diff options
Diffstat (limited to 'drivers/staging/lustre')
149 files changed, 1346 insertions, 1603 deletions
diff --git a/drivers/staging/lustre/TODO b/drivers/staging/lustre/TODO index 22742d6d62a8..0a2b6cb3775e 100644 --- a/drivers/staging/lustre/TODO +++ b/drivers/staging/lustre/TODO @@ -9,5 +9,6 @@ * Other minor misc cleanups... Please send any patches to Greg Kroah-Hartman <greg@kroah.com>, Andreas Dilger -<andreas.dilger@intel.com> and Peng Tao <tao.peng@emc.com>. CCing -hpdd-discuss <hpdd-discuss@lists.01.org> would be great too. +<andreas.dilger@intel.com>, Oleg Drokin <oleg.drokin@intel.com> and +Peng Tao <tao.peng@emc.com>. CCing hpdd-discuss <hpdd-discuss@lists.01.org> +would be great too. diff --git a/drivers/staging/lustre/include/linux/libcfs/curproc.h b/drivers/staging/lustre/include/linux/libcfs/curproc.h index 507d16b9213c..8fd47c98fd33 100644 --- a/drivers/staging/lustre/include/linux/libcfs/curproc.h +++ b/drivers/staging/lustre/include/linux/libcfs/curproc.h @@ -44,13 +44,6 @@ #define __LIBCFS_CURPROC_H__ /* - * Portable API to access common characteristics of "current" UNIX process. - * - * Implemented in portals/include/libcfs/<os>/ - */ -int cfs_curproc_groups_nr(void); - -/* * Plus, platform-specific constant * * CFS_CURPROC_COMM_MAX, @@ -91,8 +84,6 @@ void cfs_cap_raise(cfs_cap_t cap); void cfs_cap_lower(cfs_cap_t cap); int cfs_cap_raised(cfs_cap_t cap); cfs_cap_t cfs_curproc_cap_pack(void); -void cfs_curproc_cap_unpack(cfs_cap_t cap); -int cfs_capable(cfs_cap_t cap); /* __LIBCFS_CURPROC_H__ */ #endif diff --git a/drivers/staging/lustre/include/linux/libcfs/libcfs_hash.h b/drivers/staging/lustre/include/linux/libcfs/libcfs_hash.h index 9d5ee1a69c0c..e5d5db255622 100644 --- a/drivers/staging/lustre/include/linux/libcfs/libcfs_hash.h +++ b/drivers/staging/lustre/include/linux/libcfs/libcfs_hash.h @@ -65,8 +65,6 @@ #include <linux/hash.h> -#define cfs_hash_long(val, bits) hash_long(val, bits) - /** disable debug */ #define CFS_HASH_DEBUG_NONE 0 /** record hash depth and output to console when it's too deep, diff --git a/drivers/staging/lustre/include/linux/libcfs/libcfs_ioctl.h b/drivers/staging/lustre/include/linux/libcfs/libcfs_ioctl.h index 74dda57b98a8..49ba62a4daa8 100644 --- a/drivers/staging/lustre/include/linux/libcfs/libcfs_ioctl.h +++ b/drivers/staging/lustre/include/linux/libcfs/libcfs_ioctl.h @@ -211,10 +211,10 @@ static inline int libcfs_ioctl_is_invalid(struct libcfs_ioctl_data *data) } -extern int libcfs_register_ioctl(struct libcfs_ioctl_handler *hand); -extern int libcfs_deregister_ioctl(struct libcfs_ioctl_handler *hand); -extern int libcfs_ioctl_getdata(char *buf, char *end, void *arg); -extern int libcfs_ioctl_popdata(void *arg, void *buf, int size); +int libcfs_register_ioctl(struct libcfs_ioctl_handler *hand); +int libcfs_deregister_ioctl(struct libcfs_ioctl_handler *hand); +int libcfs_ioctl_getdata(char *buf, char *end, void *arg); +int libcfs_ioctl_popdata(void *arg, void *buf, int size); #endif /* __LIBCFS_IOCTL_H__ */ diff --git a/drivers/staging/lustre/include/linux/libcfs/libcfs_kernelcomm.h b/drivers/staging/lustre/include/linux/libcfs/libcfs_kernelcomm.h index 596a15fc8996..037ae8a6d531 100644 --- a/drivers/staging/lustre/include/linux/libcfs/libcfs_kernelcomm.h +++ b/drivers/staging/lustre/include/linux/libcfs/libcfs_kernelcomm.h @@ -61,6 +61,8 @@ struct kuc_hdr { __u16 kuc_msglen; /* Including header */ } __attribute__((aligned(sizeof(__u64)))); +#define KUC_CHANGELOG_MSG_MAXSIZE (sizeof(struct kuc_hdr)+CR_MAXSIZE) + #define KUC_MAGIC 0x191C /*Lustre9etLinC */ #define KUC_FL_BLOCK 0x01 /* Wait for send */ diff --git a/drivers/staging/lustre/include/linux/libcfs/libcfs_prim.h b/drivers/staging/lustre/include/linux/libcfs/libcfs_prim.h index e6e417aeefd5..7cf34aa78f79 100644 --- a/drivers/staging/lustre/include/linux/libcfs/libcfs_prim.h +++ b/drivers/staging/lustre/include/linux/libcfs/libcfs_prim.h @@ -41,19 +41,10 @@ #define __LIBCFS_PRIM_H__ /* - * Schedule - */ -void cfs_pause(cfs_duration_t ticks); - -/* * Timer */ typedef void (cfs_timer_func_t)(ulong_ptr_t); -void schedule_timeout_and_set_state(long, int64_t); -void init_waitqueue_entry_current(wait_queue_t *link); -int64_t waitq_timedwait(wait_queue_t *, long, int64_t); -void waitq_wait(wait_queue_t *, long); void add_wait_queue_exclusive_head(wait_queue_head_t *, wait_queue_t *); void cfs_init_timer(struct timer_list *t); diff --git a/drivers/staging/lustre/include/linux/libcfs/libcfs_private.h b/drivers/staging/lustre/include/linux/libcfs/libcfs_private.h index d0d942ced01a..dddccca120c9 100644 --- a/drivers/staging/lustre/include/linux/libcfs/libcfs_private.h +++ b/drivers/staging/lustre/include/linux/libcfs/libcfs_private.h @@ -120,7 +120,7 @@ do { \ do { \ LASSERT(!in_interrupt() || \ ((size) <= LIBCFS_VMALLOC_SIZE && \ - ((mask) & GFP_ATOMIC)) != 0); \ + ((mask) & __GFP_WAIT) == 0)); \ } while (0) #define LIBCFS_ALLOC_POST(ptr, size) \ diff --git a/drivers/staging/lustre/include/linux/libcfs/libcfs_string.h b/drivers/staging/lustre/include/linux/libcfs/libcfs_string.h index a6bac9c36339..509dc1e5c3b1 100644 --- a/drivers/staging/lustre/include/linux/libcfs/libcfs_string.h +++ b/drivers/staging/lustre/include/linux/libcfs/libcfs_string.h @@ -49,18 +49,6 @@ int cfs_strncasecmp(const char *s1, const char *s2, size_t n); /* Convert a text string to a bitmask */ int cfs_str2mask(const char *str, const char *(*bit2str)(int bit), int *oldmask, int minmask, int allmask); - -/* Allocate space for and copy an existing string. - * Must free with kfree(). - */ -char *cfs_strdup(const char *str, u_int32_t flags); - -/* safe vsnprintf */ -int cfs_vsnprintf(char *buf, size_t size, const char *fmt, va_list args); - -/* safe snprintf */ -int cfs_snprintf(char *buf, size_t size, const char *fmt, ...); - /* trim leading and trailing space characters */ char *cfs_firststr(char *str, size_t size); @@ -90,27 +78,10 @@ struct cfs_expr_list { struct list_head el_exprs; }; -static inline int -cfs_iswhite(char c) -{ - switch (c) { - case ' ': - case '\t': - case '\n': - case '\r': - return 1; - default: - break; - } - return 0; -} - char *cfs_trimwhite(char *str); int cfs_gettok(struct cfs_lstr *next, char delim, struct cfs_lstr *res); int cfs_str2num_check(char *str, int nob, unsigned *num, unsigned min, unsigned max); -int cfs_range_expr_parse(struct cfs_lstr *src, unsigned min, unsigned max, - int single_tok, struct cfs_range_expr **expr); int cfs_expr_list_match(__u32 value, struct cfs_expr_list *expr_list); int cfs_expr_list_values(struct cfs_expr_list *expr_list, int max, __u32 **values); @@ -124,7 +95,6 @@ cfs_expr_list_values_free(__u32 *values, int num) } void cfs_expr_list_free(struct cfs_expr_list *expr_list); -void cfs_expr_list_print(struct cfs_expr_list *expr_list); int cfs_expr_list_parse(char *str, int len, unsigned min, unsigned max, struct cfs_expr_list **elpp); void cfs_expr_list_free_list(struct list_head *list); diff --git a/drivers/staging/lustre/include/linux/lnet/lib-lnet.h b/drivers/staging/lustre/include/linux/lnet/lib-lnet.h index 3ac2bb5fd2db..856fcfaafafa 100644 --- a/drivers/staging/lustre/include/linux/lnet/lib-lnet.h +++ b/drivers/staging/lustre/include/linux/lnet/lib-lnet.h @@ -628,7 +628,7 @@ void lnet_ni_free(lnet_ni_t *ni); static inline int lnet_nid2peerhash(lnet_nid_t nid) { - return cfs_hash_long(nid, LNET_PEER_HASH_BITS); + return hash_long(nid, LNET_PEER_HASH_BITS); } static inline struct list_head * diff --git a/drivers/staging/lustre/include/linux/lnet/lib-types.h b/drivers/staging/lustre/include/linux/lnet/lib-types.h index dd8edcf1b5c0..1c13ef7df80e 100644 --- a/drivers/staging/lustre/include/linux/lnet/lib-types.h +++ b/drivers/staging/lustre/include/linux/lnet/lib-types.h @@ -282,16 +282,14 @@ typedef struct lnet_libmd { #define LNET_MD_FLAG_AUTO_UNLINK (1 << 1) #ifdef LNET_USE_LIB_FREELIST -typedef struct -{ +typedef struct { void *fl_objs; /* single contiguous array of objects */ int fl_nobjs; /* the number of them */ int fl_objsize; /* the size (including overhead) of each of them */ struct list_head fl_list; /* where they are enqueued */ } lnet_freelist_t; -typedef struct -{ +typedef struct { struct list_head fo_list; /* enqueue on fl_list */ void *fo_contents; /* aligned contents */ } lnet_freeobj_t; @@ -312,8 +310,7 @@ typedef struct { struct lnet_ni; /* forward ref */ -typedef struct lnet_lnd -{ +typedef struct lnet_lnd { /* fields managed by portals */ struct list_head lnd_list; /* stash in the LND table */ int lnd_refcount; /* # active instances */ @@ -321,8 +318,8 @@ typedef struct lnet_lnd /* fields initialised by the LND */ unsigned int lnd_type; - int (*lnd_startup) (struct lnet_ni *ni); - void (*lnd_shutdown) (struct lnet_ni *ni); + int (*lnd_startup)(struct lnet_ni *ni); + void (*lnd_shutdown)(struct lnet_ni *ni); int (*lnd_ctl)(struct lnet_ni *ni, unsigned int cmd, void *arg); /* In data movement APIs below, payload buffers are described as a set @@ -668,8 +665,7 @@ struct lnet_msg_container { #define LNET_RC_STATE_RUNNING 1 /* started up OK */ #define LNET_RC_STATE_STOPPING 2 /* telling thread to stop */ -typedef struct -{ +typedef struct { /* CPU partition table of LNet */ struct cfs_cpt_table *ln_cpt_table; /* number of CPTs in ln_cpt_table */ diff --git a/drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd.c b/drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd.c index 644a0000130a..0061c8afa206 100644 --- a/drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd.c +++ b/drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd.c @@ -41,7 +41,7 @@ #include "o2iblnd.h" #include <asm/div64.h> -lnd_t the_o2iblnd = { +static lnd_t the_o2iblnd = { .lnd_type = O2IBLND, .lnd_startup = kiblnd_startup, .lnd_shutdown = kiblnd_shutdown, @@ -53,8 +53,8 @@ lnd_t the_o2iblnd = { kib_data_t kiblnd_data; -__u32 -kiblnd_cksum (void *ptr, int nob) +static __u32 +kiblnd_cksum(void *ptr, int nob) { char *c = ptr; __u32 sum = 0; @@ -429,8 +429,8 @@ kiblnd_unlink_peer_locked (kib_peer_t *peer) kiblnd_peer_decref(peer); } -int -kiblnd_get_peer_info (lnet_ni_t *ni, int index, +static int +kiblnd_get_peer_info(lnet_ni_t *ni, int index, lnet_nid_t *nidp, int *count) { kib_peer_t *peer; @@ -468,8 +468,8 @@ kiblnd_get_peer_info (lnet_ni_t *ni, int index, return -ENOENT; } -void -kiblnd_del_peer_locked (kib_peer_t *peer) +static void +kiblnd_del_peer_locked(kib_peer_t *peer) { struct list_head *ctmp; struct list_head *cnxt; @@ -489,8 +489,8 @@ kiblnd_del_peer_locked (kib_peer_t *peer) * last ref on it. */ } -int -kiblnd_del_peer (lnet_ni_t *ni, lnet_nid_t nid) +static int +kiblnd_del_peer(lnet_ni_t *ni, lnet_nid_t nid) { LIST_HEAD (zombies); struct list_head *ptmp; @@ -543,8 +543,8 @@ kiblnd_del_peer (lnet_ni_t *ni, lnet_nid_t nid) return rc; } -kib_conn_t * -kiblnd_get_conn_by_idx (lnet_ni_t *ni, int index) +static kib_conn_t * +kiblnd_get_conn_by_idx(lnet_ni_t *ni, int index) { kib_peer_t *peer; struct list_head *ptmp; @@ -584,74 +584,6 @@ kiblnd_get_conn_by_idx (lnet_ni_t *ni, int index) return NULL; } -void -kiblnd_debug_rx (kib_rx_t *rx) -{ - CDEBUG(D_CONSOLE, " %p status %d msg_type %x cred %d\n", - rx, rx->rx_status, rx->rx_msg->ibm_type, - rx->rx_msg->ibm_credits); -} - -void -kiblnd_debug_tx (kib_tx_t *tx) -{ - CDEBUG(D_CONSOLE, " %p snd %d q %d w %d rc %d dl %lx " - "cookie "LPX64" msg %s%s type %x cred %d\n", - tx, tx->tx_sending, tx->tx_queued, tx->tx_waiting, - tx->tx_status, tx->tx_deadline, tx->tx_cookie, - tx->tx_lntmsg[0] == NULL ? "-" : "!", - tx->tx_lntmsg[1] == NULL ? "-" : "!", - tx->tx_msg->ibm_type, tx->tx_msg->ibm_credits); -} - -void -kiblnd_debug_conn (kib_conn_t *conn) -{ - struct list_head *tmp; - int i; - - spin_lock(&conn->ibc_lock); - - CDEBUG(D_CONSOLE, "conn[%d] %p [version %x] -> %s: \n", - atomic_read(&conn->ibc_refcount), conn, - conn->ibc_version, libcfs_nid2str(conn->ibc_peer->ibp_nid)); - CDEBUG(D_CONSOLE, " state %d nposted %d/%d cred %d o_cred %d r_cred %d\n", - conn->ibc_state, conn->ibc_noops_posted, - conn->ibc_nsends_posted, conn->ibc_credits, - conn->ibc_outstanding_credits, conn->ibc_reserved_credits); - CDEBUG(D_CONSOLE, " comms_err %d\n", conn->ibc_comms_error); - - CDEBUG(D_CONSOLE, " early_rxs:\n"); - list_for_each(tmp, &conn->ibc_early_rxs) - kiblnd_debug_rx(list_entry(tmp, kib_rx_t, rx_list)); - - CDEBUG(D_CONSOLE, " tx_noops:\n"); - list_for_each(tmp, &conn->ibc_tx_noops) - kiblnd_debug_tx(list_entry(tmp, kib_tx_t, tx_list)); - - CDEBUG(D_CONSOLE, " tx_queue_nocred:\n"); - list_for_each(tmp, &conn->ibc_tx_queue_nocred) - kiblnd_debug_tx(list_entry(tmp, kib_tx_t, tx_list)); - - CDEBUG(D_CONSOLE, " tx_queue_rsrvd:\n"); - list_for_each(tmp, &conn->ibc_tx_queue_rsrvd) - kiblnd_debug_tx(list_entry(tmp, kib_tx_t, tx_list)); - - CDEBUG(D_CONSOLE, " tx_queue:\n"); - list_for_each(tmp, &conn->ibc_tx_queue) - kiblnd_debug_tx(list_entry(tmp, kib_tx_t, tx_list)); - - CDEBUG(D_CONSOLE, " active_txs:\n"); - list_for_each(tmp, &conn->ibc_active_txs) - kiblnd_debug_tx(list_entry(tmp, kib_tx_t, tx_list)); - - CDEBUG(D_CONSOLE, " rxs:\n"); - for (i = 0; i < IBLND_RX_MSGS(conn->ibc_version); i++) - kiblnd_debug_rx(&conn->ibc_rxs[i]); - - spin_unlock(&conn->ibc_lock); -} - int kiblnd_translate_mtu(int value) { @@ -1039,8 +971,8 @@ kiblnd_close_stale_conns_locked (kib_peer_t *peer, return count; } -int -kiblnd_close_matching_conns (lnet_ni_t *ni, lnet_nid_t nid) +static int +kiblnd_close_matching_conns(lnet_ni_t *ni, lnet_nid_t nid) { kib_peer_t *peer; struct list_head *ptmp; @@ -1440,7 +1372,7 @@ kiblnd_find_rd_dma_mr(kib_hca_dev_t *hdev, kib_rdma_desc_t *rd) return mr; } -void +static void kiblnd_destroy_fmr_pool(kib_fmr_pool_t *pool) { LASSERT (pool->fpo_map_count == 0); @@ -1454,7 +1386,7 @@ kiblnd_destroy_fmr_pool(kib_fmr_pool_t *pool) LIBCFS_FREE(pool, sizeof(kib_fmr_pool_t)); } -void +static void kiblnd_destroy_fmr_pool_list(struct list_head *head) { kib_fmr_pool_t *pool; @@ -1480,7 +1412,7 @@ static int kiblnd_fmr_flush_trigger(int ncpts) return max(IBLND_FMR_POOL_FLUSH, size); } -int +static int kiblnd_create_fmr_pool(kib_fmr_poolset_t *fps, kib_fmr_pool_t **pp_fpo) { /* FMR pool for RDMA */ @@ -1719,7 +1651,7 @@ kiblnd_init_pool(kib_poolset_t *ps, kib_pool_t *pool, int size) pool->po_size = size; } -void +static void kiblnd_destroy_pool_list(struct list_head *head) { kib_pool_t *pool; @@ -2192,7 +2124,7 @@ kiblnd_tx_init(kib_pool_t *pool, struct list_head *node) tx->tx_cookie = tps->tps_next_tx_cookie ++; } -void +static void kiblnd_net_fini_pools(kib_net_t *net) { int i; @@ -2234,7 +2166,7 @@ kiblnd_net_fini_pools(kib_net_t *net) } } -int +static int kiblnd_net_init_pools(kib_net_t *net, __u32 *cpts, int ncpts) { unsigned long flags; @@ -2408,7 +2340,7 @@ kiblnd_hdev_get_attr(kib_hca_dev_t *hdev) return -EINVAL; } -void +static void kiblnd_hdev_cleanup_mrs(kib_hca_dev_t *hdev) { int i; @@ -2442,7 +2374,7 @@ kiblnd_hdev_destroy(kib_hca_dev_t *hdev) LIBCFS_FREE(hdev, sizeof(*hdev)); } -int +static int kiblnd_hdev_setup_mrs(kib_hca_dev_t *hdev) { struct ib_mr *mr; @@ -2746,7 +2678,7 @@ kiblnd_destroy_dev (kib_dev_t *dev) LIBCFS_FREE(dev, sizeof(*dev)); } -kib_dev_t * +static kib_dev_t * kiblnd_create_dev(char *ifname) { struct net_device *netdev; @@ -2800,7 +2732,7 @@ kiblnd_create_dev(char *ifname) return dev; } -void +static void kiblnd_base_shutdown(void) { struct kib_sched_info *sched; @@ -2842,7 +2774,8 @@ kiblnd_base_shutdown(void) CDEBUG(((i & (-i)) == i) ? D_WARNING : D_NET, /* power of 2? */ "Waiting for %d threads to terminate\n", atomic_read(&kiblnd_data.kib_nthreads)); - cfs_pause(cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); } /* fall through */ @@ -2903,7 +2836,8 @@ kiblnd_shutdown (lnet_ni_t *ni) "%s: waiting for %d peers to disconnect\n", libcfs_nid2str(ni->ni_nid), atomic_read(&net->ibn_npeers)); - cfs_pause(cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); } kiblnd_net_fini_pools(net); @@ -2940,7 +2874,7 @@ out: return; } -int +static int kiblnd_base_startup(void) { struct kib_sched_info *sched; @@ -3030,7 +2964,7 @@ kiblnd_base_startup(void) return -ENETDOWN; } -int +static int kiblnd_start_schedulers(struct kib_sched_info *sched) { int rc = 0; @@ -3071,7 +3005,7 @@ kiblnd_start_schedulers(struct kib_sched_info *sched) return rc; } -int +static int kiblnd_dev_start_threads(kib_dev_t *dev, int newdev, __u32 *cpts, int ncpts) { int cpt; @@ -3097,7 +3031,7 @@ kiblnd_dev_start_threads(kib_dev_t *dev, int newdev, __u32 *cpts, int ncpts) return 0; } -kib_dev_t * +static kib_dev_t * kiblnd_dev_search(char *ifname) { kib_dev_t *alias = NULL; @@ -3226,13 +3160,13 @@ failed: return -ENETDOWN; } -void __exit +static void __exit kiblnd_module_fini (void) { lnet_unregister_lnd(&the_o2iblnd); } -int __init +static int __init kiblnd_module_init (void) { int rc; diff --git a/drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c b/drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c index 93648632ba26..6173e74d7492 100644 --- a/drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c +++ b/drivers/staging/lustre/lnet/klnds/o2iblnd/o2iblnd_cb.c @@ -529,7 +529,7 @@ kiblnd_kvaddr_to_page (unsigned long vaddr) { struct page *page; - if (is_vmalloc_addr(vaddr)) { + if (is_vmalloc_addr((void *)vaddr)) { page = vmalloc_to_page ((void *)vaddr); LASSERT (page != NULL); return page; @@ -3127,7 +3127,7 @@ kiblnd_connd (void *arg) cfs_block_allsigs (); - init_waitqueue_entry_current (&wait); + init_waitqueue_entry(&wait, current); kiblnd_data.kib_connd = current; spin_lock_irqsave(&kiblnd_data.kib_connd_lock, flags); @@ -3208,7 +3208,7 @@ kiblnd_connd (void *arg) add_wait_queue(&kiblnd_data.kib_connd_waitq, &wait); spin_unlock_irqrestore(&kiblnd_data.kib_connd_lock, flags); - waitq_timedwait(&wait, TASK_INTERRUPTIBLE, timeout); + schedule_timeout(timeout); set_current_state(TASK_RUNNING); remove_wait_queue(&kiblnd_data.kib_connd_waitq, &wait); @@ -3324,7 +3324,7 @@ kiblnd_scheduler(void *arg) cfs_block_allsigs(); - init_waitqueue_entry_current(&wait); + init_waitqueue_entry(&wait, current); sched = kiblnd_data.kib_scheds[KIB_THREAD_CPT(id)]; @@ -3423,7 +3423,7 @@ kiblnd_scheduler(void *arg) add_wait_queue_exclusive(&sched->ibs_waitq, &wait); spin_unlock_irqrestore(&sched->ibs_lock, flags); - waitq_wait(&wait, TASK_INTERRUPTIBLE); + schedule(); busy_loops = 0; remove_wait_queue(&sched->ibs_waitq, &wait); @@ -3450,7 +3450,7 @@ kiblnd_failover_thread(void *arg) cfs_block_allsigs (); - init_waitqueue_entry_current(&wait); + init_waitqueue_entry(&wait, current); write_lock_irqsave(glock, flags); while (!kiblnd_data.kib_shutdown) { diff --git a/drivers/staging/lustre/lnet/klnds/socklnd/socklnd.c b/drivers/staging/lustre/lnet/klnds/socklnd/socklnd.c index 8f74d0be32f1..21d36ee5378a 100644 --- a/drivers/staging/lustre/lnet/klnds/socklnd/socklnd.c +++ b/drivers/staging/lustre/lnet/klnds/socklnd/socklnd.c @@ -2336,7 +2336,8 @@ ksocknal_base_shutdown(void) "waiting for %d threads to terminate\n", ksocknal_data.ksnd_nthreads); read_unlock(&ksocknal_data.ksnd_global_lock); - cfs_pause(cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); read_lock(&ksocknal_data.ksnd_global_lock); } read_unlock(&ksocknal_data.ksnd_global_lock); @@ -2584,7 +2585,8 @@ ksocknal_shutdown (lnet_ni_t *ni) CDEBUG(((i & (-i)) == i) ? D_WARNING : D_NET, /* power of 2? */ "waiting for %d peers to disconnect\n", net->ksnn_npeers); - cfs_pause(cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); ksocknal_debug_peerhash(ni); diff --git a/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_cb.c b/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_cb.c index 68a4f52ec998..bdf95ea8a54d 100644 --- a/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_cb.c +++ b/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_cb.c @@ -189,7 +189,8 @@ ksocknal_transmit (ksock_conn_t *conn, ksock_tx_t *tx) int bufnob; if (ksocknal_data.ksnd_stall_tx != 0) { - cfs_pause(cfs_time_seconds(ksocknal_data.ksnd_stall_tx)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(ksocknal_data.ksnd_stall_tx)); } LASSERT (tx->tx_resid != 0); @@ -345,7 +346,8 @@ ksocknal_receive (ksock_conn_t *conn) int rc; if (ksocknal_data.ksnd_stall_rx != 0) { - cfs_pause(cfs_time_seconds (ksocknal_data.ksnd_stall_rx)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(ksocknal_data.ksnd_stall_rx)); } rc = ksocknal_connsock_addref(conn); @@ -924,7 +926,7 @@ ksocknal_launch_packet (lnet_ni_t *ni, ksock_tx_t *tx, lnet_process_id_t id) int ksocknal_send(lnet_ni_t *ni, void *private, lnet_msg_t *lntmsg) { - int mpflag = 0; + int mpflag = 1; int type = lntmsg->msg_type; lnet_process_id_t target = lntmsg->msg_target; unsigned int payload_niov = lntmsg->msg_niov; @@ -993,8 +995,9 @@ ksocknal_send(lnet_ni_t *ni, void *private, lnet_msg_t *lntmsg) /* The first fragment will be set later in pro_pack */ rc = ksocknal_launch_packet(ni, tx, target); - if (lntmsg->msg_vmflush) + if (!mpflag) cfs_memory_pressure_restore(mpflag); + if (rc == 0) return (0); @@ -2139,7 +2142,7 @@ ksocknal_connd (void *arg) cfs_block_allsigs (); - init_waitqueue_entry_current (&wait); + init_waitqueue_entry(&wait, current); spin_lock_bh(connd_lock); @@ -2228,7 +2231,7 @@ ksocknal_connd (void *arg) spin_unlock_bh(connd_lock); nloops = 0; - waitq_timedwait(&wait, TASK_INTERRUPTIBLE, timeout); + schedule_timeout(timeout); set_current_state(TASK_RUNNING); remove_wait_queue(&ksocknal_data.ksnd_connd_waitq, &wait); @@ -2531,7 +2534,7 @@ ksocknal_reaper (void *arg) cfs_block_allsigs (); INIT_LIST_HEAD(&enomem_conns); - init_waitqueue_entry_current (&wait); + init_waitqueue_entry(&wait, current); spin_lock_bh(&ksocknal_data.ksnd_reaper_lock); @@ -2638,8 +2641,7 @@ ksocknal_reaper (void *arg) if (!ksocknal_data.ksnd_shuttingdown && list_empty (&ksocknal_data.ksnd_deathrow_conns) && list_empty (&ksocknal_data.ksnd_zombie_conns)) - waitq_timedwait (&wait, TASK_INTERRUPTIBLE, - timeout); + schedule_timeout(timeout); set_current_state (TASK_RUNNING); remove_wait_queue (&ksocknal_data.ksnd_reaper_waitq, &wait); diff --git a/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_lib-linux.c b/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_lib-linux.c index 80141aa32c21..a54b506ba7ca 100644 --- a/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_lib-linux.c +++ b/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_lib-linux.c @@ -373,7 +373,8 @@ ksocknal_lib_recv_kiov (ksock_conn_t *conn) /* NB we can't trust socket ops to either consume our iovs * or leave them alone. */ - if ((addr = ksocknal_lib_kiov_vmap(kiov, niov, scratchiov, pages)) != NULL) { + addr = ksocknal_lib_kiov_vmap(kiov, niov, scratchiov, pages); + if (addr != NULL) { nob = scratchiov[0].iov_len; msg.msg_iovlen = 1; diff --git a/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_proto.c b/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_proto.c index 71205e2015ce..2d91571cbab2 100644 --- a/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_proto.c +++ b/drivers/staging/lustre/lnet/klnds/socklnd/socklnd_proto.c @@ -165,7 +165,8 @@ ksocknal_queue_tx_zcack_v3(ksock_conn_t *conn, LASSERT (tx_ack == NULL || tx_ack->tx_msg.ksm_type == KSOCK_MSG_NOOP); - if ((tx = conn->ksnc_tx_carrier) == NULL) { + tx = conn->ksnc_tx_carrier; + if (tx == NULL) { if (tx_ack != NULL) { list_add_tail(&tx_ack->tx_list, &conn->ksnc_tx_queue); @@ -392,7 +393,8 @@ ksocknal_handle_zcreq(ksock_conn_t *c, __u64 cookie, int remote) if (tx == NULL) return -ENOMEM; - if ((rc = ksocknal_launch_packet(peer->ksnp_ni, tx, peer->ksnp_id)) == 0) + rc = ksocknal_launch_packet(peer->ksnp_ni, tx, peer->ksnp_id); + if (rc == 0) return 0; ksocknal_free_tx(tx); diff --git a/drivers/staging/lustre/lnet/lnet/acceptor.c b/drivers/staging/lustre/lnet/lnet/acceptor.c index cb2ecd717714..09ea6cb1492c 100644 --- a/drivers/staging/lustre/lnet/lnet/acceptor.c +++ b/drivers/staging/lustre/lnet/lnet/acceptor.c @@ -371,7 +371,8 @@ lnet_acceptor(void *arg) if (rc != 0) { if (rc != -EAGAIN) { CWARN("Accept error %d: pausing...\n", rc); - cfs_pause(cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); } continue; } diff --git a/drivers/staging/lustre/lnet/lnet/api-ni.c b/drivers/staging/lustre/lnet/lnet/api-ni.c index c562ff3e9283..45c23194081b 100644 --- a/drivers/staging/lustre/lnet/lnet/api-ni.c +++ b/drivers/staging/lustre/lnet/lnet/api-ni.c @@ -770,7 +770,7 @@ lnet_nid_cpt_hash(lnet_nid_t nid, unsigned int number) if (number == 1) return 0; - val = cfs_hash_long(key, LNET_CPT_BITS); + val = hash_long(key, LNET_CPT_BITS); /* NB: LNET_CP_NUMBER doesn't have to be PO2 */ if (val < number) return val; @@ -994,7 +994,8 @@ lnet_shutdown_lndnis (void) "Waiting for zombie LNI %s\n", libcfs_nid2str(ni->ni_nid)); } - cfs_pause(cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); lnet_net_lock(LNET_LOCK_EX); continue; } diff --git a/drivers/staging/lustre/lnet/lnet/config.c b/drivers/staging/lustre/lnet/lnet/config.c index 6a07b0a65d12..d97464e95ddb 100644 --- a/drivers/staging/lustre/lnet/lnet/config.c +++ b/drivers/staging/lustre/lnet/lnet/config.c @@ -443,7 +443,7 @@ lnet_str2tbs_sep(struct list_head *tbs, char *str) /* Split 'str' into separate commands */ for (;;) { /* skip leading whitespace */ - while (cfs_iswhite(*str)) + while (isspace(*str)) str++; /* scan for separator or comment */ @@ -460,7 +460,7 @@ lnet_str2tbs_sep(struct list_head *tbs, char *str) } for (i = 0; i < nob; i++) - if (cfs_iswhite(str[i])) + if (isspace(str[i])) ltb->ltb_text[i] = ' '; else ltb->ltb_text[i] = str[i]; @@ -667,7 +667,7 @@ lnet_parse_route(char *str, int *im_a_router) sep = str; for (;;) { /* scan for token start */ - while (cfs_iswhite(*sep)) + while (isspace(*sep)) sep++; if (*sep == 0) { if (ntokens < (got_hops ? 3 : 2)) @@ -679,7 +679,7 @@ lnet_parse_route(char *str, int *im_a_router) token = sep++; /* scan for token end */ - while (*sep != 0 && !cfs_iswhite(*sep)) + while (*sep != 0 && !isspace(*sep)) sep++; if (*sep != 0) *sep++ = 0; @@ -858,7 +858,7 @@ lnet_match_network_tokens(char *net_entry, __u32 *ipaddrs, int nip) sep = tokens; for (;;) { /* scan for token start */ - while (cfs_iswhite(*sep)) + while (isspace(*sep)) sep++; if (*sep == 0) break; @@ -866,7 +866,7 @@ lnet_match_network_tokens(char *net_entry, __u32 *ipaddrs, int nip) token = sep++; /* scan for token end */ - while (*sep != 0 && !cfs_iswhite(*sep)) + while (*sep != 0 && !isspace(*sep)) sep++; if (*sep != 0) *sep++ = 0; diff --git a/drivers/staging/lustre/lnet/lnet/lib-eq.c b/drivers/staging/lustre/lnet/lnet/lib-eq.c index 4ce68d3b0931..7ce07f61b2e1 100644 --- a/drivers/staging/lustre/lnet/lnet/lib-eq.c +++ b/drivers/staging/lustre/lnet/lnet/lib-eq.c @@ -334,21 +334,20 @@ lnet_eq_wait_locked(int *timeout_ms) if (tms == 0) return -1; /* don't want to wait and no new event */ - init_waitqueue_entry_current(&wl); + init_waitqueue_entry(&wl, current); set_current_state(TASK_INTERRUPTIBLE); add_wait_queue(&the_lnet.ln_eq_waitq, &wl); lnet_eq_wait_unlock(); if (tms < 0) { - waitq_wait(&wl, TASK_INTERRUPTIBLE); + schedule(); } else { struct timeval tv; now = cfs_time_current(); - waitq_timedwait(&wl, TASK_INTERRUPTIBLE, - cfs_time_seconds(tms) / 1000); + schedule_timeout(cfs_time_seconds(tms) / 1000); cfs_duration_usec(cfs_time_sub(cfs_time_current(), now), &tv); tms -= (int)(tv.tv_sec * 1000 + tv.tv_usec / 1000); if (tms < 0) /* no more wait but may have new event */ diff --git a/drivers/staging/lustre/lnet/lnet/lib-ptl.c b/drivers/staging/lustre/lnet/lnet/lib-ptl.c index 6fffd5e96f9c..920df69960a5 100644 --- a/drivers/staging/lustre/lnet/lnet/lib-ptl.c +++ b/drivers/staging/lustre/lnet/lnet/lib-ptl.c @@ -366,7 +366,7 @@ lnet_mt_match_head(struct lnet_match_table *mtable, unsigned long hash = mbits + id.nid + id.pid; LASSERT(lnet_ptl_is_unique(ptl)); - hash = cfs_hash_long(hash, LNET_MT_HASH_BITS); + hash = hash_long(hash, LNET_MT_HASH_BITS); return &mtable->mt_mhash[hash]; } } diff --git a/drivers/staging/lustre/lnet/lnet/peer.c b/drivers/staging/lustre/lnet/lnet/peer.c index 286977691393..72802b0404a4 100644 --- a/drivers/staging/lustre/lnet/lnet/peer.c +++ b/drivers/staging/lustre/lnet/lnet/peer.c @@ -145,7 +145,8 @@ lnet_peer_tables_cleanup(void) "Waiting for %d peers on peer table\n", ptable->pt_number); } - cfs_pause(cfs_time_seconds(1) / 2); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1) / 2); lnet_net_lock(i); } list_splice_init(&ptable->pt_deathrow, &deathrow); diff --git a/drivers/staging/lustre/lnet/lnet/router.c b/drivers/staging/lustre/lnet/lnet/router.c index d1ee44232eef..995f50976c42 100644 --- a/drivers/staging/lustre/lnet/lnet/router.c +++ b/drivers/staging/lustre/lnet/lnet/router.c @@ -779,7 +779,8 @@ lnet_wait_known_routerstate(void) if (all_known) return; - cfs_pause(cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); } } @@ -1147,7 +1148,8 @@ lnet_prune_rc_data(int wait_unlink) i++; CDEBUG(((i & (-i)) == i) ? D_WARNING : D_NET, "Waiting for rc buffers to unlink\n"); - cfs_pause(cfs_time_seconds(1) / 4); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1) / 4); lnet_net_lock(LNET_LOCK_EX); } @@ -1206,11 +1208,11 @@ rescan: lnet_prune_rc_data(0); /* don't wait for UNLINK */ - /* Call cfs_pause() here always adds 1 to load average + /* Call schedule_timeout() here always adds 1 to load average * because kernel counts # active tasks as nr_running * + nr_uninterruptible. */ - schedule_timeout_and_set_state(TASK_INTERRUPTIBLE, - cfs_time_seconds(1)); + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); } LASSERT(the_lnet.ln_rc_state == LNET_RC_STATE_STOPPING); diff --git a/drivers/staging/lustre/lnet/lnet/router_proc.c b/drivers/staging/lustre/lnet/lnet/router_proc.c index 20d53e08705e..0cbd9fc98e07 100644 --- a/drivers/staging/lustre/lnet/lnet/router_proc.c +++ b/drivers/staging/lustre/lnet/lnet/router_proc.c @@ -837,8 +837,8 @@ static int __proc_lnet_portal_rotor(void *data, int write, rc = -EINVAL; lnet_res_lock(0); for (i = 0; portal_rotors[i].pr_name != NULL; i++) { - if (cfs_strncasecmp(portal_rotors[i].pr_name, tmp, - strlen(portal_rotors[i].pr_name)) == 0) { + if (strncasecmp(portal_rotors[i].pr_name, tmp, + strlen(portal_rotors[i].pr_name)) == 0) { portal_rotor = portal_rotors[i].pr_value; rc = 0; break; diff --git a/drivers/staging/lustre/lnet/selftest/conrpc.c b/drivers/staging/lustre/lnet/selftest/conrpc.c index 53d58924737b..8d1eea4cef6f 100644 --- a/drivers/staging/lustre/lnet/selftest/conrpc.c +++ b/drivers/staging/lustre/lnet/selftest/conrpc.c @@ -1346,7 +1346,8 @@ lstcon_rpc_cleanup_wait(void) mutex_unlock(&console_session.ses_mutex); CWARN("Session is shutting down, waiting for termination of transactions\n"); - cfs_pause(cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); mutex_lock(&console_session.ses_mutex); } diff --git a/drivers/staging/lustre/lnet/selftest/console.c b/drivers/staging/lustre/lnet/selftest/console.c index 2a8eddc7db52..352fc96398d2 100644 --- a/drivers/staging/lustre/lnet/selftest/console.c +++ b/drivers/staging/lustre/lnet/selftest/console.c @@ -61,7 +61,7 @@ do { \ lstcon_session_t console_session; -void +static void lstcon_node_get(lstcon_node_t *nd) { LASSERT (nd->nd_ref >= 1); @@ -114,7 +114,7 @@ lstcon_node_find(lnet_process_id_t id, lstcon_node_t **ndpp, int create) return 0; } -void +static void lstcon_node_put(lstcon_node_t *nd) { lstcon_ndlink_t *ndl; @@ -344,7 +344,7 @@ lstcon_group_move(lstcon_group_t *old, lstcon_group_t *new) } } -int +static int lstcon_sesrpc_condition(int transop, lstcon_node_t *nd, void *arg) { lstcon_group_t *grp = (lstcon_group_t *)arg; @@ -373,7 +373,7 @@ lstcon_sesrpc_condition(int transop, lstcon_node_t *nd, void *arg) return 1; } -int +static int lstcon_sesrpc_readent(int transop, srpc_msg_t *msg, lstcon_rpc_ent_t *ent_up) { @@ -830,7 +830,7 @@ lstcon_group_info(char *name, lstcon_ndlist_ent_t *gents_p, return 0; } -int +static int lstcon_batch_find(const char *name, lstcon_batch_t **batpp) { lstcon_batch_t *bat; @@ -998,7 +998,7 @@ lstcon_batch_info(char *name, lstcon_test_batch_ent_t *ent_up, int server, return rc; } -int +static int lstcon_batrpc_condition(int transop, lstcon_node_t *nd, void *arg) { switch (transop) { @@ -1141,7 +1141,7 @@ lstcon_batch_destroy(lstcon_batch_t *bat) LIBCFS_FREE(bat, sizeof(lstcon_batch_t)); } -int +static int lstcon_testrpc_condition(int transop, lstcon_node_t *nd, void *arg) { lstcon_test_t *test; @@ -1370,7 +1370,7 @@ out: return rc; } -int +static int lstcon_test_find(lstcon_batch_t *batch, int idx, lstcon_test_t **testpp) { lstcon_test_t *test; @@ -1385,7 +1385,7 @@ lstcon_test_find(lstcon_batch_t *batch, int idx, lstcon_test_t **testpp) return -ENOENT; } -int +static int lstcon_tsbrpc_readent(int transop, srpc_msg_t *msg, lstcon_rpc_ent_t *ent_up) { @@ -1464,7 +1464,7 @@ lstcon_test_batch_query(char *name, int testidx, int client, return rc; } -int +static int lstcon_statrpc_readent(int transop, srpc_msg_t *msg, lstcon_rpc_ent_t *ent_up) { @@ -1488,7 +1488,7 @@ lstcon_statrpc_readent(int transop, srpc_msg_t *msg, return 0; } -int +static int lstcon_ndlist_stat(struct list_head *ndlist, int timeout, struct list_head *result_up) { @@ -1577,7 +1577,7 @@ lstcon_nodes_stat(int count, lnet_process_id_t *ids_up, return rc; } -int +static int lstcon_debug_ndlist(struct list_head *ndlist, struct