Re: [lng-odp] [API-NEXT PATCHv4 0/6] Add HMAC-SHA-256-128 and AES128 CBC support

Reviewed-by: Petri Savolainen > -Original Message- > From: lng-odp [mailto:lng-odp-boun...@lists.linaro.org] On Behalf Of > EXT Nicolas Morey-Chaisemartin > Sent: Tuesday, November 10, 2015 5:30 PM > To: lng-odp@lists.linaro.org > Subject: [lng-odp] [API-NEXT PATCHv4 0/6] Add HMAC-SHA-25

Re: [lng-odp] [API-NEXT PATCH 1/3] api: config: add odp_config_threads() API

There's already API for this: /** * Maximum thread count * * Returns the maximum thread count, which is a constant value and set in * ODP initialization phase. * * @return Maximum thread count */ int odp_thread_count_max(void); Also new #define ODP_CONFIG_THREADS should not be added. -Pe

[lng-odp] [API-NEXT PATCHv1] helper: table: fix the doxygen warning

Signed-off-by: huanggaoyang --- helper/hashtable.c| 14 ++--- helper/include/odp/helper/table.h | 111 +- helper/lineartable.c | 9 ++-- 3 files changed, 72 insertions(+), 62 deletions(-) diff --git a/helper/hashtable.c b/helper

[lng-odp] [API-NEXT PATCHv1] helper: table:fix doxygen warning

fix the doxygen warning in table.h and update some comment. huanggaoyang (1): helper: table: fix the doxygen warning helper/hashtable.c| 14 ++--- helper/include/odp/helper/table.h | 111 +- helper/lineartable.c | 9 ++-- 3 fi

Re: [lng-odp] [PATCH] doc: improve asciidoc presentation

On 10 November 2015 at 19:11, Bill Fischofer wrote: > OK, however the patches did not apply to master or api-next, but did apply > to next. > Yes I did re-base there since we stalled master work to make a bug fix release, but I feel it is the wrong direction as I have thought more about it. >

[lng-odp] [API-NEXT PATCH 3/3] linux-generic: locks: revise implementation of recursive rwlocks

Change linux-generic implementation of rwlock_recursive types to reference the externally visible ODP_CONFIG_THREADS symbol rather than the internal _ODP_INTERNAL_MAX_THREADS symbol. This avoids causing issues with installed copies of ODP that do not export internal symbols. Signed-off-by: Bill Fi

[lng-odp] [API-NEXT PATCH 1/3] api: config: add odp_config_threads() API

Signed-off-by: Bill Fischofer --- include/odp/api/config.h | 6 ++ 1 file changed, 6 insertions(+) diff --git a/include/odp/api/config.h b/include/odp/api/config.h index c9879d6..beb69d3 100644 --- a/include/odp/api/config.h +++ b/include/odp/api/config.h @@ -32,6 +32,12 @@ extern "C" { */

[lng-odp] [API-NEXT PATCH 2/3] linux-generic: config: add implementation of odp_config_threads()

Signed-off-by: Bill Fischofer --- platform/linux-generic/include/odp/config.h | 9 + 1 file changed, 9 insertions(+) diff --git a/platform/linux-generic/include/odp/config.h b/platform/linux-generic/include/odp/config.h index ee23df3..58c55c0 100644 --- a/platform/linux-generic/include/

Re: [lng-odp] [PATCH] doc: improve asciidoc presentation

OK, however the patches did not apply to master or api-next, but did apply to next. On Tue, Nov 10, 2015 at 3:03 PM, Mike Holmes wrote: > Nothing should be targeted at next imho, next catches api next code ready > for a release. Master takes all improvements to the tests and Linux generic > alth

[lng-odp] [Bug 1894] New: make distcheck does not catch inability to build stand alone apps

https://bugs.linaro.org/show_bug.cgi?id=1894 Bug ID: 1894 Summary: make distcheck does not catch inability to build stand alone apps Product: OpenDataPlane - linux- generic reference Version: api-next Hardware: Other

Re: [lng-odp] [PATCH] doc: improve asciidoc presentation

Nothing should be targeted at next imho, next catches api next code ready for a release. Master takes all improvements to the tests and Linux generic although I think this will change when the api is its own repo with its own list. On Nov 8, 2015 4:31 PM, "Bill Fischofer" wrote: > Shouldn't thes

Re: [lng-odp] [API-NEXT PATCHv4 8/8] validation: schedule: add chaos test

No, the only difference between v3 and v4 was the fixes to the CUnit test in part 8. Thanks. On Tue, Nov 10, 2015 at 10:28 AM, Maxim Uvarov wrote: > Bill, I merged that 8/8 patch from v4. All others are from v3 which I > merged before. > > Please let me know if there something changed and I nee

