/* nvram_write_error_log
*
* We need to buffer the error logs into nvram to ensure that we have
* the failure information to decode.
*/
int nvram_write_error_log(char * buff, int length,
unsigned int err_type, unsigned int error_log_cnt)
{
int rc = nvram_write_os_partition(&rtas_log_partition, buff, length,
err_type, error_log_cnt);
if (!rc) {
last_unread_rtas_event = ktime_get_real_seconds();
#ifdef CONFIG_PSTORE
last_rtas_event = ktime_get_real_seconds();
#endif
}
return rc;
}
static int test_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm)
{
struct rtc_test_data *rtd = dev_get_drvdata(dev);
ktime_t timeout;
u64 expires;
timeout = rtc_tm_to_time64(&alrm->time) - ktime_get_real_seconds();
timeout -= rtd->offset;
del_timer(&rtd->alarm);
expires = jiffies + timeout * HZ;
if (expires > U32_MAX)
expires = U32_MAX;
pr_err("ABE: %s +%d %s\n", __FILE__, __LINE__, __func__);
rtd->alarm.expires = expires;
if (alrm->enabled)
add_timer(&rtd->alarm);
rtd->alarm_en = alrm->enabled;
return 0;
}
static int llog_read_header(const struct lu_env *env,
struct llog_handle *handle,
struct obd_uuid *uuid)
{
struct llog_operations *lop;
int rc;
rc = llog_handle2ops(handle, &lop);
if (rc)
return rc;
if (lop->lop_read_header == NULL)
return -EOPNOTSUPP;
rc = lop->lop_read_header(env, handle);
if (rc == LLOG_EEMPTY) {
struct llog_log_hdr *llh = handle->lgh_hdr;
handle->lgh_last_idx = 0; /* header is record with index 0 */
llh->llh_count = 1; /* for the header record */
llh->llh_hdr.lrh_type = LLOG_HDR_MAGIC;
llh->llh_hdr.lrh_len = llh->llh_tail.lrt_len = LLOG_CHUNK_SIZE;
llh->llh_hdr.lrh_index = llh->llh_tail.lrt_index = 0;
llh->llh_timestamp = ktime_get_real_seconds();
if (uuid)
memcpy(&llh->llh_tgtuuid, uuid,
sizeof(llh->llh_tgtuuid));
llh->llh_bitmap_offset = offsetof(typeof(*llh), llh_bitmap);
ext2_set_bit(0, llh->llh_bitmap);
rc = 0;
}
return rc;
}
/**
* Disconnect a bulk desc from the network. Idempotent. Not
* thread-safe (i.e. only interlocks with completion callback).
* Returns 1 on success or 0 if network unregistration failed for whatever
* reason.
*/
int ptlrpc_unregister_bulk(struct ptlrpc_request *req, int async)
{
struct ptlrpc_bulk_desc *desc = req->rq_bulk;
wait_queue_head_t *wq;
int rc;
LASSERT(!in_interrupt()); /* might sleep */
/* Let's setup deadline for reply unlink. */
if (OBD_FAIL_CHECK(OBD_FAIL_PTLRPC_LONG_BULK_UNLINK) &&
async && req->rq_bulk_deadline == 0 && cfs_fail_val == 0)
req->rq_bulk_deadline = ktime_get_real_seconds() + LONG_UNLINK;
if (ptlrpc_client_bulk_active(req) == 0) /* completed or */
return 1; /* never registered */
LASSERT(desc->bd_req == req); /* bd_req NULL until registered */
/* the unlink ensures the callback happens ASAP and is the last
* one. If it fails, it must be because completion just happened,
* but we must still wait_event() in this case to give liblustre
* a chance to run client_bulk_callback()
*/
mdunlink_iterate_helper(desc->bd_mds, desc->bd_md_max_brw);
if (ptlrpc_client_bulk_active(req) == 0) /* completed or */
return 1; /* never registered */
/* Move to "Unregistering" phase as bulk was not unlinked yet. */
ptlrpc_rqphase_move(req, RQ_PHASE_UNREG_BULK);
/* Do not wait for unlink to finish. */
if (async)
return 0;
if (req->rq_set)
wq = &req->rq_set->set_waitq;
else
wq = &req->rq_reply_waitq;
for (;;) {
/* Network access will complete in finite time but the HUGE
* timeout lets us CWARN for visibility of sluggish LNDs
*/
int cnt = 0;
while (cnt < LONG_UNLINK &&
(rc = wait_event_idle_timeout(*wq,
!ptlrpc_client_bulk_active(req),
HZ)) == 0)
cnt += 1;
if (rc > 0) {
ptlrpc_rqphase_move(req, req->rq_next_phase);
return 1;
}
DEBUG_REQ(D_WARNING, req, "Unexpectedly long timeout: desc %p",
desc);
}
return 0;
}
/*
* Are we using the ibm,rtas-log for oops/panic reports? And if so,
* would logging this oops/panic overwrite an RTAS event that rtas_errd
* hasn't had a chance to read and process? Return 1 if so, else 0.
*
* We assume that if rtas_errd hasn't read the RTAS event in
* NVRAM_RTAS_READ_TIMEOUT seconds, it's probably not going to.
