Re: [lng-odp] odp dpdk

[Bogdan Pricope]

t;>> dev_info.tx_desc_lim.nb_max
>> >>> >>> > 1024 ? 1024 : dev_info.tx_desc_lim.nb_max,
>> >>> >>>
>> >>> >>> rte_eth_dev_socket_id(port_id),
>> >>> >>>  txconf);
>> >>> >>>
>> >>> >>> +   ret = rte_eth_rx_queue_setup(port_id, i,
>> >>> >>> +
>> >>> >>> dev_info.rx_desc_lim.nb_max
>> >>> >>> > 1024 ? 1024 : dev_info.rx_desc_lim.nb_max,
>> >>> >>>
>> >>> >>> rte_eth_dev_socket_id(port_id),
>> >>> >>>  NULL,
>> >>> >>> pkt_dpdk->pkt_pool);
>> >>> >>>
>> >>> >>>
>> >>> >>>
>> >>> >>>
>> >>> >>> Maxim.
>> >>> >>>
>> >>> >>> On 5 December 2017 at 11:20, Elo, Matias (Nokia - FI/Espoo)
>> >>> >>>  wrote:
>> >>> >>>>
>> >>> >>>> When I tested enabling HW checksum with Fortville NICs (i40e) the
>> >>> >>>> slower
>> >>> >>>> driver path alone caused ~20% throughput drop on l2fwd test. This
>> >>> >>>> was
>> >>> >>>> without actually calculating the checksums, I simply forced the
>> >>> >>>> slower
>> >>> >>>> driver path (no vectorization).
>> >>> >>>>
>> >>> >>>> -Matias
>> >>> >>>>
>> >>> >>>>
>> >>> >>>> > On 5 Dec 2017, at 8:59, Bogdan Pricope
>> >>> >>>> > 
>> >>> >>>> > wrote:
>> >>> >>>> >
>> >>> >>>> > On RX side is kind-of expected result since it uses scheduler
>> >>> >>>> > mode.
>> >>> >>>> >
>> >>> >>>> > On TX side there is this drop from 10 mpps to 7.69 mpps that is
>> >>> >>>> > unexpected.
>> >>> >>>> >
>> >>> >>>> > So Petri, when you said:
>> >>> >>>> > "DPDK uses less optimized driver code (on Intel NICs at least)
>> >>> >>>> > when
>> >>> >>>> > any of the L4 checksum offloads is enabled."
>> >>> >>>> >
>> >>> >>>> > you were referring to this kind of drop in performance?
>> >>> >>>> >
>> >>> >>>> > There is that 'folklore' that SW csum is faster on small
>> >>> >>>> > packets
>> >>> >>>> > while
>> >>> >>>> > HW csum is faster on bigger packets. Do you have this kind of
>> >>> >>>> > data?
>> >>> >>>> >
>> >>> >>>> > Anyway, for this particular case (odp_generator), since UDP
>> >>> >>>> > header/payload is not changing during the test (for now), csum
>> >>> >>>> > is
>> >>> >>>> > calculated only once at the beginning of the test: so we are
>> >>> >>>> > comparing
>> >>> >>>> > HW IPv4 + HW UDP csum vs. SW IPv4 csum yet, the differences
>> >>> >>>> > in
>> >>> >>>> > performance is huge...
>> >>> >>>> >
>> >>> >>>> >
>> >>> >>>> > On 4 December 2017 at 20:37, Maxim Uvarov
>> >>> >>>> > 
>> >>> >>>> > wrote:
>> >>> >>>> >> I added isocpus and mounted huge page TX became more stable at
>> >>> >>>> >> 7.6M.
>> >>> >>>> >> But
>> >>> >>>> >> anyway it's better to test performance for this PR because
>> >>> >>>> >> previous
>> >>> >>>> >> speed was 10M.
>> >>> >>>> >>
>> >>> >>>> >> Maxim.
>> >

Re: [lng-odp] odp dpdk

[Bill Fischofer]

 On 5 December 2017 at 11:20, Elo, Matias (Nokia - FI/Espoo)
> >>> >>>  wrote:
> >>> >>>>
> >>> >>>> When I tested enabling HW checksum with Fortville NICs (i40e) the
> >>> >>>> slower
> >>> >>>> driver path alone caused ~20% throughput drop on l2fwd test. This
> was
> >>> >>>> without actually calculating the checksums, I simply forced the
> >>> >>>> slower
> >>> >>>> driver path (no vectorization).
> >>> >>>>
> >>> >>>> -Matias
> >>> >>>>
> >>> >>>>
> >>> >>>> > On 5 Dec 2017, at 8:59, Bogdan Pricope <
> bogdan.pric...@linaro.org>
> >>> >>>> > wrote:
> >>> >>>> >
> >>> >>>> > On RX side is kind-of expected result since it uses scheduler
> mode.
> >>> >>>> >
> >>> >>>> > On TX side there is this drop from 10 mpps to 7.69 mpps that is
> >>> >>>> > unexpected.
> >>> >>>> >
> >>> >>>> > So Petri, when you said:
> >>> >>>> > "DPDK uses less optimized driver code (on Intel NICs at least)
> when
> >>> >>>> > any of the L4 checksum offloads is enabled."
> >>> >>>> >
> >>> >>>> > you were referring to this kind of drop in performance?
> >>> >>>> >
> >>> >>>> > There is that 'folklore' that SW csum is faster on small packets
> >>> >>>> > while
> >>> >>>> > HW csum is faster on bigger packets. Do you have this kind of
> data?
> >>> >>>> >
> >>> >>>> > Anyway, for this particular case (odp_generator), since UDP
> >>> >>>> > header/payload is not changing during the test (for now), csum
> is
> >>> >>>> > calculated only once at the beginning of the test: so we are
> >>> >>>> > comparing
> >>> >>>> > HW IPv4 + HW UDP csum vs. SW IPv4 csum yet, the differences
> in
> >>> >>>> > performance is huge...
> >>> >>>> >
> >>> >>>> >
> >>> >>>> > On 4 December 2017 at 20:37, Maxim Uvarov <
> maxim.uva...@linaro.org>
> >>> >>>> > wrote:
> >>> >>>> >> I added isocpus and mounted huge page TX became more stable at
> >>> >>>> >> 7.6M.
> >>> >>>> >> But
> >>> >>>> >> anyway it's better to test performance for this PR because
> >>> >>>> >> previous
> >>> >>>> >> speed was 10M.
> >>> >>>> >>
> >>> >>>> >> Maxim.
> >>> >>>> >>
> >>> >>>> >> On 12/04/17 19:42, Honnappa Nagarahalli wrote:
> >>> >>>> >>> Can you run with Linux-DPDK in ODP 2.0?
> >>> >>>> >>>
> >>> >>>> >>> On 4 December 2017 at 09:54, Maxim Uvarov
> >>> >>>> >>> 
> >>> >>>> >>> wrote:
> >>> >>>> >>>> after clean patches apply and fix in run scripts I made it
> run.
> >>> >>>> >>>>
> >>> >>>> >>>> But results is really bad. --enable-dpdk-zero-copy
> >>> >>>> >>>>
> >>> >>>> >>>> TX rate is:
> >>> >>>> >>>> 7673155 pps
> >>> >>>> >>>>
> >>> >>>> >>>> RX rate is:
> >>> >>>> >>>> 5989846 pps
> >>> >>>> >>>>
> >>> >>>> >>>>
> >>> >>>> >>>> Before patch PR 313 TX was 10M pps.
> >>> >>>> >>>>
> >>> >>>> >>>> I re run task and TX is 3.3M pps. All tests are single core.
> So
> >>> >>>> >>>> something strange happens in lava or this PR.
> >>> >>>> >>>>
> >>

Re: [lng-odp] odp dpdk

[Honnappa Nagarahalli]

gt;>> >
>>> >>>> > So Petri, when you said:
>>> >>>> > "DPDK uses less optimized driver code (on Intel NICs at least) when
>>> >>>> > any of the L4 checksum offloads is enabled."
>>> >>>> >
>>> >>>> > you were referring to this kind of drop in performance?
>>> >>>> >
>>> >>>> > There is that 'folklore' that SW csum is faster on small packets
>>> >>>> > while
>>> >>>> > HW csum is faster on bigger packets. Do you have this kind of data?
>>> >>>> >
>>> >>>> > Anyway, for this particular case (odp_generator), since UDP
>>> >>>> > header/payload is not changing during the test (for now), csum is
>>> >>>> > calculated only once at the beginning of the test: so we are
>>> >>>> > comparing
>>> >>>> > HW IPv4 + HW UDP csum vs. SW IPv4 csum yet, the differences in
>>> >>>> > performance is huge...
>>> >>>> >
>>> >>>> >
>>> >>>> > On 4 December 2017 at 20:37, Maxim Uvarov 
>>> >>>> > wrote:
>>> >>>> >> I added isocpus and mounted huge page TX became more stable at
>>> >>>> >> 7.6M.
>>> >>>> >> But
>>> >>>> >> anyway it's better to test performance for this PR because
>>> >>>> >> previous
>>> >>>> >> speed was 10M.
>>> >>>> >>
>>> >>>> >> Maxim.
>>> >>>> >>
>>> >>>> >> On 12/04/17 19:42, Honnappa Nagarahalli wrote:
>>> >>>> >>> Can you run with Linux-DPDK in ODP 2.0?
>>> >>>> >>>
>>> >>>> >>> On 4 December 2017 at 09:54, Maxim Uvarov
>>> >>>> >>> 
>>> >>>> >>> wrote:
>>> >>>> >>>> after clean patches apply and fix in run scripts I made it run.
>>> >>>> >>>>
>>> >>>> >>>> But results is really bad. --enable-dpdk-zero-copy
>>> >>>> >>>>
>>> >>>> >>>> TX rate is:
>>> >>>> >>>> 7673155 pps
>>> >>>> >>>>
>>> >>>> >>>> RX rate is:
>>> >>>> >>>> 5989846 pps
>>> >>>> >>>>
>>> >>>> >>>>
>>> >>>> >>>> Before patch PR 313 TX was 10M pps.
>>> >>>> >>>>
>>> >>>> >>>> I re run task and TX is 3.3M pps. All tests are single core. So
>>> >>>> >>>> something strange happens in lava or this PR.
>>> >>>> >>>>
>>> >>>> >>>> Maxim.
>>> >>>> >>>>
>>> >>>> >>>>
>>> >>>> >>>> On 12/04/17 17:03, Bogdan Pricope wrote:
>>> >>>> >>>>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0)
>>> >>>> >>>>> I
>>> >>>> >>>>> see:
>>> >>>> >>>>>
>>> >>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
>>> >>>> >>>>> ODP_BRANCH='api-next'
>>> >>>> >>>>>
>>> >>>> >>>>>
>>> >>>> >>>>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1)
>>> >>>> >>>>> I
>>> >>>> >>>>> see:
>>> >>>> >>>>>
>>> >>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
>>> >>>> >>>>> ODP_BRANCH='devel/api-next_shsum'
>>> >>>> >>>>>
>>> >>>> >>>>>
>>> >>>> >>>>> or are you referring to other test?
>>> >>>> >>>>>
>>> >>>> >>>>>
>>> >>>> >>>>> On 4 De

Re: [lng-odp] odp dpdk

[Maxim Uvarov]

 > wrote:
> >> >>>> >
> >> >>>> > On RX side is kind-of expected result since it uses scheduler
> mode.
> >> >>>> >
> >> >>>> > On TX side there is this drop from 10 mpps to 7.69 mpps that is
> >> >>>> > unexpected.
> >> >>>> >
> >> >>>> > So Petri, when you said:
> >> >>>> > "DPDK uses less optimized driver code (on Intel NICs at least)
> when
> >> >>>> > any of the L4 checksum offloads is enabled."
> >> >>>> >
> >> >>>> > you were referring to this kind of drop in performance?
> >> >>>> >
> >> >>>> > There is that 'folklore' that SW csum is faster on small packets
> >> >>>> > while
> >> >>>> > HW csum is faster on bigger packets. Do you have this kind of
> data?
> >> >>>> >
> >> >>>> > Anyway, for this particular case (odp_generator), since UDP
> >> >>>> > header/payload is not changing during the test (for now), csum is
> >> >>>> > calculated only once at the beginning of the test: so we are
> >> >>>> > comparing
> >> >>>> > HW IPv4 + HW UDP csum vs. SW IPv4 csum yet, the differences
> in
> >> >>>> > performance is huge...
> >> >>>> >
> >> >>>> >
> >> >>>> > On 4 December 2017 at 20:37, Maxim Uvarov <
> maxim.uva...@linaro.org>
> >> >>>> > wrote:
> >> >>>> >> I added isocpus and mounted huge page TX became more stable at
> >> >>>> >> 7.6M.
> >> >>>> >> But
> >> >>>> >> anyway it's better to test performance for this PR because
> >> >>>> >> previous
> >> >>>> >> speed was 10M.
> >> >>>> >>
> >> >>>> >> Maxim.
> >> >>>> >>
> >> >>>> >> On 12/04/17 19:42, Honnappa Nagarahalli wrote:
> >> >>>> >>> Can you run with Linux-DPDK in ODP 2.0?
> >> >>>> >>>
> >> >>>> >>> On 4 December 2017 at 09:54, Maxim Uvarov
> >> >>>> >>> 
> >> >>>> >>> wrote:
> >> >>>> >>>> after clean patches apply and fix in run scripts I made it
> run.
> >> >>>> >>>>
> >> >>>> >>>> But results is really bad. --enable-dpdk-zero-copy
> >> >>>> >>>>
> >> >>>> >>>> TX rate is:
> >> >>>> >>>> 7673155 pps
> >> >>>> >>>>
> >> >>>> >>>> RX rate is:
> >> >>>> >>>> 5989846 pps
> >> >>>> >>>>
> >> >>>> >>>>
> >> >>>> >>>> Before patch PR 313 TX was 10M pps.
> >> >>>> >>>>
> >> >>>> >>>> I re run task and TX is 3.3M pps. All tests are single core.
> So
> >> >>>> >>>> something strange happens in lava or this PR.
> >> >>>> >>>>
> >> >>>> >>>> Maxim.
> >> >>>> >>>>
> >> >>>> >>>>
> >> >>>> >>>> On 12/04/17 17:03, Bogdan Pricope wrote:
> >> >>>> >>>>> On TX (https://lng.validation.linaro.org/scheduler/job/
> 23252.0)
> >> >>>> >>>>> I
> >> >>>> >>>>> see:
> >> >>>> >>>>>
> >> >>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
> >> >>>> >>>>> ODP_BRANCH='api-next'
> >> >>>> >>>>>
> >> >>>> >>>>>
> >> >>>> >>>>> On RX (https://lng.validation.linaro.org/scheduler/job/
> 23252.1)
> >> >>>> >>>>> I
> >> >>>> >>>>> see:
> >> >>>> >>>>>
> >> >>>> >&

