_addr); return 0; } @@ -704,7 +590,7 @@ static int afs_proc_cell_servers_open(struct inode *inode, struct file *file) struct seq_file *m; int ret; - cell = PDE(inode)->data; + cell = PDE_DATA(inode); if (!cell) return -ENOENT; @@ -718,41 +604,19 @@ static int afs_proc_cell_servers_open(struct inode *inode, struct file *file) } /* - * close the file and release the ref to the cell - */ -static int afs_proc_cell_servers_release(struct inode *inode, - struct file *file) -{ - return seq_release(inode, file); -} - -/* * set up the iterator to start reading from the cells list and return the * first item */ static void *afs_proc_cell_servers_start(struct seq_file *m, loff_t *_pos) __acquires(m->private->servers_lock) { - struct list_head *_p; struct afs_cell *cell = m->private; - loff_t pos = *_pos; _enter("cell=%p pos=%Ld", cell, *_pos); /* lock the list against modification */ read_lock(&cell->servers_lock); - - /* allow for the header line */ - if (!pos) - return (void *) 1; - pos--; - - /* find the n'th element in the list */ - list_for_each(_p, &cell->servers) - if (!pos--) - break; - - return _p != &cell->servers ? _p : NULL; + return seq_list_start_head(&cell->servers, *_pos); } /* @@ -761,17 +625,10 @@ static void *afs_proc_cell_servers_start(struct seq_file *m, loff_t *_pos) static void *afs_proc_cell_servers_next(struct seq_file *p, void *v, loff_t *_pos) { - struct list_head *_p; struct afs_cell *cell = p->private; _enter("cell=%p pos=%Ld", cell, *_pos); - - (*_pos)++; - - _p = v; - _p = v == (void *) 1 ? cell->servers.next : _p->next; - - return _p != &cell->servers ? _p : NULL; + return seq_list_next(v, &cell->servers, _pos); } /* @@ -790,17 +647,18 @@ static void afs_proc_cell_servers_stop(struct seq_file *p, void *v) */ static int afs_proc_cell_servers_show(struct seq_file *m, void *v) { + struct afs_cell *cell = m->private; struct afs_server *server = list_entry(v, struct afs_server, link); char ipaddr[20]; /* display header on line 1 */ - if (v == (void *) 1) { + if (v == &cell->servers) { seq_puts(m, "USE ADDR STATE\n"); return 0; } /* display one cell per line on subsequent lines */ - sprintf(ipaddr, "%u.%u.%u.%u", NIPQUAD(server->addr)); + sprintf(ipaddr, "%pI4", &server->addr); seq_printf(m, "%3d %-15.15s %5d\n", atomic_read(&server->usage), ipaddr, server->fs_state); diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index 222c1a3abbb..03a3beb1700 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -9,6 +9,7 @@ * 2 of the License, or (at your option) any later version. */ +#include <linux/slab.h> #include <net/sock.h> #include <net/af_rxrpc.h> #include <rxrpc/packet.h> @@ -24,7 +25,7 @@ static void afs_wake_up_call_waiter(struct afs_call *); static int afs_wait_for_call_to_complete(struct afs_call *); static void afs_wake_up_async_call(struct afs_call *); static int afs_dont_wait_for_call_to_complete(struct afs_call *); -static void afs_process_async_call(struct work_struct *); +static void afs_process_async_call(struct afs_call *); static void afs_rx_interceptor(struct sock *, unsigned long, struct sk_buff *); static int afs_deliver_cm_op_id(struct afs_call *, struct sk_buff *, bool); @@ -57,6 +58,13 @@ static void afs_collect_incoming_call(struct work_struct *); static struct sk_buff_head afs_incoming_calls; static DECLARE_WORK(afs_collect_incoming_call_work, afs_collect_incoming_call); +static void afs_async_workfn(struct work_struct *work) +{ + struct afs_call *call = container_of(work, struct afs_call, async_work); + + call->async_workfn(call); +} + /* * open an RxRPC socket and bind it to be a server for callback notifications * - the socket is left in blocking mode and non-blocking ops use MSG_DONTWAIT @@ -99,6 +107,7 @@ int afs_open_socket(void) ret = kernel_bind(socket, (struct sockaddr *) &srx, sizeof(srx)); if (ret < 0) { sock_release(socket); + destroy_workqueue(afs_async_calls); _leave(" = %d [bind]", ret); return ret; } @@ -182,6 +191,28 @@ static void afs_free_call(struct afs_call *call) } /* + * End a call but do not free it + */ +static void afs_end_call_nofree(struct afs_call *call) +{ + if (call->rxcall) { + rxrpc_kernel_end_call(call->rxcall); + call->rxcall = NULL; + } + if (call->type->destructor) + call->type->destructor(call); +} + +/* + * End a call and free it + */ +static void afs_end_call(struct afs_call *call) +{ + afs_end_call_nofree(call); + afs_free_call(call); +} + +/* * allocate a call with flat request and reply buffers */ struct afs_call *afs_alloc_flat_call(const struct afs_call_type *type, @@ -237,6 +268,71 @@ void afs_flat_call_destructor(struct afs_call *call) } /* + * attach the data from a bunch of pages on an inode to a call + */ +static int afs_send_pages(struct afs_call *call, struct msghdr *msg, + struct kvec *iov) +{ + struct page *pages[8]; + unsigned count, n, loop, offset, to; + pgoff_t first = call->first, last = call->last; + int ret; + + _enter(""); + + offset = call->first_offset; + call->first_offset = 0; + + do { + _debug("attach %lx-%lx", first, last); + + count = last - first + 1; + if (count > ARRAY_SIZE(pages)) + count = ARRAY_SIZE(pages); + n = find_get_pages_contig(call->mapping, first, count, pages); + ASSERTCMP(n, ==, count); + + loop = 0; + do { + msg->msg_flags = 0; + to = PAGE_SIZE; + if (first + loop >= last) + to = call->last_to; + else + msg->msg_flags = MSG_MORE; + iov->iov_base = kmap(pages[loop]) + offset; + iov->iov_len = to - offset; + offset = 0; + + _debug("- range %u-%u%s", + offset, to, msg->msg_flags ? " [more]" : ""); + msg->msg_iov = (struct iovec *) iov; + msg->msg_iovlen = 1; + + /* have to change the state *before* sending the last + * packet as RxRPC might give us the reply before it + * returns from sending the request */ + if (first + loop >= last) + call->state = AFS_CALL_AWAIT_REPLY; + ret = rxrpc_kernel_send_data(call->rxcall, msg, + to - offset); + kunmap(pages[loop]); + if (ret < 0) + break; + } while (++loop < count); + first += count; + + for (loop = 0; loop < count; loop++) + put_page(pages[loop]); + if (ret < 0) + break; + } while (first <= last); + + _leave(" = %d", ret); + return ret; +} + +/* * initiate a call */ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp, @@ -247,17 +343,20 @@ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp, struct msghdr msg; struct kvec iov[1]; int ret; + struct sk_buff *skb; _enter("%x,{%d},", addr->s_addr, ntohs(call->port)); ASSERT(call->type != NULL); ASSERT(call->type->name != NULL); - _debug("MAKE %p{%s} [%d]", - call, call->type->name, atomic_read(&afs_outstanding_calls)); + _debug("____MAKE %p{%s,%x} [%d]____", + call, call->type->name, key_serial(call->key), + atomic_read(&afs_outstanding_calls)); call->wait_mode = wait_mode; - INIT_WORK(&call->async_work, afs_process_async_call); + call->async_workfn = afs_process_async_call; + INIT_WORK(&call->async_work, afs_async_workfn); memset(&srx, 0, sizeof(srx)); srx.srx_family = AF_RXRPC; @@ -289,27 +388,33 @@ int afs_make_call(struct in_addr *addr, struct afs_call *call, gfp_t gfp, msg.msg_iovlen = 1; msg.msg_control = NULL; msg.msg_controllen = 0; - msg.msg_flags = 0; + msg.msg_flags = (call->send_pages ? MSG_MORE : 0); /* have to change the state *before* sending the last packet as RxRPC * might give us the reply before it returns from sending the * request */ - call->state = AFS_CALL_AWAIT_REPLY; + if (!call->send_pages) + call->state = AFS_CALL_AWAIT_REPLY; ret = rxrpc_kernel_send_data(rxcall, &msg, call->request_size); if (ret < 0) goto error_do_abort; + if (call->send_pages) { + ret = afs_send_pages(call, &msg, iov); + if (ret < 0) + goto error_do_abort; + } + /* at this point, an async call may no longer exist as it may have * already completed */ return wait_mode->wait(call); error_do_abort: rxrpc_kernel_abort_call(rxcall, RX_USER_ABORT); - rxrpc_kernel_end_call(rxcall); - call->rxcall = NULL; + while ((skb = skb_dequeue(&call->rx_queue))) + afs_free_skb(skb); error_kill_call: - call->type->destructor(call); - afs_free_call(call); + afs_end_call(call); _leave(" = %d", ret); return ret; } @@ -335,7 +440,7 @@ static void afs_rx_interceptor(struct sock *sk, unsigned long user_call_ID, if (!call) { /* its an incoming call for our callback service */ skb_queue_tail(&afs_incoming_calls, skb); - schedule_work(&afs_collect_incoming_call_work); + queue_work(afs_wq, &afs_collect_incoming_call_work); } else { /* route the messages directly to the appropriate call */ skb_queue_tail(&call->rx_queue, skb); @@ -431,12 +536,8 @@ static void afs_deliver_to_call(struct afs_call *call) if (call->state >= AFS_CALL_COMPLETE) { while ((skb = skb_dequeue(&call->rx_queue))) afs_free_skb(skb); - if (call->incoming) { - rxrpc_kernel_end_call(call->rxcall); - call->rxcall = NULL; - call->type->destructor(call); - afs_free_call(call); - } + if (call->incoming) + afs_end_call(call); } _leave(""); @@ -486,10 +587,7 @@ static int afs_wait_for_call_to_complete(struct afs_call *call) } _debug("call complete"); - rxrpc_kernel_end_call(call->rxcall); - call->rxcall = NULL; - call->type->destructor(call); - afs_free_call(call); + afs_end_call(call); _leave(" = %d", ret); return ret; } @@ -525,11 +623,8 @@ static int afs_dont_wait_for_call_to_complete(struct afs_call *call) /* * delete an asynchronous call */ -static void afs_delete_async_call(struct work_struct *work) +static void afs_delete_async_call(struct afs_call *call) { - struct afs_call *call = - container_of(work, struct afs_call, async_work); - _enter(""); afs_free_call(call); @@ -542,11 +637,8 @@ static void afs_delete_async_call(struct work_struct *work) * - on a multiple-thread workqueue this work item may try to run on several * CPUs at the same time */ -static void afs_process_async_call(struct work_struct *work) +static void afs_process_async_call(struct afs_call *call) { - struct afs_call *call = - container_of(work, struct afs_call, async_work); - _enter(""); if (!skb_queue_empty(&call->rx_queue)) @@ -559,14 +651,11 @@ static void afs_process_async_call(struct work_struct *work) call->reply = NULL; /* kill the call */ - rxrpc_kernel_end_call(call->rxcall); - call->rxcall = NULL; - if (call->type->destructor) - call->type->destructor(call); + afs_end_call_nofree(call); /* we can't just delete the call because the work item may be * queued */ - PREPARE_WORK(&call->async_work, afs_delete_async_call); + call->async_workfn = afs_delete_async_call; queue_work(afs_async_calls, &call->async_work); } @@ -607,7 +696,8 @@ static void afs_collect_incoming_call(struct work_struct *work) return; } - INIT_WORK(&call->async_work, afs_process_async_call); + call->async_workfn = afs_process_async_call; + INIT_WORK(&call->async_work, afs_async_workfn); call->wait_mode = &afs_async_incoming_call; call->type = &afs_RXCMxxxx; init_waitqueue_head(&call->waitq); @@ -704,10 +794,7 @@ void afs_send_empty_reply(struct afs_call *call) _debug("oom"); rxrpc_kernel_abort_call(call->rxcall, RX_USER_ABORT); default: - rxrpc_kernel_end_call(call->rxcall); - call->rxcall = NULL; - call->type->destructor(call); - afs_free_call(call); + afs_end_call(call); _leave(" [error]"); return; } @@ -720,6 +807,7 @@ void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len) { struct msghdr msg; struct iovec iov[1]; + int n; _enter(""); @@ -734,22 +822,19 @@ void afs_send_simple_reply(struct afs_call *call, const void *buf, size_t len) msg.msg_flags = 0; call->state = AFS_CALL_AWAIT_ACK; - switch (rxrpc_kernel_send_data(call->rxcall, &msg, len)) { - case 0: + n = rxrpc_kernel_send_data(call->rxcall, &msg, len); + if (n >= 0) { + /* Success */ _leave(" [replied]"); return; + } - case -ENOMEM: + if (n == -ENOMEM) { _debug("oom"); rxrpc_kernel_abort_call(call->rxcall, RX_USER_ABORT); - default: - rxrpc_kernel_end_call(call->rxcall); - call->rxcall = NULL; - call->type->destructor(call); - afs_free_call(call); - _leave(" [error]"); - return; } + afs_end_call(call); + _leave(" [error]"); } /* diff --git a/fs/afs/security.c b/fs/afs/security.c index f9f424d8045..8d010422dc8 100644 --- a/fs/afs/security.c +++ b/fs/afs/security.c @@ -13,6 +13,7 @@ #include <linux/slab.h> #include <linux/fs.h> #include <linux/ctype.h> +#include <linux/sched.h> #include <keys/rxrpc-type.h> #include "internal.h" @@ -94,7 +95,7 @@ static struct afs_vnode *afs_get_auth_inode(struct afs_vnode *vnode, auth_inode = afs_iget(vnode->vfs_inode.i_sb, key, &vnode->status.parent, NULL, NULL); if (IS_ERR(auth_inode)) - return ERR_PTR(PTR_ERR(auth_inode)); + return ERR_CAST(auth_inode); } auth_vnode = AFS_FS_I(auth_inode); @@ -109,7 +110,7 @@ void afs_clear_permits(struct afs_vnode *vnode) { struct afs_permits *permits; - _enter("{%x}", vnode->fid.vnode); + _enter("{%x:%u}", vnode->fid.vid, vnode->fid.vnode); mutex_lock(&vnode->permits_lock); permits = vnode->permits; @@ -132,7 +133,8 @@ void afs_cache_permit(struct afs_vnode *vnode, struct key *key, long acl_order) struct afs_vnode *auth_vnode; int count, loop; - _enter("{%x},%x,%lx", vnode->fid.vnode, key_serial(key), acl_order); + _enter("{%x:%u},%x,%lx", + vnode->fid.vid, vnode->fid.vnode, key_serial(key), acl_order); auth_vnode = afs_get_auth_inode(vnode, key); if (IS_ERR(auth_vnode)) { @@ -187,8 +189,9 @@ void afs_cache_permit(struct afs_vnode *vnode, struct key *key, long acl_order) if (!permits) goto out_unlock; - memcpy(permits->permits, xpermits->permits, - count * sizeof(struct afs_permit)); + if (xpermits) + memcpy(permits->permits, xpermits->permits, + count * sizeof(struct afs_permit)); _debug("key %x access %x", key_serial(key), vnode->status.caller_access); @@ -220,7 +223,8 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key, bool valid; int loop, ret; - _enter(""); + _enter("{%x:%u},%x", + vnode->fid.vid, vnode->fid.vnode, key_serial(key)); auth_vnode = afs_get_auth_inode(vnode, key); if (IS_ERR(auth_vnode)) { @@ -268,9 +272,9 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key, _leave(" = %d", ret); return ret; } + *_access = vnode->status.caller_access; } - *_access = vnode->status.caller_access; iput(&auth_vnode->vfs_inode); _leave(" = 0 [access %x]", *_access); return 0; @@ -281,14 +285,17 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key, * - AFS ACLs are attached to directories only, and a file is controlled by its * parent directory's ACL */ -int afs_permission(struct inode *inode, int mask, struct nameidata *nd) +int afs_permission(struct inode *inode, int mask) { struct afs_vnode *vnode = AFS_FS_I(inode); - afs_access_t access; + afs_access_t uninitialized_var(access); struct key *key; int ret; - _enter("{{%x:%x},%lx},%x,", + if (mask & MAY_NOT_BLOCK) + return -ECHILD; + + _enter("{{%x:%u},%lx},%x,", vnode->fid.vid, vnode->fid.vnode, vnode->flags, mask); key = afs_request_key(vnode->volume->cell); @@ -343,7 +350,7 @@ int afs_permission(struct inode *inode, int mask, struct nameidata *nd) } key_put(key); - ret = generic_permission(inode, mask, NULL); + ret = generic_permission(inode, mask); _leave(" = %d", ret); return ret; diff --git a/fs/afs/server.c b/fs/afs/server.c index 96bb23b476a..f342acf3547 100644 --- a/fs/afs/server.c +++ b/fs/afs/server.c @@ -13,7 +13,7 @@ #include <linux/slab.h> #include "internal.h" -unsigned afs_server_timeout = 10; /* server timeout in seconds */ +static unsigned afs_server_timeout = 10; /* server timeout in seconds */ static void afs_reap_server(struct work_struct *); @@ -91,9 +91,10 @@ static struct afs_server *afs_alloc_server(struct afs_cell *cell, memcpy(&server->addr, addr, sizeof(struct in_addr)); server->addr.s_addr = addr->s_addr; + _leave(" = %p{%d}", server, atomic_read(&server->usage)); + } else { + _leave(" = NULL [nomem]"); } - - _leave(" = %p{%d}", server, atomic_read(&server->usage)); return server; } @@ -105,7 +106,7 @@ struct afs_server *afs_lookup_server(struct afs_cell *cell, { struct