*/
int clobbering_unread_rtas_event(void)
{
return (oops_log_partition.index == rtas_log_partition.index
&& last_unread_rtas_event
&& ktime_get_real_seconds() - last_unread_rtas_event <=
NVRAM_RTAS_READ_TIMEOUT);
}
void btrfs_dev_replace_suspend_for_unmount(struct btrfs_fs_info *fs_info)
{
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
mutex_lock(&dev_replace->lock_finishing_cancel_unmount);
down_write(&dev_replace->rwsem);
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
dev_replace->replace_state =
BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED;
dev_replace->time_stopped = ktime_get_real_seconds();
dev_replace->item_needs_writeback = 1;
btrfs_info(fs_info, "suspending dev_replace for unmount");
break;
}
up_write(&dev_replace->rwsem);
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
}
int nilfs_commit_super(struct super_block *sb, int flag)
{
struct the_nilfs *nilfs = sb->s_fs_info;
struct nilfs_super_block **sbp = nilfs->ns_sbp;
time64_t t;
/* nilfs->ns_sem must be locked by the caller. */
t = ktime_get_real_seconds();
nilfs->ns_sbwtime = t;
sbp[0]->s_wtime = cpu_to_le64(t);
sbp[0]->s_sum = 0;
sbp[0]->s_sum = cpu_to_le32(crc32_le(nilfs->ns_crc_seed,
(unsigned char *)sbp[0],
nilfs->ns_sbsize));
if (flag == NILFS_SB_COMMIT_ALL && sbp[1]) {
sbp[1]->s_wtime = sbp[0]->s_wtime;
sbp[1]->s_sum = 0;
sbp[1]->s_sum = cpu_to_le32(crc32_le(nilfs->ns_crc_seed,
(unsigned char *)sbp[1],
nilfs->ns_sbsize));
}
clear_nilfs_sb_dirty(nilfs);
nilfs->ns_flushed_device = 1;
/* make sure store to ns_flushed_device cannot be reordered */
smp_wmb();
return nilfs_sync_super(sb, flag);
}
static int test_rtc_set_mmss64(struct device *dev, time64_t secs)
{
struct rtc_test_data *rtd = dev_get_drvdata(dev);
rtd->offset = secs - ktime_get_real_seconds();
return 0;
}
static int test_rtc_read_time(struct device *dev, struct rtc_time *tm)
{
struct rtc_test_data *rtd = dev_get_drvdata(dev);
rtc_time64_to_tm(ktime_get_real_seconds() + rtd->offset, tm);
return 0;
}
static void setup_inj_struct(struct mce *m)
{
memset(m, 0, sizeof(struct mce));
m->cpuvendor = boot_cpu_data.x86_vendor;
m->time = ktime_get_real_seconds();
m->cpuid = cpuid_eax(1);
m->microcode = boot_cpu_data.microcode;
}
static int test_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm)
{
struct rtc_test_data *rtd = dev_get_drvdata(dev);
time64_t alarm;
alarm = (rtd->alarm.expires - jiffies) / HZ;
alarm += ktime_get_real_seconds() + rtd->offset;
rtc_time64_to_tm(alarm, &alrm->time);
alrm->enabled = rtd->alarm_en;
return 0;
}
/*
* CPER record ID need to be unique even after reboot, because record
* ID is used as index for ERST storage, while CPER records from
* multiple boot may co-exist in ERST.
*/
u64 cper_next_record_id(void)
{
static atomic64_t seq;
if (!atomic64_read(&seq)) {
time64_t time = ktime_get_real_seconds();
/*
* This code is unlikely to still be needed in year 2106,
* but just in case, let's use a few more bits for timestamps
* after y2038 to be sure they keep increasing monotonically
* for the next few hundred years...
*/
if (time < 0x80000000)
atomic64_set(&seq, (ktime_get_real_seconds()) << 32);
else
atomic64_set(&seq, 0x8000000000000000ull |
ktime_get_real_seconds() << 24);
}
return atomic64_inc_return(&seq);
}
static int osc_io_read_start(const struct lu_env *env,
const struct cl_io_slice *slice)
{
struct cl_object *obj = slice->cis_obj;
struct cl_attr *attr = &osc_env_info(env)->oti_attr;
int rc = 0;
if (!slice->cis_io->ci_noatime) {
cl_object_attr_lock(obj);
attr->cat_atime = ktime_get_real_seconds();
rc = cl_object_attr_set(env, obj, attr, CAT_ATIME);
cl_object_attr_unlock(obj);
}
return rc;
}
static int osc_io_write_start(const struct lu_env *env,
const struct cl_io_slice *slice)
{
struct cl_object *obj = slice->cis_obj;
struct cl_attr *attr = &osc_env_info(env)->oti_attr;
int rc = 0;
OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_DELAY_SETTIME, 1);
cl_object_attr_lock(obj);
attr->cat_mtime = attr->cat_ctime = ktime_get_real_seconds();
rc = cl_object_attr_set(env, obj, attr, CAT_MTIME | CAT_CTIME);
cl_object_attr_unlock(obj);
return rc;
}
/*
* Schedule a garbage collection run.