Re: [lng-odp] odp dpdk

[Bogdan Pricope]

SW csum is faster on small packets
>> >>>> > while
>> >>>> > HW csum is faster on bigger packets. Do you have this kind of data?
>> >>>> >
>> >>>> > Anyway, for this particular case (odp_generator), since UDP
>> >>>> > header/payload is not changing during the test (for now), csum is
>> >>>> > calculated only once at the beginning of the test: so we are
>> >>>> > comparing
>> >>>> > HW IPv4 + HW UDP csum vs. SW IPv4 csum yet, the differences in
>> >>>> > performance is huge...
>> >>>> >
>> >>>> >
>> >>>> > On 4 December 2017 at 20:37, Maxim Uvarov 
>> >>>> > wrote:
>> >>>> >> I added isocpus and mounted huge page TX became more stable at
>> >>>> >> 7.6M.
>> >>>> >> But
>> >>>> >> anyway it's better to test performance for this PR because
>> >>>> >> previous
>> >>>> >> speed was 10M.
>> >>>> >>
>> >>>> >> Maxim.
>> >>>> >>
>> >>>> >> On 12/04/17 19:42, Honnappa Nagarahalli wrote:
>> >>>> >>> Can you run with Linux-DPDK in ODP 2.0?
>> >>>> >>>
>> >>>> >>> On 4 December 2017 at 09:54, Maxim Uvarov
>> >>>> >>> 
>> >>>> >>> wrote:
>> >>>> >>>> after clean patches apply and fix in run scripts I made it run.
>> >>>> >>>>
>> >>>> >>>> But results is really bad. --enable-dpdk-zero-copy
>> >>>> >>>>
>> >>>> >>>> TX rate is:
>> >>>> >>>> 7673155 pps
>> >>>> >>>>
>> >>>> >>>> RX rate is:
>> >>>> >>>> 5989846 pps
>> >>>> >>>>
>> >>>> >>>>
>> >>>> >>>> Before patch PR 313 TX was 10M pps.
>> >>>> >>>>
>> >>>> >>>> I re run task and TX is 3.3M pps. All tests are single core. So
>> >>>> >>>> something strange happens in lava or this PR.
>> >>>> >>>>
>> >>>> >>>> Maxim.
>> >>>> >>>>
>> >>>> >>>>
>> >>>> >>>> On 12/04/17 17:03, Bogdan Pricope wrote:
>> >>>> >>>>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0)
>> >>>> >>>>> I
>> >>>> >>>>> see:
>> >>>> >>>>>
>> >>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
>> >>>> >>>>> ODP_BRANCH='api-next'
>> >>>> >>>>>
>> >>>> >>>>>
>> >>>> >>>>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1)
>> >>>> >>>>> I
>> >>>> >>>>> see:
>> >>>> >>>>>
>> >>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
>> >>>> >>>>> ODP_BRANCH='devel/api-next_shsum'
>> >>>> >>>>>
>> >>>> >>>>>
>> >>>> >>>>> or are you referring to other test?
>> >>>> >>>>>
>> >>>> >>>>>
>> >>>> >>>>> On 4 December 2017 at 15:53, Maxim Uvarov
>> >>>> >>>>>  wrote:
>> >>>> >>>>>>
>> >>>> >>>>>>
>> >>>> >>>>>> On 4 December 2017 at 15:11, Bogdan Pricope
>> >>>> >>>>>> 
>> >>>> >>>>>> wrote:
>> >>>> >>>>>>>
>> >>>> >>>>>>> You need to put 313 on TX side (not RX).
>> >>>> >>>>>>
>> >>>> >>>>>>
>> >>>> >>>>>>
>> >>>> >>>>>> both rx and tx have patches from 313. l2fw

Re: [lng-odp] odp dpdk

[Maxim Uvarov]

 TX became more stable at
> 7.6M.
> >>>> >> But
> >>>> >> anyway it's better to test performance for this PR because previous
> >>>> >> speed was 10M.
> >>>> >>
> >>>> >> Maxim.
> >>>> >>
> >>>> >> On 12/04/17 19:42, Honnappa Nagarahalli wrote:
> >>>> >>> Can you run with Linux-DPDK in ODP 2.0?
> >>>> >>>
> >>>> >>> On 4 December 2017 at 09:54, Maxim Uvarov <
> maxim.uva...@linaro.org>
> >>>> >>> wrote:
> >>>> >>>> after clean patches apply and fix in run scripts I made it run.
> >>>> >>>>
> >>>> >>>> But results is really bad. --enable-dpdk-zero-copy
> >>>> >>>>
> >>>> >>>> TX rate is:
> >>>> >>>> 7673155 pps
> >>>> >>>>
> >>>> >>>> RX rate is:
> >>>> >>>> 5989846 pps
> >>>> >>>>
> >>>> >>>>
> >>>> >>>> Before patch PR 313 TX was 10M pps.
> >>>> >>>>
> >>>> >>>> I re run task and TX is 3.3M pps. All tests are single core. So
> >>>> >>>> something strange happens in lava or this PR.
> >>>> >>>>
> >>>> >>>> Maxim.
> >>>> >>>>
> >>>> >>>>
> >>>> >>>> On 12/04/17 17:03, Bogdan Pricope wrote:
> >>>> >>>>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0)
> I
> >>>> >>>>> see:
> >>>> >>>>>
> >>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
> >>>> >>>>> ODP_BRANCH='api-next'
> >>>> >>>>>
> >>>> >>>>>
> >>>> >>>>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1)
> I
> >>>> >>>>> see:
> >>>> >>>>>
> >>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
> >>>> >>>>> ODP_BRANCH='devel/api-next_shsum'
> >>>> >>>>>
> >>>> >>>>>
> >>>> >>>>> or are you referring to other test?
> >>>> >>>>>
> >>>> >>>>>
> >>>> >>>>> On 4 December 2017 at 15:53, Maxim Uvarov
> >>>> >>>>>  wrote:
> >>>> >>>>>>
> >>>> >>>>>>
> >>>> >>>>>> On 4 December 2017 at 15:11, Bogdan Pricope
> >>>> >>>>>> 
> >>>> >>>>>> wrote:
> >>>> >>>>>>>
> >>>> >>>>>>> You need to put 313 on TX side (not RX).
> >>>> >>>>>>
> >>>> >>>>>>
> >>>> >>>>>>
> >>>> >>>>>> both rx and tx have patches from 313. l2fwd works on recv side.
> >>>> >>>>>> Generator
> >>>> >>>>>> does not work.
> >>>> >>>>>>
> >>>> >>>>>> Maxim.
> >>>> >>>>>>
> >>>> >>>>>>
> >>>> >>>>>>>
> >>>> >>>>>>>
> >>>> >>>>>>> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia -
> >>>> >>>>>>> FI/Espoo)
> >>>> >>>>>>>  wrote:
> >>>> >>>>>>>> Is the DPDK version 17.08 ? Other versions might not work
> >>>> >>>>>>>> properly.
> >>>> >>>>>>>>
> >>>> >>>>>>>>
> >>>> >>>>>>>>
> >>>> >>>>>>>> -Petri
> >>>> >>>>>>>>
> >>>> >>>>>>>>
> >>>> >>>>>>>>
> >>>> >>>>>>>> Fr

Re: [lng-odp] odp dpdk

[Bogdan Pricope]

n run scripts I made it run.
>>>> >>>>
>>>> >>>> But results is really bad. --enable-dpdk-zero-copy
>>>> >>>>
>>>> >>>> TX rate is:
>>>> >>>> 7673155 pps
>>>> >>>>
>>>> >>>> RX rate is:
>>>> >>>> 5989846 pps
>>>> >>>>
>>>> >>>>
>>>> >>>> Before patch PR 313 TX was 10M pps.
>>>> >>>>
>>>> >>>> I re run task and TX is 3.3M pps. All tests are single core. So
>>>> >>>> something strange happens in lava or this PR.
>>>> >>>>
>>>> >>>> Maxim.
>>>> >>>>
>>>> >>>>
>>>> >>>> On 12/04/17 17:03, Bogdan Pricope wrote:
>>>> >>>>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I
>>>> >>>>> see:
>>>> >>>>>
>>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
>>>> >>>>> ODP_BRANCH='api-next'
>>>> >>>>>
>>>> >>>>>
>>>> >>>>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I
>>>> >>>>> see:
>>>> >>>>>
>>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
>>>> >>>>> ODP_BRANCH='devel/api-next_shsum'
>>>> >>>>>
>>>> >>>>>
>>>> >>>>> or are you referring to other test?
>>>> >>>>>
>>>> >>>>>
>>>> >>>>> On 4 December 2017 at 15:53, Maxim Uvarov
>>>> >>>>>  wrote:
>>>> >>>>>>
>>>> >>>>>>
>>>> >>>>>> On 4 December 2017 at 15:11, Bogdan Pricope
>>>> >>>>>> 
>>>> >>>>>> wrote:
>>>> >>>>>>>
>>>> >>>>>>> You need to put 313 on TX side (not RX).
>>>> >>>>>>
>>>> >>>>>>
>>>> >>>>>>
>>>> >>>>>> both rx and tx have patches from 313. l2fwd works on recv side.
>>>> >>>>>> Generator
>>>> >>>>>> does not work.
>>>> >>>>>>
>>>> >>>>>> Maxim.
>>>> >>>>>>
>>>> >>>>>>
>>>> >>>>>>>
>>>> >>>>>>>
>>>> >>>>>>> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia -
>>>> >>>>>>> FI/Espoo)
>>>> >>>>>>>  wrote:
>>>> >>>>>>>> Is the DPDK version 17.08 ? Other versions might not work
>>>> >>>>>>>> properly.
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>> -Petri
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>> From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
>>>> >>>>>>>> Sent: Monday, December 04, 2017 1:10 PM
>>>> >>>>>>>> To: Savolainen, Petri (Nokia - FI/Espoo)
>>>> >>>>>>>> 
>>>> >>>>>>>> Cc: Bogdan Pricope ; lng-odp-forward
>>>> >>>>>>>> 
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>> Subject: Re: [lng-odp] odp dpdk
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>> 313 does not work also:
>>>> >>>>>>>>
>>>> >>>>>>>> https://lng.validation.linaro.org/scheduler/job/23242.1
>>>> >>>>>>>>
>>>> >>>>>>>> I will replace RX side to l2fwd and see that will be there.
>>>> >>>>>>>>
>>>> >>>>>>>> Maxim.
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>> On 4 December 2017 at 13:46, Savolainen, Petri (Nokia -
>>>> >>>>>>>> FI/Espoo)
>>>> >>>>>>>>  wrote:
>>>> >>>>>>>>
>>>> >>>>>>>> Maxim, try https://github.com/Linaro/odp/pull/313 It has been
>>>> >>>>>>>> tested to
>>>> >>>>>>>> fix
>>>> >>>>>>>> checksum insert for 10/40GE Intel NICs.
>>>> >>>>>>>>
>>>> >>>>>>>> -Petri
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>>>> -Original Message-
>>>> >>>>>>>>> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On
>>>> >>>>>>>>> Behalf Of
>>>> >>>>>>>>> Bogdan Pricope
>>>> >>>>>>>>> Sent: Monday, December 04, 2017 12:21 PM
>>>> >>>>>>>>> To: Maxim Uvarov 
>>>> >>>>>>>>> Cc: lng-odp-forward 
>>>> >>>>>>>>> Subject: Re: [lng-odp] odp dpdk
>>>> >>>>>>>>>
>>>> >>>>>>>>> I suspect this is actually caused by csum issue in TX side: on
>>>> >>>>>>>>> RX,
>>>> >>>>>>>>> socket pktio does not validate csum (and accept the packets)
>>>> >>>>>>>>> but on
>>>> >>>>>>>>> dpdk pktio the csum is validated and packets are dropped.
>>>> >>>>>>>>>
>>>> >>>>>>>>> I am not seeing this in my setup because default txq_flags for
>>>> >>>>>>>>> igb
>>>> >>>>>>>>> driver (1G interface) is
>>>> >>>>>>>>>
>>>> >>>>>>>>> .txq_flags = 0
>>>> >>>>>>>>>
>>>> >>>>>>>>> while for ixgbe (10G interface) is:
>>>> >>>>>>>>>
>>>> >>>>>>>>> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
>>>> >>>>>>>>>ETH_TXQ_FLAGS_NOOFFLOADS,
>>>> >>>>>>>>>
>>>> >>>>>>>>>
>>>> >>>>>>>>> /B
>>>> >>>>>>>>>
>>>> >>>>>>>>>
>>>> >>>>>>>>>
>>>> >>>>>>>>>
>>>> >>>>>>>>> On 1 December 2017 at 23:47, Maxim Uvarov
>>>> >>>>>>>>> 
>>>> >>>>>>>>> wrote:
>>>> >>>>>>>>>>
>>>> >>>>>>>>>> Looking to dpdk pktio support and generator. It looks like
>>>> >>>>>>>>>> receive
>>>> >>>>>>>>>> part
>>>> >>>>>>>>>> is broken. If for receive I use sockets it works well but
>>>> >>>>>>>>>> receive
>>>> >>>>>>>>>> with
>>>> >>>>>>>>>> dpdk does not get any packets. For both master and api-next.
>>>> >>>>>>>>>> Can
>>>> >>>>>>>>>> somebody confirm please that it's so. Lava is not supper
>>>> >>>>>>>>>> friendly to
>>>> >>>>>>>>>> debug issue.
>>>> >>>>>>>>>>
>>>> >>>>>>>>>>
>>>> >>>>>>>>>> 1. Recv
>>>> >>>>>>>>>> odp_generator -I 0 -m r -c 0x4
>>>> >>>>>>>>>>
>>>> >>>>>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.1
>>>> >>>>>>>>>> Network devices using DPDK-compatible driver
>>>> >>>>>>>>>> 
>>>> >>>>>>>>>> :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection
>>>> >>>>>>>>>> 10fb'
>>>> >>>>>>>>>> drv=igb_uio unused=
>>>> >>>>>>>>>>
>>>> >>>>>>>>>>
>>>> >>>>>>>>>>
>>>> >>>>>>>>>> 2. Send
>>>> >>>>>>>>>> odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac
>>>> >>>>>>>>>> 38:ea:a7:93:83:a0
>>>> >>>>>>>>>> --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8
>>>> >>>>>>>>>> -p 18 -e
>>>> >>>>>>>>>> 5000 -f 5001 -n 8
>>>> >>>>>>>>>>
>>>> >>>>>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.0
>>>> >>>>>>>>>>
>>>> >>>>>>>>>> Thank you,
>>>> >>>>>>>>>> Maxim.
>>>> >>>>>>>>
>>>> >>>>>>>>
>>>> >>>>>>
>>>> >>>>>>
>>>> >>>>
>>>> >>
>>>>
>>>
>>
>

