Incorrect buffer was used in size of and error printout.
This patch corrects this.
Signed-off-by: Anders Selhammer
---
sk.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/sk.c b/sk.c
index 0fe009a..69492d4 100644
--- a/sk.c
+++ b/sk.c
@@ -203,10 +203,10 @@ static int sk_
This patch adds network namespace as a configuration parameter for
each interface. This parameter is only needed when configuring
interfaces in more than one namespace.
Signed-off-by: Anders Selhammer
---
config.c | 1 +
ptp4l.8 | 9 +
2 files changed, 10 insertions(+)
diff --git a/con
When your network is designed, it might be situations where you want
interfaces to operate in different namespaces. If those interfaces are
used in a Boundary Clock (BC), the ptp4l program needs o be able to
handle configuration in several namespaces. It is not enough just to
open the ptp4l program
When more than one namespace is used by the interfaces, the ptp4l program
needs the possibility to switch between namespaces when configuring
those interfaces. This patch adds that possibility.
The patch require the ptp4l program to start in default namespace so that
the program can switch back to
Initialization of this global attribute is made from the configuration file.
Initialization in sk.c is never used, therefore removed.
Signed-off-by: Anders Selhammer
---
sk.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/sk.c b/sk.c
index e2b1f28..0fe009a 100644
--- a/s
When interfaces are configured the configuring program needs to be in same
namespace as the interface operates in. This patch make this switch just
before any file descriptor is opened. As long as the file descriptor remain
open, it does not matter if program later on switch to other namespace.
Si
Good explanation.
Sound good to me.
Erez
From: Anders Selhammer [anders.selham...@est.tech]
Sent: 13 November 2018 11:34
To: linuxptp-devel@lists.sourceforge.net
Subject: [Linuxptp-devel] [PATCH v2 0/5] Possibility to use ports in different
namespaces
Whe
net_tstamp.h in recent kernel versions requires time.h for clockid_t.
Signed-off-by: Miroslav Lichvar
---
clock.c | 1 +
sk.c | 1 +
timemaster.c | 1 +
3 files changed, 3 insertions(+)
diff --git a/clock.c b/clock.c
index 56bc79b..464cc83 100644
--- a/clock.c
+++ b/clock.c
@@ -17,
net_tstamp.h in recent kernel versions requires time.h for clockid_t.
Signed-off-by: Miroslav Lichvar
---
clock.c | 2 +-
sk.c | 1 +
timemaster.c | 1 +
3 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/clock.c b/clock.c
index 56bc79b..ef9a008 100644
--- a/clock.c
+++ b
Added check so that convertion from a positive tmv not result in a negative
TimeInterval and vice versa.
Signed-off-by: Anders Selhammer
---
tmv.h | 7 +++
1 file changed, 7 insertions(+)
diff --git a/tmv.h b/tmv.h
index b8339e6..cca4da7 100644
--- a/tmv.h
+++ b/tmv.h
@@ -26,6 +26,8 @@
#i
On Tue, Nov 13, 2018 at 05:48:46AM +, Anders Selhammer wrote:
> Yes, thats true. Forgot that I have an own TLV on the side, making it
> possible to change
> settings by reading the configuration file again and change whats needed.
And remember, the TLV can set the MAC address option.
> Lets
On Tue, Nov 13, 2018 at 01:16:08PM +0100, Miroslav Lichvar wrote:
> net_tstamp.h in recent kernel versions requires time.h for clockid_t.
Can you tell us which kernel exactly?
(I'll put it into the README.)
Thanks,
Richard
___
Linuxptp-devel mailing
On Tue, Nov 13, 2018 at 07:09:22AM -0800, Richard Cochran wrote:
> On Tue, Nov 13, 2018 at 01:16:08PM +0100, Miroslav Lichvar wrote:
> > net_tstamp.h in recent kernel versions requires time.h for clockid_t.
>
> Can you tell us which kernel exactly?
>
> (I'll put it into the README.)
It seems the
On Tue, Nov 06, 2018 at 08:47:47PM +, Patel, Vedang wrote:
> About the criterion, we can check the last 'n' offsets and see if all those
> are below a threshold. If that is true, we can send the signaling message to
> increase the sync interval. Both, the offset threshold and the number of
On Tue, 13 Nov 2018 10:34:20 +, Anders Selhammer wrote:
> + snprintf(path, sizeof path, "/var/run/netns/%s", ns);
This path is just a convention used by iproute2. There's nothing that
forces a network name space to be mounted in /var/run/netns. It can be
mounted elsewhere or even n
15 matches
Mail list logo