* - time precision isn't particularly important
*/
void key_schedule_gc(time64_t gc_at)
{
unsigned long expires;
time64_t now = ktime_get_real_seconds();
kenter("%lld", gc_at - now);
if (gc_at <= now || test_bit(KEY_GC_REAP_KEYTYPE, &key_gc_flags)) {
kdebug("IMMEDIATE");
schedule_work(&key_gc_work);
} else if (gc_at < key_gc_next_run) {
kdebug("DEFERRED");
key_gc_next_run = gc_at;
expires = jiffies + (gc_at - now) * HZ;
mod_timer(&key_gc_timer, expires);
}
}
static int osd_scrub_post(const struct lu_env *env, struct osd_device *dev,
int result)
{
struct lustre_scrub *scrub = &dev->od_scrub;
struct scrub_file *sf = &scrub->os_file;
int rc;
ENTRY;
CDEBUG(D_LFSCK, "%s: OI scrub post with result = %d\n",
scrub->os_name, result);
down_write(&scrub->os_rwsem);
spin_lock(&scrub->os_lock);
thread_set_flags(&scrub->os_thread, SVC_STOPPING);
spin_unlock(&scrub->os_lock);
if (scrub->os_new_checked > 0) {
sf->sf_items_checked += scrub->os_new_checked;
scrub->os_new_checked = 0;
sf->sf_pos_last_checkpoint = scrub->os_pos_current;
}
sf->sf_time_last_checkpoint = ktime_get_real_seconds();
if (result > 0) {
sf->sf_status = SS_COMPLETED;
if (!(sf->sf_param & SP_DRYRUN)) {
memset(sf->sf_oi_bitmap, 0, SCRUB_OI_BITMAP_SIZE);
sf->sf_flags &= ~(SF_RECREATED | SF_INCONSISTENT |
SF_UPGRADE | SF_AUTO);
}
sf->sf_time_last_complete = sf->sf_time_last_checkpoint;
sf->sf_success_count++;
} else if (result == 0) {
if (scrub->os_paused)
sf->sf_status = SS_PAUSED;
else
sf->sf_status = SS_STOPPED;
} else {
sf->sf_status = SS_FAILED;
}
sf->sf_run_time += ktime_get_seconds() -
scrub->os_time_last_checkpoint;
rc = scrub_file_store(env, scrub);
up_write(&scrub->os_rwsem);
RETURN(rc < 0 ? rc : result);
}
/**
* Dump Lustre log to ::debug_file_path by calling tracefile_dump_all_pages()
*/
void libcfs_debug_dumplog_internal(void *arg)
{
void *journal_info;
journal_info = current->journal_info;
current->journal_info = NULL;
if (strncmp(libcfs_debug_file_path_arr, "NONE", 4) != 0) {
snprintf(debug_file_name, sizeof(debug_file_name) - 1,
"%s.%lld.%ld", libcfs_debug_file_path_arr,
(s64)ktime_get_real_seconds(), (long_ptr_t)arg);
pr_alert("LustreError: dumping log to %s\n", debug_file_name);
cfs_tracefile_dump_all_pages(debug_file_name);
libcfs_run_debug_log_upcall(debug_file_name);
}
current->journal_info = journal_info;
}
static void ptlrpc_at_set_reply(struct ptlrpc_request *req, int flags)
{
struct ptlrpc_service_part *svcpt = req->rq_rqbd->rqbd_svcpt;
struct ptlrpc_service *svc = svcpt->scp_service;
int service_time = max_t(int, ktime_get_real_seconds() -
req->rq_arrival_time.tv_sec, 1);
if (!(flags & PTLRPC_REPLY_EARLY) &&
(req->rq_type != PTL_RPC_MSG_ERR) && req->rq_reqmsg &&
!(lustre_msg_get_flags(req->rq_reqmsg) &
(MSG_RESENT | MSG_REPLAY |
MSG_REQ_REPLAY_DONE | MSG_LOCK_REPLAY_DONE))) {
/* early replies, errors and recovery requests don't count
* toward our service time estimate
*/
int oldse = at_measured(&svcpt->scp_at_estimate, service_time);
if (oldse != 0) {
DEBUG_REQ(D_ADAPTTO, req,
"svc %s changed estimate from %d to %d",
svc->srv_name, oldse,
at_get(&svcpt->scp_at_estimate));
}
}
/* Report actual service time for client latency calc */
lustre_msg_set_service_time(req->rq_repmsg, service_time);
/* Report service time estimate for future client reqs, but report 0
* (to be ignored by client) if it's a error reply during recovery.