Re: [lng-odp] odp dpdk

[Maxim Uvarov]

t;>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I
>>> see:
>>> >>>>>
>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
>>> >>>>> ODP_BRANCH='api-next'
>>> >>>>>
>>> >>>>>
>>> >>>>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I
>>> see:
>>> >>>>>
>>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
>>> >>>>> ODP_BRANCH='devel/api-next_shsum'
>>> >>>>>
>>> >>>>>
>>> >>>>> or are you referring to other test?
>>> >>>>>
>>> >>>>>
>>> >>>>> On 4 December 2017 at 15:53, Maxim Uvarov 
>>> wrote:
>>> >>>>>>
>>> >>>>>>
>>> >>>>>> On 4 December 2017 at 15:11, Bogdan Pricope <
>>> bogdan.pric...@linaro.org>
>>> >>>>>> wrote:
>>> >>>>>>>
>>> >>>>>>> You need to put 313 on TX side (not RX).
>>> >>>>>>
>>> >>>>>>
>>> >>>>>>
>>> >>>>>> both rx and tx have patches from 313. l2fwd works on recv side.
>>> Generator
>>> >>>>>> does not work.
>>> >>>>>>
>>> >>>>>> Maxim.
>>> >>>>>>
>>> >>>>>>
>>> >>>>>>>
>>> >>>>>>>
>>> >>>>>>> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia - FI/Espoo)
>>> >>>>>>>  wrote:
>>> >>>>>>>> Is the DPDK version 17.08 ? Other versions might not work
>>> properly.
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>> -Petri
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>> From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
>>> >>>>>>>> Sent: Monday, December 04, 2017 1:10 PM
>>> >>>>>>>> To: Savolainen, Petri (Nokia - FI/Espoo) <
>>> petri.savolai...@nokia.com>
>>> >>>>>>>> Cc: Bogdan Pricope ; lng-odp-forward
>>> >>>>>>>> 
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>> Subject: Re: [lng-odp] odp dpdk
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>> 313 does not work also:
>>> >>>>>>>>
>>> >>>>>>>> https://lng.validation.linaro.org/scheduler/job/23242.1
>>> >>>>>>>>
>>> >>>>>>>> I will replace RX side to l2fwd and see that will be there.
>>> >>>>>>>>
>>> >>>>>>>> Maxim.
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>> On 4 December 2017 at 13:46, Savolainen, Petri (Nokia -
>>> FI/Espoo)
>>> >>>>>>>>  wrote:
>>> >>>>>>>>
>>> >>>>>>>> Maxim, try https://github.com/Linaro/odp/pull/313 It has been
>>> tested to
>>> >>>>>>>> fix
>>> >>>>>>>> checksum insert for 10/40GE Intel NICs.
>>> >>>>>>>>
>>> >>>>>>>> -Petri
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>>>> -Original Message-
>>> >>>>>>>>> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On
>>> Behalf Of
>>> >>>>>>>>> Bogdan Pricope
>>> >>>>>>>>> Sent: Monday, December 04, 2017 12:21 PM
>>> >>>>>>>>> To: Maxim Uvarov 
>>> >>>>>>>>> Cc: lng-odp-forward 
>>> >>>>>>>>> Subject: Re: [lng-odp] odp dpdk
>>> >>>>>>>>>
>>> >>>>>>>>> I suspect this is actually caused by csum issue in TX side: on
>>> RX,
>>> >>>>>>>>> socket pktio does not validate csum (and accept the packets)
>>> but on
>>> >>>>>>>>> dpdk pktio the csum is validated and packets are dropped.
>>> >>>>>>>>>
>>> >>>>>>>>> I am not seeing this in my setup because default txq_flags for
>>> igb
>>> >>>>>>>>> driver (1G interface) is
>>> >>>>>>>>>
>>> >>>>>>>>> .txq_flags = 0
>>> >>>>>>>>>
>>> >>>>>>>>> while for ixgbe (10G interface) is:
>>> >>>>>>>>>
>>> >>>>>>>>> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
>>> >>>>>>>>>ETH_TXQ_FLAGS_NOOFFLOADS,
>>> >>>>>>>>>
>>> >>>>>>>>>
>>> >>>>>>>>> /B
>>> >>>>>>>>>
>>> >>>>>>>>>
>>> >>>>>>>>>
>>> >>>>>>>>>
>>> >>>>>>>>> On 1 December 2017 at 23:47, Maxim Uvarov <
>>> maxim.uva...@linaro.org>
>>> >>>>>>>>> wrote:
>>> >>>>>>>>>>
>>> >>>>>>>>>> Looking to dpdk pktio support and generator. It looks like
>>> receive
>>> >>>>>>>>>> part
>>> >>>>>>>>>> is broken. If for receive I use sockets it works well but
>>> receive
>>> >>>>>>>>>> with
>>> >>>>>>>>>> dpdk does not get any packets. For both master and api-next.
>>> Can
>>> >>>>>>>>>> somebody confirm please that it's so. Lava is not supper
>>> friendly to
>>> >>>>>>>>>> debug issue.
>>> >>>>>>>>>>
>>> >>>>>>>>>>
>>> >>>>>>>>>> 1. Recv
>>> >>>>>>>>>> odp_generator -I 0 -m r -c 0x4
>>> >>>>>>>>>>
>>> >>>>>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.1
>>> >>>>>>>>>> Network devices using DPDK-compatible driver
>>> >>>>>>>>>> 
>>> >>>>>>>>>> :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection
>>> 10fb'
>>> >>>>>>>>>> drv=igb_uio unused=
>>> >>>>>>>>>>
>>> >>>>>>>>>>
>>> >>>>>>>>>>
>>> >>>>>>>>>> 2. Send
>>> >>>>>>>>>> odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac
>>> >>>>>>>>>> 38:ea:a7:93:83:a0
>>> >>>>>>>>>> --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8
>>> -p 18 -e
>>> >>>>>>>>>> 5000 -f 5001 -n 8
>>> >>>>>>>>>>
>>> >>>>>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.0
>>> >>>>>>>>>>
>>> >>>>>>>>>> Thank you,
>>> >>>>>>>>>> Maxim.
>>> >>>>>>>>
>>> >>>>>>>>
>>> >>>>>>
>>> >>>>>>
>>> >>>>
>>> >>
>>>
>>>
>>
>

Re: [lng-odp] odp dpdk

[Maxim Uvarov]

skip this message. I will recheck. Pushed to lava wrong branch.

On 6 December 2017 at 10:42, Maxim Uvarov  wrote:

> Ilias was right yesterday. If number of descriptors increased to 1024 then
> TX became again 10M.
>
> +   ret = rte_eth_tx_queue_setup(port_id, i,
> +dev_info.tx_desc_lim.nb_max
> > 1024 ? 1024 : dev_info.tx_desc_lim.nb_max,
>
> rte_eth_dev_socket_id(port_id),
>  txconf);
>
> +   ret = rte_eth_rx_queue_setup(port_id, i,
> +dev_info.rx_desc_lim.nb_max
> > 1024 ? 1024 : dev_info.rx_desc_lim.nb_max,
>
> rte_eth_dev_socket_id(port_id),
>  NULL, pkt_dpdk->pkt_pool);
>
>
>
>
> Maxim.
>
> On 5 December 2017 at 11:20, Elo, Matias (Nokia - FI/Espoo) <
> matias@nokia.com> wrote:
>
>> When I tested enabling HW checksum with Fortville NICs (i40e) the slower
>> driver path alone caused ~20% throughput drop on l2fwd test. This was
>> without actually calculating the checksums, I simply forced the slower
>> driver path (no vectorization).
>>
>> -Matias
>>
>>
>> > On 5 Dec 2017, at 8:59, Bogdan Pricope 
>> wrote:
>> >
>> > On RX side is kind-of expected result since it uses scheduler mode.
>> >
>> > On TX side there is this drop from 10 mpps to 7.69 mpps that is
>> unexpected.
>> >
>> > So Petri, when you said:
>> > "DPDK uses less optimized driver code (on Intel NICs at least) when
>> > any of the L4 checksum offloads is enabled."
>> >
>> > you were referring to this kind of drop in performance?
>> >
>> > There is that 'folklore' that SW csum is faster on small packets while
>> > HW csum is faster on bigger packets. Do you have this kind of data?
>> >
>> > Anyway, for this particular case (odp_generator), since UDP
>> > header/payload is not changing during the test (for now), csum is
>> > calculated only once at the beginning of the test: so we are comparing
>> > HW IPv4 + HW UDP csum vs. SW IPv4 csum yet, the differences in
>> > performance is huge...
>> >
>> >
>> > On 4 December 2017 at 20:37, Maxim Uvarov 
>> wrote:
>> >> I added isocpus and mounted huge page TX became more stable at 7.6M.
>> But
>> >> anyway it's better to test performance for this PR because previous
>> >> speed was 10M.
>> >>
>> >> Maxim.
>> >>
>> >> On 12/04/17 19:42, Honnappa Nagarahalli wrote:
>> >>> Can you run with Linux-DPDK in ODP 2.0?
>> >>>
>> >>> On 4 December 2017 at 09:54, Maxim Uvarov 
>> wrote:
>> >>>> after clean patches apply and fix in run scripts I made it run.
>> >>>>
>> >>>> But results is really bad. --enable-dpdk-zero-copy
>> >>>>
>> >>>> TX rate is:
>> >>>> 7673155 pps
>> >>>>
>> >>>> RX rate is:
>> >>>> 5989846 pps
>> >>>>
>> >>>>
>> >>>> Before patch PR 313 TX was 10M pps.
>> >>>>
>> >>>> I re run task and TX is 3.3M pps. All tests are single core. So
>> >>>> something strange happens in lava or this PR.
>> >>>>
>> >>>> Maxim.
>> >>>>
>> >>>>
>> >>>> On 12/04/17 17:03, Bogdan Pricope wrote:
>> >>>>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I
>> see:
>> >>>>>
>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
>> >>>>> ODP_BRANCH='api-next'
>> >>>>>
>> >>>>>
>> >>>>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I
>> see:
>> >>>>>
>> >>>>> ODP_REPO='https://github.com/muvarov/odp'
>> >>>>> ODP_BRANCH='devel/api-next_shsum'
>> >>>>>
>> >>>>>
>> >>>>> or are you referring to other test?
>> >>>>>
>> >>>>>
>> >>>>> On 4 December 2017 at 15:53, Maxim Uvarov 
>> wrote:
>> >>>>>>
>> >>>>>>
>> >>>>>> On 4 December 2017 at 15:11, Bogdan P

Re: [lng-odp] odp dpdk

[Maxim Uvarov]

Ilias was right yesterday. If number of descriptors increased to 1024 then
TX became again 10M.

+   ret = rte_eth_tx_queue_setup(port_id, i,
+dev_info.tx_desc_lim.nb_max >
1024 ? 1024 : dev_info.tx_desc_lim.nb_max,
 rte_eth_dev_socket_id(port_id),
 txconf);

+   ret = rte_eth_rx_queue_setup(port_id, i,
+dev_info.rx_desc_lim.nb_max >
1024 ? 1024 : dev_info.rx_desc_lim.nb_max,
 rte_eth_dev_socket_id(port_id),
 NULL, pkt_dpdk->pkt_pool);




Maxim.

On 5 December 2017 at 11:20, Elo, Matias (Nokia - FI/Espoo) <
matias@nokia.com> wrote:

> When I tested enabling HW checksum with Fortville NICs (i40e) the slower
> driver path alone caused ~20% throughput drop on l2fwd test. This was
> without actually calculating the checksums, I simply forced the slower
> driver path (no vectorization).
>
> -Matias
>
>
> > On 5 Dec 2017, at 8:59, Bogdan Pricope 
> wrote:
> >
> > On RX side is kind-of expected result since it uses scheduler mode.
> >
> > On TX side there is this drop from 10 mpps to 7.69 mpps that is
> unexpected.
> >
> > So Petri, when you said:
> > "DPDK uses less optimized driver code (on Intel NICs at least) when
> > any of the L4 checksum offloads is enabled."
> >
> > you were referring to this kind of drop in performance?
> >
> > There is that 'folklore' that SW csum is faster on small packets while
> > HW csum is faster on bigger packets. Do you have this kind of data?
> >
> > Anyway, for this particular case (odp_generator), since UDP
> > header/payload is not changing during the test (for now), csum is
> > calculated only once at the beginning of the test: so we are comparing
> > HW IPv4 + HW UDP csum vs. SW IPv4 csum yet, the differences in
> > performance is huge...
> >
> >
> > On 4 December 2017 at 20:37, Maxim Uvarov 
> wrote:
> >> I added isocpus and mounted huge page TX became more stable at 7.6M. But
> >> anyway it's better to test performance for this PR because previous
> >> speed was 10M.
> >>
> >> Maxim.
> >>
> >> On 12/04/17 19:42, Honnappa Nagarahalli wrote:
> >>> Can you run with Linux-DPDK in ODP 2.0?
> >>>
> >>> On 4 December 2017 at 09:54, Maxim Uvarov 
> wrote:
> >>>> after clean patches apply and fix in run scripts I made it run.
> >>>>
> >>>> But results is really bad. --enable-dpdk-zero-copy
> >>>>
> >>>> TX rate is:
> >>>> 7673155 pps
> >>>>
> >>>> RX rate is:
> >>>> 5989846 pps
> >>>>
> >>>>
> >>>> Before patch PR 313 TX was 10M pps.
> >>>>
> >>>> I re run task and TX is 3.3M pps. All tests are single core. So
> >>>> something strange happens in lava or this PR.
> >>>>
> >>>> Maxim.
> >>>>
> >>>>
> >>>> On 12/04/17 17:03, Bogdan Pricope wrote:
> >>>>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I
> see:
> >>>>>
> >>>>> ODP_REPO='https://github.com/muvarov/odp'
> >>>>> ODP_BRANCH='api-next'
> >>>>>
> >>>>>
> >>>>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I
> see:
> >>>>>
> >>>>> ODP_REPO='https://github.com/muvarov/odp'
> >>>>> ODP_BRANCH='devel/api-next_shsum'
> >>>>>
> >>>>>
> >>>>> or are you referring to other test?
> >>>>>
> >>>>>
> >>>>> On 4 December 2017 at 15:53, Maxim Uvarov 
> wrote:
> >>>>>>
> >>>>>>
> >>>>>> On 4 December 2017 at 15:11, Bogdan Pricope <
> bogdan.pric...@linaro.org>
> >>>>>> wrote:
> >>>>>>>
> >>>>>>> You need to put 313 on TX side (not RX).
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>> both rx and tx have patches from 313. l2fwd works on recv side.
> Generator
> >>>>>> does not work.
> >>>>>>
> >>&

Re: [lng-odp] odp dpdk

[Elo, Matias (Nokia - FI/Espoo)]

When I tested enabling HW checksum with Fortville NICs (i40e) the slower driver 
path alone caused ~20% throughput drop on l2fwd test. This was without actually 
calculating the checksums, I simply forced the slower driver path (no 
vectorization).

