The sysoff_measure() function gained an additional parameter some time
ago, but the legacy method for older kernels without PTP_SYS_OFFSET was
overlooked. This patch brings the legacy method up to date again.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
In port_synchronize(), when the clock returned SERVO_UNLOCKED and
port_dispatch() triggered a switch of the PHC, the variable "s" would
point to a servo which was already freed and the following call of
servo_offset_threshold() would read invalid memory.
Don't save the servo before dispatching the port.
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
Fixes: 3f5f5653d7 ("port: Add interval update mechanism.")
TLV_PORT_PROPERTIES_NP messages serve for querying of port properties, such
as timestamp type and, prominently, netdevice name associated with the
port. pmc however does not support this query, which makes it difficult to
access this information e.g. from scripts. Add this support to pmc.
Signed-off-by: Mykola Zhuravel <mykola@mellanox.com>
Signed-off-by: Petr Machata <petrm@mellanox.com>
TLV_PORT_PROPERTIES_NP carries, among other attributes, a timestamp type
used for that port. In order to make it possible to format the value for
user consumption, introduce a new function ts_str().
Signed-off-by: Petr Machata <petrm@mellanox.com>
In commit 2b5bec8d27 ("pmc: Add a new TLV to obtain per-port
statistics"), pmc gained a new TLV to obtain port stats from ptp4l. Mention
the TLV in the pmc man page.
Signed-off-by: Petr Machata <petrm@mellanox.com>
mc6_addr holds the parsed multicast address to which messages should be sent.
But since each port can have a different scope, and the primary MC address
depends on the scope, it really can't be stored in a global variable. Move both
to struct udp6.
Additionally, document the fact that the primary multicast address is changed at
runtime.
Reported-by: Alex Veber <alexve@mellanox.com>
Signed-off-by: Petr Machata <petrm@mellanox.com>
Add struct PortStats to keep per-port number of messages sent and received,
split by message type. Bump TX counters after messages are sent
successfully, and RX counters after a message is received. To keep things
simple, reserve one counter for each theoretically possible message type,
including the reserved ones.
Signed-off-by: Petr Machata <petrm@mellanox.com>
The two programs, phc2sys and phc_ctl, use nearly identical subroutines to
open an instance of a dynamic posix clock. In anticipation of yet a third
program with similar needs, this patch refactors the common code into the
utilities.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
The order of the options list has gone a bit astray. This patch
restores the ordering.
Fixes: 450b1ed844 ("port: Add inhibit_delay_req.")
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
ptp4l -f default.cfg complains:
unknown option inhibit_pdelay_req at line 41 in global section
failed to parse configuration file /etc/linuxptp/default.cfg
Fixes: 450b1ed844 ("port: Add inhibit_delay_req.")
Signed-off-by: Vladimir Oltean <olteanv@gmail.com>
Users may need to use different socket priorities for ptp4l traffic for
the purpose of traffic shaping. An example is to route ptp4l traffic
through a specific Linux egress queue using the mqprio qdisc.
- Update raw.c open_socket() to accept a socket_priority parameter
- Add the socket_priority option to config.c and the default.cfg config
file. The option defaults to 0.
CC: "Ong, Boon Leong" <boon.leong.ong@intel.com>
CC: "Wong, Vincent Por Yin" <vincent.por.yin.wong@intel.com>
Signed-off-by: Khor, Isaac Shi Yan <isaac.shi.yan.khor@intel.com>
struct hwtstamp_config.rx_filter passed to SIOCSHWTSTAMP can be updated by
the kernel with the value of HWTSTAMP_FILTER_SOME. That indicates that all
requested packets will be timestamped, and some others as well.
Update hwts_init() to recognize this as a valid response in HWTS_FILTER_NORMAL
mode, instead of rejecting it as mismatch.
Cc: "Keller, Jacob E" <jacob.e.keller@intel.com>
Cc: "Geva, Erez" <erez.geva.ext@siemens.com>
Signed-off-by: Petr Machata <petrm@mellanox.com>
Set the announce, sync and delay request intervals in the example
configs per the G.8275.1 and G.8275.2 recommendations. Enable the L2
transport in G.8275.1.