* (bz15815)
*/
if (req->rq_type == PTL_RPC_MSG_ERR && !req->rq_export)
lustre_msg_set_timeout(req->rq_repmsg, 0);
else
lustre_msg_set_timeout(req->rq_repmsg,
at_get(&svcpt->scp_at_estimate));
if (req->rq_reqmsg &&
!(lustre_msghdr_get_flags(req->rq_reqmsg) & MSGHDR_AT_SUPPORT)) {
CDEBUG(D_ADAPTTO, "No early reply support: flags=%#x req_flags=%#x magic=%x/%x len=%d\n",
flags, lustre_msg_get_flags(req->rq_reqmsg),
lustre_msg_get_magic(req->rq_reqmsg),
lustre_msg_get_magic(req->rq_repmsg), req->rq_replen);
}
}
static int llog_read_header(const struct lu_env *env,
struct llog_handle *handle,
struct obd_uuid *uuid)
{
struct llog_operations *lop;
int rc;
rc = llog_handle2ops(handle, &lop);
if (rc)
return rc;
if (!lop->lop_read_header)
return -EOPNOTSUPP;
rc = lop->lop_read_header(env, handle);
if (rc == LLOG_EEMPTY) {
struct llog_log_hdr *llh = handle->lgh_hdr;
size_t len;
/* lrh_len should be initialized in llog_init_handle */
handle->lgh_last_idx = 0; /* header is record with index 0 */
llh->llh_count = 1; /* for the header record */
llh->llh_hdr.lrh_type = LLOG_HDR_MAGIC;
LASSERT(handle->lgh_ctxt->loc_chunk_size >= LLOG_MIN_CHUNK_SIZE);
llh->llh_hdr.lrh_len = handle->lgh_ctxt->loc_chunk_size;
llh->llh_hdr.lrh_index = 0;
llh->llh_timestamp = ktime_get_real_seconds();
if (uuid)
memcpy(&llh->llh_tgtuuid, uuid,
sizeof(llh->llh_tgtuuid));
llh->llh_bitmap_offset = offsetof(typeof(*llh), llh_bitmap);
/*
* Since update llog header might also call this function,
* let's reset the bitmap to 0 here
*/
len = llh->llh_hdr.lrh_len - llh->llh_bitmap_offset;
memset(LLOG_HDR_BITMAP(llh), 0, len - sizeof(llh->llh_tail));
ext2_set_bit(0, LLOG_HDR_BITMAP(llh));
LLOG_HDR_TAIL(llh)->lrt_len = llh->llh_hdr.lrh_len;
LLOG_HDR_TAIL(llh)->lrt_index = llh->llh_hdr.lrh_index;
rc = 0;
}
return rc;
}
static int iostat_info_seq_show(struct seq_file *seq, void *offset)
{
struct super_block *sb = seq->private;
struct f2fs_sb_info *sbi = F2FS_SB(sb);
time64_t now = ktime_get_real_seconds();
if (!sbi->iostat_enable)
return 0;
seq_printf(seq, "time: %-16llu\n", now);
/* print app IOs */
seq_printf(seq, "app buffered: %-16llu\n",
sbi->write_iostat[APP_BUFFERED_IO]);
seq_printf(seq, "app direct: %-16llu\n",
sbi->write_iostat[APP_DIRECT_IO]);
seq_printf(seq, "app mapped: %-16llu\n",
sbi->write_iostat[APP_MAPPED_IO]);
/* print fs IOs */
seq_printf(seq, "fs data: %-16llu\n",
sbi->write_iostat[FS_DATA_IO]);
seq_printf(seq, "fs node: %-16llu\n",
sbi->write_iostat[FS_NODE_IO]);
seq_printf(seq, "fs meta: %-16llu\n",
sbi->write_iostat[FS_META_IO]);
seq_printf(seq, "fs gc data: %-16llu\n",
sbi->write_iostat[FS_GC_DATA_IO]);
seq_printf(seq, "fs gc node: %-16llu\n",
sbi->write_iostat[FS_GC_NODE_IO]);
seq_printf(seq, "fs cp data: %-16llu\n",
sbi->write_iostat[FS_CP_DATA_IO]);
seq_printf(seq, "fs cp node: %-16llu\n",
sbi->write_iostat[FS_CP_NODE_IO]);
seq_printf(seq, "fs cp meta: %-16llu\n",
sbi->write_iostat[FS_CP_META_IO]);
seq_printf(seq, "fs discard: %-16llu\n",
sbi->write_iostat[FS_DISCARD]);
return 0;
}
/*
* Begin iteration through a server list, starting with the last used server if
* possible, or the last recorded good server if not.
*/
static bool afs_start_vl_iteration(struct afs_vl_cursor *vc)
{
struct afs_cell *cell = vc->cell;
unsigned int dns_lookup_count;
if (cell->dns_source == DNS_RECORD_UNAVAILABLE ||
cell->dns_expiry <= ktime_get_real_seconds()) {
dns_lookup_count = smp_load_acquire(&cell->dns_lookup_count);
set_bit(AFS_CELL_FL_DO_LOOKUP, &cell->flags);
queue_work(afs_wq, &cell->manager);
if (cell->dns_source == DNS_RECORD_UNAVAILABLE) {
if (wait_var_event_interruptible(
&cell->dns_lookup_count,
smp_load_acquire(&cell->dns_lookup_count)
!= dns_lookup_count) < 0) {
vc->error = -ERESTARTSYS;
return false;
}
}
/* Status load is ordered after lookup counter load */
if (cell->dns_source == DNS_RECORD_UNAVAILABLE) {
vc->error = -EDESTADDRREQ;
return false;
}
}
read_lock(&cell->vl_servers_lock);
vc->server_list = afs_get_vlserverlist(
rcu_dereference_protected(cell->vl_servers,
lockdep_is_held(&cell->vl_servers_lock)));
read_unlock(&cell->vl_servers_lock);
if (!vc->server_list->nr_servers)
return false;
vc->untried = (1UL << vc->server_list->nr_servers) - 1;
vc->index = -1;
return true;
}
/*
* Drop a reference on a cell record.
