Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 14 +-
1 file changed, 5 insertions(+), 9 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 24717cc..060f5e0 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -430,8
Instead of performing every operation inside do_port_add() it seems
clearer to introduce port_create(), since we already have
port_destroy().
No functional change.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.
Found by inspection.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 1e8a37c..24717cc 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -986,6
the problem by removing the first signal() from the
pmd thread.
This is hardly a problem on current master, because the main thread
will call the first wait() a long time after the creation of a pmd
thread. It becomes a problem with the next commits.
Signed-off-by: Daniele Di Proietto <dipro
tion have been merged.
* Fixed netdev_dpdk_set_config(): dev->requested_n_rxq is now accessed
while holding the appropriate mutex.
* Fixed some outdated comments about rx queue configuration.
Daniele Di Proietto (15):
dpif-netdev: Destroy 'port_mutex' in dp_netdev_free().
dpif-netdev: Remove unused 'index
Thanks for writing this up, applied to master!
2016-04-19 3:35 GMT-07:00 Kevin Traynor :
> For a given release this is listed in the INSTALL.DPDK.md
> but it gets asked quite a bit on the mailing list, so create
> a table in the FAQ.
>
> Signed-off-by: Kevin Traynor
Thanks for adding this tests!
Acked-by: Daniele Di Proietto <diproiet...@vmware.com>
On 20/04/2016 16:07, "Joe Stringer" <j...@ovn.org> wrote:
>Signed-off-by: Joe Stringer <j...@ovn.org>
>---
> tests/system-common-macros.at | 4
Acked-by: Daniele Di Proietto <diproiet...@vmware.com>
On 21/04/2016 13:29, "Joe Stringer" <j...@ovn.org> wrote:
>On 20 April 2016 at 16:07, Joe Stringer <j...@ovn.org> wrote:
>> Signed-off-by: Joe Stringer <j...@ovn.org>
>> ---
>
>
>&
Thanks for fixing this
Acked-by: Daniele Di Proietto <diproiet...@vmware.com>
On 21/04/2016 14:10, "Joe Stringer" <j...@ovn.org> wrote:
>This was missed before somehow, which would cause the test to fail
>(rather than being skipped) if iproute2 didn't suppor
On 21/04/2016 14:54, "Ben Pfaff" <b...@ovn.org> wrote:
>On Thu, Apr 21, 2016 at 09:41:03PM +, Daniele Di Proietto wrote:
>>
>>
>> On 21/04/2016 11:28, "Ben Pfaff" <b...@ovn.org> wrote:
>>
>> >On Tue, Apr 19, 2016 at 0
On 21/04/2016 11:28, "Ben Pfaff" <b...@ovn.org> wrote:
>On Tue, Apr 19, 2016 at 03:28:39PM -0700, Daniele Di Proietto wrote:
>> Makes popping each member of the hmap a bit easier.
>>
>> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
>
On 20/04/2016 07:21, "Ilya Maximets" <i.maxim...@samsung.com> wrote:
>On 20.04.2016 01:28, diproiettod at vmware.com (Daniele Di Proietto)
>wrote:
>> netdev objects are hard to use with RCU, because it's not possible to
>> split removal and reclamat
ailed analysis and the suggested fix.
I applied the suggested incremental, but kept emc_cache_uninit()
where it is right now.
I sent an updated version here:
http://openvswitch.org/pipermail/dev/2016-April/069835.html
Thanks,
Daniele
>
>
>> On 08.04.2016 06:13, Daniele D
to the proper NUMA socket.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 69 +-
lib/netdev-bsd.c | 2 +-
lib/netdev-dpdk.c | 195 ++
lib/netdev-dummy.c| 2 +-
lib/netdev-l
An upcoming commit will make dpif_poll_threads_set() record the
requested configuration and dpif_run() apply it, so it makes sense to
change the order.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
Tested-by: Ilya Maximets <i.maxim...@samsung.com>
Acked-by: Ilya Maxim
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 144 ++--
lib/dpif-provider.h | 3 +-
2 files changed, 84 insertions(+), 63 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 8
rx and tx queues in
netdev-dpdk.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
Tested-by: Ilya Maximets <i.maxim...@samsung.com>
Acked-by: Ilya Maximets <i.maxim...@samsung.com>
Acked-by: Mark Kavanagh <mark.b.kavan...@intel.com>
---
lib/netdev-bsd.c |
Errors returned by netdev_set_multiq() and netdev_rxq_open() weren't
handled properly in reconfigure_pmd_threads(). In case of error now we
remove the port from the datapath.
