On 9 September 2015 at 09:29, Maxim Uvarov wrote:
> On 09/08/15 16:04, Alexandru Badicioiu wrote:
>
>> I agree, at least in my case some CPUs are assigned at boot time for
>> dataplane work and cannot be changed.
>>
>> Alex
>>
>>
> Alex, how do you run validation tests then? Do you have your own
On 09/08/15 16:04, Alexandru Badicioiu wrote:
I agree, at least in my case some CPUs are assigned at boot time for
dataplane work and cannot be changed.
Alex
Alex, how do you run validation tests then? Do you have your own variant
of odp helper to create threads?
Maxim.
On 8 September
Petri, do you want to review these patches?
Thanks,
Maxim.
On 08/31/15 16:05, Ivan Khoronzhuk wrote:
This series contains fixes and is required before changing time API.
It includes previous patches:
- "[lng-odp] [Patch v2 0/2] performance: odp_pktio_perf: fix potentials ticks
overflows"
Merged,
Maxim.
On 09/08/15 18:08, Nicolas Morey-Chaisemartin wrote:
On 09/07/2015 01:53 PM, Christophe Milard wrote:
Renaming module libs in /test/validation/. Module test libs are
renamed from lib.la to libtestmodule.la.
This is done to avoid libname clashes, such as libcrypto, which seems to
Merged.
On 09/08/15 23:34, Bill Fischofer wrote:
Maxim: I believe this one should be ready to merge. I added my
reviewed to the v2 version of this patch last week.
Yes. Not it's good. I did not include it to 1.3 to not break anything
which we can miss.
Maxim.
On Tue, Sep 8, 2015 at 10:3
Thanks Bob, having all OPNFV setup up is definitely interesting
project. I guess it even can be part of lng openembedded image or some
debian repository.
So that everyone can download and start paying with it. And of course
more people will learn what is ODP.
Best regards,
Maxim.
On 09/08/15
Thanks very much for your support!
Robert (Bob) Monkman
Enterprise Segment Marketing Manager
150 Rose Orchard Way
San Jose, Ca 95134
M: +1.510.676.5490
From: Cuiaiguo (Aiguo Cui, WN) [mailto:ag@huawei.com]
Sent: Tuesday, September 08, 2015 7:40 PM
To: Bob Monkman; lng...@linaro.org; lng-odp@l
Hi Bob
Congratulations. I think we LNG can update some requiremets in coming SFO15.
CAG
Sent from HUAWEI AnyOffice
发件人:Bob.Monkman
收件人:Linaro Networking Group Steering Committee (lng...@linaro.org),LNG ODP
Mailman List (lng-odp@lists.linaro.org),
抄送:Robert Dimond,Bob.Monkman,Charlene Marini,Co
Maxim: I believe this one should be ready to merge. I added my reviewed to
the v2 version of this patch last week.
On Tue, Sep 8, 2015 at 10:37 AM, Nicolas Morey-Chaisemartin <
nmo...@kalray.eu> wrote:
> ping ?
>
> On 09/01/2015 02:51 PM, Bill Fischofer wrote:
> > On Tue, Sep 1, 2015 at 7:50 AM,
On 7 September 2015 at 06:01, Stuart Haslam
wrote:
> On Mon, Sep 07, 2015 at 10:42:13AM +0300, Maxim Uvarov wrote:
> > On 09/04/15 20:37, Bala Manoharan wrote:
> > >This method of using pcap file to generate packets is fine.
> > >But why should we use a dedicated interface with "pcap" as the name
On 7 September 2015 at 06:13, Maxim Uvarov wrote:
> On 09/07/15 12:18, Christophe Milard wrote:
>
>> I agree with this Patch: I don't see any reason to keep non working code
>> in the repos'head. This code is still in git, if we want to get it later.
>> /Christophe.
>>
>> my point is: or remove t
Cool!
Sent from my mobile. Excuse any typos.