*/
void afs_put_cell(struct afs_net *net, struct afs_cell *cell)
{
time64_t now, expire_delay;
if (!cell)
return;
_enter("%s", cell->name);
now = ktime_get_real_seconds();
cell->last_inactive = now;
expire_delay = 0;
if (!test_bit(AFS_CELL_FL_DNS_FAIL, &cell->flags) &&
!test_bit(AFS_CELL_FL_NOT_FOUND, &cell->flags))
expire_delay = afs_cell_gc_delay;
if (atomic_dec_return(&cell->usage) > 1)
return;
/* 'cell' may now be garbage collected. */
afs_set_cell_timer(net, expire_delay);
}
/*
* Display the list of cells known to the namespace.
*/
static int afs_proc_cells_show(struct seq_file *m, void *v)
{
struct afs_vlserver_list *vllist;
struct afs_cell *cell;
if (v == SEQ_START_TOKEN) {
/* display header on line 1 */
seq_puts(m, "USE TTL SV NAME\n");
return 0;
}
cell = list_entry(v, struct afs_cell, proc_link);
vllist = rcu_dereference(cell->vl_servers);
/* display one cell per line on subsequent lines */
seq_printf(m, "%3u %6lld %2u %s\n",
atomic_read(&cell->usage),
cell->dns_expiry - ktime_get_real_seconds(),
vllist ? vllist->nr_servers : 0,
cell->name);
return 0;
}
static int nilfs_setup_super(struct super_block *sb, int is_mount)
{
struct the_nilfs *nilfs = sb->s_fs_info;
struct nilfs_super_block **sbp;
int max_mnt_count;
int mnt_count;
/* nilfs->ns_sem must be locked by the caller. */
sbp = nilfs_prepare_super(sb, 0);
if (!sbp)
return -EIO;
if (!is_mount)
goto skip_mount_setup;
max_mnt_count = le16_to_cpu(sbp[0]->s_max_mnt_count);
mnt_count = le16_to_cpu(sbp[0]->s_mnt_count);
if (nilfs->ns_mount_state & NILFS_ERROR_FS) {
nilfs_msg(sb, KERN_WARNING, "mounting fs with errors");
#if 0
} else if (max_mnt_count >= 0 && mnt_count >= max_mnt_count) {
nilfs_msg(sb, KERN_WARNING, "maximal mount count reached");
#endif
}
if (!max_mnt_count)
sbp[0]->s_max_mnt_count = cpu_to_le16(NILFS_DFL_MAX_MNT_COUNT);
sbp[0]->s_mnt_count = cpu_to_le16(mnt_count + 1);
sbp[0]->s_mtime = cpu_to_le64(ktime_get_real_seconds());
skip_mount_setup:
sbp[0]->s_state =
cpu_to_le16(le16_to_cpu(sbp[0]->s_state) & ~NILFS_VALID_FS);
/* synchronize sbp[1] with sbp[0] */
if (sbp[1])
memcpy(sbp[1], sbp[0], nilfs->ns_sbsize);
return nilfs_commit_super(sb, NILFS_SB_COMMIT_ALL);
}
/**
* Dump Lustre log to ::debug_file_path by calling tracefile_dump_all_pages()
*/
void libcfs_debug_dumplog_internal(void *arg)
{
static time64_t last_dump_time;
time64_t current_time;
void *journal_info;
journal_info = current->journal_info;
current->journal_info = NULL;
current_time = ktime_get_real_seconds();
if (strncmp(libcfs_debug_file_path_arr, "NONE", 4) &&
current_time > last_dump_time) {
last_dump_time = current_time;
snprintf(debug_file_name, sizeof(debug_file_name) - 1,
"%s.%lld.%ld", libcfs_debug_file_path_arr,
(s64)current_time, (long)arg);
pr_alert("LustreError: dumping log to %s\n", debug_file_name);
cfs_tracefile_dump_all_pages(debug_file_name);
libcfs_run_debug_log_upcall(debug_file_name);
}
current->journal_info = journal_info;
}
static int sptlrpc_info_lprocfs_seq_show(struct seq_file *seq, void *v)
{
struct obd_device *dev = seq->private;
struct client_obd *cli = &dev->u.cli;
struct ptlrpc_sec *sec = NULL;
char str[32];
LASSERT(strcmp(dev->obd_type->typ_name, LUSTRE_OSC_NAME) == 0 ||
strcmp(dev->obd_type->typ_name, LUSTRE_MDC_NAME) == 0 ||
strcmp(dev->obd_type->typ_name, LUSTRE_MGC_NAME) == 0);
if (cli->cl_import)
sec = sptlrpc_import_sec_ref(cli->cl_import);
if (sec == NULL)
goto out;
sec_flags2str(sec->ps_flvr.sf_flags, str, sizeof(str));
seq_printf(seq, "rpc flavor: %s\n",
sptlrpc_flavor2name_base(sec->ps_flvr.sf_rpc));
seq_printf(seq, "bulk flavor: %s\n",
sptlrpc_flavor2name_bulk(&sec->ps_flvr, str, sizeof(str)));
seq_printf(seq, "flags: %s\n",
sec_flags2str(sec->ps_flvr.sf_flags, str, sizeof(str)));
seq_printf(seq, "id: %d\n", sec->ps_id);
seq_printf(seq, "refcount: %d\n",
atomic_read(&sec->ps_refcount));
seq_printf(seq, "nctx: %d\n", atomic_read(&sec->ps_nctx));
seq_printf(seq, "gc internal %ld\n", sec->ps_gc_interval);
seq_printf(seq, "gc next %lld\n",
sec->ps_gc_interval ?
