Signed-off-by: Alex Wang
---
ofproto/ofproto-dpif-upcall.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/ofproto/ofproto-dpif-upcall.c b/ofproto/ofproto-dpif-upcall.c
index f00c17f..654fbd3 100644
--- a/ofproto/ofproto-dpif-upcall.c
+++ b/ofproto/ofproto-dpif
sorry for the confusion, please ignore my patch~
On Sun, Jul 27, 2014 at 6:29 PM, Justin Pettit wrote:
> Your patch had two fixes, so he should probably grab yours.
>
> --Justin
>
>
> > On Jul 27, 2014, at 5:37 PM, Ben Pfaff wrote:
> >
> > Yeah, but I'm happy if Alex wants to push it since I d
Signed-off-by: Alex Wang
---
ofproto/connmgr.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/ofproto/connmgr.c b/ofproto/connmgr.c
index b7d5d3b..c6432bf 100644
--- a/ofproto/connmgr.c
+++ b/ofproto/connmgr.c
@@ -1496,7 +1496,7 @@ ofconn_wants_packet_in_on_miss(struct
connmgr_wants_packet_in_on_miss() is called by multiple threads
and thusly should be protected by the mutex.
Signed-off-by: Alex Wang
---
ofproto/connmgr.c |6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/ofproto/connmgr.c b/ofproto/connmgr.c
index c6432bf..89af6b6
The connmgr_wants_packet_in_on_miss() should only check 'ofconn's
of type OFCONN_PRIMARY (i.e. controller connections). This commit
makes it happen.
Signed-off-by: Alex Wang
---
ofproto/connmgr.c | 16 +---
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git
the ofctl monitor to simulate controller and capture the
packet_ins.
So, could you help me clarify if we need consider ofctl conn as a
controller connection?
Thanks,
Alex Wang,
On Tue, Jul 29, 2014 at 11:24 AM, Alex Wang wrote:
> The connmgr_wants_packet_in_on_miss() should only chec
use this fix. Do you think it is okay?
Thanks,
Alex Wang,
On Tue, Jul 29, 2014 at 1:23 PM, Andy Zhou wrote:
> Should we also consider lock annotate the ofproto->connmgr variable?
>
> Also, it seems to me the lock can be per ofproto, instead of a global
> lock, but I am not s
On Tue, Jul 29, 2014 at 2:16 PM, Andy Zhou wrote:
> On Tue, Jul 29, 2014 at 2:09 PM, Alex Wang wrote:
> > Thx for the suggestion, I think it will make the connmgr module safer
> > if we annotate the ofproto->connmgr. With a quick look, I think it will
> > take more ch
Thx, pushed both patches to master and branch-2.3.
I'll try improving the thread-safety of connmgr module later,
On Tue, Jul 29, 2014 at 2:32 PM, Alex Wang wrote:
>
>
>
> On Tue, Jul 29, 2014 at 2:16 PM, Andy Zhou wrote:
>
>> On Tue, Jul 29, 2014 at 2:09 PM, Alex
Hey Ben,
Sorry for the confusion,
On Tue, Jul 29, 2014 at 2:52 PM, Ben Pfaff wrote:
> On Tue, Jul 29, 2014 at 11:30:12AM -0700, Alex Wang wrote:
> > I think we should not check 'ofconn's for OFCONN_SERVICE... This
> > change broke several tests:
> >
> > *
27; to
controller, connmgr_wants_packet_in_on_miss() should always returns false,
and 'ofctl monitor' should not work.
After checking the code further, I think we treat 'ofconn' from 'ovs-ofctl
monitor' as 'controller&
This commit flips the default value of bfd ip source and destination,
so that they match the default value of ip destination and source
of vtep schema.
Signed-off-by: Alex Wang
---
lib/bfd.c|4 ++--
vswitchd/vswitch.xml |4 ++--
2 files changed, 4 insertions(+), 4 deletions
matching of MAC address on recevied BFD packets.