Alkuperäinen viesti
Lähettäjä: ext Bob Monkman
Päivämäärä: 08.09.2015 17.57 (GMT+02:00)
Saaja: "Linaro Networking Group Steering Committee (lng...@linaro.org)"
, "LNG ODP Mailman List (lng-odp@lists.linaro.org)"
Kopio: Robert D
https://bugs.linaro.org/show_bug.cgi?id=1752
Ola Liljedahl changed:
What|Removed |Added
CC||ola.liljed...@linaro.org
--- Comment #3 from O
ping ?
On 09/01/2015 02:51 PM, Bill Fischofer wrote:
> On Tue, Sep 1, 2015 at 7:50 AM, Nicolas Morey-Chaisemartin > wrote:
>> Signed-off-by: Nicolas Morey-Chaisemartin
>>
> Reviewed-by: Bill Fischofer
>
>
>> ---
>> v3:
>> Packet for add_rem_data is now packet_len so we can add tail data when
>>
This one won't apply on master anymore:
Applying: linux-generic: pktio: extend maximum devname length
error: patch failed: platform/linux-generic/odp_packet_io.c:239
error: platform/linux-generic/odp_packet_io.c: patch does not apply
Patch failed at 0001 linux-generic: pktio: extend maximum devnam
On 09/04/2015 03:20 PM, Stuart Haslam wrote:
> Create a new pktio type that allows for reading from and writing to a
> pcap capture file. This is intended to be used as a simple way of
> injecting test packets into an application for functional testing and
> can be used as it is with some of the
On 09/07/2015 01:53 PM, Christophe Milard wrote:
> Renaming module libs in /test/validation/. Module test libs are
> renamed from lib.la to libtestmodule.la.
> This is done to avoid libname clashes, such as libcrypto, which seems to
> confuse libtool, and should therefore fix bug ID 1787.
>
> Sig
All,
I am pleased to announce that OPNFV ARMBand project, which I
sent a description of out on the LNG list last week, has been approved as an
official project in OPNFV and approved to target the Brahmaputra Release,
tentatively scheduled for February 2016. This was voted and app
Yes. Basically you want a test that creates 10,000+ queues that have work
on them and then measure things like average queue wait time, etc.
On Tue, Sep 8, 2015 at 9:03 AM, Mike Holmes wrote:
> Do we need to add a specific test to the performance suite to aid
> assessing these issues ?
>
> On 8
Do we need to add a specific test to the performance suite to aid assessing
these issues ?
On 8 September 2015 at 09:57, Nicolas Morey-Chaisemartin
wrote:
>
>
> On 09/08/2015 03:33 PM, Ola Liljedahl wrote:
> > Sorry I missed this discussion. It is really interesting. IMO the
> linux-generic sche
On 09/08/2015 03:33 PM, Ola Liljedahl wrote:
> Sorry I missed this discussion. It is really interesting. IMO the
> linux-generic scheduler is too simplistic to be used as is or its behaviour
> copied. We have seen some undesirable behaviour in our internal work where we
> use ODP. Very simplif
Sorry I missed this discussion. It is really interesting. IMO the
linux-generic scheduler is too simplistic to be used as is or its behaviour
copied. We have seen some undesirable behaviour in our internal work where
we use ODP. Very simplified, we essentially have two levels of processing,
both fe
> > +#ifndef ODP_API_RWLOCK_RECURSIVE_H_
> > +#define ODP_API_RWLOCK_RECURSIVE_H_
> > +
> > +#ifdef __cplusplus
> > +extern "C" {
> > +#endif
> > +
> > +/** @addtogroup odp_synchronizers
> > + * Operations on recursive rwlocks.
> > + * @{
> > + */
> > +
> > +/**
> > + * @typedef odp_rwlock_recurs
On 8 September 2015 at 15:57, Ola Liljedahl
wrote:
> On 1 September 2015 at 13:09, wrote:
>
>> From: Alexandru Badicioiu
>>
>> This patch provides the applications with helpers to create, update
>> and read counters associated with application defined objects - e.g.
>> IPSec security associatio
I agree, at least in my case some CPUs are assigned at boot time for
dataplane work and cannot be changed.
