On Sun, Dec 8, 2019 at 5:18 AM Russell Bryant wrote:
>
> A previous commit added more detail to this log message. Fix the test
> to reflect the new text.
>
> Signed-off-by: Russell Bryant
Hi Russell,
Thanks for fixing the failing test case.
Acked-by: Dumitru Ceara
> ---
> tests/ovn.at |
On 12/9/2019 9:15 AM, 贺鹏 wrote:
> Hi,
>
>
> cmap support multiple readers and one writer.
> since all write operations are performed in the offload thread,
> why need a mutex here? Maybe I miss some patches.
Collecting statistics using netdev_flow_get in patch 10/20 is another
thread. There, it
Hi,
cmap support multiple readers and one writer.
since all write operations are performed in the offload thread,
why need a mutex here? Maybe I miss some patches.
Eli Britstein 于2019年12月8日周日 下午9:23写道:
>
> Flow deletion and dumping for statistics collection are called from
> different
This seems useful as I'm usually making a lot of typing mistakes.
Also, few commonly used words added to the extended dictionary.
Signed-off-by: Ilya Maximets
---
utilities/checkpatch.py | 33 ++---
1 file changed, 22 insertions(+), 11 deletions(-)
diff --git
Words like 'br0' are common and usually references some code or
database objects that should not be targets for spell checking.
So, it's better to skip all the words that has digits inside instead
of ones that starts with numbers.
Signed-off-by: Ilya Maximets
---
utilities/checkpatch.py | 4
Abbreviations of Latin expressions like 'i.e.' or 'e.g.' are common
and known by the dictionary. However, our spell checker is not able
to recognize them because it strips dots out of them. To avoid this
issue we could pass non-stripped version of the word to the dictionary
checker too.
Ilya Maximets (3):
checkpatch: Allow common abbreviations for spell checking.
checkpatch: Skip words containing numbers.
checkpatch: Check spelling in commit messages.
utilities/checkpatch.py | 41 +++--
1 file changed, 27 insertions(+), 14 deletions(-)
Ilya Maximets (2):
dpif: Turn dpif_flow_hash function into generic odp_flow_key_hash.
dpif-netdev: Get rid of broken dpif pointer in dp_netdev structure.
lib/dpif-netdev.c | 12 +---
lib/dpif-netlink.c | 27 ---
lib/dpif.c | 16
This pointer was introduced in July 2014 by commit
6b31e07347ad ("dpif-netdev: Polling threads directly call ofproto upcall
functions.")
and it was broken right from this point because dpif_netdev_open()
updates it on each call with the pointer to a newly allocated
'dpif' structure that becomes
Current implementation of dpif_flow_hash() doesn't depend on datapath
interface and only complicates the callers by forcing them to figure
out what is their current 'dpif'. If we'll need different hashing
for different 'dpif's we'll implement an API for dpif-providers
and each dpif implementation
On 08.12.2019 15:42, Ilya Maximets wrote:
> On 08.12.2019 15:29, Ophir Munk wrote:
>> In case a pmd pointer (struct dp_netdev_pmd_thread *) needs to retrieve
>> the dpif_class it points at - it can access it as: pmd->dp->class. A
>> second option is to access it as: pmd->dp->dpif->dpif_class.
Hi Ilya,
I think the root cause for the issues mentioned below is the usage of
pmd->dp->dpif->dpif_class
rather than pmd->dp->class. Please see a fixing patch
https://patchwork.ozlabs.org/patch/1205684/
("dpif-netdev: Retrieve dpif_class from struct dp_netdev ").
When running dpctl statistics
On 08.12.2019 15:29, Ophir Munk wrote:
> In case a pmd pointer (struct dp_netdev_pmd_thread *) needs to retrieve
> the dpif_class it points at - it can access it as: pmd->dp->class. A
> second option is to access it as: pmd->dp->dpif->dpif_class. The first
> option is safe since there is one dp
In case a pmd pointer (struct dp_netdev_pmd_thread *) needs to retrieve
the dpif_class it points at - it can access it as: pmd->dp->class. A
second option is to access it as: pmd->dp->dpif->dpif_class. The first
option is safe since there is one dp netdev with a constant pointer to
the dpif
Bleep bloop. Greetings Eli Britstein, I am a robot and I have tried out your
patch.
Thanks for your contribution.
I encountered some error that I wasn't expecting. See the details below.
checkpatch:
WARNING: Line is 81 characters long (recommended limit is 79)
#88 FILE:
Bleep bloop. Greetings Eli Britstein, I am a robot and I have tried out your
patch.
Thanks for your contribution.
I encountered some error that I wasn't expecting. See the details below.
checkpatch:
WARNING: Line is 80 characters long (recommended limit is 79)
#159 FILE:
Fixes: e8a2b5bf92bb ("netdev-dpdk: implement flow offload with rte flow")
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
lib/netdev-offload-dpdk.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/netdev-offload-dpdk.c b/lib/netdev-offload-dpdk.c
index
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
Documentation/howto/dpdk.rst | 1 +
NEWS | 3 +-
lib/netdev-dpdk.c| 15 ++
lib/netdev-offload-dpdk.c| 107 +++
4 files changed, 125 insertions(+), 1
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
Documentation/howto/dpdk.rst | 1 +
NEWS | 2 +-
lib/netdev-dpdk.c| 2 ++
lib/netdev-offload-dpdk.c| 4 +---
4 files changed, 5 insertions(+), 4 deletions(-)
diff --git
From: Ophir Munk
Use netdev flow get API in order to update the statistics of fully
offloaded flows.
