From doc: "A return value of 0 meansinactive,
anything else is active.".
Signed-off-by: Maxim Uvarov
---
test/validation/system/system.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/test/validation/system/system.c b/test/validation/system/system.c
index 5b5a3f0..5d
Since scmversion gets populated during the configure process, the only
way for the cache to be updated through commits is by either running
make distclean or by forcing autoconf updates when bootstrap gets called
(e.g. builddeb).
Signed-off-by: Ricardo Salveti
---
bootstrap | 2 +-
1 file change
There is no single OpenDataPlane product anymore, so pointing out to the
generic implementation instead.
Signed-off-by: Ricardo Salveti
---
README | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/README b/README
index cad5138..d8583be 100644
--- a/README
+++ b/README
@@ -45,4
Allow the user to easily produce and use the source package when the
binary build is not desired (e.g. to be uploaded in extra
repositories).
Signed-off-by: Ricardo Salveti
---
scripts/builddeb | 10 +-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/scripts/builddeb b/scrip
OK, thanks. After applying that prereq series this series looks OK, so for
it:
Reviewed-and-tested-by: Bill Fischofer
However, as I noted on it's thread, the odp_cache_line_size patch
introduces a doxygen error that should be corrected before that series can
be applied.
On Tue, Feb 2, 2016 at
After applying this patch there are doxygen issues:
bill@Ubuntu15:~/linaro/hongbo$ make doxygen-html
rm -rf doc/output
make --directory=./doc/images
make[1]: Entering directory '/home/bill/linaro/hongbo/doc/images'
mscgen -T svg -i resource_management.msc -o resource_management.svg
make[1]: Leavin
Now is in CI for api-next we get following error:
Test: system_test_odp_cpu_hz ...FAILED
1. ../../../../test/validation/system/system.c:235 - hz > 1 * KILO_HZ
(too late to understand roots today).
Maxim.
On 02/03/2016 12:37, Hongbo Zhang wrote:
On 3 February 2016 at 17:13, Maxim Uvarov
Switch timer to use SIGEV_THREAD_ID instead of SIGEV_THREAD.
I.e. do not start timer handle thread on each timer action.
Start timer handle manually and wait for signal there.
This patch also fixes nasty bug with hanging timer threads,
which wants to access to timer pool and free shm on pool
destro
For the v5 series:
Reviewed-and-tested-by: Bill Fischofer
On Wed, Feb 3, 2016 at 11:45 AM, Balasubramanian Manoharan <
bala.manoha...@linaro.org> wrote:
> Packet match rule creation is modified to include source and destination
> class of service. Removes the ability to add any class of service
New version sent.
Thanks for such quick reviews Bill. :)
Regards,
Bala
On 3 February 2016 at 23:12, Bala Manoharan wrote:
> My mistake. I am sending v5 shortly.
>
> Regards,
> Bala
>
>
> On 3 February 2016 at 22:51, Bill Fischofer wrote:
>> This now applies and runs, however there are doxygen
Adds validation suite for packet match rule create functions.
Signed-off-by: Balasubramanian Manoharan
---
test/validation/classification/classification.h| 1 -
.../classification/odp_classification_basic.c | 141 +++-
.../classification/odp_classification_common.c | 26 +
Changes for packet match rule create function.
Signed-off-by: Balasubramanian Manoharan
---
example/classifier/odp_classifier.c| 38 --
test/validation/classification/classification.h| 4 +--
.../classification/odp_classification_basic.c | 14 ---
Implements packet match rule create functions.
If a packet match rule needs to be applied at the pktio level it has to be
configured to the default class of service.
Signed-off-by: Balasubramanian Manoharan
---
.../include/odp/plat/classification_types.h| 9 -
.../include/odp_classific
Packet match rule creation is modified to include source and destination
class of service. Removes the ability to add any class of service directly
with pktio. If a PMR needs to be applied at the pktio level the same
should be applied to default class of service.
Packet match rule destroy function
Packet match rule creation is modified to include source and destination
class of service. Removes the ability to add any class of service directly
with pktio. If a PMR needs to be applied at the pktio level the same
should be applied to default class of service.
Packet match rule destroy function
My mistake. I am sending v5 shortly.