-Matias


> On 5 Dec 2017, at 8:59, Bogdan Pricope  wrote:
> 
> On RX side is kind-of expected result since it uses scheduler mode.
> 
> On TX side there is this drop from 10 mpps to 7.69 mpps that is unexpected.
> 
> So Petri, when you said:
> "DPDK uses less optimized driver code (on Intel NICs at least) when
> any of the L4 checksum offloads is enabled."
> 
> you were referring to this kind of drop in performance?
> 
> There is that 'folklore' that SW csum is faster on small packets while
> HW csum is faster on bigger packets. Do you have this kind of data?
> 
> Anyway, for this particular case (odp_generator), since UDP
> header/payload is not changing during the test (for now), csum is
> calculated only once at the beginning of the test: so we are comparing
> HW IPv4 + HW UDP csum vs. SW IPv4 csum yet, the differences in
> performance is huge...
> 
> 
> On 4 December 2017 at 20:37, Maxim Uvarov  wrote:
>> I added isocpus and mounted huge page TX became more stable at 7.6M. But
>> anyway it's better to test performance for this PR because previous
>> speed was 10M.
>> 
>> Maxim.
>> 
>> On 12/04/17 19:42, Honnappa Nagarahalli wrote:
>>> Can you run with Linux-DPDK in ODP 2.0?
>>> 
>>> On 4 December 2017 at 09:54, Maxim Uvarov  wrote:
>>>> after clean patches apply and fix in run scripts I made it run.
>>>> 
>>>> But results is really bad. --enable-dpdk-zero-copy
>>>> 
>>>> TX rate is:
>>>> 7673155 pps
>>>> 
>>>> RX rate is:
>>>> 5989846 pps
>>>> 
>>>> 
>>>> Before patch PR 313 TX was 10M pps.
>>>> 
>>>> I re run task and TX is 3.3M pps. All tests are single core. So
>>>> something strange happens in lava or this PR.
>>>> 
>>>> Maxim.
>>>> 
>>>> 
>>>> On 12/04/17 17:03, Bogdan Pricope wrote:
>>>>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I see:
>>>>> 
>>>>> ODP_REPO='https://github.com/muvarov/odp'
>>>>> ODP_BRANCH='api-next'
>>>>> 
>>>>> 
>>>>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I see:
>>>>> 
>>>>> ODP_REPO='https://github.com/muvarov/odp'
>>>>> ODP_BRANCH='devel/api-next_shsum'
>>>>> 
>>>>> 
>>>>> or are you referring to other test?
>>>>> 
>>>>> 
>>>>> On 4 December 2017 at 15:53, Maxim Uvarov  wrote:
>>>>>> 
>>>>>> 
>>>>>> On 4 December 2017 at 15:11, Bogdan Pricope 
>>>>>> wrote:
>>>>>>> 
>>>>>>> You need to put 313 on TX side (not RX).
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> both rx and tx have patches from 313. l2fwd works on recv side. Generator
>>>>>> does not work.
>>>>>> 
>>>>>> Maxim.
>>>>>> 
>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia - FI/Espoo)
>>>>>>>  wrote:
>>>>>>>> Is the DPDK version 17.08 ? Other versions might not work properly.
>>>>>>>> 
>>>>>>>> 
>>>>>>>> 
>>>>>>>> -Petri
>>>>>>>> 
>>>>>>>> 
>>>>>>>> 
>>>>>>>> From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
>>>>>>>> Sent: Monday, December 04, 2017 1:10 PM
>>>>>>>> To: Savolainen, Petri (Nokia - FI/Espoo) 
>>>>>>>> Cc: Bogdan Pricope ; lng-odp-forward
>>>>>>>> 
>>>>>>>> 
>>>>>>>> 
>>>>>>>> Subject: Re: [lng-odp] odp dpdk
>>>>>>>> 
>>>>>>>> 
>>>>>>>> 
>>>>>>>> 313 does not work also:
>>>>>>>> 
>>>>>>>> https://lng.validation.linaro.org/schedu

Re: [lng-odp] odp dpdk

[Bogdan Pricope]

On RX side is kind-of expected result since it uses scheduler mode.

On TX side there is this drop from 10 mpps to 7.69 mpps that is unexpected.

So Petri, when you said:
"DPDK uses less optimized driver code (on Intel NICs at least) when
any of the L4 checksum offloads is enabled."

you were referring to this kind of drop in performance?

There is that 'folklore' that SW csum is faster on small packets while
HW csum is faster on bigger packets. Do you have this kind of data?

Anyway, for this particular case (odp_generator), since UDP
header/payload is not changing during the test (for now), csum is
calculated only once at the beginning of the test: so we are comparing
HW IPv4 + HW UDP csum vs. SW IPv4 csum yet, the differences in
performance is huge...


On 4 December 2017 at 20:37, Maxim Uvarov  wrote:
> I added isocpus and mounted huge page TX became more stable at 7.6M. But
> anyway it's better to test performance for this PR because previous
> speed was 10M.
>
> Maxim.
>
> On 12/04/17 19:42, Honnappa Nagarahalli wrote:
>> Can you run with Linux-DPDK in ODP 2.0?
>>
>> On 4 December 2017 at 09:54, Maxim Uvarov  wrote:
>>> after clean patches apply and fix in run scripts I made it run.
>>>
>>> But results is really bad. --enable-dpdk-zero-copy
>>>
>>> TX rate is:
>>> 7673155 pps
>>>
>>> RX rate is:
>>> 5989846 pps
>>>
>>>
>>> Before patch PR 313 TX was 10M pps.
>>>
>>> I re run task and TX is 3.3M pps. All tests are single core. So
>>> something strange happens in lava or this PR.
>>>
>>> Maxim.
>>>
>>>
>>> On 12/04/17 17:03, Bogdan Pricope wrote:
>>>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I see:
>>>>
>>>> ODP_REPO='https://github.com/muvarov/odp'
>>>> ODP_BRANCH='api-next'
>>>>
>>>>
>>>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I see:
>>>>
>>>> ODP_REPO='https://github.com/muvarov/odp'
>>>> ODP_BRANCH='devel/api-next_shsum'
>>>>
>>>>
>>>> or are you referring to other test?
>>>>
>>>>
>>>> On 4 December 2017 at 15:53, Maxim Uvarov  wrote:
>>>>>
>>>>>
>>>>> On 4 December 2017 at 15:11, Bogdan Pricope 
>>>>> wrote:
>>>>>>
>>>>>> You need to put 313 on TX side (not RX).
>>>>>
>>>>>
>>>>>
>>>>> both rx and tx have patches from 313. l2fwd works on recv side. Generator
>>>>> does not work.
>>>>>
>>>>> Maxim.
>>>>>
>>>>>
>>>>>>
>>>>>>
>>>>>> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia - FI/Espoo)
>>>>>>  wrote:
>>>>>>> Is the DPDK version 17.08 ? Other versions might not work properly.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> -Petri
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
>>>>>>> Sent: Monday, December 04, 2017 1:10 PM
>>>>>>> To: Savolainen, Petri (Nokia - FI/Espoo) 
>>>>>>> Cc: Bogdan Pricope ; lng-odp-forward
>>>>>>> 
>>>>>>>
>>>>>>>
>>>>>>> Subject: Re: [lng-odp] odp dpdk
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> 313 does not work also:
>>>>>>>
>>>>>>> https://lng.validation.linaro.org/scheduler/job/23242.1
>>>>>>>
>>>>>>> I will replace RX side to l2fwd and see that will be there.
>>>>>>>
>>>>>>> Maxim.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On 4 December 2017 at 13:46, Savolainen, Petri (Nokia - FI/Espoo)
>>>>>>>  wrote:
>>>>>>>
>>>>>>> Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to
>>>>>>> fix
>>>>>>> checksum insert for 10/40GE Intel NICs.
>>>>>>>
>>>>>>> -Petri
>>>>>>>
>>>>>&g

Re: [lng-odp] odp dpdk

[Maxim Uvarov]

I added isocpus and mounted huge page TX became more stable at 7.6M. But
anyway it's better to test performance for this PR because previous
speed was 10M.

Maxim.

On 12/04/17 19:42, Honnappa Nagarahalli wrote:
> Can you run with Linux-DPDK in ODP 2.0?
> 
> On 4 December 2017 at 09:54, Maxim Uvarov  wrote:
>> after clean patches apply and fix in run scripts I made it run.
>>
>> But results is really bad. --enable-dpdk-zero-copy
>>
>> TX rate is:
>> 7673155 pps
>>
>> RX rate is:
>> 5989846 pps
>>
>>
>> Before patch PR 313 TX was 10M pps.
>>
>> I re run task and TX is 3.3M pps. All tests are single core. So
>> something strange happens in lava or this PR.
>>
>> Maxim.
>>
>>
>> On 12/04/17 17:03, Bogdan Pricope wrote:
>>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I see:
>>>
>>> ODP_REPO='https://github.com/muvarov/odp'
>>> ODP_BRANCH='api-next'
>>>
>>>
>>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I see:
>>>
>>> ODP_REPO='https://github.com/muvarov/odp'
>>> ODP_BRANCH='devel/api-next_shsum'
>>>
>>>
>>> or are you referring to other test?
>>>
>>>
>>> On 4 December 2017 at 15:53, Maxim Uvarov  wrote:
>>>>
>>>>
>>>> On 4 December 2017 at 15:11, Bogdan Pricope 
>>>> wrote:
>>>>>
>>>>> You need to put 313 on TX side (not RX).
>>>>
>>>>
>>>>
>>>> both rx and tx have patches from 313. l2fwd works on recv side. Generator
>>>> does not work.
>>>>
>>>> Maxim.
>>>>
>>>>
>>>>>
>>>>>
>>>>> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia - FI/Espoo)
>>>>>  wrote:
>>>>>> Is the DPDK version 17.08 ? Other versions might not work properly.
>>>>>>
>>>>>>
>>>>>>
>>>>>> -Petri
>>>>>>
>>>>>>
>>>>>>
>>>>>> From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
>>>>>> Sent: Monday, December 04, 2017 1:10 PM
>>>>>> To: Savolainen, Petri (Nokia - FI/Espoo) 
>>>>>> Cc: Bogdan Pricope ; lng-odp-forward
>>>>>> 
>>>>>>
>>>>>>
>>>>>> Subject: Re: [lng-odp] odp dpdk
>>>>>>
>>>>>>
>>>>>>
>>>>>> 313 does not work also:
>>>>>>
>>>>>> https://lng.validation.linaro.org/scheduler/job/23242.1
>>>>>>
>>>>>> I will replace RX side to l2fwd and see that will be there.
>>>>>>
>>>>>> Maxim.
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On 4 December 2017 at 13:46, Savolainen, Petri (Nokia - FI/Espoo)
>>>>>>  wrote:
>>>>>>
>>>>>> Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to
>>>>>> fix
>>>>>> checksum insert for 10/40GE Intel NICs.
>>>>>>
>>>>>> -Petri
>>>>>>
>>>>>>
>>>>>>> -Original Message-
>>>>>>> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
>>>>>>> Bogdan Pricope
>>>>>>> Sent: Monday, December 04, 2017 12:21 PM
>>>>>>> To: Maxim Uvarov 
>>>>>>> Cc: lng-odp-forward 
>>>>>>> Subject: Re: [lng-odp] odp dpdk
>>>>>>>
>>>>>>> I suspect this is actually caused by csum issue in TX side: on RX,
>>>>>>> socket pktio does not validate csum (and accept the packets) but on
>>>>>>> dpdk pktio the csum is validated and packets are dropped.
>>>>>>>
>>>>>>> I am not seeing this in my setup because default txq_flags for igb
>>>>>>> driver (1G interface) is
>>>>>>>
>>>>>>> .txq_flags = 0
>>>>>>>
>>>>>>> while for ixgbe (10G interface) is:
>>>>>>>
>>>>>>> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
>>>>>>> ETH_TXQ_FLAGS_NOOFFLOADS,
>>>>>>>
>>>>>>>
>>>>>>> /B
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On 1 December 2017 at 23:47, Maxim Uvarov 
>>>>>>> wrote:
>>>>>>>>
>>>>>>>> Looking to dpdk pktio support and generator. It looks like receive
>>>>>>>> part
>>>>>>>> is broken. If for receive I use sockets it works well but receive
>>>>>>>> with
>>>>>>>> dpdk does not get any packets. For both master and api-next. Can
>>>>>>>> somebody confirm please that it's so. Lava is not supper friendly to
>>>>>>>> debug issue.
>>>>>>>>
>>>>>>>>
>>>>>>>> 1. Recv
>>>>>>>> odp_generator -I 0 -m r -c 0x4
>>>>>>>>
>>>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.1
>>>>>>>> Network devices using DPDK-compatible driver
>>>>>>>> 
>>>>>>>> :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
>>>>>>>> drv=igb_uio unused=
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> 2. Send
>>>>>>>> odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac
>>>>>>>> 38:ea:a7:93:83:a0
>>>>>>>> --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
>>>>>>>> 5000 -f 5001 -n 8
>>>>>>>>
>>>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.0
>>>>>>>>
>>>>>>>> Thank you,
>>>>>>>> Maxim.
>>>>>>
>>>>>>
>>>>
>>>>
>>

Re: [lng-odp] odp dpdk

[Honnappa Nagarahalli]

Can you run with Linux-DPDK in ODP 2.0?