Co-authored-by: Eli Britstein
Signed-off-by: Ophir Munk
Reviewed-by: Oz Shlomo
Signed-off-by: Eli Britstein
---
lib/dpif-netdev.c | 47 +--
1
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
Documentation/howto/dpdk.rst | 2 ++
NEWS | 4 ++--
lib/netdev-dpdk.c| 14 +
lib/netdev-offload-dpdk.c| 48
4 files changed, 66
Implement the flow get method for DPDK, to get the statistics of the
provided ufid, towards reading statistics of fully offloaded flows.
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
lib/netdev-offload-dpdk.c | 40
1 file changed, 40
Introduce a rte flow query function as a pre-step towards reading HW
statistics of fully offloaded flows.
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
lib/netdev-dpdk.c | 25 +
lib/netdev-dpdk.h | 6 ++
2 files changed, 31 insertions(+)
diff --git
Flow deletion and dumping for statistics collection are called from
different threads. As a pre-step towards collecting HW statistics,
protect the UFID map by mutex to make it thread safe.
Signed-off-by: Eli Britstein
---
lib/netdev-offload-dpdk.c | 7 ++-
1 file changed, 6 insertions(+), 1
Signed-off-by: Eli Britstein
---
lib/netdev-offload-provider.h | 1 +
lib/netdev-offload.c | 12
2 files changed, 13 insertions(+)
diff --git a/lib/netdev-offload-provider.h b/lib/netdev-offload-provider.h
index 4e1c4251d..5a809c0cd 100644
---
Add debug prints when creating and destroying rte flows, with all the
flow details (attributes, patterns, actions).
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
lib/netdev-dpdk.c | 260 ++
lib/netdev-offload-dpdk.c | 207
Support offload of output action, also configuring count action for
allowing query statistics of HW offloaded flows.
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
Documentation/howto/dpdk.rst | 17 ---
NEWS | 1 +
lib/netdev-dpdk.c| 22
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
Documentation/howto/dpdk.rst | 1 +
NEWS | 4 ++--
lib/netdev-dpdk.c| 24
lib/netdev-offload-dpdk.c| 26 ++
4 files changed, 53 insertions(+), 2
Refactor the flow patterns code to a helper function for better
readability and towards supporting more matches.
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
lib/netdev-offload-dpdk.c | 208 +-
1 file changed, 112 insertions(+), 96
Currently HW offload is accelerating only the rule matching sequence.
Introduce a framework for offloading rule actions as a pre-step for
processing the rule actions in HW. In case of a failure, fallback to the
legacy partial offload scheme.
Note: a flow will be fully offloaded only if it can
Instead of statically allocated pattern items on the stack, dynamically
allocate only the required items while parsing the matches, to simplify
the parsing and make it self-contained, without need of external types,
making it easier to support more matches in the future.
Signed-off-by: Eli
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
Documentation/howto/dpdk.rst | 1 +
NEWS | 4 ++--
lib/netdev-dpdk.c| 14 ++
lib/netdev-offload-dpdk.c| 34 ++
4 files changed, 51 insertions(+), 2
Populate dpif class field in offload struct to be used in offloading
flow put.
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
lib/dpif-netdev.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 37e7e5c38..1d1de94de 100644
---
Currently, netdev datapath offload only accelerates the flow match
sequence by associating a mark per flow. This series introduces the full
offload of netdev datapath flows by having the HW also perform the flow
actions.
This series adds HW offload for output, drop, set MAC, set IPv4, set
TCP/UDP
Action item data structures are pointed by rte_flow_action items.
Refactor the code to free the data structures when freeing the
rte_flow_action items, allowing simpler future actions simpler to add to
the code.
Signed-off-by: Eli Britstein
---
lib/netdev-offload-dpdk.c | 92
Add a getter function for using the dpdk port id outside the scope of
netdev-dpdk.c to be used for HW offload.
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
lib/netdev-dpdk.c | 18 ++
lib/netdev-dpdk.h | 2 ++
2 files changed, 20 insertions(+)
diff --git
On Sat, Dec 7, 2019 at 4:53 PM Han Zhou wrote:
>
>
>
> On Sat, Dec 7, 2019 at 3:21 AM Dumitru Ceara wrote:
> >
> > On Sat, Dec 7, 2019 at 1:36 AM Han Zhou wrote:
> > >
> > > Thanks Dumitru for the patch. Please see my comments below.
> > >
> >
> > Hi Han,
> >
> > Thanks for the review.
> >
> >
On 12/6/2019 4:34 PM, Ilya Maximets wrote:
> On 06.12.2019 14:09, Ophir Munk wrote:
>> Hi Ilya,
>> I applied this series on master ("dpdk: Update to use DPDK 19.11.") and
>> PINGed between two "dpdk" ports (hw-offload=true).
>> It worked fine.
>> Then I watched flow statistics (dpif based) by
According to the microsoft doc:
https://docs.microsoft.com/en-us/windows-hardware/drivers/network/hyper-v-extensible-switch-port-and-network-adapter-states
Below OID request sequence is validation:
OID_SWITCH_NIC_CONNECT -> OID_SWITCH_NIC_DISCONNECT
^
On 12/5/2019 6:46 PM, Sriharsha Basavapatna wrote:
> On Wed, Dec 4, 2019 at 8:55 PM Eli Britstein wrote:
>>
>> On 12/3/2019 5:19 PM, Sriharsha Basavapatna wrote:
>>> On Wed, Nov 20, 2019 at 9:07 PM Eli Britstein wrote:
Signed-off-by: Eli Britstein
Reviewed-by: Oz Shlomo
---
Hi William,
GTP-U header size is not constant, you *must* take into account the flags,
mainly
the sequence. The use of sequence in GTP-U is optional but some devices do use
it. see from 3GPP definition:
"For PGW, SGW and eNodeB the usage of sequence numbers in G-PDUs is optional,
but
if
41 matches
Mail list logo