Regards,
Bala
On 3 February 2016 at 22:51, Bill Fischofer wrote:
> This now applies and runs, however there are doxygen issues:
>
> bill@Ubuntu15:~/linaro/balapmr$ make doxygen-html
> rm -rf doc/output
> make --directory=./doc/images
> make[1]: Entering dire
This now applies and runs, however there are doxygen issues:
bill@Ubuntu15:~/linaro/balapmr$ make doxygen-html
rm -rf doc/output
make --directory=./doc/images
make[1]: Entering directory '/home/bill/linaro/balapmr/doc/images'
mscgen -T svg -i resource_management.msc -o resource_management.svg
make
Adds validation suite for packet match rule create functions.
Signed-off-by: Balasubramanian Manoharan
---
test/validation/classification/classification.h| 1 -
.../classification/odp_classification_basic.c | 141 +++-
.../classification/odp_classification_common.c | 26 +
Changes for packet match rule create function.
Signed-off-by: Balasubramanian Manoharan
---
example/classifier/odp_classifier.c | 38 -
1 file changed, 21 insertions(+), 17 deletions(-)
diff --git a/example/classifier/odp_classifier.c
b/example/classifier/od
Implements packet match rule create functions.
If a packet match rule needs to be applied at the pktio level it has to be
configured to the default class of service.
Signed-off-by: Balasubramanian Manoharan
---
.../include/odp_classification_datamodel.h | 48 +---
.../include/odp_classi
Packet match rule creation is modified to include source and destination
class of service. Removes the ability to add any class of service directly
with pktio. If a PMR needs to be applied at the pktio level the same
should be applied to default class of service.
Packet match rule destroy function
Packet match rule creation is modified to include source and destination
class of service. Removes the ability to add any class of service directly
with pktio. If a PMR needs to be applied at the pktio level the same
should be applied to default class of service.
Packet match rule destroy function
Merged to api-next.
But test still fails.
Maxim.
On 02/03/2016 17:54, Savolainen, Petri (Nokia - FI/Espoo) wrote:
Also this 2/2
Reviewed-by: Petri Savolainen
-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT
Maxim Uvarov
Sent: Wednesday, F
I just confirmed I get the same behavior if I apply and run the patch
against api-next.
On Wed, Feb 3, 2016 at 11:04 AM, Bill Fischofer
wrote:
> OK, however it was marked PATCH and addressed a bug, not an API change, so
> I assumed it was intended for master.
>
> On Wed, Feb 3, 2016 at 11:03 AM,
OK, however it was marked PATCH and addressed a bug, not an API change, so
I assumed it was intended for master.
On Wed, Feb 3, 2016 at 11:03 AM, Maxim Uvarov
wrote:
> btw, it's for api-next, I did not test it on master. But I think result
> will be the same.
>
> Maxim.
>
> On 02/03/2016 18:48,
btw, it's for api-next, I did not test it on master. But I think result
will be the same.
Maxim.
On 02/03/2016 18:48, Bill Fischofer wrote:
I'm not sure this is working properly. Running
examples/timer/odp_timer_test without this patch I see this sort of
output:
bill@Ubuntu15:~/linaro/odp/
Ok, I'll review/test that as soon as it gets posted. Thanks.
On Wed, Feb 3, 2016 at 10:59 AM, Bala Manoharan
wrote:
> Thanks Bill.
> Looks like the latest merge in api-next has created conflict in my patch.
> I will rebase to HEAD and post V4.
>
> Regards,
> Bala
>
>
> On 3 February 2016 at 21:
Thanks Bill.
Looks like the latest merge in api-next has created conflict in my patch.
I will rebase to HEAD and post V4.
Regards,
Bala
On 3 February 2016 at 21:37, Bill Fischofer wrote:
> Part 3 of this patch fails to apply to the head of api-next for me:
>
> bill@Ubuntu15:~/linaro/balapmr$ gi
On 02/03/2016 18:48, Bill Fischofer wrote:
Alarm clock
You should not see that. It has to be masked. I guess under gdb test passes.
Will check what went wrong. Might be deleted something on patch clean up.
Maxim.
___
lng-odp mailing list
lng-odp@list
Part 3 of this patch fails to apply to the head of api-next for me:
bill@Ubuntu15:~/linaro/balapmr$ git am --reject ~/Mail/Incoming/Bala/7
Applying: validation: classification: adds validation suite for pmr create
api
Checking patch test/validation/classification/classification.h...