afs_server *server, *candidate; - _enter("%p,"NIPQUAD_FMT, cell, NIPQUAD(addr->s_addr)); + _enter("%p,%pI4", cell, &addr->s_addr); /* quick scan of the list to see if we already have the server */ read_lock(&cell->servers_lock); @@ -168,9 +169,8 @@ found_server: server_in_two_cells: write_unlock(&cell->servers_lock); kfree(candidate); - printk(KERN_NOTICE "kAFS:" - " Server "NIPQUAD_FMT" appears to be in two cells\n", - NIPQUAD(*addr)); + printk(KERN_NOTICE "kAFS: Server %pI4 appears to be in two cells\n", + addr); _leave(" = -EEXIST"); return ERR_PTR(-EEXIST); } @@ -184,7 +184,7 @@ struct afs_server *afs_find_server(const struct in_addr *_addr) struct rb_node *p; struct in_addr addr = *_addr; - _enter(NIPQUAD_FMT, NIPQUAD(addr.s_addr)); + _enter("%pI4", &addr.s_addr); read_lock(&afs_servers_lock); @@ -238,8 +238,8 @@ void afs_put_server(struct afs_server *server) if (atomic_read(&server->usage) == 0) { list_move_tail(&server->grave, &afs_server_graveyard); server->time_of_death = get_seconds(); - schedule_delayed_work(&afs_server_reaper, - afs_server_timeout * HZ); + queue_delayed_work(afs_wq, &afs_server_reaper, + afs_server_timeout * HZ); } spin_unlock(&afs_server_graveyard_lock); _leave(" [dead]"); @@ -252,6 +252,9 @@ static void afs_destroy_server(struct afs_server *server) { _enter("%p", server); + ASSERTIF(server->cb_break_head != server->cb_break_tail, + delayed_work_pending(&server->cb_break_work)); + ASSERTCMP(server->fs_vnodes.rb_node, ==, NULL); ASSERTCMP(server->cb_promises.rb_node, ==, NULL); ASSERTCMP(server->cb_break_head, ==, server->cb_break_tail); @@ -282,11 +285,7 @@ static void afs_reap_server(struct work_struct *work) expiry = server->time_of_death + afs_server_timeout; if (expiry > now) { delay = (expiry - now) * HZ; - if (!schedule_delayed_work(&afs_server_reaper, delay)) { - cancel_delayed_work(&afs_server_reaper); - schedule_delayed_work(&afs_server_reaper, - delay); - } + mod_delayed_work(afs_wq, &afs_server_reaper, delay); break; } @@ -319,6 +318,5 @@ static void afs_reap_server(struct work_struct *work) void __exit afs_purge_servers(void) { afs_server_timeout = 0; - cancel_delayed_work(&afs_server_reaper); - schedule_delayed_work(&afs_server_reaper, 0); + mod_delayed_work(afs_wq, &afs_server_reaper, 0); } diff --git a/fs/afs/super.c b/fs/afs/super.c index 7030d76155f..c4861557e38 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -10,50 +10,50 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. * * Authors: David Howells <dhowells@redhat.com> - * David Woodhouse <dwmw2@redhat.com> + * David Woodhouse <dwmw2@infradead.org> * */ #include <linux/kernel.h> #include <linux/module.h> +#include <linux/mount.h> #include <linux/init.h> #include <linux/slab.h> #include <linux/fs.h> #include <linux/pagemap.h> #include <linux/parser.h> +#include <linux/statfs.h> +#include <linux/sched.h> +#include <linux/nsproxy.h> +#include <net/net_namespace.h> #include "internal.h" #define AFS_FS_MAGIC 0x6B414653 /* 'kAFS' */ -static void afs_i_init_once(void *foo, struct kmem_cache *cachep, - unsigned long flags); - -static int afs_get_sb(struct file_system_type *fs_type, - int flags, const char *dev_name, - void *data, struct vfsmount *mnt); - +static void afs_i_init_once(void *foo); +static struct dentry *afs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data); +static void afs_kill_super(struct super_block *sb); static struct inode *afs_alloc_inode(struct super_block *sb); - -static void afs_put_super(struct super_block *sb); - static void afs_destroy_inode(struct inode *inode); +static int afs_statfs(struct dentry *dentry, struct kstatfs *buf); struct file_system_type afs_fs_type = { .owner = THIS_MODULE, .name = "afs", - .get_sb = afs_get_sb, - .kill_sb = kill_anon_super, + .mount = afs_mount, + .kill_sb = afs_kill_super, .fs_flags = 0, }; +MODULE_ALIAS_FS("afs"); static const struct super_operations afs_super_ops = { - .statfs = simple_statfs, + .statfs = afs_statfs, .alloc_inode = afs_alloc_inode, - .drop_inode = generic_delete_inode, + .drop_inode = afs_drop_inode, .destroy_inode = afs_destroy_inode, - .clear_inode = afs_clear_inode, - .umount_begin = afs_umount_begin, - .put_super = afs_put_super, + .evict_inode = afs_evict_inode, + .show_options = generic_show_options, }; static struct kmem_cache *afs_inode_cachep; @@ -64,12 +64,14 @@ enum { afs_opt_cell, afs_opt_rwpath, afs_opt_vol, + afs_opt_autocell, }; static const match_table_t afs_options_list = { { afs_opt_cell, "cell=%s" }, { afs_opt_rwpath, "rwpath" }, { afs_opt_vol, "vol=%s" }, + { afs_opt_autocell, "autocell" }, { afs_no_opt, NULL }, }; @@ -90,8 +92,7 @@ int __init afs_fs_init(void) sizeof(struct afs_vnode), 0, SLAB_HWCACHE_ALIGN, - afs_i_init_once, - NULL); + afs_i_init_once); if (!afs_inode_cachep) { printk(KERN_NOTICE "kAFS: Failed to allocate inode cache\n"); return ret; @@ -125,6 +126,11 @@ void __exit afs_fs_exit(void) BUG(); } + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(afs_inode_cachep); _leave(""); } @@ -154,7 +160,8 @@ static int afs_parse_options(struct afs_mount_params *params, switch (token) { case afs_opt_cell: cell = afs_cell_lookup(args[0].from, - args[0].to - args[0].from); + args[0].to - args[0].from, + false); if (IS_ERR(cell)) return PTR_ERR(cell); afs_put_cell(params->cell); @@ -169,6 +176,10 @@ static int afs_parse_options(struct afs_mount_params *params, *devname = args[0].from; break; + case afs_opt_autocell: + params->autocell = 1; + break; + default: printk(KERN_ERR "kAFS:" " Unknown or invalid mount option: '%s'\n", p); @@ -255,10 +266,10 @@ static int afs_parse_device_name(struct afs_mount_params *params, /* lookup the cell record */ if (cellname || !params->cell) { - cell = afs_cell_lookup(cellname, cellnamesz); + cell = afs_cell_lookup(cellname, cellnamesz, true); if (IS_ERR(cell)) { - printk(KERN_ERR "kAFS: unable to lookup cell '%s'\n", - cellname ?: ""); + printk(KERN_ERR "kAFS: unable to lookup cell '%*.*s'\n", + cellnamesz, cellnamesz, cellname ?: ""); return PTR_ERR(cell); } afs_put_cell(params->cell); @@ -278,42 +289,38 @@ static int afs_parse_device_name(struct afs_mount_params *params, */ static int afs_test_super(struct super_block *sb, void *data) { - struct afs_mount_params *params = data; + struct afs_super_info *as1 = data; struct afs_super_info *as = sb->s_fs_info; - return as->volume == params->volume; + return as->volume == as1->volume; +} + +static int afs_set_super(struct super_block *sb, void *data) +{ + sb->s_fs_info = data; + return set_anon_super(sb, NULL); } /* * fill in the superblock */ -static int afs_fill_super(struct super_block *sb, void *data) +static int afs_fill_super(struct super_block *sb, + struct afs_mount_params *params) { - struct afs_mount_params *params = data; - struct afs_super_info *as = NULL; + struct afs_super_info *as = sb->s_fs_info; struct afs_fid fid; - struct dentry *root = NULL; struct inode *inode = NULL; int ret; _enter(""); - /* allocate a superblock info record */ - as = kzalloc(sizeof(struct afs_super_info), GFP_KERNEL); - if (!as) { - _leave(" = -ENOMEM"); - return -ENOMEM; - } - - afs_get_volume(params->volume); - as->volume = params->volume; - /* fill in the superblock */ sb->s_blocksize = PAGE_CACHE_SIZE; sb->s_blocksize_bits = PAGE_CACHE_SHIFT; sb->s_magic = AFS_FS_MAGIC; sb->s_op = &afs_super_ops; - sb->s_fs_info = as; + sb->s_bdi = &as->volume->bdi; + strlcpy(sb->s_id, as->volume->vlocation->vldb.name, sizeof(sb->s_id)); /* allocate the root inode and dentry */ fid.vid = as->volume->vid; @@ -321,28 +328,22 @@ static int afs_fill_super(struct super_block *sb, void *data) fid.unique = 1; inode = afs_iget(sb, params->key, &fid, NULL, NULL); if (IS_ERR(inode)) - goto error_inode; + return PTR_ERR(inode); + + if (params->autocell) + set_bit(AFS_VNODE_AUTOCELL, &AFS_FS_I(inode)->flags); ret = -ENOMEM; - root = d_alloc_root(inode); - if (!root) + sb->s_root = d_make_root(inode); + if (!sb->s_root) goto error; - sb->s_root = root; + sb->s_d_op = &afs_fs_dentry_operations; _leave(" = 0"); return 0; -error_inode: - ret = PTR_ERR(inode); - inode = NULL; error: - iput(inode); - afs_put_volume(as->volume); - kfree(as); - - sb->s_fs_info = NULL; - _leave(" = %d", ret); return ret; } @@ -350,22 +351,25 @@ error: /* * get an AFS superblock */ -static int afs_get_sb(struct file_system_type *fs_type, - int flags, - const char *dev_name, - void *options, - struct vfsmount *mnt) +static struct dentry *afs_mount(struct file_system_type *fs_type, + int flags, const char *dev_name, void *options) { struct afs_mount_params params; struct super_block *sb; struct afs_volume *vol; struct key *key; + char *new_opts = kstrdup(options, GFP_KERNEL); + struct afs_super_info *as; int ret; _enter(",,%s,%p", dev_name, options); memset(¶ms, 0, sizeof(params)); + ret = -EINVAL; + if (current->nsproxy->net_ns != &init_net) + goto error; + /* parse the options and device name */ if (options) { ret = afs_parse_options(¶ms, options, &dev_name); @@ -392,76 +396,82 @@ static int afs_get_sb(struct file_system_type *fs_type, ret = PTR_ERR(vol); goto error; } - params.volume = vol; + + /* allocate a superblock info record */ + as = kzalloc(sizeof(struct afs_super_info), GFP_KERNEL); + if (!as) { + ret = -ENOMEM; + afs_put_volume(vol); + goto error; + } + as->volume = vol; /* allocate a deviceless superblock */ - sb = sget(fs_type, afs_test_super, set_anon_super, ¶ms); + sb = sget(fs_type, afs_test_super, afs_set_super, flags, as); if (IS_ERR(sb)) { ret = PTR_ERR(sb); + afs_put_volume(vol); + kfree(as); goto error; } if (!sb->s_root) { /* initial superblock/root creation */ _debug("create"); - sb->s_flags = flags; ret = afs_fill_super(sb, ¶ms); if (ret < 0) { - up_write(&sb->s_umount); - deactivate_super(sb); + deactivate_locked_super(sb); goto error; } + save_mount_options(sb, new_opts); sb->s_flags |= MS_ACTIVE; } else { _debug("reuse"); ASSERTCMP(sb->s_flags, &, MS_ACTIVE); + afs_put_volume(vol); + kfree(as); } - simple_set_mnt(mnt, sb); - afs_put_volume(params.volume); afs_put_cell(params.cell); + kfree(new_opts); _leave(" = 0 [%p]", sb); - return 0; + return dget(sb->s_root); error: - afs_put_volume(params.volume); afs_put_cell(params.cell); key_put(params.key); + kfree(new_opts); _leave(" = %d", ret); - return ret; + return ERR_PTR(ret); } -/* - * finish the unmounting process on the superblock - */ -static void afs_put_super(struct super_block *sb) +static void afs_kill_super(struct super_block *sb) { struct afs_super_info *as = sb->s_fs_info; - - _enter(""); - + kill_anon_super(sb); afs_put_volume(as->volume); - - _leave(""); + kfree(as); } /* * initialise an inode cache slab element prior to any use */ -static void afs_i_init_once(void *_vnode, struct kmem_cache *cachep, - unsigned long flags) +static void afs_i_init_once(void *_vnode) { struct afs_vnode *vnode = _vnode; - if (flags & SLAB_CTOR_CONSTRUCTOR) { - memset(vnode, 0, sizeof(*vnode)); - inode_init_once(&vnode->vfs_inode); - init_waitqueue_head(&vnode->update_waitq); - mutex_init(&vnode->permits_lock); - mutex_init(&vnode->validate_lock); - spin_lock_init(&vnode->lock); - INIT_WORK(&vnode->cb_broken_work, afs_broken_callback_work); - } + memset(vnode, 0, sizeof(*vnode)); + inode_init_once(&vnode->vfs_inode); + init_waitqueue_head(&vnode->update_waitq); + mutex_init(&vnode->permits_lock); + mutex_init(&vnode->validate_lock); + spin_lock_init(&vnode->writeback_lock); + spin_lock_init(&vnode->lock); + INIT_LIST_HEAD(&vnode->writebacks); + INIT_LIST_HEAD(&vnode->pending_locks); + INIT_LIST_HEAD(&vnode->granted_locks); + INIT_DELAYED_WORK(&vnode->lock_work, afs_lock_work); + INIT_WORK(&vnode->cb_broken_work, afs_broken_callback_work); } /* @@ -485,9 +495,17 @@ static struct inode *afs_alloc_inode(struct super_block *sb) vnode->flags = 1 << AFS_VNODE_UNSET; vnode->cb_promised = false; + _leave(" = %p", &vnode->vfs_inode); return &vnode->vfs_inode; } +static void afs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + struct afs_vnode *vnode = AFS_FS_I(inode); + kmem_cache_free(afs_inode_cachep, vnode); +} + /* * destroy an AFS inode struct */ @@ -495,12 +513,45 @@ static void afs_destroy_inode(struct inode *inode) { struct afs_vnode *vnode = AFS_FS_I(inode); - _enter("{%lu}", inode->i_ino); + _enter("%p{%x:%u}", inode, vnode->fid.vid, vnode->fid.vnode); _debug("DESTROY INODE %p", inode); ASSERTCMP(vnode->server, ==, NULL); - kmem_cache_free(afs_inode_cachep, vnode); + call_rcu(&inode->i_rcu, afs_i_callback); atomic_dec(&afs_count_active_inodes); } + +/* + * return information about an AFS volume + */ +static int afs_statfs(struct dentry *dentry, struct kstatfs *buf) +{ + struct afs_volume_status vs; + struct afs_vnode *vnode = AFS_FS_I(dentry->d_inode); + struct key *key; + int ret; + + key = afs_request_key(vnode->volume->cell); + if (IS_ERR(key)) + return PTR_ERR(key); + + ret = afs_vnode_get_volume_status(vnode, key, &vs); + key_put(key); + if (ret < 0) { + _leave(" = %d", ret); + return ret; + } + + buf->f_type = dentry->d_sb->s_magic; + buf->f_bsize = AFS_BLOCK_SIZE; + buf->f_namelen = AFSNAMEMAX - 1; + + if (vs.max_quota == 0) + buf->f_blocks = vs.part_max_blocks; + else + buf->f_blocks = vs.max_quota; + buf->f_bavail = buf->f_bfree = buf->f_blocks - vs.blocks_in_use; + return 0; +} diff --git a/fs/afs/vlclient.c b/fs/afs/vlclient.c index 36c1306e09e..340afd0cd18 100644 --- a/fs/afs/vlclient.c +++ b/fs/afs/vlclient.c @@ -9,6 +9,7 @@ * 2 of the License, or (at your option) any later version. */ +#include <linux/gfp.h> #include <linux/init.h> #include <linux/sched.h> #include "internal.h" diff --git a/fs/afs/vlocation.c b/fs/afs/vlocation.c index 3370cdb7256..b6df2e83809 100644 --- a/fs/afs/vlocation.c +++ b/fs/afs/vlocation.c @@ -11,11 +11,13 @@ #include <linux/kernel.h> #include <linux/module.h> +#include <linux/slab.h> #include <linux/init.h> +#include <linux/sched.h> #include "internal.h" -unsigned afs_vlocation_timeout = 10; /* volume location timeout in seconds */ -unsigned afs_vlocation_update_timeout = 10 * 60; +static unsigned afs_vlocation_timeout = 10; /* volume location timeout in seconds */ +static unsigned afs_vlocation_update_timeout = 10 * 60; static void afs_vlocation_reaper(struct work_struct *); static void afs_vlocation_updater(struct work_struct *); @@ -64,6 +66,8 @@ static int afs_vlocation_access_vl_by_name(struct afs_vlocation *vl, goto out; goto rotate; case -ENOMEDIUM: + case -EKEYREJECTED: + case -EKEYEXPIRED: goto out; default: ret = -EIO; @@ -280,9 +284,8 @@ static void afs_vlocation_apply_update(struct afs_vlocation *vl, vl->vldb = *vldb; -#ifdef AFS_CACHING_SUPPORT - /* update volume entry in local cache */ - cachefs_update_cookie(vl->cache); +#ifdef CONFIG_AFS_FSCACHE + fscache_update_cookie(vl->cache); #endif } @@ -303,11 +306,10 @@ static int afs_vlocation_fill_in_record(struct afs_vlocation *vl, memset(&vldb, 0, sizeof(vldb)); /* see if we have an in-cache copy (will set vl->valid if there is) */ -#ifdef AFS_CACHING_SUPPORT - cachefs_acquire_cookie(cell->cache, - &afs_volume_cache_index_def, - vlocation, - &vl->cache); +#ifdef CONFIG_AFS_FSCACHE + vl->cache = fscache_acquire_cookie(vl->cell->cache, + &afs_vlocation_cache_index_def, vl, + true); #endif if (vl->valid) { @@ -334,7 +336,7 @@ static int afs_vlocation_fill_in_record(struct afs_vlocation *vl, /* * queue a vlocation record for updates */ -void afs_vlocation_queue_for_updates(struct afs_vlocation *vl) +static void afs_vlocation_queue_for_updates(struct afs_vlocation *vl) { struct afs_vlocation *xvl; @@ -381,7 +383,7 @@ struct afs_vlocation *afs_vlocation_lookup(struct afs_cell *cell, cell->name, key_serial(key), (int) namesz, (int) namesz, name, namesz); - if (namesz > sizeof(vl->vldb.name)) { + if (namesz >= sizeof(vl->vldb.name)) { _leave(" = -ENAMETOOLONG"); return