Signed-off-by: Alex Wang
---
lib/bfd.c| 60 +++---
vswitchd/vswitch.xml | 32 +++
2 files changed, 75 insertions(+), 17 deletions(-)
diff --git a/lib/bfd.c b/lib/b
This commit adds options for configuring the MAC addresses
in BFD state machine. Therein, the "bfd_local_src_mac" and
"bfd_local_dst_mac" configure the MAC address of sent BFD
packets. The "bfd_remote_dst_mac" configure the matching
of MAC address on recevied BFD
Signed-off-by: Alex Wang
---
lib/netdev-dpdk.c | 14 ++
lib/netdev-dpdk.h |8
2 files changed, 22 insertions(+)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index 6ee9803..7298334 100644
--- a/lib/netdev-dpdk.c
+++ b/lib/netdev-dpdk.c
@@ -1442,6 +1442,20
The pmd threads are pinned to available cpu cores on the
corresponding cpu socket. Note, core 0 is not pinnable
and is reserved for all non-pmd threads.
Signed-off-by: Alex Wang
---
lib/dpif-netdev.c | 254 +
lib/dpif-netdev.h |2 +-
lib
This commit makes the memory pool name contain the socket id.
Since dpdk library do not allow creation of memory pool with
same name, this commit serves as a simple way of making each
name unique.
Signed-off-by: Alex Wang
Acked-by: Thomas Graf
---
lib/netdev-dpdk.c |3 ++-
1 file changed
Later patch will use this patch to configure the pmd thread
cpu affinity.
Signed-off-by: Alex Wang
---
lib/ovs-numa.c | 87
lib/ovs-numa.h |7 +
2 files changed, 82 insertions(+), 12 deletions(-)
diff --git a/lib/ovs-numa.c b
Later patch will use this function to configure rx queues.
Signed-off-by: Alex Wang
---
lib/netdev-dpdk.c | 42 +-
lib/netdev-dpdk.h |7 +++
2 files changed, 48 insertions(+), 1 deletion(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
32 udp 64 |32 udp 64-2|
| Tx.Load |100|100|
| Tx.Saturation| 99.9809428128 | 99.9806658144 |
|Tx.DropRate | 0.000304743297568 | 0.000304743297568 |
++---+---+
Alex Wang (8):
nfigurable to help distribution of rx packets among multiple
pmd threads.
Signed-off-by: Alex Wang
---
lib/dpif-linux.c |1 +
lib/dpif-netdev.c | 130 ++--
lib/dpif-provider.h|7 +++
lib/dpif.c | 17 +
on the cpu
socket. Also each dpdk interface will have one tx queue for
each cpu core, even though not all of those queues will be
used.
Signed-off-by: Alex Wang
---
lib/dpif-netdev.h |1 -
lib/netdev-dpdk.c | 55 +
2 files changed, 47
Signed-off-by: Alex Wang
---
Changes since first post:
- use readdir() for the buffer safety.
- only compile ovs-numa.c in linux.
- make ovs_numa_get_n_sockets/cores() return *UNSPEC if there
is no sockets or cores extracted.
- refine code based on Ben's review.
- Address review comments
Looks good to me,
Acked-by: Alex Wang
On Mon, Aug 4, 2014 at 1:02 PM, Ben Pfaff wrote:
> On Wed, Jul 16, 2014 at 03:00:15PM -0700, Ben Pfaff wrote:
> > The controller setup in my personal test environment has been broken for
> a
> > while. I figured that it wasn't any
Looks good to me,
On Mon, Aug 11, 2014 at 2:36 PM, Ben Pfaff wrote:
> These error codes are proposed in the ONF extensibility working group as an
> OpenFlow 1.3 extension.
>
> Error codes are also proposed for the other three "bad property" error
> codes, but those already have standardized Ope
Thanks Ben, for the review,
> > +
> > +/* To avoid sparse warning. */
> > +#ifdef __linux__
> ...
> > +#endif /* __linux__ */
>
> I didn't see the connection to sparse at first. I'm still not sure
> there is one. Maybe a better comment:
> /* On non-Linux, these functions are defined inline in
Signed-off-by: Alex Wang
---
Changes since first post:
- use readdir() for the buffer safety.
- only compile ovs-numa.c in linux.
- make ovs_numa_get_n_sockets/cores() return *UNSPEC if there
is no sockets or cores extracted.
- refine code based on Ben's review.
- Address review comments
On Wed, Aug 13, 2014 at 12:42 PM, Ben Pfaff wrote:
> Acked-by: Ben Pfaff
>
Thx, applied to master!
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Thx for the fix, Acked-by: Alex Wang
On Wed, Aug 13, 2014 at 2:19 PM, Ben Pfaff wrote:
> list_size() yields a size_t, not uint64_t, and these types are different on
> 32-bit targets.