On 4 December 2017 at 09:54, Maxim Uvarov  wrote:
> after clean patches apply and fix in run scripts I made it run.
>
> But results is really bad. --enable-dpdk-zero-copy
>
> TX rate is:
> 7673155 pps
>
> RX rate is:
> 5989846 pps
>
>
> Before patch PR 313 TX was 10M pps.
>
> I re run task and TX is 3.3M pps. All tests are single core. So
> something strange happens in lava or this PR.
>
> Maxim.
>
>
> On 12/04/17 17:03, Bogdan Pricope wrote:
>> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I see:
>>
>> ODP_REPO='https://github.com/muvarov/odp'
>> ODP_BRANCH='api-next'
>>
>>
>> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I see:
>>
>> ODP_REPO='https://github.com/muvarov/odp'
>> ODP_BRANCH='devel/api-next_shsum'
>>
>>
>> or are you referring to other test?
>>
>>
>> On 4 December 2017 at 15:53, Maxim Uvarov  wrote:
>>>
>>>
>>> On 4 December 2017 at 15:11, Bogdan Pricope 
>>> wrote:
>>>>
>>>> You need to put 313 on TX side (not RX).
>>>
>>>
>>>
>>> both rx and tx have patches from 313. l2fwd works on recv side. Generator
>>> does not work.
>>>
>>> Maxim.
>>>
>>>
>>>>
>>>>
>>>> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia - FI/Espoo)
>>>>  wrote:
>>>>> Is the DPDK version 17.08 ? Other versions might not work properly.
>>>>>
>>>>>
>>>>>
>>>>> -Petri
>>>>>
>>>>>
>>>>>
>>>>> From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
>>>>> Sent: Monday, December 04, 2017 1:10 PM
>>>>> To: Savolainen, Petri (Nokia - FI/Espoo) 
>>>>> Cc: Bogdan Pricope ; lng-odp-forward
>>>>> 
>>>>>
>>>>>
>>>>> Subject: Re: [lng-odp] odp dpdk
>>>>>
>>>>>
>>>>>
>>>>> 313 does not work also:
>>>>>
>>>>> https://lng.validation.linaro.org/scheduler/job/23242.1
>>>>>
>>>>> I will replace RX side to l2fwd and see that will be there.
>>>>>
>>>>> Maxim.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 4 December 2017 at 13:46, Savolainen, Petri (Nokia - FI/Espoo)
>>>>>  wrote:
>>>>>
>>>>> Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to
>>>>> fix
>>>>> checksum insert for 10/40GE Intel NICs.
>>>>>
>>>>> -Petri
>>>>>
>>>>>
>>>>>> -Original Message-
>>>>>> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
>>>>>> Bogdan Pricope
>>>>>> Sent: Monday, December 04, 2017 12:21 PM
>>>>>> To: Maxim Uvarov 
>>>>>> Cc: lng-odp-forward 
>>>>>> Subject: Re: [lng-odp] odp dpdk
>>>>>>
>>>>>> I suspect this is actually caused by csum issue in TX side: on RX,
>>>>>> socket pktio does not validate csum (and accept the packets) but on
>>>>>> dpdk pktio the csum is validated and packets are dropped.
>>>>>>
>>>>>> I am not seeing this in my setup because default txq_flags for igb
>>>>>> driver (1G interface) is
>>>>>>
>>>>>> .txq_flags = 0
>>>>>>
>>>>>> while for ixgbe (10G interface) is:
>>>>>>
>>>>>> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
>>>>>> ETH_TXQ_FLAGS_NOOFFLOADS,
>>>>>>
>>>>>>
>>>>>> /B
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On 1 December 2017 at 23:47, Maxim Uvarov 
>>>>>> wrote:
>>>>>>>
>>>>>>> Looking to dpdk pktio support and generator. It looks like receive
>>>>>>> part
>>>>>>> is broken. If for receive I use sockets it works well but receive
>>>>>>> with
>>>>>>> dpdk does not get any packets. For both master and api-next. Can
>>>>>>> somebody confirm please that it's so. Lava is not supper friendly to
>>>>>>> debug issue.
>>>>>>>
>>>>>>>
>>>>>>> 1. Recv
>>>>>>> odp_generator -I 0 -m r -c 0x4
>>>>>>>
>>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.1
>>>>>>> Network devices using DPDK-compatible driver
>>>>>>> 
>>>>>>> :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
>>>>>>> drv=igb_uio unused=
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> 2. Send
>>>>>>> odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac
>>>>>>> 38:ea:a7:93:83:a0
>>>>>>> --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
>>>>>>> 5000 -f 5001 -n 8
>>>>>>>
>>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.0
>>>>>>>
>>>>>>> Thank you,
>>>>>>> Maxim.
>>>>>
>>>>>
>>>
>>>
>

Re: [lng-odp] odp dpdk

[Maxim Uvarov]

after clean patches apply and fix in run scripts I made it run.

But results is really bad. --enable-dpdk-zero-copy

TX rate is:
7673155 pps

RX rate is:
5989846 pps


Before patch PR 313 TX was 10M pps.

I re run task and TX is 3.3M pps. All tests are single core. So
something strange happens in lava or this PR.

Maxim.


On 12/04/17 17:03, Bogdan Pricope wrote:
> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I see:
> 
> ODP_REPO='https://github.com/muvarov/odp'
> ODP_BRANCH='api-next'
> 
> 
> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I see:
> 
> ODP_REPO='https://github.com/muvarov/odp'
> ODP_BRANCH='devel/api-next_shsum'
> 
> 
> or are you referring to other test?
> 
> 
> On 4 December 2017 at 15:53, Maxim Uvarov  wrote:
>>
>>
>> On 4 December 2017 at 15:11, Bogdan Pricope 
>> wrote:
>>>
>>> You need to put 313 on TX side (not RX).
>>
>>
>>
>> both rx and tx have patches from 313. l2fwd works on recv side. Generator
>> does not work.
>>
>> Maxim.
>>
>>
>>>
>>>
>>> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia - FI/Espoo)
>>>  wrote:
>>>> Is the DPDK version 17.08 ? Other versions might not work properly.
>>>>
>>>>
>>>>
>>>> -Petri
>>>>
>>>>
>>>>
>>>> From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
>>>> Sent: Monday, December 04, 2017 1:10 PM
>>>> To: Savolainen, Petri (Nokia - FI/Espoo) 
>>>> Cc: Bogdan Pricope ; lng-odp-forward
>>>> 
>>>>
>>>>
>>>> Subject: Re: [lng-odp] odp dpdk
>>>>
>>>>
>>>>
>>>> 313 does not work also:
>>>>
>>>> https://lng.validation.linaro.org/scheduler/job/23242.1
>>>>
>>>> I will replace RX side to l2fwd and see that will be there.
>>>>
>>>> Maxim.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On 4 December 2017 at 13:46, Savolainen, Petri (Nokia - FI/Espoo)
>>>>  wrote:
>>>>
>>>> Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to
>>>> fix
>>>> checksum insert for 10/40GE Intel NICs.
>>>>
>>>> -Petri
>>>>
>>>>
>>>>> -Original Message-
>>>>> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
>>>>> Bogdan Pricope
>>>>> Sent: Monday, December 04, 2017 12:21 PM
>>>>> To: Maxim Uvarov 
>>>>> Cc: lng-odp-forward 
>>>>> Subject: Re: [lng-odp] odp dpdk
>>>>>
>>>>> I suspect this is actually caused by csum issue in TX side: on RX,
>>>>> socket pktio does not validate csum (and accept the packets) but on
>>>>> dpdk pktio the csum is validated and packets are dropped.
>>>>>
>>>>> I am not seeing this in my setup because default txq_flags for igb
>>>>> driver (1G interface) is
>>>>>
>>>>> .txq_flags = 0
>>>>>
>>>>> while for ixgbe (10G interface) is:
>>>>>
>>>>> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
>>>>> ETH_TXQ_FLAGS_NOOFFLOADS,
>>>>>
>>>>>
>>>>> /B
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 1 December 2017 at 23:47, Maxim Uvarov 
>>>>> wrote:
>>>>>>
>>>>>> Looking to dpdk pktio support and generator. It looks like receive
>>>>>> part
>>>>>> is broken. If for receive I use sockets it works well but receive
>>>>>> with
>>>>>> dpdk does not get any packets. For both master and api-next. Can
>>>>>> somebody confirm please that it's so. Lava is not supper friendly to
>>>>>> debug issue.
>>>>>>
>>>>>>
>>>>>> 1. Recv
>>>>>> odp_generator -I 0 -m r -c 0x4
>>>>>>
>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.1
>>>>>> Network devices using DPDK-compatible driver
>>>>>> 
>>>>>> :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
>>>>>> drv=igb_uio unused=
>>>>>>
>>>>>>
>>>>>>
>>>>>> 2. Send
>>>>>> odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac
>>>>>> 38:ea:a7:93:83:a0
>>>>>> --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
>>>>>> 5000 -f 5001 -n 8
>>>>>>
>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.0
>>>>>>
>>>>>> Thank you,
>>>>>> Maxim.
>>>>
>>>>
>>
>>

Re: [lng-odp] odp dpdk

[Maxim Uvarov]

On 12/04/17 17:03, Bogdan Pricope wrote:
> On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I see:
> 
> ODP_REPO='https://github.com/muvarov/odp'
> ODP_BRANCH='api-next'
> 
> 
> On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I see:
> 
> ODP_REPO='https://github.com/muvarov/odp'
> ODP_BRANCH='devel/api-next_shsum'
> 
> 
> or are you referring to other test?
> 

here I used l2fwd which works:
https://lng.validation.linaro.org/scheduler/job/23246.1

now I think that I did not apply patches cleanly to I re-applied them
and testing now:
https://lng.validation.linaro.org/scheduler/job/23255.1

If not I will comment out bad check summ drop in generator and do next
debug...

Maxim.


> 
> On 4 December 2017 at 15:53, Maxim Uvarov  wrote:
>>
>>
>> On 4 December 2017 at 15:11, Bogdan Pricope 
>> wrote:
>>>
>>> You need to put 313 on TX side (not RX).
>>
>>
>>
>> both rx and tx have patches from 313. l2fwd works on recv side. Generator
>> does not work.
>>
>> Maxim.
>>
>>
>>>
>>>
>>> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia - FI/Espoo)
>>>  wrote:
>>>> Is the DPDK version 17.08 ? Other versions might not work properly.
>>>>
>>>>
>>>>
>>>> -Petri
>>>>
>>>>
>>>>
>>>> From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
>>>> Sent: Monday, December 04, 2017 1:10 PM
>>>> To: Savolainen, Petri (Nokia - FI/Espoo) 
>>>> Cc: Bogdan Pricope ; lng-odp-forward
>>>> 
>>>>
>>>>
>>>> Subject: Re: [lng-odp] odp dpdk
>>>>
>>>>
>>>>
>>>> 313 does not work also:
>>>>
>>>> https://lng.validation.linaro.org/scheduler/job/23242.1
>>>>
>>>> I will replace RX side to l2fwd and see that will be there.
>>>>
>>>> Maxim.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On 4 December 2017 at 13:46, Savolainen, Petri (Nokia - FI/Espoo)
>>>>  wrote:
>>>>
>>>> Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to
>>>> fix
>>>> checksum insert for 10/40GE Intel NICs.
>>>>
>>>> -Petri
>>>>
>>>>
>>>>> -Original Message-
>>>>> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
>>>>> Bogdan Pricope
>>>>> Sent: Monday, December 04, 2017 12:21 PM
>>>>> To: Maxim Uvarov 
>>>>> Cc: lng-odp-forward 
>>>>> Subject: Re: [lng-odp] odp dpdk
>>>>>
>>>>> I suspect this is actually caused by csum issue in TX side: on RX,
>>>>> socket pktio does not validate csum (and accept the packets) but on
>>>>> dpdk pktio the csum is validated and packets are dropped.
>>>>>
>>>>> I am not seeing this in my setup because default txq_flags for igb
>>>>> driver (1G interface) is
>>>>>
>>>>> .txq_flags = 0
>>>>>
>>>>> while for ixgbe (10G interface) is:
>>>>>
>>>>> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
>>>>> ETH_TXQ_FLAGS_NOOFFLOADS,
>>>>>
>>>>>
>>>>> /B
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 1 December 2017 at 23:47, Maxim Uvarov 
>>>>> wrote:
>>>>>>
>>>>>> Looking to dpdk pktio support and generator. It looks like receive
>>>>>> part
>>>>>> is broken. If for receive I use sockets it works well but receive
>>>>>> with
>>>>>> dpdk does not get any packets. For both master and api-next. Can
>>>>>> somebody confirm please that it's so. Lava is not supper friendly to
>>>>>> debug issue.
>>>>>>
>>>>>>
>>>>>> 1. Recv
>>>>>> odp_generator -I 0 -m r -c 0x4
>>>>>>
>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.1
>>>>>> Network devices using DPDK-compatible driver
>>>>>> 
>>>>>> :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
>>>>>> drv=igb_uio unused=
>>>>>>
>>>>>>
>>>>>>
>>>>>> 2. Send
>>>>>> odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac
>>>>>> 38:ea:a7:93:83:a0
>>>>>> --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
>>>>>> 5000 -f 5001 -n 8
>>>>>>
>>>>>> https://lng.validation.linaro.org/scheduler/job/23206.0
>>>>>>
>>>>>> Thank you,
>>>>>> Maxim.
>>>>
>>>>
>>
>>

Re: [lng-odp] odp dpdk

[Bogdan Pricope]

On TX (https://lng.validation.linaro.org/scheduler/job/23252.0) I see:

ODP_REPO='https://github.com/muvarov/odp'
ODP_BRANCH='api-next'


On RX (https://lng.validation.linaro.org/scheduler/job/23252.1) I see:

ODP_REPO='https://github.com/muvarov/odp'
ODP_BRANCH='devel/api-next_shsum'


or are you referring to other test?