(s64)(sec->ps_gc_next - ktime_get_real_seconds()) : 0ll);
sptlrpc_sec_put(sec);
out:
return 0;
}
/**
* Send request \a request.
* if \a noreply is set, don't expect any reply back and don't set up
* reply buffers.
* Returns 0 on success or error code.
*/
int ptl_send_rpc(struct ptlrpc_request *request, int noreply)
{
int rc;
int rc2;
int mpflag = 0;
struct ptlrpc_connection *connection;
lnet_handle_me_t reply_me_h;
lnet_md_t reply_md;
struct obd_device *obd = request->rq_import->imp_obd;
if (OBD_FAIL_CHECK(OBD_FAIL_PTLRPC_DROP_RPC))
return 0;
LASSERT(request->rq_type == PTL_RPC_MSG_REQUEST);
LASSERT(request->rq_wait_ctx == 0);
/* If this is a re-transmit, we're required to have disengaged
* cleanly from the previous attempt */
LASSERT(!request->rq_receiving_reply);
LASSERT(!((lustre_msg_get_flags(request->rq_reqmsg) & MSG_REPLAY) &&
(request->rq_import->imp_state == LUSTRE_IMP_FULL)));
if (unlikely(obd != NULL && obd->obd_fail)) {
CDEBUG(D_HA, "muting rpc for failed imp obd %s\n",
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;
}
connection = request->rq_import->imp_connection;
lustre_msg_set_handle(request->rq_reqmsg,
&request->rq_import->imp_remote_handle);
lustre_msg_set_type(request->rq_reqmsg, PTL_RPC_MSG_REQUEST);
lustre_msg_set_conn_cnt(request->rq_reqmsg,
request->rq_import->imp_conn_cnt);
lustre_msghdr_set_flags(request->rq_reqmsg,
request->rq_import->imp_msghdr_flags);
if (request->rq_resend)
lustre_msg_add_flags(request->rq_reqmsg, MSG_RESENT);
if (request->rq_memalloc)
mpflag = cfs_memory_pressure_get_and_set();
rc = sptlrpc_cli_wrap_request(request);
if (rc)
goto out;
/* bulk register should be done after wrap_request() */
if (request->rq_bulk != NULL) {
rc = ptlrpc_register_bulk(request);
if (rc != 0)
goto out;
}
if (!noreply) {
LASSERT(request->rq_replen != 0);
if (request->rq_repbuf == NULL) {
LASSERT(request->rq_repdata == NULL);
LASSERT(request->rq_repmsg == NULL);
rc = sptlrpc_cli_alloc_repbuf(request,
request->rq_replen);
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;
}
} else {
request->rq_repdata = NULL;
request->rq_repmsg = NULL;
}
rc = LNetMEAttach(request->rq_reply_portal,/*XXX FIXME bug 249*/
connection->c_peer, request->rq_xid, 0,
LNET_UNLINK, LNET_INS_AFTER, &reply_me_h);
if (rc != 0) {
CERROR("LNetMEAttach failed: %d\n", rc);
LASSERT(rc == -ENOMEM);
rc = -ENOMEM;
goto cleanup_bulk;
}
}
spin_lock(&request->rq_lock);
/* If the MD attach succeeds, there _will_ be a reply_in callback */
request->rq_receiving_reply = !noreply;
request->rq_req_unlink = 1;
/* We are responsible for unlinking the reply buffer */
request->rq_reply_unlink = !noreply;
/* Clear any flags that may be present from previous sends. */
request->rq_replied = 0;
request->rq_err = 0;
request->rq_timedout = 0;
request->rq_net_err = 0;
request->rq_resend = 0;
request->rq_restart = 0;
request->rq_reply_truncate = 0;
spin_unlock(&request->rq_lock);
if (!noreply) {
reply_md.start = request->rq_repbuf;
reply_md.length = request->rq_repbuf_len;
/* Allow multiple early replies */
reply_md.threshold = LNET_MD_THRESH_INF;
/* Manage remote for early replies */
reply_md.options = PTLRPC_MD_OPTIONS | LNET_MD_OP_PUT |
LNET_MD_MANAGE_REMOTE |
LNET_MD_TRUNCATE; /* allow to make EOVERFLOW error */
reply_md.user_ptr = &request->rq_reply_cbid;
reply_md.eq_handle = ptlrpc_eq_h;
/* We must see the unlink callback to unset rq_reply_unlink,
so we can't auto-unlink */
rc = LNetMDAttach(reply_me_h, reply_md, LNET_RETAIN,
&request->rq_reply_md_h);
if (rc != 0) {
CERROR("LNetMDAttach failed: %d\n", rc);
LASSERT(rc == -ENOMEM);
spin_lock(&request->rq_lock);
/* ...but the MD attach didn't succeed... */
request->rq_receiving_reply = 0;
spin_unlock(&request->rq_lock);
rc = -ENOMEM;
goto cleanup_me;
}
CDEBUG(D_NET, "Setup reply buffer: %u bytes, xid %llu, portal %u\n",
request->rq_repbuf_len, request->rq_xid,
request->rq_reply_portal);
}
/* add references on request for request_out_callback */
ptlrpc_request_addref(request);
if (obd != NULL && obd->obd_svc_stats != NULL)
lprocfs_counter_add(obd->obd_svc_stats, PTLRPC_REQACTIVE_CNTR,
atomic_read(&request->rq_import->imp_inflight));