Also, part of the code is moved in a new function, port_reconfigure().
Signed-off-by: Daniele Di Proietto <dipro
state, so explicit calls to
ovsrcu_quiesce_start() and ovsrcu_quiesce_end() are added there.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/ovs-rcu.h | 3 +--
lib/ovs-thread.c| 2 --
vswitchd/system-stats.c | 6 ++
3 files changed, 7 insertions
to the RCU model we use in userspace.
This commit changes the port container from cmap to hmap. 'port_mutex'
must be held by readers and writers. This shouldn't have performace
impact, as readers in the fast path use a thread local cache.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.
Makes popping each member of the hmap a bit easier.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/cfm.c| 5 ++---
lib/hmap.h | 4
lib/id-pool.c| 5 ++---
lib/learning-switch.c| 5 ++---
lib/
The interface will be more similar to the cmap.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/hmap.c | 26 --
lib/hmap.h | 7 ++-
lib/sset.c | 12 +---
lib/sset.h | 7 ++-
o
eds to synchronize with
every pmd thread.
Among the advantages, keeping a per thread port mapping could allow
greater control over the txq assigment.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 249 +++---
1 file cha
Instead of duplicating code to add ports in
dp_netdev_set_pmds_on_numa(), we can always use
dp_netdev_add_port_to_pmds__().
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 58 +--
1 file chang
the problem by removing the first signal() from the
pmd thread.
This is hardly a problem on current master, because the main thread
will call the first wait() a long time after the creation of a pmd
thread. It becomes a problem with the next commits.
Signed-off-by: Daniele Di Proietto <dipro
Instead of performing every operation inside do_port_add() it seems
clearer to introduce port_create(), since we already have
port_destroy().
No functional change.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.
This commit introduces some functions to add/remove rxqs from pmd
threads without reloading them. They will be used by next commits.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 77 ---
1 file chang
Found by inspection.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 1e8a37c..24717cc 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -986,6
outdated comments about rx queue configuration.
Daniele Di Proietto (16):
dpif-netdev: Destroy 'port_mutex' in dp_netdev_free().
dpif-netdev: Remove unused 'index' in dp_netdev_pmd_thread.
dpif-netdev: Factor out port_create() from do_add_port().
dpif-netdev: Add functions to modify rx
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 14 +-
1 file changed, 5 insertions(+), 9 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 24717cc..060f5e0 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -430,8
On 18/04/2016 14:26, "Ben Pfaff" <b...@ovn.org> wrote:
>On Mon, Apr 18, 2016 at 02:02:37PM -0700, Daniele Di Proietto wrote:
>> CodingStyle.md says:
>>
>> "Use form feeds (control+L) to divide long source files into logical
>> pieces. A form fe
On 18/04/2016 07:50, "Ilya Maximets" <i.maxim...@samsung.com> wrote:
>On 08.04.2016 06:13, Daniele Di Proietto wrote:
>> Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
>> ---
>> lib/dpif-netdev.c | 243
>>
vs_fatal(0, "batch_size must be between 1 and
>NETDEV_MAX_BURST(%u)",
You're right there are several genuine errors reported by checkpatch.py.
I'll fix them.
>
>patches 7/15 and 8/15, 10/15, 11/15 - as there's no comment into the
>patch, maybe
>they just need an empty
W(199): Line has trailing whitespace
+
This commit suppresses the two warnings for lines with form feeds as the
only character.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
utilities/checkpatch.py | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a
of 'unsigned' for 'len', since the value can be
negative.
CC: Sairam Venugopal <vsai...@vmware.com>
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
I tested a similar fix on the userspace connection tracker, but I didn't
compile this for the windows datapath.