ERR_PTR(-ENAMETOOLONG); } @@ -419,6 +421,11 @@ fill_in_record: spin_unlock(&vl->lock); wake_up(&vl->waitq); + /* update volume entry in local cache */ +#ifdef CONFIG_AFS_FSCACHE + fscache_update_cookie(vl->cache); +#endif + /* schedule for regular updates */ afs_vlocation_queue_for_updates(vl); goto success; @@ -464,7 +471,7 @@ found_in_memory: spin_unlock(&vl->lock); success: - _leave(" = %p",vl); + _leave(" = %p", vl); return vl; error_abandon: @@ -501,8 +508,8 @@ void afs_put_vlocation(struct afs_vlocation *vl) _debug("buried"); list_move_tail(&vl->grave, &afs_vlocation_graveyard); vl->time_of_death = get_seconds(); - schedule_delayed_work(&afs_vlocation_reap, - afs_vlocation_timeout * HZ); + queue_delayed_work(afs_wq, &afs_vlocation_reap, + afs_vlocation_timeout * HZ); /* suspend updates on this record */ if (!list_empty(&vl->update)) { @@ -522,10 +529,9 @@ static void afs_vlocation_destroy(struct afs_vlocation *vl) { _enter("%p", vl); -#ifdef AFS_CACHING_SUPPORT - cachefs_relinquish_cookie(vl->cache, 0); +#ifdef CONFIG_AFS_FSCACHE + fscache_relinquish_cookie(vl->cache, 0); #endif - afs_put_cell(vl->cell); kfree(vl); } @@ -556,12 +562,7 @@ static void afs_vlocation_reaper(struct work_struct *work) if (expiry > now) { delay = (expiry - now) * HZ; _debug("delay %lu", delay); - if (!schedule_delayed_work(&afs_vlocation_reap, - delay)) { - cancel_delayed_work(&afs_vlocation_reap); - schedule_delayed_work(&afs_vlocation_reap, - delay); - } + mod_delayed_work(afs_wq, &afs_vlocation_reap, delay); break; } @@ -609,13 +610,10 @@ void afs_vlocation_purge(void) spin_lock(&afs_vlocation_updates_lock); list_del_init(&afs_vlocation_updates); spin_unlock(&afs_vlocation_updates_lock); - cancel_delayed_work(&afs_vlocation_update); - queue_delayed_work(afs_vlocation_update_worker, - &afs_vlocation_update, 0); + mod_delayed_work(afs_vlocation_update_worker, &afs_vlocation_update, 0); destroy_workqueue(afs_vlocation_update_worker); - cancel_delayed_work(&afs_vlocation_reap); - schedule_delayed_work(&afs_vlocation_reap, 0); + mod_delayed_work(afs_wq, &afs_vlocation_reap, 0); } /* diff --git a/fs/afs/vnode.c b/fs/afs/vnode.c index a1904ab8426..25cf4c3f4ff 100644 --- a/fs/afs/vnode.c +++ b/fs/afs/vnode.c @@ -12,8 +12,8 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/init.h> -#include <linux/slab.h> #include <linux/fs.h> +#include <linux/sched.h> #include "internal.h" #if 0 @@ -175,24 +175,33 @@ static void afs_vnode_deleted_remotely(struct afs_vnode *vnode) { struct afs_server *server; + _enter("{%p}", vnode->server); + set_bit(AFS_VNODE_DELETED, &vnode->flags); server = vnode->server; - if (vnode->cb_promised) { - spin_lock(&server->cb_lock); + if (server) { if (vnode->cb_promised) { - rb_erase(&vnode->cb_promise, &server->cb_promises); - vnode->cb_promised = false; + spin_lock(&server->cb_lock); + if (vnode->cb_promised) { + rb_erase(&vnode->cb_promise, + &server->cb_promises); + vnode->cb_promised = false; + } + spin_unlock(&server->cb_lock); } - spin_unlock(&server->cb_lock); - } - spin_lock(&vnode->server->fs_lock); - rb_erase(&vnode->server_rb, &vnode->server->fs_vnodes); - spin_unlock(&vnode->server->fs_lock); + spin_lock(&server->fs_lock); + rb_erase(&vnode->server_rb, &server->fs_vnodes); + spin_unlock(&server->fs_lock); - vnode->server = NULL; - afs_put_server(server); + vnode->server = NULL; + afs_put_server(server); + } else { + ASSERT(!vnode->cb_promised); + } + + _leave(""); } /* @@ -225,7 +234,7 @@ void afs_vnode_finalise_status_update(struct afs_vnode *vnode, */ static void afs_vnode_status_update_failed(struct afs_vnode *vnode, int ret) { - _enter("%p,%d", vnode, ret); + _enter("{%x:%u},%d", vnode->fid.vid, vnode->fid.vnode, ret); spin_lock(&vnode->lock); @@ -261,7 +270,7 @@ int afs_vnode_fetch_status(struct afs_vnode *vnode, DECLARE_WAITQUEUE(myself, current); - _enter("%s,{%u,%u,%u}", + _enter("%s,{%x:%u.%u}", vnode->volume->vlocation->vldb.name, vnode->fid.vid, vnode->fid.vnode, vnode->fid.unique); @@ -389,7 +398,7 @@ int afs_vnode_fetch_data(struct afs_vnode *vnode, struct key *key, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%x,,,", + _enter("%s{%x:%u.%u},%x,,,", vnode->volume->vlocation->vldb.name, vnode->fid.vid, vnode->fid.vnode, @@ -446,7 +455,7 @@ int afs_vnode_create(struct afs_vnode *vnode, struct key *key, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%x,%s,,", + _enter("%s{%x:%u.%u},%x,%s,,", vnode->volume->vlocation->vldb.name, vnode->fid.vid, vnode->fid.vnode, @@ -502,7 +511,7 @@ int afs_vnode_remove(struct afs_vnode *vnode, struct key *key, const char *name, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%x,%s", + _enter("%s{%x:%u.%u},%x,%s", vnode->volume->vlocation->vldb.name, vnode->fid.vid, vnode->fid.vnode, @@ -551,13 +560,13 @@ no_server: /* * create a hard link */ -extern int afs_vnode_link(struct afs_vnode *dvnode, struct afs_vnode *vnode, +int afs_vnode_link(struct afs_vnode *dvnode, struct afs_vnode *vnode, struct key *key, const char *name) { struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%s{%u,%u,%u},%x,%s", + _enter("%s{%x:%u.%u},%s{%x:%u.%u},%x,%s", dvnode->volume->vlocation->vldb.name, dvnode->fid.vid, dvnode->fid.vnode, @@ -628,7 +637,7 @@ int afs_vnode_symlink(struct afs_vnode *vnode, struct key *key, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%x,%s,%s,,,", + _enter("%s{%x:%u.%u},%x,%s,%s,,,", vnode->volume->vlocation->vldb.name, vnode->fid.vid, vnode->fid.vnode, @@ -687,7 +696,7 @@ int afs_vnode_rename(struct afs_vnode *orig_dvnode, struct afs_server *server; int ret; - _enter("%s{%u,%u,%u},%s{%u,%u,%u},%x,%s,%s", + _enter("%s{%x:%u.%u},%s{%u,%u,%u},%x,%s,%s", orig_dvnode->volume->vlocation->vldb.name, orig_dvnode->fid.vid, orig_dvnode->fid.vnode, @@ -753,3 +762,264 @@ no_server: _leave(" = %ld [cnt %d]", PTR_ERR(server), orig_dvnode->update_cnt); return PTR_ERR(server); } + +/* + * write to a file + */ +int afs_vnode_store_data(struct afs_writeback *wb, pgoff_t first, pgoff_t last, + unsigned offset, unsigned to) +{ + struct afs_server *server; + struct afs_vnode *vnode = wb->vnode; + int ret; + + _enter("%s{%x:%u.%u},%x,%lx,%lx,%x,%x", + vnode->volume->vlocation->vldb.name, + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + key_serial(wb->key), + first, last, offset, to); + + /* this op will fetch the status */ + spin_lock(&vnode->lock); + vnode->update_cnt++; + spin_unlock(&vnode->lock); + + do { + /* pick a server to query */ + server = afs_volume_pick_fileserver(vnode); + if (IS_ERR(server)) + goto no_server; + + _debug("USING SERVER: %08x\n", ntohl(server->addr.s_addr)); + + ret = afs_fs_store_data(server, wb, first, last, offset, to, + &afs_sync_call); + + } while (!afs_volume_release_fileserver(vnode, server, ret)); + + /* adjust the flags */ + if (ret == 0) { + afs_vnode_finalise_status_update(vnode, server); + afs_put_server(server); + } else { + afs_vnode_status_update_failed(vnode, ret); + } + + _leave(" = %d", ret); + return ret; + +no_server: + spin_lock(&vnode->lock); + vnode->update_cnt--; + ASSERTCMP(vnode->update_cnt, >=, 0); + spin_unlock(&vnode->lock); + return PTR_ERR(server); +} + +/* + * set the attributes on a file + */ +int afs_vnode_setattr(struct afs_vnode *vnode, struct key *key, + struct iattr *attr) +{ + struct afs_server *server; + int ret; + + _enter("%s{%x:%u.%u},%x", + vnode->volume->vlocation->vldb.name, + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + key_serial(key)); + + /* this op will fetch the status */ + spin_lock(&vnode->lock); + vnode->update_cnt++; + spin_unlock(&vnode->lock); + + do { + /* pick a server to query */ + server = afs_volume_pick_fileserver(vnode); + if (IS_ERR(server)) + goto no_server; + + _debug("USING SERVER: %08x\n", ntohl(server->addr.s_addr)); + + ret = afs_fs_setattr(server, key, vnode, attr, &afs_sync_call); + + } while (!afs_volume_release_fileserver(vnode, server, ret)); + + /* adjust the flags */ + if (ret == 0) { + afs_vnode_finalise_status_update(vnode, server); + afs_put_server(server); + } else { + afs_vnode_status_update_failed(vnode, ret); + } + + _leave(" = %d", ret); + return ret; + +no_server: + spin_lock(&vnode->lock); + vnode->update_cnt--; + ASSERTCMP(vnode->update_cnt, >=, 0); + spin_unlock(&vnode->lock); + return PTR_ERR(server); +} + +/* + * get the status of a volume + */ +int afs_vnode_get_volume_status(struct afs_vnode *vnode, struct key *key, + struct afs_volume_status *vs) +{ + struct afs_server *server; + int ret; + + _enter("%s{%x:%u.