Re: [lng-odp] [API-NEXT PATCH v2 1/6] api: pktio: add config options for pktin timestamping

Merged, Maxim. On 04/01/16 20:00, Balasubramanian Manoharan wrote: For the series: Reviewed-by: Balasubramanian Manoharan On 31/03/16 9:41 PM, Petri Savolainen wrote: Added API calls and definitions for pktio interface level configuration options. Timestamping is

[lng-odp] [Bug 2159] Compiling with gcc flag -Og fails

https://bugs.linaro.org/show_bug.cgi?id=2159 Bill Fischofer changed: What|Removed |Added Ever confirmed|0 |1

[lng-odp] [PATCH] linux-generic: queue: avoid false positive when compiling with -Og

Resolve Bug https://bugs.linaro.org/show_bug.cgi?id=2159 by adding an extraneous variable initialization to avoid a false positive error when compiling with gcc -Og Signed-off-by: Bill Fischofer --- platform/linux-generic/odp_queue.c | 2 +- 1 file changed, 1

Re: [lng-odp] [PATCH 1/2] configure: support out-of-tree CUnit

On 04/01 22:15:34, Anders Roxell wrote: > On 25 March 2016 at 20:25, Brian Brooks wrote: > > If --with-cunit-path=DIR is used, skip the AC_CHECK_LIB and modify the > > linker > > flags accordingly. > > Why should we skip the AC_CHECK_LIB and AC_CHECK_HEADERS step >

[lng-odp] [PATCHv2] configure: remove separate so_version file

Set versioning in configure.ac directly, folks familiar with autotools will expect it to be set there. Suggested-by: Fathi Boudra Signed-off-by: Anders Roxell --- .so_version | 1 - Makefile.am | 2 +- configure.ac | 10

[lng-odp] [PATCH] configure: reorder m4_includes

Reorder the m4_include due to LIBS get contaminated while checking for libraries. Configure error: checking for GCC atomic builtins... no GCC-style __atomic builtins not supported by the compiler. Use newer version. For gcc > 4.7.0 Suggested-by: Ricardo Salveti

Re: [lng-odp] [PATCH 1/2] configure: support out-of-tree CUnit

On 25 March 2016 at 20:25, Brian Brooks wrote: > If --with-cunit-path=DIR is used, skip the AC_CHECK_LIB and modify the linker > flags accordingly. Why should we skip the AC_CHECK_LIB and AC_CHECK_HEADERS step don't feel right. > > Signed-off-by: Brian Brooks

Re: [lng-odp] [PATCH 2/2] configure: remove dead code

On 25 March 2016 at 20:25, Brian Brooks wrote: > Remove dead code which potentially acted as a workaround for making > AC_CHECK_LIB to work with --with-cunit-path. Workaround you may say that. I would put it, one way to do checks for libs and headers that is in the path

Re: [lng-odp] [PATCH] configure: remove separate so_version file

On Fri, Apr 1, 2016 at 12:47 PM, Maxim Uvarov wrote: > On 04/01/16 13:38, Anders Roxell wrote: >> >> Set versioning in configure.ac directly, folks familiar with autotools >> will expect it to be set there. >> >> Suggested-by: Fathi Boudra >>

[lng-odp] [Bug 2160] Building natively on AARch64 fails

https://bugs.linaro.org/show_bug.cgi?id=2160 Mike Holmes changed: What|Removed |Added Assignee|anders.rox...@linaro.org|lng-odp@lists.linaro.org

[lng-odp] [Bug 2160] New: Building natively on AARch64 fails

https://bugs.linaro.org/show_bug.cgi?id=2160 Bug ID: 2160 Summary: Building natively on AARch64 fails Product: OpenDataPlane - linux- generic reference Version: 1.8 Hardware: Other OS: Linux Status: UNCONFIRMED

[lng-odp] [Bug 2159] New: Compiling with gcc flag -Og fails

https://bugs.linaro.org/show_bug.cgi?id=2159 Bug ID: 2159 Summary: Compiling with gcc flag -Og fails Product: OpenDataPlane - linux- generic reference Version: 1.8 Hardware: Other OS: Linux Status: UNCONFIRMED

Re: [lng-odp] [API-NEXT PATCHv7 0/5] fix BUG 2027 & add init cpumasks

that is applied. Maxim. On 03/31/16 16:08, Bill Fischofer wrote: This patch series replaces the patches previously submitted with subjects: ODP API: add control/worker cpumasks to init data -and- linux-generic: add support for initial cpumasks The addition of the ability for ODP to accept

Re: [lng-odp] [API-NEXT PATCH v2 1/6] api: pktio: add config options for pktin timestamping

For the series: Reviewed-by: Balasubramanian Manoharan On 31/03/16 9:41 PM, Petri Savolainen wrote: Added API calls and definitions for pktio interface level configuration options. Timestamping is one of those options. Added definitions to query and enable packet

Re: [lng-odp] flush_in_queues() on devices never started

I've sent a patch to fix that. I've also found that this error handling could be necessary. start() has finished, so I think we should call stop(). And the state should be either STOPPED or OPENED, I'm not sure in that. @@ -395,6 +396,11 @@ int odp_pktio_start(odp_pktio_t id)

[lng-odp] [PATCH 1/2] configure: support out-of-tree CUnit

If --with-cunit-path=DIR is used, skip the AC_CHECK_LIB and modify the linker flags accordingly. Signed-off-by: Brian Brooks --- test/m4/validation.m4 | 8 ++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/test/m4/validation.m4

[lng-odp] [PATCH 2/2] configure: remove dead code

Remove dead code which potentially acted as a workaround for making AC_CHECK_LIB to work with --with-cunit-path. Signed-off-by: Brian Brooks --- configure.ac | 14 -- 1 file changed, 14 deletions(-) diff --git a/configure.ac b/configure.ac index

[lng-odp] [PATCH] linux-generic: packet_io: fix state handling

The current two state doesn't tell odp_pktio_close() whether the device has ever been started, which is important to decide whether it should flush anything or not. The new OPENED state only happen after odp_pktio_open(), after a start -> stop transition it goes to a STOPPED state. Both valid for

Re: [lng-odp] [PATCH] example: packet: add odp_term_global

On 1 April 2016 at 12:12, Maxim Uvarov wrote: > On 04/01/16 19:00, Mike Holmes wrote: > >> How does this example terminate in a "make check" situation ? >> > > we do not run it in make check > We should reinforce that it is unusual for our examples to run indefinitely

Re: [lng-odp] [PATCH] example: packet: add odp_term_global

On 04/01/16 19:00, Mike Holmes wrote: How does this example terminate in a "make check" situation ? we do not run it in make check Maxim. On 1 April 2016 at 11:53, Maxim Uvarov > wrote: ping. Any feedback is valuable.

Re: [lng-odp] [API-NEXT PATCHv4] api: packet: add support for multi-segment push/pull operations

ping . review is needed. Maxim. On 03/22/16 14:27, Bill Fischofer wrote: Add support for multi-segment push/pull operations for Monarch compliance. If a push for more than the available headroom/tailroom is requested, then allocate additional head/tail segments if possible to complete the

Re: [lng-odp] [PATCH] linux-generic: pktio: avoid coverity issues by adding explicit cast

On 03/27/16 00:21, Bill Fischofer wrote: Resolve bug https://bugs.linaro.org/show_bug.cgi?id=2138 by adding an explicit cast to odp_pktio_capability(). This routine cannot return a bad RC since the error condition is already guarded earlier in the code, however coverity doesn't see this.

Re: [lng-odp] [PATCH] example: packet: add odp_term_global

How does this example terminate in a "make check" situation ? On 1 April 2016 at 11:53, Maxim Uvarov wrote: > ping. Any feedback is valuable. > > Maxim. > > > On 03/28/16 18:36, Maxim Uvarov wrote: > >> Add odp_term_global and comment that it will never >> be called. >>

Re: [lng-odp] [PATCH] linux-generic: pktio: handle transient output queue nonempty conditions

On 03/27/16 03:05, Bill Fischofer wrote: Out queues are normally empty but any non-empty state should be very transient. Retry odp_queue_destroy() operation when performing destroy_out_queues() function to handle such cases. This addresses bug https://bugs.linaro.org/show_bug.cgi?id=2089

Re: [lng-odp] [API-NEXT PATCHv2] doc: application: add missing .gitignore file for doc output

that was applied, probably I did not wrote applied message. Maxim. On 03/27/16 19:10, Bill Fischofer wrote: Signed-off-by: Bill Fischofer --- doc/application-api-guide/.gitignore | 1 + 1 file changed, 1 insertion(+) create mode 100644

Re: [lng-odp] [PATCH] example: packet: add odp_term_global

ping. Any feedback is valuable. Maxim. On 03/28/16 18:36, Maxim Uvarov wrote: Add odp_term_global and comment that it will never be called. https://bugs.linaro.org/show_bug.cgi?id=1706 Signed-off-by: Maxim Uvarov --- example/packet/odp_pktio.c | 9 - 1

Re: [lng-odp] [PATCH] configure: remove separate so_version file

On 04/01/16 13:38, Anders Roxell wrote: Set versioning in configure.ac directly, folks familiar with autotools will expect it to be set there. Suggested-by: Fathi Boudra Signed-off-by: Anders Roxell --- .so_version | 1 - Makefile.am

Re: [lng-odp] [PATCHv2] linux-generic: test: shmem: atomic check+open fifo

On 03/30/16 10:54, Christophe Milard wrote: Fix for bug 2146, CID 159395: The open system call is directely used to check the presence of the fifo and open it at the same time. Signed-off-by: Christophe Milard --- since v1: changed loop to avoid open() line

Re: [lng-odp] [API-NEXT PATCHv7 0/5] fix BUG 2027 & add init cpumasks

Merged, Maxim. On 03/31/16 16:08, Bill Fischofer wrote: This patch series replaces the patches previously submitted with subjects: ODP API: add control/worker cpumasks to init data -and- linux-generic: add support for initial cpumasks The addition of the ability for ODP to accept control and

Re: [lng-odp] [PATCHv5 0/2] scripts: add builddpdk

Merged :) Maxim. On 03/31/16 19:31, Mike Holmes wrote: On 31 March 2016 at 11:44, Maxim Uvarov > wrote: On 03/29/16 15:20, Mike Holmes wrote: Were you able to test this on ARM64 ? The only other thing I can

Re: [lng-odp] [PATCH 0/2] linux-generic: pool: add more debug print for pool_destroy

Thanks, Merged. On 04/01/16 09:07, Balasubramanian Manoharan wrote: For the series: Reviewed-by: Balasubramanian Manoharan On Tuesday 29 March 2016 07:29 PM, Maxim Uvarov wrote: Maxim Uvarov (2): linux-generic: pool: add more debug print for pool_destroy

[lng-odp] [Bug 2158] CID 159488: Uninitialized variables: shmem_odp.c

https://bugs.linaro.org/show_bug.cgi?id=2158 Mike Holmes changed: What|Removed |Added Assignee|lng-odp@lists.linaro.org

[lng-odp] [Bug 2158] New: CID 159488: Uninitialized variables: shmem_odp.c

https://bugs.linaro.org/show_bug.cgi?id=2158 Bug ID: 2158 Summary: CID 159488: Uninitialized variables: shmem_odp.c Product: OpenDataPlane - linux- generic reference Version: 1.8 Hardware: Other OS: Linux Status:

[lng-odp] [PATCHv2] linux-generic: test: shmem: close fifo

Fixes: https://bugs.linaro.org/show_bug.cgi?id=2147 (CID 159394) The fifo is closed at end of test Signed-off-by: Christophe Milard --- since v1: bug URL added in commit msg (Anders) platform/linux-generic/test/shmem/shmem_odp.c | 1 + 1 file changed, 1

[lng-odp] [PATCHv3] linux-generic: test: shmem: atomic check+open fifo

Fixes: https://bugs.linaro.org/show_bug.cgi?id=2146 (CID 159395) The open system call is directely used to check the presence of the fifo and open it at the same time. Signed-off-by: Christophe Milard --- since v2: bug URL added (Anders) since v1: changed loop to

Re: [lng-odp] flush_in_queues() on devices never started

Hi, The issue is in odp-linux implementation of odp_pktio_close(). It should try to flush only if the interface was ever started (there would be something to flush). We'd need to extend pktio interface states so that close call can tell the difference of open() -> close() (state ==

[lng-odp] [PATCH] linux-generic: test: shmem: close fifo

Fix for bug 2147, CID 159394: The fifo is closed at end of test Signed-off-by: Christophe Milard --- platform/linux-generic/test/shmem/shmem_odp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/platform/linux-generic/test/shmem/shmem_odp.c

Re: [lng-odp] [PATCH 0/2] linux-generic: pool: add more debug print for pool_destroy

For the series: Reviewed-by: Balasubramanian Manoharan On Tuesday 29 March 2016 07:29 PM, Maxim Uvarov wrote: Maxim Uvarov (2): linux-generic: pool: add more debug print for pool_destroy example: classifier: add termination path