---
datapath-wi
-by: Daniele Di Proietto <diproiet...@vmware.com>
---
tests/system-kmod-macros.at | 28 +++
tests/system-traffic.at | 49 ++--
tests/system-userspace-macros.at | 45 +---
3 files changed, 107 inse
and to allow the pickup of already
established connections.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
COPYING | 1 +
debian/copyright.in | 4 +
lib/automake.mk | 5 +
lib/conntrack-other.c | 91 ++
lib/conntrack-private.h
around.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
tests/system-common-macros.at| 1 +
tests/system-kmod-macros.at | 7 +++
tests/system-userspace-macros.at | 18 ++
3 files changed, 26 insertions(+)
diff --git a/tests/system-common-macros
While the system testsuite already has connection tracking tests, it
will be still useful to add some to the standard testsuite because:
* They're run more often by developers.
* Some of them are more interesting for the userspace datapath.
Signed-off-by: Daniele Di Proietto <dipro
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/conntrack-private.h | 3 ++
lib/conntrack-tcp.c | 34 +
lib/conntrack.c | 125
lib/conntrack.h | 16 +++
4 files changed, 178 inse
-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
tests/test-conntrack.c | 65 ++
1 file changed, 65 insertions(+)
diff --git a/tests/test-conntrack.c b/tests/test-conntrack.c
index 414d7dc..5292320 100644
--- a/tests/test-conntrack.c
Useful to test the datapath ability to forward tcp packets without the
complexity of connection tracking.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
tests/system-traffic.at | 20
1 file changed, 20 insertions(+)
diff --git a/tests/system-traffi
This commit implements the OVS_ACTION_ATTR_CT action in dpif-netdev.
To allow ofproto-dpif to detect the conntrack feature, flow_put will not
discard anymore flows with ct_* fields set. We still shouldn't allow
flows with NAT bits set, since there is no support for NAT.
Signed-off-by: Daniele Di
different batching and locking strategies.
E.g. the line:
`./test/ovstest test-conntrack benchmark 1 1488 32`
starts 1 thread that will send 1488 packets to the connection
tracker, 32 at a time. It will print the time taken to process them.
Signed-off-by: Daniele Di Proietto <dipro
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 10 +-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 5ac2bf3..995cbc0 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -3991,6 +3
Linux and FreeBSD have slightly different names for these constants.
Windows doesn't define them. It is simpler to redefine them from
scratch for OVS. The new names are different than those used in Linux
and FreeBSD.
These definitions will be used by a future commit.
Signed-off-by: Daniele Di
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 60 ---
1 file changed, 57 insertions(+), 3 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 436359a..5ac2bf3 100644
--- a/li
This will be used by a future commit.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/flow.c | 140 ++---
lib/flow.h | 3 ++
2 files changed, 81 insertions(+), 62 deletions(-)
diff --git a/lib/flow.c b/lib/
The function simply returns the ethernet type of the packet (after
eventually discarding the VLAN tag). It will be used by a following
commit.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/flow.c | 14 --
lib/flow.h | 1 +
2 files changed, 13 insertions
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/conntrack.c | 21 +
lib/conntrack.h | 2 ++
2 files changed, 23 insertions(+)
diff --git a/lib/conntrack.c b/lib/conntrack.c
index 840335b..7913e76 100644
--- a/lib/conntrack.c
+++ b/lib/connt
egy in conntrack_execute() to allow a newly
created connection to be picked up by packets in the same batch.
* Added an ovs-test module to throw pcap files at the connection tracker.
* Added a workaround for the userspace testsuite on new kernels and a tcp
non-conntrack test.
Daniele Di Proie
d call reconfigure().
>>
>>This mechanism can also be used to:
>>* Automatically match the number of rxq with the one provided by qemu
>> via the new_device callback.
>>* Provide a way to change the MTU of dpdk devices at runtime.
>>* Move a DPDK vhost de
gt;
You're right, I removed the reference to ovs_mutex_cond_wait() there.
Thanks for noticing this!
>
>>
>>In system_stats_thread_func() the code relied on ovs_mutex_cond_wait()
>>to introduce a quiescent state, so explicit calls to
>>ovsrcu_quiesce_start() and ovsrcu_qui
Hi Mark,
On 14/04/2016 05:36, "Kavanagh, Mark B" <mark.b.kavan...@intel.com> wrote:
>Hi Daniele,
>
>One minor comment inline.