list_head *translist, int timeout, struct list_head *result_up) diff --git a/drivers/staging/lustre/lnet/selftest/rpc.c b/drivers/staging/lustre/lnet/selftest/rpc.c index d838985f51cb..9fc0429e8296 100644 --- a/drivers/staging/lustre/lnet/selftest/rpc.c +++ b/drivers/staging/lustre/lnet/selftest/rpc.c @@ -1585,7 +1585,8 @@ srpc_startup (void) spin_lock_init(&srpc_data.rpc_glock); /* 1 second pause to avoid timestamp reuse */ - cfs_pause(cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); srpc_data.rpc_matchbits = ((__u64) cfs_time_current_sec()) << 48; srpc_data.rpc_state = SRPC_STATE_NONE; diff --git a/drivers/staging/lustre/lnet/selftest/selftest.h b/drivers/staging/lustre/lnet/selftest/selftest.h index 228927e0f962..f4806a6bc942 100644 --- a/drivers/staging/lustre/lnet/selftest/selftest.h +++ b/drivers/staging/lustre/lnet/selftest/selftest.h @@ -572,7 +572,11 @@ swi_state2str (int state) #undef STATE2STR } -#define selftest_wait_events() cfs_pause(cfs_time_seconds(1) / 10) +#define selftest_wait_events() \ + do { \ + set_current_state(TASK_UNINTERRUPTIBLE); \ + schedule_timeout(cfs_time_seconds(1) / 10); \ + } while (0) #define lst_wait_until(cond, lock, fmt, ...) \ diff --git a/drivers/staging/lustre/lustre/fid/fid_request.c b/drivers/staging/lustre/lustre/fid/fid_request.c index 79fc2fe131a2..3401c9ad42ac 100644 --- a/drivers/staging/lustre/lustre/fid/fid_request.c +++ b/drivers/staging/lustre/lustre/fid/fid_request.c @@ -196,8 +196,8 @@ static int seq_client_alloc_seq(const struct lu_env *env, if (range_is_exhausted(&seq->lcs_space)) { rc = seq_client_alloc_meta(env, seq); if (rc) { - CERROR("%s: Can't allocate new meta-sequence," - "rc %d\n", seq->lcs_name, rc); + CERROR("%s: Can't allocate new meta-sequence, rc %d\n", + seq->lcs_name, rc); return rc; } else { CDEBUG(D_INFO, "%s: New range - "DRANGE"\n", @@ -225,7 +225,7 @@ static int seq_fid_alloc_prep(struct lu_client_seq *seq, set_current_state(TASK_UNINTERRUPTIBLE); mutex_unlock(&seq->lcs_mutex); - waitq_wait(link, TASK_UNINTERRUPTIBLE); + schedule(); mutex_lock(&seq->lcs_mutex); remove_wait_queue(&seq->lcs_waitq, link); @@ -256,7 +256,7 @@ int seq_client_get_seq(const struct lu_env *env, LASSERT(seqnr != NULL); mutex_lock(&seq->lcs_mutex); - init_waitqueue_entry_current(&link); + init_waitqueue_entry(&link, current); while (1) { rc = seq_fid_alloc_prep(seq, &link); @@ -266,15 +266,15 @@ int seq_client_get_seq(const struct lu_env *env, rc = seq_client_alloc_seq(env, seq, seqnr); if (rc) { - CERROR("%s: Can't allocate new sequence, " - "rc %d\n", seq->lcs_name, rc); + CERROR("%s: Can't allocate new sequence, rc %d\n", + seq->lcs_name, rc); seq_fid_alloc_fini(seq); mutex_unlock(&seq->lcs_mutex); return rc; } - CDEBUG(D_INFO, "%s: allocate sequence " - "[0x%16.16"LPF64"x]\n", seq->lcs_name, *seqnr); + CDEBUG(D_INFO, "%s: allocate sequence [0x%16.16"LPF64"x]\n", + seq->lcs_name, *seqnr); /* Since the caller require the whole seq, * so marked this seq to be used */ @@ -306,7 +306,7 @@ int seq_client_alloc_fid(const struct lu_env *env, LASSERT(seq != NULL); LASSERT(fid != NULL); - init_waitqueue_entry_current(&link); + init_waitqueue_entry(&link, current); mutex_lock(&seq->lcs_mutex); if (OBD_FAIL_CHECK(OBD_FAIL_SEQ_EXHAUST)) @@ -329,15 +329,15 @@ int seq_client_alloc_fid(const struct lu_env *env, rc = seq_client_alloc_seq(env, seq, &seqnr); if (rc) { - CERROR("%s: Can't allocate new sequence, " - "rc %d\n", seq->lcs_name, rc); + CERROR("%s: Can't allocate new sequence, rc %d\n", + seq->lcs_name, rc); seq_fid_alloc_fini(seq); mutex_unlock(&seq->lcs_mutex); return rc; } - CDEBUG(D_INFO, "%s: Switch to sequence " - "[0x%16.16"LPF64"x]\n", seq->lcs_name, seqnr); + CDEBUG(D_INFO, "%s: Switch to sequence [0x%16.16"LPF64"x]\n", + seq->lcs_name, seqnr); seq->lcs_fid.f_oid = LUSTRE_FID_INIT_OID; seq->lcs_fid.f_seq = seqnr; @@ -370,7 +370,7 @@ void seq_client_flush(struct lu_client_seq *seq) wait_queue_t link; LASSERT(seq != NULL); - init_waitqueue_entry_current(&link); + init_waitqueue_entry(&link, current); mutex_lock(&seq->lcs_mutex); while (seq->lcs_update) { @@ -378,7 +378,7 @@ void seq_client_flush(struct lu_client_seq *seq) set_current_state(TASK_UNINTERRUPTIBLE); mutex_unlock(&seq->lcs_mutex); - waitq_wait(&link, TASK_UNINTERRUPTIBLE); + schedule(); mutex_lock(&seq->lcs_mutex); remove_wait_queue(&seq->lcs_waitq, &link); @@ -428,8 +428,8 @@ static int seq_client_proc_init(struct lu_client_seq *seq) rc = lprocfs_add_vars(seq->lcs_proc_dir, seq_client_proc_list, seq); if (rc) { - CERROR("%s: Can't init sequence manager " - "proc, rc %d\n", seq->lcs_name, rc); + CERROR("%s: Can't init sequence manager proc, rc %d\n", + seq->lcs_name, rc); GOTO(out_cleanup, rc); } diff --git a/drivers/staging/lustre/lustre/fld/fld_cache.c b/drivers/staging/lustre/lustre/fld/fld_cache.c index 6c379301df82..a06a642f549e 100644 --- a/drivers/staging/lustre/lustre/fld/fld_cache.c +++ b/drivers/staging/lustre/lustre/fld/fld_cache.c @@ -236,8 +236,8 @@ static int fld_cache_shrink(struct fld_cache *cache) num++; } - CDEBUG(D_INFO, "%s: FLD cache - Shrunk by " - "%d entries\n", cache->fci_name, num); + CDEBUG(D_INFO, "%s: FLD cache - Shrunk by %d entries\n", + cache->fci_name, num); return 0; } @@ -355,7 +355,7 @@ static void fld_cache_overlap_handle(struct fld_cache *cache, fld_cache_entry_add(cache, f_new, &f_curr->fce_list); } else CERROR("NEW range ="DRANGE" curr = "DRANGE"\n", - PRANGE(range),PRANGE(&f_curr->fce_range)); + PRANGE(range), PRANGE(&f_curr->fce_range)); } struct fld_cache_entry diff --git a/drivers/staging/lustre/lustre/fld/fld_internal.h b/drivers/staging/lustre/lustre/fld/fld_internal.h index 5f3935cc0fd7..8661a788f120 100644 --- a/drivers/staging/lustre/lustre/fld/fld_internal.h +++ b/drivers/staging/lustre/lustre/fld/fld_internal.h @@ -90,7 +90,7 @@ struct fld_cache { int fci_threshold; /** - * Prefered number of cached entries */ + * Preferred number of cached entries */ int fci_cache_size; /** diff --git a/drivers/staging/lustre/lustre/fld/fld_request.c b/drivers/staging/lustre/lustre/fld/fld_request.c index 896f9fe83ffd..1f8abba31428 100644 --- a/drivers/staging/lustre/lustre/fld/fld_request.c +++ b/drivers/staging/lustre/lustre/fld/fld_request.c @@ -138,9 +138,8 @@ fld_rrb_scan(struct lu_client_fld *fld, seqno_t seq) return target; } - CERROR("%s: Can't find target by hash %d (seq "LPX64"). " - "Targets (%d):\n", fld->lcf_name, hash, seq, - fld->lcf_count); + CERROR("%s: Can't find target by hash %d (seq "LPX64"). Targets (%d):\n", + fld->lcf_name, hash, seq, fld->lcf_count); list_for_each_entry(target, &fld->lcf_targets, ft_chain) { const char *srv_name = target->ft_srv != NULL ? @@ -209,9 +208,8 @@ int fld_client_add_target(struct lu_client_fld *fld, LASSERT(tar->ft_srv != NULL || tar->ft_exp != NULL); if (fld->lcf_flags != LUSTRE_FLD_INIT) { - CERROR("%s: Attempt to add target %s (idx "LPU64") " - "on fly - skip it\n", fld->lcf_name, name, - tar->ft_idx); + CERROR("%s: Attempt to add target %s (idx "LPU64") on fly - skip it\n", + fld->lcf_name, name, tar->ft_idx); return 0; } else { CDEBUG(D_INFO, "%s: Adding target %s (idx " @@ -476,9 +474,8 @@ int fld_client_lookup(struct lu_client_fld *fld, seqno_t seq, mdsno_t *mds, target = fld_client_get_target(fld, seq); LASSERT(target != NULL); - CDEBUG(D_INFO, "%s: Lookup fld entry (seq: "LPX64") on " - "target %s (idx "LPU64")\n", fld->lcf_name, seq, - fld_target_name(target), target->ft_idx); + CDEBUG(D_INFO, "%s: Lookup fld entry (seq: "LPX64") on target %s (idx "LPU64")\n", + fld->lcf_name, seq, fld_target_name(target), target->ft_idx); res.lsr_start = seq; fld_range_set_type(&res, flags); diff --git a/drivers/staging/lustre/lustre/include/cl_object.h b/drivers/staging/lustre/lustre/include/cl_object.h index 4d692dcd96cf..c809239c0866 100644 --- a/drivers/staging/lustre/lustre/include/cl_object.h +++ b/drivers/staging/lustre/lustre/include/cl_object.h @@ -1314,7 +1314,7 @@ static inline int __page_in_use(const struct cl_page *page, int refc) * calls. To achieve this, every layer can implement ->clo_fits_into() method, * that is called by lock matching code (cl_lock_lookup()), and that can be * used to selectively disable matching of certain locks for certain IOs. For - * exmaple, lov layer implements lov_lock_fits_into() that allow multi-stripe + * example, lov layer implements lov_lock_fits_into() that allow multi-stripe * locks to be matched only for truncates and O_APPEND writes. * * Interaction with DLM @@ -2385,14 +2385,18 @@ struct cl_io { * Check if layout changed after the IO finishes. Mainly for HSM * requirement. If IO occurs to openning files, it doesn't need to * verify layout because HSM won't release openning files. - * Right now, only two opertaions need to verify layout: glimpse + * Right now, only two operations need to verify layout: glimpse * and setattr. */ ci_verify_layout:1, /** * file is released, restore has to to be triggered by vvp layer */ - ci_restore_needed:1; + ci_restore_needed:1, + /** + * O_NOATIME + */ + ci_noatime:1; /** * Number of pages owned by this IO. For invariant checking. */ @@ -2552,7 +2556,7 @@ struct cl_req_obj { */ struct cl_req { enum cl_req_type crq_type; - /** A list of pages being transfered */ + /** A list of pages being transferred */ struct list_head crq_pages; /** Number of pages in cl_req::crq_pages */ unsigned crq_nrpages; @@ -3224,7 +3228,7 @@ void cl_sync_io_note(struct cl_sync_io *anchor, int ioret); * * - call chains have no non-lustre portions inserted between lustre code. * - * On a client both these assumtpion fails, because every user thread can + * On a client both these assumption fails, because every user thread can * potentially execute lustre code as part of a system call, and lustre calls * into VFS or MM that call back into lustre. * diff --git a/drivers/staging/lustre/lustre/include/ioctl.h b/drivers/staging/lustre/lustre/include/ioctl.h index 227c261b2ae9..b986920926bd 100644 --- a/drivers/staging/lustre/lustre/include/ioctl.h +++ b/drivers/staging/lustre/lustre/include/ioctl.h @@ -38,7 +38,7 @@ * and on newer kernels this header is shared as _ASM_GENERIC_IOCTL_H. * * We can avoid any problems with the kernel header being included again by - * defining _ASM_I386_IOCTL_H here so that a later occurence of <asm/ioctl.h> + * defining _ASM_I386_IOCTL_H here so that a later occurrence of <asm/ioctl.h> * does not include the kernel's ioctl.h after this one. b=14746 */ #define _ASM_I386_IOCTL_H #define _ASM_GENERIC_IOCTL_H diff --git a/drivers/staging/lustre/lustre/include/lclient.h b/drivers/staging/lustre/lustre/include/lclient.h index 27316f7b7a70..827209ea6bd0 100644 --- a/drivers/staging/lustre/lustre/include/lclient.h +++ b/drivers/staging/lustre/lustre/include/lclient.h @@ -118,8 +118,8 @@ struct ccc_io { }; /** - * True, if \a io is a normal io, False for other (sendfile, splice*). - * must be impementated in arch specific code. + * True, if \a io is a normal io, False for splice_{read,write}. + * must be implemented in arch specific code. */ int cl_is_normalio(const struct lu_env *env, const struct cl_io *io); diff --git a/drivers/staging/lustre/lustre/include/linux/lustre_acl.h b/drivers/staging/lustre/lustre/include/linux/lustre_acl.h index 778b123ce31e..a91c5497d22c 100644 --- a/drivers/staging/lustre/lustre/include/linux/lustre_acl.h +++ b/drivers/staging/lustre/lustre/include/linux/lustre_acl.h @@ -44,7 +44,7 @@ #define _LUSTRE_LINUX_ACL_H #ifndef _LUSTRE_ACL_H -#error Shoud not include direectly. use #include <lustre_acl.h> instead +#error Should not include directly. use #include <lustre_acl.h> instead #endif #include <linux/fs.h> diff --git a/drivers/staging/lustre/lustre/include/linux/obd.h b/drivers/staging/lustre/lustre/include/linux/obd.h index 01a50265239d..dc36f75eb635 100644 --- a/drivers/staging/lustre/lustre/include/linux/obd.h +++ b/drivers/staging/lustre/lustre/include/linux/obd.h @@ -96,7 +96,8 @@ static inline void __client_obd_list_lock(client_obd_lock_t *lock, LCONSOLE_WARN("====== for current process =====\n"); dump_stack(); LCONSOLE_WARN("====== end =======\n"); - cfs_pause(1000 * HZ); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(1000 * HZ); } cpu_relax(); } diff --git a/drivers/staging/lustre/lustre/include/lustre/lustre_idl.h b/drivers/staging/lustre/lustre/include/lustre/lustre_idl.h index 5da31c54924a..87905bbc68e1 100644 --- a/drivers/staging/lustre/lustre/include/lustre/lustre_idl.h +++ b/drivers/staging/lustre/lustre/include/lustre/lustre_idl.h @@ -91,8 +91,8 @@ #ifndef _LUSTRE_IDL_H_ #define _LUSTRE_IDL_H_ -#if !defined(LASSERT) && !defined(LPU64) -#include <linux/libcfs/libcfs.h> /* for LASSERT, LPUX64, etc */ +#if !defined(LPU64) +#include <linux/libcfs/libcfs.h> /* for LPUX64, etc */ #endif /* Defn's shared with user-space. */ @@ -232,7 +232,6 @@ static inline unsigned fld_range_is_any(const struct lu_seq_range *range) static inline void fld_range_set_type(struct lu_seq_range *range, unsigned flags) { - LASSERT(!(flags & ~LU_SEQ_RANGE_MASK)); range->lsr_flags |= flags; } @@ -351,7 +350,7 @@ struct som_attrs { /** Bitfield for supported data in this structure. For future use. */ __u32 som_compat; - /** Incompat feature list. The supported feature mask is availabe in + /** Incompat feature list. The supported feature mask is available in * SOM_INCOMPAT_SUPP */ __u32 som_incompat; @@ -615,7 +614,6 @@ static inline obd_id fid_idif_id(obd_seq seq, __u32 oid, __u32 ver) /* extract ost index from IDIF FID */ static inline __u32 fid_idif_ost_idx(const struct lu_fid *fid) { - LASSERT(fid_is_idif(fid)); return (fid_seq(fid) >> 16) & 0xffff; } @@ -833,11 +831,6 @@ static inline void lu_igif_build(struct lu_fid *fid, __u32 ino, __u32 gen) */ static inline void fid_cpu_to_le(struct lu_fid *dst, const struct lu_fid *src) { - /* check that all fields are converted */ - CLASSERT(sizeof(*src) == - sizeof(fid_seq(src)) + - sizeof(fid_oid(src)) + - sizeof(fid_ver(src))); dst->f_seq = cpu_to_le64(fid_seq(src)); dst->f_oid = cpu_to_le32(fid_oid(src)); dst->f_ver = cpu_to_le32(fid_ver(src)); @@ -845,11 +838,6 @@ static inline void fid_cpu_to_le(struct lu_fid *dst, const struct lu_fid *src) static inline void fid_le_to_cpu(struct lu_fid *dst, const struct lu_fid *src) { - /* check that all fields are converted */ - CLASSERT(sizeof(*src) == - sizeof(fid_seq(src)) + - sizeof(fid_oid(src)) + - sizeof(fid_ver(src))); dst->f_seq = le64_to_cpu(fid_seq(src)); dst->f_oid = le32_to_cpu(fid_oid(src)); dst->f_ver = le32_to_cpu(fid_ver(src)); @@ -857,11 +845,6 @@ static inline void fid_le_to_cpu(struct lu_fid *dst, const struct lu_fid *src) static inline void fid_cpu_to_be(struct lu_fid *dst, const struct lu_fid *src) { - /* check that all fields are converted */ - CLASSERT(sizeof(*src) == - sizeof(fid_seq(src)) + - sizeof(fid_oid(src)) + - sizeof(fid_ver(src))); dst->f_seq = cpu_to_be64(fid_seq(src)); dst->f_oid = cpu_to_be32(fid_oid(src)); dst->f_ver = cpu_to_be32(fid_ver(src)); @@ -869,11 +852,6 @@ static inline void fid_cpu_to_be(struct lu_fid *dst, const struct lu_fid *src) static inline void fid_be_to_cpu(struct lu_fid *dst, const struct lu_fid *src) { - /* check that all fields are converted */ - CLASSERT(sizeof(*src) == - sizeof(fid_seq(src)) + - sizeof(fid_oid(src)) + - sizeof(fid_ver(src))); dst->f_seq = be64_to_cpu(fid_seq(src)); dst->f_oid = be32_to_cpu(fid_oid(src)); dst->f_ver = be32_to_cpu(fid_ver(src)); @@ -897,11 +875,6 @@ extern void lustre_swab_lu_seq_range(struct lu_seq_range *range); static inline int lu_fid_eq(const struct lu_fid *f0, const struct lu_fid *f1) { - /* Check that there is no alignment padding. */ - CLASSERT(sizeof(*f0) == - sizeof(f0->f_seq) + - sizeof(f0->f_oid) + - sizeof(f0->f_ver)); return memcmp(f0, f1, sizeof(*f0)) == 0; } @@ -960,7 +933,7 @@ enum lu_dirent_attrs { LUDA_TYPE = 0x0002, LUDA_64BITHASH = 0x0004, - /* The following attrs are used for MDT interanl only, + /* The following attrs are used for MDT internal only, * not visible to client */ /* Verify the dirent consistency */ @@ -1331,6 +1304,9 @@ extern void lustre_swab_ptlrpc_body(struct ptlrpc_body *pb); #define OBD_CONNECT_LIGHTWEIGHT 0x1000000000000ULL/* lightweight connection */ #define OBD_CONNECT_SHORTIO 0x2000000000000ULL/* short io */ #define OBD_CONNECT_PINGLESS 0x4000000000000ULL/* pings not required */ +#define OBD_CONNECT_FLOCK_DEAD 0x8000000000000ULL/* flock deadlock detection */ +#define OBD_CONNECT_DISP_STRIPE 0x10000000000000ULL/*create stripe disposition*/ + /* XXX README XXX: * Please DO NOT add flag values here before first ensuring that this same * flag value is not in use on some other branch. Please clear any such @@ -1368,7 +1344,10 @@ extern void lustre_swab_ptlrpc_body(struct ptlrpc_body *pb); OBD_CONNECT_EINPROGRESS | \ OBD_CONNECT_LIGHTWEIGHT | OBD_CONNECT_UMASK | \ OBD_CONNECT_LVB_TYPE | OBD_CONNECT_LAYOUTLOCK |\ - OBD_CONNECT_PINGLESS | OBD_CONNECT_MAX_EASIZE) + OBD_CONNECT_PINGLESS | OBD_CONNECT_MAX_EASIZE |\ + OBD_CONNECT_FLOCK_DEAD | \ + OBD_CONNECT_DISP_STRIPE) + #define OST_CONNECT_SUPPORTED (OBD_CONNECT_SRVLOCK | OBD_CONNECT_GRANT | \ OBD_CONNECT_REQPORTAL | OBD_CONNECT_VERSION | \ OBD_CONNECT_TRUNCLOCK | OBD_CONNECT_INDEX | \ @@ -1771,7 +1750,6 @@ static inline __u32 lov_mds_md_size(__u16 stripes, __u32 lmm_magic) OBD_MD_FLGID | OBD_MD_FLFLAGS | OBD_MD_FLNLINK | \ OBD_MD_FLGENER | OBD_MD_FLRDEV | OBD_MD_FLGROUP) -#define OBD_MD_FLXATTRLOCKED OBD_MD_FLGETATTRLOCK #define OBD_MD_FLXATTRALL (OBD_MD_FLXATTR | OBD_MD_FLXATTRLS) /* don't forget obdo_fid which is way down at the bottom so it can @@ -2134,19 +2112,32 @@ extern void lustre_swab_generic_32s (__u32 *val); #define DISP_LOOKUP_POS 0x00000008 #define DISP_OPEN_CREATE 0x00000010 #define DISP_OPEN_OPEN 0x00000020 -#define DISP_ENQ_COMPLETE 0x00400000 +#define DISP_ENQ_COMPLETE 0x00400000 /* obsolete and unused */ #define DISP_ENQ_OPEN_REF 0x00800000 #define DISP_ENQ_CREATE_REF 0x01000000 #define DISP_OPEN_LOCK 0x02000000 #define DISP_OPEN_LEASE 0x04000000 +#define DISP_OPEN_STRIPE 0x08000000 /* INODE LOCK PARTS */ -#define MDS_INODELOCK_LOOKUP 0x000001 /* dentry, mode, owner, group */ -#define MDS_INODELOCK_UPDATE 0x000002 /* size, links, timestamps */ -#define MDS_INODELOCK_OPEN 0x000004 /* For opened files */ -#define MDS_INODELOCK_LAYOUT 0x000008 /* for layout */ -#define MDS_INODELOCK_PERM 0x000010 /* for permission */ -#define MDS_INODELOCK_XATTR 0x000020 /* extended attributes */ +#define MDS_INODELOCK_LOOKUP 0x000001 /* For namespace, dentry etc, and also + * was used to protect permission (mode, + * owner, group etc) before 2.4. */ +#define MDS_INODELOCK_UPDATE 0x000002 /* size, links, timestamps */ +#define MDS_INODELOCK_OPEN 0x000004 /* For opened files */ +#define MDS_INODELOCK_LAYOUT 0x000008 /* for layout */ + +/* The PERM bit is added int 2.4, and it is used to protect permission(mode, + * owner, group, acl etc), so to separate the permission from LOOKUP lock. + * Because for remote directories(in DNE), these locks will be granted by + * different MDTs(different ldlm namespace). + * + * For local directory, MDT will always grant UPDATE_LOCK|PERM_LOCK together. + * For Remote directory, the master MDT, where the remote directory is, will + * grant UPDATE_LOCK|PERM_LOCK, and the remote MDT, where the name entry is, + * will grant LOOKUP_LOCK. */ +#define MDS_INODELOCK_PERM 0x000010 +#define MDS_INODELOCK_XATTR 0x000020 /* extended attributes */ #define MDS_INODELOCK_MAXSHIFT 5 /* This FULL lock is useful to take on unlink sort of operations */ @@ -2595,7 +2586,7 @@ struct mdt_rec_setxattr { * Do NOT change the size of various members, otherwise the value * will be broken in lustre_swab_mdt_rec_reint(). * - * If you add new members in other mdt_reint_xxx structres and need to use the + * If you add new members in other mdt_reint_xxx structures and need to use the * rr_padding_x fields, then update lustre_swab_mdt_rec_reint() also. */ struct mdt_rec_reint { @@ -3328,9 +3319,10 @@ struct obdo { #define o_grant_used o_data_version static inline void lustre_set_wire_obdo(struct obd_connect_data *ocd, - struct obdo *wobdo, struct obdo *lobdo) + struct obdo *wobdo, + const struct obdo *lobdo) { - memcpy(wobdo, lobdo, sizeof(*lobdo)); + *wobdo = *lobdo; wobdo->o_flags &= ~OBD_FL_LOCAL_MASK; if (ocd == NULL) return; @@ -3345,16 +3337,15 @@ static inline void lustre_set_wire_obdo(struct obd_connect_data *ocd, } static inline void lustre_get_wire_obdo(struct obd_connect_data *ocd, - struct obdo *lobdo, struct obdo *wobdo) + struct obdo *lobdo, + const struct obdo *wobdo) { obd_flag local_flags = 0; if (lobdo->o_valid & OBD_MD_FLFLAGS) local_flags = lobdo->o_flags & OBD_FL_LOCAL_MASK; - LASSERT(!(wobdo->o_flags & OBD_FL_LOCAL_MASK)); - - memcpy(lobdo, wobdo, sizeof(*lobdo)); + *lobdo = *wobdo; if (local_flags != 0) { lobdo->o_valid |= OBD_MD_FLFLAGS; lobdo->o_flags &= ~OBD_FL_LOCAL_MASK; diff --git a/drivers/staging/lustre/lustre/include/lustre/lustre_user.h b/drivers/staging/lustre/lustre/include/lustre/lustre_user.h index 6b6c0240e824..f5f369e603de 100644 --- a/drivers/staging/lustre/lustre/include/lustre/lustre_user.h +++ b/drivers/staging/lustre/lustre/include/lustre/lustre_user.h @@ -265,13 +265,11 @@ struct ost_id { #define MAX_OBD_NAME 128 /* If this changes, a NEW ioctl must be added */ -/* Hopefully O_LOV_DELAY_CREATE does not conflict with standard O_xxx flags. - * Previously it was defined as 0100000000 and conflicts with FMODE_NONOTIFY - * which was added since kernel 2.6.36, so we redefine it as 020000000. - * To be compatible with old version's statically linked binary, finally we - * define it as (020000000 | 0100000000). - * */ -#define O_LOV_DELAY_CREATE 0120000000 +/* Define O_LOV_DELAY_CREATE to be a mask that is not useful for regular + * files, but are unlikely to be used in practice and are not harmful if + * used incorrectly. O_NOCTTY and FASYNC are only meaningful for character + * devices and are safe for use on new files (See LU-812, LU-4209). */ +#define O_LOV_DELAY_CREATE (O_NOCTTY | FASYNC) #define LL_FILE_IGNORE_LOCK 0x00000001 #define LL_FILE_GROUP_LOCKED 0x00000002 @@ -300,7 +298,7 @@ struct ost_id { #define LOV_MAX_STRIPE_COUNT_OLD 160 /* This calculation is crafted so that input of 4096 will result in 160 * which in turn is equal to old maximal stripe count. - * XXX: In fact this is too simpified for now, what it also need is to get + * XXX: In fact this is too simplified for now, what it also need is to get * ea_type argument to clearly know how much space each stripe consumes. * * The limit of 12 pages is somewhat arbitrary, but is a reasonably large @@ -760,7 +758,8 @@ static inline void hsm_set_cl_error(int *flags, int error) *flags |= (error << CLF_HSM_ERR_L); } -#define CR_MAXSIZE cfs_size_round(2*NAME_MAX + 1 + sizeof(struct changelog_rec)) +#define CR_MAXSIZE cfs_size_round(2*NAME_MAX + 1 + \ + sizeof(struct changelog_ext_rec)) struct changelog_rec { __u16 cr_namelen; @@ -929,7 +928,7 @@ struct hsm_state_set_ioc { /* * This structure describes the current in-progress action for a file. - * it is retuned to user space and send over the wire + * it is returned to user space and send over the wire */ struct hsm_current_action { /** The current undergoing action, if there is one */ @@ -1158,12 +1157,6 @@ struct hsm_progress { __u32 padding; }; -/** - * Use by copytool during any hsm request they handled. - * This structure is initialized by llapi_hsm_copy_start() - * which is an helper over the ioctl() interface - * Store Lustre, internal use only, data. - */ struct hsm_copy { __u64 hc_data_version; __u16 hc_flags; diff --git a/drivers/staging/lustre/lustre/include/lustre_cfg.h b/drivers/staging/lustre/lustre/include/lustre_cfg.h index e14a5f674e85..3680668a8920 100644 --- a/drivers/staging/lustre/lustre/include/lustre_cfg.h +++ b/drivers/staging/lustre/lustre/include/lustre_cfg.h @@ -88,6 +88,8 @@ enum lcfg_command_type { LCFG_SET_LDLM_TIMEOUT = 0x00ce030, /**< set ldlm_timeout */ LCFG_PRE_CLEANUP = 0x00cf031, /**< call type-specific pre * cleanup cleanup */ + LCFG_SET_PARAM = 0x00ce032, /**< use set_param syntax to set + *a proc parameters */ }; struct lustre_cfg_bufs { diff --git a/drivers/staging/lustre/lustre/include/lustre_disk.h b/drivers/staging/lustre/lustre/include/lustre_disk.h index 1de9a8bed497..ac08164793cb 100644 --- a/drivers/staging/lustre/lustre/include/lustre_disk.h +++ b/drivers/staging/lustre/lustre/include/lustre_disk.h @@ -99,6 +99,8 @@ #define LDD_F_IR_CAPABLE 0x2000 /** the MGS refused to register the target. */ #define LDD_F_ERROR 0x4000 +/** process at lctl conf_param */ +#define LDD_F_PARAM2 0x8000 /* opc for target register */ #define LDD_F_OPC_REG 0x10000000 diff --git a/drivers/staging/lustre/lustre/include/lustre_dlm.h b/drivers/staging/lustre/lustre/include/lustre_dlm.h index ec4bb5e3c13e..3e25f001c07d 100644 --- a/drivers/staging/lustre/lustre/include/lustre_dlm.h +++ b/drivers/staging/lustre/lustre/include/lustre_dlm.h @@ -330,7 +330,7 @@ enum { }; typedef enum { - /** invalide type */ + /** invalid type */ LDLM_NS_TYPE_UNKNOWN = 0, /** mdc namespace */ LDLM_NS_TYPE_MDC, @@ -1185,7 +1185,7 @@ ldlm_handle2lock_long(const struct lustre_handle *h, __u64 flags) /** * Update Lock Value Block Operations (LVBO) on a resource taking into account - * data from reqest \a r + * data from request \a r */ static inline int ldlm_res_lvbo_update(struct ldlm_resource *res, struct ptlrpc_request *r, int increase) diff --git a/drivers/staging/lustre/lustre/include/lustre_dlm_flags.h b/drivers/staging/lustre/lustre/include/lustre_dlm_flags.h index 75716f17f64b..16dcdbfae689 100644 --- a/drivers/staging/lustre/lustre/include/lustre_dlm_flags.h +++ b/drivers/staging/lustre/lustre/include/lustre_dlm_flags.h @@ -35,10 +35,10 @@ #ifndef LDLM_ALL_FLAGS_MASK /** l_flags bits marked as "all_flags" bits */ -#define LDLM_FL_ALL_FLAGS_MASK 0x00FFFFFFC08F132FULL +#define LDLM_FL_ALL_FLAGS_MASK 0x00FFFFFFC08F932FULL /** l_flags bits marked as "ast" bits */ -#define LDLM_FL_AST_MASK 0x0000000080000000ULL +#define LDLM_FL_AST_MASK 0x0000000080008000ULL /** l_flags bits marked as "blocked" bits */ #define LDLM_FL_BLOCKED_MASK 0x000000000000000EULL @@ -56,7 +56,7 @@ #define LDLM_FL_LOCAL_ONLY_MASK 0x00FFFFFF00000000ULL /** l_flags bits marked as "on_wire" bits */ -#define LDLM_FL_ON_WIRE_MASK 0x00000000C08F132FULL +#define LDLM_FL_ON_WIRE_MASK 0x00000000C08F932FULL /** extent, mode, or resource changed */ #define LDLM_FL_LOCK_CHANGED 0x0000000000000001ULL // bit 0 @@ -114,6 +114,12 @@ #define ldlm_set_has_intent(_l) LDLM_SET_FLAG(( _l), 1ULL << 12) #define ldlm_clear_has_intent(_l) LDLM_CLEAR_FLAG((_l), 1ULL << 12) +/** flock deadlock detected */ +#define LDLM_FL_FLOCK_DEADLOCK 0x0000000000008000ULL /* bit 15 */ +#define ldlm_is_flock_deadlock(_l) LDLM_TEST_FLAG((_l), 1ULL << 15) +#define ldlm_set_flock_deadlock(_l) LDLM_SET_FLAG((_l), 1ULL << 15) +#define ldlm_clear_flock_deadlock(_l) LDLM_CLEAR_FLAG((_l), 1ULL << 15) + /** discard (no writeback) on cancel */ #define LDLM_FL_DISCARD_DATA 0x0000000000010000ULL // bit 16 #define ldlm_is_discard_data(_l) LDLM_TEST_FLAG(( _l), 1ULL << 16) @@ -141,7 +147,7 @@ #define ldlm_clear_test_lock(_l) LDLM_CLEAR_FLAG((_l), 1ULL << 19) /** - * Immediatelly cancel such locks when they block some other locks. Send + * Immediately cancel such locks when they block some other locks. Send * cancel notification to original lock holder, but expect no reply. This * is for clients (like liblustre) that cannot be expected to reliably * response to blocking AST. */ @@ -242,7 +248,7 @@ /** * A lock contributes to the known minimum size (KMS) calculation until it - * has finished the part of its cancelation that performs write back on its + * has finished the part of its cancellation that performs write back on its * dirty pages. It can remain on the granted list during this whole time. * Threads racing to update the KMS after performing their writeback need * to know to exclude each other's locks from the calculation as they walk @@ -390,6 +396,7 @@ static int hf_lustre_ldlm_fl_ast_sent = -1; static int hf_lustre_ldlm_fl_replay = -1; static int hf_lustre_ldlm_fl_intent_only = -1; static int hf_lustre_ldlm_fl_has_intent = -1; +static int hf_lustre_ldlm_fl_flock_deadlock = -1; static int hf_lustre_ldlm_fl_discard_data = -1; static int hf_lustre_ldlm_fl_no_timeout = -1; static int hf_lustre_ldlm_fl_block_nowait = -1; @@ -431,6 +438,7 @@ const value_string lustre_ldlm_flags_vals[] = { {LDLM_FL_REPLAY, "LDLM_FL_REPLAY"}, {LDLM_FL_INTENT_ONLY, "LDLM_FL_INTENT_ONLY"}, {LDLM_FL_HAS_INTENT, "LDLM_FL_HAS_INTENT"}, + {LDLM_FL_FLOCK_DEADLOCK, "LDLM_FL_FLOCK_DEADLOCK"}, {LDLM_FL_DISCARD_DATA, "LDLM_FL_DISCARD_DATA"}, {LDLM_FL_NO_TIMEOUT, "LDLM_FL_NO_TIMEOUT"}, {LDLM_FL_BLOCK_NOWAIT, "LDLM_FL_BLOCK_NOWAIT"}, diff --git a/drivers/staging/lustre/lustre/include/lustre_export.h b/drivers/staging/lustre/lustre/include/lustre_export.h index 2feb38b51af2..103f7a8bd83f 100644 --- a/drivers/staging/lustre/lustre/include/lustre_export.h +++ b/drivers/staging/lustre/lustre/include/lustre_export.h @@ -192,9 +192,9 @@ struct obd_export { struct obd_import *exp_imp_reverse; struct nid_stat *exp_nid_stats; struct lprocfs_stats *exp_md_stats; - /** Active connetion */ + /** Active connection */ struct ptlrpc_connection *exp_connection; - /** Connection count value from last succesful reconnect rpc */ + /** Connection count value from last successful reconnect rpc */ __u32 exp_conn_cnt; /** Hash list of all ldlm locks granted on this export */ struct cfs_hash *exp_lock_hash; @@ -380,6 +380,23 @@ static inline bool imp_connect_lvb_type(struct obd_import *imp) return false; } +static inline __u64 exp_connect_ibits(struct obd_export *exp) +{ + struct obd_connect_data *ocd; + + ocd = &exp->exp_connect_data; + return ocd->ocd_ibits_known; +} + +static inline bool imp_connect_disp_stripe(struct obd_import *imp) +{ + struct obd_connect_data *ocd; + + LASSERT(imp != NULL); + ocd = &imp->imp_connect_data; + return ocd->ocd_connect_flags & OBD_CONNECT_DISP_STRIPE; +} + extern struct obd_export *class_conn2export(struct lustre_handle *conn); extern struct obd_device *class_conn2obd(struct lustre_handle *conn); diff --git a/drivers/staging/lustre/lustre/include/lustre_fid.h b/drivers/staging/lustre/lustre/include/lustre_fid.h index 84a897eed1df..5e7b3165a851 100644 --- a/drivers/staging/lustre/lustre/include/lustre_fid.h +++ b/drivers/staging/lustre/lustre/include/lustre_fid.h @@ -339,7 +339,7 @@ struct lu_client_seq { struct mutex lcs_mutex; /* - * Range of allowed for allocation sequeces. When using lu_client_seq on + * Range of allowed for allocation sequences. When using lu_client_seq on * clients, this contains meta-sequence range. And for servers this * contains super-sequence range. */ @@ -398,7 +398,7 @@ struct lu_server_seq { /* LUSTRE_SEQ_SERVER or LUSTRE_SEQ_CONTROLLER */ enum lu_mgr_type lss_type; - /* Client interafce to request controller */ + /* Client interface to request controller */ struct lu_client_seq *lss_cli; /* Mutex for protecting allocation */ @@ -568,14 +568,14 @@ fid_build_pdo_res_name(const struct lu_fid *fid, unsigned int hash, * finally, when we replace ost_id with FID in data stack. * * Currently, resid from the old client, whose res[0] = object_id, - * res[1] = object_seq, is just oposite with Metatdata + * res[1] = object_seq, is just opposite with Metatdata * resid, where, res[0] = fid->f_seq, res[1] = fid->f_oid. - * To unifiy the resid identification, we will reverse the data + * To unify the resid identification, we will reverse the data * resid to keep it same with Metadata resid, i.e. * * For resid from the old client, * res[0] = objid, res[1] = 0, still keep the original order, - * for compatiblity. + * for compatibility. * * For new resid * res will be built from normal FID directly, i.e. res[0] = f_seq, @@ -685,7 +685,7 @@ static inline __u32 fid_hash(const struct lu_fid *f, int bits) { /* all objects with same id and different versions will belong to same * collisions list. */ - return cfs_hash_long(fid_flatten(f), bits); + return hash_long(fid_flatten(f), bits); } /** diff --git a/drivers/staging/lustre/lustre/include/lustre_import.h b/drivers/staging/lustre/lustre/include/lustre_import.h index 67259eb43cde..01ed786b40b9 100644 --- a/drivers/staging/lustre/lustre/include/lustre_import.h +++ b/drivers/staging/lustre/lustre/include/lustre_import.h @@ -152,7 +152,7 @@ struct import_state_hist { }; /** - * Defintion of PortalRPC import structure. + * Definition of PortalRPC import structure. * Imports are representing client-side view to remote target. */ struct obd_import { @@ -180,6 +180,17 @@ struct obd_import { struct list_head imp_delayed_list; /** @} */ + /** + * List of requests that are retained for committed open replay. Once + * open is committed, open replay request will be moved from the + * imp_replay_list into the imp_committed_list. + * The imp_replay_cursor is for accelerating searching during replay. + * @{ + */ + struct list_head imp_committed_list; + struct list_head *imp_replay_cursor; + /** @} */ + /** obd device for this import */ struct obd_device *imp_obd; @@ -219,7 +230,7 @@ struct obd_import { * after a check to save on unnecessary replay