Checking patch
I'm not sure this is working properly. Running
examples/timer/odp_timer_test without this patch I see this sort of output:
bill@Ubuntu15:~/linaro/odp/example/timer$ ./odp_timer_test
ODP timer example starts
odp_pool.c:103:odp_pool_init_global():
Pool init global
odp_pool.c:104:odp_pool_init_glob
Also this 2/2
Reviewed-by: Petri Savolainen
> -Original Message-
> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT
> Maxim Uvarov
> Sent: Wednesday, February 03, 2016 4:46 PM
> To: lng-odp@lists.linaro.org
> Subject: [lng-odp] [API-NEXT PATCHv3 2/2] validation:
Signed-off-by: Maxim Uvarov
---
test/validation/system/system.c | 31 +--
test/validation/system/system.h | 1 +
2 files changed, 30 insertions(+), 2 deletions(-)
diff --git a/test/validation/system/system.c b/test/validation/system/system.c
index 942905a..5b5a3f0 10
Do not fail if odp_cpu_hz() returns 0, just skip the test.
Suggested-by: Petri Savolainen
Reviewed-by: Petri Savolainen
Signed-off-by: Maxim Uvarov
---
test/validation/system/system.c | 22 ++
test/validation/system/system.h | 4
2 files changed, 22 insertions(+), 4 d
Hi all,
Do we need this patch set?
This is for bug https://bugs.linaro.org/show_bug.cgi?id=1881
On 29 January 2016 at 19:22, wrote:
> From: Hongbo Zhang
>
> v1-v2 change:
>
> Add macro __OCTEON for patch 2.
>
> v1 notes:
>
> This is for bug https://bugs.linaro.org/show_bug.cgi?id=1881
>
> This
Patch itself is ok. I merged this patch as it fixes current problem.
Then will work on Stuart patches and probably we can move macs to other
place.
Maxim.
On 01/22/2016 20:22, Stuart Haslam wrote:
On Thu, Jan 14, 2016 at 12:56:31PM +0200, Ivan Khoronzhuk wrote:
In case of one loop pktio the
On 02/03/2016 17:21, Hongbo Zhang wrote:
Maxim,
you only handled system_test_odp_cpu_hz(), you should do the same for
system_test_odp_cpu_hz_id()
ok, if it's the same issue will do it.
Maxim.
On 3 February 2016 at 21:55, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
-Original Message--
Please wait,
As I replied in the v1 patch, system_test_odp_cpu_hz_id() should be
done with the same solution.
On 3 February 2016 at 22:22, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
> Reviewed-by: Petri Savolainen
>
>> -Original Message-
>> From: lng-odp [mailto:lng-odp-boun...@lists.l
Reviewed-by: Petri Savolainen
> -Original Message-
> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT
> Maxim Uvarov
> Sent: Wednesday, February 03, 2016 4:10 PM
> To: lng-odp@lists.linaro.org
> Subject: [lng-odp] [API-NEXT PATCHv2] validation: system: make odp_cpu
Maxim,
you only handled system_test_odp_cpu_hz(), you should do the same for
system_test_odp_cpu_hz_id()
On 3 February 2016 at 21:55, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
>
>
>> -Original Message-
>> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT
>> Maxim
Merged,
Maxim.
On 02/03/2016 16:01, Mike Holmes wrote:
Ok, thanks
On 3 February 2016 at 07:59, Elo, Matias (Nokia - FI/Espoo)
mailto:matias@nokia.com>> wrote:
The l2fwd application in api-next is already ported to use the new
multi-queue pktio api. This fix is relevant to that ve
Do not fail if odp_cpu_hz() returns 0, just skip the test.
Suggested-by: Petri Savolainen
Signed-off-by: Maxim Uvarov
---
test/validation/system/system.c | 22 ++
test/validation/system/system.h | 4
2 files changed, 22 insertions(+), 4 deletions(-)
diff --git a/test/
> -Original Message-
> From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT
> Maxim Uvarov
> Sent: Wednesday, February 03, 2016 3:42 PM
> To: lng-odp@lists.linaro.org
> Subject: [lng-odp] [API-NEXT PATCH] validation: system: make odp_cpu_hz
> optional in validation tes
Not that thread. Please skip previous comment. But it's not needed to
resend it.
Maxim.
On 02/03/2016 16:37, Savolainen, Petri (Nokia - FI/Espoo) wrote:
Similar to the patch from Matias, this would not likely apply to
master since the code base is substantially modified (functions
re-organi
No need to resend it. I will take that to 1.7 as bug fix.