>
>Cheers,
>Mark
>
>>
>>The interface will be more similar to the cmap.
>>
>>Signed-off-by: Dan
Thanks for the patch! I pushed this to master
2016-04-14 9:40 GMT-07:00 mweglicx :
> Following changes are applied:
> - INSTALL.DPDK.md: CONFIG_RTE_BUILD_COMBINE_LIBS step has been
>removed because it is no longer present in DPDK configuration
>(combined
On 14/04/2016 14:51, "Ben Pfaff" wrote:
>On Tue, Apr 12, 2016 at 11:44:15AM +0100, Bhanuprakash Bodireddy wrote:
>> When using DPDK datapath, the OVS configure script requires the DPDK
>> build directory passed on --with-dpdk. This can be avoided if DPDK
>> library, headers are
On 10/04/2016 12:23, "Ben Pfaff" <b...@ovn.org> wrote:
>On Fri, Apr 08, 2016 at 03:12:59AM +, Daniele Di Proietto wrote:
>>
>>
>> On 01/04/2016 09:52, "Jarno Rajahalme" <ja...@ovn.org> wrote:
>>
>> >
>> >&g
2016-04-13 9:21 GMT-07:00 Traynor, Kevin :
> > -Original Message-
> > From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Panu
> Matilainen
> > Sent: Wednesday, April 13, 2016 8:50 AM
> > To: Weglicki, MichalX ;
Thanks for the patch, I have a couple of comments:
DPDK 16.04 enables by default checksum offloads and TSO for vhostuser
device. While this seem to work ok, there seem to be a few problems with
this:
* OVS in userspace assumes that a packet is stored using a single mbuf
(it is not aware of
Hi Ian,
On 07/04/2016 06:00, "Stokes, Ian" wrote:
>> > >71034a0..faf3583 100644
>> > >--- a/lib/netdev-dpdk.c
>> > >+++ b/lib/netdev-dpdk.c
>> > >@@ -53,6 +53,7 @@
>> > >
>> > > #include "rte_config.h"
>> > > #include "rte_mbuf.h"
>> > >+#include "rte_meter.h"
>> > >
to the proper NUMA socket.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 69 +-
lib/netdev-dpdk.c | 195 ++
lib/netdev-provider.h | 23 +++---
lib/netdev.c | 34 +++-
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 144 ++--
lib/dpif-provider.h | 3 +-
2 files changed, 84 insertions(+), 63 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index f
Makes popping each member of the hmap a bit easier.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/cfm.c| 5 ++---
lib/hmap.h | 4
lib/id-pool.c| 5 ++---
lib/learning-switch.c| 5 ++---
lib/
Instead of performing every operation inside do_port_add() it seems
clearer to introduce port_create(), since we already have
port_destroy().
No functional change.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.
rx and tx queues in
netdev-dpdk.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
Tested-by: Ilya Maximets <i.maxim...@samsung.com>
Acked-by: Ilya Maximets <i.maxim...@samsung.com>
Acked-by: Mark Kavanagh <mark.b.kavan...@intel.com>
---
lib/netdev-bsd.c |
An upcoming commit will make dpif_poll_threads_set() record the
requested configuration and dpif_run() apply it, so it makes sense to
change the order.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
Tested-by: Ilya Maximets <i.maxim...@samsung.com>
Acked-by: Ilya Maxim
Errors returned by netdev_set_multiq() and netdev_rxq_open() weren't
handled properly in reconfigure_pmd_threads(). In case of error now we
remove the port from the datapath.
Also, part of the code is moved in a new function, port_reconfigure().