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
This patch add team interface activebackup mode support. As linux team use
genl netlink message, when we get a rtnl link change notify, we have to setup
a new genl socket and request the current active port.
v2: check nlmsg_len before copy rta_data
v3: a) Do not make rtnl_buf global as it may be freed by calling rtnl_close()
while we are using it in rtnl_link_status()
b) Reorder declarations of variables as reversed Christmas tree for
function rtnl_link_status()
c) remove rtnl_len
v4: Remove the first !rtnl_buf check in rtnl_link_status as it's alway true
v5: a) Re-order {nl, rtnl}_open and add function nl_close()
b) revert the v3_{a,c}, v4 changes, use nl_close to close genl fd
c) do not use len in get_team_active_iface() as it may mislead reader
v6: Return index at the end to fix fd leak in get_team_active_iface()
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Add provision to disable delay requests. Designated masters who do not
need to calculate pdelay can use this option. This is required by
Automotive Profile to reduce network load.
Signed-off-by: Vedang Patel <vedang.patel@intel.com>
This commit adds functionality to increase the sync and pdelay request
intervals once gptp synchronization has been achieved. This is useful
while running Automotive Profile where the network is usually static.
Here, we are assuming that the gPTP synchronization is achieved whenever
the last 'n' offsets calculated using the Sync messages are below a
certain threshold. Both, the number of offsets to consider and the offset
threshold are configurable.
In order to support the interval update, a new clock servo state
CLOCK_SERVO_STABLE has been added. The clock will transition to this
state whenever the above condition is satisfied. When this transition
occurs, the slave will send an "interval request" signaling message
asking the master to change the sync interval. It will also update the
pdelay request interval locally.
Four new config options have been added to provide this functionality:
- servo_offset_threshold: All the offset values being considered should
be below this value.
- servo_num_offset_values: number of previously received offset values
to consider.
- operLogSyncInterval: slave will ask the master to switch to this
interval for sending sync messages when the offset stabilizes.
- operLogPdelayReqInterval: the slave will set this interval for pdelay
request messages when the offset stabilizes.
Signed-off-by: Vedang Patel <vedang.patel@intel.com>
This implements transmission and receipt of message interval requests.
Using this mechanism, a time-aware device can request change in sync,
link delay and announce intervals on the device connected on the other
end.
As part of the commit, we are also saving the initial values of Announce
and Sync Interval.
For more information look at sections 10.2.4.4, 10.3.9.5, 10.3.14, 10.4,
10.5.4 and 11.2.17 of 802.1AS standard
Please note that this commit does not implement logic related to
computeNeighborRateRatio and computeNeighborPropDelay flags mentioned in
Section 10.5.4.3.9 of 802.1AS Standard.
[ RC - memset the targetPortIdentity to all ones. ]
Signed-off-by: Vedang Patel <vedang.patel@intel.com>
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
It will be used to store the current value of Pdelay Request Interval.
It will be represented as power of two in seconds.
Signed-off-by: Vedang Patel <vedang.patel@intel.com>
Make port_signaling_construct() to create a generic signaling
ptp_message. Move the unicast specific code to
port_signaling_uc_construct().
This is done to support the creation of Message Interval Request TLV (as
defined in 802.1AS Section 10.5.4.3). This TLV can be used to request
change in Announce Interval, Sync Interval or Link Delay Interval on a
remote system.
[ RC - fix trivial rebase conflict on top of targetPortIdentity check. ]
Signed-off-by: Vedang Patel <vedang.patel@intel.com>
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
IEEE 1588 specifies dropping signaling messages if the
targetPortIdentity does not match the receiving port. Up until now
the unicast code did check that field, but only for the unicast
transmission request and cancel operations.
This patch moves the target port identity check so that it applies to
each and every received signaling message.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Ever since commit d88b4ff229 ("util: Added IPv6 support in help
funtions addreq and str2addr"), the unicast master table supports
entries with IPv6 addresses. This patch updates the man page to
reflect this fact.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
In logic, when we want to switch phc, we should check if the new phc
index is valid instead of checking the previous one.