Maxim.
On 02/03/2016 16:37, Savolainen, Petri (Nokia - FI/Espoo) wrote:
Similar to the patch from Matias, this would not likely apply to
master since the code base is substantially modified (functions
re-organized, etc). It’s a bonus
Do not fail if odp_cpu_hz() returns 0, just skip the test.
Suggested-by: Petri Savolainen
Signed-off-by: Maxim Uvarov
---
test/validation/system/system.c | 22 ++
test/validation/system/system.h | 4
2 files changed, 22 insertions(+), 4 deletions(-)
diff --git a/test/
Hi,
I added odp_dpdk_pktio_init_local() function, which is called in
odp_init_local(). It does the same ('RTE_PER_LCORE(_lcore_id) = cpu;') what is
done in odp_thread_init_local() in odp-dpdk.
-Matias
> -Original Message-
> From: EXT Zoltan Kiss [mailto:zoltan.k...@linaro.org]
> Sent:
Similar to the patch from Matias, this would not likely apply to master since
the code base is substantially modified (functions re-organized, etc). It’s a
bonus if it applies on both, but api-next should be the main target.
-Petri
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Beh
On 02/03/2016 16:22, Hongbo Zhang wrote:
OK.
Will follow Petri's suggestion.
But will send patch out my next morning, I don't have a Linux PC at
home right now.
Thank you all.
I will send a patch. We need it ASAP.
Maxim.
On 3 February 2016 at 19:31, Maxim Uvarov wrote:
On 02/03/2016 14:
Hi,
I haven't checked it yet, but how does your code handles setting up
lcore id's?
Zoli
On 29/01/16 11:14, Matias Elo wrote:
V2:
- Check the number of mbuf segments in mbuf_to_pkt() (Zoltan Kiss)
- Copy DPDK RSS hash to ODP packet header in mbuf_to_pkt() (Zoltan Kiss)
- Compare packet lengt
OK.
Will follow Petri's suggestion.
But will send patch out my next morning, I don't have a Linux PC at
home right now.
Thank you all.
On 3 February 2016 at 19:31, Maxim Uvarov wrote:
> On 02/03/2016 14:23, Savolainen, Petri (Nokia - FI/Espoo) wrote:
>>
>> I still think that the test should pass
On 29/01/16 12:30, Savolainen, Petri (Nokia - FI/Espoo) wrote:
-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT
Zoltan Kiss
Sent: Friday, January 29, 2016 2:13 PM
To: Elo, Matias (Nokia - FI/Espoo); EXT Ola Liljedahl
Cc: lng-odp@lists.linaro
Ok, should I submit a new version or can you apply the patch straight to master?
I used api-next when developing this because it has the netmap multi queue
support.
From: EXT Mike Holmes [mailto:mike.hol...@linaro.org]
Sent: Wednesday, February 03, 2016 2:51 PM
To: Tilli, Juha-Matti (Nokia - FI/
Merged,
Maxim.
On 02/02/2016 23:44, Bill Fischofer wrote:
Add missing validation test to resolve bug
https://bugs.linaro.org/show_bug.cgi?id=2017
Signed-off-by: Bill Fischofer
---
test/validation/queue/queue.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/test/validation/queue/queue
needs update tp head.
But with update make check under root fails for tap and normal pktio
somewhere in termination path.
We need to fix it.
Maxim.
On 02/03/2016 03:10, Bill Fischofer wrote:
Add coverage for the following new multi-queue pktio APIs:
- odp_pktio_input_queue_param_init()
- odp
Agree that this is not a perfect tool, it is based on gcov and you do have
to check manually, it also may still have bugs.
I will be extending this tool to grep for the apis in the test directory to
highlight potential mismatch for inspection
But as with the bug just fixed yesterday - not committ
On 3 February 2016 at 16:02, Savolainen, Petri (Nokia - FI/Espoo) <
petri.savolai...@nokia.com> wrote:
>
>
>
>
> *From:* lng-odp [mailto:lng-odp-boun...@lists.linaro.org] *On Behalf Of *EXT
> Mike Holmes
> *Sent:* Wednesday, February 03, 2016 2:48 PM
> *To:* lng-odp
> *Subject:* [lng-odp] ODP next
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT Mike
Holmes
Sent: Wednesday, February 03, 2016 2:48 PM
To: lng-odp
Subject: [lng-odp] ODP next branch test coverage status for v1.7
GIT_BRANCH=next UNTESTED=1 ./build.sh
Untested Public API
odp_atomic_fetch_add_u64
stati
Ok, thanks
On 3 February 2016 at 07:59, Elo, Matias (Nokia - FI/Espoo) <
matias@nokia.com> wrote:
> The l2fwd application in api-next is already ported to use the new
> multi-queue pktio api. This fix is relevant to that version only.