Signed-off-by: Daniele Di Proietto <dipro
state, so explicit calls to
ovsrcu_quiesce_start() and ovsrcu_quiesce_end() are added there.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/ovs-thread.c| 2 --
vswitchd/system-stats.c | 6 ++
2 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/l
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 243 +++---
1 file changed, 175 insertions(+), 68 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 8c5893d..5d1cc43 100644
--- a/li
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 93 ---
1 file changed, 55 insertions(+), 38 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 5d1cc43..fba6592 100644
--- a/li
The interface will be more similar to the cmap.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/hmap.c | 26 --
lib/hmap.h | 7 ++-
lib/sset.c | 12 +---
lib/sset.h | 7 ++-
o
This commit introduces some functions to add/remove rxqs from pmd
threads without reloading them. They will be used by next commits.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 77 ---
1 file chang
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 14 +-
1 file changed, 5 insertions(+), 9 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 7959342..78e4e35 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -431,8
outdated comments about rx queue configuration.
Daniele Di Proietto (16):
dpif-netdev: Destroy 'port_mutex' in dp_netdev_free().
dpif-netdev: Remove unused 'index' in dp_netdev_pmd_thread.
dpif-netdev: Factor out port_create() from do_add_port().
dpif-netdev: Add functions to modify rx
Found by inspection.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 2870951..7959342 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -987,6
Instead of duplicating code to add ports in
dp_netdev_set_pmds_on_numa(), we can always use
dp_netdev_add_port_to_pmds__().
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
---
lib/dpif-netdev.c | 58 +--
1 file chang
the problem by removing the first signal() from the
pmd thread.
This is hardly a problem on current master, because the main thread
will call the first wait() a long time after the creation of a pmd
thread. It becomes a problem with the next commits.
Signed-off-by: Daniele Di Proietto <dipro
On 01/04/2016 09:52, "Jarno Rajahalme" <ja...@ovn.org> wrote:
>
>> On Mar 30, 2016, at 8:08 PM, Daniele Di Proietto
>><diproiet...@vmware.com> wrote:
>>
>>
>> On 30/03/2016 16:01, "Ben Pfaff" <b...@ovn.org> wrote:
>>
Thanks for fixing this!
Acked-by: Daniele Di Proietto <diproiet...@vmware.com>
On 06/04/2016 15:07, "Joe Stringer" <j...@ovn.org> wrote:
>Since continuations were introduced, the system-traffic tests which use
>OpenFlow monitors to check the results of datapath
Thanks for the fix!
I've applied this to master and branch-2.5
2016-04-06 16:28 GMT-07:00 William Tu :
> Valgrind reports "Conditional jump or move depends on uninitialised value"
> and "Use of uninitialised value" at case 2016 ovn -- 3 HVs, 1 LS, 3
> lports/HV. It is caused
2016-04-06 10:09 GMT-07:00 Darrell Ball :
> On Wed, Apr 6, 2016 at 9:37 AM, William Tu wrote:
>
>> Valgrind reports "Conditional jump or move depends on uninitialised value"
>> and "Use of uninitialised value" at case 2016 ovn -- 3 HVs, 1 LS, 3
>> lports/HV.
On 01/04/2016 09:52, "Jarno Rajahalme" <ja...@ovn.org> wrote:
>
>> On Mar 30, 2016, at 8:08 PM, Daniele Di Proietto
>><diproiet...@vmware.com> wrote:
>>
>>
>> On 30/03/2016 16:01, "Ben Pfaff" <b...@ovn.org> wrote:
>>
On 30/03/2016 16:01, "Ben Pfaff" <b...@ovn.org> wrote:
>(I'm taking a look at this patch specifically because Daniele asked me;
>I'm not planning to review the whole series.)
>
>On Mon, Mar 28, 2016 at 12:41:40PM -0700, Daniele Di Proietto wrote:
>> The dpif-net
On 29/03/2016 06:44, "Karl Rister" <kris...@redhat.com> wrote:
>On 03/29/2016 08:08 AM, Flavio Leitner wrote:
>> On Tue, Mar 29, 2016 at 02:13:18AM +, Daniele Di Proietto wrote:
>>> Hi Flavio and Karl,
>>>
>>> thanks for the patch! I
On 29/03/2016 06:08, "Flavio Leitner" <f...@redhat.com> wrote:
>On Tue, Mar 29, 2016 at 02:13:18AM +, Daniele Di Proietto wrote:
>> Hi Flavio and Karl,
>>
>> thanks for the patch! I have a couple of comments:
>>
>> Can you point out a c
Thanks!
