int do_clock_gettime(clockid_t which_clock, struct timespec64 *tp)
{
switch (which_clock) {
case CLOCK_REALTIME:
ktime_get_real_ts64(tp);
break;
case CLOCK_MONOTONIC:
ktime_get_ts64(tp);
break;
case CLOCK_BOOTTIME:
ktime_get_boottime_ts64(tp);
break;
default:
return -EINVAL;
}
return 0;
}
SYSCALL_DEFINE2(clock_gettime, const clockid_t, which_clock,
struct timespec __user *,tp)
{
struct timespec64 kernel_tp64;
struct timespec kernel_tp;
switch (which_clock) {
case CLOCK_REALTIME: ktime_get_real_ts64(&kernel_tp64); break;
case CLOCK_MONOTONIC: ktime_get_ts64(&kernel_tp64); break;
case CLOCK_BOOTTIME: get_monotonic_boottime64(&kernel_tp64); break;
default: return -EINVAL;
}
kernel_tp = timespec64_to_timespec(kernel_tp64);
if (copy_to_user(tp, &kernel_tp, sizeof (kernel_tp)))
return -EFAULT;
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;
}
long ptp_ioctl(struct posix_clock *pc, unsigned int cmd, unsigned long arg)
{
struct ptp_clock *ptp = container_of(pc, struct ptp_clock, clock);
struct ptp_sys_offset_extended *extoff = NULL;
struct ptp_sys_offset_precise precise_offset;
struct system_device_crosststamp xtstamp;
struct ptp_clock_info *ops = ptp->info;
struct ptp_sys_offset *sysoff = NULL;
struct ptp_system_timestamp sts;
struct ptp_clock_request req;
struct ptp_clock_caps caps;
struct ptp_clock_time *pct;
unsigned int i, pin_index;
struct ptp_pin_desc pd;
struct timespec64 ts;
int enable, err = 0;
switch (cmd) {
case PTP_CLOCK_GETCAPS:
memset(&caps, 0, sizeof(caps));
caps.max_adj = ptp->info->max_adj;
caps.n_alarm = ptp->info->n_alarm;
caps.n_ext_ts = ptp->info->n_ext_ts;
caps.n_per_out = ptp->info->n_per_out;
caps.pps = ptp->info->pps;
caps.n_pins = ptp->info->n_pins;
caps.cross_timestamping = ptp->info->getcrosststamp != NULL;
if (copy_to_user((void __user *)arg, &caps, sizeof(caps)))
err = -EFAULT;
break;
case PTP_EXTTS_REQUEST:
if (copy_from_user(&req.extts, (void __user *)arg,
sizeof(req.extts))) {
err = -EFAULT;
break;
}
if (req.extts.index >= ops->n_ext_ts) {
err = -EINVAL;
break;
}
req.type = PTP_CLK_REQ_EXTTS;
enable = req.extts.flags & PTP_ENABLE_FEATURE ? 1 : 0;
err = ops->enable(ops, &req, enable);
break;
case PTP_PEROUT_REQUEST:
if (copy_from_user(&req.perout, (void __user *)arg,
sizeof(req.perout))) {
err = -EFAULT;
break;
}
if (req.perout.index >= ops->n_per_out) {
err = -EINVAL;
break;
}
req.type = PTP_CLK_REQ_PEROUT;
enable = req.perout.period.sec || req.perout.period.nsec;
err = ops->enable(ops, &req, enable);
break;
case PTP_ENABLE_PPS:
if (!capable(CAP_SYS_TIME))
return -EPERM;
req.type = PTP_CLK_REQ_PPS;
enable = arg ? 1 : 0;
err = ops->enable(ops, &req, enable);
break;
case PTP_SYS_OFFSET_PRECISE:
if (!ptp->info->getcrosststamp) {
err = -EOPNOTSUPP;
break;
}
err = ptp->info->getcrosststamp(ptp->info, &xtstamp);
if (err)
break;
memset(&precise_offset, 0, sizeof(precise_offset));
ts = ktime_to_timespec64(xtstamp.device);
precise_offset.device.sec = ts.tv_sec;
precise_offset.device.nsec = ts.tv_nsec;
ts = ktime_to_timespec64(xtstamp.sys_realtime);
precise_offset.sys_realtime.sec = ts.tv_sec;
precise_offset.sys_realtime.nsec = ts.tv_nsec;
ts = ktime_to_timespec64(xtstamp.sys_monoraw);
precise_offset.sys_monoraw.sec = ts.tv_sec;
precise_offset.sys_monoraw.nsec = ts.tv_nsec;