>
> Signed-off-by: Ben Pfaff
> ---
> lib/ovs-numa.c |4 ++--
> 1 file change
Thx, applied both patches to master and branch-2.3.
On Wed, Aug 13, 2014 at 6:49 PM, Ethan Jackson wrote:
> Acked-by: Ethan Jackson
>
>
> On Mon, Aug 11, 2014 at 6:34 PM, Alex Wang wrote:
> > This commit adds options for configuring the MAC addresses
> > in BFD stat
This commit adds more explanation for the bfd diagnostic and
bfd remote diagnostic in vswitch and vtep documentation.
Requested-by:Bruce Davie
Signed-off-by: Alex Wang
---
vswitchd/vswitch.xml | 10 ++
vtep/vtep.xml| 10 ++
2 files changed, 12 insertions(+), 8
Thx, applied to master and branch-2.3.
On Thu, Aug 14, 2014 at 1:59 PM, Bruce Davie wrote:
> Looks good, thanks.
>
> Bruce
>
> On Aug 14, 2014, at 11:02 AM, Alex Wang wrote:
>
> > This commit adds more explanation for the bfd diagnostic and
> > bfd remote
cause ovs abort.
This commit fixes the above issue by defining an array of the
variable on the function stack.
Signed-off-by: Alex Wang
---
ofproto/ofproto-dpif-upcall.c | 27 ++-
1 file changed, 14 insertions(+), 13 deletions(-)
diff --git a/ofproto/ofproto-dpif
Thx, applied to master,
On Fri, Aug 15, 2014 at 1:33 PM, Ethan Jackson wrote:
> Acked-by: Ethan Jackson
>
> Thanks for taking care of this.
>
> Ethan
>
> On Fri, Aug 15, 2014 at 1:15 AM, Alex Wang wrote:
> > Commit cc377352d (ofproto: Reorganize in preparation
This commit clarifies the documentation for 'bfd:enable'
when it is not specified.
Signed-off-by: Alex Wang
---
vswitchd/vswitch.xml |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml
index ab738da..6c49250 100644
---
Thx, applied to master and branch-2.3
On Tue, Aug 19, 2014 at 11:52 AM, Ben Pfaff wrote:
> On Tue, Aug 19, 2014 at 11:06:04AM -0700, Alex Wang wrote:
> > This commit clarifies the documentation for 'bfd:enable'
> > when it is not specified.
> >
> > Signed
Signed-off-by: Alex Wang
---
vswitchd/vswitch.xml |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml
index 6c49250..42e944b 100644
--- a/vswitchd/vswitch.xml
+++ b/vswitchd/vswitch.xml
@@ -2096,7 +2096,7 @@
default is 00
Thx, applied to master and branch-2.3,
On Tue, Aug 26, 2014 at 12:26 AM, Justin Pettit wrote:
> Acked-by: Justin Pettit
>
> --Justin
>
>
> On August 25, 2014 at 10:42:24 PM, Alex Wang (al...@nicira.com) wrote:
> > Signed-off-by: Alex Wang
> > ---
> > vs
>
>
> This should be part of netdev_class api rather than dpdk specific API.
> we should try to minimize dpdk specific call in dpif-netdev layer.
>
> One more issue I noticed in this series is that socket and numa is
> used interchangeably, can we use just one?
>
>
I'll add it to netdev_class api.
On Sat, Aug 30, 2014 at 12:02 PM, Pravin Shelar wrote:
> On Mon, Aug 11, 2014 at 9:56 PM, Alex Wang wrote:
> > Before this commit, ovs only creates one tx/rx queue for each
> > dpdk interface and uses only one poll thread for handling the
> > I/O of all dpdk interfaces
>
> > static struct dp_netdev_port *dp_netdev_lookup_port(const struct
> dp_netdev *dp,
> > @@ -281,6 +281,15 @@ struct dp_netdev_actions
> *dp_netdev_flow_get_actions(
> > const struct dp_netdev_flow *);
> > static void dp_netdev_actions_free(struct dp_netdev_actions *);
> >
> > +/* Represe
>
>
> > diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
> > index 012ee68..3013df5 100644
> > --- a/lib/netdev-dpdk.c
> > +++ b/lib/netdev-dpdk.c
> > @@ -130,6 +130,8 @@ enum { DRAIN_TSC = 20ULL };
> >
> > static int rte_eal_init_ret = ENODEV;
> >
> > +static size_t dpdk_rx_queues OVS_GUARD
>
> > > Generally we should describe subject in less than 70 characters.