[lng-odp] [PATCHv3 9/9] scripts/common_pkg_build: don't create pkg's if no repo

We shouldn't support building packages from a tarball. Suggested-by: Fathi Boudra Signed-off-by: Anders Roxell --- scripts/common_pkg_build.sh | 8 +++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/scripts/common_pkg_build.sh b/scripts/common_pkg_build.sh index cf4f6e3..eefc

[lng-odp] [PATCHv3 8/9] scripts: make rpm packages

Signed-off-by: Anders Roxell Reviewed-by: Fathi Boudra --- scripts/builddeb| 14 -- scripts/buildrpm| 30 ++ scripts/common_pkg_build.sh | 17 + 3 files changed, 51 insertions(+), 10 deletions(-) create mode 100755

[lng-odp] [PATCHv3 7/9] move debian into pkg/

Signed-off-by: Anders Roxell --- {debian => pkg/debian}/README.Debian| 0 {debian => pkg/debian}/changelog| 0 {debian => pkg/debian}/compat | 0 {debian => pkg/debian}/control | 0 {debian => pkg/debian}/copyright| 0

[lng-odp] [PATCHv3 5/9] doc: Makefile: add docs to the tarball

Signed-off-by: Anders Roxell --- doc/Makefile.am | 8 1 file changed, 8 insertions(+) diff --git a/doc/Makefile.am b/doc/Makefile.am index 73bd8e2..bb0ef44 100644 --- a/doc/Makefile.am +++ b/doc/Makefile.am @@ -6,3 +6,11 @@ endif if user_guide SUBDIRS = implementers-guide endif + +EX

[lng-odp] [PATCHv3 6/9] pkg: add odp.spec to generate rpm packages

Signed-off-by: Anders Roxell Reviewed-by: Fathi Boudra --- pkg/rpm/odp.spec | 65 1 file changed, 65 insertions(+) create mode 100644 pkg/rpm/odp.spec diff --git a/pkg/rpm/odp.spec b/pkg/rpm/odp.spec new file mode 100644 index 000..8

[lng-odp] [PATCHv3 4/9] scripts: cleanup git_hash and use it instead of odp_version.

This makes the version depend on git describe so there will be new deb packages names. Signed-off-by: Anders Roxell --- configure.ac| 2 +- platform/Makefile.inc | 2 +- scripts/Makefile.am | 2 +- scripts/builddeb| 9 - scripts/get_impl_str.sh | 11

[lng-odp] [PATCHv3 2/9] configure: set tar-pax AM_INIT_AUTOMAKE option

tar-pax is needed for long filename support Signed-off-by: Anders Roxell --- configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/configure.ac b/configure.ac index 05054df..b4fae43 100644 --- a/configure.ac +++ b/configure.ac @@ -1,6 +1,6 @@ AC_PREREQ([2.5]) AC_INIT

[lng-odp] [PATCHv3 3/9] scripts/git_hash: fix bug where dirty was always set

Signed-off-by: Anders Roxell --- scripts/git_hash.sh | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/scripts/git_hash.sh b/scripts/git_hash.sh index e15094e..fc0b057 100755 --- a/scripts/git_hash.sh +++ b/scripts/git_hash.sh @@ -9,7 +9,8 @@ ROOTDIR=${1} CUSTOM_STR=${CUSTOM_

[lng-odp] [PATCHv3 1/9] move scmversion to correct Makefile

Signed-off-by: Anders Roxell --- Makefile.am | 2 +- scripts/Makefile.am | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/Makefile.am b/Makefile.am index 6c2a9a4..1eb1a9f 100644 --- a/Makefile.am +++ b/Makefile.am @@ -13,4 +13,4 @@ SUBDIRS = @platform_with_platform@

[lng-odp] [PATCHv3 0/9] add RPM packaging for ODP

Hi, This patchset enables RPM packaging for ODP and it creates the correct version number for every commit. v3: - fix short log spelling correction that Maxim found - fix the comment maxim had about odp.spec v2: - fixed line over 80 characters - added -M to format-patch... =/ Cheers, Anders

Re: [lng-odp] [PATCHv2 6/8] pkg: add odp.spec to generate rpm packages

On 2015-11-02 10:14, Maxim Uvarov wrote: > On 10/30/2015 23:57, Anders Roxell wrote: > >Signed-off-by: Anders Roxell > >Reviewed-by: Fathi Boudra > >--- > > pkg/rpm/odp.spec | 65 > > > > 1 file changed, 65 insertions(+) > > create mode

Re: [lng-odp] [PATCHv2 8/8] scritps/common_pkg_build: don't create pkg's if no repo

On 2015-11-02 10:16, Maxim Uvarov wrote: > scritps -> scripts Will fix. Thanks, Anders > > On 10/30/2015 23:57, Anders Roxell wrote: > >We shouldn't support building packages from a tarball. > > > >Suggested-by:Fathi Boudra > >Signed-off-by: Anders Roxell > >--- > > scripts/common_pkg_build

Re: [lng-odp] Runtime inlining

> On Nov 10, 2015, at 7:28 AM, