[lng-odp] api-next drvshmem_main validation test failing

The api-next make check is failing for me on drvshmem validation test. Anyone else having the same problem? -Matias HW setup: Intel Haswell EP - Xeon E5-2697v3 Ubuntu 16.04 - 4.4.0-38-generic GRUB_CMDLINE_LINUX_DEFAULT="text default_hugepagesz=2M hugepagesz=2M

[lng-odp] [API-NEXT PATCH 2/5] linux-gen: schedule: fix creating event group with no name

Previously trying to create an event group with no name (=NULL) caused a segfault. Fix this and test it in the validation suite. Signed-off-by: Matias Elo --- platform/linux-generic/odp_schedule.c | 26 +++---

[lng-odp] [API-NEXT PATCH 1/5] api: improve name argument definitions in *_create() functions

The current APIs don't always define valid name argument values. Fix this by stating when NULL is a valid value and when the name string doesn't have to be unique. Signed-off-by: Matias Elo --- include/odp/api/spec/classification.h | 10 ++

[lng-odp] [API-NEXT PATCH 5/5] linux-gen: timer: fix creating timer pool with no name

Previously trying to create a timer pool with no name (=NULL) caused a segfault. Signed-off-by: Matias Elo --- platform/linux-generic/odp_timer.c | 14 ++ 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/platform/linux-generic/odp_timer.c

[lng-odp] [API-NEXT PATCH 0/5] name argument definitions in *_create() functions

This patch set improves name argument definitions in *_create() functions to clearly state when NULL is a valid value and when the name string doesn't have to be unique. The related function implementions are fixed to handle NULL name values correctly. NULL values are also tested in the

[lng-odp] [API-NEXT PATCH 4/5] linux-gen: classification: fix creating cos with no name

Previously trying to create a class-of-service with no name (=NULL) caused a segfault. Fix this and test it in the validation suite. Signed-off-by: Matias Elo --- platform/linux-generic/odp_classification.c | 11 ---

[lng-odp] [API-NEXT PATCH 3/5] linux-gen: queue: fix creating queue with no name

Previously trying to create a queue with no name (=NULL) caused a segfault. Fix this and test it in the validation suite. Signed-off-by: Matias Elo --- platform/linux-generic/odp_queue.c| 8 ++-- test/common_plat/validation/api/queue/queue.c | 10

Re: [lng-odp] [API-NEXT PATCH 1/5] api: improve name argument definitions in *_create() functions

> -Original Message- > From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of Maxim > Uvarov > Sent: Friday, October 14, 2016 12:53 PM > To: lng-odp@lists.linaro.org > Subject: Re: [lng-odp] [API-NEXT PATCH 1/5] api: improve name argument > definitions in *_create() functions

Re: [lng-odp] [API-NEXT PATCH 1/5] api: improve name argument definitions in *_create() functions

On 10/14/16 11:49, Matias Elo wrote: The current APIs don't always define valid name argument values. Fix this by stating when NULL is a valid value and when the name string doesn't have to be unique. Signed-off-by: Matias Elo --- include/odp/api/spec/classification.h |

Re: [lng-odp] [PATCH] example: generator: actually use specified default

Retested this patch again. Now I can not reproduce issue. Merged. Maxim. On 10/12/16 21:42, Mike Holmes wrote: On 22 September 2016 at 10:33, Maxim Uvarov > wrote: Reverted this patch due to have issues with

Re: [lng-odp] [PATCHv2] test: skip pktio_perf tests on 1 and 2 cpus machines

Merged. I think it will be good to commit .travis.yml into repo. So that any git push to github will run tests. Maxim. On 10/13/16 23:57, Mike Holmes wrote: On 13 October 2016 at 12:37, Maxim Uvarov > wrote: Make check should

Re: [lng-odp] [PATCHv2] test: skip pktio_perf tests on 1 and 2 cpus machines

Reverted patch. I must be more careful with review my own patches. On 10/13/16 23:57, Mike Holmes wrote: On 13 October 2016 at 12:37, Maxim Uvarov > wrote: Make check should skip the test instead of failing it. Test splits

Re: [lng-odp] [PATCH] timers: fix off by one tick in timer expiration processing

also need to put Fixes https://bugs.linaro.org/show_bug.cgi?id=2552 Into the patch description, maybe that can happen as it is pushed ? On 13 October 2016 at 17:46, Bill Fischofer wrote: > Since this is a bug fix, please open a Bug for it so that this can be >

[lng-odp] [PATCHv3] test: skip pktio_perf tests on 1 and 2 cpus machines

Make check should skip the test instead of failing it. Test splits RX and TX cores for packet processing. Core 0 bind to control thread. So running machine should have at least 2 worker threads which is not enough on 1 and 2 cpus machine. CUnit uses special value 77 to mark test as SKIPPED and not

Re: [lng-odp] [PATCH] linux-generic: ticketlock: add missing doxygen for ticketlock_inlines.h

On 22 September 2016 at 15:04, Bill Fischofer wrote: > Add the missing internal doxygen documentation for the ticketlock_inlines > functions used to accelerate odp-linux even when building with > --enable-abi-compat=yes > > Signed-off-by: Bill Fischofer

[lng-odp] [Bug 2553] New: doxygen: static_inline.h:21

https://bugs.linaro.org/show_bug.cgi?id=2553 Bug ID: 2553 Summary: doxygen: static_inline.h:21 Product: OpenDataPlane - linux- generic reference Version: v1.11.0.0 Hardware: Other OS: Linux Status: UNCONFIRMED

Re: [lng-odp] [PATCHv2] test: skip pktio_perf tests on 1 and 2 cpus machines

On 14 October 2016 at 07:02, Maxim Uvarov wrote: > Merged. > > I think it will be good to commit .travis.yml into repo. So that any git > push to github will run tests. > The travis will be pushed as a patch, a second one for coverity als. We have to alter how we do