On 4 December 2017 at 15:53, Maxim Uvarov  wrote:
>
>
> On 4 December 2017 at 15:11, Bogdan Pricope 
> wrote:
>>
>> You need to put 313 on TX side (not RX).
>
>
>
> both rx and tx have patches from 313. l2fwd works on recv side. Generator
> does not work.
>
> Maxim.
>
>
>>
>>
>> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia - FI/Espoo)
>>  wrote:
>> > Is the DPDK version 17.08 ? Other versions might not work properly.
>> >
>> >
>> >
>> > -Petri
>> >
>> >
>> >
>> > From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
>> > Sent: Monday, December 04, 2017 1:10 PM
>> > To: Savolainen, Petri (Nokia - FI/Espoo) 
>> > Cc: Bogdan Pricope ; lng-odp-forward
>> > 
>> >
>> >
>> > Subject: Re: [lng-odp] odp dpdk
>> >
>> >
>> >
>> > 313 does not work also:
>> >
>> > https://lng.validation.linaro.org/scheduler/job/23242.1
>> >
>> > I will replace RX side to l2fwd and see that will be there.
>> >
>> > Maxim.
>> >
>> >
>> >
>> >
>> >
>> > On 4 December 2017 at 13:46, Savolainen, Petri (Nokia - FI/Espoo)
>> >  wrote:
>> >
>> > Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to
>> > fix
>> > checksum insert for 10/40GE Intel NICs.
>> >
>> > -Petri
>> >
>> >
>> >> -Original Message-
>> >> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
>> >> Bogdan Pricope
>> >> Sent: Monday, December 04, 2017 12:21 PM
>> >> To: Maxim Uvarov 
>> >> Cc: lng-odp-forward 
>> >> Subject: Re: [lng-odp] odp dpdk
>> >>
>> >> I suspect this is actually caused by csum issue in TX side: on RX,
>> >> socket pktio does not validate csum (and accept the packets) but on
>> >> dpdk pktio the csum is validated and packets are dropped.
>> >>
>> >> I am not seeing this in my setup because default txq_flags for igb
>> >> driver (1G interface) is
>> >>
>> >> .txq_flags = 0
>> >>
>> >> while for ixgbe (10G interface) is:
>> >>
>> >> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
>> >> ETH_TXQ_FLAGS_NOOFFLOADS,
>> >>
>> >>
>> >> /B
>> >>
>> >>
>> >>
>> >>
>> >> On 1 December 2017 at 23:47, Maxim Uvarov 
>> >> wrote:
>> >> >
>> >> > Looking to dpdk pktio support and generator. It looks like receive
>> >> > part
>> >> > is broken. If for receive I use sockets it works well but receive
>> >> > with
>> >> > dpdk does not get any packets. For both master and api-next. Can
>> >> > somebody confirm please that it's so. Lava is not supper friendly to
>> >> > debug issue.
>> >> >
>> >> >
>> >> > 1. Recv
>> >> > odp_generator -I 0 -m r -c 0x4
>> >> >
>> >> > https://lng.validation.linaro.org/scheduler/job/23206.1
>> >> > Network devices using DPDK-compatible driver
>> >> > 
>> >> > :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
>> >> > drv=igb_uio unused=
>> >> >
>> >> >
>> >> >
>> >> > 2. Send
>> >> > odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac
>> >> > 38:ea:a7:93:83:a0
>> >> > --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
>> >> > 5000 -f 5001 -n 8
>> >> >
>> >> > https://lng.validation.linaro.org/scheduler/job/23206.0
>> >> >
>> >> > Thank you,
>> >> > Maxim.
>> >
>> >
>
>

Re: [lng-odp] odp dpdk

[Maxim Uvarov]

On 4 December 2017 at 15:11, Bogdan Pricope 
wrote:

> You need to put 313 on TX side (not RX).
>


both rx and tx have patches from 313. l2fwd works on recv side. Generator
does not work.

Maxim.



>
> On 4 December 2017 at 13:19, Savolainen, Petri (Nokia - FI/Espoo)
>  wrote:
> > Is the DPDK version 17.08 ? Other versions might not work properly.
> >
> >
> >
> > -Petri
> >
> >
> >
> > From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
> > Sent: Monday, December 04, 2017 1:10 PM
> > To: Savolainen, Petri (Nokia - FI/Espoo) 
> > Cc: Bogdan Pricope ; lng-odp-forward
> > 
> >
> >
> > Subject: Re: [lng-odp] odp dpdk
> >
> >
> >
> > 313 does not work also:
> >
> > https://lng.validation.linaro.org/scheduler/job/23242.1
> >
> > I will replace RX side to l2fwd and see that will be there.
> >
> > Maxim.
> >
> >
> >
> >
> >
> > On 4 December 2017 at 13:46, Savolainen, Petri (Nokia - FI/Espoo)
> >  wrote:
> >
> > Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to
> fix
> > checksum insert for 10/40GE Intel NICs.
> >
> > -Petri
> >
> >
> >> -Original Message-
> >> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
> >> Bogdan Pricope
> >> Sent: Monday, December 04, 2017 12:21 PM
> >> To: Maxim Uvarov 
> >> Cc: lng-odp-forward 
> >> Subject: Re: [lng-odp] odp dpdk
> >>
> >> I suspect this is actually caused by csum issue in TX side: on RX,
> >> socket pktio does not validate csum (and accept the packets) but on
> >> dpdk pktio the csum is validated and packets are dropped.
> >>
> >> I am not seeing this in my setup because default txq_flags for igb
> >> driver (1G interface) is
> >>
> >> .txq_flags = 0
> >>
> >> while for ixgbe (10G interface) is:
> >>
> >> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
> >> ETH_TXQ_FLAGS_NOOFFLOADS,
> >>
> >>
> >> /B
> >>
> >>
> >>
> >>
> >> On 1 December 2017 at 23:47, Maxim Uvarov 
> wrote:
> >> >
> >> > Looking to dpdk pktio support and generator. It looks like receive
> part
> >> > is broken. If for receive I use sockets it works well but receive with
> >> > dpdk does not get any packets. For both master and api-next. Can
> >> > somebody confirm please that it's so. Lava is not supper friendly to
> >> > debug issue.
> >> >
> >> >
> >> > 1. Recv
> >> > odp_generator -I 0 -m r -c 0x4
> >> >
> >> > https://lng.validation.linaro.org/scheduler/job/23206.1
> >> > Network devices using DPDK-compatible driver
> >> > 
> >> > :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
> >> > drv=igb_uio unused=
> >> >
> >> >
> >> >
> >> > 2. Send
> >> > odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac
> 38:ea:a7:93:83:a0
> >> > --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
> >> > 5000 -f 5001 -n 8
> >> >
> >> > https://lng.validation.linaro.org/scheduler/job/23206.0
> >> >
> >> > Thank you,
> >> > Maxim.
> >
> >
>

Re: [lng-odp] odp dpdk

[Bogdan Pricope]

You need to put 313 on TX side (not RX).

On 4 December 2017 at 13:19, Savolainen, Petri (Nokia - FI/Espoo)
 wrote:
> Is the DPDK version 17.08 ? Other versions might not work properly.
>
>
>
> -Petri
>
>
>
> From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
> Sent: Monday, December 04, 2017 1:10 PM
> To: Savolainen, Petri (Nokia - FI/Espoo) 
> Cc: Bogdan Pricope ; lng-odp-forward
> 
>
>
> Subject: Re: [lng-odp] odp dpdk
>
>
>
> 313 does not work also:
>
> https://lng.validation.linaro.org/scheduler/job/23242.1
>
> I will replace RX side to l2fwd and see that will be there.
>
> Maxim.
>
>
>
>
>
> On 4 December 2017 at 13:46, Savolainen, Petri (Nokia - FI/Espoo)
>  wrote:
>
> Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to fix
> checksum insert for 10/40GE Intel NICs.
>
> -Petri
>
>
>> -Original Message-
>> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
>> Bogdan Pricope
>> Sent: Monday, December 04, 2017 12:21 PM
>> To: Maxim Uvarov 
>> Cc: lng-odp-forward 
>> Subject: Re: [lng-odp] odp dpdk
>>
>> I suspect this is actually caused by csum issue in TX side: on RX,
>> socket pktio does not validate csum (and accept the packets) but on
>> dpdk pktio the csum is validated and packets are dropped.
>>
>> I am not seeing this in my setup because default txq_flags for igb
>> driver (1G interface) is
>>
>> .txq_flags = 0
>>
>> while for ixgbe (10G interface) is:
>>
>> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
>> ETH_TXQ_FLAGS_NOOFFLOADS,
>>
>>
>> /B
>>
>>
>>
>>
>> On 1 December 2017 at 23:47, Maxim Uvarov  wrote:
>> >
>> > Looking to dpdk pktio support and generator. It looks like receive part
>> > is broken. If for receive I use sockets it works well but receive with
>> > dpdk does not get any packets. For both master and api-next. Can
>> > somebody confirm please that it's so. Lava is not supper friendly to
>> > debug issue.
>> >
>> >
>> > 1. Recv
>> > odp_generator -I 0 -m r -c 0x4
>> >
>> > https://lng.validation.linaro.org/scheduler/job/23206.1
>> > Network devices using DPDK-compatible driver
>> > 
>> > :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
>> > drv=igb_uio unused=
>> >
>> >
>> >
>> > 2. Send
>> > odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac 38:ea:a7:93:83:a0
>> > --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
>> > 5000 -f 5001 -n 8
>> >
>> > https://lng.validation.linaro.org/scheduler/job/23206.0
>> >
>> > Thank you,
>> > Maxim.
>
>

Re: [lng-odp] odp dpdk

[Savolainen, Petri (Nokia - FI/Espoo)]

Is the DPDK version 17.08 ? Other versions might not work properly.

-Petri

From: Maxim Uvarov [mailto:maxim.uva...@linaro.org]
Sent: Monday, December 04, 2017 1:10 PM
To: Savolainen, Petri (Nokia - FI/Espoo) 
Cc: Bogdan Pricope ; lng-odp-forward 

Subject: Re: [lng-odp] odp dpdk

313 does not work also:
https://lng.validation.linaro.org/scheduler/job/23242.1<https://www.google.com/url?q=https%3A%2F%2Flng.validation.linaro.org%2Fscheduler%2Fjob%2F23242.1&sa=D&sntz=1&usg=AFQjCNEsQmMQuPpbnJWtu_0-P8DCOlVKGg>
I will replace RX side to l2fwd and see that will be there.
Maxim.


On 4 December 2017 at 13:46, Savolainen, Petri (Nokia - FI/Espoo) 
mailto:petri.savolai...@nokia.com>> wrote:
Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to fix 
checksum insert for 10/40GE Intel NICs.

-Petri

> -Original Message-
> From: lng-odp 
> [mailto:lng-odp-boun...@lists.linaro.org<mailto:lng-odp-boun...@lists.linaro.org>]
>  On Behalf Of
> Bogdan Pricope
> Sent: Monday, December 04, 2017 12:21 PM
> To: Maxim Uvarov mailto:maxim.uva...@linaro.org>>
> Cc: lng-odp-forward 
> mailto:lng-odp@lists.linaro.org>>
> Subject: Re: [lng-odp] odp dpdk
>
> I suspect this is actually caused by csum issue in TX side: on RX,
> socket pktio does not validate csum (and accept the packets) but on
> dpdk pktio the csum is validated and packets are dropped.
>
> I am not seeing this in my setup because default txq_flags for igb
> driver (1G interface) is
>
> .txq_flags = 0
>
> while for ixgbe (10G interface) is:
>
> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
> ETH_TXQ_FLAGS_NOOFFLOADS,
>
>
> /B
>
>
>
>
> On 1 December 2017 at 23:47, Maxim Uvarov 
> mailto:maxim.uva...@linaro.org>> wrote:
> >
> > Looking to dpdk pktio support and generator. It looks like receive part
> > is broken. If for receive I use sockets it works well but receive with
> > dpdk does not get any packets. For both master and api-next. Can
> > somebody confirm please that it's so. Lava is not supper friendly to
> > debug issue.
> >
> >
> > 1. Recv
> > odp_generator -I 0 -m r -c 0x4
> >
> > https://lng.validation.linaro.org/scheduler/job/23206.1
> > Network devices using DPDK-compatible driver
> > 
> > :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
> > drv=igb_uio unused=
> >
> >
> >
> > 2. Send
> > odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac 38:ea:a7:93:83:a0
> > --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
> > 5000 -f 5001 -n 8
> >
> > https://lng.validation.linaro.org/scheduler/job/23206.0
> >
> > Thank you,
> > Maxim.

Re: [lng-odp] odp dpdk

[Maxim Uvarov]

313 does not work also:
https://lng.validation.linaro.org/scheduler/job/23242.1
<https://www.google.com/url?q=https%3A%2F%2Flng.validation.linaro.org%2Fscheduler%2Fjob%2F23242.1&sa=D&sntz=1&usg=AFQjCNEsQmMQuPpbnJWtu_0-P8DCOlVKGg>

I will replace RX side to l2fwd and see that will be there.

Maxim.


On 4 December 2017 at 13:46, Savolainen, Petri (Nokia - FI/Espoo) <
petri.savolai...@nokia.com> wrote:

> Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to
> fix checksum insert for 10/40GE Intel NICs.
>
> -Petri
>
> > -Original Message-
> > From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
> > Bogdan Pricope
> > Sent: Monday, December 04, 2017 12:21 PM
> > To: Maxim Uvarov 
> > Cc: lng-odp-forward 
> > Subject: Re: [lng-odp] odp dpdk
> >
> > I suspect this is actually caused by csum issue in TX side: on RX,
> > socket pktio does not validate csum (and accept the packets) but on
> > dpdk pktio the csum is validated and packets are dropped.
> >
> > I am not seeing this in my setup because default txq_flags for igb
> > driver (1G interface) is
> >
> > .txq_flags = 0
> >
> > while for ixgbe (10G interface) is:
> >
> > .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
> > ETH_TXQ_FLAGS_NOOFFLOADS,
> >
> >
> > /B
> >
> >
> >
> >
> > On 1 December 2017 at 23:47, Maxim Uvarov 
> wrote:
> > >
> > > Looking to dpdk pktio support and generator. It looks like receive part
> > > is broken. If for receive I use sockets it works well but receive with
> > > dpdk does not get any packets. For both master and api-next. Can
> > > somebody confirm please that it's so. Lava is not supper friendly to
> > > debug issue.
> > >
> > >
> > > 1. Recv
> > > odp_generator -I 0 -m r -c 0x4
> > >
> > > https://lng.validation.linaro.org/scheduler/job/23206.1
> > > Network devices using DPDK-compatible driver
> > > 
> > > :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
> > > drv=igb_uio unused=
> > >
> > >
> > >
> > > 2. Send
> > > odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac
> 38:ea:a7:93:83:a0
> > > --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
> > > 5000 -f 5001 -n 8
> > >
> > > https://lng.validation.linaro.org/scheduler/job/23206.0
> > >
> > > Thank you,
> > > Maxim.
>

Re: [lng-odp] odp dpdk

[Savolainen, Petri (Nokia - FI/Espoo)]

Maxim, try https://github.com/Linaro/odp/pull/313 It has been tested to fix 
checksum insert for 10/40GE Intel NICs.