Alex
On 8 September 2015 at 15:57, Savolainen, Petri (Nokia - FI/Espoo) <
petri.savolai...@nokia.com> wrote:
> I'm wondering the use case for asking only the default number of CPUs
> suita
I'm wondering the use case for asking only the default number of CPUs suitable
for worker threads (without asking CPU IDs).
I think application should ask the mask (and number of cpus) once and store the
information for later use.
-Petri
> -Original Message-
> From: lng-odp [mailto:l
On 1 September 2015 at 13:09, wrote:
> From: Alexandru Badicioiu
>
> This patch provides the applications with helpers to create, update
> and read counters associated with application defined objects - e.g.
> IPSec security associations in a multi-threading scenario where
> multiple threads use
https://bugs.linaro.org/show_bug.cgi?id=1615
--- Comment #18 from Ola Liljedahl ---
We don't want the periodic timer thread to perform the teardown (as some
unspecified time). The teardown needs to be complete (all resources freed) when
odp_timer_pool_destroy (?) returns.
Stuart suggested we cre
https://bugs.linaro.org/show_bug.cgi?id=1615
--- Comment #17 from Bill Fischofer ---
To properly clean up the following is required.
1. Be able to quiesce the context so that no new events will be accepted by the
APIs.
2. Modify the context so that it tracks the number of events outstanding
aga
https://bugs.linaro.org/show_bug.cgi?id=1615
--- Comment #16 from Stuart Haslam ---
I thought about just setting some flag in odp_timer_pool_del and waiting for
the next timer expiry to action on it, but assuming that it's possible for
there to be more than 1 outstanding event (in different threa
https://bugs.linaro.org/show_bug.cgi?id=1615
--- Comment #15 from Ola Liljedahl ---
Synchronize with what? We don't know if there is any timer signal in-flight. If
we could inject our own signal into the kernel signal queues, we could wait for
that signal to be handled (assuming it would be handl
On 8 September 2015 at 13:32, Zoltan Kiss wrote:
> I've meant it to the other patch for ODP-DPDK :)
> I wanted to rather ask Anders what's going on with this patch?
I was surprised to see two patches in one patch, so I couldn't apply this one
until you applied the patch to odp-dpdk repo.
Applied
Applied
On 02/09/15 16:45, Zoltan Kiss wrote:
And remove the restriction about IP fragment support.
Signed-off-by: Zoltan Kiss
---
helper/platform/dpdk | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/helper/platform/dpdk b/helper/platform/dpdk
index ae545a2..58cd6eb 10
I've meant it to the other patch for ODP-DPDK :)
I wanted to rather ask Anders what's going on with this patch?
Zoli
On 08/09/15 12:30, Zoltan Kiss wrote:
Applied
On 02/09/15 16:45, Zoltan Kiss wrote:
And remove the restriction about IP fragment support.
Signed-off-by: Zoltan Kiss
---
hel
On 08.09.15 14:03, Maxim Uvarov wrote:
On 09/08/15 13:55, Ivan Khoronzhuk wrote:
Maxim,
Could you please clarify.
I expect that after patch adding some implementation should build.
If you adding new API it doesn't have impact.
But if API is modified, probably that project has build problem.
T
Merged,
Maxim.
On 09/02/15 11:25, Nicolas Morey-Chaisemartin wrote:
Actual check/changes of the pktio state is now done by the pktio
global interface.
Each pktio can still implement a start/stop function to do additional
actions.
Signed-off-by: Nicolas Morey-Chaisemartin
---
platform/linux-g
On 4 September 2015 at 15:44, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
> Cpu mask API mainly manipulate the mask. So, odp_cpumask_max can be mixed to
> return “max” or “last” cpu ID. Same problem with plain odp_cpumask_all –
> it’s too close to “set all cpu IDs in the mask”.
>
>
>
> I think the
https://bugs.linaro.org/show_bug.cgi?id=1615
Bill Fischofer changed:
What|Removed |Added
CC||bill.fischo...@linaro.org
--- Comment #14 fro
ODP has api to request available number of workers. Now
no need limit that inside application.
