> I think it would be more better for test-stp.c to just call
> vlog_set_pattern() itself, rather than each use of it including it on
> the command line. You could also have it disable logging to syslog
> with vlog_set_levels(), since that just makes the unit tests spam
> syslog. If you make thos
This can be useful when testing.
Suggested-by: Reid Price
Signed-off-by: Ethan Jackson
---
lib/cfm.c | 47 +++-
vswitchd/ovs-vswitchd.8.in |4 +++
2 files changed, 50 insertions(+), 1 deletions(-)
diff --git a/lib/cfm.c b/lib/cfm.c
Here's an incremental.
---
NEWS |2 ++
vswitchd/vswitch.xml |6 +++---
2 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/NEWS b/NEWS
index d7332f8..2565b78 100644
--- a/NEWS
+++ b/NEWS
@@ -6,6 +6,8 @@ post-v1.5.0
- The default bond_mode changed from S
On Thu, Jan 26, 2012 at 12:35, Ben Pfaff wrote:
> On Wed, Jan 25, 2012 at 06:12:25PM -0800, Ethan Jackson wrote:
>> Bond rebalances come with a risk of packet reordering which some
>> users may find unacceptable.
>>
>> Requested-by: Ben Basler
>> Signed-off-by: Ethan Jackson
>
> Add a NEWS item?
Bug #8827.
Signed-off-by: Ben Pfaff
---
NEWS |3 ++
include/openflow/nicira-ext.h | 10 -
lib/classifier.c | 77 +++
lib/classifier.h |6 +++-
lib/flow.c| 44 ++
Signed-off-by: Ben Pfaff
---
lib/nx-match.c | 175 +---
1 files changed, 66 insertions(+), 109 deletions(-)
diff --git a/lib/nx-match.c b/lib/nx-match.c
index 63e5e5b..f42bb9a 100644
--- a/lib/nx-match.c
+++ b/lib/nx-match.c
@@ -407,6 +407,55
These patches add support for bitwise matching on TCP and UDP ports,
which is a useful building block that controllers can use to match
on TCP and UDP port ranges.
Ben Pfaff (3):
flow: Add a couple of missing build assertions on FLOW_WC_SEQ.
nx-match: Factor redundant code out of nx_put_match(
Every piece of code that may need to change whenever struct flow or struct
flow_wildcards changes, but might easily get overlooked, should have a
build assertion on the value of FLOW_WC_SEQ, but these functions did not.
Signed-off-by: Ben Pfaff
---
lib/flow.c |6 ++
1 files changed, 6 in
On Fri, Jan 27, 2012 at 03:55:17PM -0800, Gurucharan Shetty wrote:
> Change the default timestamp for console and file logs to
> UTC in a format that satisfies timestamp requirements in RFC 5424.
>
> Also, add the ability for ovs-appctl to log timestamps in UTC.
>
> Bug #9052.
>
> Signed-off-by:
Change the default timestamp for console and file logs to
UTC in a format that satisfies timestamp requirements in RFC 5424.
Also, add the ability for ovs-appctl to log timestamps in UTC.
Bug #9052.
Signed-off-by: Gurucharan Shetty
---
NEWS |3 +++
lib/dynamic-string.c
> Minor variants in comment style, like whether the /* and */ are on
> lines by themselves, just don't bother me much, which is why
> CodingStyle says, "You may put the /* and */ on the same line as
> comment text if you prefer." I normally write them on the same line,
> so that's probably why a l
On Fri, Jan 27, 2012 at 02:15:18PM -0800, Ethan Jackson wrote:
> However, the CodingStyle document says to write comments the way that
> you've done in this patch. Ben, is there a particular reason for the
> way the CodingStyle document says to write block comments? If not,
> I'd be inclined to c
> +/*
> + * Writes the time 'tm' to 'string' based on 'template'. A Null 'tm'
> + * writes the current utc time or localtime to 'string' based on 'utc'
> + */
Two minor nits. First, this comment requieres a period at the end of
the second sentence.
Second. In most of the code we don't do a newl
Looks good.
Ethan
On Fri, Jan 27, 2012 at 12:50, Ben Pfaff wrote:
> Signed-off-by: Ben Pfaff
> ---
> lib/table.c | 4 +++-
> 1 files changed, 3 insertions(+), 1 deletions(-)
>
> diff --git a/lib/table.c b/lib/table.c
> index ff11e78..537fae9 100644
> --- a/lib/table.c
> +++ b/lib/table.c
>
Signed-off-by: Ben Pfaff
---
lib/table.c |4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)
diff --git a/lib/table.c b/lib/table.c
index ff11e78..537fae9 100644
--- a/lib/table.c
+++ b/lib/table.c
@@ -480,8 +480,10 @@ table_print_json__(const struct table *table, const struct
table_s
This eliminates a kluge that was duplicated in three different daemons.
Signed-off-by: Ben Pfaff
---
lib/daemon.c | 36 ++--
lib/daemon.h |3 ++-
ovsdb/ovsdb-client.c |8 ++--
tests/test-netflow.c | 11 ++-
utilities/ovs
Suggestion #9347.
Suggested-by: Alan Shieh
Signed-off-by: Ben Pfaff
---
NEWS|3 +++
lib/table.c | 43 ++-
lib/table.h |4 +++-
ovsdb/ovsdb-client.1.in |7 +++
ovsdb/ovsdb-client.c| 13
Signed-off-by: Ben Pfaff
---
lib/table.man |2 +-
ovsdb/ovsdb-client.1.in |4
2 files changed, 5 insertions(+), 1 deletions(-)
diff --git a/lib/table.man b/lib/table.man
index 70ef778..a8f1094 100644
--- a/lib/table.man
+++ b/lib/table.man
@@ -35,7 +35,7 @@ as text strings
On Fri, Jan 27, 2012 at 11:39:15AM -0800, Gurucharan Shetty wrote:
> Change the default timestamp for console and file logs to
> UTC in a format that satisfies timestamp requirements in RFC 5424.
>
> Also, add the ability for ovs-appctl to log timestamps in UTC.
>
> Bug #9052.
>
> Signed-off-by:
On Fri, Jan 27, 2012 at 11:39:14AM -0800, Gurucharan Shetty wrote:
> The ovs-appctl man page has missing information about the
> facility - "file". This patch adds that information.
>
> Signed-off-by: Gurucharan Shetty
Good catch. I wonder how we missed documenting that?
Thanks,
Ben.
Change the default timestamp for console and file logs to
UTC in a format that satisfies timestamp requirements in RFC 5424.
Also, add the ability for ovs-appctl to log timestamps in UTC.
Bug #9052.
Signed-off-by: Gurucharan Shetty
---
lib/dynamic-string.c | 13 +++--
lib/dynami
The ovs-appctl man page has missing information about the
facility - "file". This patch adds that information.
Signed-off-by: Gurucharan Shetty
---
utilities/ovs-appctl.8.in | 17 +
1 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/utilities/ovs-appctl.8.in b/utili
I updated the documentation. Thanks for pointing out that it was
needed. I remember reviewing that and didn't notice the problem at
the time.
A unit test is a good idea.
Thanks,
Ben.
On Thu, Jan 26, 2012 at 05:21:17PM -0800, Ethan Jackson wrote:
> Thanks for fixing this Ben.
>
> I think some
23 matches
Mail list logo