list iterations */ int imp_last_generation_checked; - /** Last tranno we replayed */ + /** Last transno we replayed */ __u64 imp_last_replay_transno; /** Last transno committed on remote side */ __u64 imp_peer_committed_transno; @@ -237,7 +248,7 @@ struct obd_import { struct lustre_handle imp_remote_handle; /** When to perform next ping. time in jiffies. */ cfs_time_t imp_next_ping; - /** When we last succesfully connected. time in 64bit jiffies */ + /** When we last successfully connected. time in 64bit jiffies */ __u64 imp_last_success_conn; /** List of all possible connection for import. */ @@ -268,7 +279,7 @@ struct obd_import { imp_no_lock_replay:1, /* recovery by versions was failed */ imp_vbr_failed:1, - /* force an immidiate ping */ + /* force an immediate ping */ imp_force_verify:1, /* force a scheduled ping */ imp_force_next_verify:1, @@ -281,7 +292,7 @@ struct obd_import { /* need IR MNE swab */ imp_need_mne_swab:1, /* import must be reconnected instead of - * chouse new connection */ + * chose new connection */ imp_force_reconnect:1, /* import has tried to connect with server */ imp_connect_tried:1; diff --git a/drivers/staging/lustre/lustre/include/lustre_lib.h b/drivers/staging/lustre/lustre/include/lustre_lib.h index 609a090484a6..0368ca6ffcc1 100644 --- a/drivers/staging/lustre/lustre/include/lustre_lib.h +++ b/drivers/staging/lustre/lustre/include/lustre_lib.h @@ -536,7 +536,7 @@ do { \ if (condition) \ break; \ \ - init_waitqueue_entry_current(&__wait); \ + init_waitqueue_entry(&__wait, current); \ l_add_wait(&wq, &__wait); \ \ /* Block all signals (just the non-fatal ones if no timeout). */ \ @@ -558,15 +558,13 @@ do { \ break; \ \ if (__timeout == 0) { \ - waitq_wait(&__wait, __wstate); \ + schedule(); \ } else { \ cfs_duration_t interval = info->lwi_interval? \ min_t(cfs_duration_t, \ info->lwi_interval,__timeout):\ __timeout; \ - cfs_duration_t remaining = waitq_timedwait(&__wait,\ - __wstate, \ - interval); \ + cfs_duration_t remaining = schedule_timeout(interval);\ __timeout = cfs_time_sub(__timeout, \ cfs_time_sub(interval, remaining));\ if (__timeout == 0) { \ diff --git a/drivers/staging/lustre/lustre/include/lustre_linkea.h b/drivers/staging/lustre/lustre/include/lustre_linkea.h index 5790be913bf6..500ace30cfbf 100644 --- a/drivers/staging/lustre/lustre/include/lustre_linkea.h +++ b/drivers/staging/lustre/lustre/include/lustre_linkea.h @@ -33,7 +33,7 @@ struct linkea_data { */ struct lu_buf *ld_buf; /** - * The matched header, entry and its lenght in the EA + * The matched header, entry and its length in the EA */ struct link_ea_header *ld_leh; struct link_ea_entry *ld_lee; diff --git a/drivers/staging/lustre/lustre/include/lustre_mdc.h b/drivers/staging/lustre/lustre/include/lustre_mdc.h index c1e02702b931..468f36344a34 100644 --- a/drivers/staging/lustre/lustre/include/lustre_mdc.h +++ b/drivers/staging/lustre/lustre/include/lustre_mdc.h @@ -166,6 +166,17 @@ void it_clear_disposition(struct lookup_intent *it, int flag); void it_set_disposition(struct lookup_intent *it, int flag); int it_open_error(int phase, struct lookup_intent *it); +static inline bool cl_is_lov_delay_create(unsigned int flags) +{ + return (flags & O_LOV_DELAY_CREATE) == O_LOV_DELAY_CREATE; +} + +static inline void cl_lov_delay_create_clear(unsigned int *flags) +{ + if ((*flags & O_LOV_DELAY_CREATE) == O_LOV_DELAY_CREATE) + *flags &= ~O_LOV_DELAY_CREATE; +} + /** @} mdc */ #endif diff --git a/drivers/staging/lustre/lustre/include/lustre_net.h b/drivers/staging/lustre/lustre/include/lustre_net.h index d8d088035428..745adbb74cfc 100644 --- a/drivers/staging/lustre/lustre/include/lustre_net.h +++ b/drivers/staging/lustre/lustre/include/lustre_net.h @@ -445,7 +445,7 @@ struct ptlrpc_reply_state { lnet_handle_md_t rs_md_h; atomic_t rs_refcount; - /** Context for the sevice thread */ + /** Context for the service thread */ struct ptlrpc_svc_ctx *rs_svc_ctx; /** Reply buffer (actually sent to the client), encoded if needed */ struct lustre_msg *rs_repbuf; /* wrapper */ @@ -454,9 +454,9 @@ struct ptlrpc_reply_state { /** Size of the reply message */ int rs_repdata_len; /* wrapper msg length */ /** - * Actual reply message. Its content is encrupted (if needed) to + * Actual reply message. Its content is encrypted (if needed) to * produce reply buffer for actual sending. In simple case - * of no network encryption we jus set \a rs_repbuf to \a rs_msg + * of no network encryption we just set \a rs_repbuf to \a rs_msg */ struct lustre_msg *rs_msg; /* reply message */ @@ -497,7 +497,7 @@ struct ptlrpc_request_pool { spinlock_t prp_lock; /** list of ptlrpc_request structs */ struct list_head prp_req_list; - /** Maximum message size that would fit into a rquest from this pool */ + /** Maximum message size that would fit into a request from this pool */ int prp_rq_size; /** Function to allocate more requests for this pool */ void (*prp_populate)(struct ptlrpc_request_pool *, int); @@ -904,7 +904,7 @@ struct ptlrpc_nrs_pol_conf { */ struct module *nc_owner; /** - * Policy registration flags; a bitmast of \e nrs_policy_flags + * Policy registration flags; a bitmask of \e nrs_policy_flags */ unsigned nc_flags; }; @@ -1351,7 +1351,7 @@ struct nrs_orr_data { */ enum nrs_orr_supp od_supp; /** - * Round Robin quantum; the maxium number of RPCs that each request + * Round Robin quantum; the maximum number of RPCs that each request * batch for each object or OST can have in a scheduling round. */ __u16 od_quantum; @@ -1486,7 +1486,7 @@ struct ptlrpc_nrs_request { */ struct nrs_fifo_req fifo; /** - * CRR-N request defintion + * CRR-N request definition */ struct nrs_crrn_req crr; /** ORR and TRR share the same request definition */ @@ -1550,7 +1550,7 @@ struct ptlrpc_request { * requests in time */ struct list_head rq_timed_list; - /** server-side history, used for debuging purposes. */ + /** server-side history, used for debugging purposes. */ struct list_head rq_history_list; /** server-side per-export list */ struct list_head rq_exp_list; @@ -1611,7 +1611,7 @@ struct ptlrpc_request { enum rq_phase rq_phase; /* one of RQ_PHASE_* */ enum rq_phase rq_next_phase; /* one of RQ_PHASE_* to be used next */ atomic_t rq_refcount;/* client-side refcount for SENT race, - server-side refcounf for multiple replies */ + server-side refcount for multiple replies */ /** Portal to which this request would be sent */ short rq_request_portal; /* XXX FIXME bug 249 */ @@ -1637,7 +1637,7 @@ struct ptlrpc_request { /** xid */ __u64 rq_xid; /** - * List item to for replay list. Not yet commited requests get linked + * List item to for replay list. Not yet committed requests get linked * there. * Also see \a rq_replay comment above. */ @@ -1952,7 +1952,7 @@ void _debug_req(struct ptlrpc_request *req, __attribute__ ((format (printf, 3, 4))); /** - * Helper that decides if we need to print request accordig to current debug + * Helper that decides if we need to print request according to current debug * level settings */ #define debug_req(msgdata, mask, cdls, req, fmt, a...) \ @@ -1966,7 +1966,7 @@ do { \ } while(0) /** - * This is the debug print function you need to use to print request sturucture + * This is the debug print function you need to use to print request structure * content into lustre debug log. * for most callers (level is a constant) this is resolved at compile time */ #define DEBUG_REQ(level, req, fmt, args...) \ @@ -2621,6 +2621,8 @@ int ptlrpc_register_rqbd(struct ptlrpc_request_buffer_desc *rqbd); * request queues, request management, etc. * @{ */ +void ptlrpc_request_committed(struct ptlrpc_request *req, int force); + void ptlrpc_init_client(int req_portal, int rep_portal, char *name, struct ptlrpc_client *); void ptlrpc_cleanup_client(struct obd_import *imp); diff --git a/drivers/staging/lustre/lustre/include/lustre_quota.h b/drivers/staging/lustre/lustre/include/lustre_quota.h index 71b5d97e0343..07cb7c310bcc 100644 --- a/drivers/staging/lustre/lustre/include/lustre_quota.h +++ b/drivers/staging/lustre/lustre/include/lustre_quota.h @@ -140,7 +140,7 @@ struct qsd_instance; * (i.e. when ->ldo_recovery_complete is called). This is used * to notify the qsd layer that quota should now be enforced * again via the qsd_op_begin/end functions. The last step of the - * reintegration prodecure (namely usage reconciliation) will be + * reintegration procedure (namely usage reconciliation) will be * completed during start. * * - qsd_fini(): is used to release a qsd_instance structure allocated with diff --git a/drivers/staging/lustre/lustre/include/lustre_sec.h b/drivers/staging/lustre/lustre/include/lustre_sec.h index 885247d28b3a..bf3ee3915c28 100644 --- a/drivers/staging/lustre/lustre/include/lustre_sec.h +++ b/drivers/staging/lustre/lustre/include/lustre_sec.h @@ -572,7 +572,7 @@ struct ptlrpc_sec_cops { /** * Called then the reference of \a ctx dropped to 0. The policy module * is supposed to destroy this context or whatever else according to - * its cache maintainance mechamism. + * its cache maintenance mechanism. * * \param sync if zero, we shouldn't wait for the context being * destroyed completely. @@ -1002,7 +1002,7 @@ struct ptlrpc_sec *sptlrpc_sec_get(struct ptlrpc_sec *sec); void sptlrpc_sec_put(struct ptlrpc_sec *sec); /* - * internal apis which only used by policy impelentation + * internal apis which only used by policy implementation */ int sptlrpc_get_next_secid(void); void sptlrpc_sec_destroy(struct ptlrpc_sec *sec); diff --git a/drivers/staging/lustre/lustre/include/md_object.h b/drivers/staging/lustre/lustre/include/md_object.h index 7b45b47b48f9..ef46b2c461a6 100644 --- a/drivers/staging/lustre/lustre/include/md_object.h +++ b/drivers/staging/lustre/lustre/include/md_object.h @@ -35,7 +35,7 @@ * * lustre/include/md_object.h * - * Extention of lu_object.h for metadata objects + * Extension of lu_object.h for metadata objects */ #ifndef _LUSTRE_MD_OBJECT_H diff --git a/drivers/staging/lustre/lustre/include/obd.h b/drivers/staging/lustre/lustre/include/obd.h index c3470ce62cff..72cf3fe4b0c9 100644 --- a/drivers/staging/lustre/lustre/include/obd.h +++ b/drivers/staging/lustre/lustre/include/obd.h @@ -158,7 +158,7 @@ struct obd_info { /* statfs data specific for every OSC, if needed at all. */ struct obd_statfs *oi_osfs; /* An update callback which is called to update some data on upper - * level. E.g. it is used for update lsm->lsm_oinfo at every recieved + * level. E.g. it is used for update lsm->lsm_oinfo at every received * request in osc level for enqueue requests. It is also possible to * update some caller data from LOV layer if needed. */ obd_enqueue_update_f oi_cb_up; @@ -1042,8 +1042,8 @@ static inline int it_to_lock_mode(struct lookup_intent *it) } struct md_op_data { - struct lu_fid op_fid1; /* operation fid1 (usualy parent) */ - struct lu_fid op_fid2; /* operation fid2 (usualy child) */ + struct lu_fid op_fid1; /* operation fid1 (usually parent) */ + struct lu_fid op_fid2; /* operation fid2 (usually child) */ struct lu_fid op_fid3; /* 2 extra fids to find conflicting */ struct lu_fid op_fid4; /* to the operation locks. */ mdsno_t op_mds; /* what mds server open will go to */ @@ -1323,7 +1323,8 @@ struct md_open_data { struct obd_client_handle *mod_och; struct ptlrpc_request *mod_open_req; struct ptlrpc_request *mod_close_req; - atomic_t mod_refcount; + atomic_t mod_refcount; + bool mod_is_create; }; struct lookup_intent; @@ -1392,7 +1393,7 @@ struct md_ops { int (*m_set_open_replay_data)(struct obd_export *, struct obd_client_handle *, - struct ptlrpc_request *); + struct lookup_intent *); int (*m_clear_open_replay_data)(struct obd_export *, struct obd_client_handle *); int (*m_set_lock_data)(struct obd_export *, __u64 *, void *, __u64 *); diff --git a/drivers/staging/lustre/lustre/include/obd_class.h b/drivers/staging/lustre/lustre/include/obd_class.h index 983718fe1e55..9d1f266e55e4 100644 --- a/drivers/staging/lustre/lustre/include/obd_class.h +++ b/drivers/staging/lustre/lustre/include/obd_class.h @@ -175,9 +175,13 @@ enum { CONFIG_T_CONFIG = 0, CONFIG_T_SPTLRPC = 1, CONFIG_T_RECOVER = 2, - CONFIG_T_MAX = 3 + CONFIG_T_PARAMS = 3, + CONFIG_T_MAX = 4 }; +#define PARAMS_FILENAME "params" +#define LCTL_UPCALL "lctl" + /* list of active configuration logs */ struct config_llog_data { struct ldlm_res_id cld_resid; @@ -185,7 +189,8 @@ struct config_llog_data { struct list_head cld_list_chain; atomic_t cld_refcount; struct config_llog_data *cld_sptlrpc;/* depended sptlrpc log */ - struct config_llog_data *cld_recover; /* imperative recover log */ + struct config_llog_data *cld_params; /* common parameters log */ + struct config_llog_data *cld_recover;/* imperative recover log */ struct obd_export *cld_mgcexp; struct mutex cld_lock; int cld_type; @@ -1626,7 +1631,7 @@ static inline int obd_health_check(const struct lu_env *env, { /* returns: 0 on healthy * >0 on unhealthy + reason code/flag - * however the only suppored reason == 1 right now + * however the only supported reason == 1 right now * We'll need to define some better reasons * or flags in the future. * <0 on error @@ -1996,11 +2001,11 @@ static inline int md_getxattr(struct obd_export *exp, static inline int md_set_open_replay_data(struct obd_export *exp, struct obd_client_handle *och, - struct ptlrpc_request *open_req) + struct lookup_intent *it) { EXP_CHECK_MD_OP(exp, set_open_replay_data); EXP_MD_COUNTER_INCREMENT(exp, set_open_replay_data); - return MDP(exp->exp_obd, set_open_replay_data)(exp, och, open_req); + return MDP(exp->exp_obd, set_open_replay_data)(exp, och, it); } static inline int md_clear_open_replay_data(struct obd_export *exp, diff --git a/drivers/staging/lustre/lustre/include/obd_support.h b/drivers/staging/lustre/lustre/include/obd_support.h index 977bc231df9d..5ec336968fc8 100644 --- a/drivers/staging/lustre/lustre/include/obd_support.h +++ b/drivers/staging/lustre/lustre/include/obd_support.h @@ -681,7 +681,7 @@ do { \ * * Be very careful when changing this value, especially when decreasing it, * since vmalloc in Linux doesn't perform well on multi-cores system, calling - * vmalloc in critical path would hurt peformance badly. See LU-66. + * vmalloc in critical path would hurt performance badly. See LU-66. */ #define OBD_ALLOC_BIG (4 * PAGE_CACHE_SIZE) diff --git a/drivers/staging/lustre/lustre/lclient/lcommon_cl.c b/drivers/staging/lustre/lustre/lclient/lcommon_cl.c index 94b164127e0c..6907a16dbbd1 100644 --- a/drivers/staging/lustre/lustre/lclient/lcommon_cl.c +++ b/drivers/staging/lustre/lustre/lclient/lcommon_cl.c @@ -1196,14 +1196,14 @@ static void cl_object_put_last(struct lu_env *env, struct cl_object *obj) bkt = lu_site_bkt_from_fid(site, &header->loh_fid); - init_waitqueue_entry_current(&waiter); + init_waitqueue_entry(&waiter, current); add_wait_queue(&bkt->lsb_marche_funebre, &waiter); while (1) { set_current_state(TASK_UNINTERRUPTIBLE); if (atomic_read(&header->loh_ref) == 1) break; - waitq_wait(&waiter, TASK_UNINTERRUPTIBLE); + schedule(); } set_current_state(TASK_RUNNING); diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_flock.c b/drivers/staging/lustre/lustre/ldlm/ldlm_flock.c index c9aae132f98a..986bf384bff7 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_flock.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_flock.c @@ -205,6 +205,26 @@ ldlm_flock_deadlock(struct ldlm_lock *req, struct ldlm_lock *bl_lock) return 0; } +static void ldlm_flock_cancel_on_deadlock(struct ldlm_lock *lock, + struct list_head *work_list) +{ + CDEBUG(D_INFO, "reprocess deadlock req=%p\n", lock); + + if ((exp_connect_flags(lock->l_export) & + OBD_CONNECT_FLOCK_DEAD) == 0) { + CERROR( + "deadlock found, but client doesn't support flock canceliation\n"); + } else { + LASSERT(lock->l_completion_ast); + LASSERT((lock->l_flags & LDLM_FL_AST_SENT) == 0); + lock->l_flags |= LDLM_FL_AST_SENT | LDLM_FL_CANCEL_ON_BLOCK | + LDLM_FL_FLOCK_DEADLOCK; + ldlm_flock_blocking_unlink(lock); + ldlm_resource_unlink_lock(lock); + ldlm_add_ast_work_item(lock, NULL, work_list); + } +} + /** * Process a granting attempt for flock lock. * Must be called under ns lock held. @@ -272,6 +292,7 @@ reprocess: } } } else { + int reprocess_failed = 0; lockmode_verify(mode); /* This loop determines if there are existing locks @@ -293,8 +314,15 @@ reprocess: if (!ldlm_flocks_overlap(lock, req)) continue; - if (!first_enq) - return LDLM_ITER_CONTINUE; + if (!first_enq) { + reprocess_failed = 1; + if (ldlm_flock_deadlock(req, lock)) { + ldlm_flock_cancel_on_deadlock(req, + work_list); + return LDLM_ITER_CONTINUE; + } + continue; + } if (*flags & LDLM_FL_BLOCK_NOWAIT) { ldlm_flock_destroy(req, mode, *flags); @@ -330,6 +358,8 @@ reprocess: *flags |= LDLM_FL_BLOCK_GRANTED; return LDLM_ITER_STOP; } + if (reprocess_failed) + return LDLM_ITER_CONTINUE; } if (*flags & LDLM_FL_TEST_LOCK) { @@ -646,7 +676,10 @@ granted: /* ldlm_lock_enqueue() has already placed lock on the granted list. */ list_del_init(&lock->l_res_link); - if (flags & LDLM_FL_TEST_LOCK) { + if (lock->l_flags & LDLM_FL_FLOCK_DEADLOCK) { + LDLM_DEBUG(lock, "client-side enqueue deadlock received"); + rc = -EDEADLK; + } else if (flags & LDLM_FL_TEST_LOCK) { /* fcntl(F_GETLK) request */ /* The old mode was saved in getlk->fl_type so that if the mode * in the lock changes we can decref the appropriate refcount.*/ @@ -672,7 +705,7 @@ granted: ldlm_process_flock_lock(lock, &noreproc, 1, &err, NULL); } unlock_res_and_lock(lock); - return 0; + return rc; } EXPORT_SYMBOL(ldlm_flock_completion_ast); diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_lock.c b/drivers/staging/lustre/lustre/ldlm/ldlm_lock.c index 692623beee12..0548aca45e8e 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_lock.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_lock.c @@ -145,8 +145,6 @@ char *ldlm_it2str(int it) return "getxattr"; case IT_LAYOUT: return "layout"; - case IT_SETXATTR: - return "setxattr"; default: CERROR("Unknown intent %d\n", it); return "UNKNOWN"; diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_lockd.c b/drivers/staging/lustre/lustre/ldlm/ldlm_lockd.c index 3ed020eb89c0..7e63cf355cd9 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_lockd.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_lockd.c @@ -192,8 +192,8 @@ static void ldlm_handle_cp_callback(struct ptlrpc_request *req, if (OBD_FAIL_CHECK(OBD_FAIL_LDLM_CANCEL_BL_CB_RACE)) { int to = cfs_time_seconds(1); while (to > 0) { - schedule_timeout_and_set_state( - TASK_INTERRUPTIBLE, to); + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(to); if (lock->l_granted_mode == lock->l_req_mode || lock->l_flags & LDLM_FL_DESTROYED) break; @@ -228,6 +228,7 @@ static void ldlm_handle_cp_callback(struct ptlrpc_request *req, lock_res_and_lock(lock); LASSERT(lock->l_lvb_data == NULL); + lock->l_lvb_type = LVB_T_LAYOUT; lock->l_lvb_data = lvb_data; lock->l_lvb_len = lvb_len; unlock_res_and_lock(lock); diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c index c0e54aead2ce..fcc7a99ce395 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c @@ -299,7 +299,7 @@ EXPORT_SYMBOL(ldlm_completion_ast); * A helper to build a blocking AST function * * Perform a common operation for blocking ASTs: - * defferred lock cancellation. + * deferred lock cancellation. * * \param lock the lock blocking or canceling AST was called on * \retval 0 diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_resource.c b/drivers/staging/lustre/lustre/ldlm/ldlm_resource.c index 5f89864cda14..2824d4a9a85b 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_resource.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_resource.c @@ -421,9 +421,9 @@ static unsigned ldlm_res_hop_fid_hash(struct cfs_hash *hs, } else { val = fid_oid(&fid); } - hash = cfs_hash_long(hash, hs->hs_bkt_bits); + hash = hash_long(hash, hs->hs_bkt_bits); /* give me another random factor */ - hash -= cfs_hash_long((unsigned long)hs, val % 11 + 3); + hash -= hash_long((unsigned long)hs, val % 11 + 3); hash <<= hs->hs_cur_bits - hs->hs_bkt_bits; hash |= ldlm_res_hop_hash(hs, key, CFS_HASH_NBKT(hs) - 1); diff --git a/drivers/staging/lustre/lustre/libcfs/debug.c b/drivers/staging/lustre/lustre/libcfs/debug.c index f30c84f195aa..1e4c5ad26157 100644 --- a/drivers/staging/lustre/lustre/libcfs/debug.c +++ b/drivers/staging/lustre/lustre/libcfs/debug.c @@ -368,7 +368,7 @@ void libcfs_debug_dumplog(void) /* we're being careful to ensure that the kernel thread is * able to set our state to running as it exits before we * get to schedule() */ - init_waitqueue_entry_current(&wait); + init_waitqueue_entry(&wait, current); set_current_state(TASK_INTERRUPTIBLE); add_wait_queue(&debug_ctlwq, &wait); @@ -379,7 +379,7 @@ void libcfs_debug_dumplog(void) printk(KERN_ERR "LustreError: cannot start log dump thread:" " %ld\n", PTR_ERR(dumper)); else - waitq_wait(&wait, TASK_INTERRUPTIBLE); + schedule(); /* be sure to teardown if cfs_create_thread() failed */ remove_wait_queue(&debug_ctlwq, &wait); diff --git a/drivers/staging/lustre/lustre/libcfs/fail.c b/drivers/staging/lustre/lustre/libcfs/fail.c index c54448d69008..ba43ff7f7900 100644 --- a/drivers/staging/lustre/lustre/libcfs/fail.c +++ b/drivers/staging/lustre/lustre/libcfs/fail.c @@ -127,8 +127,8 @@ int __cfs_fail_timeout_set(__u32 id, __u32 value, int ms, int set) if (ret) { CERROR("cfs_fail_timeout id %x sleeping for %dms\n", id, ms); - schedule_timeout_and_set_state(TASK_UNINTERRUPTIBLE, - cfs_time_seconds(ms) / 1000); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(ms) / 1000); set_current_state(TASK_RUNNING); CERROR("cfs_fail_timeout id %x awake\n", id); } diff --git a/drivers/staging/lustre/lustre/libcfs/kernel_user_comm.c b/drivers/staging/lustre/lustre/libcfs/kernel_user_comm.c index 7b2c31599327..b6ddc998f750 100644 --- a/drivers/staging/lustre/lustre/libcfs/kernel_user_comm.c +++ b/drivers/staging/lustre/lustre/libcfs/kernel_user_comm.c @@ -44,106 +44,6 @@ #include <linux/libcfs/libcfs.h> -#ifdef LUSTRE_UTILS -/* This is the userspace side. */ - -/** Start the userspace side of a KUC pipe. - * @param link Private descriptor for pipe/socket. - * @param groups KUC broadcast group to listen to - * (can be null for unicast to this pid) - */ -int libcfs_ukuc_start(lustre_kernelcomm *link, int group) -{ - int pfd[2]; - - if (pipe(pfd) < 0) - return -errno; - - memset(link, 0, sizeof(*link)); - link->lk_rfd = pfd[0]; - link->lk_wfd = pfd[1]; - link->lk_group = group; - link->lk_uid = getpid(); - return 0; -} - -int libcfs_ukuc_stop(lustre_kernelcomm *link) -{ - if (link->lk_wfd > 0) - close(link->lk_wfd); - return close(link->lk_rfd); -} - -#define lhsz sizeof(*kuch) - -/** Read a message from the link. - * Allocates memory, returns handle - * - * @param link Private descriptor for pipe/socket. - * @param buf Buffer to read into, must include size for kuc_hdr - * @param maxsize Maximum message size allowed - * @param transport Only listen to messages on this transport - * (and the generic transport) - */ -int libcfs_ukuc_msg_get(lustre_kernelcomm *link, char *buf, int maxsize, - int transport) -{ - struct kuc_hdr *kuch; - int rc = 0; - - memset(buf, 0, maxsize); - - CDEBUG(D_KUC, "Waiting for message from kernel on fd %d\n", - link->lk_rfd); - - while (1) { - /* Read header first to get message size */ - rc = read(link->lk_rfd, buf, lhsz); - if (rc <= 0) { - rc = -errno; - break; - } - kuch = (struct kuc_hdr *)buf; - - CDEBUG(D_KUC, "Received message mg=%x t=%d m=%d l=%d\n", - kuch->kuc_magic, kuch->kuc_transport, kuch->kuc_msgtype, - kuch->kuc_msglen); - - if (kuch->kuc_magic != KUC_MAGIC) { - CERROR("bad message magic %x != %x\n", - kuch->kuc_magic, KUC_MAGIC); - rc = -EPROTO; - break; - } - - if (kuch->kuc_msglen > maxsize) { - rc = -EMSGSIZE; - break; - } - - /* Read payload */ - rc = read(link->lk_rfd, buf + lhsz, kuch->kuc_msglen - lhsz); - if (rc < 0) { - rc = -errno; - break; - } - if (rc < (kuch->kuc_msglen - lhsz)) { - CERROR("short read: got %d of %d bytes\n", - rc, kuch->kuc_msglen); - rc = -EPROTO; - break; - } - - if (kuch->kuc_transport == transport || - kuch->kuc_transport == KUC_TRANSPORT_GENERIC) { - return 0; - } - /* Drop messages for other transports */ - } - return rc; -} - -#else /* LUSTRE_UTILS */ /* This is the kernel side (liblustre as well). */ /** @@ -338,5 +238,3 @@ int libcfs_kkuc_group_foreach(int group, libcfs_kkuc_cb_t cb_func, return rc; } EXPORT_SYMBOL(libcfs_kkuc_group_foreach); - -#endif /* LUSTRE_UTILS */ diff --git a/drivers/staging/lustre/lustre/libcfs/libcfs_string.c b/drivers/staging/lustre/lustre/libcfs/libcfs_string.c index 922debd0a412..ed0a6b531058 100644 --- a/drivers/staging/lustre/lustre/libcfs/libcfs_string.c +++ b/drivers/staging/lustre/lustre/libcfs/libcfs_string.c @@ -42,26 +42,6 @@ #include <linux/libcfs/libcfs.h> -/* non-0 = don't match */ -int cfs_strncasecmp(const char *s1, const char *s2, size_t n) -{ - if (s1 == NULL || s2 == NULL) - return 1; - - if (n == 0) - return 0; - - while (n-- != 0 && tolower(*s1) == tolower(*s2)) { - if (n == 0 || *s1 == '\0' || *s2 == '\0') - break; - s1++; - s2++; - } - - return tolower(*(unsigned char *)s1) - tolower(*(unsigned char *)s2); -} -EXPORT_SYMBOL(cfs_strncasecmp); - /* Convert a text string to a bitmask */ int cfs_str2mask(const char *str, const char *(*bit2str)(int bit), int *oldmask, int minmask, int allmask) @@ -101,7 +81,7 @@ int cfs_str2mask(const char *str, const char *(*bit2str)(int bit), debugstr = bit2str(i); if (debugstr != NULL && strlen(debugstr) == len && - cfs_strncasecmp(str, debugstr, len) == 0) { + strncasecmp(str, debugstr, len) == 0) { if (op == '-') newmask &= ~(1 << i); else @@ -111,7 +91,7 @@ int cfs_str2mask(const char *str, const char *(*bit2str)(int bit), } } if (!found && len == 3 && - (cfs_strncasecmp(str, "ALL", len) == 0)) { + (strncasecmp(str, "ALL", len) == 0)) { if (op == '-') newmask = minmask; else @@ -129,7 +109,6 @@ int cfs_str2mask(const char *str, const char *(*bit2str)(int bit), *oldmask = newmask; return 0; } -EXPORT_SYMBOL(cfs_str2mask); /* get the first string out of @str */ char *cfs_firststr(char *str, size_t size) @@ -164,12 +143,12 @@ cfs_trimwhite(char *str) { char *end; - while (cfs_iswhite(*str)) + while (isspace(*str)) str++; end = str + strlen(str); while (end > str) { - if (!cfs_iswhite(end[-1])) + if (!isspace(end[-1])) break; end--; } @@ -199,7 +178,7 @@ cfs_gettok(struct cfs_lstr *next, char delim, struct cfs_lstr *res) /* skip leading white spaces */ while (next->ls_len) { - if (!cfs_iswhite(*next->ls_str)) + if (!isspace(*next->ls_str)) break; next->ls_str++; next->ls_len--; @@ -226,14 +205,13 @@ cfs_gettok(struct cfs_lstr *next, char delim, struct cfs_lstr *res) /* skip ending whitespaces */ while (--end != res->ls_str) { - if (!cfs_iswhite(*end)) + if (!isspace(*end)) break; } res->ls_len = end - res->ls_str + 1; return 1; } -EXPORT_SYMBOL(cfs_gettok); /** * Converts string to integer. @@ -256,13 +234,12 @@ cfs_str2num_check(char *str, int nob, unsigned *num, return 0; for (; endp < str + nob; endp++) { - if (!cfs_iswhite(*endp)) + if (!isspace(*endp)) return 0; } return (*num >= min && *num <= max); } -EXPORT_SYMBOL(cfs_str2num_check); /** * Parses \<range_expr\> token of the syntax. If \a bracketed is false, @@ -277,7 +254,7 @@ EXPORT_SYMBOL(cfs_str2num_check); * \retval 0 will be returned if it can be parsed, otherwise -EINVAL or * -ENOMEM will be returned. */ -int +static int cfs_range_expr_parse(struct cfs_lstr *src, unsigned min, unsigned max, int bracketed, struct cfs_range_expr **expr) { @@ -340,7 +317,6 @@ cfs_range_expr_parse(struct cfs_lstr *src, unsigned min, unsigned max, LIBCFS_FREE(re, sizeof(*re)); return -EINVAL; } -EXPORT_SYMBOL(cfs_range_expr_parse); /** * Matches value (\a value) against ranges expression list \a expr_list. @@ -361,7 +337,6 @@ cfs_expr_list_match(__u32 value, struct cfs_expr_list *expr_list) return 0; } -EXPORT_SYMBOL(cfs_expr_list_match); /** * Convert express list (\a expr_list) to an array of all matched values @@ -432,18 +407,6 @@ cfs_expr_list_free(struct cfs_expr_list *expr_list) } EXPORT_SYMBOL(cfs_expr_list_free); -void -cfs_expr_list_print(struct cfs_expr_list *expr_list) -{ - struct cfs_range_expr *expr; - - list_for_each_entry(expr, &expr_list->el_exprs, re_link) { - CDEBUG(D_WARNING, "%d-%d/%d\n", - expr->re_lo, expr->re_hi, expr->re_stride); - } -} -EXPORT_SYMBOL(cfs_expr_list_print); - /** * Parses \<cfs_expr_list\> token of the syntax. * @@ -526,7 +489,6 @@ cfs_expr_list_free_list(struct list_head *list) cfs_expr_list_free(el); } } -EXPORT_SYMBOL(cfs_expr_list_free_list); int cfs_ip_addr_parse(char *str, int len, struct list_head *list) diff --git a/drivers/staging/lustre/lustre/libcfs/linux/linux-cpu.c b/drivers/staging/lustre/lustre/libcfs/linux/linux-cpu.c index 58bb256ee047..77b1ef64ecc0 100644 --- a/drivers/staging/lustre/lustre/libcfs/linux/linux-cpu.c +++ b/drivers/staging/lustre/lustre/libcfs/linux/linux-cpu.c @@ -952,6 +952,7 @@ static int cfs_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; + bool warn; switch (action) { case CPU_DEAD: @@ -962,9 +963,21 @@ cfs_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) cpt_data.cpt_version++; spin_unlock(&cpt_data.cpt_lock); default: - CWARN("Lustre: can't support CPU hotplug well now, " - "performance and stability could be impacted" - "[CPU %u notify: %lx]\n", cpu, action); + if (action != CPU_DEAD && action != CPU_DEAD_FROZEN) { + CDEBUG(D_INFO, "CPU changed [cpu %u action %lx]\n", + cpu, action); + break; + } + + down(&cpt_data.cpt_mutex); + /* if all HTs in a core are offline, it may break affinity */ + cfs_cpu_ht_siblings(cpu, cpt_data.cpt_cpumask); + warn = any_online_cpu(*cpt_data.cpt_cpumask) >= nr_cpu_ids; + up(&cpt_data.cpt_mutex); + CDEBUG(warn ? D_WARNING : D_INFO, + "Lustre: can't support CPU plug-out well now, " + "performance and stability could be impacted " + "[CPU %u action: %lx]\n", cpu, action); } return NOTIFY_OK; diff --git a/drivers/staging/lustre/lustre/libcfs/linux/linux-curproc.c b/drivers/staging/lustre/lustre/libcfs/linux/linux-curproc.c index a2ef64c3403d..e74c3e28a972 100644 --- a/drivers/staging/lustre/lustre/libcfs/linux/linux-curproc.c +++ b/drivers/staging/lustre/lustre/libcfs/linux/linux-curproc.c @@ -55,25 +55,13 @@ * for Linux kernel. */ -int cfs_curproc_groups_nr(void) -{ - int nr; - - task_lock(current); - nr = current_cred()->group_info->ngroups; - task_unlock(current); - return nr; -} - -/* Currently all the CFS_CAP_* defines match CAP_* ones. */ -#define cfs_cap_pack(cap) (cap) -#define cfs_cap_unpack(cap) (cap) - void cfs_cap_raise(cfs_cap_t cap) { struct cred *cred; - if ((cred = prepare_creds())) { - cap_raise(cred->cap_effective, cfs_cap_unpack(cap)); + + cred = prepare_creds(); + if (cred) { + cap_raise(cred->cap_effective, cap); commit_creds(cred); } } @@ -81,42 +69,28 @@ void cfs_cap_raise(cfs_cap_t cap) void cfs_cap_lower(cfs_cap_t cap) { struct cred *cred; - if ((cred = prepare_creds())) { - cap_lower(cred->cap_effective, cfs_cap_unpack(cap)); + + cred = prepare_creds(); + if (cred) { + cap_lower(cred->cap_effective, cap); commit_creds(cred); } } int cfs_cap_raised(cfs_cap_t cap) { - return cap_raised(current_cap(), cfs_cap_unpack(cap)); + return cap_raised(current_cap(), cap); } void cfs_kernel_cap_pack(kernel_cap_t kcap, cfs_cap_t *cap) { -#if defined (_LINUX_CAPABILITY_VERSION) && _LINUX_CAPABILITY_VERSION == 0x19980330 - *cap = cfs_cap_pack(kcap); -#elif defined (_LINUX_CAPABILITY_VERSION) && _LINUX_CAPABILITY_VERSION == 0x20071026 - *cap = cfs_cap_pack(kcap[0]); -#elif defined(_KERNEL_CAPABILITY_VERSION) && _KERNEL_CAPABILITY_VERSION == 0x20080522 /* XXX lost high byte */ - *cap = cfs_cap_pack(kcap.cap[0]); -#else - #error "need correct _KERNEL_CAPABILITY_VERSION " -#endif + *cap = kcap.cap[0]; } void cfs_kernel_cap_unpack(kernel_cap_t *kcap, cfs_cap_t cap) { -#if defined (_LINUX_CAPABILITY_VERSION) && _LINUX_CAPABILITY_VERSION == 0x19980330 - *kcap = cfs_cap_unpack(cap); -#elif defined (_LINUX_CAPABILITY_VERSION) && _LINUX_CAPABILITY_VERSION == 0x20071026 - (*kcap)[0] = cfs_cap_unpack(cap); -#elif defined(_KERNEL_CAPABILITY_VERSION) && _KERNEL_CAPABILITY_VERSION == 0x20080522 - kcap->cap[0] = cfs_cap_unpack(cap); -#else - #error "need correct _KERNEL_CAPABILITY_VERSION " -#endif + kcap->cap[0] = cap; } cfs_cap_t cfs_curproc_cap_pack(void) @@ -126,20 +100,6 @@ cfs_cap_t cfs_curproc_cap_pack(void) return cap; } -void cfs_curproc_cap_unpack(cfs_cap_t cap) -{ - struct cred *cred; - if ((cred = prepare_creds())) { - cfs_kernel_cap_unpack(&cred->cap_effective, cap); - commit_creds(cred); - } -} - -int cfs_capable(cfs_cap_t cap) -{ - return capable(cfs_cap_unpack(cap)); -} - static int cfs_access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write) { @@ -292,13 +252,10 @@ out: } EXPORT_SYMBOL(cfs_get_environ); -EXPORT_SYMBOL(cfs_curproc_groups_nr); EXPORT_SYMBOL(cfs_cap_raise); EXPORT_SYMBOL(cfs_cap_lower); EXPORT_SYMBOL(cfs_cap_raised); EXPORT_SYMBOL(cfs_curproc_cap_pack); -EXPORT_SYMBOL(cfs_curproc_cap_unpack); -EXPORT_SYMBOL(cfs_capable); /* * Local variables: diff --git a/drivers/staging/lustre/lustre/libcfs/linux/linux-module.c b/drivers/staging/lustre/lustre/libcfs/linux/linux-module.c index 55296a3591d5..e6eae0666f0d 100644 --- a/drivers/staging/lustre/lustre/libcfs/linux/linux-module.c +++ b/drivers/staging/lustre/lustre/libcfs/linux/linux-module.c @@ -150,12 +150,12 @@ static long libcfs_ioctl(struct file *file, /* Handle platform-dependent IOC requests */ switch (cmd) { case IOC_LIBCFS_PANIC: - if (!cfs_capable(CFS_CAP_SYS_BOOT)) + if (!capable(CFS_CAP_SYS_BOOT)) return (-EPERM); panic("debugctl-invoked panic"); return (0); case IOC_LIBCFS_MEMHOG: - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) return -EPERM; /* go thought */ } diff --git a/drivers/staging/lustre/lustre/libcfs/linux/linux-prim.c b/drivers/staging/lustre/lustre/libcfs/linux/linux-prim.c index c7bc7fcccb8e..9a40d1415a65 100644 --- a/drivers/staging/lustre/lustre/libcfs/linux/linux-prim.c +++ b/drivers/staging/lustre/lustre/libcfs/linux/linux-prim.c @@ -46,13 +46,6 @@ #include <asm/kgdb.h> #endif -void -init_waitqueue_entry_current(wait_queue_t *link) -{ - init_waitqueue_entry(link, current); -} -EXPORT_SYMBOL(init_waitqueue_entry_current); - /** * wait_queue_t of Linux (version < 2.6.34) is a FIFO list for exclusively * waiting threads, which is not always desirable because all threads will @@ -77,37 +70,6 @@ add_wait_queue_exclusive_head(wait_queue_head_t *waitq, wait_queue_t *link) } EXPORT_SYMBOL(add_wait_queue_exclusive_head); -void -waitq_wait(wait_queue_t *link, long state) -{ - schedule(); -} -EXPORT_SYMBOL(waitq_wait); - -int64_t -waitq_timedwait(wait_queue_t *link, long state, int64_t timeout) -{ - return schedule_timeout(timeout); -} -EXPORT_SYMBOL(waitq_timedwait); - -void -schedule_timeout_and_set_state(long state, int64_t timeout) -{ - set_current_state(state); - schedule_timeout(timeout); -} -EXPORT_SYMBOL(schedule_timeout_and_set_state); - -/* deschedule for a bit... */ -void -cfs_pause(cfs_duration_t ticks) -{ - set_current_state(TASK_UNINTERRUPTIBLE); - schedule_timeout(ticks); -} -EXPORT_SYMBOL(cfs_pause); - void cfs_init_timer(struct timer_list *t) { init_timer(t); diff --git a/drivers/staging/lustre/lustre/libcfs/nidstrings.c b/drivers/staging/lustre/lustre/libcfs/nidstrings.c index 732ae5540bf4..cfb274fc9cbd 100644 --- a/drivers/staging/lustre/lustre/libcfs/nidstrings.c +++ b/drivers/staging/lustre/lustre/libcfs/nidstrings.c @@ -65,23 +65,20 @@ void libcfs_init_nidstrings(void) spin_lock_init(&libcfs_nidstring_lock); } -# define NIDSTR_LOCK(f) spin_lock_irqsave(&libcfs_nidstring_lock, f) -# define NIDSTR_UNLOCK(f) spin_unlock_irqrestore(&libcfs_nidstring_lock, f) - static char * libcfs_next_nidstring(void) { char *str; unsigned long flags; - NIDSTR_LOCK(flags); + spin_lock_irqsave(&libcfs_nidstring_lock, flags); str = libcfs_nidstrings[libcfs_nidstring_idx++]; if (libcfs_nidstring_idx == sizeof(libcfs_nidstrings)/sizeof(libcfs_nidstrings[0])) libcfs_nidstring_idx = 0; - NIDSTR_UNLOCK(flags); + spin_unlock_irqrestore(&libcfs_nidstring_lock, flags); return str; } diff --git a/drivers/staging/lustre/lustre/libcfs/tracefile.c b/drivers/staging/lustre/lustre/libcfs/tracefile.c index 54290ce6bb43..c8599eefeb76 100644 --- a/drivers/staging/lustre/lustre/libcfs/tracefile.c +++ b/drivers/staging/lustre/lustre/libcfs/tracefile.c @@ -1076,11 +1076,10 @@ end_loop: break; } } - init_waitqueue_entry_current(&__wait); + init_waitqueue_entry(&__wait, current); add_wait_queue(&tctl->tctl_waitq, &__wait); set_current_state(TASK_INTERRUPTIBLE); - waitq_timedwait(&__wait, TASK_INTERRUPTIBLE, - cfs_time_seconds(1)); + schedule_timeout(cfs_time_seconds(1)); remove_wait_queue(&tctl->tctl_waitq, &__wait); } complete(&tctl->tctl_stop); diff --git a/drivers/staging/lustre/lustre/libcfs/upcall_cache.c b/drivers/staging/lustre/lustre/libcfs/upcall_cache.c index 245b46f0dd96..8085e32e5e7a 100644 --- a/drivers/staging/lustre/lustre/libcfs/upcall_cache.c +++ b/drivers/staging/lustre/lustre/libcfs/upcall_cache.c @@ -218,13 +218,12 @@ find_again: MAX_SCHEDULE_TIMEOUT; long left; - init_waitqueue_entry_current(&wait); + init_waitqueue_entry(&wait, current); add_wait_queue(&entry->ue_waitq, &wait); set_current_state(TASK_INTERRUPTIBLE); spin_unlock(&cache->uc_lock); - left = waitq_timedwait(&wait, TASK_INTERRUPTIBLE, - expiry); + left = schedule_timeout(expiry); spin_lock(&cache->uc_lock); remove_wait_queue(&entry->ue_waitq, &wait); diff --git a/drivers/staging/lustre/lustre/libcfs/workitem.c b/drivers/staging/lustre/lustre/libcfs/workitem.c index 1a55c81892e0..ba16fd5db704 100644 --- a/drivers/staging/lustre/lustre/libcfs/workitem.c +++ b/drivers/staging/lustre/lustre/libcfs/workitem.c @@ -334,7 +334,8 @@ cfs_wi_sched_destroy(struct cfs_wi_sched *sched) sched->ws_nthreads, sched->ws_name); spin_unlock(&cfs_wi_data.wi_glock); - cfs_pause(cfs_time_seconds(1) / 20); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1) / 20); spin_lock(&cfs_wi_data.wi_glock); } @@ -389,11 +390,11 @@ cfs_wi_sched_create(char *name, struct cfs_cpt_table *cptab, spin_unlock(&cfs_wi_data.wi_glock); if (sched->ws_cptab != NULL && sched->ws_cpt >= 0) { - snprintf(name, sizeof(name), "%s_%02d_%02d", + snprintf(name, sizeof(name), "%s_%02d_%02u", sched->ws_name, sched->ws_cpt, sched->ws_nthreads); } else { - snprintf(name, sizeof(name), "%s_%02d", + snprintf(name, sizeof(name), "%s_%02u", sched->ws_name, sched->ws_nthreads); } @@ -459,7 +460,8 @@ cfs_wi_shutdown (void) while (sched->ws_nthreads != 0) { spin_unlock(&cfs_wi_data.wi_glock); - cfs_pause(cfs_time_seconds(1) / 20); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1) / 20); spin_lock(&cfs_wi_data.wi_glock); } spin_unlock(&cfs_wi_data.wi_glock); diff --git a/drivers/staging/lustre/lustre/llite/dcache.c b/drivers/staging/lustre/lustre/llite/dcache.c index cbd663ed030c..8b5508086174 100644 --- a/drivers/staging/lustre/lustre/llite/dcache.c +++ b/drivers/staging/lustre/lustre/llite/dcache.c @@ -160,7 +160,7 @@ static int ll_ddelete(const struct dentry *de) /* kernel >= 2.6.38 last refcount is decreased after this function. */ LASSERT(d_count(de) == 1); - /* Disable this piece of code temproarily because this is called + /* Disable this piece of code temporarily because this is called * inside dcache_lock so it's not appropriate to do lots of work * here. ATTENTION: Before this piece of code enabling, LU-2487 must be * resolved. */ @@ -176,7 +176,7 @@ static int ll_ddelete(const struct dentry *de) return 0; } -static int ll_set_dd(struct dentry *de) +int ll_d_init(struct dentry *de) { LASSERT(de != NULL); @@ -190,40 +190,22 @@ static int ll_set_dd(struct dentry *de) OBD_ALLOC_PTR(lld); if (likely(lld != NULL)) { spin_lock(&de->d_lock); - if (likely(de->d_fsdata == NULL)) + if (likely(de->d_fsdata == NULL)) { de->d_fsdata = lld; - else + __d_lustre_invalidate(de); + } else { OBD_FREE_PTR(lld); + } spin_unlock(&de->d_lock); } else { return -ENOMEM; } } + LASSERT(de->d_op == &ll_d_ops); return 0; } -int ll_dops_init(struct dentry *de, int block, int init_sa) -{ - struct ll_dentry_data *lld = ll_d2d(de); - int rc = 0; - - if (lld == NULL && block != 0) { - rc = ll_set_dd(de); - if (rc) - return rc; - - lld = ll_d2d(de); - } - - if (lld != NULL && init_sa != 0) - lld->lld_sa_generation = 0; - - /* kernel >= 2.6.38 d_op is set in d_alloc() */ - LASSERT(de->d_op == &ll_d_ops); - return rc; -} - void ll_intent_drop_lock(struct lookup_intent *it) { if (it->it_op && it->d.lustre.it_lock_mode) { @@ -259,9 +241,6 @@ void ll_intent_release(struct lookup_intent *it) ptlrpc_req_finished(it->d.lustre.it_data); /* ll_file_open */ if (it_disposition(it, DISP_ENQ_CREATE_REF)) /* create rec */ ptlrpc_req_finished(it->d.lustre.it_data); - if (it_disposition(it, DISP_ENQ_COMPLETE)) /* saved req from revalidate - * to lookup */ - ptlrpc_req_finished(it->d.lustre.it_data); it->d.lustre.it_disposition = 0; it->d.lustre.it_data = NULL; @@ -346,268 +325,32 @@ void ll_frob_intent(struct lookup_intent **itp, struct lookup_intent *deft) } -int ll_revalidate_it(struct dentry *de, int lookup_flags, - struct lookup_intent *it) +static int ll_revalidate_dentry(struct dentry *dentry, + unsigned int lookup_flags) { - struct md_op_data *op_data; - struct ptlrpc_request *req = NULL; - struct lookup_intent lookup_it = { .it_op = IT_LOOKUP }; - struct obd_export *exp; - struct inode *parent = de->d_parent->d_inode; - int rc; - - CDEBUG(D_VFSTRACE, "VFS Op:name=%s,intent=%s\n", de->d_name.name, - LL_IT2STR(it)); + struct inode *dir = dentry->d_parent->d_inode; - if (de->d_inode == NULL) { - __u64 ibits; - - /* We can only use negative dentries if this is stat or lookup, - for opens and stuff we do need to query server. */ - /* If there is IT_CREAT in intent op set, then we must throw - away this negative dentry and actually do the request to - kernel to create whatever needs to be created (if possible)*/ - if (it && (it->it_op & IT_CREAT)) - return 0; - - if (d_lustre_invalid(de)) - return 0; - - ibits = MDS_INODELOCK_UPDATE; - rc = ll_have_md_lock(parent, &ibits, LCK_MINMODE); - GOTO(out_sa, rc); - } - - /* Never execute intents for mount points. - * Attributes will be fixed up in ll_inode_revalidate_it */ - if (d_mountpoint(de)) - GOTO(out_sa, rc = 1); - - /* need to get attributes in case root got changed from other client */ - if (de == de->d_sb->s_root) { - rc = __ll_inode_revalidate_it(de, it, MDS_INODELOCK_LOOKUP); - if (rc == 0) - rc = 1; - GOTO(out_sa, rc); - } - - exp = ll_i2mdexp(de->d_inode); - - OBD_FAIL_TIMEOUT(OBD_FAIL_MDC_REVALIDATE_PAUSE, 5); - ll_frob_intent(&it, &lookup_it); - LASSERT(it); + /* + * if open&create is set, talk to MDS to make sure file is created if + * necessary, because we can't do this in ->open() later since that's + * called on an inode. return 0 here to let lookup to handle this. + */ + if ((lookup_flags & (LOOKUP_OPEN | LOOKUP_CREATE)) == + (LOOKUP_OPEN | LOOKUP_CREATE)) + return 0; - if (it->it_op == IT_LOOKUP && !d_lustre_invalid(de)) + if (lookup_flags & (LOOKUP_PARENT | LOOKUP_OPEN | LOOKUP_CREATE)) return 1; - if (it->it_op == IT_OPEN) { - struct inode *inode = de->d_inode; - struct ll_inode_info *lli = ll_i2info(inode); - struct obd_client_handle **och_p; - __u64 ibits; - - /* - * We used to check for MDS_INODELOCK_OPEN here, but in fact - * just having LOOKUP lock is enough to justify inode is the - * same. And if inode is the same and we have suitable - * openhandle, then there is no point in doing another OPEN RPC - * just to throw away newly received openhandle. There are no - * security implications too, if file owner or access mode is - * change, LOOKUP lock is revoked. - */ - - - if (it->it_flags & FMODE_WRITE) - och_p = &lli->lli_mds_write_och; - else if (it->it_flags & FMODE_EXEC) - och_p = &lli->lli_mds_exec_och; - else - och_p = &lli->lli_mds_read_och; - - /* Check for the proper lock. */ - ibits = MDS_INODELOCK_LOOKUP; - if (!ll_have_md_lock(inode, &ibits, LCK_MINMODE)) - goto do_lock; - mutex_lock(&lli->lli_och_mutex); - if (*och_p) { /* Everything is open already, do nothing */ - /* Originally it was idea to do not let them steal our - * open handle from under us by (*och_usecount)++ here. - * But in case we have the handle, but we cannot use it - * due to later checks (e.g. O_CREAT|O_EXCL flags set), - * nobody would decrement counter increased here. So we - * just hope the lock won't be invalidated in between. - * But if it would be, we'll reopen the open request to - * MDS later during file open path. - */ - mutex_unlock(&lli->lli_och_mutex); - return 1; - } - mutex_unlock(&lli->lli_och_mutex); - } - - if (it->it_op == IT_GETATTR) { - rc = ll_statahead_enter(parent, &de, 0); - if (rc == 1) - goto mark; - else if (rc != -EAGAIN && rc != 0) - GOTO(out, rc = 0); - } - -do_lock: - op_data = ll_prep_md_op_data(NULL, parent, de->d_inode, - de->d_name.name, de->d_name.len, - 0, LUSTRE_OPC_ANY, NULL); - if (IS_ERR(op_data)) - return PTR_ERR(op_data); - - if (!IS_POSIXACL(parent) || !exp_connect_umask(exp)) - it->it_create_mode &= ~current_umask(); - it->it_create_mode |= M_CHECK_STALE; - rc = md_intent_lock(exp, op_data, NULL, 0, it, - lookup_flags, - &req, ll_md_blocking_ast, 0); - it->it_create_mode &= ~M_CHECK_STALE; - ll_finish_md_op_data(op_data); - - /* If req is NULL, then md_intent_lock only tried to do a lock match; - * if all was well, it will return 1 if it found locks, 0 otherwise. */ - if (req == NULL && rc >= 0) { - if (!rc) - goto do_lookup; - GOTO(out, rc); - } - - if (rc < 0) { - if (rc != -ESTALE) { - CDEBUG(D_INFO, "ll_intent_lock: rc %d : it->it_status " - "%d\n", rc, it->d.lustre.it_status); - } - GOTO(out, rc = 0); - } - -revalidate_finish: - rc = ll_revalidate_it_finish(req, it, de); - if (rc != 0) { - if (rc != -ESTALE && rc != -ENOENT) - ll_intent_release(it); - GOTO(out, rc = 0); - } - - if ((it->it_op & IT_OPEN) && de->d_inode && - !S_ISREG(de->d_inode->i_mode) && - !S_ISDIR(de->d_inode->i_mode)) { - ll_release_openhandle(de, it); - } - rc = 1; - -out: - /* We do not free request as it may be reused during following lookup - * (see comment in mdc/mdc_locks.c::mdc_intent_lock()), request will - * be freed in ll_lookup_it or in ll_intent_release. But if - * request was not completed, we need to free it. (bug 5154, 9903) */ - if (req != NULL && !it_disposition(it, DISP_ENQ_COMPLETE)) - ptlrpc_req_finished(req); - if (rc == 0) { - /* mdt may grant layout lock for the newly created file, so - * release the lock to avoid leaking */ - ll_intent_drop_lock(it); - ll_invalidate_aliases(de->d_inode); - } else { - __u64 bits = 0; - __u64 matched_bits = 0; - - CDEBUG(D_DENTRY, "revalidated dentry %.*s (%p) parent %p " - "inode %p refc %d\n", de->d_name.len, - de->d_name.name, de, de->d_parent, de->d_inode, - d_count(de)); - - ll_set_lock_data(exp, de->d_inode, it, &bits); - - /* Note: We have to match both LOOKUP and PERM lock - * here to make sure the dentry is valid and no one - * changing the permission. - * But if the client connects < 2.4 server, which will - * only grant LOOKUP lock, so we can only Match LOOKUP - * lock for old server */ - if (exp_connect_flags(ll_i2mdexp(de->d_inode)) && - OBD_CONNECT_LVB_TYPE) - matched_bits = - MDS_INODELOCK_LOOKUP | MDS_INODELOCK_PERM; - else - matched_bits = MDS_INODELOCK_LOOKUP; - - if (((bits & matched_bits) == matched_bits) && - d_lustre_invalid(de)) - d_lustre_revalidate(de); - ll_lookup_finish_locks(it, de); - } - -mark: - if (it != NULL && it->it_op == IT_GETATTR && rc > 0) - ll_statahead_mark(parent, de); - return rc; - - /* - * This part is here to combat evil-evil race in real_lookup on 2.6 - * kernels. The race details are: We enter do_lookup() looking for some - * name, there is nothing in dcache for this name yet and d_lookup() - * returns NULL. We proceed to real_lookup(), and while we do this, - * another process does open on the same file we looking up (most simple - * reproducer), open succeeds and the dentry is added. Now back to - * us. In real_lookup() we do d_lookup() again and suddenly find the - * dentry, so we call d_revalidate on it, but there is no lock, so - * without this code we would return 0, but unpatched real_lookup just - * returns -ENOENT in such a case instead of retrying the lookup. Once - * this is dealt with in real_lookup(), all of this ugly mess can go and - * we can just check locks in ->d_revalidate without doing any RPCs - * ever. - */ -do_lookup: - if (it != &lookup_it) { - /* MDS_INODELOCK_UPDATE needed for IT_GETATTR case. */ - if (it->it_op == IT_GETATTR) - lookup_it.it_op = IT_GETATTR; - ll_lookup_finish_locks(it, de); - it = &lookup_it; - } + if (d_need_statahead(dir, dentry) <= 0) + return 1; - /* Do real lookup here. */ - op_data = ll_prep_md_op_data(NULL, parent, NULL, de->d_name.name, - de->d_name.len, 0, (it->it_op & IT_CREAT ? - LUSTRE_OPC_CREATE : - LUSTRE_OPC_ANY), NULL); - if (IS_ERR(op_data)) - return PTR_ERR(op_data); - - rc = md_intent_lock(exp, op_data, NULL, 0, it, 0, &req, - ll_md_blocking_ast, 0); - if (rc >= 0) { - struct mdt_body *mdt_body; - struct lu_fid fid = {.f_seq = 0, .f_oid = 0, .f_ver = 0}; - mdt_body = req_capsule_server_get(&req->rq_pill, &RMF_MDT_BODY); - - if (de->d_inode) - fid = *ll_inode2fid(de->d_inode); - - /* see if we got same inode, if not - return error */ - if (lu_fid_eq(&fid, &mdt_body->fid1)) { - ll_finish_md_op_data(op_data); - op_data = NULL; - goto revalidate_finish; - } - ll_intent_release(it); - } - ll_finish_md_op_data(op_data); - GOTO(out, rc = 0); + if (lookup_flags & LOOKUP_RCU) + return -ECHILD; -out_sa: - /* - * For rc == 1 case, should not return directly to prevent losing - * statahead windows; for rc == 0 case, the "lookup" will be done later. - */ - if (it != NULL && it->it_op == IT_GETATTR && rc == 1) - ll_statahead_enter(parent, &de, 1); - goto mark; + do_statahead_enter(dir, &dentry, dentry->d_inode == NULL); + ll_statahead_mark(dir, dentry); + return 1; } /* @@ -615,24 +358,13 @@ out_sa: */ int ll_revalidate_nd(struct dentry *dentry, unsigned int flags) { - struct inode *parent = dentry->d_parent->d_inode; - int unplug = 0; + int rc; - CDEBUG(D_VFSTRACE, "VFS Op:name=%s,flags=%u\n", + CDEBUG(D_VFSTRACE, "VFS Op:name=%s, flags=%u\n", dentry->d_name.name, flags); - if (!(flags & (LOOKUP_PARENT|LOOKUP_OPEN|LOOKUP_CREATE)) && - ll_need_statahead(parent, dentry) > 0) { - if (flags & LOOKUP_RCU) - return -ECHILD; - - if (dentry->d_inode == NULL) - unplug = 1; - do_statahead_enter(parent, &dentry, unplug); - ll_statahead_mark(parent, dentry); - } - - return 1; + rc = ll_revalidate_dentry(dentry, flags); + return rc; } diff --git a/drivers/staging/lustre/lustre/llite/dir.c b/drivers/staging/lustre/lustre/llite/dir.c index 22d0acc95bc5..7fbc18e3e654 100644 --- a/drivers/staging/lustre/lustre/llite/dir.c +++ b/drivers/staging/lustre/lustre/llite/dir.c @@ -362,7 +362,7 @@ struct page *ll_get_dir_page(struct inode *dir, __u64 hash, struct ptlrpc_request *request; struct md_op_data *op_data; - op_data = ll_prep_md_op_data(NULL, dir, NULL, NULL, 0, 0, + op_data = ll_prep_md_op_data(NULL, dir, dir, NULL, 0, 0, LUSTRE_OPC_ANY, NULL); if (IS_ERR(op_data)) return (void *)op_data; @@ -1048,20 +1048,25 @@ progress: } -static int copy_and_ioctl(int cmd, struct obd_export *exp, void *data, int len) +static int copy_and_ioctl(int cmd, struct obd_export *exp, + const void __user *data, size_t size) { - void *ptr; + void *copy; int rc; - OBD_ALLOC(ptr, len); - if (ptr == NULL) + OBD_ALLOC(copy, size); + if (copy == NULL) return -ENOMEM; - if (copy_from_user(ptr, data, len)) { - OBD_FREE(ptr, len); - return -EFAULT; + + if (copy_from_user(copy, data, size)) { + rc = -EFAULT; + goto out; } - rc = obd_iocontrol(cmd, exp, len, data, NULL); - OBD_FREE(ptr, len); + + rc = obd_iocontrol(cmd, exp, size, copy, NULL); +out: + OBD_FREE(copy, size); + return rc; } @@ -1080,16 +1085,16 @@ static int quotactl_ioctl(struct ll_sb_info *sbi, struct if_quotactl *qctl) case Q_QUOTAOFF: case Q_SETQUOTA: case Q_SETINFO: - if (!cfs_capable(CFS_CAP_SYS_ADMIN) || + if (!capable(CFS_CAP_SYS_ADMIN) || sbi->ll_flags & LL_SBI_RMT_CLIENT) return -EPERM; break; case Q_GETQUOTA: if (((type == USRQUOTA && - uid_eq(current_euid(), make_kuid(&init_user_ns, id))) || + !uid_eq(current_euid(), make_kuid(&init_user_ns, id))) || (type == GRPQUOTA && !in_egroup_p(make_kgid(&init_user_ns, id)))) && - (!cfs_capable(CFS_CAP_SYS_ADMIN) || + (!capable(CFS_CAP_SYS_ADMIN) || sbi->ll_flags & LL_SBI_RMT_CLIENT)) return -EPERM; break; @@ -1395,7 +1400,7 @@ lmv_out_free: if (tmp == NULL) GOTO(free_lmv, rc = -ENOMEM); - memcpy(tmp, &lum, sizeof(lum)); + *tmp = lum; tmp->lum_type = LMV_STRIPE_TYPE; tmp->lum_stripe_count = 1; mdtindex = ll_get_mdt_idx(inode); @@ -1597,7 +1602,7 @@ out_rmdir: struct obd_quotactl *oqctl; int error = 0; - if (!cfs_capable(CFS_CAP_SYS_ADMIN) || + if (!capable(CFS_CAP_SYS_ADMIN) || sbi->ll_flags & LL_SBI_RMT_CLIENT) return -EPERM; @@ -1621,7 +1626,7 @@ out_rmdir: case OBD_IOC_POLL_QUOTACHECK: { struct if_quotacheck *check; - if (!cfs_capable(CFS_CAP_SYS_ADMIN) || + if (!capable(CFS_CAP_SYS_ADMIN) || sbi->ll_flags & LL_SBI_RMT_CLIENT) return -EPERM; diff --git a/drivers/staging/lustre/lustre/llite/file.c b/drivers/staging/lustre/lustre/llite/file.c index c12821aedc2f..8e844a6371e0 100644 --- a/drivers/staging/lustre/lustre/llite/file.c +++ b/drivers/staging/lustre/lustre/llite/file.c @@ -205,7 +205,7 @@ out: return rc; } -int ll_md_real_close(struct inode *inode, int flags) +int ll_md_real_close(struct inode *inode, fmode_t fmode) { struct ll_inode_info *lli = ll_i2info(inode); struct obd_client_handle **och_p; @@ -213,30 +213,33 @@ int ll_md_real_close(struct inode *inode, int flags) __u64 *och_usecount; int rc = 0; - if (flags & FMODE_WRITE) { + if (fmode & FMODE_WRITE) { och_p = &lli->lli_mds_write_och; och_usecount = &lli->lli_open_fd_write_count; - } else if (flags & FMODE_EXEC) { + } else if (fmode & FMODE_EXEC) { och_p = &lli->lli_mds_exec_och; och_usecount = &lli->lli_open_fd_exec_count; } else { - LASSERT(flags & FMODE_READ); + LASSERT(fmode & FMODE_READ); och_p = &lli->lli_mds_read_och; och_usecount = &lli->lli_open_fd_read_count; } mutex_lock(&lli->lli_och_mutex); - if (*och_usecount) { /* There are still users of this handle, so - skip freeing it. */ + if (*och_usecount > 0) { + /* There are still users of this handle, so skip + * freeing it. */ mutex_unlock(&lli->lli_och_mutex); return 0; } + och=*och_p; *och_p = NULL; mutex_unlock(&lli->lli_och_mutex); - if (och) { /* There might be a race and somebody have freed this och - already */ + if (och != NULL) { + /* There might be a race and this handle may already + be closed. */ rc = ll_close_inode_openhandle(ll_i2sbi(inode)->ll_md_exp, inode, och, NULL); } @@ -443,8 +446,7 @@ static int ll_intent_file_open(struct file *file, void *lmm, itp, NULL); out: - ptlrpc_req_finished(itp->d.lustre.it_data); - it_clear_disposition(itp, DISP_ENQ_COMPLETE); + ptlrpc_req_finished(req); ll_intent_drop_lock(itp); return rc; @@ -477,7 +479,7 @@ static int ll_och_fill(struct obd_export *md_exp, struct lookup_intent *it, och->och_magic = OBD_CLIENT_HANDLE_MAGIC; och->och_flags = it->it_flags; - return md_set_open_replay_data(md_exp, och, req); + return md_set_open_replay_data(md_exp, och, it); } int ll_local_open(struct file *file, struct lookup_intent *it, @@ -671,14 +673,13 @@ restart: ll_capa_open(inode); - if (!lli->lli_has_smd) { - if (file->f_flags & O_LOV_DELAY_CREATE || - !(file->f_mode & FMODE_WRITE)) { - CDEBUG(D_INODE, "object creation was delayed\n"); - GOTO(out_och_free, rc); - } + if (!lli->lli_has_smd && + (cl_is_lov_delay_create(file->f_flags) || + (file->f_mode & FMODE_WRITE) == 0)) { + CDEBUG(D_INODE, "object creation was delayed\n"); + GOTO(out_och_free, rc); } - file->f_flags &= ~O_LOV_DELAY_CREATE; + cl_lov_delay_create_clear(&file->f_flags); GOTO(out_och_free, rc); out_och_free: @@ -813,10 +814,7 @@ struct obd_client_handle *ll_lease_open(struct inode *inode, struct file *file, * doesn't deal with openhandle, so normal openhandle will be leaked. */ LDLM_FL_NO_LRU | LDLM_FL_EXCL); ll_finish_md_op_data(op_data); - if (req != NULL) { - ptlrpc_req_finished(req); - it_clear_disposition(&it, DISP_ENQ_COMPLETE); - } + ptlrpc_req_finished(req); if (rc < 0) GOTO(out_release_it, rc); @@ -1033,6 +1031,33 @@ int ll_glimpse_ioctl(struct ll_sb_info *sbi, struct lov_stripe_md *lsm, return rc; } +static bool file_is_noatime(const struct file *file) +{ + const struct vfsmount *mnt = file->f_path.mnt; + const struct inode *inode = file->f_path.dentry->d_inode; + + /* Adapted from file_accessed() and touch_atime().*/ + if (file->f_flags & O_NOATIME) + return true; + + if (inode->i_flags & S_NOATIME) + return true; + + if (IS_NOATIME(inode)) + return true; + + if (mnt->mnt_flags & (MNT_NOATIME | MNT_READONLY)) + return true; + + if ((mnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode)) + return true; + + if ((inode->i_sb->s_flags & MS_NODIRATIME) && S_ISDIR(inode->i_mode)) + return true; + + return false; +} + void ll_io_init(struct cl_io *io, const struct file *file, int write) { struct inode *inode = file->f_dentry->d_inode; @@ -1052,6 +1077,8 @@ void ll_io_init(struct cl_io *io, const struct file *file, int write) } else if (file->f_flags & O_APPEND) { io->ci_lockreq = CILR_MANDATORY; } + + io->ci_noatime = file_is_noatime(file); } static ssize_t @@ -1092,16 +1119,12 @@ restart: down_read(&lli->lli_trunc_sem); } break; - case IO_SENDFILE: - vio->u.sendfile.cui_actor = args->u.sendfile.via_actor; - vio->u.sendfile.cui_target = args->u.sendfile.via_target; - break; case IO_SPLICE: vio->u.splice.cui_pipe = args->u.splice.via_pipe; vio->u.splice.cui_flags = args->u.splice.via_flags; break; default: - CERROR("Unknow IO type - %u\n", vio->cui_io_subtype); + CERROR("Unknown IO type - %u\n", vio->cui_io_subtype); LBUG(); } result = cl_io_loop(env, io); @@ -1340,7 +1363,7 @@ static int ll_lov_recreate_obj(struct inode *inode, unsigned long arg) struct ll_recreate_obj ucreat; struct ost_id oi; - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) return -EPERM; if (copy_from_user(&ucreat, (struct ll_recreate_obj *)arg, @@ -1358,7 +1381,7 @@ static int ll_lov_recreate_fid(struct inode *inode, unsigned long arg) struct ost_id oi; obd_count ost_idx; - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) return -EPERM; if (copy_from_user(&fid, (struct lu_fid *)arg, sizeof(fid))) @@ -1381,23 +1404,25 @@ int ll_lov_setstripe_ea_info(struct inode *inode, struct file *file, ccc_inode_lsm_put(inode, lsm); CDEBUG(D_IOCTL, "stripe already exists for ino %lu\n", inode->i_ino); - return -EEXIST; + GOTO(out, rc = -EEXIST); } ll_inode_size_lock(inode); rc = ll_intent_file_open(file, lum, lum_size, &oit); if (rc) - GOTO(out, rc); + GOTO(out_unlock, rc); rc = oit.d.lustre.it_status; if (rc < 0) GOTO(out_req_free, rc); ll_release_openhandle(file->f_dentry, &oit); - out: +out_unlock: ll_inode_size_unlock(inode); ll_intent_release(&oit); ccc_inode_lsm_put(inode, lsm); +out: + cl_lov_delay_create_clear(&file->f_flags); return rc; out_req_free: ptlrpc_req_finished((struct ptlrpc_request *) oit.d.lustre.it_data); @@ -1497,7 +1522,7 @@ static int ll_lov_setea(struct inode *inode, struct file *file, sizeof(struct lov_user_ost_data); int rc; - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) return -EPERM; OBD_ALLOC_LARGE(lump, lum_size); @@ -1747,7 +1772,7 @@ int ll_fid2path(struct inode *inode, void *arg) struct getinfo_fid2path *gfout, *gfin; int outsize, rc; - if (!cfs_capable(CFS_CAP_DAC_READ_SEARCH) && + if (!capable(CFS_CAP_DAC_READ_SEARCH) && !(ll_i2sbi(inode)->ll_flags & LL_SBI_USER_FID2PATH)) return -EPERM; @@ -2005,7 +2030,7 @@ static int ll_swap_layouts(struct file *file1, struct file *file2, llss->ia2.ia_valid = ATTR_MTIME | ATTR_ATIME; } - /* ultimate check, before swaping the layouts we check if + /* ultimate check, before swapping the layouts we check if * dataversion has changed (if requested) */ if (llss->check_dv1) { rc = ll_data_version(llss->inode1, &dv, 0); @@ -2093,7 +2118,7 @@ static int ll_hsm_state_set(struct inode *inode, struct hsm_state_set *hss) /* Non-root users are forbidden to set or clear flags which are * NOT defined in HSM_USER_MASK. */ if (((hss->hss_setmask | hss->hss_clearmask) & ~HSM_USER_MASK) && - !cfs_capable(CFS_CAP_SYS_ADMIN)) + !capable(CFS_CAP_SYS_ADMIN)) return -EPERM; op_data = ll_prep_md_op_data(NULL, inode, NULL, NULL, 0, 0, @@ -2670,7 +2695,7 @@ int ll_file_flock(struct file *file, int cmd, struct file_lock *file_lock) LASSERT((cmd == F_SETLKW) || (cmd == F_SETLK)); /* flocks are whole-file locks */ flock.l_flock.end = OFFSET_MAX; - /* For flocks owner is determined by the local file desctiptor*/ + /* For flocks owner is determined by the local file descriptor*/ flock.l_flock.owner = (unsigned long)file_lock->fl_file; } else if (file_lock->fl_flags & FL_POSIX) { flock.l_flock.owner = (unsigned long)file_lock->fl_owner; @@ -2891,7 +2916,7 @@ int __ll_inode_revalidate_it(struct dentry *dentry, struct lookup_intent *it, oit.it_op = IT_LOOKUP; /* Call getattr by fid, so do not provide name at all. */ - op_data = ll_prep_md_op_data(NULL, dentry->d_parent->d_inode, + op_data = ll_prep_md_op_data(NULL, dentry->d_inode, dentry->d_inode, NULL, 0, 0, LUSTRE_OPC_ANY, NULL); if (IS_ERR(op_data)) @@ -3175,7 +3200,7 @@ struct inode_operations ll_file_inode_operations = { .get_acl = ll_get_acl, }; -/* dynamic ioctl number support routins */ +/* dynamic ioctl number support routines */ static struct llioc_ctl_data { struct rw_semaphore ioc_sem; struct list_head ioc_head; @@ -3299,7 +3324,7 @@ int ll_layout_conf(struct inode *inode, const struct cl_object_conf *conf) if (result == 0) { /* it can only be allowed to match after layout is * applied to inode otherwise false layout would be - * seen. Applying layout shoud happen before dropping + * seen. Applying layout should happen before dropping * the intent lock. */ ldlm_lock_allow_match(lock); } diff --git a/drivers/staging/lustre/lustre/llite/llite_close.c b/drivers/staging/lustre/lustre/llite/llite_close.c index e2996c46b2c0..38c2d0e947db 100644 --- a/drivers/staging/lustre/lustre/llite/llite_close.c +++ b/drivers/staging/lustre/lustre/llite/llite_close.c @@ -348,7 +348,7 @@ static int ll_close_thread(void *arg) break; inode = ll_info2i(lli); - CDEBUG(D_INFO, "done_writting for inode %lu/%u\n", + CDEBUG(D_INFO, "done_writing for inode %lu/%u\n", inode->i_ino, inode->i_generation); ll_done_writing(inode); iput(inode); diff --git a/drivers/staging/lustre/lustre/llite/llite_internal.h b/drivers/staging/lustre/lustre/llite/llite_internal.h index 7ee5c02783f9..69aba0afca41 100644 --- a/drivers/staging/lustre/lustre/llite/llite_internal.h +++ b/drivers/staging/lustre/lustre/llite/llite_internal.h @@ -296,13 +296,6 @@ int ll_xattr_cache_get(struct inode *inode, size_t size, __u64 valid); -int ll_xattr_cache_update(struct inode *inode, - const char *name, - const char *newval, - size_t size, - __u64 valid, - int flags); - /* * Locking to guarantee consistency of non-atomic updates to long long i_size, * consistency between file size and KMS. @@ -532,7 +525,7 @@ struct ll_sb_info { atomic_t ll_agl_total; /* AGL thread started count */ dev_t ll_sdev_orig; /* save s_dev before assign for - * clustred nfs */ + * clustered nfs */ struct rmtacl_ctl_table ll_rct; struct eacl_table ll_et; __kernel_fsid_t ll_fsid; @@ -782,7 +775,7 @@ int ll_local_open(struct file *file, int ll_release_openhandle(struct dentry *, struct lookup_intent *); int ll_md_close(struct obd_export *md_exp, struct inode *inode, struct file *file); -int ll_md_real_close(struct inode *inode, int flags); +int ll_md_real_close(struct inode *inode, fmode_t fmode); void ll_ioepoch_close(struct inode *inode, struct md_op_data *op_data, struct obd_client_handle **och, unsigned long flags); void ll_done_writing_attr(struct inode *inode, struct md_op_data *op_data); @@ -828,7 +821,7 @@ int ll_lease_close(struct obd_client_handle *och, struct inode *inode, /* llite/dcache.c */ -int ll_dops_init(struct dentry *de, int block, int init_sa); +int ll_d_init(struct dentry *de); extern struct dentry_operations ll_d_ops; void ll_intent_drop_lock(struct lookup_intent *); void ll_intent_release(struct lookup_intent *); @@ -915,12 +908,10 @@ struct ccc_object *cl_inode2ccc(struct inode *inode); void vvp_write_pending (struct ccc_object *club, struct ccc_page *page); void vvp_write_complete(struct ccc_object *club, struct ccc_page *page); -/* specific achitecture can implement only part of this list */ +/* specific architecture can implement only part of this list */ enum vvp_io_subtype { /** normal IO */ IO_NORMAL, - /** io called from .sendfile */ - IO_SENDFILE, /** io started from splice_{read|write} */ IO_SPLICE }; @@ -932,10 +923,6 @@ struct vvp_io { union { struct { - read_actor_t cui_actor; - void *cui_target; - } sendfile; - struct { struct pipe_inode_info *cui_pipe; unsigned int cui_flags; } splice; @@ -981,7 +968,7 @@ struct vvp_io { * IO arguments for various VFS I/O interfaces. */ struct vvp_io_args { - /** normal/sendfile/splice */ + /** normal/splice */ enum vvp_io_subtype via_io_subtype; union { @@ -991,10 +978,6 @@ struct vvp_io_args { unsigned long via_nrsegs; } normal; struct { - read_actor_t via_actor; - void *via_target; - } sendfile; - struct { struct pipe_inode_info *via_pipe; unsigned int via_flags; } splice; @@ -1320,12 +1303,13 @@ ll_statahead_mark(struct inode *dir, struct dentry *dentry) if (lli->lli_opendir_pid != current_pid()) return; - if (sai != NULL && ldd != NULL) + LASSERT(ldd != NULL); + if (sai != NULL) ldd->lld_sa_generation = sai->sai_generation; } static inline int -ll_need_statahead(struct inode *dir, struct dentry *dentryp) +d_need_statahead(struct inode *dir, struct dentry *dentryp) { struct ll_inode_info *lli; struct ll_dentry_data *ldd; @@ -1370,14 +1354,14 @@ ll_statahead_enter(struct inode *dir, struct dentry **dentryp, int only_unplug) { int ret; - ret = ll_need_statahead(dir, *dentryp); + ret = d_need_statahead(dir, *dentryp); if (ret <= 0) return ret; return do_statahead_enter(dir, dentryp, only_unplug); } -/* llite ioctl register support rountine */ +/* llite ioctl register support routine */ enum llioc_iter { LLIOC_CONT = 0, LLIOC_STOP @@ -1389,7 +1373,7 @@ enum llioc_iter { * Rules to write a callback function: * * Parameters: - * @magic: Dynamic ioctl call routine will feed this vaule with the