-Petri

> -Original Message-
> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
> Bogdan Pricope
> Sent: Monday, December 04, 2017 12:21 PM
> To: Maxim Uvarov 
> Cc: lng-odp-forward 
> Subject: Re: [lng-odp] odp dpdk
> 
> I suspect this is actually caused by csum issue in TX side: on RX,
> socket pktio does not validate csum (and accept the packets) but on
> dpdk pktio the csum is validated and packets are dropped.
> 
> I am not seeing this in my setup because default txq_flags for igb
> driver (1G interface) is
> 
> .txq_flags = 0
> 
> while for ixgbe (10G interface) is:
> 
> .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
> ETH_TXQ_FLAGS_NOOFFLOADS,
> 
> 
> /B
> 
> 
> 
> 
> On 1 December 2017 at 23:47, Maxim Uvarov  wrote:
> >
> > Looking to dpdk pktio support and generator. It looks like receive part
> > is broken. If for receive I use sockets it works well but receive with
> > dpdk does not get any packets. For both master and api-next. Can
> > somebody confirm please that it's so. Lava is not supper friendly to
> > debug issue.
> >
> >
> > 1. Recv
> > odp_generator -I 0 -m r -c 0x4
> >
> > https://lng.validation.linaro.org/scheduler/job/23206.1
> > Network devices using DPDK-compatible driver
> > 
> > :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
> > drv=igb_uio unused=
> >
> >
> >
> > 2. Send
> > odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac 38:ea:a7:93:83:a0
> > --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
> > 5000 -f 5001 -n 8
> >
> > https://lng.validation.linaro.org/scheduler/job/23206.0
> >
> > Thank you,
> > Maxim.

Re: [lng-odp] odp dpdk

[Bogdan Pricope]

I suspect this is actually caused by csum issue in TX side: on RX,
socket pktio does not validate csum (and accept the packets) but on
dpdk pktio the csum is validated and packets are dropped.

I am not seeing this in my setup because default txq_flags for igb
driver (1G interface) is

.txq_flags = 0

while for ixgbe (10G interface) is:

.txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
ETH_TXQ_FLAGS_NOOFFLOADS,


/B




On 1 December 2017 at 23:47, Maxim Uvarov  wrote:
>
> Looking to dpdk pktio support and generator. It looks like receive part
> is broken. If for receive I use sockets it works well but receive with
> dpdk does not get any packets. For both master and api-next. Can
> somebody confirm please that it's so. Lava is not supper friendly to
> debug issue.
>
>
> 1. Recv
> odp_generator -I 0 -m r -c 0x4
>
> https://lng.validation.linaro.org/scheduler/job/23206.1
> Network devices using DPDK-compatible driver
> 
> :07:00.1 '82599ES 10-Gigabit SFI/SFP+ Network Connection 10fb'
> drv=igb_uio unused=
>
>
>
> 2. Send
> odp_generator -I 0 --srcmac 38:ea:a7:93:98:94 --dstmac 38:ea:a7:93:83:a0
> --srcip 192.168.100.2 --dstip 192.168.100.1 -m u -i 0 -c 0x8 -p 18 -e
> 5000 -f 5001 -n 8
>
> https://lng.validation.linaro.org/scheduler/job/23206.0
>
> Thank you,
> Maxim.

Re: [lng-odp] odp-dpdk gives error with "configure" command

[Gyanesh Patra]

Thank you.

It works fine now.

P Gyanesh Patra

On Mon, 30 Jan 2017 at 05:54 Elo Matias (Nokia - FI/Espoo)

<
mailto:Elo Matias (Nokia - FI/Espoo) 
> wrote:

a, pre, code, a:link, body { word-wrap: break-word !important; }

> On 28 Jan 2017, at 23:38, Gyanesh Patra <
mailto:pgyanesh.pa...@gmail.com
> wrote:

>

> odp-dpdk repo gives error for “configure” command when tried with dpdk 16.07 
> and dpdk 17. I am running on ubuntu16 LTS.

>

> ./configure --with-dpdk-path=./dpdk/x86_64-native-linuxapp-gcc

>

> checking rte_config.h usability... no

>

> checking rte_config.h presence... no

>

> checking for rte_config.h... no

>

> configure: error: in `/home/macsad/pktio/odp-dpdk':

>

> configure: error: "can't find DPDK headers"

>

> See `config.log' for more details

>

> macsad@india:~/pktio/odp-dpdk$

>

> P Gyanesh Patra

Hi Gyanesh,

I’m installing odp-dpdk as follows:

# DPDK install

cd

git checkout v16.07

make config T=x86_64-native-linuxapp-gcc O=x86_64-native-linuxapp-gcc

cd x86_64-native-linuxapp-gcc/

sed -ri 's,(CONFIG_RTE_LIBRTE_PMD_PCAP=).*,\1y,' .config

cd ..

make install T=x86_64-native-linuxapp-gcc DESTDIR=./install EXTRA_CFLAGS="-fPIC"

# Odp-dpdk install

cd

./bootstrap

./configure --with-platform=linux-dpdk --with-sdk-install-path=

/x86_64-native-linuxapp-gcc

make

Did this help?

-Matias

Re: [lng-odp] odp-dpdk gives error with "configure" command

[Elo, Matias (Nokia - FI/Espoo)]

> On 28 Jan 2017, at 23:38, Gyanesh Patra  wrote:
> 
> odp-dpdk repo gives error for “configure” command when tried with dpdk 16.07 
> and dpdk 17. I am running on ubuntu16 LTS. 
> 
> ./configure --with-dpdk-path=./dpdk/x86_64-native-linuxapp-gcc
> 
> checking rte_config.h usability... no
> 
> checking rte_config.h presence... no
> 
> checking for rte_config.h... no
> 
> configure: error: in `/home/macsad/pktio/odp-dpdk':
> 
> configure: error: "can't find DPDK headers"
> 
> See `config.log' for more details
> 
> macsad@india:~/pktio/odp-dpdk$ 
> 
> P Gyanesh Patra

Hi Gyanesh,

I’m installing odp-dpdk as follows:

# DPDK install
cd 
git checkout v16.07
make config T=x86_64-native-linuxapp-gcc O=x86_64-native-linuxapp-gcc
cd x86_64-native-linuxapp-gcc/
sed -ri 's,(CONFIG_RTE_LIBRTE_PMD_PCAP=).*,\1y,' .config
cd ..
make install T=x86_64-native-linuxapp-gcc DESTDIR=./install EXTRA_CFLAGS="-fPIC"

# Odp-dpdk install
cd 
./bootstrap
./configure --with-platform=linux-dpdk 
--with-sdk-install-path=/x86_64-native-linuxapp-gcc
make


Did this help?

-Matias

Re: [lng-odp] ODP dpdk pktio build broken in api-next

[Elo, Matias (Nokia - FI/Espoo)]

Hi,

Good that the problem is fixed. It would help a bit (and we should really do it 
anyway) if we at least build all the pktio types (netmap, dpdk etc.) in the CI 
system. Implementing this shouldn't take much effort. The build has been broken 
with optional pktio types way too often in the recent weeks.

Regards,
Matias

> -Original Message-
> From: EXT Maxim Uvarov [mailto:maxim.uva...@linaro.org]
> Sent: Wednesday, April 13, 2016 2:23 PM
> To: Elo, Matias (Nokia - FI/Espoo) 
> Subject: Re: ODP dpdk pktio build broken in api-next
> 
> On 04/13/16 13:11, Elo, Matias (Nokia - FI/Espoo) wrote:
> > Hi Maxim,
> >
> > The dpdk pktio build in api-next is currently broken due to a merge error. 
> > I just
> sent a fix for this to the list.
> >
> > Regards,
> > Matias
> yes, applying, thanks! It's not trivial now to sync both branches. Might
> be it will be more easy to merge everything to one branch
> and back port to master
> 
> Maxim.
___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] [ODP-DPDK] odp_generator ":PMD requires >32 buffers burst"

[Zoltan Kiss]

Hi Nicolas,

The issue about QUEUE_MULTI_MAX being 8 will disappear when you start 
using DPDK 2.2. The ixgbe driver lowered the requirement for minimum 
receive batch size to 4. I'll push the patches for 2.2 support on 
ODP-DPDK probably tomorrow, if nothing objects on the patches.


Regards,

Zoltan

On 12/10/15 18:11, Zoltan Kiss wrote:

Hi,

Note, there is a lng-odp-d...@lists.linaro.org list for ODP-DPDK
specific issues.

On 06/10/15 12:03, Kury Nicolas wrote:

Hi


I have to NIC. A HP560SFP+ 10G (82599ES) and a Intel EXPI9301CT 1G
(82574L), I have some problems when I run odp_generator from
ODP-DPDK v1.2.0.0_DPDK_2.1.0-rc3.


First problem.

With the HP560SFP+ 10G, there is the messages "PMD requires >32 buffers
burst. Current 8" when I receive paquets. With the other NIC, no problem.


The ixgbe vector poll mode driver requires at least 32 buffers passed to
the receive call, otherwise it won't return any packet (and even if you
give it more it will only return 32 btw.) It is the default driver now
in 2.1, so we'll use go with it too. There is a patch upstream which
decrease this minimum to 4, but as ODP can't impose such requirement to
the application, we have a workaround in place which makes sure the DPDK
function is called with the appropriate minimum amount of buffers, then
drops the surplus. It's not nice, but so far noone had a better idea.
For this particular problem we should change the odp_pktio_recv() calls
in platform/linux-dpdk/odp_packet_io.c to use the proper minimum instead
of QUEUE_MULTI_MAX. As a workaround you can overwrite that value in your
local copy to 32.

___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] [ODP-DPDK] SpanningTree and block a port ?

[Kury Nicolas]

Hi!

Thank you very much for the answer!

Regards
Nicolas



De : Brian Brooks 
Envoyé : mercredi 7 octobre 2015 15:49
À : Kury Nicolas; lng-odp@lists.linaro.org
Objet : RE: [ODP-DPDK] SpanningTree and block a port ?

Hi Nicolas,

In your use case, a STP 'application' and a traffic generator 'application' 
need to be designed to run as separate processes that work together.

Multi-process aside, odp_pktio_start/stop() is documented as a way to 
start/stop tx & rx on a packet IO interface. Depending on the implementation, 
this may be usable for STP.

For MSTP, perhaps you could utilize the classification APIs to match on pktio + 
VLAN and drop for ingress blocking. However, the application would need to be 
aware of the spanning tree state to prevent transmitting packets destined for 
blocked ports.

Note the API for polling link state: 
https://lists.linaro.org/pipermail/lng-odp/2015-September/015557.html

Regards,
Brian

From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of Kury 
Nicolas
Sent: Friday, October 02, 2015 8:50 AM
To: lng-odp@lists.linaro.org
Subject: [lng-odp] [ODP-DPDK] SpanningTree and block a port ?

Hi

I would like to implement an application with spanning tree protocol. But can 
the application "spanningtree" stop and start a port used by another 
applications (exemple odp_generator) ?
Here a diagram:  http://s8.postimg.org/yis5flhed/spanningtree.png

For exemple, if the link #1 is down, spanningtree application enables port 0.

Thank you
Nicolas
Transmitting



-- IMPORTANT NOTICE: The contents of this email and any attachments are 
confidential and may also be privileged. If you are not the intended recipient, 
please notify the sender immediately and do not disclose the contents to any 
other person, use it for any purpose, or store or copy the information in any 
medium. Thank you.

___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] [ODP-DPDK] odp_generator ":PMD requires >32 buffers burst"

[Zoltan Kiss]

Hi,

Note, there is a lng-odp-d...@lists.linaro.org list for ODP-DPDK 
specific issues.


On 06/10/15 12:03, Kury Nicolas wrote:

Hi


I have to NIC. A HP560SFP+ 10G (82599ES) and a Intel EXPI9301CT 1G
(82574L), I have some problems when I run odp_generator from
ODP-DPDK v1.2.0.0_DPDK_2.1.0-rc3.


First problem.

With the HP560SFP+ 10G, there is the messages "PMD requires >32 buffers
burst. Current 8" when I receive paquets. With the other NIC, no problem.


The ixgbe vector poll mode driver requires at least 32 buffers passed to 
the receive call, otherwise it won't return any packet (and even if you 
give it more it will only return 32 btw.) It is the default driver now 
in 2.1, so we'll use go with it too. There is a patch upstream which 
decrease this minimum to 4, but as ODP can't impose such requirement to 
the application, we have a workaround in place which makes sure the DPDK 
function is called with the appropriate minimum amount of buffers, then 
drops the surplus. It's not nice, but so far noone had a better idea.
For this particular problem we should change the odp_pktio_recv() calls 
in platform/linux-dpdk/odp_packet_io.c to use the proper minimum instead 
of QUEUE_MULTI_MAX. As a workaround you can overwrite that value in your 
local copy to 32.



sudo ./odp_generator -I 0 -srcmac fe:0f:97:c9:e0:44  --dstmac
32:cb:9b:27:2f:1a --srcip 192.168.0.1 --dstip 192.168.0.2 -m u

odp_packet_dpdk.c:188:recv_pkt_dpdk():PMD requires >32 buffers burst.
Current 8, dropped 24
odp_packet_dpdk.c:188:recv_pkt_dpdk():PMD requires >32 buffers burst.
Current 8, dropped 24
odp_packet_dpdk.c:188:recv_pkt_dpdk():PMD requires >32 buffers burst.
Current 8, dropped 24
...


_

The second problem I have (with both NIC) is that I must give the
interval argument (exemple 2000ms).

sudo ./odp_generator -I 2 -srcmac fe:0f:97:c9:e0:44  --dstmac
32:cb:9b:27:2f:1a --srcip 192.168.0.1 --dstip 192.168.0.2 -m u -i 2000


If the interval is zero or a small value (exemple 1ms), the sending
thread stops. I have also tried with 7 workers (-w 7). With 1000ms it's
okay but I would like to test odp_generator to reach the 1,488Mpps on
line 1Gb/s...


sudo ./odp_generator -I 2 -srcmac fe:0f:97:c9:e0:44  --dstmac
32:cb:9b:27:2f:1a --srcip 192.168.0.1 --dstip 192.168.0.2 -m u -i 1

   [01] send pkt no:571 seq 571
   [01] send pkt no:572 seq 572
   [01] send pkt no:573 seq 573
   [01] send pkt no:574 seq 574
   [01] send pkt no:575 seq 575
odp_generator.c:415:gen_send_thread():  [01] send pkt err!


