deeacbcdf3
Enhanced phc2sys to support multiple PTP timing clock sources, to select the best clock source available according to the clocks statuses. A series of changes were patched back from a newer version of the linuxptp package. They contain the support for multiple pmc nodes, necessary to communicate with multiple ptp4l instances. The phc2sys is now able to automatically select the best clock according to the clocks statuses, and also monitors any clock status change and act to keep the system synchronized to the best clock source available. A new set of configuration options were added to control the behavior of the automatic clock selection algorithm and to configure the clock source interfaces. And a new command interface was added to phc2sys to gather statuses about the clocks configured and the HA algorithm. This interface also provides commands to control the clock source selection for debug and maintenance proposes. Test plan: PASS: Verify the linuxptp package is built containing all the patches. PASS: Verify the .deb package produced in the build install and that all the PTP different deployments new and existing are OK. PASS: Verify that a new image containing the linuxptp packed can be installed and that all the PTP different deployments new and existing are OK. Story: 2010723 Task: 48642 Change-Id: Ic8c4d9bd335c582a91f1f4418947a81fc5e8b4f9 Signed-off-by: Andre Mauricio Zelak <andre.zelak@windriver.com>
46 lines
1.6 KiB
Diff
46 lines
1.6 KiB
Diff
From 5caa4d0a9161e6a33e269c8e445b322e4437e6b3 Mon Sep 17 00:00:00 2001
|
|
Message-Id: <5caa4d0a9161e6a33e269c8e445b322e4437e6b3.1630418391.git.Jim.Somerville@windriver.com>
|
|
In-Reply-To: <0389752e3aecf8d2b2743f16ce1408a58088bea9.1630418391.git.Jim.Somerville@windriver.com>
|
|
References: <0389752e3aecf8d2b2743f16ce1408a58088bea9.1630418391.git.Jim.Somerville@windriver.com>
|
|
From: Miroslav Lichvar <mlichvar@redhat.com>
|
|
Date: Mon, 31 May 2021 11:07:55 +0200
|
|
Subject: [PATCH 4/47] port: Don't renew raw transport.
|
|
|
|
Renewing of the transport on announce/sync timeout is needed in the
|
|
client-only mode to avoid getting stuck with a broken multicast socket
|
|
when the link goes down.
|
|
|
|
This shouldn't be necessary with the raw transport. Closing and binding
|
|
of raw sockets can apparently be so slow that it triggers a false
|
|
positive in the clock check.
|
|
|
|
Reported-by: Amar Subramanyam <asubramanyam@altiostar.com>
|
|
Signed-off-by: Miroslav Lichvar <mlichvar@redhat.com>
|
|
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
|
|
[commit 6df84259647757bc53818a039734f8ff85618c02 upstream]
|
|
Signed-off-by: Jim Somerville <Jim.Somerville@windriver.com>
|
|
---
|
|
port.c | 6 ++++++
|
|
1 file changed, 6 insertions(+)
|
|
|
|
diff --git a/port.c b/port.c
|
|
index 387d5a2..d26b87f 100644
|
|
--- a/port.c
|
|
+++ b/port.c
|
|
@@ -1805,6 +1805,12 @@ static int port_renew_transport(struct port *p)
|
|
if (!port_is_enabled(p)) {
|
|
return 0;
|
|
}
|
|
+
|
|
+ /* Closing and binding of raw sockets is too slow and unnecessary */
|
|
+ if (transport_type(p->trp) == TRANS_IEEE_802_3) {
|
|
+ return 0;
|
|
+ }
|
|
+
|
|
transport_close(p->trp, &p->fda);
|
|
port_clear_fda(p, FD_FIRST_TIMER);
|
|
res = transport_open(p->trp, p->iface, &p->fda, p->timestamping);
|
|
--
|
|
2.29.2
|
|
|