pointer + * @magic: Dynamic ioctl call routine will feed this value with the pointer * returned to ll_iocontrol_register. Callback functions should use this * data to check the potential collasion of ioctl cmd. If collasion is * found, callback function should return LLIOC_CONT. @@ -1414,7 +1398,7 @@ enum llioc_iter ll_iocontrol_call(struct inode *inode, struct file *file, * @cb: callback function, it will be called if an ioctl command is found to * belong to the command list @cmd. * - * Return vaule: + * Return value: * A magic pointer will be returned if success; * otherwise, NULL will be returned. * */ @@ -1524,7 +1508,7 @@ static inline void ll_set_lock_data(struct obd_export *exp, struct inode *inode, * separate locks in different namespaces, Master MDT, * where the name entry is, will grant LOOKUP lock, * remote MDT, where the object is, will grant - * UPDATE|PERM lock. The inode will be attched to both + * UPDATE|PERM lock. The inode will be attached to both * LOOKUP and PERM locks, so revoking either locks will * case the dcache being cleared */ if (it->d.lustre.it_remote_lock_mode) { diff --git a/drivers/staging/lustre/lustre/llite/llite_lib.c b/drivers/staging/lustre/lustre/llite/llite_lib.c index 6cfdb9e4b74b..26003d3c1be7 100644 --- a/drivers/staging/lustre/lustre/llite/llite_lib.c +++ b/drivers/staging/lustre/lustre/llite/llite_lib.c @@ -155,11 +155,6 @@ void ll_free_sbi(struct super_block *sb) } } -static struct dentry_operations ll_d_root_ops = { - .d_compare = ll_dcompare, - .d_revalidate = ll_revalidate_nd, -}; - static int client_common_fill_super(struct super_block *sb, char *md, char *dt, struct vfsmount *mnt) { @@ -211,7 +206,10 @@ static int client_common_fill_super(struct super_block *sb, char *md, char *dt, OBD_CONNECT_EINPROGRESS | OBD_CONNECT_JOBSTATS | OBD_CONNECT_LVB_TYPE | OBD_CONNECT_LAYOUTLOCK | - OBD_CONNECT_PINGLESS | OBD_CONNECT_MAX_EASIZE; + OBD_CONNECT_PINGLESS | + OBD_CONNECT_MAX_EASIZE | + OBD_CONNECT_FLOCK_DEAD | + OBD_CONNECT_DISP_STRIPE; if (sbi->ll_flags & LL_SBI_SOM_PREVIEW) data->ocd_connect_flags |= OBD_CONNECT_SOM; @@ -281,7 +279,7 @@ static int client_common_fill_super(struct super_block *sb, char *md, char *dt, /* For mount, we only need fs info from MDT0, and also in DNE, it * can make sure the client can be mounted as long as MDT0 is - * avaible */ + * available */ err = obd_statfs(NULL, sbi->ll_md_exp, osfs, cfs_time_shift_64(-OBD_STATFS_CACHE_SECONDS), OBD_STATFS_FOR_MDT0); @@ -579,10 +577,6 @@ static int client_common_fill_super(struct super_block *sb, char *md, char *dt, GOTO(out_root, err = -ENOMEM); } - /* kernel >= 2.6.38 store dentry operations in sb->s_d_op. */ - d_set_d_op(sb->s_root, &ll_d_root_ops); - sb->s_d_op = &ll_d_ops; - sbi->ll_sdev_orig = sb->s_dev; /* We set sb->s_dev equal on all lustre clients in order to support @@ -723,7 +717,7 @@ void ll_kill_super(struct super_block *sb) return; sbi = ll_s2sbi(sb); - /* we need restore s_dev from changed for clustred NFS before put_super + /* we need to restore s_dev from changed for clustered NFS before put_super * because new kernels have cached s_dev and change sb->s_dev in * put_super not affected real removing devices */ if (sbi) { @@ -740,7 +734,8 @@ char *ll_read_opt(const char *opt, char *data) CDEBUG(D_SUPER, "option: %s, data %s\n", opt, data); if (strncmp(opt, data, strlen(opt))) return NULL; - if ((value = strchr(data, '=')) == NULL) + value = strchr(data, '='); + if (value == NULL) return NULL; value++; @@ -1013,6 +1008,8 @@ int ll_fill_super(struct super_block *sb, struct vfsmount *mnt) GOTO(out_free, err); sb->s_bdi = &lsi->lsi_bdi; + /* kernel >= 2.6.38 store dentry operations in sb->s_d_op. */ + sb->s_d_op = &ll_d_ops; /* Generate a string unique to this super, in case some joker tries to mount the same fs at two mount points. @@ -1067,7 +1064,7 @@ out_free: void ll_put_super(struct super_block *sb) { - struct config_llog_instance cfg; + struct config_llog_instance cfg, params_cfg; struct obd_device *obd; struct lustre_sb_info *lsi = s2lsi(sb); struct ll_sb_info *sbi = ll_s2sbi(sb); @@ -1081,6 +1078,9 @@ void ll_put_super(struct super_block *sb) cfg.cfg_instance = sb; lustre_end_log(sb, profilenm, &cfg); + params_cfg.cfg_instance = sb; + lustre_end_log(sb, PARAMS_FILENAME, ¶ms_cfg); + if (sbi->ll_md_exp) { obd = class_exp2obd(sbi->ll_md_exp); if (obd) @@ -1405,7 +1405,7 @@ int ll_setattr_raw(struct dentry *dentry, struct iattr *attr, bool hsm_import) /* POSIX: check before ATTR_*TIME_SET set (from inode_change_ok) */ if (attr->ia_valid & TIMES_SET_FLAGS) { if ((!uid_eq(current_fsuid(), inode->i_uid)) && - !cfs_capable(CFS_CAP_FOWNER)) + !capable(CFS_CAP_FOWNER)) return -EPERM; } diff --git a/drivers/staging/lustre/lustre/llite/llite_mmap.c b/drivers/staging/lustre/lustre/llite/llite_mmap.c index caed6423e4ef..90b2c0d275f9 100644 --- a/drivers/staging/lustre/lustre/llite/llite_mmap.c +++ b/drivers/staging/lustre/lustre/llite/llite_mmap.c @@ -285,7 +285,7 @@ static inline int to_fault_error(int result) * Lustre implementation of a vm_operations_struct::fault() method, called by * VM to server page fault (both in kernel and user space). * - * \param vma - is virtiual area struct related to page fault + * \param vma - is virtual area struct related to page fault * \param vmf - structure which describe type and address where hit fault * * \return allocated and filled _locked_ page for address @@ -370,7 +370,7 @@ restart: goto restart; } - result |= VM_FAULT_LOCKED; + result = VM_FAULT_LOCKED; } cfs_restore_sigs(set); return result; diff --git a/drivers/staging/lustre/lustre/llite/llite_nfs.c b/drivers/staging/lustre/lustre/llite/llite_nfs.c index 1767c741fb72..3580069789fc 100644 --- a/drivers/staging/lustre/lustre/llite/llite_nfs.c +++ b/drivers/staging/lustre/lustre/llite/llite_nfs.c @@ -167,10 +167,10 @@ ll_iget_for_nfs(struct super_block *sb, struct lu_fid *fid, struct lu_fid *paren } result = d_obtain_alias(inode); - if (IS_ERR(result)) + if (IS_ERR(result)) { + iput(inode); return result; - - ll_dops_init(result, 1, 0); + } return result; } diff --git a/drivers/staging/lustre/lustre/llite/lloop.c b/drivers/staging/lustre/lustre/llite/lloop.c index 0718905adeb2..f78eda235c7a 100644 --- a/drivers/staging/lustre/lustre/llite/lloop.c +++ b/drivers/staging/lustre/lustre/llite/lloop.c @@ -255,7 +255,7 @@ static int do_bio_lustrebacked(struct lloop_device *lo, struct bio *head) * to store parity; * 2. Reserve the # of (page_count * depth) cl_pages from the reserved * pool. Afterwards, the clio would allocate the pages from reserved - * pool, this guarantees we neeedn't allocate the cl_pages from + * pool, this guarantees we needn't allocate the cl_pages from * generic cl_page slab cache. * Of course, if there is NOT enough pages in the pool, we might * be asked to write less pages once, this purely depends on @@ -325,7 +325,7 @@ static unsigned int loop_get_bio(struct lloop_device *lo, struct bio **req) bio = &(*bio)->bi_next; } if (*bio) { - /* Some of bios can't be mergable. */ + /* Some of bios can't be mergeable. */ lo->lo_bio = *bio; *bio = NULL; } else { @@ -658,7 +658,7 @@ static struct block_device_operations lo_fops = { * ll_iocontrol_call. * * This is a llite regular file ioctl function. It takes the responsibility - * of attaching or detaching a file by a lloop's device numner. + * of attaching or detaching a file by a lloop's device number. */ static enum llioc_iter lloop_ioctl(struct inode *unused, struct file *file, unsigned int cmd, unsigned long arg, diff --git a/drivers/staging/lustre/lustre/llite/namei.c b/drivers/staging/lustre/lustre/llite/namei.c index fc8d264f6c9a..25a6ea580f00 100644 --- a/drivers/staging/lustre/lustre/llite/namei.c +++ b/drivers/staging/lustre/lustre/llite/namei.c @@ -195,101 +195,107 @@ static void ll_invalidate_negative_children(struct inode *dir) int ll_md_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *desc, void *data, int flag) { - int rc; struct lustre_handle lockh; + int rc; switch (flag) { case LDLM_CB_BLOCKING: ldlm_lock2handle(lock, &lockh); rc = ldlm_cli_cancel(&lockh, LCF_ASYNC); if (rc < 0) { - CDEBUG(D_INODE, "ldlm_cli_cancel: %d\n", rc); + CDEBUG(D_INODE, "ldlm_cli_cancel: rc = %d\n", rc); return rc; } break; case LDLM_CB_CANCELING: { struct inode *inode = ll_inode_from_resource_lock(lock); - struct ll_inode_info *lli; __u64 bits = lock->l_policy_data.l_inodebits.bits; - struct lu_fid *fid; - ldlm_mode_t mode = lock->l_req_mode; /* Inode is set to lock->l_resource->lr_lvb_inode * for mdc - bug 24555 */ LASSERT(lock->l_ast_data == NULL); - /* Invalidate all dentries associated with this inode */ if (inode == NULL) break; + /* Invalidate all dentries associated with this inode */ LASSERT(lock->l_flags & LDLM_FL_CANCELING); - if (bits & MDS_INODELOCK_XATTR) + if (!fid_res_name_eq(ll_inode2fid(inode), + &lock->l_resource->lr_name)) { + LDLM_ERROR(lock, "data mismatch with object "DFID"(%p)", + PFID(ll_inode2fid(inode)), inode); + LBUG(); + } + + if (bits & MDS_INODELOCK_XATTR) { ll_xattr_cache_destroy(inode); + bits &= ~MDS_INODELOCK_XATTR; + } /* For OPEN locks we differentiate between lock modes * LCK_CR, LCK_CW, LCK_PR - bug 22891 */ - if (bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_UPDATE | - MDS_INODELOCK_LAYOUT | MDS_INODELOCK_PERM)) - ll_have_md_lock(inode, &bits, LCK_MINMODE); - if (bits & MDS_INODELOCK_OPEN) - ll_have_md_lock(inode, &bits, mode); - - fid = ll_inode2fid(inode); - if (!fid_res_name_eq(fid, &lock->l_resource->lr_name)) - LDLM_ERROR(lock, "data mismatch with object " - DFID" (%p)", PFID(fid), inode); + ll_have_md_lock(inode, &bits, lock->l_req_mode); if (bits & MDS_INODELOCK_OPEN) { - int flags = 0; + fmode_t fmode; + switch (lock->l_req_mode) { case LCK_CW: - flags = FMODE_WRITE; + fmode = FMODE_WRITE; break; case LCK_PR: - flags = FMODE_EXEC; + fmode = FMODE_EXEC; break; case LCK_CR: - flags = FMODE_READ; + fmode = FMODE_READ; break; default: - CERROR("Unexpected lock mode for OPEN lock " - "%d, inode %ld\n", lock->l_req_mode, - inode->i_ino); + LDLM_ERROR(lock, "bad lock mode for OPEN lock"); + LBUG(); } - ll_md_real_close(inode, flags); + + ll_md_real_close(inode, fmode); } - lli = ll_i2info(inode); + if (bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_UPDATE | + MDS_INODELOCK_LAYOUT | MDS_INODELOCK_PERM)) + ll_have_md_lock(inode, &bits, LCK_MINMODE); + if (bits & MDS_INODELOCK_LAYOUT) { - struct cl_object_conf conf = { { 0 } }; + struct cl_object_conf conf = { + .coc_opc = OBJECT_CONF_INVALIDATE, + .coc_inode = inode, + }; - conf.coc_opc = OBJECT_CONF_INVALIDATE; - conf.coc_inode = inode; rc = ll_layout_conf(inode, &conf); - if (rc) - CDEBUG(D_INODE, "invaliding layout %d.\n", rc); + if (rc < 0) + CDEBUG(D_INODE, "cannot invalidate layout of " + DFID": rc = %d\n", + PFID(ll_inode2fid(inode)), rc); } if (bits & MDS_INODELOCK_UPDATE) { + struct ll_inode_info *lli = ll_i2info(inode); + spin_lock(&lli->lli_lock); lli->lli_flags &= ~LLIF_MDS_SIZE_LOCK; spin_unlock(&lli->lli_lock); } - if (S_ISDIR(inode->i_mode) && - (bits & MDS_INODELOCK_UPDATE)) { + if ((bits & MDS_INODELOCK_UPDATE) && S_ISDIR(inode->i_mode)) { CDEBUG(D_INODE, "invalidating inode %lu\n", inode->i_ino); truncate_inode_pages(inode->i_mapping, 0); ll_invalidate_negative_children(inode); } - if (inode->i_sb->s_root && - inode != inode->i_sb->s_root->d_inode && - (bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_PERM))) + if ((bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_PERM)) && + inode->i_sb->s_root != NULL && + inode != inode->i_sb->s_root->d_inode) ll_invalidate_aliases(inode); + iput(inode); break; } @@ -400,11 +406,16 @@ static struct dentry *ll_find_alias(struct inode *inode, struct dentry *dentry) struct dentry *ll_splice_alias(struct inode *inode, struct dentry *de) { struct dentry *new; + int rc; if (inode) { new = ll_find_alias(inode, de); if (new) { - ll_dops_init(new, 1, 1); + rc = ll_d_init(new); + if (rc < 0) { + dput(new); + return ERR_PTR(rc); + } d_move(new, de); iput(inode); CDEBUG(D_DENTRY, @@ -413,8 +424,9 @@ struct dentry *ll_splice_alias(struct inode *inode, struct dentry *de) return new; } } - ll_dops_init(de, 1, 1); - __d_lustre_invalidate(de); + rc = ll_d_init(de); + if (rc < 0) + return ERR_PTR(rc); d_add(de, inode); CDEBUG(D_DENTRY, "Add dentry %p inode %p refc %d flags %#x\n", de, de->d_inode, d_count(de), de->d_flags); @@ -453,10 +465,22 @@ int ll_lookup_it_finish(struct ptlrpc_request *request, } /* Only hash *de if it is unhashed (new dentry). - * Atoimc_open may passin hashed dentries for open. + * Atoimc_open may passing hashed dentries for open. */ - if (d_unhashed(*de)) - *de = ll_splice_alias(inode, *de); + if (d_unhashed(*de)) { + struct dentry *alias; + + alias = ll_splice_alias(inode, *de); + if (IS_ERR(alias)) + return PTR_ERR(alias); + *de = alias; + } else if (!it_disposition(it, DISP_LOOKUP_NEG) && + !it_disposition(it, DISP_OPEN_CREATE)) { + /* With DISP_OPEN_CREATE dentry will + instantiated in ll_create_it. */ + LASSERT((*de)->d_inode == NULL); + d_instantiate(*de, inode); + } if (!it_disposition(it, DISP_LOOKUP_NEG)) { /* we have lookup look - unhide dentry */ @@ -505,16 +529,6 @@ static struct dentry *ll_lookup_it(struct inode *parent, struct dentry *dentry, ll_frob_intent(&it, &lookup_it); - /* As do_lookup is called before follow_mount, root dentry may be left - * not valid, revalidate it here. */ - if (parent->i_sb->s_root && (parent->i_sb->s_root->d_inode == parent) && - (it->it_op & (IT_OPEN | IT_CREAT))) { - rc = ll_inode_revalidate_it(parent->i_sb->s_root, it, - MDS_INODELOCK_LOOKUP); - if (rc) - return ERR_PTR(rc); - } - if (it->it_op == IT_GETATTR) { rc = ll_statahead_enter(parent, &dentry, 0); if (rc == 1) { @@ -584,12 +598,8 @@ static struct dentry *ll_lookup_nd(struct inode *parent, struct dentry *dentry, parent->i_generation, parent, flags); /* Optimize away (CREATE && !OPEN). Let .create handle the race. */ - if ((flags & LOOKUP_CREATE ) && !(flags & LOOKUP_OPEN)) { - ll_dops_init(dentry, 1, 1); - __d_lustre_invalidate(dentry); - d_add(dentry, NULL); + if ((flags & LOOKUP_CREATE) && !(flags & LOOKUP_OPEN)) return NULL; - } if (flags & (LOOKUP_PARENT|LOOKUP_OPEN|LOOKUP_CREATE)) itp = NULL; diff --git a/drivers/staging/lustre/lustre/llite/rw.c b/drivers/staging/lustre/lustre/llite/rw.c index e9ba38a553cf..416f7a094a6d 100644 --- a/drivers/staging/lustre/lustre/llite/rw.c +++ b/drivers/staging/lustre/lustre/llite/rw.c @@ -135,7 +135,7 @@ static struct ll_cl_context *ll_cl_init(struct file *file, } /* - * Loop-back driver calls ->prepare_write() and ->sendfile() + * Loop-back driver calls ->prepare_write(). * methods directly, bypassing file system ->write() operation, * so cl_io has to be created here. */ @@ -558,10 +558,10 @@ static int ll_read_ahead_page(const struct lu_env *env, struct cl_io *io, * striped over, rather than having a constant value for all files here. */ /* RAS_INCREASE_STEP should be (1UL << (inode->i_blkbits - PAGE_CACHE_SHIFT)). - * Temprarily set RAS_INCREASE_STEP to 1MB. After 4MB RPC is enabled + * Temporarily set RAS_INCREASE_STEP to 1MB. After 4MB RPC is enabled * by default, this should be adjusted corresponding with max_read_ahead_mb * and max_read_ahead_per_file_mb otherwise the readahead budget can be used - * up quickly which will affect read performance siginificantly. See LU-2816 */ + * up quickly which will affect read performance significantly. See LU-2816 */ #define RAS_INCREASE_STEP(inode) (ONE_MB_BRW_SIZE >> PAGE_CACHE_SHIFT) static inline int stride_io_mode(struct ll_readahead_state *ras) @@ -570,7 +570,7 @@ static inline int stride_io_mode(struct ll_readahead_state *ras) } /* The function calculates how much pages will be read in * [off, off + length], in such stride IO area, - * stride_offset = st_off, stride_lengh = st_len, + * stride_offset = st_off, stride_length = st_len, * stride_pages = st_pgs * * |------------------|*****|------------------|*****|------------|*****|.... @@ -1090,7 +1090,7 @@ void ras_update(struct ll_sb_info *sbi, struct inode *inode, ras_set_start(inode, ras, index); if (stride_io_mode(ras)) - /* Since stride readahead is sentivite to the offset + /* Since stride readahead is sensitive to the offset * of read-ahead, so we use original offset here, * instead of ras_window_start, which is RPC aligned */ ras->ras_next_readahead = max(index, ras->ras_next_readahead); diff --git a/drivers/staging/lustre/lustre/llite/statahead.c b/drivers/staging/lustre/lustre/llite/statahead.c index f6b5f4b95f37..c8624b5a9875 100644 --- a/drivers/staging/lustre/lustre/llite/statahead.c +++ b/drivers/staging/lustre/lustre/llite/statahead.c @@ -577,7 +577,7 @@ static void ll_agl_trigger(struct inode *inode, struct ll_statahead_info *sai) * Someone triggered glimpse within 1 sec before. * 1) The former glimpse succeeded with glimpse lock granted by OST, and * if the lock is still cached on client, AGL needs to do nothing. If - * it is cancelled by other client, AGL maybe cannot obtaion new lock + * it is cancelled by other client, AGL maybe cannot obtain new lock * for no glimpse callback triggered by AGL. * 2) The former glimpse succeeded, but OST did not grant glimpse lock. * Under such case, it is quite possible that the OST will not grant @@ -877,9 +877,6 @@ static int do_sa_revalidate(struct inode *dir, struct ll_sa_entry *entry, if (d_mountpoint(dentry)) return 1; - if (unlikely(dentry == dentry->d_sb->s_root)) - return 1; - entry->se_inode = igrab(inode); rc = md_revalidate_lock(ll_i2mdexp(dir), &it, ll_inode2fid(inode),NULL); if (rc == 1) { @@ -1588,8 +1585,15 @@ int do_statahead_enter(struct inode *dir, struct dentry **dentryp, ll_inode2fid(inode), &bits); if (rc == 1) { if ((*dentryp)->d_inode == NULL) { - *dentryp = ll_splice_alias(inode, + struct dentry *alias; + + alias = ll_splice_alias(inode, *dentryp); + if (IS_ERR(alias)) { + ll_sai_unplug(sai, entry); + return PTR_ERR(alias); + } + *dentryp = alias; } else if ((*dentryp)->d_inode != inode) { /* revalidate, but inode is recreated */ CDEBUG(D_READA, diff --git a/drivers/staging/lustre/lustre/llite/vvp_io.c b/drivers/staging/lustre/lustre/llite/vvp_io.c index 93cbfbb7e7f7..c7d70091246e 100644 --- a/drivers/staging/lustre/lustre/llite/vvp_io.c +++ b/drivers/staging/lustre/lustre/llite/vvp_io.c @@ -51,7 +51,7 @@ static struct vvp_io *cl2vvp_io(const struct lu_env *env, const struct cl_io_slice *slice); /** - * True, if \a io is a normal io, False for sendfile() / splice_{read|write} + * True, if \a io is a normal io, False for splice_{read,write} */ int cl_is_normalio(const struct lu_env *env, const struct cl_io *io) { @@ -474,20 +474,6 @@ static void vvp_io_setattr_fini(const struct lu_env *env, vvp_io_fini(env, ios); } -static ssize_t lustre_generic_file_read(struct file *file, - struct ccc_io *vio, loff_t *ppos) -{ - return generic_file_aio_read(vio->cui_iocb, vio->cui_iov, - vio->cui_nrsegs, *ppos); -} - -static ssize_t lustre_generic_file_write(struct file *file, - struct ccc_io *vio, loff_t *ppos) -{ - return generic_file_aio_write(vio->cui_iocb, vio->cui_iov, - vio->cui_nrsegs, *ppos); -} - static int vvp_io_read_start(const struct lu_env *env, const struct cl_io_slice *ios) { @@ -540,8 +526,11 @@ static int vvp_io_read_start(const struct lu_env *env, file_accessed(file); switch (vio->cui_io_subtype) { case IO_NORMAL: - result = lustre_generic_file_read(file, cio, &pos); - break; + LASSERT(cio->cui_iocb->ki_pos == pos); + result = generic_file_aio_read(cio->cui_iocb, + cio->cui_iov, cio->cui_nrsegs, + cio->cui_iocb->ki_pos); + break; case IO_SPLICE: result = generic_file_splice_read(file, &pos, vio->u.splice.cui_pipe, cnt, @@ -586,7 +575,6 @@ static int vvp_io_write_start(const struct lu_env *env, struct cl_io *io = ios->cis_io; struct cl_object *obj = io->ci_obj; struct inode *inode = ccc_object_inode(obj); - struct file *file = cio->cui_fd->fd_file; ssize_t result = 0; loff_t pos = io->u.ci_wr.wr.crw_pos; size_t cnt = io->u.ci_wr.wr.crw_count; @@ -601,6 +589,8 @@ static int vvp_io_write_start(const struct lu_env *env, */ pos = io->u.ci_wr.wr.crw_pos = i_size_read(inode); cio->cui_iocb->ki_pos = pos; + } else { + LASSERT(cio->cui_iocb->ki_pos == pos); } CDEBUG(D_VFSTRACE, "write: [%lli, %lli)\n", pos, pos + (long long)cnt); @@ -608,8 +598,9 @@ static int vvp_io_write_start(const struct lu_env *env, if (cio->cui_iov == NULL) /* from a temp io in ll_cl_init(). */ result = 0; else - result = lustre_generic_file_write(file, cio, &pos); - + result = generic_file_aio_write(cio->cui_iocb, + cio->cui_iov, cio->cui_nrsegs, + cio->cui_iocb->ki_pos); if (result > 0) { if (result < cnt) io->ci_continue = 0; @@ -655,7 +646,7 @@ static int vvp_io_kernel_fault(struct vvp_fault_io *cfio) if (cfio->fault.ft_flags & VM_FAULT_RETRY) return -EAGAIN; - CERROR("unknow error in page fault %d!\n", cfio->fault.ft_flags); + CERROR("Unknown error in page fault %d!\n", cfio->fault.ft_flags); return -EINVAL; } @@ -1201,7 +1192,7 @@ int vvp_io_init(const struct lu_env *env, struct cl_object *obj, if (result == -ENOENT) /* If the inode on MDS has been removed, but the objects * on OSTs haven't been destroyed (async unlink), layout - * fetch will return -ENOENT, we'd ingore this error + * fetch will return -ENOENT, we'd ignore this error * and continue with dirty flush. LU-3230. */ result = 0; if (result < 0) @@ -1216,7 +1207,7 @@ int vvp_io_init(const struct lu_env *env, struct cl_object *obj, static struct vvp_io *cl2vvp_io(const struct lu_env *env, const struct cl_io_slice *slice) { - /* Caling just for assertion */ + /* Calling just for assertion */ cl2ccc_io(env, slice); return vvp_env_io(env); } diff --git a/drivers/staging/lustre/lustre/llite/xattr.c b/drivers/staging/lustre/lustre/llite/xattr.c index 3a7d03c12dd9..b1ed4d9ea6be 100644 --- a/drivers/staging/lustre/lustre/llite/xattr.c +++ b/drivers/staging/lustre/lustre/llite/xattr.c @@ -95,7 +95,7 @@ int xattr_type_filter(struct ll_sb_info *sbi, int xattr_type) if (xattr_type == XATTR_USER_T && !(sbi->ll_flags & LL_SBI_USER_XATTR)) return -EOPNOTSUPP; - if (xattr_type == XATTR_TRUSTED_T && !cfs_capable(CFS_CAP_SYS_ADMIN)) + if (xattr_type == XATTR_TRUSTED_T && !capable(CFS_CAP_SYS_ADMIN)) return -EPERM; if (xattr_type == XATTR_OTHER_T) return -EOPNOTSUPP; @@ -183,17 +183,11 @@ int ll_setxattr_common(struct inode *inode, const char *name, valid |= rce_ops2valid(rce->rce_ops); } #endif - if (sbi->ll_xattr_cache_enabled && - (rce == NULL || rce->rce_ops == RMT_LSETFACL)) { - rc = ll_xattr_cache_update(inode, name, pv, size, valid, flags); - } else { oc = ll_mdscapa_get(inode); rc = md_setxattr(sbi->ll_md_exp, ll_inode2fid(inode), oc, valid, name, pv, size, 0, flags, ll_i2suppgid(inode), &req); capa_put(oc); - } - #ifdef CONFIG_FS_POSIX_ACL if (new_value != NULL) lustre_posix_acl_xattr_free(new_value, size); @@ -292,6 +286,7 @@ int ll_getxattr_common(struct inode *inode, const char *name, void *xdata; struct obd_capa *oc; struct rmtacl_ctl_entry *rce = NULL; + struct ll_inode_info *lli = ll_i2info(inode); CDEBUG(D_VFSTRACE, "VFS Op:inode=%lu/%u(%p)\n", inode->i_ino, inode->i_generation, inode); @@ -339,7 +334,7 @@ int ll_getxattr_common(struct inode *inode, const char *name, */ if (xattr_type == XATTR_ACL_ACCESS_T && !(sbi->ll_flags & LL_SBI_RMT_CLIENT)) { - struct ll_inode_info *lli = ll_i2info(inode); + struct posix_acl *acl; spin_lock(&lli->lli_lock); @@ -358,13 +353,27 @@ int ll_getxattr_common(struct inode *inode, const char *name, #endif do_getxattr: - if (sbi->ll_xattr_cache_enabled && (rce == NULL || - rce->rce_ops == RMT_LGETFACL || - rce->rce_ops == RMT_LSETFACL)) { + if (sbi->ll_xattr_cache_enabled && xattr_type != XATTR_ACL_ACCESS_T) { rc = ll_xattr_cache_get(inode, name, buffer, size, valid); + if (rc == -EAGAIN) + goto getxattr_nocache; if (rc < 0) GOTO(out_xattr, rc); + + /* Add "system.posix_acl_access" to the list */ + if (lli->lli_posix_acl != NULL && valid & OBD_MD_FLXATTRLS) { + if (size == 0) { + rc += sizeof(XATTR_NAME_ACL_ACCESS); + } else if (size - rc >= sizeof(XATTR_NAME_ACL_ACCESS)) { + memcpy(buffer + rc, XATTR_NAME_ACL_ACCESS, + sizeof(XATTR_NAME_ACL_ACCESS)); + rc += sizeof(XATTR_NAME_ACL_ACCESS); + } else { + GOTO(out_xattr, rc = -ERANGE); + } + } } else { +getxattr_nocache: oc = ll_mdscapa_get(inode); rc = md_getxattr(sbi->ll_md_exp, ll_inode2fid(inode), oc, valid | (rce ? rce_ops2valid(rce->rce_ops) : 0), diff --git a/drivers/staging/lustre/lustre/llite/xattr_cache.c b/drivers/staging/lustre/lustre/llite/xattr_cache.c index 3e3be1f13502..4defa2fd83b3 100644 --- a/drivers/staging/lustre/lustre/llite/xattr_cache.c +++ b/drivers/staging/lustre/lustre/llite/xattr_cache.c @@ -98,13 +98,13 @@ static int ll_xattr_cache_find(struct list_head *cache, } /** - * This adds or updates an xattr. + * This adds an xattr. * * Add @xattr_name attr with @xattr_val value and @xattr_val_len length, - * if the attribute already exists, then update its value. * * \retval 0 success * \retval -ENOMEM if no memory could be allocated for the cached attr + * \retval -EPROTO if duplicate xattr is being added */ static int ll_xattr_cache_add(struct list_head *cache, const char *xattr_name, @@ -116,27 +116,8 @@ static int ll_xattr_cache_add(struct list_head *cache, if (ll_xattr_cache_find(cache, xattr_name, &xattr) == 0) { - /* Found a cached EA, update it */ - - if (xattr_val_len != xattr->xe_vallen) { - char *val; - OBD_ALLOC(val, xattr_val_len); - if (val == NULL) { - CDEBUG(D_CACHE, - "failed to allocate %u bytes for xattr %s update\n", - xattr_val_len, xattr_name); - return -ENOMEM; - } - OBD_FREE(xattr->xe_value, xattr->xe_vallen); - xattr->xe_value = val; - xattr->xe_vallen = xattr_val_len; - } - memcpy(xattr->xe_value, xattr_val, xattr_val_len); - - CDEBUG(D_CACHE, "update: [%s]=%.*s\n", xattr_name, - xattr_val_len, xattr_val); - - return 0; + CDEBUG(D_CACHE, "duplicate xattr: [%s]\n", xattr_name); + return -EPROTO; } OBD_SLAB_ALLOC_PTR_GFP(xattr, xattr_kmem, __GFP_IO); @@ -261,7 +242,7 @@ int ll_xattr_cache_valid(struct ll_inode_info *lli) * * Free all xattr memory. @lli is the inode info pointer. * - * \retval 0 no error occured + * \retval 0 no error occurred */ static int ll_xattr_cache_destroy_locked(struct ll_inode_info *lli) { @@ -292,14 +273,14 @@ int ll_xattr_cache_destroy(struct inode *inode) } /** - * Match or enqueue a PR or PW LDLM lock. + * Match or enqueue a PR lock. * * Find or request an LDLM lock with xattr data. * Since LDLM does not provide API for atomic match_or_enqueue, * the function handles it with a separate enq lock. * If successful, the function exits with the list lock held. * - * \retval 0 no error occured + * \retval 0 no error occurred * \retval -ENOMEM not enough memory */ static int ll_xattr_find_get_lock(struct inode *inode, @@ -322,9 +303,7 @@ static int ll_xattr_find_get_lock(struct inode *inode, mutex_lock(&lli->lli_xattrs_enq_lock); /* Try matching first. */ - mode = ll_take_md_lock(inode, MDS_INODELOCK_XATTR, &lockh, 0, - oit->it_op == IT_SETXATTR ? LCK_PW : - (LCK_PR | LCK_PW)); + mode = ll_take_md_lock(inode, MDS_INODELOCK_XATTR, &lockh, 0, LCK_PR); if (mode != 0) { /* fake oit in mdc_revalidate_lock() manner */ oit->d.lustre.it_lock_handle = lockh.cookie; @@ -340,13 +319,7 @@ static int ll_xattr_find_get_lock(struct inode *inode, return PTR_ERR(op_data); } - op_data->op_valid = OBD_MD_FLXATTR | OBD_MD_FLXATTRLS | - OBD_MD_FLXATTRLOCKED; -#ifdef CONFIG_FS_POSIX_ACL - /* If working with ACLs, we would like to cache local ACLs */ - if (sbi->ll_flags & LL_SBI_RMT_CLIENT) - op_data->op_valid |= OBD_MD_FLRMTLGETFACL; -#endif + op_data->op_valid = OBD_MD_FLXATTR | OBD_MD_FLXATTRLS; rc = md_enqueue(exp, &einfo, oit, op_data, &lockh, NULL, 0, NULL, 0); ll_finish_md_op_data(op_data); @@ -374,7 +347,7 @@ out: * a read or a write xattr lock depending on operation in @oit. * Intent is dropped on exit unless the operation is setxattr. * - * \retval 0 no error occured + * \retval 0 no error occurred * \retval -EPROTO network protocol error * \retval -ENOMEM not enough memory for the cache */ @@ -409,7 +382,11 @@ static int ll_xattr_cache_refill(struct inode *inode, struct lookup_intent *oit) if (oit->d.lustre.it_status < 0) { CDEBUG(D_CACHE, "getxattr intent returned %d for fid "DFID"\n", oit->d.lustre.it_status, PFID(ll_inode2fid(inode))); - GOTO(out_destroy, rc = oit->d.lustre.it_status); + rc = oit->d.lustre.it_status; + /* xattr data is so large that we don't want to cache it */ + if (rc == -ERANGE) + rc = -EAGAIN; + GOTO(out_destroy, rc); } body = req_capsule_server_get(&req->rq_pill, &RMF_MDT_BODY); @@ -447,6 +424,11 @@ static int ll_xattr_cache_refill(struct inode *inode, struct lookup_intent *oit) rc = -EPROTO; } else if (OBD_FAIL_CHECK(OBD_FAIL_LLITE_XATTR_ENOMEM)) { rc = -ENOMEM; + } else if (!strcmp(xdata, XATTR_NAME_ACL_ACCESS)) { + /* Filter out ACL ACCESS since it's cached separately */ + CDEBUG(D_CACHE, "not caching %s\n", + XATTR_NAME_ACL_ACCESS); + rc = 0; } else { rc = ll_xattr_cache_add(&lli->lli_xattrs, xdata, xval, *xsizes); @@ -467,8 +449,7 @@ static int ll_xattr_cache_refill(struct inode *inode, struct lookup_intent *oit) GOTO(out_maybe_drop, rc); out_maybe_drop: - /* drop lock on error or getxattr */ - if (rc != 0 || oit->it_op != IT_SETXATTR) + ll_intent_drop_lock(oit); if (rc != 0) @@ -496,7 +477,7 @@ out_destroy: * The resulting value/list is stored in @buffer if the former * is not larger than @size. * - * \retval 0 no error occured + * \retval 0 no error occurred * \retval -EPROTO network protocol error * \retval -ENOMEM not enough memory for the cache * \retval -ERANGE the buffer is not large enough @@ -553,65 +534,3 @@ out: return rc; } - - -/** - * Set/update an xattr value or remove xattr using the write-through cache. - * - * Set/update the xattr value (if @valid has OBD_MD_FLXATTR) of @name to @newval - * or - * remove the xattr @name (@valid has OBD_MD_FLXATTRRM set) from @inode. - * @flags is either XATTR_CREATE or XATTR_REPLACE as defined by setxattr(2) - * - * \retval 0 no error occured - * \retval -EPROTO network protocol error - * \retval -ENOMEM not enough memory for the cache - * \retval -ERANGE the buffer is not large enough - * \retval -ENODATA no such attr (in the removal case) - */ -int ll_xattr_cache_update(struct inode *inode, - const char *name, - const char *newval, - size_t size, - __u64 valid, - int flags) -{ - struct lookup_intent oit = { .it_op = IT_SETXATTR }; - struct ll_sb_info *sbi = ll_i2sbi(inode); - struct ptlrpc_request *req = NULL; - struct ll_inode_info *lli = ll_i2info(inode); - struct obd_capa *oc; - int rc; - - - - LASSERT(!!(valid & OBD_MD_FLXATTR) ^ !!