The queueing part of the ODP-DPDK code is not tested too much I'm 
afraid. You can try and debug where the error code comes from (somewhere 
in linux-generic/odp_queue.c::queue_pktout_enq(), I guess, my bet would 
be that odp_pktio_send() returns it in the end)


It comes form this line. What is the reason ? Can I just comment these
lines ?

I'm afraid it wouldn't help.


 err = odp_queue_enq(outq_def, odp_packet_to_event(pkt));
/* if (err != 0) {
 EXAMPLE_ERR("  [%02i] send pkt err!\n", thr);
 odp_packet_free(pkt);
 return NULL;
 }*/


_


Thank you!

Nicolas







___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp


___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] [ODP-DPDK] SpanningTree and block a port ?

[Brian Brooks]

Hi Nicolas,

In your use case, a STP 'application' and a traffic generator 'application' 
need to be designed to run as separate processes that work together.

Multi-process aside, odp_pktio_start/stop() is documented as a way to 
start/stop tx & rx on a packet IO interface. Depending on the implementation, 
this may be usable for STP.

For MSTP, perhaps you could utilize the classification APIs to match on pktio + 
VLAN and drop for ingress blocking. However, the application would need to be 
aware of the spanning tree state to prevent transmitting packets destined for 
blocked ports.

Note the API for polling link state: 
https://lists.linaro.org/pipermail/lng-odp/2015-September/015557.html

Regards,
Brian

From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of Kury 
Nicolas
Sent: Friday, October 02, 2015 8:50 AM
To: lng-odp@lists.linaro.org
Subject: [lng-odp] [ODP-DPDK] SpanningTree and block a port ?

Hi

I would like to implement an application with spanning tree protocol. But can 
the application "spanningtree" stop and start a port used by another 
applications (exemple odp_generator) ?
Here a diagram:  http://s8.postimg.org/yis5flhed/spanningtree.png

For exemple, if the link #1 is down, spanningtree application enables port 0.

Thank you
Nicolas
Transmitting



-- IMPORTANT NOTICE: The contents of this email and any attachments are 
confidential and may also be privileged. If you are not the intended recipient, 
please notify the sender immediately and do not disclose the contents to any 
other person, use it for any purpose, or store or copy the information in any 
medium. Thank you.

___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] [ODP-DPDK] Question about arguments for odp_generator

[Zoltan Kiss]

Hi,

There is a separate list for ODP-DPDK, I've put it in the Cc.

You can pass your parameters via ODP_PLATFORM_PARAMS environment variable:

https://git.linaro.org/lng/odp-dpdk.git/blob/HEAD:/platform/linux-dpdk/README#l225

Regards,

Zoltan

On 14/09/15 12:13, Kury Nicolas wrote:

Hi


I have DPDK 2.1.0 and ODP-DPDK  v1.2.0.0_DPDK_2.1.0.


I would like to try the exemple odp_generator but I didn't find how tu
use the arguments of DPDK. It seems that -c (bitmask of cores to run on)
and -n ( Number of memory channels) aren't recognized.


Exemple with a application from DPDK package. It works, I see
received/sent packets.

*sudo ./l2fwd -c 1 -n 1 -- -p 4*


With the application odp_generator from ODP-DPDK:

*sudo ./odp_generator -c 1 -n 1 --p 3*
But I have the error "CPU cores must be enabled with options -c, -l or
--lcores". See
http://s10.postimg.org/hi2jkt9gp/Capture_du_2015_09_14_13_10_34.png


How schould I do ?


Thank you

Nicolas




___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp


___
lng-odp mailing list
lng-odp@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] ODP-DPDK multi queue support

[Maxim Uvarov]

On 09/03/2014 02:45 PM, Santosh Shukla wrote:

"Patch to test multi-queue support on ODP-DPDK without " with multi-queue.

Tested-by : Santosh Shukla 


Santosh thanks for testing this! Unfortunately I can not modify commits 
in public repo,

so this record will be here in mailing list.

Best regards,
Maxim.




On 3 September 2014 16:10, Santosh Shukla  wrote:

Confirmed. It is working.

On 1 September 2014 14:09, Venkatesh Vivekanandan
 wrote:



On 28 August 2014 12:04, Santosh Shukla  wrote:

No hw availability for this week, Next week I will have x86 boxes with
me so expect to see multi-queue update from me. Thanks.


It is better if you can close this off when your setup is back. It takes
hardly 10-15 min once you have the setup.


On 28 August 2014 12:01, Venkatesh Vivekanandan
 wrote:

Hi Santosh/Mike,

I would like to get this to closure. Can you please check the
multi-queue
support for odp-dpdk?. It is just that you have to generate packets with
different dstip's from odp_generator. Please refer "Patch to test
multi-queue support on ODP-DPDK without Ixia" email for the test setup.
Please let me know if you need any help.

Thanks,
Venkatesh.

___
lng-odp mailing list
lng-odp@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/lng-odp




___
lng-odp mailing list
lng-odp@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/lng-odp



___
lng-odp mailing list
lng-odp@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] ODP-DPDK multi queue support

[Maxim Uvarov]

On 09/03/2014 02:48 PM, Venkatesh Vivekanandan wrote:

Hi Maxim,

Can you please add the Tested-by to the below commit,



I can not change existence commits. But we can have this for record in 
mailing list.


Maxim.


commit e9e7bc237086d6d07dc004e591063838a12563ed
Author: Venkatesh Vivekanandan >

Date:   Tue Aug 5 20:24:05 2014 +0530

ODP-DPDK multi-queue support

Thanks,
Venky


On 3 September 2014 16:15, Santosh Shukla > wrote:


"Patch to test multi-queue support on ODP-DPDK without " with
multi-queue.

Tested-by : Santosh Shukla mailto:santosh.shu...@linaro.org>>

On 3 September 2014 16:10, Santosh Shukla
mailto:santosh.shu...@linaro.org>> wrote:
> Confirmed. It is working.
>
> On 1 September 2014 14:09, Venkatesh Vivekanandan
> mailto:venkatesh.vivekanan...@linaro.org>> wrote:
>>
>>
>>
>> On 28 August 2014 12:04, Santosh Shukla
mailto:santosh.shu...@linaro.org>> wrote:
>>>
>>> No hw availability for this week, Next week I will have x86
boxes with
>>> me so expect to see multi-queue update from me. Thanks.
>>>
>>
>> It is better if you can close this off when your setup is back.
It takes
>> hardly 10-15 min once you have the setup.
>>
>>>
>>> On 28 August 2014 12:01, Venkatesh Vivekanandan
>>> mailto:venkatesh.vivekanan...@linaro.org>> wrote:
>>> > Hi Santosh/Mike,
>>> >
>>> > I would like to get this to closure. Can you please check the
>>> > multi-queue
>>> > support for odp-dpdk?. It is just that you have to generate
packets with
>>> > different dstip's from odp_generator. Please refer "Patch to
test
>>> > multi-queue support on ODP-DPDK without Ixia" email for the
test setup.
>>> > Please let me know if you need any help.
>>> >
>>> > Thanks,
>>> > Venkatesh.
>>> >
>>> > ___
>>> > lng-odp mailing list
>>> > lng-odp@lists.linaro.org 
>>> > http://lists.linaro.org/mailman/listinfo/lng-odp
>>> >
>>
>>




___
lng-odp mailing list
lng-odp@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/lng-odp



___
lng-odp mailing list
lng-odp@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] ODP-DPDK multi queue support

[Venkatesh Vivekanandan]

Hi Maxim,

Can you please add the Tested-by to the below commit,

commit e9e7bc237086d6d07dc004e591063838a12563ed
Author: Venkatesh Vivekanandan 
Date:   Tue Aug 5 20:24:05 2014 +0530

ODP-DPDK multi-queue support

Thanks,
Venky


On 3 September 2014 16:15, Santosh Shukla  wrote:

> "Patch to test multi-queue support on ODP-DPDK without " with multi-queue.
>
> Tested-by : Santosh Shukla 
>
> On 3 September 2014 16:10, Santosh Shukla 
> wrote:
> > Confirmed. It is working.
> >
> > On 1 September 2014 14:09, Venkatesh Vivekanandan
> >  wrote:
> >>
> >>
> >>
> >> On 28 August 2014 12:04, Santosh Shukla 
> wrote:
> >>>
> >>> No hw availability for this week, Next week I will have x86 boxes with
> >>> me so expect to see multi-queue update from me. Thanks.
> >>>
> >>
> >> It is better if you can close this off when your setup is back. It takes
> >> hardly 10-15 min once you have the setup.
> >>
> >>>
> >>> On 28 August 2014 12:01, Venkatesh Vivekanandan
> >>>  wrote:
> >>> > Hi Santosh/Mike,
> >>> >
> >>> > I would like to get this to closure. Can you please check the
> >>> > multi-queue
> >>> > support for odp-dpdk?. It is just that you have to generate packets
> with
> >>> > different dstip's from odp_generator. Please refer "Patch to test
> >>> > multi-queue support on ODP-DPDK without Ixia" email for the test
> setup.
> >>> > Please let me know if you need any help.
> >>> >
> >>> > Thanks,
> >>> > Venkatesh.
> >>> >
> >>> > ___
> >>> > lng-odp mailing list
> >>> > lng-odp@lists.linaro.org
> >>> > http://lists.linaro.org/mailman/listinfo/lng-odp
> >>> >
> >>
> >>
>
___
lng-odp mailing list
lng-odp@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] ODP-DPDK multi queue support

[Santosh Shukla]

"Patch to test multi-queue support on ODP-DPDK without " with multi-queue.

Tested-by : Santosh Shukla 

On 3 September 2014 16:10, Santosh Shukla  wrote:
> Confirmed. It is working.
>
> On 1 September 2014 14:09, Venkatesh Vivekanandan
>  wrote:
>>
>>
>>
>> On 28 August 2014 12:04, Santosh Shukla  wrote:
>>>
>>> No hw availability for this week, Next week I will have x86 boxes with
>>> me so expect to see multi-queue update from me. Thanks.
>>>
>>
>> It is better if you can close this off when your setup is back. It takes
>> hardly 10-15 min once you have the setup.
>>
>>>
>>> On 28 August 2014 12:01, Venkatesh Vivekanandan
>>>  wrote:
>>> > Hi Santosh/Mike,
>>> >
>>> > I would like to get this to closure. Can you please check the
>>> > multi-queue
>>> > support for odp-dpdk?. It is just that you have to generate packets with
>>> > different dstip's from odp_generator. Please refer "Patch to test
>>> > multi-queue support on ODP-DPDK without Ixia" email for the test setup.
>>> > Please let me know if you need any help.
>>> >
>>> > Thanks,
>>> > Venkatesh.
>>> >
>>> > ___
>>> > lng-odp mailing list
>>> > lng-odp@lists.linaro.org
>>> > http://lists.linaro.org/mailman/listinfo/lng-odp
>>> >
>>
>>

___
lng-odp mailing list
lng-odp@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] ODP-DPDK multi queue support

[Santosh Shukla]

Confirmed. It is working.

On 1 September 2014 14:09, Venkatesh Vivekanandan
 wrote:
>
>
>
> On 28 August 2014 12:04, Santosh Shukla  wrote:
>>
>> No hw availability for this week, Next week I will have x86 boxes with
>> me so expect to see multi-queue update from me. Thanks.
>>
>
> It is better if you can close this off when your setup is back. It takes
> hardly 10-15 min once you have the setup.
>
>>
>> On 28 August 2014 12:01, Venkatesh Vivekanandan
>>  wrote:
>> > Hi Santosh/Mike,
>> >
>> > I would like to get this to closure. Can you please check the
>> > multi-queue
>> > support for odp-dpdk?. It is just that you have to generate packets with
>> > different dstip's from odp_generator. Please refer "Patch to test
>> > multi-queue support on ODP-DPDK without Ixia" email for the test setup.
>> > Please let me know if you need any help.
>> >
>> > Thanks,
>> > Venkatesh.
>> >
>> > ___
>> > lng-odp mailing list
>> > lng-odp@lists.linaro.org
>> > http://lists.linaro.org/mailman/listinfo/lng-odp
>> >
>
>

___
lng-odp mailing list
lng-odp@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] ODP-DPDK multi queue support

[Venkatesh Vivekanandan]

On 28 August 2014 12:04, Santosh Shukla  wrote:

> No hw availability for this week, Next week I will have x86 boxes with
> me so expect to see multi-queue update from me. Thanks.
>
>
It is better if you can close this off when your setup is back. It takes
hardly 10-15 min once you have the setup.


> On 28 August 2014 12:01, Venkatesh Vivekanandan
>  wrote:
> > Hi Santosh/Mike,
> >
> > I would like to get this to closure. Can you please check the multi-queue
> > support for odp-dpdk?. It is just that you have to generate packets with
> > different dstip's from odp_generator. Please refer "Patch to test
> > multi-queue support on ODP-DPDK without Ixia" email for the test setup.
> > Please let me know if you need any help.
> >
> > Thanks,
> > Venkatesh.
> >
> > ___
> > lng-odp mailing list
> > lng-odp@lists.linaro.org
> > http://lists.linaro.org/mailman/listinfo/lng-odp
> >
>
___
lng-odp mailing list
lng-odp@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/lng-odp

Re: [lng-odp] ODP-DPDK multi queue support

[Santosh Shukla]

No hw availability for this week, Next week I will have x86 boxes with
me so expect to see multi-queue update from me. Thanks.

On 28 August 2014 12:01, Venkatesh Vivekanandan
 wrote:
> Hi Santosh/Mike,
>
> I would like to get this to closure. Can you please check the multi-queue
> support for odp-dpdk?. It is just that you have to generate packets with
> different dstip's from odp_generator. Please refer "Patch to test
> multi-queue support on ODP-DPDK without Ixia" email for the test setup.
> Please let me know if you need any help.
>
> Thanks,
> Venkatesh.
>
> ___
> lng-odp mailing list
> lng-odp@lists.linaro.org
> http://lists.linaro.org/mailman/listinfo/lng-odp
>

___
lng-odp mailing list
lng-odp@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/lng-odp