OBD_FAIL_TIMEOUT(OBD_FAIL_PTLRPC_DELAY_SEND, request->rq_timeout + 5);
ktime_get_real_ts64(&request->rq_arrival_time);
request->rq_sent = ktime_get_real_seconds();
/* We give the server rq_timeout secs to process the req, and
add the network latency for our local timeout. */
request->rq_deadline = request->rq_sent + request->rq_timeout +
ptlrpc_at_get_net_latency(request);
ptlrpc_pinger_sending_on_import(request->rq_import);
DEBUG_REQ(D_INFO, request, "send flg=%x",
lustre_msg_get_flags(request->rq_reqmsg));
rc = ptl_send_buf(&request->rq_req_md_h,
request->rq_reqbuf, request->rq_reqdata_len,
LNET_NOACK_REQ, &request->rq_req_cbid,
connection,
request->rq_request_portal,
request->rq_xid, 0);
if (rc == 0)
goto out;
ptlrpc_req_finished(request);
if (noreply)
goto out;
cleanup_me:
/* MEUnlink is safe; the PUT didn't even get off the ground, and
* nobody apart from the PUT's target has the right nid+XID to
* access the reply buffer. */
rc2 = LNetMEUnlink(reply_me_h);
LASSERT(rc2 == 0);
/* UNLINKED callback called synchronously */
LASSERT(!request->rq_receiving_reply);
cleanup_bulk:
/* We do sync unlink here as there was no real transfer here so
* the chance to have long unlink to sluggish net is smaller here. */
ptlrpc_unregister_bulk(request, 0);
out:
if (request->rq_memalloc)
cfs_memory_pressure_restore(mpflag);
return rc;
}
int
lstcon_ioctl_entry(unsigned int cmd, struct libcfs_ioctl_hdr *hdr)
{
char *buf;
struct libcfs_ioctl_data *data;
int opc;
int rc;
if (cmd != IOC_LIBCFS_LNETST)
return -EINVAL;
data = container_of(hdr, struct libcfs_ioctl_data, ioc_hdr);
opc = data->ioc_u32[0];
if (data->ioc_plen1 > PAGE_SIZE)
return -EINVAL;
LIBCFS_ALLOC(buf, data->ioc_plen1);
if (!buf)
return -ENOMEM;
/* copy in parameter */
if (copy_from_user(buf, data->ioc_pbuf1, data->ioc_plen1)) {
LIBCFS_FREE(buf, data->ioc_plen1);
return -EFAULT;
}
mutex_lock(&console_session.ses_mutex);
console_session.ses_laststamp = ktime_get_real_seconds();
if (console_session.ses_shutdown) {
rc = -ESHUTDOWN;
goto out;
}
if (console_session.ses_expired)
lstcon_session_end();
if (opc != LSTIO_SESSION_NEW &&
console_session.ses_state == LST_SESSION_NONE) {
CDEBUG(D_NET, "LST no active session\n");
rc = -ESRCH;
goto out;
}
memset(&console_session.ses_trans_stat, 0, sizeof(lstcon_trans_stat_t));
switch (opc) {
case LSTIO_SESSION_NEW:
rc = lst_session_new_ioctl((lstio_session_new_args_t *)buf);
break;
case LSTIO_SESSION_END:
rc = lst_session_end_ioctl((lstio_session_end_args_t *)buf);
break;
case LSTIO_SESSION_INFO:
rc = lst_session_info_ioctl((lstio_session_info_args_t *)buf);
break;
case LSTIO_DEBUG:
rc = lst_debug_ioctl((lstio_debug_args_t *)buf);
break;
case LSTIO_GROUP_ADD:
rc = lst_group_add_ioctl((lstio_group_add_args_t *)buf);
break;
case LSTIO_GROUP_DEL:
rc = lst_group_del_ioctl((lstio_group_del_args_t *)buf);
break;
case LSTIO_GROUP_UPDATE:
rc = lst_group_update_ioctl((lstio_group_update_args_t *)buf);
break;
case LSTIO_NODES_ADD:
rc = lst_nodes_add_ioctl((lstio_group_nodes_args_t *)buf);
break;
case LSTIO_GROUP_LIST:
rc = lst_group_list_ioctl((lstio_group_list_args_t *)buf);
break;
case LSTIO_GROUP_INFO:
rc = lst_group_info_ioctl((lstio_group_info_args_t *)buf);
break;
case LSTIO_BATCH_ADD:
rc = lst_batch_add_ioctl((lstio_batch_add_args_t *)buf);
break;
case LSTIO_BATCH_START:
rc = lst_batch_run_ioctl((lstio_batch_run_args_t *)buf);
break;
case LSTIO_BATCH_STOP:
rc = lst_batch_stop_ioctl((lstio_batch_stop_args_t *)buf);
break;
case LSTIO_BATCH_QUERY:
rc = lst_batch_query_ioctl((lstio_batch_query_args_t *)buf);
break;
case LSTIO_BATCH_LIST:
rc = lst_batch_list_ioctl((lstio_batch_list_args_t *)buf);
break;
case LSTIO_BATCH_INFO:
rc = lst_batch_info_ioctl((lstio_batch_info_args_t *)buf);
break;
case LSTIO_TEST_ADD:
rc = lst_test_add_ioctl((lstio_test_args_t *)buf);
break;
case LSTIO_STAT_QUERY:
rc = lst_stat_query_ioctl((lstio_stat_args_t *)buf);
break;
default:
rc = -EINVAL;
}
if (copy_to_user(data->ioc_pbuf2, &console_session.ses_trans_stat,
sizeof(lstcon_trans_stat_t)))
rc = -EFAULT;
out:
mutex_unlock(&console_session.ses_mutex);
LIBCFS_FREE(buf, data->ioc_plen1);
return rc;
}
/**
* Send request reply from request \a req reply buffer.