Signed-off-by: Maxim Uvarov
---
v2: timer.c: do not substract -1 from number of workers.
test/validation/common/odp_cunit_common.c | 7 ++--
test/validation/common/odp_cunit_common.h | 2
On 09/08/15 13:55, Ivan Khoronzhuk wrote:
Maxim,
Could you please clarify.
I expect that after patch adding some implementation should build.
If you adding new API it doesn't have impact.
But if API is modified, probably that project has build problem.
This patch, can have examples/validations i
Maxim,
Could you please clarify.
I expect that after patch adding some implementation should build.
If you adding new API it doesn't have impact.
But if API is modified, probably that project has build problem.
This patch, can have examples/validations in one patch in order to allow every
patch
https://bugs.linaro.org/show_bug.cgi?id=1615
--- Comment #13 from Stuart Haslam ---
In my case the tp in the signal handler is a valid timer pool pointer, or at
least it was a short time ago, so it's stale rather than corrupted.
Commenting out the odp_shm_free() in odp_timer_pool_del() makes the
ename odp_cpumask_def_worker/control to _default_worker/control
and make mask argument optinal.
Maxim.
Maxim Uvarov (4):
api: rename odp_cpumask_def to _default
linux-generic/validation/example/helper: rename to
odp_cpumask_default_worker
api: odp_cpumask_default_ mask argument can be
Functions odp_cpumask_default_worker and odp_cpumask_default_control
can be used for calculation number of worker and control threads. In
that case mask parameter can be optional.
Signed-off-by: Maxim Uvarov
---
include/odp/api/cpumask.h | 2 +-
platform/linux-generic/odp_cpumas
Signed-off-by: Maxim Uvarov
---
test/validation/synchronizers/synchronizers.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/test/validation/synchronizers/synchronizers.c
b/test/validation/synchronizers/synchronizers.c
index 06977f9..53ddc33 100644
--- a/test/validation/sy
Use full default word in api to make function
name more clear.
https://bugs.linaro.org/show_bug.cgi?id=1745
Signed-off-by: Maxim Uvarov
---
include/odp/api/cpumask.h | 4 ++--
platform/linux-generic/odp_cpumask_task.c | 4 ++--
2 files changed, 4 insertions(+), 4 deletions(-)
di
Do renames to odp_cpumask_default_worker and odp_cpumask_default_control.
Signed-off-by: Maxim Uvarov
---
example/classifier/odp_classifier.c | 2 +-
example/generator/odp_generator.c | 4 ++--
example/ipsec/odp_ipsec.c | 2 +-
example/packet/odp_pktio.
On 08.09.15 11:31, Petri Savolainen wrote:
Use CPU cycle count API instead of time API to measure CPU
cycles.
Signed-off-by: Petri Savolainen
Reviewed-by: Ivan Khoronzhuk
Minor comments below.
---
test/performance/odp_scheduling.c | 147 ++
1 file c
On 08.09.15 11:31, Petri Savolainen wrote:
Added implementation for CPU cycle diff, max and resolution.
Signed-off-by: Petri Savolainen
Reviewed-by: Ivan Khoronzhuk
---
platform/linux-generic/Makefile.am | 1 +
platform/linux-generic/arch/linux/odp_cpu_cycles.c | 10
On 08.09.15 11:31, Petri Savolainen wrote:
Implemented odp_cpu_cycles() be renaming odp_time_cycles()
implementation. Time implementation uses odp_cpu_cycles
temporarely.
temporarely -> temporarily
Reviewed-by: Ivan Khoronzhuk
Signed-off-by: Petri Savolainen
---
platform/linux-generic
On 08.09.15 11:31, Petri Savolainen wrote:
Renamed time_cycles files to cpu_cycles, since those files
implement cpu cycle counter read needed by cpu API but not
needed by new time API.
Signed-off-by: Petri Savolainen
Reviewed-by: Ivan Khoronzhuk
---
platform/linux-generic/Makefile.am
On 08.09.15 11:31, Petri Savolainen wrote:
Raw CPU cycle counts can be used to measure performance
in CPU cycles. These functions will replace some usage of
odp_time_cycles() of odp_time_diff_cycles().
Signed-off-by: Petri Savolainen
I'm only hesitate about 32bits, for sum & cmp, but it can
On 08.09.15 12:37, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: ext Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 08, 2015 12:20 PM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp] [API-NE
> -Original Message-
> From: ext Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
> Sent: Tuesday, September 08, 2015 12:20 PM
> To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
> Subject: Re: [lng-odp] [API-NEXT PATCH 3/5] linux-generic: cpu: rename
> time_cycles to
On 09/07/15 17:35, Stuart Haslam wrote:
On Mon, Sep 07, 2015 at 02:05:29PM +0300, Maxim Uvarov wrote:
When I added that print I expected that person who executed test for
specific pktio can verify that
printed mac address is the same as read pktio (device).
Hrm, I've never done that.
With rem
On 08.09.15 11:35, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: ext Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
Sent: Tuesday, September 08, 2015 10:26 AM
To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
Subject: Re: [lng-odp] [API-NE
> -Original Message-
> From: ext Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org]
> Sent: Tuesday, September 08, 2015 10:26 AM
> To: Savolainen, Petri (Nokia - FI/Espoo); lng-odp@lists.linaro.org
> Subject: Re: [lng-odp] [API-NEXT PATCH 3/5] linux-generic: cpu: rename
> time_cycles to
Renamed time_cycles files to cpu_cycles, since those files
implement cpu cycle counter read needed by cpu API but not
needed by new time API.
Signed-off-by: Petri Savolainen
---
platform/linux-generic/Makefile.am | 8 ++---
platform/linux-generic/arch/linux/odp_cpu_cycles.c | 38
Use CPU cycle count API instead of time API to measure CPU
cycles.
Signed-off-by: Petri Savolainen
---
test/performance/odp_scheduling.c | 147 ++
1 file changed, 68 insertions(+), 79 deletions(-)
diff --git a/test/performance/odp_scheduling.c
b/test/perform
Implemented odp_cpu_cycles() be renaming odp_time_cycles()
implementation. Time implementation uses odp_cpu_cycles
temporarely.
Signed-off-by: Petri Savolainen
---
platform/linux-generic/arch/linux/odp_cpu_cycles.c | 4 ++--
platform/linux-generic/arch/mips64/odp_cpu_cycles.c | 4 ++--
platform
Added implementation for CPU cycle diff, max and resolution.
Signed-off-by: Petri Savolainen
---
platform/linux-generic/Makefile.am | 1 +
platform/linux-generic/arch/linux/odp_cpu_cycles.c | 10 ++
platform/linux-generic/arch/mips64/odp_cpu_cycles.c | 10 ++
p
Raw CPU cycle counts can be used to measure performance
in CPU cycles. These functions will replace some usage of
odp_time_cycles() of odp_time_diff_cycles().
Signed-off-by: Petri Savolainen
---
include/odp/api/cpu.h | 52 +++
1 file changed, 52 in
Raw CPU cycle count API is needed for measuring CPU cycle consumption. This can
replace many of the odp_time_cycles() when time API moves to real time from
CPU cycles.
v2:
* corrected comparison in diff
* code style changes in patch 5
Petri Savolainen (5):
api: cpu: added cpu cycle count
> -Original Message-
> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
> ext Ivan Khoronzhuk
> Sent: Monday, September 07, 2015 7:15 PM
> To: lng-odp@lists.linaro.org
> Subject: Re: [lng-odp] [API-NEXT PATCH 5/5] performance: sched: update
> scheduling test to use cyc
On 07.09.15 15:41, Petri Savolainen wrote:
Implemented odp_cpu_cycles() be renaming odp_time_cycles()
implementation. Time implementation uses odp_cpu_cycles
temporarely.
Right. You didn't add functional changes here.
Currently It's not documented that odp_time_cycles returns global cycles ).
65 matches
Mail list logo