>
>
>
> -Matias
>
>
>
> *From:* EXT Mike Holmes [mailto:mi
The l2fwd application in api-next is already ported to use the new multi-queue
pktio api. This fix is relevant to that version only.
-Matias
From: EXT Mike Holmes [mailto:mike.hol...@linaro.org]
Sent: Wednesday, February 03, 2016 2:53 PM
To: Elo, Matias (Nokia - FI/Espoo)
Cc: lng-odp@lists.lina
This can go straight to master, there is no ODP api change.
On 3 February 2016 at 02:51, Elo, Matias (Nokia - FI/Espoo) <
matias@nokia.com> wrote:
> ping
>
> > -Original Message-
> > From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of
> EXT Matias
> > Elo
> > Sent: Th
This can go straight to linux-generic becasue it does not touch the API.
On 3 February 2016 at 04:06, Juha-Matti Tilli
wrote:
> Added option (--dst_addr or -r) to l2fwd that allows the user to specify
> what destination MAC addresses to use per port. The MAC addresses are
> separated by commas a
GIT_BRANCH=next UNTESTED=1 ./build.sh
Untested Public API
odp_atomic_fetch_add_u64
odp_atomic_sub_u32
odp_be_to_cpu_32
odp_cos_drop
odp_cos_with_l3_qos
odp_cpu_pause
odp_override_abort
odp_pktio_capability
odp_pktio_input_queue_param_init
odp_pktio_input_queues_config
odp_pktio_in_queues
odp_pktio
Merged,
Maxim.
On 02/02/2016 16:01, Bill Fischofer wrote:
i agree. Best to get these changes out sooner rather than later.
On Tue, Feb 2, 2016 at 6:31 AM, Savolainen, Petri (Nokia - FI/Espoo)
mailto:petri.savolai...@nokia.com>> wrote:
It would be great to get it in v1.7. Otherwise, mast
Merged.
Maxim.
On 02/02/2016 19:13, Savolainen, Petri (Nokia - FI/Espoo) wrote:
Reviewed-by: Petri Savolainen
-Original Message-
From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of EXT
Maxim Uvarov
Sent: Tuesday, February 02, 2016 6:08 PM
To: lng-odp@lists.linaro.or
On 02/03/2016 14:23, Savolainen, Petri (Nokia - FI/Espoo) wrote:
I still think that the test should pass with a warning when hz == 0, and
otherwise check that hz inside some sane range. If / when all platforms can
return a valid (non-zero) value, we can update the test to fail on hz == 0.
Toda
I still think that the test should pass with a warning when hz == 0, and
otherwise check that hz inside some sane range. If / when all platforms can
return a valid (non-zero) value, we can update the test to fail on hz == 0.
Today, the main thing is to call the function at least once (and get v1
On 02/03/2016 13:16, Hongbo Zhang wrote:
I mean, use ODP_TEST_INFO_INACTIVE currently, and when everything
becomes ready, we will come back to use ODP_TEST_INFO again.
On 3 February 2016 at 17:56, Hongbo Zhang wrote:
I prefer ODP_TEST_INFO_INACTIVE at this stage, agree?
I don't agree. Accept
I send this out today before going home, when at home I won't have a
Linux PC to work.
On 3 February 2016 at 18:48, wrote:
> From: Hongbo Zhang
>
> This patch disable validation of CPU current frequency because it isn't
> fully implemented on every platform, then unnecesary failures are reporte
From: Hongbo Zhang
This patch disable validation of CPU current frequency because it isn't
fully implemented on every platform, then unnecesary failures are reported.
This validation can be re-enable again in furture if every thing is ready.
Signed-off-by: Hongbo Zhang
---
test/validation/syst
I mean, use ODP_TEST_INFO_INACTIVE currently, and when everything
becomes ready, we will come back to use ODP_TEST_INFO again.