* \a flags defines reply types
* Returns 0 on success or error code
*/
int ptlrpc_send_reply(struct ptlrpc_request *req, int flags)
{
struct ptlrpc_reply_state *rs = req->rq_reply_state;
struct ptlrpc_connection *conn;
int rc;
/* We must already have a reply buffer (only ptlrpc_error() may be
* called without one). The reply generated by sptlrpc layer (e.g.
* error notify, etc.) might have NULL rq->reqmsg; Otherwise we must
* have a request buffer which is either the actual (swabbed) incoming
* request, or a saved copy if this is a req saved in
* target_queue_final_reply().
*/
LASSERT(req->rq_no_reply == 0);
LASSERT(req->rq_reqbuf != NULL);
LASSERT(rs != NULL);
LASSERT((flags & PTLRPC_REPLY_MAYBE_DIFFICULT) || !rs->rs_difficult);
LASSERT(req->rq_repmsg != NULL);
LASSERT(req->rq_repmsg == rs->rs_msg);
LASSERT(rs->rs_cb_id.cbid_fn == reply_out_callback);
LASSERT(rs->rs_cb_id.cbid_arg == rs);
/* There may be no rq_export during failover */
if (unlikely(req->rq_export && req->rq_export->exp_obd &&
req->rq_export->exp_obd->obd_fail)) {
/* Failed obd's only send ENODEV */
req->rq_type = PTL_RPC_MSG_ERR;
req->rq_status = -ENODEV;
CDEBUG(D_HA, "sending ENODEV from failed obd %d\n",
req->rq_export->exp_obd->obd_minor);
}
/* In order to keep interoperability with the client (< 2.3) which
* doesn't have pb_jobid in ptlrpc_body, We have to shrink the
* ptlrpc_body in reply buffer to ptlrpc_body_v2, otherwise, the
* reply buffer on client will be overflow.
*
* XXX Remove this whenever we drop the interoperability with
* such client.
*/
req->rq_replen = lustre_shrink_msg(req->rq_repmsg, 0,
sizeof(struct ptlrpc_body_v2), 1);
if (req->rq_type != PTL_RPC_MSG_ERR)
req->rq_type = PTL_RPC_MSG_REPLY;
lustre_msg_set_type(req->rq_repmsg, req->rq_type);
lustre_msg_set_status(req->rq_repmsg,
ptlrpc_status_hton(req->rq_status));
lustre_msg_set_opc(req->rq_repmsg,
req->rq_reqmsg ? lustre_msg_get_opc(req->rq_reqmsg) : 0);
target_pack_pool_reply(req);
ptlrpc_at_set_reply(req, flags);
if (req->rq_export == NULL || req->rq_export->exp_connection == NULL)
conn = ptlrpc_connection_get(req->rq_peer, req->rq_self, NULL);
else
conn = ptlrpc_connection_addref(req->rq_export->exp_connection);
if (unlikely(conn == NULL)) {
CERROR("not replying on NULL connection\n"); /* bug 9635 */
return -ENOTCONN;
}
ptlrpc_rs_addref(rs); /* +1 ref for the network */
rc = sptlrpc_svc_wrap_reply(req);
if (unlikely(rc))
goto out;
req->rq_sent = ktime_get_real_seconds();
rc = ptl_send_buf(&rs->rs_md_h, rs->rs_repbuf, rs->rs_repdata_len,
(rs->rs_difficult && !rs->rs_no_ack) ?
LNET_ACK_REQ : LNET_NOACK_REQ,
&rs->rs_cb_id, conn,
ptlrpc_req2svc(req)->srv_rep_portal,
req->rq_xid, req->rq_reply_off);
out:
if (unlikely(rc != 0))
ptlrpc_req_drop_rs(req);
ptlrpc_connection_put(conn);
return rc;
}
static int test_rtc_read_time(struct device *dev,
struct rtc_time *tm)
{
rtc_time64_to_tm(ktime_get_real_seconds(), tm);
return 0;
}