I applied this to master and branch-2.5
On 28/03/2016 23:20, "Ilya Maximets" wrote:
>According to QEMU documentation (docs/specs/vhost-user.txt) one queue
>should be enabled initially. More queues are enabled dynamically, by
>sending message
Hi Flavio and Karl,
thanks for the patch! I have a couple of comments:
Can you point out a configuration where this is the bottleneck?
I'm interested in reproducing this.
I think the implementation would look simpler if we could
avoid explicitly taking the mutex in dpif-netdev and instead
Thanks for the patch, Ilya
In __netdev_dpdk_vhost_send() we have
qid = vhost_dev->tx_q[qid % vhost_dev->real_n_txq].map;
if (OVS_UNLIKELY(!is_vhost_running(virtio_dev) || qid == -1)) {
Should we change -1 to OVS_VHOST_QUEUE_MAP_UNKNOWN and handle also
OVS_VHOST_QUEUE_DISABLED?
On
On 09/03/2016 09:38, "Aaron Conole" wrote:
>A previous patch introduced the ability to pass arbitrary EAL command
>line options via the dpdk_extras database entry. This commit enhances
>that by warning the user when such a configuration is detected and
>prefering the value
I still have some comment:
dpdk-mem-channels: This is not required by DPDK anymore, so I still
think that's not necessary and could be removed. If someone want
I think we shouldn't abort if we fail something during the initialization.
I know that rte_eal_init() can still abort, but I want to
I'm ok with the idea and the implementation and I'm willing to apply
this patch despite the issue depicted below.
I just want to mention that with just this patch applied I experience
a lower throughput for the single flow phy test on my system (~14 Mpps
instead of 14.88 Mpps).
I thought that
On 09/03/2016 10:38, "Aaron Conole" wrote:
>A previous change moved some commonly used arguments from commandline to
>the database, and with it the ability to pass arbitrary arguments to
>EAL. This change allows arbitrary eal arguments to be provided
>via a new db entry
On 09/03/2016 09:38, "Aaron Conole" wrote:
>The user has control over the DPDK internal lcore coremask, but this
>parameter can be autofilled with a bit more intelligence. If the user
>does not fill this parameter in, we use the lowest set bit in the
>current task CPU
Hi Aaron,
apologies for the delay. I'm finally convinced that this is the
best way for OVS to initialize DPDK.
I've put some comments on the patches, mostly minor fixes, other than
those I think this is ready to be merged
Thanks for all your work,
Daniele
>> Aaron Conole
of returning prematurely from the function without
restarting the pmd threads).
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
Tested-by: Ilya Maximets <i.maxim...@samsung.com>
Acked-by: Ilya Maximets <i.maxim...@samsung.com>
---
lib/
An upcoming commit will make dpif_poll_threads_set() record the
requested configuration and dpif_run() apply it, so it makes sense to
change the order.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
Tested-by: Ilya Maximets <i.maxim...@samsung.com>
Acked-by: Ilya Maxim
This just documents what the current code already assumes.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
Tested-by: Ilya Maximets <i.maxim...@samsung.com>
Acked-by: Ilya Maximets <i.maxim...@samsung.com>
---
lib/dpif-netdev.c | 1 +
1 file changed, 1 insertion(+)
rx and tx queues in
netdev-dpdk.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
Tested-by: Ilya Maximets <i.maxim...@samsung.com>
Acked-by: Ilya Maximets <i.maxim...@samsung.com>
---
lib/netdev-bsd.c | 1 +
lib/netdev-dpdk.c | 1 +
lib/netdev-dummy.c
Hi Mark,
thanks for your comment, I replied inline
On 24/03/2016 10:17, "Kavanagh, Mark B" <mark.b.kavan...@intel.com> wrote:
>Hi Daniele,
>
>One comment inline.
>
>Cheers,
>Mark
>
>>-----Original Message-
>>From: Daniele Di Proietto [mailto
RCU is more in line with other cmap users.
* We might want to allow port removal and queue reconfiguration without
stopping completely all the pmd threads.
Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
Acked-by: Ilya Maximets <i.maxim...@samsung.com>
---
lib/dpif-ne
601 - 700 of 1432 matches
Mail list logo