%u},%x,", + vnode->volume->vlocation->vldb.name, + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + key_serial(key)); + + do { + /* pick a server to query */ + server = afs_volume_pick_fileserver(vnode); + if (IS_ERR(server)) + goto no_server; + + _debug("USING SERVER: %08x\n", ntohl(server->addr.s_addr)); + + ret = afs_fs_get_volume_status(server, key, vnode, vs, &afs_sync_call); + + } while (!afs_volume_release_fileserver(vnode, server, ret)); + + /* adjust the flags */ + if (ret == 0) + afs_put_server(server); + + _leave(" = %d", ret); + return ret; + +no_server: + return PTR_ERR(server); +} + +/* + * get a lock on a file + */ +int afs_vnode_set_lock(struct afs_vnode *vnode, struct key *key, + afs_lock_type_t type) +{ + struct afs_server *server; + int ret; + + _enter("%s{%x:%u.%u},%x,%u", + vnode->volume->vlocation->vldb.name, + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + key_serial(key), type); + + do { + /* pick a server to query */ + server = afs_volume_pick_fileserver(vnode); + if (IS_ERR(server)) + goto no_server; + + _debug("USING SERVER: %08x\n", ntohl(server->addr.s_addr)); + + ret = afs_fs_set_lock(server, key, vnode, type, &afs_sync_call); + + } while (!afs_volume_release_fileserver(vnode, server, ret)); + + /* adjust the flags */ + if (ret == 0) + afs_put_server(server); + + _leave(" = %d", ret); + return ret; + +no_server: + return PTR_ERR(server); +} + +/* + * extend a lock on a file + */ +int afs_vnode_extend_lock(struct afs_vnode *vnode, struct key *key) +{ + struct afs_server *server; + int ret; + + _enter("%s{%x:%u.%u},%x", + vnode->volume->vlocation->vldb.name, + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + key_serial(key)); + + do { + /* pick a server to query */ + server = afs_volume_pick_fileserver(vnode); + if (IS_ERR(server)) + goto no_server; + + _debug("USING SERVER: %08x\n", ntohl(server->addr.s_addr)); + + ret = afs_fs_extend_lock(server, key, vnode, &afs_sync_call); + + } while (!afs_volume_release_fileserver(vnode, server, ret)); + + /* adjust the flags */ + if (ret == 0) + afs_put_server(server); + + _leave(" = %d", ret); + return ret; + +no_server: + return PTR_ERR(server); +} + +/* + * release a lock on a file + */ +int afs_vnode_release_lock(struct afs_vnode *vnode, struct key *key) +{ + struct afs_server *server; + int ret; + + _enter("%s{%x:%u.%u},%x", + vnode->volume->vlocation->vldb.name, + vnode->fid.vid, + vnode->fid.vnode, + vnode->fid.unique, + key_serial(key)); + + do { + /* pick a server to query */ + server = afs_volume_pick_fileserver(vnode); + if (IS_ERR(server)) + goto no_server; + + _debug("USING SERVER: %08x\n", ntohl(server->addr.s_addr)); + + ret = afs_fs_release_lock(server, key, vnode, &afs_sync_call); + + } while (!afs_volume_release_fileserver(vnode, server, ret)); + + /* adjust the flags */ + if (ret == 0) + afs_put_server(server); + + _leave(" = %d", ret); + return ret; + +no_server: + return PTR_ERR(server); +} diff --git a/fs/afs/volume.c b/fs/afs/volume.c index dd160cada45..2b607257820 100644 --- a/fs/afs/volume.c +++ b/fs/afs/volume.c @@ -15,6 +15,7 @@ #include <linux/slab.h> #include <linux/fs.h> #include <linux/pagemap.h> +#include <linux/sched.h> #include "internal.h" static const char *afs_voltypes[] = { "R/W", "R/O", "BAK" }; @@ -105,6 +106,10 @@ struct afs_volume *afs_volume_lookup(struct afs_mount_params *params) volume->cell = params->cell; volume->vid = vlocation->vldb.vid[params->type]; + ret = bdi_setup_and_register(&volume->bdi, "afs", BDI_CAP_MAP_COPY); + if (ret) + goto error_bdi; + init_rwsem(&volume->server_sem); /* look up all the applicable server records */ @@ -123,13 +128,11 @@ struct afs_volume *afs_volume_lookup(struct afs_mount_params *params) } /* attach the cache and volume location */ -#ifdef AFS_CACHING_SUPPORT - cachefs_acquire_cookie(vlocation->cache, - &afs_vnode_cache_index_def, - volume, - &volume->cache); +#ifdef CONFIG_AFS_FSCACHE + volume->cache = fscache_acquire_cookie(vlocation->cache, + &afs_volume_cache_index_def, + volume, true); #endif - afs_get_vlocation(vlocation); volume->vlocation = vlocation; @@ -152,6 +155,8 @@ error: return ERR_PTR(ret); error_discard: + bdi_destroy(&volume->bdi); +error_bdi: up_write(¶ms->cell->vl_sem); for (loop = volume->nservers - 1; loop >= 0; loop--) @@ -193,14 +198,15 @@ void afs_put_volume(struct afs_volume *volume) up_write(&vlocation->cell->vl_sem); /* finish cleaning up the volume */ -#ifdef AFS_CACHING_SUPPORT - cachefs_relinquish_cookie(volume->cache, 0); +#ifdef CONFIG_AFS_FSCACHE + fscache_relinquish_cookie(volume->cache, 0); #endif afs_put_vlocation(vlocation); for (loop = volume->nservers - 1; loop >= 0; loop--) afs_put_server(volume->servers[loop]); + bdi_destroy(&volume->bdi); kfree(volume); _leave(" [destroyed]"); diff --git a/fs/afs/write.c b/fs/afs/write.c new file mode 100644 index 00000000000..ab6adfd5251 --- /dev/null +++ b/fs/afs/write.c @@ -0,0 +1,763 @@ +/* handling of writes to regular files and writing back to the server + * + * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ +#include <linux/backing-dev.h> +#include <linux/slab.h> +#include <linux/fs.h> +#include <linux/pagemap.h> +#include <linux/writeback.h> +#include <linux/pagevec.h> +#include <linux/aio.h> +#include "internal.h" + +static int afs_write_back_from_locked_page(struct afs_writeback *wb, + struct page *page); + +/* + * mark a page as having been made dirty and thus needing writeback + */ +int afs_set_page_dirty(struct page *page) +{ + _enter(""); + return __set_page_dirty_nobuffers(page); +} + +/* + * unlink a writeback record because its usage has reached zero + * - must be called with the wb->vnode->writeback_lock held + */ +static void afs_unlink_writeback(struct afs_writeback *wb) +{ + struct afs_writeback *front; + struct afs_vnode *vnode = wb->vnode; + + list_del_init(&wb->link); + if (!list_empty(&vnode->writebacks)) { + /* if an fsync rises to the front of the queue then wake it + * up */ + front = list_entry(vnode->writebacks.next, + struct afs_writeback, link); + if (front->state == AFS_WBACK_SYNCING) { + _debug("wake up sync"); + front->state = AFS_WBACK_COMPLETE; + wake_up(&front->waitq); + } + } +} + +/* + * free a writeback record + */ +static void afs_free_writeback(struct afs_writeback *wb) +{ + _enter(""); + key_put(wb->key); + kfree(wb); +} + +/* + * dispose of a reference to a writeback record + */ +void afs_put_writeback(struct afs_writeback *wb) +{ + struct afs_vnode *vnode = wb->vnode; + + _enter("{%d}", wb->usage); + + spin_lock(&vnode->writeback_lock); + if (--wb->usage == 0) + afs_unlink_writeback(wb); + else + wb = NULL; + spin_unlock(&vnode->writeback_lock); + if (wb) + afs_free_writeback(wb); +} + +/* + * partly or wholly fill a page that's under preparation for writing + */ +static int afs_fill_page(struct afs_vnode *vnode, struct key *key, + loff_t pos, struct page *page) +{ + loff_t i_size; + int ret; + int len; + + _enter(",,%llu", (unsigned long long)pos); + + i_size = i_size_read(&vnode->vfs_inode); + if (pos + PAGE_CACHE_SIZE > i_size) + len = i_size - pos; + else + len = PAGE_CACHE_SIZE; + + ret = afs_vnode_fetch_data(vnode, key, pos, len, page); + if (ret < 0) { + if (ret == -ENOENT) { + _debug("got NOENT from server" + " - marking file deleted and stale"); + set_bit(AFS_VNODE_DELETED, &vnode->flags); + ret = -ESTALE; + } + } + + _leave(" = %d", ret); + return ret; +} + +/* + * prepare to perform part of a write to a page + */ +int afs_write_begin(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned flags, + struct page **pagep, void **fsdata) +{ + struct afs_writeback *candidate, *wb; + struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); + struct page *page; + struct key *key = file->private_data; + unsigned from = pos & (PAGE_CACHE_SIZE - 1); + unsigned to = from + len; + pgoff_t index = pos >> PAGE_CACHE_SHIFT; + int ret; + + _enter("{%x:%u},{%lx},%u,%u", + vnode->fid.vid, vnode->fid.vnode, index, from, to); + + candidate = kzalloc(sizeof(*candidate), GFP_KERNEL); + if (!candidate) + return -ENOMEM; + candidate->vnode = vnode; + candidate->first = candidate->last = index; + candidate->offset_first = from; + candidate->to_last = to; + INIT_LIST_HEAD(&candidate->link); + candidate->usage = 1; + candidate->state = AFS_WBACK_PENDING; + init_waitqueue_head(&candidate->waitq); + + page = grab_cache_page_write_begin(mapping, index, flags); + if (!page) { + kfree(candidate); + return -ENOMEM; + } + *pagep = page; + /* page won't leak in error case: it eventually gets cleaned off LRU */ + + if (!PageUptodate(page) && len != PAGE_CACHE_SIZE) { + ret = afs_fill_page(vnode, key, index << PAGE_CACHE_SHIFT, page); + if (ret < 0) { + kfree(candidate); + _leave(" = %d [prep]", ret); + return ret; + } + SetPageUptodate(page); + } + +try_again: + spin_lock(&vnode->writeback_lock); + + /* see if this page is already pending a writeback under a suitable key + * - if so we can just join onto that one */ + wb = (struct afs_writeback *) page_private(page); + if (wb) { + if (wb->key == key && wb->state == AFS_WBACK_PENDING) + goto subsume_in_current_wb; + goto flush_conflicting_wb; + } + + if (index > 0) { + /* see if we can find an already pending writeback that we can + * append this page to */ + list_for_each_entry(wb, &vnode->writebacks, link) { + if (wb->last == index - 1 && wb->key == key && + wb->state == AFS_WBACK_PENDING) + goto append_to_previous_wb; + } + } + + list_add_tail(&candidate->link, &vnode->writebacks); + candidate->key = key_get(key); + spin_unlock(&vnode->writeback_lock); + SetPagePrivate(page); + set_page_private(page, (unsigned long) candidate); + _leave(" = 0 [new]"); + return 0; + +subsume_in_current_wb: + _debug("subsume"); + ASSERTRANGE(wb->first, <=, index, <=, wb->last); + if (index == wb->first && from < wb->offset_first) + wb->offset_first = from; + if (index == wb->last && to > wb->to_last) + wb->to_last = to; + spin_unlock(&vnode->writeback_lock); + kfree(candidate); + _leave(" = 0 [sub]"); + return 0; + +append_to_previous_wb: + _debug("append into %lx-%lx", wb->first, wb->last); + wb->usage++; + wb->last++; + wb->to_last = to; + spin_unlock(&vnode->writeback_lock); + SetPagePrivate(page); + set_page_private(page, (unsigned long) wb); + kfree(candidate); + _leave(" = 0 [app]"); + return 0; + + /* the page is currently bound to another context, so if it's dirty we + * need to flush it before we can use the new context */ +flush_conflicting_wb: + _debug("flush conflict"); + if (wb->state == AFS_WBACK_PENDING) + wb->state = AFS_WBACK_CONFLICTING; + spin_unlock(&vnode->writeback_lock); + if (PageDirty(page)) { + ret = afs_write_back_from_locked_page(wb, page); + if (ret < 0) { + afs_put_writeback(candidate); + _leave(" = %d", ret); + return ret; + } + } + + /* the page holds a ref on the writeback record */ + afs_put_writeback(wb); + set_page_private(page, 0); + ClearPagePrivate(page); + goto try_again; +} + +/* + * finalise part of a write to a page + */ +int afs_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ + struct afs_vnode *vnode = AFS_FS_I(file_inode(file)); + loff_t i_size, maybe_i_size; + + _enter("{%x:%u},{%lx}", + vnode->fid.vid, vnode->fid.vnode, page->index); + + maybe_i_size = pos + copied; + + i_size = i_size_read(&vnode->vfs_inode); + if (maybe_i_size > i_size) { + spin_lock(&vnode->writeback_lock); + i_size = i_size_read(&vnode->vfs_inode); + if (maybe_i_size > i_size) + i_size_write(&vnode->vfs_inode, maybe_i_size); + spin_unlock(&vnode->writeback_lock); + } + + set_page_dirty(page); + if (PageDirty(page)) + _debug("dirtied"); + unlock_page(page); + page_cache_release(page); + + return copied; +} + +/* + * kill all the pages in the given range + */ +static void afs_kill_pages(struct afs_vnode *vnode, bool error, + pgoff_t first, pgoff_t last) +{ + struct pagevec pv; + unsigned count, loop; + + _enter("{%x:%u},%lx-%lx", + vnode->fid.vid, vnode->fid.vnode, first, last); + + pagevec_init(&pv, 0); + + do { + _debug("kill %lx-%lx", first, last); + + count = last - first + 1; + if (count > PAGEVEC_SIZE) + count = PAGEVEC_SIZE; + pv.nr = find_get_pages_contig(vnode->vfs_inode.i_mapping, + first, count, pv.pages); + ASSERTCMP(pv.nr, ==, count); + + for (loop = 0; loop < count; loop++) { + ClearPageUptodate(pv.pages[loop]); + if (error) + SetPageError(pv.pages[loop]); + end_page_writeback(pv.pages[loop]); + } + + __pagevec_release(&pv); + } while (first < last); + + _leave(""); +} + +/* + * synchronously write back the locked page and any subsequent non-locked dirty + * pages also covered by the same writeback record + */ +static int afs_write_back_from_locked_page(struct afs_writeback *wb, + struct page *primary_page) +{ + struct page *pages[8], *page; + unsigned long count; + unsigned n, offset, to; + pgoff_t start, first, last; + int loop, ret; + + _enter(",%lx", primary_page->index); + + count = 1; + if (!clear_page_dirty_for_io(primary_page)) + BUG(); + if (test_set_page_writeback(primary_page)) + BUG(); + + /* find all consecutive lockable dirty pages, stopping when we find a + * page that is not immediately lockable, is not dirty or is missing, + * or we reach the end of the range */ + start = primary_page->index; + if (start >= wb->last) + goto no_more; + start++; + do { + _debug("more %lx [%lx]", start, count); + n = wb->last - start + 1; + if (n > ARRAY_SIZE(pages)) + n = ARRAY_SIZE(pages); + n = find_get_pages_contig(wb->vnode->vfs_inode.i_mapping, + start, n, pages); + _debug("fgpc %u", n); + if (n == 0) + goto no_more; + if (pages[0]->index != start) { + do { + put_page(pages[--n]); + } while (n > 0); + goto no_more; + } + + for (loop = 0; loop < n; loop++) { + page = pages[loop]; + if (page->index > wb->last) + break; + if (!trylock_page(page)) + break; + if (!PageDirty(page) || + page_private(page) != (unsigned long) wb) { + unlock_page(page); + break; + } + if (!clear_page_dirty_for_io(page)) + BUG(); + if (test_set_page_writeback(page)) + BUG(); + unlock_page(page); + put_page(page); + } + count += loop; + if (loop < n) { + for (; loop < n; loop++) + put_page(pages[loop]); + goto no_more; + } + + start += loop; + } while (start <= wb->last && count < 65536); + +no_more: + /* we now have a contiguous set of dirty pages, each with writeback set + * and the dirty mark cleared; the first page is locked and must remain + * so, all the rest are unlocked */ + first = primary_page->index; + last = first + count - 1; + + offset = (first == wb->first) ? wb->offset_first : 0; + to = (last == wb->last) ? wb->to_last : PAGE_SIZE; + + _debug("write back %lx[%u..] to %lx[..