if (copy_to_user((void __user *)arg, &precise_offset,
sizeof(precise_offset)))
err = -EFAULT;
break;
case PTP_SYS_OFFSET_EXTENDED:
if (!ptp->info->gettimex64) {
err = -EOPNOTSUPP;
break;
}
extoff = memdup_user((void __user *)arg, sizeof(*extoff));
if (IS_ERR(extoff)) {
err = PTR_ERR(extoff);
extoff = NULL;
break;
}
if (extoff->n_samples > PTP_MAX_SAMPLES
|| extoff->rsv[0] || extoff->rsv[1] || extoff->rsv[2]) {
err = -EINVAL;
break;
}
for (i = 0; i < extoff->n_samples; i++) {
err = ptp->info->gettimex64(ptp->info, &ts, &sts);
if (err)
goto out;
extoff->ts[i][0].sec = sts.pre_ts.tv_sec;
extoff->ts[i][0].nsec = sts.pre_ts.tv_nsec;
extoff->ts[i][1].sec = ts.tv_sec;
extoff->ts[i][1].nsec = ts.tv_nsec;
extoff->ts[i][2].sec = sts.post_ts.tv_sec;
extoff->ts[i][2].nsec = sts.post_ts.tv_nsec;
}
if (copy_to_user((void __user *)arg, extoff, sizeof(*extoff)))
err = -EFAULT;
break;
case PTP_SYS_OFFSET:
sysoff = memdup_user((void __user *)arg, sizeof(*sysoff));
if (IS_ERR(sysoff)) {
err = PTR_ERR(sysoff);
sysoff = NULL;
break;
}
if (sysoff->n_samples > PTP_MAX_SAMPLES) {
err = -EINVAL;
break;
}
pct = &sysoff->ts[0];
for (i = 0; i < sysoff->n_samples; i++) {
ktime_get_real_ts64(&ts);
pct->sec = ts.tv_sec;
pct->nsec = ts.tv_nsec;
pct++;
if (ops->gettimex64)
err = ops->gettimex64(ops, &ts, NULL);
else
err = ops->gettime64(ops, &ts);
if (err)
goto out;
pct->sec = ts.tv_sec;
pct->nsec = ts.tv_nsec;
pct++;
}
ktime_get_real_ts64(&ts);
pct->sec = ts.tv_sec;
pct->nsec = ts.tv_nsec;
if (copy_to_user((void __user *)arg, sysoff, sizeof(*sysoff)))
err = -EFAULT;
break;
case PTP_PIN_GETFUNC:
if (copy_from_user(&pd, (void __user *)arg, sizeof(pd))) {
err = -EFAULT;
break;
}
pin_index = pd.index;
if (pin_index >= ops->n_pins) {
err = -EINVAL;
break;
}
pin_index = array_index_nospec(pin_index, ops->n_pins);
if (mutex_lock_interruptible(&ptp->pincfg_mux))
return -ERESTARTSYS;
pd = ops->pin_config[pin_index];
mutex_unlock(&ptp->pincfg_mux);
if (!err && copy_to_user((void __user *)arg, &pd, sizeof(pd)))
err = -EFAULT;
break;
case PTP_PIN_SETFUNC:
if (copy_from_user(&pd, (void __user *)arg, sizeof(pd))) {
err = -EFAULT;
break;
}
pin_index = pd.index;
if (pin_index >= ops->n_pins) {
err = -EINVAL;
break;
}
pin_index = array_index_nospec(pin_index, ops->n_pins);
if (mutex_lock_interruptible(&ptp->pincfg_mux))
return -ERESTARTSYS;
err = ptp_set_pinfunc(ptp, pin_index, pd.func, pd.chan);
mutex_unlock(&ptp->pincfg_mux);
break;
default:
err = -ENOTTY;
break;
}
out:
kfree(extoff);
kfree(sysoff);
return err;
}
/**
* 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_import *imp = request->rq_import;
struct obd_device *obd = imp->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) &&
(imp->imp_state == LUSTRE_IMP_FULL)));
if (unlikely(obd && 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 = imp->imp_connection;
lustre_msg_set_handle(request->rq_reqmsg,
&imp->imp_remote_handle);
lustre_msg_set_type(request->rq_reqmsg, PTL_RPC_MSG_REQUEST);
lustre_msg_set_conn_cnt(request->rq_reqmsg, imp->imp_conn_cnt);
lustre_msghdr_set_flags(request->rq_reqmsg, imp->imp_msghdr_flags);
/*
* If it's the first time to resend the request for EINPROGRESS,
* we need to allocate a new XID (see after_reply()), it's different
* from the resend for reply timeout.