> > > Commit msg should explain why we are introducing this change. It is
> > > not clear from the patch the relation between number of core on socket
> > > and rx queues.
> > >
> > >
> >
> > Thx, I'll use the 70 characters ru
On Tue, Sep 2, 2014 at 11:27 AM, wrote:
> It also brings the usage
> of the BFD diagnostic keys into line with the recent clarifications
> made for OVS.
>
Does this commit miss this part? did not see any change/line move related
to diagnostic,
___
dev
Thx, applied this to master.
On Sat, Aug 30, 2014 at 12:01 PM, Pravin Shelar wrote:
> On Mon, Aug 11, 2014 at 9:56 PM, Alex Wang wrote:
> > This commit makes the memory pool name contain the socket id.
> > Since dpdk library do not allow creation of memory pool with
> > s
rt_try_ref()
which uses ovs_refcount_try_ref_rcu(). And the pmd thread
will only load the port's queue, if port_try_ref() returns
true.
Found by inspection.
Signed-off-by: Alex Wang
---
lib/dpif-netdev.c | 17 -
1 file changed, 16 insertions(+), 1 deletion(-)
diff -
This commit fixes a bug which prevents the display of interface
status for dpdk0.
Found by inspection.
Signed-off-by: Alex Wang
---
lib/netdev-dpdk.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index 049ea5e..a4ed355 100644
--- a
>
> > @@ -1864,7 +1874,10 @@ pmd_load_queues(struct pmd_thread *f,
> > index = 0;
> >
> > CMAP_FOR_EACH (port, node, &f->dp->ports) {
> > -if (netdev_is_pmd(port->netdev)) {
> > +/* Calls port_try_ref() to prevent the main thread
> > + * from deleting the port. */
Thx, applied to master
On Tue, Sep 2, 2014 at 7:27 PM, Pravin Shelar wrote:
> On Tue, Sep 2, 2014 at 6:44 PM, Alex Wang wrote:
> > This commit fixes a bug which prevents the display of interface
> > status for dpdk0.
> >
> > Found by inspection.
> >
> >
index++;
}
-index++;
}
+/* Unrefs the port_try_ref(). */
+port_unref(port);
}
Will apply to master soon,
On Tue, Sep 2, 2014 at 9:39 PM, Alex Wang wrote:
> > @@ -1864,7 +1874,10 @@ pmd_load_queues(struct pmd_thread *
Many of the ovs_numa_*() functions abort the program when the
input cpu socket or core id is invalid. This commit relaxes
the input check and makes these functions return OVS_*_UNSPEC
when the check fails.
Signed-off-by: Alex Wang
---
lib/ovs-numa.c | 77
Signed-off-by: Alex Wang
---
lib/ovs-numa.c | 18 ++
lib/ovs-numa.h |7 +++
2 files changed, 25 insertions(+)
diff --git a/lib/ovs-numa.c b/lib/ovs-numa.c
index 545b075..2de2aa2 100644
--- a/lib/ovs-numa.c
+++ b/lib/ovs-numa.c
@@ -168,6 +168,24 @@ ovs_numa_get_n_cores
Add a short description of the module and its assumption.
Signed-off-by: Alex Wang
---
lib/ovs-numa.c | 12
1 file changed, 12 insertions(+)
diff --git a/lib/ovs-numa.c b/lib/ovs-numa.c
index 2de2aa2..ad44d95 100644
--- a/lib/ovs-numa.c
+++ b/lib/ovs-numa.c
@@ -36,6 +36,18
Thx for the clarification, I'll replace the 'socket' with 'numa' and repost
the series.
On Thu, Sep 4, 2014 at 1:32 PM, Pravin Shelar wrote:
> On Thu, Sep 4, 2014 at 1:26 PM, Alex Wang wrote:
> > Add a short description of the module and its assumption
'numa' and 'socket' are currently used interchangeably in ovs-numa.
But they are not always equivalent as some platform can have multiple
sockets on a numa node. To avoid confusion, this commit renames all
the 'cpu_socket' to 'numa_node'.
Signed-off-b
Many of the ovs_numa_*() functions abort the program when the
input cpu socket or core id is invalid. This commit relaxes
the input check and makes these functions return OVS_*_UNSPEC
when the check fails.