%u]", first, offset, last, to); + + ret = afs_vnode_store_data(wb, first, last, offset, to); + if (ret < 0) { + switch (ret) { + case -EDQUOT: + case -ENOSPC: + set_bit(AS_ENOSPC, + &wb->vnode->vfs_inode.i_mapping->flags); + break; + case -EROFS: + case -EIO: + case -EREMOTEIO: + case -EFBIG: + case -ENOENT: + case -ENOMEDIUM: + case -ENXIO: + afs_kill_pages(wb->vnode, true, first, last); + set_bit(AS_EIO, &wb->vnode->vfs_inode.i_mapping->flags); + break; + case -EACCES: + case -EPERM: + case -ENOKEY: + case -EKEYEXPIRED: + case -EKEYREJECTED: + case -EKEYREVOKED: + afs_kill_pages(wb->vnode, false, first, last); + break; + default: + break; + } + } else { + ret = count; + } + + _leave(" = %d", ret); + return ret; +} + +/* + * write a page back to the server + * - the caller locked the page for us + */ +int afs_writepage(struct page *page, struct writeback_control *wbc) +{ + struct afs_writeback *wb; + int ret; + + _enter("{%lx},", page->index); + + wb = (struct afs_writeback *) page_private(page); + ASSERT(wb != NULL); + + ret = afs_write_back_from_locked_page(wb, page); + unlock_page(page); + if (ret < 0) { + _leave(" = %d", ret); + return 0; + } + + wbc->nr_to_write -= ret; + + _leave(" = 0"); + return 0; +} + +/* + * write a region of pages back to the server + */ +static int afs_writepages_region(struct address_space *mapping, + struct writeback_control *wbc, + pgoff_t index, pgoff_t end, pgoff_t *_next) +{ + struct afs_writeback *wb; + struct page *page; + int ret, n; + + _enter(",,%lx,%lx,", index, end); + + do { + n = find_get_pages_tag(mapping, &index, PAGECACHE_TAG_DIRTY, + 1, &page); + if (!n) + break; + + _debug("wback %lx", page->index); + + if (page->index > end) { + *_next = index; + page_cache_release(page); + _leave(" = 0 [%lx]", *_next); + return 0; + } + + /* at this point we hold neither mapping->tree_lock nor lock on + * the page itself: the page may be truncated or invalidated + * (changing page->mapping to NULL), or even swizzled back from + * swapper_space to tmpfs file mapping + */ + lock_page(page); + + if (page->mapping != mapping) { + unlock_page(page); + page_cache_release(page); + continue; + } + + if (wbc->sync_mode != WB_SYNC_NONE) + wait_on_page_writeback(page); + + if (PageWriteback(page) || !PageDirty(page)) { + unlock_page(page); + continue; + } + + wb = (struct afs_writeback *) page_private(page); + ASSERT(wb != NULL); + + spin_lock(&wb->vnode->writeback_lock); + wb->state = AFS_WBACK_WRITING; + spin_unlock(&wb->vnode->writeback_lock); + + ret = afs_write_back_from_locked_page(wb, page); + unlock_page(page); + page_cache_release(page); + if (ret < 0) { + _leave(" = %d", ret); + return ret; + } + + wbc->nr_to_write -= ret; + + cond_resched(); + } while (index < end && wbc->nr_to_write > 0); + + *_next = index; + _leave(" = 0 [%lx]", *_next); + return 0; +} + +/* + * write some of the pending data back to the server + */ +int afs_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + pgoff_t start, end, next; + int ret; + + _enter(""); + + if (wbc->range_cyclic) { + start = mapping->writeback_index; + end = -1; + ret = afs_writepages_region(mapping, wbc, start, end, &next); + if (start > 0 && wbc->nr_to_write > 0 && ret == 0) + ret = afs_writepages_region(mapping, wbc, 0, start, + &next); + mapping->writeback_index = next; + } else if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) { + end = (pgoff_t)(LLONG_MAX >> PAGE_CACHE_SHIFT); + ret = afs_writepages_region(mapping, wbc, 0, end, &next); + if (wbc->nr_to_write > 0) + mapping->writeback_index = next; + } else { + start = wbc->range_start >> PAGE_CACHE_SHIFT; + end = wbc->range_end >> PAGE_CACHE_SHIFT; + ret = afs_writepages_region(mapping, wbc, start, end, &next); + } + + _leave(" = %d", ret); + return ret; +} + +/* + * completion of write to server + */ +void afs_pages_written_back(struct afs_vnode *vnode, struct afs_call *call) +{ + struct afs_writeback *wb = call->wb; + struct pagevec pv; + unsigned count, loop; + pgoff_t first = call->first, last = call->last; + bool free_wb; + + _enter("{%x:%u},{%lx-%lx}", + vnode->fid.vid, vnode->fid.vnode, first, last); + + ASSERT(wb != NULL); + + pagevec_init(&pv, 0); + + do { + _debug("done %lx-%lx", first, last); + + count = last - first + 1; + if (count > PAGEVEC_SIZE) + count = PAGEVEC_SIZE; + pv.nr = find_get_pages_contig(call->mapping, first, count, + pv.pages); + ASSERTCMP(pv.nr, ==, count); + + spin_lock(&vnode->writeback_lock); + for (loop = 0; loop < count; loop++) { + struct page *page = pv.pages[loop]; + end_page_writeback(page); + if (page_private(page) == (unsigned long) wb) { + set_page_private(page, 0); + ClearPagePrivate(page); + wb->usage--; + } + } + free_wb = false; + if (wb->usage == 0) { + afs_unlink_writeback(wb); + free_wb = true; + } + spin_unlock(&vnode->writeback_lock); + first += count; + if (free_wb) { + afs_free_writeback(wb); + wb = NULL; + } + + __pagevec_release(&pv); + } while (first <= last); + + _leave(""); +} + +/* + * write to an AFS file + */ +ssize_t afs_file_write(struct kiocb *iocb, struct iov_iter *from) +{ + struct afs_vnode *vnode = AFS_FS_I(file_inode(iocb->ki_filp)); + ssize_t result; + size_t count = iov_iter_count(from); + + _enter("{%x.%u},{%zu},", + vnode->fid.vid, vnode->fid.vnode, count); + + if (IS_SWAPFILE(&vnode->vfs_inode)) { + printk(KERN_INFO + "AFS: Attempt to write to active swap file!\n"); + return -EBUSY; + } + + if (!count) + return 0; + + result = generic_file_write_iter(iocb, from); + if (IS_ERR_VALUE(result)) { + _leave(" = %zd", result); + return result; + } + + _leave(" = %zd", result); + return result; +} + +/* + * flush the vnode to the fileserver + */ +int afs_writeback_all(struct afs_vnode *vnode) +{ + struct address_space *mapping = vnode->vfs_inode.i_mapping; + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = LONG_MAX, + .range_cyclic = 1, + }; + int ret; + + _enter(""); + + ret = mapping->a_ops->writepages(mapping, &wbc); + __mark_inode_dirty(mapping->host, I_DIRTY_PAGES); + + _leave(" = %d", ret); + return ret; +} + +/* + * flush any dirty pages for this process, and check for write errors. + * - the return status from this call provides a reliable indication of + * whether any write errors occurred for this process. + */ +int afs_fsync(struct file *file, loff_t start, loff_t end, int datasync) +{ + struct dentry *dentry = file->f_path.dentry; + struct inode *inode = file->f_mapping->host; + struct afs_writeback *wb, *xwb; + struct afs_vnode *vnode = AFS_FS_I(dentry->d_inode); + int ret; + + _enter("{%x:%u},{n=%s},%d", + vnode->fid.vid, vnode->fid.vnode, dentry->d_name.name, + datasync); + + ret = filemap_write_and_wait_range(inode->i_mapping, start, end); + if (ret) + return ret; + mutex_lock(&inode->i_mutex); + + /* use a writeback record as a marker in the queue - when this reaches + * the front of the queue, all the outstanding writes are either + * completed or rejected */ + wb = kzalloc(sizeof(*wb), GFP_KERNEL); + if (!wb) { + ret = -ENOMEM; + goto out; + } + wb->vnode = vnode; + wb->first = 0; + wb->last = -1; + wb->offset_first = 0; + wb->to_last = PAGE_SIZE; + wb->usage = 1; + wb->state = AFS_WBACK_SYNCING; + init_waitqueue_head(&wb->waitq); + + spin_lock(&vnode->writeback_lock); + list_for_each_entry(xwb, &vnode->writebacks, link) { + if (xwb->state == AFS_WBACK_PENDING) + xwb->state = AFS_WBACK_CONFLICTING; + } + list_add_tail(&wb->link, &vnode->writebacks); + spin_unlock(&vnode->writeback_lock); + + /* push all the outstanding writebacks to the server */ + ret = afs_writeback_all(vnode); + if (ret < 0) { + afs_put_writeback(wb); + _leave(" = %d [wb]", ret); + goto out; + } + + /* wait for the preceding writes to actually complete */ + ret = wait_event_interruptible(wb->waitq, + wb->state == AFS_WBACK_COMPLETE || + vnode->writebacks.next == &wb->link); + afs_put_writeback(wb); + _leave(" = %d", ret); +out: + mutex_unlock(&inode->i_mutex); + return ret; +} + +/* + * notification that a previously read-only page is about to become writable + * - if it returns an error, the caller will deliver a bus error signal + */ +int afs_page_mkwrite(struct vm_area_struct *vma, struct page *page) +{ + struct afs_vnode *vnode = AFS_FS_I(vma->vm_file->f_mapping->host); + + _enter("{{%x:%u}},{%lx}", + vnode->fid.vid, vnode->fid.vnode, page->index); + + /* wait for the page to be written to the cache before we allow it to + * be modified */ +#ifdef CONFIG_AFS_FSCACHE + fscache_wait_on_page_write(vnode->cache, page); +#endif + + _leave(" = 0"); + return 0; +} |