In reality, if we use linux team interface with activebackup mode. As
teamd is a userspace tool, it sets the new slave as active port after
receiving link change message. If we set current active port down and
another slave up. There is a race that we receive the new slave's link
up message while active port(ts_index) is still the old one. This means
we may use a link down interface as ts_index and get phc_index with -1.
If we update the p->phc_index to -1, there will be no possibility to
change it back to other value as we swith phc only when p->phc_index >= 0.
With this fix, we will not switch phc_index until receiving the real
active port(p->iface->ts_info.phc_index >= 0) update message.
Reported-by: Miroslav Lichvar <mlichvar@redhat.com>
Fixes: 536a71031d ("ptp4l: use ts label to get ts info")
Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
The manpage for recvmsg says -1 will be returned on error, Zero indicates an
"orderly shutdown", presumably only in case of stream sockets.
Further, UNIX Network Programming, Vol 1 says ".. a return value of 0 from
recvfrom is acceptable for a datagram protocol"
Such packets have been observed in the wild, aimed at PTP's multicast
address and port, possibly related to malformed management queries.
Patch to properly check return from recvmesg and not trigger the fault
codepath. Instead, such packets are treated as "Bad Message" the same as
non-zero but still too-short UDP payloads.
Signed-off-by: David Mirabito <davidjm@arista.com>
Commit c8107c8d ("config: Add a configuration option for TC mode.")
added a new TC mode. An early version of the patch series included a
new '-t' command line option, but this was dropped considering the
fact that long options are always available. IOW the TC user can
simply add '--clock_type=P2P_TC' to the command line, making the
single dash option redundant.
This patch fixes the misleading usage message by removing the
non-existing option.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Now that PTP support support is well established, remove the text that
attempts to explain the (at the time) incomplete kernel support. Instead
just give a couple of ethtool examples.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Adjust checks of logMessageInterval in received packet to prevent 1 of
type int being shifted by values >= 31.
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
A clock that cannot be adjusted has no servo. Check if a servo is
present before updating the clock.
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
If synchronizing a PHC to the system clock, use one of the
PTP_SYS_OFFSET ioctls (if supported) to measure the offset between the
two clocks. Negate the offset and switch the timestamp before passing
them to the servo.
This makes the synchronization between PHC and system clock symmetric.
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
This is a more accurate variant of the the PTP_SYS_OFFSET ioctl, which
will probably be supported in future kernel versions.
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
This ioctl uses cross timestamping for a more accurate measurement of
the offset. It is supported on some onboard Intel NICs using the e1000e
driver and a virtual PHC with the ptp_kvm driver.
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
The kernel supports different PTP_SYS_OFFSET* ioctls. Use the sysoff
enum to allow selecting between them in sysoff_measure().
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
Added check so that convertion from a positive tmv not result in a negative TimeInterval and vice versa.
Signed-off-by: Anders Selhammer <anders.selhammer@est.tech>
I have reduced the requirements for the -p option to accept
any type of device. With this it should be possible to use
ptp4l to syncronize any clock device that implements the kernel
POSIX clock interface.
[ RC - Initialize struct timex. ]
Signed-off-by: Dimitrios Katsaros <patcherwork@gmail.com>
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
When inhibit_announce is set, ANNOUNCE_RECEIPT_TIMEOUT_EXPIRES event is
still sent once every state change. Avoid this by explicitly sending
EV_NONE instead.
[ RC - Simplify and keep all of the timer's state maintenance. ]
Signed-off-by: Vedang Patel <vedang.patel@intel.com>
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
The "inhibit_announce" and "ignore_source_id" options were introduced
in order to support the automotive profile. That profile happens to
specify using the peer to peer delay mechanism, but there is no reason
to restrict them to that mechanism. As it now stands, enabling these
two options prevents E2E slaves from synchronizing. This patch fixes
the issue by checking the "ignore_source_id" option when accepting E2E
delay response messages.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Richard Cochran
Miroslav Lichvar
Petr Machata
Vladimir Oltean
Khor, Isaac Shi Yan
Hubert Feurstein
Hangbin Liu
Vedang Patel
David Mirabito via Linuxptp-devel
Anders Selhammer
Dimitrios Katsaros