Signed-off-by: Alex Wang
---
lib/ovs-numa.c | 77
Add a short description of the module and its assumption.
Signed-off-by: Alex Wang
---
lib/ovs-numa.c | 12
1 file changed, 12 insertions(+)
diff --git a/lib/ovs-numa.c b/lib/ovs-numa.c
index 4698967..c909d5a 100644
--- a/lib/ovs-numa.c
+++ b/lib/ovs-numa.c
@@ -36,6 +36,18
Signed-off-by: Alex Wang
---
lib/ovs-numa.c | 17 +
lib/ovs-numa.h |7 +++
2 files changed, 24 insertions(+)
diff --git a/lib/ovs-numa.c b/lib/ovs-numa.c
index bcdea3b..4698967 100644
--- a/lib/ovs-numa.c
+++ b/lib/ovs-numa.c
@@ -168,6 +168,23 @@ ovs_numa_get_n_cores
Thx for pointing it out, I'll make a note in the comment.
On Fri, Sep 5, 2014 at 2:19 PM, Pravin Shelar wrote:
> On Thu, Sep 4, 2014 at 3:17 PM, Alex Wang wrote:
> > Many of the ovs_numa_*() functions abort the program when the
> > input cpu socket or core id is invalid.
Thanks, applied all to master
On Fri, Sep 5, 2014 at 2:20 PM, Pravin Shelar wrote:
> On Thu, Sep 4, 2014 at 3:17 PM, Alex Wang wrote:
> > Add a short description of the module and its assumption.
> >
> > Signed-off-by: Alex Wang
>
>
> LG
This commit updates the pointer to 'struct numa_node'
when initializing the 'struct cpu_core'.
Signed-off-by: Alex Wang
---
lib/ovs-numa.c |1 +
1 file changed, 1 insertion(+)
diff --git a/lib/ovs-numa.c b/lib/ovs-numa.c
index 4f0ba0d..7da1407 100644
--- a/lib/ovs-
mmit fixes the issue by making 'upcall_receive()' always
initialize the 'upcall->put_actions'.
Found by inspection.
Signed-off-by: Alex Wang
---
ofproto/ofproto-dpif-upcall.c |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/ofproto/ofproto-dpif-u
Thx Ben,
> It's a little unusual for an initialization function that fails to
> still leave the object that it initializes ready to be destroyed. If
> upcall_receive() fails, is there other data in 'upcall' that needs to
> be destroyed?
No, I don't think there is other data to be destroyed..
ommit fixes the issue by making the caller not conduct
the uninitialize of the 'upcall' when there is error.
Found by inspection.
Signed-off-by: Alex Wang
---
PATCH -> V2:
1. make the caller just return the error.
2. fix the comment to warn the user.
---
ofproto/ofproto-dpif-up
ad'
for NON_PMD_CORE_ID.
- the received packet processing functions are refactored to use
'struct dp_netdev_pmd_thread' as input.
- the 'netdev_send()' function will be called with the proper
queue id.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase
x27; which is
set if the queue is to be used by a non-local pmd thread.
Then, when enqueueing the tx pkts, if the flag is set, the tx
queue will always be flushed immediately after the enqueue.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refactor the code.
---
lib/netdev-dpdk.c |
will be used, each poll thread will
have its own queue for transmission on the dpdk interface.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refactor the code.
---
lib/dpif-netdev.h |1 -
lib/netdev-dpdk.c | 54 -
2 files changed,
ues is not supported. Upcomping patches will start
using it and create multiple tx queues for dpdk netdev by default.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refactor the code.
---
lib/dpif-netdev.c |2 +-
lib/dpif-netdev.h |2 ++
lib/netdev-bsd.c |6 ++
er needed. And this commit
removes it from 'struct dpdk_tx_queue'.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refactor the code.
---
lib/netdev-dpdk.c |6 --
1 file changed, 6 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index d6bf0bd..26b15
This commit adds a new API to the 'struct netdev_class' which
allows user to query the numa node id the 'netdev' is on.
Currently, only netdev-dpdk module implements this function.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refactor the code.
---
lib/netdev-b
Thx, applied to master,
On Mon, Sep 8, 2014 at 10:52 AM, Ben Pfaff wrote:
> On Mon, Sep 08, 2014 at 08:29:22AM -0700, Alex Wang wrote:
> > This commit updates the pointer to 'struct numa_node'
> > when initializing the 'struct cpu_core'.