*/
if (request->rq_nr_resend && list_empty(&request->rq_unreplied_list)) {
__u64 min_xid = 0;
/*
* resend for EINPROGRESS, allocate new xid to avoid reply
* reconstruction
*/
spin_lock(&imp->imp_lock);
ptlrpc_assign_next_xid_nolock(request);
min_xid = ptlrpc_known_replied_xid(imp);
spin_unlock(&imp->imp_lock);
lustre_msg_set_last_xid(request->rq_reqmsg, min_xid);
DEBUG_REQ(D_RPCTRACE, request, "Allocating new xid for resend on EINPROGRESS");
}
if (request->rq_bulk) {
ptlrpc_set_bulk_mbits(request);
lustre_msg_set_mbits(request->rq_reqmsg, request->rq_mbits);
}
if (list_empty(&request->rq_unreplied_list) ||
request->rq_xid <= imp->imp_known_replied_xid) {
DEBUG_REQ(D_ERROR, request,
"xid: %llu, replied: %llu, list_empty:%d\n",
request->rq_xid, imp->imp_known_replied_xid,
list_empty(&request->rq_unreplied_list));
LBUG();
}
/**
* For enabled AT all request should have AT_SUPPORT in the
* FULL import state when OBD_CONNECT_AT is set
*/
LASSERT(AT_OFF || imp->imp_state != LUSTRE_IMP_FULL ||
(imp->imp_msghdr_flags & MSGHDR_AT_SUPPORT) ||
!(imp->imp_connect_data.ocd_connect_flags &
OBD_CONNECT_AT));
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) {
/*
* set rq_sent so that this request is treated
* as a delayed send in the upper layers
*/
if (rc == -ENOMEM)
request->rq_sent = ktime_get_seconds();
goto out;
}
/* bulk register should be done after wrap_request() */
if (request->rq_bulk) {
rc = ptlrpc_register_bulk(request);
if (rc != 0)
goto out;
}
if (!noreply) {
LASSERT(request->rq_replen != 0);
if (!request->rq_repbuf) {
LASSERT(!request->rq_repdata);
LASSERT(!request->rq_repmsg);
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);
/* We are responsible for unlinking the reply buffer */
request->rq_reply_unlinked = noreply;
request->rq_receiving_reply = !noreply;
/* Clear any flags that may be present from previous sends. */
request->rq_req_unlinked = 0;
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_truncated = 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 set rq_reply_unlinked,
* 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 && obd->obd_svc_stats)
lprocfs_counter_add(obd->obd_svc_stats, PTLRPC_REQACTIVE_CNTR,
atomic_read(&imp->imp_inflight));
OBD_FAIL_TIMEOUT(OBD_FAIL_PTLRPC_DELAY_SEND, request->rq_timeout + 5);
ktime_get_real_ts64(&request->rq_sent_tv);
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(imp);
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 (likely(rc == 0))
goto out;
request->rq_req_unlinked = 1;
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;
}