(valid & OBD_MD_FLXATTRRM)); - - rc = ll_xattr_cache_refill(inode, &oit); - if (rc) - return rc; - - oc = ll_mdscapa_get(inode); - rc = md_setxattr(sbi->ll_md_exp, ll_inode2fid(inode), oc, - valid | OBD_MD_FLXATTRLOCKED, name, newval, - size, 0, flags, ll_i2suppgid(inode), &req); - capa_put(oc); - - if (rc) { - ll_intent_drop_lock(&oit); - GOTO(out, rc); - } - - if (valid & OBD_MD_FLXATTR) - rc = ll_xattr_cache_add(&lli->lli_xattrs, name, newval, size); - else if (valid & OBD_MD_FLXATTRRM) - rc = ll_xattr_cache_del(&lli->lli_xattrs, name); - - ll_intent_drop_lock(&oit); - GOTO(out, rc); -out: - up_write(&lli->lli_xattrs_list_rwsem); - ptlrpc_req_finished(req); - - return rc; -} diff --git a/drivers/staging/lustre/lustre/lmv/lmv_intent.c b/drivers/staging/lustre/lustre/lmv/lmv_intent.c index 56dedceaf0a0..9ba5a0a57390 100644 --- a/drivers/staging/lustre/lustre/lmv/lmv_intent.c +++ b/drivers/staging/lustre/lustre/lmv/lmv_intent.c @@ -119,7 +119,6 @@ static int lmv_intent_remote(struct obd_export *exp, void *lmm, CDEBUG(D_INODE, "REMOTE_INTENT with fid="DFID" -> mds #%d\n", PFID(&body->fid1), tgt->ltd_idx); - it->d.lustre.it_disposition &= ~DISP_ENQ_COMPLETE; rc = md_intent_lock(tgt->ltd_exp, op_data, lmm, lmmsize, it, flags, &req, cb_blocking, extra_lock_flags); if (rc) diff --git a/drivers/staging/lustre/lustre/lmv/lmv_obd.c b/drivers/staging/lustre/lustre/lmv/lmv_obd.c index 1bddd8f62fbf..3ba0a0a1d945 100644 --- a/drivers/staging/lustre/lustre/lmv/lmv_obd.c +++ b/drivers/staging/lustre/lustre/lmv/lmv_obd.c @@ -1744,7 +1744,6 @@ lmv_enqueue_remote(struct obd_export *exp, struct ldlm_enqueue_info *einfo, it->d.lustre.it_data = NULL; fid1 = body->fid1; - it->d.lustre.it_disposition &= ~DISP_ENQ_COMPLETE; ptlrpc_req_finished(req); tgt = lmv_find_target(lmv, &fid1); @@ -2593,7 +2592,7 @@ int lmv_free_lustre_md(struct obd_export *exp, struct lustre_md *md) int lmv_set_open_replay_data(struct obd_export *exp, struct obd_client_handle *och, - struct ptlrpc_request *open_req) + struct lookup_intent *it) { struct obd_device *obd = exp->exp_obd; struct lmv_obd *lmv = &obd->u.lmv; @@ -2603,7 +2602,7 @@ int lmv_set_open_replay_data(struct obd_export *exp, if (IS_ERR(tgt)) return PTR_ERR(tgt); - return md_set_open_replay_data(tgt->ltd_exp, och, open_req); + return md_set_open_replay_data(tgt->ltd_exp, och, it); } int lmv_clear_open_replay_data(struct obd_export *exp, diff --git a/drivers/staging/lustre/lustre/lmv/lproc_lmv.c b/drivers/staging/lustre/lustre/lmv/lproc_lmv.c index b355d01410e4..5d5c3081c467 100644 --- a/drivers/staging/lustre/lustre/lmv/lproc_lmv.c +++ b/drivers/staging/lustre/lustre/lmv/lproc_lmv.c @@ -87,8 +87,9 @@ static int lmv_placement_seq_show(struct seq_file *m, void *v) #define MAX_POLICY_STRING_SIZE 64 -static ssize_t lmv_placement_seq_write(struct file *file, const char *buffer, - size_t count, loff_t *off) +static ssize_t lmv_placement_seq_write(struct file *file, + const char __user *buffer, + size_t count, loff_t *off) { struct obd_device *dev = ((struct seq_file *)file->private_data)->private; char dummy[MAX_POLICY_STRING_SIZE + 1]; diff --git a/drivers/staging/lustre/lustre/lov/lov_ea.c b/drivers/staging/lustre/lustre/lov/lov_ea.c index e6c60151dc65..6f356e025543 100644 --- a/drivers/staging/lustre/lustre/lov/lov_ea.c +++ b/drivers/staging/lustre/lustre/lov/lov_ea.c @@ -98,7 +98,7 @@ struct lov_stripe_md *lsm_alloc_plain(__u16 stripe_count, int *size) OBD_ALLOC_LARGE(lsm, *size); if (!lsm) - return NULL;; + return NULL; for (i = 0; i < stripe_count; i++) { OBD_SLAB_ALLOC_PTR_GFP(loi, lov_oinfo_slab, __GFP_IO); diff --git a/drivers/staging/lustre/lustre/lov/lov_io.c b/drivers/staging/lustre/lustre/lov/lov_io.c index 5a6ab70ed0a1..65133ea308b6 100644 --- a/drivers/staging/lustre/lustre/lov/lov_io.c +++ b/drivers/staging/lustre/lustre/lov/lov_io.c @@ -194,6 +194,7 @@ static int lov_io_sub_init(const struct lu_env *env, struct lov_io *lio, sub_io->ci_lockreq = io->ci_lockreq; sub_io->ci_type = io->ci_type; sub_io->ci_no_srvlock = io->ci_no_srvlock; + sub_io->ci_noatime = io->ci_noatime; lov_sub_enter(sub); result = cl_io_sub_init(sub->sub_env, sub_io, diff --git a/drivers/staging/lustre/lustre/lov/lov_obd.c b/drivers/staging/lustre/lustre/lov/lov_obd.c index 50a77c5ef69a..02509d0cb106 100644 --- a/drivers/staging/lustre/lustre/lov/lov_obd.c +++ b/drivers/staging/lustre/lustre/lov/lov_obd.c @@ -339,7 +339,7 @@ static int lov_disconnect(struct obd_export *exp) for (i = 0; i < lov->desc.ld_tgt_count; i++) { if (lov->lov_tgts[i] && lov->lov_tgts[i]->ltd_exp) { /* Disconnection is the last we know about an obd */ - lov_del_target(obd, i, 0, lov->lov_tgts[i]->ltd_gen); + lov_del_target(obd, i, NULL, lov->lov_tgts[i]->ltd_gen); } } obd_putref(obd); @@ -644,7 +644,7 @@ out: if (rc) { CERROR("add failed (%d), deleting %s\n", rc, obd_uuid2str(&tgt->ltd_uuid)); - lov_del_target(obd, index, 0, 0); + lov_del_target(obd, index, NULL, 0); } obd_putref(obd); return rc; @@ -768,7 +768,7 @@ void lov_fix_desc(struct lov_desc *desc) int lov_setup(struct obd_device *obd, struct lustre_cfg *lcfg) { - struct lprocfs_static_vars lvars = { 0 }; + struct lprocfs_static_vars lvars = { NULL }; struct lov_desc *desc; struct lov_obd *lov = &obd->u.lov; int rc; diff --git a/drivers/staging/lustre/lustre/lov/lov_object.c b/drivers/staging/lustre/lustre/lov/lov_object.c index df8b5b5b7cf4..d6b2cb45b938 100644 --- a/drivers/staging/lustre/lustre/lov/lov_object.c +++ b/drivers/staging/lustre/lustre/lov/lov_object.c @@ -122,8 +122,8 @@ static struct cl_object *lov_sub_find(const struct lu_env *env, } static int lov_init_sub(const struct lu_env *env, struct lov_object *lov, - struct cl_object *stripe, - struct lov_layout_raid0 *r0, int idx) + struct cl_object *stripe, struct lov_layout_raid0 *r0, + int idx) { struct cl_object_header *hdr; struct cl_object_header *subhdr; @@ -144,7 +144,6 @@ static int lov_init_sub(const struct lu_env *env, struct lov_object *lov, hdr = cl_object_header(lov2cl(lov)); subhdr = cl_object_header(stripe); - parent = subhdr->coh_parent; oinfo = lov->lo_lsm->lsm_oinfo[idx]; CDEBUG(D_INODE, DFID"@%p[%d] -> "DFID"@%p: ostid: "DOSTID @@ -153,8 +152,12 @@ static int lov_init_sub(const struct lu_env *env, struct lov_object *lov, PFID(&hdr->coh_lu.loh_fid), hdr, POSTID(&oinfo->loi_oi), oinfo->loi_ost_idx, oinfo->loi_ost_gen); + /* reuse ->coh_attr_guard to protect coh_parent change */ + spin_lock(&subhdr->coh_attr_guard); + parent = subhdr->coh_parent; if (parent == NULL) { subhdr->coh_parent = hdr; + spin_unlock(&subhdr->coh_attr_guard); subhdr->coh_nesting = hdr->coh_nesting + 1; lu_object_ref_add(&stripe->co_lu, "lov-parent", lov); r0->lo_sub[idx] = cl2lovsub(stripe); @@ -166,6 +169,7 @@ static int lov_init_sub(const struct lu_env *env, struct lov_object *lov, struct lov_object *old_lov; unsigned int mask = D_INODE; + spin_unlock(&subhdr->coh_attr_guard); old_obj = lu_object_locate(&parent->coh_lu, &lov_device_type); LASSERT(old_obj != NULL); old_lov = cl2lov(lu2cl(old_obj)); @@ -306,7 +310,7 @@ static void lov_subobject_kill(const struct lu_env *env, struct lov_object *lov, * ->lo_sub[] slot in lovsub_object_fini() */ if (r0->lo_sub[idx] == los) { waiter = &lov_env_info(env)->lti_waiter; - init_waitqueue_entry_current(waiter); + init_waitqueue_entry(waiter, current); add_wait_queue(&bkt->lsb_marche_funebre, waiter); set_current_state(TASK_UNINTERRUPTIBLE); while (1) { @@ -316,7 +320,7 @@ static void lov_subobject_kill(const struct lu_env *env, struct lov_object *lov, spin_lock(&r0->lo_sub_lock); if (r0->lo_sub[idx] == los) { spin_unlock(&r0->lo_sub_lock); - waitq_wait(waiter, TASK_UNINTERRUPTIBLE); + schedule(); } else { spin_unlock(&r0->lo_sub_lock); set_current_state(TASK_RUNNING); @@ -508,7 +512,7 @@ static int lov_attr_get_raid0(const struct lu_env *env, struct cl_object *obj, return result; } -const static struct lov_layout_operations lov_dispatch[] = { +static const struct lov_layout_operations lov_dispatch[] = { [LLT_EMPTY] = { .llo_init = lov_init_empty, .llo_delete = lov_delete_empty, diff --git a/drivers/staging/lustre/lustre/lov/lov_pack.c b/drivers/staging/lustre/lustre/lov/lov_pack.c index 27ed27e6fa6a..74200cf1b331 100644 --- a/drivers/staging/lustre/lustre/lov/lov_pack.c +++ b/drivers/staging/lustre/lustre/lov/lov_pack.c @@ -339,7 +339,8 @@ int lov_free_memmd(struct lov_stripe_md **lsmp) *lsmp = NULL; LASSERT(atomic_read(&lsm->lsm_refc) > 0); - if ((refc = atomic_dec_return(&lsm->lsm_refc)) == 0) { + refc = atomic_dec_return(&lsm->lsm_refc); + if (refc == 0) { LASSERT(lsm_op_find(lsm->lsm_magic) != NULL); lsm_op_find(lsm->lsm_magic)->lsm_free(lsm); } diff --git a/drivers/staging/lustre/lustre/lov/lov_request.c b/drivers/staging/lustre/lustre/lov/lov_request.c index ca81cac9041c..a5481d7eb5d6 100644 --- a/drivers/staging/lustre/lustre/lov/lov_request.c +++ b/drivers/staging/lustre/lustre/lov/lov_request.c @@ -50,7 +50,7 @@ static void lov_init_set(struct lov_request_set *set) atomic_set(&set->set_completes, 0); atomic_set(&set->set_success, 0); atomic_set(&set->set_finish_checked, 0); - set->set_cookies = 0; + set->set_cookies = NULL; INIT_LIST_HEAD(&set->set_list); atomic_set(&set->set_refcount, 1); init_waitqueue_head(&set->set_waitq); diff --git a/drivers/staging/lustre/lustre/lov/lovsub_dev.c b/drivers/staging/lustre/lustre/lov/lovsub_dev.c index 998ea1cbc7bb..926c35a25ceb 100644 --- a/drivers/staging/lustre/lustre/lov/lovsub_dev.c +++ b/drivers/staging/lustre/lustre/lov/lovsub_dev.c @@ -131,6 +131,10 @@ static struct lu_device *lovsub_device_free(const struct lu_env *env, struct lovsub_device *lsd = lu2lovsub_dev(d); struct lu_device *next = cl2lu_dev(lsd->acid_next); + if (atomic_read(&d->ld_ref) && d->ld_site) { + LIBCFS_DEBUG_MSG_DATA_DECL(msgdata, D_ERROR, NULL); + lu_site_print(env, d->ld_site, &msgdata, lu_cdebug_printer); + } cl_device_fini(lu2cl_dev(d)); OBD_FREE_PTR(lsd); return next; diff --git a/drivers/staging/lustre/lustre/lvfs/lvfs_linux.c b/drivers/staging/lustre/lustre/lvfs/lvfs_linux.c index 428ffd8c37b7..e44b7a532de7 100644 --- a/drivers/staging/lustre/lustre/lvfs/lvfs_linux.c +++ b/drivers/staging/lustre/lustre/lvfs/lvfs_linux.c @@ -55,7 +55,7 @@ struct lprocfs_stats *obd_memory = NULL; EXPORT_SYMBOL(obd_memory); -/* refine later and change to seqlock or simlar from libcfs */ +/* refine later and change to seqlock or similar from libcfs */ /* Debugging check only needed during development */ #ifdef OBD_CTXT_DEBUG diff --git a/drivers/staging/lustre/lustre/mdc/mdc_internal.h b/drivers/staging/lustre/lustre/mdc/mdc_internal.h index 506982996c0e..c78bf003c2c5 100644 --- a/drivers/staging/lustre/lustre/mdc/mdc_internal.h +++ b/drivers/staging/lustre/lustre/mdc/mdc_internal.h @@ -101,7 +101,7 @@ int mdc_enqueue(struct obd_export *exp, struct ldlm_enqueue_info *einfo, struct lustre_handle *lockh, void *lmm, int lmmsize, struct ptlrpc_request **req, __u64 extra_lock_flags); -int mdc_resource_get_unused(struct obd_export *exp, struct lu_fid *fid, +int mdc_resource_get_unused(struct obd_export *exp, const struct lu_fid *fid, struct list_head *cancels, ldlm_mode_t mode, __u64 bits); /* mdc/mdc_request.c */ @@ -122,7 +122,7 @@ int mdc_free_lustre_md(struct obd_export *exp, struct lustre_md *md); int mdc_set_open_replay_data(struct obd_export *exp, struct obd_client_handle *och, - struct ptlrpc_request *open_req); + struct lookup_intent *it); int mdc_clear_open_replay_data(struct obd_export *exp, struct obd_client_handle *och); diff --git a/drivers/staging/lustre/lustre/mdc/mdc_lib.c b/drivers/staging/lustre/lustre/mdc/mdc_lib.c index 91f6876dac3f..5b9f37141512 100644 --- a/drivers/staging/lustre/lustre/mdc/mdc_lib.c +++ b/drivers/staging/lustre/lustre/mdc/mdc_lib.c @@ -197,7 +197,7 @@ static __u64 mds_pack_open_flags(__u64 flags, __u32 mode) if (flags & FMODE_EXEC) cr_flags |= MDS_FMODE_EXEC; #endif - if (flags & O_LOV_DELAY_CREATE) + if (cl_is_lov_delay_create(flags)) cr_flags |= MDS_OPEN_DELAY_CREATE; if (flags & O_NONBLOCK) diff --git a/drivers/staging/lustre/lustre/mdc/mdc_locks.c b/drivers/staging/lustre/lustre/mdc/mdc_locks.c index 8aa7c80c2002..53022ec390f0 100644 --- a/drivers/staging/lustre/lustre/mdc/mdc_locks.c +++ b/drivers/staging/lustre/lustre/mdc/mdc_locks.c @@ -37,15 +37,15 @@ #define DEBUG_SUBSYSTEM S_MDC # include <linux/module.h> -# include <linux/pagemap.h> -# include <linux/miscdevice.h> -#include <lustre_acl.h> +#include <linux/lustre_intent.h> +#include <obd.h> #include <obd_class.h> #include <lustre_dlm.h> -/* fid_res_name_eq() */ -#include <lustre_fid.h> -#include <lprocfs_status.h> +#include <lustre_fid.h> /* fid_res_name_eq() */ +#include <lustre_mdc.h> +#include <lustre_net.h> +#include <lustre_req_layout.h> #include "mdc_internal.h" struct mdc_getattr_args { @@ -121,7 +121,7 @@ int mdc_set_lock_data(struct obd_export *exp, __u64 *lockh, void *data, struct ldlm_lock *lock; struct inode *new_inode = data; - if(bits) + if (bits) *bits = 0; if (!*lockh) @@ -160,6 +160,8 @@ ldlm_mode_t mdc_lock_match(struct obd_export *exp, __u64 flags, ldlm_mode_t rc; fid_build_reg_res_name(fid, &res_id); + /* LU-4405: Clear bits not supported by server */ + policy->l_inodebits.bits &= exp_connect_ibits(exp); rc = ldlm_lock_match(class_exp2obd(exp)->obd_namespace, flags, &res_id, type, policy, mode, lockh, 0); return rc; @@ -194,7 +196,7 @@ int mdc_null_inode(struct obd_export *exp, fid_build_reg_res_name(fid, &res_id); res = ldlm_resource_get(ns, NULL, &res_id, 0, 0); - if(res == NULL) + if (res == NULL) return 0; lock_res(res); @@ -334,9 +336,9 @@ static struct ptlrpc_request *mdc_intent_open_pack(struct obd_export *exp, max(lmmsize, obddev->u.cli.cl_default_mds_easize)); rc = ldlm_prep_enqueue_req(exp, req, &cancels, count); - if (rc) { + if (rc < 0) { ptlrpc_request_free(req); - return NULL; + return ERR_PTR(rc); } spin_lock(&req->rq_lock); @@ -378,13 +380,6 @@ mdc_intent_getxattr_pack(struct obd_export *exp, mdc_set_capa_size(req, &RMF_CAPA1, op_data->op_capa1); - if (it->it_op == IT_SETXATTR) - /* If we want to upgrade to LCK_PW, let's cancel LCK_PR - * locks now. This avoids unnecessary ASTs. */ - count = mdc_resource_get_unused(exp, &op_data->op_fid1, - &cancels, LCK_PW, - MDS_INODELOCK_XATTR); - rc = ldlm_prep_enqueue_req(exp, req, &cancels, count); if (rc) { ptlrpc_request_free(req); @@ -646,7 +641,7 @@ static int mdc_finish_enqueue(struct obd_export *exp, * happens immediately after swabbing below, new reply * is swabbed by that handler correctly. */ - mdc_set_open_replay_data(NULL, NULL, req); + mdc_set_open_replay_data(NULL, NULL, it); } if ((body->valid & (OBD_MD_FLDIREA | OBD_MD_FLEASIZE)) != 0) { @@ -758,6 +753,7 @@ static int mdc_finish_enqueue(struct obd_export *exp, /* install lvb_data */ lock_res_and_lock(lock); if (lock->l_lvb_data == NULL) { + lock->l_lvb_type = LVB_T_LAYOUT; lock->l_lvb_data = lmm; lock->l_lvb_len = lvb_len; lmm = NULL; @@ -842,7 +838,7 @@ resend: return -EOPNOTSUPP; req = mdc_intent_layout_pack(exp, it, op_data); lvb_type = LVB_T_LAYOUT; - } else if (it->it_op & (IT_GETXATTR | IT_SETXATTR)) { + } else if (it->it_op & IT_GETXATTR) { req = mdc_intent_getxattr_pack(exp, it, op_data); } else { LBUG(); @@ -880,7 +876,7 @@ resend: rc = ldlm_cli_enqueue(exp, &req, einfo, &res_id, policy, &flags, NULL, 0, lvb_type, lockh, 0); if (!it) { - /* For flock requests we immediatelly return without further + /* For flock requests we immediately return without further delay and let caller deal with the rest, since rest of this function metadata processing makes no sense for flock requests anyway. But in case of problem during comms with @@ -973,7 +969,6 @@ static int mdc_finish_intent_lock(struct obd_export *exp, if (fid_is_sane(&op_data->op_fid2) && it->it_create_mode & M_CHECK_STALE && it->it_op != IT_GETATTR) { - it_set_disposition(it, DISP_ENQ_COMPLETE); /* Also: did we find the same inode? */ /* sever can return one of two fids: @@ -1068,7 +1063,23 @@ int mdc_revalidate_lock(struct obd_export *exp, struct lookup_intent *it, fid_build_reg_res_name(fid, &res_id); switch (it->it_op) { case IT_GETATTR: - policy.l_inodebits.bits = MDS_INODELOCK_UPDATE; + /* File attributes are held under multiple bits: + * nlink is under lookup lock, size and times are + * under UPDATE lock and recently we've also got + * a separate permissions lock for owner/group/acl that + * were protected by lookup lock before. + * Getattr must provide all of that information, + * so we need to ensure we have all of those locks. + * Unfortunately, if the bits are split across multiple + * locks, there's no easy way to match all of them here, + * so an extra RPC would be performed to fetch all + * of those bits at once for now. */ + /* For new MDTs(> 2.4), UPDATE|PERM should be enough, + * but for old MDTs (< 2.4), permission is covered + * by LOOKUP lock, so it needs to match all bits here.*/ + policy.l_inodebits.bits = MDS_INODELOCK_UPDATE | + MDS_INODELOCK_LOOKUP | + MDS_INODELOCK_PERM; break; case IT_LAYOUT: policy.l_inodebits.bits = MDS_INODELOCK_LAYOUT; @@ -1077,10 +1088,11 @@ int mdc_revalidate_lock(struct obd_export *exp, struct lookup_intent *it, policy.l_inodebits.bits = MDS_INODELOCK_LOOKUP; break; } - mode = ldlm_lock_match(exp->exp_obd->obd_namespace, - LDLM_FL_BLOCK_GRANTED, &res_id, + + mode = mdc_lock_match(exp, LDLM_FL_BLOCK_GRANTED, fid, LDLM_IBITS, &policy, - LCK_CR|LCK_CW|LCK_PR|LCK_PW, &lockh, 0); + LCK_CR | LCK_CW | LCK_PR | LCK_PW, + &lockh); } if (mode) { @@ -1127,6 +1139,12 @@ int mdc_intent_lock(struct obd_export *exp, struct md_op_data *op_data, ldlm_blocking_callback cb_blocking, __u64 extra_lock_flags) { + struct ldlm_enqueue_info einfo = { + .ei_type = LDLM_IBITS, + .ei_mode = it_to_lock_mode(it), + .ei_cb_bl = cb_blocking, + .ei_cb_cp = ldlm_completion_ast, + }; struct lustre_handle lockh; int rc = 0; @@ -1152,42 +1170,19 @@ int mdc_intent_lock(struct obd_export *exp, struct md_op_data *op_data, return rc; } - /* lookup_it may be called only after revalidate_it has run, because - * revalidate_it cannot return errors, only zero. Returning zero causes - * this call to lookup, which *can* return an error. - * - * We only want to execute the request associated with the intent one - * time, however, so don't send the request again. Instead, skip past - * this and use the request from revalidate. In this case, revalidate - * never dropped its reference, so the refcounts are all OK */ - if (!it_disposition(it, DISP_ENQ_COMPLETE)) { - struct ldlm_enqueue_info einfo = { - .ei_type = LDLM_IBITS, - .ei_mode = it_to_lock_mode(it), - .ei_cb_bl = cb_blocking, - .ei_cb_cp = ldlm_completion_ast, - }; - - /* For case if upper layer did not alloc fid, do it now. */ - if (!fid_is_sane(&op_data->op_fid2) && it->it_op & IT_CREAT) { - rc = mdc_fid_alloc(exp, &op_data->op_fid2, op_data); - if (rc < 0) { - CERROR("Can't alloc new fid, rc %d\n", rc); - return rc; - } - } - rc = mdc_enqueue(exp, &einfo, it, op_data, &lockh, - lmm, lmmsize, NULL, extra_lock_flags); - if (rc < 0) + /* For case if upper layer did not alloc fid, do it now. */ + if (!fid_is_sane(&op_data->op_fid2) && it->it_op & IT_CREAT) { + rc = mdc_fid_alloc(exp, &op_data->op_fid2, op_data); + if (rc < 0) { + CERROR("Can't alloc new fid, rc %d\n", rc); return rc; - } else if (!fid_is_sane(&op_data->op_fid2) || - !(it->it_create_mode & M_CHECK_STALE)) { - /* DISP_ENQ_COMPLETE set means there is extra reference on - * request referenced from this intent, saved for subsequent - * lookup. This path is executed when we proceed to this - * lookup, so we clear DISP_ENQ_COMPLETE */ - it_clear_disposition(it, DISP_ENQ_COMPLETE); + } } + rc = mdc_enqueue(exp, &einfo, it, op_data, &lockh, lmm, lmmsize, NULL, + extra_lock_flags); + if (rc < 0) + return rc; + *reqp = it->d.lustre.it_data; rc = mdc_finish_intent_lock(exp, *reqp, op_data, it, &lockh); return rc; @@ -1269,8 +1264,8 @@ int mdc_intent_getattr_async(struct obd_export *exp, fid_build_reg_res_name(&op_data->op_fid1, &res_id); req = mdc_intent_getattr_pack(exp, it, op_data); - if (!req) - return -ENOMEM; + if (IS_ERR(req)) + return PTR_ERR(req); rc = mdc_enter_request(&obddev->u.cli); if (rc != 0) { diff --git a/drivers/staging/lustre/lustre/mdc/mdc_reint.c b/drivers/staging/lustre/lustre/mdc/mdc_reint.c index 9f3a345f34e0..d79aa1641fef 100644 --- a/drivers/staging/lustre/lustre/mdc/mdc_reint.c +++ b/drivers/staging/lustre/lustre/mdc/mdc_reint.c @@ -66,7 +66,7 @@ static int mdc_reint(struct ptlrpc_request *request, /* Find and cancel locally locks matched by inode @bits & @mode in the resource * found by @fid. Found locks are added into @cancel list. Returns the amount of * locks added to @cancels list. */ -int mdc_resource_get_unused(struct obd_export *exp, struct lu_fid *fid, +int mdc_resource_get_unused(struct obd_export *exp, const struct lu_fid *fid, struct list_head *cancels, ldlm_mode_t mode, __u64 bits) { @@ -165,6 +165,7 @@ int mdc_setattr(struct obd_export *exp, struct md_op_data *op_data, req->rq_cb_data = *mod; (*mod)->mod_open_req = req; req->rq_commit_cb = mdc_commit_open; + (*mod)->mod_is_create = true; /** * Take an extra reference on \var mod, it protects \var * mod from being freed on eviction (commit callback is diff --git a/drivers/staging/lustre/lustre/mdc/mdc_request.c b/drivers/staging/lustre/lustre/mdc/mdc_request.c index d1ad91c34ddc..bde9f93c149b 100644 --- a/drivers/staging/lustre/lustre/mdc/mdc_request.c +++ b/drivers/staging/lustre/lustre/mdc/mdc_request.c @@ -355,10 +355,32 @@ static int mdc_xattr_common(struct obd_export *exp,const struct req_format *fmt, input_size); } - rc = ptlrpc_request_pack(req, LUSTRE_MDS_VERSION, opcode); - if (rc) { - ptlrpc_request_free(req); - return rc; + /* Flush local XATTR locks to get rid of a possible cancel RPC */ + if (opcode == MDS_REINT && fid_is_sane(fid) && + exp->exp_connect_data.ocd_ibits_known & MDS_INODELOCK_XATTR) { + LIST_HEAD(cancels); + int count; + + /* Without that packing would fail */ + if (input_size == 0) + req_capsule_set_size(&req->rq_pill, &RMF_EADATA, + RCL_CLIENT, 0); + + count = mdc_resource_get_unused(exp, fid, + &cancels, LCK_EX, + MDS_INODELOCK_XATTR); + + rc = mdc_prep_elc_req(exp, req, MDS_REINT, &cancels, count); + if (rc) { + ptlrpc_request_free(req); + return rc; + } + } else { + rc = ptlrpc_request_pack(req, LUSTRE_MDS_VERSION, opcode); + if (rc) { + ptlrpc_request_free(req); + return rc; + } } if (opcode == MDS_REINT) { @@ -700,11 +722,12 @@ void mdc_commit_open(struct ptlrpc_request *req) int mdc_set_open_replay_data(struct obd_export *exp, struct obd_client_handle *och, - struct ptlrpc_request *open_req) + struct lookup_intent *it) { struct md_open_data *mod; struct mdt_rec_create *rec; struct mdt_body *body; + struct ptlrpc_request *open_req = it->d.lustre.it_data; struct obd_import *imp = open_req->rq_import; if (!open_req->rq_replay) @@ -738,6 +761,8 @@ int mdc_set_open_replay_data(struct obd_export *exp, spin_lock(&open_req->rq_lock); och->och_mod = mod; mod->mod_och = och; + mod->mod_is_create = it_disposition(it, DISP_OPEN_CREATE) || + it_disposition(it, DISP_OPEN_STRIPE); mod->mod_open_req = open_req; open_req->rq_cb_data = mod; open_req->rq_commit_cb = mdc_commit_open; @@ -758,6 +783,23 @@ int mdc_set_open_replay_data(struct obd_export *exp, return 0; } +static void mdc_free_open(struct md_open_data *mod) +{ + int committed = 0; + + if (mod->mod_is_create == 0 && + imp_connect_disp_stripe(mod->mod_open_req->rq_import)) + committed = 1; + + LASSERT(mod->mod_open_req->rq_replay == 0); + + DEBUG_REQ(D_RPCTRACE, mod->mod_open_req, "free open request\n"); + + ptlrpc_request_committed(mod->mod_open_req, committed); + if (mod->mod_close_req) + ptlrpc_request_committed(mod->mod_close_req, committed); +} + int mdc_clear_open_replay_data(struct obd_export *exp, struct obd_client_handle *och) { @@ -771,6 +813,8 @@ int mdc_clear_open_replay_data(struct obd_export *exp, return 0; LASSERT(mod != LP_POISON); + LASSERT(mod->mod_open_req != NULL); + mdc_free_open(mod); mod->mod_och = NULL; och->och_mod = NULL; @@ -969,6 +1013,9 @@ int mdc_done_writing(struct obd_export *exp, struct md_op_data *op_data, if (mod) { if (rc != 0) mod->mod_close_req = NULL; + LASSERT(mod->mod_open_req != NULL); + mdc_free_open(mod); + /* Since now, mod is accessed through setattr req only, * thus DW req does not keep a reference on mod anymore. */ obd_mod_put(mod); @@ -1430,7 +1477,7 @@ static struct kuc_hdr *changelog_kuc_hdr(char *buf, int len, int flags) { struct kuc_hdr *lh = (struct kuc_hdr *)buf; - LASSERT(len <= CR_MAXSIZE); + LASSERT(len <= KUC_CHANGELOG_MSG_MAXSIZE); lh->kuc_magic = KUC_MAGIC; lh->kuc_transport = KUC_TRANSPORT_CHANGELOG; @@ -1503,7 +1550,7 @@ static int mdc_changelog_send_thread(void *csdata) CDEBUG(D_CHANGELOG, "changelog to fp=%p start "LPU64"\n", cs->cs_fp, cs->cs_startrec); - OBD_ALLOC(cs->cs_buf, CR_MAXSIZE); + OBD_ALLOC(cs->cs_buf, KUC_CHANGELOG_MSG_MAXSIZE); if (cs->cs_buf == NULL) GOTO(out, rc = -ENOMEM); @@ -1527,8 +1574,8 @@ static int mdc_changelog_send_thread(void *csdata) rc = llog_cat_process(NULL, llh, changelog_kkuc_cb, cs, 0, 0); /* Send EOF no matter what our result */ - if ((kuch = changelog_kuc_hdr(cs->cs_buf, sizeof(*kuch), - cs->cs_flags))) { + kuch = changelog_kuc_hdr(cs->cs_buf, sizeof(*kuch), cs->cs_flags); + if (kuch) { kuch->kuc_msgtype = CL_EOF; libcfs_kkuc_msg_put(cs->cs_fp, kuch); } @@ -1540,7 +1587,7 @@ out: if (ctxt) llog_ctxt_put(ctxt); if (cs->cs_buf) - OBD_FREE(cs->cs_buf, CR_MAXSIZE); + OBD_FREE(cs->cs_buf, KUC_CHANGELOG_MSG_MAXSIZE); OBD_FREE_PTR(cs); return rc; } @@ -1650,11 +1697,16 @@ static int mdc_quotactl(struct obd_device *unused, struct obd_export *exp, if (rc) CERROR("ptlrpc_queue_wait failed, rc: %d\n", rc); - if (req->rq_repmsg && - (oqc = req_capsule_server_get(&req->rq_pill, &RMF_OBD_QUOTACTL))) { - *oqctl = *oqc; + if (req->rq_repmsg) { + oqc = req_capsule_server_get(&req->rq_pill, &RMF_OBD_QUOTACTL); + if (oqc) { + *oqctl = *oqc; + } else if (!rc) { + CERROR ("Can't unpack obd_quotactl\n"); + rc = -EPROTO; + } } else if (!rc) { - CERROR ("Can't unpack obd_quotactl\n"); + CERROR("Can't unpack obd_quotactl\n"); rc = -EPROTO; } ptlrpc_req_finished(req); diff --git a/drivers/staging/lustre/lustre/mgc/mgc_request.c b/drivers/staging/lustre/lustre/mgc/mgc_request.c index 3bdbb94e020f..de9fb1433edd 100644 --- a/drivers/staging/lustre/lustre/mgc/mgc_request.c +++ b/drivers/staging/lustre/lustre/mgc/mgc_request.c @@ -56,7 +56,7 @@ static int mgc_name2resid(char *name, int len, struct ldlm_res_id *res_id, { __u64 resname = 0; - if (len > 8) { + if (len > sizeof(resname)) { CERROR("name too long: %s\n", name); return -EINVAL; } @@ -76,6 +76,7 @@ static int mgc_name2resid(char *name, int len, struct ldlm_res_id *res_id, resname = 0; break; case CONFIG_T_RECOVER: + case CONFIG_T_PARAMS: resname = type; break; default: @@ -101,10 +102,13 @@ int mgc_logname2resid(char *logname, struct ldlm_res_id *res_id, int type) int len; /* logname consists of "fsname-nodetype". - * e.g. "lustre-MDT0001", "SUN-000-client" */ + * e.g. "lustre-MDT0001", "SUN-000-client" + * there is an exception: llog "params" */ name_end = strrchr(logname, '-'); - LASSERT(name_end); - len = name_end - logname; + if (!name_end) + len = strlen(logname); + else + len = name_end - logname; return mgc_name2resid(logname, len, res_id, type); } @@ -140,6 +144,8 @@ static void config_log_put(struct config_llog_data *cld) config_log_put(cld->cld_recover); if (cld->cld_sptlrpc) config_log_put(cld->cld_sptlrpc); + if (cld->cld_params) + config_log_put(cld->cld_params); if (cld_is_sptlrpc(cld)) sptlrpc_conf_log_stop(cld->cld_logname); @@ -271,6 +277,19 @@ static struct config_llog_data *config_recover_log_add(struct obd_device *obd, return cld; } +static struct config_llog_data *config_params_log_add(struct obd_device *obd, + struct config_llog_instance *cfg, struct super_block *sb) +{ + struct config_llog_instance lcfg = *cfg; + struct config_llog_data *cld; + + lcfg.cfg_instance = sb; + + cld = do_config_log_add(obd, PARAMS_FILENAME, CONFIG_T_PARAMS, + &lcfg, sb); + + return cld; +} /** Add this log to the list of active logs watched by an MGC. * Active means we're watching for updates. @@ -284,8 +303,10 @@ static int config_log_add(struct obd_device *obd, char *logname, struct lustre_sb_info *lsi = s2lsi(sb); struct config_llog_data *cld; struct config_llog_data *sptlrpc_cld; - char seclogname[32]; - char *ptr; + struct config_llog_data *params_cld; + char seclogname[32]; + char *ptr; + int rc; CDEBUG(D_MGC, "adding config log %s:%p\n", logname, cfg->cfg_instance); @@ -308,32 +329,49 @@ static int config_log_add(struct obd_device *obd, char *logname, CONFIG_T_SPTLRPC, NULL, NULL); if (IS_ERR(sptlrpc_cld)) { CERROR("can't create sptlrpc log: %s\n", seclogname); - return PTR_ERR(sptlrpc_cld); + GOTO(out_err, rc = PTR_ERR(sptlrpc_cld)); } } + params_cld = config_params_log_add(obd, cfg, sb); + if (IS_ERR(params_cld)) { + rc = PTR_ERR(params_cld); + CERROR("%s: can't create params log: rc = %d\n", + obd->obd_name, rc); + GOTO(out_err1, rc); + } cld = do_config_log_add(obd, logname, CONFIG_T_CONFIG, cfg, sb); if (IS_ERR(cld)) { CERROR("can't create log: %s\n", logname); - config_log_put(sptlrpc_cld); - return PTR_ERR(cld); + GOTO(out_err2, rc = PTR_ERR(cld)); } cld->cld_sptlrpc = sptlrpc_cld; + cld->cld_params = params_cld; LASSERT(lsi->lsi_lmd); if (!(lsi->lsi_lmd->lmd_flags & LMD_FLG_NOIR)) { struct config_llog_data *recover_cld; *strrchr(seclogname, '-') = 0; recover_cld = config_recover_log_add(obd, seclogname, cfg, sb); - if (IS_ERR(recover_cld)) { - config_log_put(cld); - return PTR_ERR(recover_cld); - } + if (IS_ERR(recover_cld)) + GOTO(out_err3, rc = PTR_ERR(recover_cld)); cld->cld_recover = recover_cld; } return 0; + +out_err3: + config_log_put(cld); + +out_err2: + config_log_put(params_cld); + +out_err1: + config_log_put(sptlrpc_cld); + +out_err: + return rc; } DEFINE_MUTEX(llog_process_lock); @@ -344,6 +382,7 @@ static int config_log_end(char *logname, struct config_llog_instance *cfg) { struct config_llog_data *cld; struct config_llog_data *cld_sptlrpc = NULL; + struct config_llog_data *cld_params = NULL; struct config_llog_data *cld_recover = NULL; int rc = 0; @@ -382,11 +421,20 @@ static int config_log_end(char *logname, struct config_llog_instance *cfg) spin_lock(&config_list_lock); cld_sptlrpc = cld->cld_sptlrpc; cld->cld_sptlrpc = NULL; + cld_params = cld->cld_params; + cld->cld_params = NULL; spin_unlock(&config_list_lock); if (cld_sptlrpc) config_log_put(cld_sptlrpc); + if (cld_params) { + mutex_lock(&cld_params->cld_lock); + cld_params->cld_stopping = 1; + mutex_unlock(&cld_params->cld_lock); + config_log_put(cld_params); + } + /* drop the ref from the find */ config_log_put(cld); /* drop the start ref */ @@ -1584,7 +1632,7 @@ static int mgc_llog_local_copy(const struct lu_env *env, /* * - copy it to backup using llog_backup() * - copy remote llog to logname using llog_backup() - * - if failed then move bakup to logname again + * - if failed then move backup to logname again */ OBD_ALLOC(temp_log, strlen(logname) + 1); @@ -1664,7 +1712,7 @@ static int mgc_process_cfg_log(struct obd_device *mgc, LCONSOLE_ERROR_MSG(0x13a, "Failed to get MGS log %s and no local copy.\n", cld->cld_logname); - GOTO(out_pop, rc = -ENOTCONN); + GOTO(out_pop, rc = -ENOENT); } CDEBUG(D_MGC, "Failed to get MGS log %s, using local copy for now, will try to update later.