> >
> > Si
Thx, applied to master,
On Mon, Sep 8, 2014 at 11:18 AM, Ben Pfaff wrote:
> On Mon, Sep 08, 2014 at 11:13:34AM -0700, Alex Wang wrote:
> > On current master, when 'upcall_receive()' returns error, the
> > ofpbuf 'upcall->put_actions' is uninitialized
run,
there will be a refactor of the use of ovs-rcu module, in which all
callback events only free the old version of objects.
Signed-off-by: Alex Wang
---
lib/ovs-rcu.c |8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/lib/ovs-rcu.c b/lib/ovs-rcu.c
index a052d6c..6fe6
This commit adds a new API to the 'struct netdev_class' which
allows user to query the numa node id the 'netdev' is on.
Currently, only netdev-dpdk module implements this function.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refactor the code.
V2 -> V3:
ues is not supported. Upcomping patches will start
using it and create multiple tx queues for dpdk netdev by default.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refactor the code.
V2 -> V3:
- rebase.
---
lib/dpif-netdev.c |2 +-
lib/dpif-netdev.h |2 ++
lib/netdev-bsd
will be used, each poll thread will
have its own queue for transmission on the dpdk interface.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refactor the code.
V2 -> V3:
- rebase.
---
lib/dpif-netdev.h |1 -
lib/netdev-dpdk.c
er needed. And this commit
removes it from 'struct dpdk_tx_queue'.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refactor the code.
V2 -> V3:
- rebase.
---
lib/netdev-dpdk.c |6 --
1 file changed, 6 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-
x27; which is
set if the queue is to be used by a non-local pmd thread.
Then, when enqueueing the tx pkts, if the flag is set, the tx
queue will always be flushed immediately after the enqueue.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refactor the code.
V2 -> V3:
- rebase.
---
ad'
for NON_PMD_CORE_ID.
- the received packet processing functions are refactored to use
'struct dp_netdev_pmd_thread' as input.
- the 'netdev_send()' function will be called with the proper
queue id.
Signed-off-by: Alex Wang
---
PATCH -> V2
- rebase and refacto
Signed-off-by: Alex Wang
---
utilities/ovs-appctl.c | 19 +++
1 file changed, 19 insertions(+)
diff --git a/utilities/ovs-appctl.c b/utilities/ovs-appctl.c
index bb17ec2..8249bf8 100644
--- a/utilities/ovs-appctl.c
+++ b/utilities/ovs-appctl.c
@@ -81,6 +81,19 @@ main(int argc
Signed-off-by: Alex Wang
---
lib/dpif-netdev.c |4 ++--
lib/netdev-dummy.c |2 +-
lib/timeval.c |2 +-
3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 072ed5d..a4dabb6 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif
There is no use case of this option in ovsdb-server. Also,
it causes dpif-dummy and netdev-dummy module register unrelated
unixctl commands.
Signed-off-by: Alex Wang
---
ovsdb/ovsdb-server.c |7 ---
tests/ovsdb-server.at |2 +-
2 files changed, 1 insertion(+), 8 deletions(-)
diff
Remove the include of vlog.h and all the related macros, since
it is not really used.
Signed-off-by: Alex Wang
---
utilities/ovs-appctl.c |7 ---
1 file changed, 7 deletions(-)
diff --git a/utilities/ovs-appctl.c b/utilities/ovs-appctl.c
index 8249bf8..174d147 100644
--- a/utilities
There is no use case of this option in ovsdb-server. Also,
it causes dpif-dummy and netdev-dummy module register unrelated
unixctl commands.
Signed-off-by: Alex Wang
---
ovsdb/ovsdb-server.c |7 ---
tests/ovsdb-server.at |2 +-
2 files changed, 1 insertion(+), 8 deletions(-)
diff
Signed-off-by: Alex Wang
---
lib/dpif-netdev.c |4 ++--
lib/netdev-dummy.c |2 +-
lib/timeval.c |2 +-
3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 072ed5d..a4dabb6 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif
Signed-off-by: Alex Wang
---
utilities/ovs-appctl.c | 19 +++
1 file changed, 19 insertions(+)
diff --git a/utilities/ovs-appctl.c b/utilities/ovs-appctl.c
index bb17ec2..f279f50 100644
--- a/utilities/ovs-appctl.c
+++ b/utilities/ovs-appctl.c
@@ -81,6 +81,19 @@ main(int argc
Just realized that I mistakenly sent this series (not ready) out. Please
ignore,
I have sent the new version (3 patches instead of 4) without adding a
version number... sorry,
On Wed, Sep 10, 2014 at 12:52 PM, Alex Wang wrote:
> Remove the include of vlog.h and all the related macros, si
ote:
> On Wed, Sep 10, 2014 at 02:12:31PM -0700, Alex Wang wrote:
> > Signed-off-by: Alex Wang
>
> What's it for?