On 3 February 2016 at 17:56, Hongbo Zhang wrote:
> I prefer ODP_TEST_INFO_INACTIVE at this stage, agree?
>
> And after 1.7, there still be something to be updated, for ex
This is a bug fix for netmat pktio MTU size. Valid packets
were being dropped in netmap_send() due to not taking into
account the ethernet header length.
Signed-off-by: Matias Elo
---
platform/linux-generic/pktio/netmap.c | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a
I prefer ODP_TEST_INFO_INACTIVE at this stage, agree?
And after 1.7, there still be something to be updated, for example,
ARM max freq is still using dummy data, and I still think the
/sys/devices/system/cpu/ should be the first place for getting cpu
freq, it is a common way of cpu freq, from here
On 3 February 2016 at 17:27, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
> The test should look like this. Zero is acceptable, otherwise hz value should
> be in a sane range.
>
> void system_test_odp_cpu_hz(void)
> {
> uint64_t hz;
>
> hz = odp_cpu_hz();
>
> if (hz == 0)
> pri
On 3 February 2016 at 17:13, Maxim Uvarov wrote:
>
>
> On 3 February 2016 at 11:34, Hongbo Zhang wrote:
>>
>> On 3 February 2016 at 16:08, Savolainen, Petri (Nokia - FI/Espoo)
>> wrote:
>> > I think we updated the API to return 0, when a CPU frequency (max or
>> > current) cannot be read. You ju
The test should look like this. Zero is acceptable, otherwise hz value should
be in a sane range.
void system_test_odp_cpu_hz(void)
{
uint64_t hz;
hz = odp_cpu_hz();
if (hz == 0)
print warning ("cpu_hz not supported or other failure") and return
(with skipped status?)
I had already tried this before sending patch, we will get compile warning:
system.c:222:14: error: comparison of unsigned expression >= 0 is
always true [-Werror=type-limits]
CU_ASSERT(0 <= hz);
On 3 February 2016 at 17:16, Maxim Uvarov wrote:
> but here :
> void system_test_odp_cpu_hz(void)
but here :
void system_test_odp_cpu_hz(void)
{
uint64_t hz;
hz = odp_cpu_hz();
CU_ASSERT(0 < hz);
}
it looks like simple change to 0 <= hz is right. You can add some comment.
Maxim.
On 3 February 2016 at 12:13, Maxim Uvarov wrote:
>
>
> On 3 February 2016 at 11:34, Hongbo Zhang w
On 3 February 2016 at 11:34, Hongbo Zhang wrote:
> On 3 February 2016 at 16:08, Savolainen, Petri (Nokia - FI/Espoo)
> wrote:
> > I think we updated the API to return 0, when a CPU frequency (max or
> current) cannot be read. You just need to return 0 on ARM (== "was not able
> to read current f
Added option (--dst_addr or -r) to l2fwd that allows the user to specify
what destination MAC addresses to use per port. The MAC addresses are
separated by commas and the number of MAC addresses, if given, must be
equal to the number of interfaces.
This makes it much easier to test l2fwd with TRex
Switch timer to use SIGEV_THREAD_ID instead of SIGEV_THREAD.
I.e. do not start timer handle thread on each timer action.
Start timer handle manually and wait for signal there.
This patch also fixes nasty bug with hanging timer threads,
which wants to access to timer pool and free shm on pool
destro
On 3 February 2016 at 16:08, Savolainen, Petri (Nokia - FI/Espoo)
wrote:
> I think we updated the API to return 0, when a CPU frequency (max or current)
> cannot be read. You just need to return 0 on ARM (== "was not able to read
> current frequency"). Validation test needs to accept both zero a
On 3 February 2016 at 15:29, Maxim Uvarov wrote:
> On 02/03/2016 06:45, Hongbo Zhang wrote:
>>
>> No, not fixed.
>>
>> One reason is there isn't such info in /proc/cpuinfo for ARM, another
>> reason is there isn't any user calling "current freq" interfaces.
>>
>> This "current freq" api was introd
I think we updated the API to return 0, when a CPU frequency (max or current)
cannot be read. You just need to return 0 on ARM (== "was not able to read
current frequency"). Validation test needs to accept both zero and non-zero
values.
-Petri
> -Original Message-
> From: lng-odp [ma
84 matches
Mail list logo