\n", @@ -1863,6 +1911,20 @@ static int mgc_process_config(struct obd_device *obd, obd_count len, void *buf) if (rc) CERROR("Cannot process recover llog %d\n", rc); } + + if (rc == 0 && cld->cld_params != NULL) { + rc = mgc_process_log(obd, cld->cld_params); + if (rc == -ENOENT) { + CDEBUG(D_MGC, + "There is no params config file yet\n"); + rc = 0; + } + /* params log is optional */ + if (rc) + CERROR( + "%s: can't process params llog: rc = %d\n", + obd->obd_name, rc); + } config_log_put(cld); break; diff --git a/drivers/staging/lustre/lustre/obdclass/cl_io.c b/drivers/staging/lustre/lustre/obdclass/cl_io.c index e048500edd60..3bebc78e7673 100644 --- a/drivers/staging/lustre/lustre/obdclass/cl_io.c +++ b/drivers/staging/lustre/lustre/obdclass/cl_io.c @@ -942,7 +942,7 @@ int cl_io_cancel(const struct lu_env *env, struct cl_io *io, struct cl_page *page; int result = 0; - CERROR("Canceling ongoing page trasmission\n"); + CERROR("Canceling ongoing page transmission\n"); cl_page_list_for_each(page, queue) { int rc; diff --git a/drivers/staging/lustre/lustre/obdclass/cl_lock.c b/drivers/staging/lustre/lustre/obdclass/cl_lock.c index 749eb082f979..d795cef3f164 100644 --- a/drivers/staging/lustre/lustre/obdclass/cl_lock.c +++ b/drivers/staging/lustre/lustre/obdclass/cl_lock.c @@ -932,7 +932,7 @@ int cl_lock_state_wait(const struct lu_env *env, struct cl_lock *lock) * LU-305 */ blocked = cfs_block_sigsinv(LUSTRE_FATAL_SIGS); - init_waitqueue_entry_current(&waiter); + init_waitqueue_entry(&waiter, current); add_wait_queue(&lock->cll_wq, &waiter); set_current_state(TASK_INTERRUPTIBLE); cl_lock_mutex_put(env, lock); @@ -943,7 +943,7 @@ int cl_lock_state_wait(const struct lu_env *env, struct cl_lock *lock) * can be restarted if signals are pending here */ result = -ERESTARTSYS; if (likely(!OBD_FAIL_CHECK(OBD_FAIL_LOCK_STATE_WAIT_INTR))) { - waitq_wait(&waiter, TASK_INTERRUPTIBLE); + schedule(); if (!cfs_signal_pending()) result = 0; } diff --git a/drivers/staging/lustre/lustre/obdclass/cl_object.c b/drivers/staging/lustre/lustre/obdclass/cl_object.c index 1a926036724b..0fc256f59e92 100644 --- a/drivers/staging/lustre/lustre/obdclass/cl_object.c +++ b/drivers/staging/lustre/lustre/obdclass/cl_object.c @@ -508,7 +508,7 @@ EXPORT_SYMBOL(cl_site_stats_print); * about journal_info. Currently following fields in task_struct are identified * can be used for this purpose: * - cl_env: for liblustre. - * - tux_info: ony on RedHat kernel. + * - tux_info: only on RedHat kernel. * - ... * \note As long as we use task_struct to store cl_env, we assume that once * called into Lustre, we'll never call into the other part of the kernel diff --git a/drivers/staging/lustre/lustre/obdclass/genops.c b/drivers/staging/lustre/lustre/obdclass/genops.c index d9f750d42c80..f2bdea33041d 100644 --- a/drivers/staging/lustre/lustre/obdclass/genops.c +++ b/drivers/staging/lustre/lustre/obdclass/genops.c @@ -1010,6 +1010,8 @@ struct obd_import *class_new_import(struct obd_device *obd) INIT_LIST_HEAD(&imp->imp_replay_list); INIT_LIST_HEAD(&imp->imp_sending_list); INIT_LIST_HEAD(&imp->imp_delayed_list); + INIT_LIST_HEAD(&imp->imp_committed_list); + imp->imp_replay_cursor = &imp->imp_committed_list; spin_lock_init(&imp->imp_lock); imp->imp_last_success_conn = 0; imp->imp_state = LUSTRE_IMP_NEW; @@ -1532,8 +1534,8 @@ void obd_exports_barrier(struct obd_device *obd) spin_lock(&obd->obd_dev_lock); while (!list_empty(&obd->obd_unlinked_exports)) { spin_unlock(&obd->obd_dev_lock); - schedule_timeout_and_set_state(TASK_UNINTERRUPTIBLE, - cfs_time_seconds(waited)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(waited)); if (waited > 5 && IS_PO2(waited)) { LCONSOLE_WARN("%s is waiting for obd_unlinked_exports " "more than %d seconds. " @@ -1625,7 +1627,7 @@ static int obd_zombie_impexp_check(void *arg) } /** - * Add export to the obd_zombe thread and notify it. + * Add export to the obd_zombie thread and notify it. */ static void obd_zombie_export_add(struct obd_export *exp) { spin_lock(&exp->exp_obd->obd_dev_lock); @@ -1641,7 +1643,7 @@ static void obd_zombie_export_add(struct obd_export *exp) { } /** - * Add import to the obd_zombe thread and notify it. + * Add import to the obd_zombie thread and notify it. */ static void obd_zombie_import_add(struct obd_import *imp) { LASSERT(imp->imp_sec == NULL); @@ -1661,7 +1663,7 @@ static void obd_zombie_import_add(struct obd_import *imp) { static void obd_zombie_impexp_notify(void) { /* - * Make sure obd_zomebie_impexp_thread get this notification. + * Make sure obd_zombie_impexp_thread get this notification. * It is possible this signal only get by obd_zombie_barrier, and * barrier gulps this notification and sleeps away and hangs ensues */ diff --git a/drivers/staging/lustre/lustre/obdclass/linux/linux-module.c b/drivers/staging/lustre/lustre/obdclass/linux/linux-module.c index 121a856d5052..ba20776ebfa1 100644 --- a/drivers/staging/lustre/lustre/obdclass/linux/linux-module.c +++ b/drivers/staging/lustre/lustre/obdclass/linux/linux-module.c @@ -184,7 +184,7 @@ static long obd_class_ioctl(struct file *filp, unsigned int cmd, int err = 0; /* Allow non-root access for OBD_IOC_PING_TARGET - used by lfs check */ - if (!cfs_capable(CFS_CAP_SYS_ADMIN) && (cmd != OBD_IOC_PING_TARGET)) + if (!capable(CFS_CAP_SYS_ADMIN) && (cmd != OBD_IOC_PING_TARGET)) return err = -EACCES; if ((cmd & 0xffffff00) == ((int)'T') << 8) /* ignore all tty ioctls */ return err = -ENOTTY; diff --git a/drivers/staging/lustre/lustre/obdclass/llog_cat.c b/drivers/staging/lustre/lustre/obdclass/llog_cat.c index c0f3af725747..1d999310ec92 100644 --- a/drivers/staging/lustre/lustre/obdclass/llog_cat.c +++ b/drivers/staging/lustre/lustre/obdclass/llog_cat.c @@ -551,9 +551,8 @@ int llog_cat_process_cb(const struct lu_env *env, struct llog_handle *cat_llh, if (rec->lrh_index < d->lpd_startcat) /* Skip processing of the logs until startcat */ - return 0; - - if (d->lpd_startidx > 0) { + rc = 0; + else if (d->lpd_startidx > 0) { struct llog_process_cat_data cd; cd.lpcd_first_idx = d->lpd_startidx; @@ -566,6 +565,7 @@ int llog_cat_process_cb(const struct lu_env *env, struct llog_handle *cat_llh, rc = llog_process_or_fork(env, llh, d->lpd_cb, d->lpd_data, NULL, false); } + llog_handle_put(llh); return rc; diff --git a/drivers/staging/lustre/lustre/obdclass/llog_lvfs.c b/drivers/staging/lustre/lustre/obdclass/llog_lvfs.c index 5385d8e658cf..d86bb8c60354 100644 --- a/drivers/staging/lustre/lustre/obdclass/llog_lvfs.c +++ b/drivers/staging/lustre/lustre/obdclass/llog_lvfs.c @@ -376,7 +376,7 @@ static void llog_skip_over(__u64 *off, int curr, int goal) /* sets: * - cur_offset to the furthest point read in the log file - * - cur_idx to the log index preceeding cur_offset + * - cur_idx to the log index preceding cur_offset * returns -EIO/-EINVAL on error */ static int llog_lvfs_next_block(const struct lu_env *env, diff --git a/drivers/staging/lustre/lustre/obdclass/llog_osd.c b/drivers/staging/lustre/lustre/obdclass/llog_osd.c index 654c8e189653..682279de8bea 100644 --- a/drivers/staging/lustre/lustre/obdclass/llog_osd.c +++ b/drivers/staging/lustre/lustre/obdclass/llog_osd.c @@ -514,7 +514,7 @@ static void llog_skip_over(__u64 *off, int curr, int goal) /* sets: * - cur_offset to the furthest point read in the log file - * - cur_idx to the log index preceeding cur_offset + * - cur_idx to the log index preceding cur_offset * returns -EIO/-EINVAL on error */ static int llog_osd_next_block(const struct lu_env *env, @@ -1073,7 +1073,7 @@ static int llog_osd_setup(const struct lu_env *env, struct obd_device *obd, LASSERT(ctxt); /* initialize data allowing to generate new fids, - * literally we need a sequece */ + * literally we need a sequence */ lgi->lgi_fid.f_seq = FID_SEQ_LLOG; lgi->lgi_fid.f_oid = 1; lgi->lgi_fid.f_ver = 0; @@ -1280,7 +1280,7 @@ int llog_osd_put_cat_list(const struct lu_env *env, struct dt_device *d, lgi->lgi_buf.lb_len = size; rc = dt_record_write(env, o, &lgi->lgi_buf, &lgi->lgi_off, th); if (rc) - CDEBUG(D_INODE, "error writeing CATALOGS: rc = %d\n", rc); + CDEBUG(D_INODE, "error writing CATALOGS: rc = %d\n", rc); out_trans: dt_trans_stop(env, d, th); out: diff --git a/drivers/staging/lustre/lustre/obdclass/local_storage.c b/drivers/staging/lustre/lustre/obdclass/local_storage.c index e79e4beb3628..e76f7d044231 100644 --- a/drivers/staging/lustre/lustre/obdclass/local_storage.c +++ b/drivers/staging/lustre/lustre/obdclass/local_storage.c @@ -737,7 +737,7 @@ int lastid_compat_check(const struct lu_env *env, struct dt_device *dev, * All dynamic fids will be generated with the same sequence and incremented * OIDs * - * Returned local_oid_storage is in-memory representaion of OID storage + * Returned local_oid_storage is in-memory representation of OID storage */ int local_oid_storage_init(const struct lu_env *env, struct dt_device *dev, const struct lu_fid *first_fid, diff --git a/drivers/staging/lustre/lustre/obdclass/lprocfs_status.c b/drivers/staging/lustre/lustre/obdclass/lprocfs_status.c index ec3b605dae6b..1432dd74fe95 100644 --- a/drivers/staging/lustre/lustre/obdclass/lprocfs_status.c +++ b/drivers/staging/lustre/lustre/obdclass/lprocfs_status.c @@ -98,6 +98,8 @@ static const char * const obd_connect_names[] = { "lightweight_conn", "short_io", "pingless", + "flock_deadlock", + "disp_stripe", "unknown", NULL }; diff --git a/drivers/staging/lustre/lustre/obdclass/lu_object.c b/drivers/staging/lustre/lustre/obdclass/lu_object.c index 9887d8fffb6e..92e8a15a5e5d 100644 --- a/drivers/staging/lustre/lustre/obdclass/lu_object.c +++ b/drivers/staging/lustre/lustre/obdclass/lu_object.c @@ -571,7 +571,7 @@ static struct lu_object *htable_lookup(struct lu_site *s, * drained), and moreover, lookup has to wait until object is freed. */ - init_waitqueue_entry_current(waiter); + init_waitqueue_entry(waiter, current); add_wait_queue(&bkt->lsb_marche_funebre, waiter); set_current_state(TASK_UNINTERRUPTIBLE); lprocfs_counter_incr(s->ls_stats, LU_SS_CACHE_DEATH_RACE); @@ -712,7 +712,7 @@ struct lu_object *lu_object_find_at(const struct lu_env *env, * lu_object_find_try() already added waiter into the * wait queue. */ - waitq_wait(&wait, TASK_UNINTERRUPTIBLE); + schedule(); bkt = lu_site_bkt_from_fid(dev->ld_site, (void *)f); remove_wait_queue(&bkt->lsb_marche_funebre, &wait); } @@ -890,10 +890,10 @@ static unsigned lu_obj_hop_hash(struct cfs_hash *hs, hash = fid_flatten32(fid); hash += (hash >> 4) + (hash << 12); /* mixing oid and seq */ - hash = cfs_hash_long(hash, hs->hs_bkt_bits); + hash = hash_long(hash, hs->hs_bkt_bits); /* give me another random factor */ - hash -= cfs_hash_long((unsigned long)hs, fid_oid(fid) % 11 + 3); + hash -= hash_long((unsigned long)hs, fid_oid(fid) % 11 + 3); hash <<= hs->hs_cur_bits - hs->hs_bkt_bits; hash |= (fid_seq(fid) + fid_oid(fid)) & (CFS_HASH_NBKT(hs) - 1); @@ -2100,7 +2100,7 @@ void lu_object_assign_fid(const struct lu_env *env, struct lu_object *o, EXPORT_SYMBOL(lu_object_assign_fid); /** - * allocates object with 0 (non-assiged) fid + * allocates object with 0 (non-assigned) fid * XXX: temporary solution to be able to assign fid in ->do_create() * till we have fully-functional OST fids */ diff --git a/drivers/staging/lustre/lustre/obdclass/md_attrs.c b/drivers/staging/lustre/lustre/obdclass/md_attrs.c index f7187829e276..f080cceb384c 100644 --- a/drivers/staging/lustre/lustre/obdclass/md_attrs.c +++ b/drivers/staging/lustre/lustre/obdclass/md_attrs.c @@ -60,7 +60,7 @@ EXPORT_SYMBOL(lustre_lma_init); */ void lustre_lma_swab(struct lustre_mdt_attrs *lma) { - /* Use LUSTRE_MSG_MAGIC to detect local endianess. */ + /* Use LUSTRE_MSG_MAGIC to detect local endianness. */ if (LUSTRE_MSG_MAGIC != cpu_to_le32(LUSTRE_MSG_MAGIC)) { __swab32s(&lma->lma_compat); __swab32s(&lma->lma_incompat); @@ -77,7 +77,7 @@ EXPORT_SYMBOL(lustre_lma_swab); */ void lustre_som_swab(struct som_attrs *attrs) { - /* Use LUSTRE_MSG_MAGIC to detect local endianess. */ + /* Use LUSTRE_MSG_MAGIC to detect local endianness. */ if (LUSTRE_MSG_MAGIC != cpu_to_le32(LUSTRE_MSG_MAGIC)) { __swab32s(&attrs->som_compat); __swab32s(&attrs->som_incompat); @@ -135,7 +135,7 @@ EXPORT_SYMBOL(lustre_buf2som); */ void lustre_hsm_swab(struct hsm_attrs *attrs) { - /* Use LUSTRE_MSG_MAGIC to detect local endianess. */ + /* Use LUSTRE_MSG_MAGIC to detect local endianness. */ if (LUSTRE_MSG_MAGIC != cpu_to_le32(LUSTRE_MSG_MAGIC)) { __swab32s(&attrs->hsm_compat); __swab32s(&attrs->hsm_flags); diff --git a/drivers/staging/lustre/lustre/obdclass/obd_config.c b/drivers/staging/lustre/lustre/obdclass/obd_config.c index 362ae541b209..2d5777699f47 100644 --- a/drivers/staging/lustre/lustre/obdclass/obd_config.c +++ b/drivers/staging/lustre/lustre/obdclass/obd_config.c @@ -61,7 +61,8 @@ int class_find_param(char *buf, char *key, char **valp) if (!buf) return 1; - if ((ptr = strstr(buf, key)) == NULL) + ptr = strstr(buf, key); + if (ptr == NULL) return 1; if (valp) @@ -592,7 +593,7 @@ int class_detach(struct obd_device *obd, struct lustre_cfg *lcfg) } EXPORT_SYMBOL(class_detach); -/** Start shutting down the obd. There may be in-progess ops when +/** Start shutting down the obd. There may be in-progress ops when * this is called. We tell them to start shutting down with a call * to class_disconnect_exports(). */ @@ -655,7 +656,7 @@ int class_cleanup(struct obd_device *obd, struct lustre_cfg *lcfg) /* The three references that should be remaining are the * obd_self_export and the attach and setup references. */ if (atomic_read(&obd->obd_refcount) > 3) { - /* refcounf - 3 might be the number of real exports + /* refcount - 3 might be the number of real exports (excluding self export). But class_incref is called by other things as well, so don't count on it. */ CDEBUG(D_IOCTL, "%s: forcing exports to disconnect: %d\n", @@ -1027,6 +1028,46 @@ struct lustre_cfg *lustre_cfg_rename(struct lustre_cfg *cfg, } EXPORT_SYMBOL(lustre_cfg_rename); +static int process_param2_config(struct lustre_cfg *lcfg) +{ + char *param = lustre_cfg_string(lcfg, 1); + char *upcall = lustre_cfg_string(lcfg, 2); + char *argv[] = { + [0] = "/usr/sbin/lctl", + [1] = "set_param", + [2] = param, + [3] = NULL + }; + struct timeval start; + struct timeval end; + int rc; + + + /* Add upcall processing here. Now only lctl is supported */ + if (strcmp(upcall, LCTL_UPCALL) != 0) { + CERROR("Unsupported upcall %s\n", upcall); + return -EINVAL; + } + + do_gettimeofday(&start); + rc = USERMODEHELPER(argv[0], argv, NULL); + do_gettimeofday(&end); + + if (rc < 0) { + CERROR( + "lctl: error invoking upcall %s %s %s: rc = %d; time %ldus\n", + argv[0], argv[1], argv[2], rc, + cfs_timeval_sub(&end, &start, NULL)); + } else { + CDEBUG(D_HA, "lctl: invoked upcall %s %s %s, time %ldus\n", + argv[0], argv[1], argv[2], + cfs_timeval_sub(&end, &start, NULL)); + rc = 0; + } + + return rc; +} + void lustre_register_quota_process_config(int (*qpc)(struct lustre_cfg *lcfg)) { quota_process_config = qpc; @@ -1142,11 +1183,14 @@ int class_process_config(struct lustre_cfg *lcfg) err = (*quota_process_config)(lcfg); GOTO(out, err); } - /* Fall through */ + break; } + case LCFG_SET_PARAM: { + err = process_param2_config(lcfg); + GOTO(out, 0); + } } - /* Commands that require a device */ obd = class_name2obd(lustre_cfg_string(lcfg, 0)); if (obd == NULL) { diff --git a/drivers/staging/lustre/lustre/obdclass/obd_mount.c b/drivers/staging/lustre/lustre/obdclass/obd_mount.c index a69a630b596e..6f8ba54e9667 100644 --- a/drivers/staging/lustre/lustre/obdclass/obd_mount.c +++ b/drivers/staging/lustre/lustre/obdclass/obd_mount.c @@ -501,7 +501,7 @@ static int lustre_stop_mgc(struct super_block *sb) } /* The MGC has no recoverable data in any case. - * force shotdown set in umount_begin */ + * force shutdown set in umount_begin */ obd->obd_no_recov = 1; if (obd->u.cli.cl_mgc_mgsexp) { @@ -754,7 +754,7 @@ int server_name2index(const char *svname, __u32 *idx, const char **endptr) } EXPORT_SYMBOL(server_name2index); -/*************** mount common betweeen server and client ***************/ +/*************** mount common between server and client ***************/ /* Common umount */ int lustre_common_put_super(struct super_block *sb) diff --git a/drivers/staging/lustre/lustre/obdclass/obdo.c b/drivers/staging/lustre/lustre/obdclass/obdo.c index 70997648a4f3..3b1b28afd6b1 100644 --- a/drivers/staging/lustre/lustre/obdclass/obdo.c +++ b/drivers/staging/lustre/lustre/obdclass/obdo.c @@ -191,7 +191,7 @@ int obdo_cmp_md(struct obdo *dst, struct obdo *src, obd_flag compare) } if ( compare & OBD_MD_FLGENER ) res = (res || (dst->o_parent_oid != src->o_parent_oid)); - /* XXX Don't know if thses should be included here - wasn't previously + /* XXX Don't know if these should be included here - wasn't previously if ( compare & OBD_MD_FLINLINE ) res = (res || memcmp(dst->o_inline, src->o_inline)); */ @@ -233,7 +233,7 @@ void obdo_from_iattr(struct obdo *oa, struct iattr *attr, unsigned int ia_valid) oa->o_mode = attr->ia_mode; oa->o_valid |= OBD_MD_FLTYPE | OBD_MD_FLMODE; if (!in_group_p(make_kgid(&init_user_ns, oa->o_gid)) && - !cfs_capable(CFS_CAP_FSETID)) + !capable(CFS_CAP_FSETID)) oa->o_mode &= ~S_ISGID; } if (ia_valid & ATTR_UID) { @@ -282,7 +282,7 @@ void iattr_from_obdo(struct iattr *attr, struct obdo *oa, obd_flag valid) attr->ia_mode = (attr->ia_mode & S_IFMT)|(oa->o_mode & ~S_IFMT); attr->ia_valid |= ATTR_MODE; if (!in_group_p(make_kgid(&init_user_ns, oa->o_gid)) && - !cfs_capable(CFS_CAP_FSETID)) + !capable(CFS_CAP_FSETID)) attr->ia_mode &= ~S_ISGID; } if (valid & OBD_MD_FLUID) { diff --git a/drivers/staging/lustre/lustre/obdecho/echo.c b/drivers/staging/lustre/lustre/obdecho/echo.c index debb9cec4900..96a807f82ec1 100644 --- a/drivers/staging/lustre/lustre/obdecho/echo.c +++ b/drivers/staging/lustre/lustre/obdecho/echo.c @@ -606,7 +606,8 @@ static int echo_cleanup(struct obd_device *obd) /* XXX Bug 3413; wait for a bit to ensure the BL callback has * happened before calling ldlm_namespace_free() */ - schedule_timeout_and_set_state(TASK_UNINTERRUPTIBLE, cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); ldlm_namespace_free(obd->obd_namespace, NULL, obd->obd_force); obd->obd_namespace = NULL; diff --git a/drivers/staging/lustre/lustre/obdecho/echo_client.c b/drivers/staging/lustre/lustre/obdecho/echo_client.c index 9b2dea292363..754aa8e55665 100644 --- a/drivers/staging/lustre/lustre/obdecho/echo_client.c +++ b/drivers/staging/lustre/lustre/obdecho/echo_client.c @@ -997,8 +997,8 @@ static struct lu_device *echo_device_free(const struct lu_env *env, spin_unlock(&ec->ec_lock); CERROR("echo_client still has objects at cleanup time, " "wait for 1 second\n"); - schedule_timeout_and_set_state(TASK_UNINTERRUPTIBLE, - cfs_time_seconds(1)); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(cfs_time_seconds(1)); lu_site_purge(env, &ed->ed_site->cs_lu, -1); spin_lock(&ec->ec_lock); } @@ -2764,7 +2764,7 @@ echo_client_iocontrol(unsigned int cmd, struct obd_export *exp, int len, switch (cmd) { case OBD_IOC_CREATE: /* may create echo object */ - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) GOTO (out, rc = -EPERM); rc = echo_create_object(env, ed, 1, oa, data->ioc_pbuf1, @@ -2778,7 +2778,7 @@ echo_client_iocontrol(unsigned int cmd, struct obd_export *exp, int len, int dirlen; __u64 id; - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) GOTO(out, rc = -EPERM); count = data->ioc_count; @@ -2806,7 +2806,7 @@ echo_client_iocontrol(unsigned int cmd, struct obd_export *exp, int len, __u64 seq; int max_count; - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) GOTO(out, rc = -EPERM); cl_env = cl_env_get(&refcheck); @@ -2838,7 +2838,7 @@ echo_client_iocontrol(unsigned int cmd, struct obd_export *exp, int len, GOTO(out, rc); } case OBD_IOC_DESTROY: - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) GOTO (out, rc = -EPERM); rc = echo_get_object(&eco, ed, oa); @@ -2863,7 +2863,7 @@ echo_client_iocontrol(unsigned int cmd, struct obd_export *exp, int len, GOTO(out, rc); case OBD_IOC_SETATTR: - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) GOTO (out, rc = -EPERM); rc = echo_get_object(&eco, ed, oa); @@ -2878,7 +2878,7 @@ echo_client_iocontrol(unsigned int cmd, struct obd_export *exp, int len, GOTO(out, rc); case OBD_IOC_BRW_WRITE: - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) GOTO (out, rc = -EPERM); rw = OBD_BRW_WRITE; @@ -2897,7 +2897,7 @@ echo_client_iocontrol(unsigned int cmd, struct obd_export *exp, int len, GOTO(out, rc); case ECHO_IOC_SET_STRIPE: - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) GOTO (out, rc = -EPERM); if (data->ioc_pbuf1 == NULL) { /* unset */ @@ -2914,7 +2914,7 @@ echo_client_iocontrol(unsigned int cmd, struct obd_export *exp, int len, GOTO (out, rc); case ECHO_IOC_ENQUEUE: - if (!cfs_capable(CFS_CAP_SYS_ADMIN)) + if (!capable(CFS_CAP_SYS_ADMIN)) GOTO (out, rc = -EPERM); rc = echo_client_enqueue(exp, oa, diff --git a/drivers/staging/lustre/lustre/osc/osc_cache.c b/drivers/staging/lustre/lustre/osc/osc_cache.c index be4511e78c04..fe9989a49f81 100644 --- a/drivers/staging/lustre/lustre/osc/osc_cache.c +++ b/drivers/staging/lustre/lustre/osc/osc_cache.c @@ -1020,7 +1020,7 @@ out: /** * This function is used to make the extent prepared for transfer. - * A race with flusing page - ll_writepage() has to be handled cautiously. + * A race with flushing page - ll_writepage() has to be handled cautiously. */ static int osc_extent_make_ready(const struct lu_env *env, struct osc_extent *ext) @@ -2146,7 +2146,7 @@ int osc_prep_async_page(struct osc_object *osc, struct osc_page *ops, oap->oap_obj_off = offset; LASSERT(!(offset & ~CFS_PAGE_MASK)); - if (!client_is_remote(exp) && cfs_capable(CFS_CAP_SYS_RESOURCE)) + if (!client_is_remote(exp) && capable(CFS_CAP_SYS_RESOURCE)) oap->oap_brw_flags = OBD_BRW_NOQUOTA; INIT_LIST_HEAD(&oap->oap_pending_item); @@ -2186,7 +2186,7 @@ int osc_queue_async_io(const struct lu_env *env, struct cl_io *io, /* Set the OBD_BRW_SRVLOCK before the page is queued. */ brw_flags |= ops->ops_srvlock ? OBD_BRW_SRVLOCK : 0; if (!client_is_remote(osc_export(osc)) && - cfs_capable(CFS_CAP_SYS_RESOURCE)) { + capable(CFS_CAP_SYS_RESOURCE)) { brw_flags |= OBD_BRW_NOQUOTA; cmd |= OBD_BRW_NOQUOTA; } @@ -2394,6 +2394,12 @@ int osc_flush_async_page(const struct lu_env *env, struct cl_io *io, * really sending the RPC. */ case OES_TRUNC: /* race with truncate, page will be redirtied */ + case OES_ACTIVE: + /* The extent is active so we need to abort and let the caller + * re-dirty the page. If we continued on here, and we were the + * one making the extent active, we could deadlock waiting for + * the page writeback to clear but it won't because the extent + * is active and won't be written out. */ GOTO(out, rc = -EAGAIN); default: break; diff --git a/drivers/staging/lustre/lustre/osc/osc_io.c b/drivers/staging/lustre/lustre/osc/osc_io.c index 681d60a7d5ef..5f3c545418d1 100644 --- a/drivers/staging/lustre/lustre/osc/osc_io.c +++ b/drivers/staging/lustre/lustre/osc/osc_io.c @@ -297,7 +297,7 @@ static int osc_io_commit_write(const struct lu_env *env, */ osc_page_touch(env, cl2osc_page(slice), to); if (!client_is_remote(osc_export(obj)) && - cfs_capable(CFS_CAP_SYS_RESOURCE)) + capable(CFS_CAP_SYS_RESOURCE)) oap->oap_brw_flags |= OBD_BRW_NOQUOTA; if (oio->oi_lockless) @@ -512,19 +512,15 @@ static int osc_io_read_start(const struct lu_env *env, struct osc_io *oio = cl2osc_io(env, slice); struct cl_object *obj = slice->cis_obj; struct cl_attr *attr = &osc_env_info(env)->oti_attr; - int result = 0; + int rc = 0; - if (oio->oi_lockless == 0) { + if (oio->oi_lockless == 0 && !slice->cis_io->ci_noatime) { cl_object_attr_lock(obj); - result = cl_object_attr_get(env, obj, attr); - if (result == 0) { - attr->cat_atime = LTIME_S(CURRENT_TIME); - result = cl_object_attr_set(env, obj, attr, - CAT_ATIME); - } + attr->cat_atime = LTIME_S(CURRENT_TIME); + rc = cl_object_attr_set(env, obj, attr, CAT_ATIME); cl_object_attr_unlock(obj); } - return result; + return rc; } static int osc_io_write_start(const struct lu_env *env, diff --git a/drivers/staging/lustre/lustre/osc/osc_page.c b/drivers/staging/lustre/lustre/osc/osc_page.c index 4909e486dc5c..96cb6e2b9c4e 100644 --- a/drivers/staging/lustre/lustre/osc/osc_page.c +++ b/drivers/staging/lustre/lustre/osc/osc_page.c @@ -561,7 +561,7 @@ void osc_page_submit(const struct lu_env *env, struct osc_page *opg, oap->oap_brw_flags = OBD_BRW_SYNC | brw_flags; if (!client_is_remote(osc_export(obj)) && - cfs_capable(CFS_CAP_SYS_RESOURCE)) { + capable(CFS_CAP_SYS_RESOURCE)) { oap->oap_brw_flags |= OBD_BRW_NOQUOTA; oap->oap_cmd |= OBD_BRW_NOQUOTA; } diff --git a/drivers/staging/lustre/lustre/osc/osc_quota.c b/drivers/staging/lustre/lustre/osc/osc_quota.c index 6045a78a2baa..0235fabaaffe 100644 --- a/drivers/staging/lustre/lustre/osc/osc_quota.c +++ b/drivers/staging/lustre/lustre/osc/osc_quota.c @@ -51,11 +51,8 @@ int osc_quota_chkdq(struct client_obd *cli, const unsigned int qid[]) oqi = cfs_hash_lookup(cli->cl_quota_hash[type], &qid[type]); if (oqi) { - obd_uid id = oqi->oqi_id; - - LASSERTF(id == qid[type], - "The ids don't match %u != %u\n", - id, qid[type]); + /* do not try to access oqi here, it could have been + * freed by osc_quota_setdq() */ /* the slot is busy, the user is about to run out of * quota space on this OST */ @@ -268,11 +265,16 @@ int osc_quotactl(struct obd_device *unused, struct obd_export *exp, if (rc) CERROR("ptlrpc_queue_wait failed, rc: %d\n", rc); - if (req->rq_repmsg && - (oqc = req_capsule_server_get(&req->rq_pill, &RMF_OBD_QUOTACTL))) { - *oqctl = *oqc; + if (req->rq_repmsg) { + oqc = req_capsule_server_get(&req->rq_pill, &RMF_OBD_QUOTACTL); + if (oqc) { + *oqctl = *oqc; + } else if (!rc) { + CERROR("Can't unpack obd_quotactl\n"); + rc = -EPROTO; + } } else if (!rc) { - CERROR ("Can't unpack obd_quotactl\n"); + CERROR("Can't unpack obd_quotactl\n"); rc = -EPROTO; } ptlrpc_req_finished(req); diff --git a/drivers/staging/lustre/lustre/ptlrpc/client.c b/drivers/staging/lustre/lustre/ptlrpc/client.c index d90efe408414..4c9e00695087 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/client.c +++ b/drivers/staging/lustre/lustre/ptlrpc/client.c @@ -48,6 +48,7 @@ #include "ptlrpc_internal.h" static int ptlrpc_send_new_req(struct ptlrpc_request *req); +static int ptlrpcd_check_work(struct ptlrpc_request *req); /** * Initialize passed in client structure \a cl. @@ -62,7 +63,7 @@ void ptlrpc_init_client(int req_portal, int rep_portal, char *name, EXPORT_SYMBOL(ptlrpc_init_client); /** - * Return PortalRPC connection for remore uud \a uuid + * Return PortalRPC connection for remote uud \a uuid */ struct ptlrpc_connection *ptlrpc_uuid_to_connection(struct obd_uuid *uuid) { @@ -127,7 +128,7 @@ struct ptlrpc_bulk_desc *ptlrpc_new_bulk(unsigned npages, unsigned max_brw, * Prepare bulk descriptor for specified outgoing request \a req that * can fit \a npages * pages. \a type is bulk type. \a portal is where * the bulk to be sent. Used on client-side. - * Returns pointer to newly allocatrd initialized bulk descriptor or NULL on + * Returns pointer to newly allocated initialized bulk descriptor or NULL on * error. */ struct ptlrpc_bulk_desc *ptlrpc_prep_bulk_imp(struct ptlrpc_request *req, @@ -631,7 +632,7 @@ int ptlrpc_request_pack(struct ptlrpc_request *request, /* For some old 1.8 clients (< 1.8.7), they will LASSERT the size of * ptlrpc_body sent from server equal to local ptlrpc_body size, so we - * have to send old ptlrpc_body to keep interoprability with these + * have to send old ptlrpc_body to keep interoperability with these * clients. * * Only three kinds of server->client RPCs so far: @@ -639,7 +640,7 @@ int ptlrpc_request_pack(struct ptlrpc_request *request, * - LDLM_CP_CALLBACK * - LDLM_GL_CALLBACK * - * XXX This should be removed whenever we drop the interoprability with + * XXX This should be removed whenever we drop the interoperability with * the these old clients. */ if (opcode == LDLM_BL_CALLBACK || opcode == LDLM_CP_CALLBACK || @@ -686,7 +687,7 @@ struct ptlrpc_request *__ptlrpc_request_alloc(struct obd_import *imp, /** * Helper function for creating a request. - * Calls __ptlrpc_request_alloc to allocate new request sturcture and inits + * Calls __ptlrpc_request_alloc to allocate new request structure and inits * buffer structures according to capsule template \a format. * Returns allocated request structure pointer or NULL on error. */ @@ -743,7 +744,7 @@ void ptlrpc_request_free(struct ptlrpc_request *request) EXPORT_SYMBOL(ptlrpc_request_free); /** - * Allocate new request for operatione \a opcode and immediatelly pack it for + * Allocate new request for operation \a opcode and immediately pack it for * network transfer. * Only used for simple requests like OBD_PING where the only important * part of the request is operation itself. @@ -768,7 +769,7 @@ struct ptlrpc_request *ptlrpc_request_alloc_pack(struct obd_import *imp, EXPORT_SYMBOL(ptlrpc_request_alloc_pack); /** - * Prepare request (fetched from pool \a poolif not NULL) on import \a imp + * Prepare request (fetched from pool \a pool if not NULL) on import \a imp * for operation \a opcode. Request would contain \a count buffers. * Sizes of buffers are described in array \a lengths and buffers themselves * are provided by a pointer \a bufs. @@ -1073,7 +1074,7 @@ static int ptlrpc_import_delay_req(struct obd_import *imp, } /** - * Decide if the eror message regarding provided request \a req + * Decide if the error message regarding provided request \a req * should be printed to the console or not. * Makes it's decision on request status and other properties. * Returns 1 to print error on the system console or 0 if not. @@ -1159,7 +1160,7 @@ static void ptlrpc_save_versions(struct ptlrpc_request *req) /** * Callback function called when client receives RPC reply for \a req. * Returns 0 on success or error code. - * The return alue would be assigned to req->rq_status by the caller + * The return value would be assigned to req->rq_status by the caller * as request processing status. * This function also decides if the request needs to be saved for later replay. */ @@ -1190,7 +1191,9 @@ static int after_reply(struct ptlrpc_request *req) * will roundup it */ req->rq_replen = req->rq_nob_received; req->rq_nob_received = 0; + spin_lock(&req->rq_lock); req->rq_resend = 1; + spin_unlock(&req->rq_lock); return 0; } @@ -1313,7 +1316,11 @@ static int after_reply(struct ptlrpc_request *req) /** version recovery */ ptlrpc_save_versions(req); ptlrpc_retain_replayable_request(req, imp); - } else if (req->rq_commit_cb != NULL) { + } else if (req->rq_commit_cb != NULL && + list_empty(&req->rq_replay_list)) { + /* NB: don't call rq_commit_cb if it's already on + * rq_replay_list, ptlrpc_free_committed() will call + * it later, see LU-3618 for details */ spin_unlock(&imp->imp_lock); req->rq_commit_cb(req); spin_lock(&imp->imp_lock); @@ -1408,7 +1415,9 @@ static int ptlrpc_send_new_req(struct ptlrpc_request *req) req->rq_status = rc; return 1; } else { + spin_lock(&req->rq_lock); req->rq_wait_ctx = 1; + spin_unlock(&req->rq_lock); return 0; } } @@ -1423,7 +1432,9 @@ static int ptlrpc_send_new_req(struct ptlrpc_request *req) rc = ptl_send_rpc(req, 0); if (rc) { DEBUG_REQ(D_HA, req, "send failed (%d); expect timeout", rc); + spin_lock(&req->rq_lock); req->rq_net_err = 1; + spin_unlock(&req->rq_lock); return rc; } return 0; @@ -1688,6 +1699,7 @@ int ptlrpc_check_set(const struct lu_env *env, struct ptlrpc_request_set *set) spin_lock(&req->rq_lock); req->rq_net_err = 1; spin_unlock(&req->rq_lock); + continue; } /* need to reset the timeout */ force_timer_recalc = 1; @@ -1773,6 +1785,10 @@ interpret: ptlrpc_req_interpret(env, req, req->rq_status); + if (ptlrpcd_check_work(req)) { + atomic_dec(&set->set_remaining); + continue; + } ptlrpc_rqphase_move(req, RQ_PHASE_COMPLETE); CDEBUG(req->rq_reqmsg != NULL ? D_RPCTRACE : 0, @@ -2031,7 +2047,7 @@ int ptlrpc_set_next_timeout(struct ptlrpc_request_set *set) EXPORT_SYMBOL(ptlrpc_set_next_timeout); /** - * Send all unset request from the set and then wait untill all + * Send all unset request from the set and then wait until all * requests in the set complete (either get a reply, timeout, get an * error or otherwise be interrupted). * Returns 0 on success or error code otherwise. @@ -2156,7 +2172,7 @@ int ptlrpc_set_wait(struct ptlrpc_request_set *set) EXPORT_SYMBOL(ptlrpc_set_wait); /** - * Helper fuction for request freeing. + * Helper function for request freeing. * Called when request count reached zero and request needs to be freed. * Removes request from all sorts of sending/replay lists it might be on, * frees network buffers if any are present. @@ -2223,7 +2239,7 @@ static void __ptlrpc_free_req(struct ptlrpc_request *request, int locked) static int __ptlrpc_req_finished(struct ptlrpc_request *request, int locked); /** * Drop one request reference. Must be called with import imp_lock held. - * When reference count drops to zero, reuqest is freed. + * When reference count drops to zero, request is freed. */ void ptlrpc_req_finished_with_imp_lock(struct ptlrpc_request *request) { @@ -2236,7 +2252,7 @@ EXPORT_SYMBOL(ptlrpc_req_finished_with_imp_lock); * Helper function * Drops one reference count for request \a request. * \a locked set indicates that caller holds import imp_lock. - * Frees the request whe reference count reaches zero. + * Frees the request when reference count reaches zero. */ static int __ptlrpc_req_finished(struct ptlrpc_request *request, int locked) { @@ -2360,19 +2376,52 @@ int ptlrpc_unregister_reply(struct ptlrpc_request *request, int async) } EXPORT_SYMBOL(ptlrpc_unregister_reply); +static void ptlrpc_free_request(struct ptlrpc_request *req) +{ + spin_lock(&req->rq_lock); + req->rq_replay = 0; + spin_unlock(&req->rq_lock); + + if (req->rq_commit_cb != NULL) + req->rq_commit_cb(req); + list_del_init(&req->rq_replay_list); + + __ptlrpc_req_finished(req, 1); +} + +/** + * the request is committed and dropped from the replay list of its import + */ +void ptlrpc_request_committed(struct ptlrpc_request *req, int force) +{ + struct obd_import *imp = req->rq_import; + + spin_lock(&imp->imp_lock); + if (list_empty(&req->rq_replay_list)) { + spin_unlock(&imp->imp_lock); + return; + } + + if (force || req->rq_transno <= imp->imp_peer_committed_transno) + ptlrpc_free_request(req); + + spin_unlock(&imp->imp_lock); +} +EXPORT_SYMBOL(ptlrpc_request_committed); + /** * Iterates through replay_list on import and prunes * all requests have transno smaller than last_committed for the * import and don't have rq_replay set. - * Since requests are sorted in transno order, stops when meetign first + * Since requests are sorted in transno order, stops when meeting first * transno bigger than last_committed. * caller must hold imp->imp_lock */ void ptlrpc_free_committed(struct obd_import *imp) { - struct list_head *tmp, *saved; - struct ptlrpc_request *req; + struct ptlrpc_request *req, *saved; struct ptlrpc_request *last_req = NULL; /* temporary fire escape */ + bool skip_committed_list = true; LASSERT(imp != NULL); @@ -2388,13 +2437,15 @@ void ptlrpc_free_committed(struct obd_import *imp) CDEBUG(D_RPCTRACE, "%s: committing for last_committed "LPU64" gen %d\n", imp->imp_obd->obd_name, imp->imp_peer_committed_transno, imp->imp_generation); + + if (imp->imp_generation != imp->imp_last_generation_checked) + skip_committed_list = false; + imp->imp_last_transno_checked = imp->imp_peer_committed_transno; imp->imp_last_generation_checked = imp->imp_generation; - list_for_each_safe(tmp, saved, &imp->imp_replay_list) { - req = list_entry(tmp, struct ptlrpc_request, - rq_replay_list); - + list_for_each_entry_safe(req, saved, &imp->imp_replay_list, + rq_replay_list) { /* XXX ok to remove when 1357 resolved - rread 05/29/03 */ LASSERT(req != last_req); last_req = req; @@ -2408,27 +2459,34 @@ void ptlrpc_free_committed(struct obd_import *imp) GOTO(free_req, 0); } - if (req->rq_replay) { - DEBUG_REQ(D_RPCTRACE, req, "keeping (FL_REPLAY)"); - continue; - } - /* not yet committed */ if (req->rq_transno > imp->imp_peer_committed_transno) { DEBUG_REQ(D_RPCTRACE, req, "stopping search"); break; } + if (req->rq_replay) { + DEBUG_REQ(D_RPCTRACE, req, "keeping (FL_REPLAY)"); + list_move_tail(&req->rq_replay_list, + &imp->imp_committed_list); + continue; + } + DEBUG_REQ(D_INFO, req, "commit (last_committed "LPU64")", imp->imp_peer_committed_transno); free_req: - spin_lock(&req->rq_lock); - req->rq_replay = 0; - spin_unlock(&req->rq_lock); - if (req->rq_commit_cb != NULL) - req->rq_commit_cb(req); - list_del_init(&req->rq_replay_list); - __ptlrpc_req_finished(req, 1); + ptlrpc_free_request(req); + } + if (skip_committed_list) + return; + + list_for_each_entry_safe(req, saved, &imp->imp_committed_list, + rq_replay_list) { + LASSERT(req->rq_transno != 0); + if (req->rq_import_generation < imp->imp_generation) { + DEBUG_REQ(D_RPCTRACE, req, "free stale open request"); + ptlrpc_free_request(req); + } } } @@ -2585,7 +2643,7 @@ struct ptlrpc_replay_async_args { /** * Callback used for replayed requests reply processing. - * In case of succesful reply calls registeresd request replay callback. + * In case of successful reply calls registered request replay callback. * In case of error restart replay process. */ static int ptlrpc_replay_interpret(const struct lu_env *env, @@ -2834,7 +2892,7 @@ void ptlrpc_init_xid(void) ptlrpc_last_xid = (__u64)now << 20; } - /* Need to always be aligned to a power-of-two for mutli-bulk BRW */ + /* Always need to be aligned to a power-of-two for multi-bulk BRW */ CLASSERT((PTLRPC_BULK_OPS_COUNT & (PTLRPC_BULK_OPS_COUNT - 1)) == 