>
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Thx Ben,
I'll adjust the commit message,
Also, i think it makes sense to generalize into a function in command-line
module.
Alex Wang,
On Thu, Sep 11, 2014 at 11:03 AM, Ben Pfaff wrote:
> Can you add that to the commit message?
>
> Does it make any sense to generalize this
Thx, applied to master,
On Thu, Sep 11, 2014 at 10:45 AM, Ben Pfaff wrote:
> On Wed, Sep 10, 2014 at 02:12:29PM -0700, Alex Wang wrote:
> > There is no use case of this option in ovsdb-server. Also,
> > it causes dpif-dummy and netdev-dummy module register unrelated
>
command.
Signed-off-by: Alex Wang
---
lib/command-line.c | 23 +++
lib/command-line.h |1 +
utilities/ovs-appctl.c | 18 ++
3 files changed, 42 insertions(+)
diff --git a/lib/command-line.c b/lib/command-line.c
index cb73a25..820f3ff 100644
--- a
the
pkt immediately (via calling dpif_netdev_execute()).
On Thu, Sep 11, 2014 at 10:39 PM, Pravin Shelar wrote:
> On Tue, Sep 9, 2014 at 5:00 PM, Alex Wang wrote:
> > With this commit, ovs by default will try creating 'number of
> > dpdk interfaces on numa node' pmd t
>
> >
> > Specifically, the default number of rx queues will be the number
> > of dpdk interfaces on the numa node. And the upcoming work
> > will assign each rx queue to a different poll thread. The default
> > number of tx queues will be the number of cpu cores on the machine.
> > Although not
;, not
for
specifying the number of queues.
On Thu, Sep 11, 2014 at 11:22 PM, Pravin Shelar wrote:
> On Thu, Sep 11, 2014 at 10:56 PM, Alex Wang wrote:
> >> >
> >> > Specifically, the default number of rx queues will be the number
> >> > of dpdk interfaces on
etion.d/
or manually run it via . ovs-appctl-compgen.bash.
Also, a unit testsuite is provided as ovs-appctl-compgen-test.bash.
It is suggested that this testsuit be run inside ovs-sandbox.
Signed-off-by: Alex Wang
---
utilities/ovs-appctl-compgen-test.bash | 586 +++
depends on this patch:
http://openvswitch.org/pipermail/dev/2014-September/045617.html
On Fri, Sep 12, 2014 at 12:14 AM, Alex Wang wrote:
> This patch is a RFC for the bash command-line completion script
> for ovs-appctl command. Right now, the script can do the following:
>
> -
, Sep 09, 2014 at 11:23:34AM -0700, Alex Wang wrote:
> > On current master, the per-thread callback event set is flushed
> > when ovsrcu_quiesce_start() is called or when the callback
> > event set is full. For threads that only call 'ovsrcu_quiesce()'
> > to ind
Applied to master with the changes mentioned in previous email.
Thanks,
On Fri, Sep 12, 2014 at 10:18 AM, Alex Wang wrote:
> Thx for pointing it out!
>
> Actually, I did 'perthread = pthread_getspecific(perthread_key);' instead
> of
> using 'ovsrcu_perthread_get
Applied the first two patches to master.
Will rebase the following ones based on the discussion and comments.
On Thu, Sep 11, 2014 at 10:38 PM, Pravin Shelar wrote:
> On Tue, Sep 9, 2014 at 5:00 PM, Alex Wang wrote:
> > This commit adds new variable n_txq to 'struct netdev&
dule implements this function.
Signed-off-by: Alex Wang
---
PATCH -> V4:
- move this hidden patch forward.
---
lib/netdev-bsd.c |1 +
lib/netdev-dpdk.c | 46 +-
lib/netdev-dummy.c|1 +
lib/netdev-linux.c|1 +
lib/n
1 - 100 of 1759 matches
Mail list logo