0); ptlrpc_last_xid &= PTLRPC_BULK_OPS_MASK; } @@ -2904,22 +2962,50 @@ EXPORT_SYMBOL(ptlrpc_sample_next_xid); * have delay before it really runs by ptlrpcd thread. */ struct ptlrpc_work_async_args { - __u64 magic; int (*cb)(const struct lu_env *, void *); void *cbdata; }; -#define PTLRPC_WORK_MAGIC 0x6655436b676f4f44ULL /* magic code */ +static void ptlrpcd_add_work_req(struct ptlrpc_request *req) +{ + /* re-initialize the req */ + req->rq_timeout = obd_timeout; + req->rq_sent = cfs_time_current_sec(); + req->rq_deadline = req->rq_sent + req->rq_timeout; + req->rq_reply_deadline = req->rq_deadline; + req->rq_phase = RQ_PHASE_INTERPRET; + req->rq_next_phase = RQ_PHASE_COMPLETE; + req->rq_xid = ptlrpc_next_xid(); + req->rq_import_generation = req->rq_import->imp_generation; + + ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1); +} static int work_interpreter(const struct lu_env *env, struct ptlrpc_request *req, void *data, int rc) { struct ptlrpc_work_async_args *arg = data; - LASSERT(arg->magic == PTLRPC_WORK_MAGIC); + LASSERT(ptlrpcd_check_work(req)); LASSERT(arg->cb != NULL); - return arg->cb(env, arg->cbdata); + rc = arg->cb(env, arg->cbdata); + + list_del_init(&req->rq_set_chain); + req->rq_set = NULL; + + if (atomic_dec_return(&req->rq_refcount) > 1) { + atomic_set(&req->rq_refcount, 2); + ptlrpcd_add_work_req(req); + } + return rc; +} + +static int worker_format; + +static int ptlrpcd_check_work(struct ptlrpc_request *req) +{ + return req->rq_pill.rc_fmt == (void *)&worker_format; } /** @@ -2952,6 +3038,7 @@ void *ptlrpcd_alloc_work(struct obd_import *imp, req->rq_receiving_reply = 0; req->rq_must_unlink = 0; req->rq_no_delay = req->rq_no_resend = 1; + req->rq_pill.rc_fmt = (void *)&worker_format; spin_lock_init(&req->rq_lock); INIT_LIST_HEAD(&req->rq_list); @@ -2965,7 +3052,6 @@ void *ptlrpcd_alloc_work(struct obd_import *imp, CLASSERT(sizeof(*args) <= sizeof(req->rq_async_args)); args = ptlrpc_req_async_args(req); - args->magic = PTLRPC_WORK_MAGIC; args->cb = cb; args->cbdata = cbdata; @@ -2995,25 +3081,8 @@ int ptlrpcd_queue_work(void *handler) * req as opaque data. - Jinshan */ LASSERT(atomic_read(&req->rq_refcount) > 0); - if (atomic_read(&req->rq_refcount) > 1) - return -EBUSY; - - if (atomic_inc_return(&req->rq_refcount) > 2) { /* race */ - atomic_dec(&req->rq_refcount); - return -EBUSY; - } - - /* re-initialize the req */ - req->rq_timeout = obd_timeout; - req->rq_sent = cfs_time_current_sec(); - req->rq_deadline = req->rq_sent + req->rq_timeout; - req->rq_reply_deadline = req->rq_deadline; - req->rq_phase = RQ_PHASE_INTERPRET; - req->rq_next_phase = RQ_PHASE_COMPLETE; - req->rq_xid = ptlrpc_next_xid(); - req->rq_import_generation = req->rq_import->imp_generation; - - ptlrpcd_add_req(req, PDL_POLICY_ROUND, -1); + if (atomic_inc_return(&req->rq_refcount) == 2) + ptlrpcd_add_work_req(req); return 0; } EXPORT_SYMBOL(ptlrpcd_queue_work); diff --git a/drivers/staging/lustre/lustre/ptlrpc/events.c b/drivers/staging/lustre/lustre/ptlrpc/events.c index f66cfea87acf..6ea0a491cfb3 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/events.c +++ b/drivers/staging/lustre/lustre/ptlrpc/events.c @@ -545,7 +545,7 @@ int ptlrpc_ni_init(void) * different depending on... */ /* kernel LNet calls our master callback when there are new event, * because we are guaranteed to get every event via callback, - * so we just set EQ size to 0 to avoid overhread of serializing + * so we just set EQ size to 0 to avoid overhead of serializing * enqueue/dequeue operations in LNet. */ rc = LNetEQAlloc(0, ptlrpc_master_callback, &ptlrpc_eq_h); if (rc == 0) diff --git a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_cli_upcall.c b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_cli_upcall.c index 55247af3910e..c279edf5b2a5 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_cli_upcall.c +++ b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_cli_upcall.c @@ -336,7 +336,7 @@ int gss_do_ctx_init_rpc(__user char *buffer, unsigned long count) if (rc) { /* If any _real_ denial be made, we expect server return * -EACCES reply or return success but indicate gss error - * inside reply messsage. All other errors are treated as + * inside reply message. All other errors are treated as * timeout, caller might try the negotiation repeatedly, * leave recovery decisions to general ptlrpc layer. * diff --git a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_generic_token.c b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_generic_token.c index 56c28286c9c1..8dc5c724958d 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_generic_token.c +++ b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_generic_token.c @@ -144,7 +144,8 @@ int der_read_length(unsigned char **buf, int *bufsize) sf = *(*buf)++; (*bufsize)--; if (sf & 0x80) { - if ((sf &= 0x7f) > ((*bufsize) - 1)) + sf &= 0x7f; + if (((*bufsize) - 1) < sf) return -1; if (sf > SIZEOF_INT) return -1; @@ -199,27 +200,32 @@ __u32 g_verify_token_header(rawobj_t *mech, int *body_size, rawobj_t toid; int ret = 0; - if ((toksize -= 1) < 0) + toksize -= 1; + if (0 > toksize) return (G_BAD_TOK_HEADER); if (*buf++ != 0x60) return (G_BAD_TOK_HEADER); - if ((seqsize = der_read_length(&buf, &toksize)) < 0) + seqsize = der_read_length(&buf, &toksize); + if (seqsize < 0) return(G_BAD_TOK_HEADER); if (seqsize != toksize) return (G_BAD_TOK_HEADER); - if ((toksize -= 1) < 0) + toksize -= 1; + if (0 > toksize) return (G_BAD_TOK_HEADER); if (*buf++ != 0x06) return (G_BAD_TOK_HEADER); - if ((toksize -= 1) < 0) + toksize -= 1; + if (0 > toksize) return (G_BAD_TOK_HEADER); toid.len = *buf++; - if ((toksize -= toid.len) < 0) + toksize -= toid.len; + if (0 > toksize) return (G_BAD_TOK_HEADER); toid.data = buf; buf += toid.len; @@ -231,7 +237,8 @@ __u32 g_verify_token_header(rawobj_t *mech, int *body_size, * important to return G_BAD_TOK_HEADER if the token header is * in fact bad */ - if ((toksize -= 2) < 0) + toksize -= 2; + if (0 > toksize) return (G_BAD_TOK_HEADER); if (ret) @@ -256,24 +263,29 @@ __u32 g_get_mech_oid(rawobj_t *mech, rawobj_t *in_buf) int ret = 0; int seqsize; - if ((len -= 1) < 0) + len -= 1; + if (0 > len) return (G_BAD_TOK_HEADER); if (*buf++ != 0x60) return (G_BAD_TOK_HEADER); - if ((seqsize = der_read_length(&buf, &len)) < 0) + seqsize = der_read_length(&buf, &len); + if (seqsize < 0) return (G_BAD_TOK_HEADER); - if ((len -= 1) < 0) + len -= 1; + if (0 > len) return (G_BAD_TOK_HEADER); if (*buf++ != 0x06) return (G_BAD_TOK_HEADER); - if ((len -= 1) < 0) + len -= 1; + if (0 > len) return (G_BAD_TOK_HEADER); mech->len = *buf++; - if ((len -= mech->len) < 0) + len -= mech->len; + if (0 > len) return (G_BAD_TOK_HEADER); OBD_ALLOC_LARGE(mech->data, mech->len); if (!mech->data) diff --git a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_keyring.c b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_keyring.c index d43a13c69669..4642bbfb9273 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_keyring.c +++ b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_keyring.c @@ -1176,7 +1176,7 @@ int gss_kt_instantiate(struct key *key, const void *data, size_t datalen) /* * called with key semaphore write locked. it means we can operate - * on the context without fear of loosing refcount. + * on the context without fear of losing refcount. */ static int gss_kt_update(struct key *key, const void *data, size_t datalen) diff --git a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_krb5_mech.c b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_krb5_mech.c index b9fa3b4a40db..d03f6c114171 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_krb5_mech.c +++ b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_krb5_mech.c @@ -54,7 +54,6 @@ #include <linux/slab.h> #include <linux/crypto.h> #include <linux/mutex.h> -#include <linux/crypto.h> #include <obd.h> #include <obd_class.h> @@ -679,7 +678,8 @@ __s32 krb5_make_checksum(__u32 enctype, __u32 code = GSS_S_FAILURE; int rc; - if (!(tfm = ll_crypto_alloc_hash(ke->ke_hash_name, 0, 0))) { + tfm = ll_crypto_alloc_hash(ke->ke_hash_name, 0, 0); + if (!tfm) { CERROR("failed to alloc TFM: %s\n", ke->ke_hash_name); return GSS_S_FAILURE; } diff --git a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_pipefs.c b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_pipefs.c index c624518c181a..7a1ff4fecbe3 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_pipefs.c +++ b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_pipefs.c @@ -85,7 +85,7 @@ static void gss_sec_pipe_upcall_fini(struct gss_sec *gsec) } /**************************************** - * internel context helpers * + * internal context helpers * ****************************************/ static @@ -652,7 +652,7 @@ __u32 mech_name2idx(const char *name) /* pipefs dentries for each mechanisms */ static struct dentry *de_pipes[MECH_MAX] = { NULL, }; -/* all upcall messgaes linked here */ +/* all upcall messages linked here */ static struct list_head upcall_lists[MECH_MAX]; /* and protected by this */ static spinlock_t upcall_locks[MECH_MAX]; diff --git a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_svc_upcall.c b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_svc_upcall.c index 5b5365b4629f..359c48ec2f5b 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/gss/gss_svc_upcall.c +++ b/drivers/staging/lustre/lustre/ptlrpc/gss/gss_svc_upcall.c @@ -102,7 +102,7 @@ static inline unsigned long hash_mem(char *buf, int length, int bits) len++; if ((len & (BITS_PER_LONG/8-1)) == 0) - hash = cfs_hash_long(hash^l, BITS_PER_LONG); + hash = hash_long(hash^l, BITS_PER_LONG); } while (len); return hash >> (BITS_PER_LONG - bits); @@ -586,7 +586,7 @@ static int rsc_parse(struct cache_detail *cd, char *mesg, int mlen) goto out; /* currently the expiry time passed down from user-space - * is invalid, here we retrive it from mech. */ + * is invalid, here we retrieve it from mech. */ if (lgss_inquire_context(rsci.ctx.gsc_mechctx, &ctx_expiry)) { CERROR("unable to get expire time, drop it\n"); goto out; @@ -880,7 +880,7 @@ int gss_svc_upcall_handle_init(struct ptlrpc_request *req, cache_get(&rsip->h); /* take an extra ref */ init_waitqueue_head(&rsip->waitq); - init_waitqueue_entry_current(&wait); + init_waitqueue_entry(&wait, current); add_wait_queue(&rsip->waitq, &wait); cache_check: @@ -1067,7 +1067,7 @@ int __init gss_init_svc_upcall(void) * the init upcall channel, otherwise there's big chance that the first * upcall issued before the channel be opened thus nfsv4 cache code will * drop the request direclty, thus lead to unnecessary recovery time. - * here we wait at miximum 1.5 seconds. */ + * here we wait at maximum 1.5 seconds. */ for (i = 0; i < 6; i++) { if (atomic_read(&rsi_cache.readers) > 0) break; diff --git a/drivers/staging/lustre/lustre/ptlrpc/gss/sec_gss.c b/drivers/staging/lustre/lustre/ptlrpc/gss/sec_gss.c index 8ce6271a5daa..383601cdd4e6 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/gss/sec_gss.c +++ b/drivers/staging/lustre/lustre/ptlrpc/gss/sec_gss.c @@ -382,7 +382,7 @@ void gss_cli_ctx_uptodate(struct gss_cli_ctx *gctx) /* At this point this ctx might have been marked as dead by * someone else, in which case nobody will make further use * of it. we don't care, and mark it UPTODATE will help - * destroying server side context when it be destroied. */ + * destroying server side context when it be destroyed. */ set_bit(PTLRPC_CTX_UPTODATE_BIT, &ctx->cc_flags); if (sec_is_reverse(ctx->cc_sec)) { @@ -676,7 +676,7 @@ redo: * lead to the sequence number fall behind the window on server and * be dropped. also applies to gss_cli_ctx_seal(). * - * Note: null mode dosen't check sequence number. */ + * Note: null mode doesn't check sequence number. */ if (svc != SPTLRPC_SVC_NULL && atomic_read(&gctx->gc_seq) - seq > GSS_SEQ_REPACK_THRESHOLD) { int behind = atomic_read(&gctx->gc_seq) - seq; @@ -1215,7 +1215,7 @@ int gss_cli_ctx_fini_common(struct ptlrpc_sec *sec, /* * remove UPTODATE flag of reverse ctx thus we won't send fini rpc, * this is to avoid potential problems of client side reverse svc ctx - * be mis-destroyed in various recovery senarios. anyway client can + * be mis-destroyed in various recovery scenarios. anyway client can * manage its reverse ctx well by associating it with its buddy ctx. */ if (sec_is_reverse(sec)) @@ -1882,7 +1882,7 @@ int gss_svc_sign(struct ptlrpc_request *req, LASSERT(rs->rs_msg == lustre_msg_buf(rs->rs_repbuf, 1, 0)); - /* embedded lustre_msg might have been shrinked */ + /* embedded lustre_msg might have been shrunk */ if (req->rq_replen != rs->rs_repbuf->lm_buflens[1]) lustre_shrink_msg(rs->rs_repbuf, 1, req->rq_replen, 1); @@ -2596,7 +2596,7 @@ static int gss_svc_seal(struct ptlrpc_request *req, int msglen, rc; /* get clear data length. note embedded lustre_msg might - * have been shrinked */ + * have been shrunk */ if (req->rq_replen != lustre_msg_buflen(rs->rs_repbuf, 0)) msglen = lustre_shrink_msg(rs->rs_repbuf, 0, req->rq_replen, 1); else @@ -2765,7 +2765,7 @@ int gss_copy_rvc_cli_ctx(struct ptlrpc_cli_ctx *cli_ctx, * replay. * * each reverse root ctx will record its latest sequence number on its - * buddy svcctx before be destroied, so here we continue use it. + * buddy svcctx before be destroyed, so here we continue use it. */ atomic_set(&cli_gctx->gc_seq, svc_gctx->gsc_rvs_seq); @@ -2836,7 +2836,7 @@ int __init sptlrpc_gss_init(void) if (rc) goto out_svc_upcall; - /* register policy after all other stuff be intialized, because it + /* register policy after all other stuff be initialized, because it * might be in used immediately after the registration. */ rc = gss_init_keyring(); diff --git a/drivers/staging/lustre/lustre/ptlrpc/import.c b/drivers/staging/lustre/lustre/ptlrpc/import.c index f465547eb95e..537aa6204a51 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/import.c +++ b/drivers/staging/lustre/lustre/ptlrpc/import.c @@ -560,17 +560,30 @@ static int ptlrpc_first_transno(struct obd_import *imp, __u64 *transno) struct ptlrpc_request *req; struct list_head *tmp; - if (list_empty(&imp->imp_replay_list)) - return 0; - tmp = imp->imp_replay_list.next; - req = list_entry(tmp, struct ptlrpc_request, rq_replay_list); - *transno = req->rq_transno; - if (req->rq_transno == 0) { - DEBUG_REQ(D_ERROR, req, "zero transno in replay"); - LBUG(); + /* The requests in committed_list always have smaller transnos than + * the requests in replay_list */ + if (!list_empty(&imp->imp_committed_list)) { + tmp = imp->imp_committed_list.next; + req = list_entry(tmp, struct ptlrpc_request, rq_replay_list); + *transno = req->rq_transno; + if (req->rq_transno == 0) { + DEBUG_REQ(D_ERROR, req, + "zero transno in committed_list"); + LBUG(); + } + return 1; } - - return 1; + if (!list_empty(&imp->imp_replay_list)) { + tmp = imp->imp_replay_list.next; + req = list_entry(tmp, struct ptlrpc_request, rq_replay_list); + *transno = req->rq_transno; + if (req->rq_transno == 0) { + DEBUG_REQ(D_ERROR, req, "zero transno in replay_list"); + LBUG(); + } + return 1; + } + return 0; } /** @@ -1042,7 +1055,7 @@ finish: if ((ocd->ocd_cksum_types & cksum_types_supported_client()) == 0) { LCONSOLE_WARN("The negotiation of the checksum " - "alogrithm to use with server %s " + "algorithm to use with server %s " "failed (%x/%x), disabling " "checksums\n", obd2cli_tgt(imp->imp_obd), @@ -1260,7 +1273,7 @@ static int ptlrpc_invalidate_import_thread(void *data) /** * This is the state machine for client-side recovery on import. * - * Typicaly we have two possibly paths. If we came to server and it is not + * Typically we have two possibly paths. If we came to server and it is not * in recovery, we just enter IMP_EVICTED state, invalidate our import * state and reconnect from scratch. * If we came to server that is in recovery, we enter IMP_REPLAY import state. diff --git a/drivers/staging/lustre/lustre/ptlrpc/layout.c b/drivers/staging/lustre/lustre/ptlrpc/layout.c index dfcb410fe485..41c12e00129f 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/layout.c +++ b/drivers/staging/lustre/lustre/ptlrpc/layout.c @@ -295,7 +295,8 @@ static const struct req_msg_field *mds_reint_setxattr_client[] = { &RMF_REC_REINT, &RMF_CAPA1, &RMF_NAME, - &RMF_EADATA + &RMF_EADATA, + &RMF_DLM_REQ }; static const struct req_msg_field *mdt_swap_layouts[] = { @@ -2154,7 +2155,7 @@ EXPORT_SYMBOL(req_capsule_server_sized_swab_get); * request (if the caller is executing on the server-side) or reply (if the * caller is executing on the client-side). * - * This function convienient for use is code that could be executed on the + * This function convenient for use is code that could be executed on the * client and server alike. */ const void *req_capsule_other_get(struct req_capsule *pill, diff --git a/drivers/staging/lustre/lustre/ptlrpc/lproc_ptlrpc.c b/drivers/staging/lustre/lustre/ptlrpc/lproc_ptlrpc.c index 1be978609c59..58f1c8bf25b0 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/lproc_ptlrpc.c +++ b/drivers/staging/lustre/lustre/ptlrpc/lproc_ptlrpc.c @@ -616,7 +616,7 @@ out: } /** - * The longest valid command string is the maxium policy name size, plus the + * The longest valid command string is the maximum policy name size, plus the * length of the " reg" substring */ #define LPROCFS_NRS_WR_MAX_CMD (NRS_POL_NAME_MAX + sizeof(" reg") - 1) @@ -1184,7 +1184,7 @@ int lprocfs_wr_evict_client(struct file *file, const char *buffer, } tmpbuf = cfs_firststr(kbuf, min_t(unsigned long, BUFLEN - 1, count)); /* Kludge code(deadlock situation): the lprocfs lock has been held - * since the client is evicted by writting client's + * since the client is evicted by writing client's * uuid/nid to procfs "evict_client" entry. However, * obd_export_evict_by_uuid() will call lprocfs_remove() to destroy * the proc entries under the being destroyed export{}, so I have diff --git a/drivers/staging/lustre/lustre/ptlrpc/niobuf.c b/drivers/staging/lustre/lustre/ptlrpc/niobuf.c index 3c6bf23415f9..a47a8d807d5b 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/niobuf.c +++ b/drivers/staging/lustre/lustre/ptlrpc/niobuf.c @@ -349,7 +349,7 @@ static void ptlrpc_at_set_reply(struct ptlrpc_request *req, int flags) /** * Send request reply from request \a req reply buffer. * \a flags defines reply types - * Returns 0 on sucess or error code + * Returns 0 on success or error code */ int ptlrpc_send_reply(struct ptlrpc_request *req, int flags) { @@ -389,7 +389,7 @@ int ptlrpc_send_reply(struct ptlrpc_request *req, int flags) * ptlrpc_body in reply buffer to ptlrpc_body_v2, otherwise, the * reply buffer on client will be overflow. * - * XXX Remove this whenver we drop the interoprability with such client. + * XXX Remove this whenever we drop the interoprability with such client. */ req->rq_replen = lustre_shrink_msg(req->rq_repmsg, 0, sizeof(struct ptlrpc_body_v2), 1); @@ -511,7 +511,9 @@ int ptl_send_rpc(struct ptlrpc_request *request, int noreply) CDEBUG(D_HA, "muting rpc for failed imp obd %s\n", request->rq_import->imp_obd->obd_name); /* this prevents us from waiting in ptlrpc_queue_wait */ + spin_lock(&request->rq_lock); request->rq_err = 1; + spin_unlock(&request->rq_lock); request->rq_status = -ENODEV; return -ENODEV; } @@ -553,7 +555,9 @@ int ptl_send_rpc(struct ptlrpc_request *request, int noreply) if (rc) { /* this prevents us from looping in * ptlrpc_queue_wait */ + spin_lock(&request->rq_lock); request->rq_err = 1; + spin_unlock(&request->rq_lock); request->rq_status = rc; GOTO(cleanup_bulk, rc); } diff --git a/drivers/staging/lustre/lustre/ptlrpc/nrs.c b/drivers/staging/lustre/lustre/ptlrpc/nrs.c index 0abcd6d82273..bcba1c8e8693 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/nrs.c +++ b/drivers/staging/lustre/lustre/ptlrpc/nrs.c @@ -1322,7 +1322,7 @@ EXPORT_SYMBOL(ptlrpc_nrs_policy_unregister); * Setup NRS heads on all service partitions of service \a svc, and register * all compatible policies on those NRS heads. * - * To be called from withing ptl + * To be called from within ptl * \param[in] svc the service to setup * * \retval -ve error, the calling logic should eventually call @@ -1736,7 +1736,7 @@ fail: } /** - * Removes all policy desciptors from nrs_core::nrs_policies, and frees the + * Removes all policy descriptors from nrs_core::nrs_policies, and frees the * policy descriptors. * * Since all PTLRPC services are stopped at this point, there are no more diff --git a/drivers/staging/lustre/lustre/ptlrpc/pack_generic.c b/drivers/staging/lustre/lustre/ptlrpc/pack_generic.c index 464479c0f00b..45c0b84621e4 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/pack_generic.c +++ b/drivers/staging/lustre/lustre/ptlrpc/pack_generic.c @@ -123,7 +123,7 @@ int lustre_msg_early_size(void) * with the old client (< 2.3) which doesn't have pb_jobid * in the ptlrpc_body. * - * XXX Remove this whenever we dorp interoprability with such + * XXX Remove this whenever we drop interoprability with such * client. */ __u32 pblen = sizeof(struct ptlrpc_body_v2); @@ -244,15 +244,7 @@ int lustre_pack_request(struct ptlrpc_request *req, __u32 magic, int count, LASSERT(lens[MSG_PTLRPC_BODY_OFF] == sizeof(struct ptlrpc_body)); /* only use new format, we don't need to be compatible with 1.4 */ - magic = LUSTRE_MSG_MAGIC_V2; - - switch (magic) { - case LUSTRE_MSG_MAGIC_V2: - return lustre_pack_request_v2(req, count, lens, bufs); - default: - LASSERTF(0, "incorrect message magic: %08x\n", magic); - return -EINVAL; - } + return lustre_pack_request_v2(req, count, lens, bufs); } EXPORT_SYMBOL(lustre_pack_request); @@ -1545,7 +1537,7 @@ void lustre_msg_set_jobid(struct lustre_msg *msg, char *jobid) __u32 opc = lustre_msg_get_opc(msg); struct ptlrpc_body *pb; - /* Don't set jobid for ldlm ast RPCs, they've been shrinked. + /* Don't set jobid for ldlm ast RPCs, they've been shrunk. * See the comment in ptlrpc_request_pack(). */ if (!opc || opc == LDLM_BL_CALLBACK || opc == LDLM_CP_CALLBACK || opc == LDLM_GL_CALLBACK) diff --git a/drivers/staging/lustre/lustre/ptlrpc/ptlrpcd.c b/drivers/staging/lustre/lustre/ptlrpc/ptlrpcd.c index 2d26fd543d46..ca734ce079c1 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/ptlrpcd.c +++ b/drivers/staging/lustre/lustre/ptlrpc/ptlrpcd.c @@ -229,7 +229,7 @@ void ptlrpcd_add_req(struct ptlrpc_request *req, pdl_policy_t policy, int idx) spin_unlock(&req->rq_lock); l_wait_event(req->rq_set_waitq, (req->rq_set == NULL), &lwi); } else if (req->rq_set) { - /* If we have a vaid "rq_set", just reuse it to avoid double + /* If we have a valid "rq_set", just reuse it to avoid double * linked. */ LASSERT(req->rq_phase == RQ_PHASE_NEW); LASSERT(req->rq_send_state == LUSTRE_IMP_REPLAY); @@ -471,7 +471,7 @@ static int ptlrpcd(void *arg) * be better. But it breaks former data transfer policy. * * So we shouldn't be blind for avoiding the data transfer. We make some - * compromise: divide the ptlrpcd threds pool into two parts. One part is + * compromise: divide the ptlrpcd threads pool into two parts. One part is * for bound mode, each ptlrpcd thread in this part is bound to some CPU * core. The other part is for free mode, all the ptlrpcd threads in the * part can be scheduled on any CPU core. We specify some partnership diff --git a/drivers/staging/lustre/lustre/ptlrpc/recover.c b/drivers/staging/lustre/lustre/ptlrpc/recover.c index 84c39e083ea4..48ae328ce24e 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/recover.c +++ b/drivers/staging/lustre/lustre/ptlrpc/recover.c @@ -105,24 +105,59 @@ int ptlrpc_replay_next(struct obd_import *imp, int *inflight) * imp_lock is being held by ptlrpc_replay, but it's not. it's * just a little race... */ - list_for_each_safe(tmp, pos, &imp->imp_replay_list) { + + /* Replay all the committed open requests on committed_list first */ + if (!list_empty(&imp->imp_committed_list)) { + tmp = imp->imp_committed_list.prev; req = list_entry(tmp, struct ptlrpc_request, rq_replay_list); - /* If need to resend the last sent transno (because a - reconnect has occurred), then stop on the matching - req and send it again. If, however, the last sent - transno has been committed then we continue replay - from the next request. */ + /* The last request on committed_list hasn't been replayed */ if (req->rq_transno > last_transno) { - if (imp->imp_resend_replay) - lustre_msg_add_flags(req->rq_reqmsg, - MSG_RESENT); - break; + /* Since the imp_committed_list is immutable before + * all of it's requests being replayed, it's safe to + * use a cursor to accelerate the search */ + imp->imp_replay_cursor = imp->imp_replay_cursor->next; + + while (imp->imp_replay_cursor != + &imp->imp_committed_list) { + req = list_entry(imp->imp_replay_cursor, + struct ptlrpc_request, + rq_replay_list); + if (req->rq_transno > last_transno) + break; + + req = NULL; + imp->imp_replay_cursor = + imp->imp_replay_cursor->next; + } + } else { + /* All requests on committed_list have been replayed */ + imp->imp_replay_cursor = &imp->imp_committed_list; + req = NULL; + } + } + + /* All the requests in committed list have been replayed, let's replay + * the imp_replay_list */ + if (req == NULL) { + list_for_each_safe(tmp, pos, &imp->imp_replay_list) { + req = list_entry(tmp, struct ptlrpc_request, + rq_replay_list); + + if (req->rq_transno > last_transno) + break; + req = NULL; } - req = NULL; } + /* If need to resend the last sent transno (because a reconnect + * has occurred), then stop on the matching req and send it again. + * If, however, the last sent transno has been committed then we + * continue replay from the next request. */ + if (req != NULL && imp->imp_resend_replay) + lustre_msg_add_flags(req->rq_reqmsg, MSG_RESENT); + spin_lock(&imp->imp_lock); imp->imp_resend_replay = 0; spin_unlock(&imp->imp_lock); diff --git a/drivers/staging/lustre/lustre/ptlrpc/sec.c b/drivers/staging/lustre/lustre/ptlrpc/sec.c index 962b31d163de..d80418057f29 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/sec.c +++ b/drivers/staging/lustre/lustre/ptlrpc/sec.c @@ -543,8 +543,8 @@ int sptlrpc_req_replace_dead_ctx(struct ptlrpc_request *req) "ctx (%p, fl %lx) doesn't switch, relax a little bit\n", newctx, newctx->cc_flags); - schedule_timeout_and_set_state(TASK_INTERRUPTIBLE, - HZ); + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(HZ); } else { /* * it's possible newctx == oldctx if we're switching @@ -779,7 +779,7 @@ again: * e.g. ptlrpc_abort_inflight(); */ if (!cli_ctx_is_refreshed(ctx)) { - /* timed out or interruptted */ + /* timed out or interrupted */ req_off_ctx_list(req, ctx); LASSERT(rc != 0); @@ -805,7 +805,7 @@ void sptlrpc_req_set_flavor(struct ptlrpc_request *req, int opcode) LASSERT(req->rq_cli_ctx->cc_sec); LASSERT(req->rq_bulk_read == 0 || req->rq_bulk_write == 0); - /* special security flags accoding to opcode */ + /* special security flags according to opcode */ switch (opcode) { case OST_READ: case MDS_READPAGE: @@ -1218,7 +1218,7 @@ static void sec_cop_destroy_sec(struct ptlrpc_sec *sec) LASSERT_ATOMIC_ZERO(&sec->ps_nctx); LASSERT(policy->sp_cops->destroy_sec); - CDEBUG(D_SEC, "%s@%p: being destroied\n", sec->ps_policy->sp_name, sec); + CDEBUG(D_SEC, "%s@%p: being destroyed\n", sec->ps_policy->sp_name, sec); policy->sp_cops->destroy_sec(sec); sptlrpc_policy_put(policy); @@ -1264,7 +1264,7 @@ void sptlrpc_sec_put(struct ptlrpc_sec *sec) EXPORT_SYMBOL(sptlrpc_sec_put); /* - * policy module is responsible for taking refrence of import + * policy module is responsible for taking reference of import */ static struct ptlrpc_sec * sptlrpc_sec_create(struct obd_import *imp, @@ -1419,7 +1419,7 @@ int sptlrpc_import_sec_adapt(struct obd_import *imp, sp = imp->imp_obd->u.cli.cl_sp_me; } else { - /* reverse import, determine flavor from incoming reqeust */ + /* reverse import, determine flavor from incoming request */ sf = *flvr; if (sf.sf_rpc != SPTLRPC_FLVR_NULL) @@ -2057,7 +2057,7 @@ int sptlrpc_svc_unwrap_request(struct ptlrpc_request *req) /* * if it's not null flavor (which means embedded packing msg), - * reset the swab mask for the comming inner msg unpacking. + * reset the swab mask for the coming inner msg unpacking. */ if (SPTLRPC_FLVR_POLICY(req->rq_flvr.sf_rpc) != SPTLRPC_POLICY_NULL) req->rq_req_swab_mask = 0; @@ -2108,7 +2108,7 @@ int sptlrpc_svc_alloc_rs(struct ptlrpc_request *req, int msglen) /** * Used by ptlrpc server, to perform transformation upon reply message. * - * \post req->rq_reply_off is set to approriate server-controlled reply offset. + * \post req->rq_reply_off is set to appropriate server-controlled reply offset. * \post req->rq_repmsg and req->rq_reply_state->rs_msg becomes inaccessible. */ int sptlrpc_svc_wrap_reply(struct ptlrpc_request *req) diff --git a/drivers/staging/lustre/lustre/ptlrpc/sec_bulk.c b/drivers/staging/lustre/lustre/ptlrpc/sec_bulk.c index 316103ab7c3c..965668132747 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/sec_bulk.c +++ b/drivers/staging/lustre/lustre/ptlrpc/sec_bulk.c @@ -113,7 +113,7 @@ static struct ptlrpc_enc_page_pool { unsigned long epp_st_missings; /* # of cache missing */ unsigned long epp_st_lowfree; /* lowest free pages reached */ unsigned int epp_st_max_wqlen; /* highest waitqueue length */ - cfs_time_t epp_st_max_wait; /* in jeffies */ + cfs_time_t epp_st_max_wait; /* in jiffies */ /* * pointers to pools */ @@ -545,11 +545,11 @@ again: page_pools.epp_waitqlen; set_current_state(TASK_UNINTERRUPTIBLE); - init_waitqueue_entry_current(&waitlink); + init_waitqueue_entry(&waitlink, current); add_wait_queue(&page_pools.epp_waitq, &waitlink); spin_unlock(&page_pools.epp_lock); - waitq_wait(&waitlink, TASK_UNINTERRUPTIBLE); + schedule(); remove_wait_queue(&page_pools.epp_waitq, &waitlink); LASSERT(page_pools.epp_waitqlen > 0); spin_lock(&page_pools.epp_lock); diff --git a/drivers/staging/lustre/lustre/ptlrpc/sec_config.c b/drivers/staging/lustre/lustre/ptlrpc/sec_config.c index 6cc3f23c27cc..bf56120abfaf 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/sec_config.c +++ b/drivers/staging/lustre/lustre/ptlrpc/sec_config.c @@ -255,7 +255,7 @@ void sptlrpc_rule_set_free(struct sptlrpc_rule_set *rset) EXPORT_SYMBOL(sptlrpc_rule_set_free); /* - * return 0 if the rule set could accomodate one more rule. + * return 0 if the rule set could accommodate one more rule. */ int sptlrpc_rule_set_expand(struct sptlrpc_rule_set *rset) { diff --git a/drivers/staging/lustre/lustre/ptlrpc/service.c b/drivers/staging/lustre/lustre/ptlrpc/service.c index 590fa8df8b7f..192adec5382a 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/service.c +++ b/drivers/staging/lustre/lustre/ptlrpc/service.c @@ -177,7 +177,7 @@ ptlrpc_grow_req_bufs(struct ptlrpc_service_part *svcpt, int post) /** * Part of Rep-Ack logic. - * Puts a lock and its mode into reply state assotiated to request reply. + * Puts a lock and its mode into reply state associated to request reply. */ void ptlrpc_save_lock(struct ptlrpc_request *req, @@ -252,7 +252,7 @@ struct rs_batch { static struct ptlrpc_hr_service ptlrpc_hr; /** - * maximum mumber of replies scheduled in one batch + * maximum number of replies scheduled in one batch */ #define MAX_SCHEDULED 256 @@ -612,7 +612,7 @@ ptlrpc_service_part_init(struct ptlrpc_service *svc, INIT_LIST_HEAD(&svcpt->scp_hist_reqs); INIT_LIST_HEAD(&svcpt->scp_hist_rqbds); - /* acitve requests and hp requests */ + /* active requests and hp requests */ spin_lock_init(&svcpt->scp_req_lock); /* reply states */ @@ -752,7 +752,7 @@ ptlrpc_register_service(struct ptlrpc_service_conf *conf, spin_lock_init(&service->srv_lock); service->srv_name = conf->psc_name; service->srv_watchdog_factor = conf->psc_watchdog_factor; - INIT_LIST_HEAD(&service->srv_list); /* for safty of cleanup */ + INIT_LIST_HEAD(&service->srv_list); /* for safety of cleanup */ /* buffer configuration */ service->srv_nbuf_per_group = test_req_buffer_pressure ? @@ -1982,7 +1982,7 @@ put_conn: do_gettimeofday(&work_end); timediff = cfs_timeval_sub(&work_end, &work_start, NULL); CDEBUG(D_RPCTRACE, "Handled RPC pname:cluuid+ref:pid:xid:nid:opc " - "%s:%s+%d:%d:x"LPU64":%s:%d Request procesed in " + "%s:%s+%d:%d:x"LPU64":%s:%d Request processed in " "%ldus (%ldus total) trans "LPU64" rc %d/%d\n", current_comm(), (request->rq_export ? @@ -2736,7 +2736,7 @@ int ptlrpc_start_thread(struct ptlrpc_service_part *svcpt, int wait) spin_lock(&svcpt->scp_lock); --svcpt->scp_nthrs_starting; if (thread_is_stopping(thread)) { - /* this ptlrpc_thread is being hanled + /* this ptlrpc_thread is being handled * by ptlrpc_svcpt_stop_threads now */ thread_add_flags(thread, SVC_STOPPED); diff --git a/drivers/staging/lustre/lustre/ptlrpc/wiretest.c b/drivers/staging/lustre/lustre/ptlrpc/wiretest.c index 3aa445952024..3c8846006a7b 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/wiretest.c +++ b/drivers/staging/lustre/lustre/ptlrpc/wiretest.c @@ -1152,6 +1152,8 @@ void lustre_assert_wire_constants(void) OBD_CONNECT_SHORTIO); LASSERTF(OBD_CONNECT_PINGLESS == 0x4000000000000ULL, "found 0x%.16llxULL\n", OBD_CONNECT_PINGLESS); + LASSERTF(OBD_CONNECT_FLOCK_DEAD == 0x8000000000000ULL, + "found 0x%.16llxULL\n", OBD_CONNECT_FLOCK_DEAD); LASSERTF(OBD_CKSUM_CRC32 == 0x00000001UL, "found 0x%.8xUL\n", (unsigned)OBD_CKSUM_CRC32); LASSERTF(OBD_CKSUM_ADLER == 0x00000002UL, "found 0x%.8xUL\n", |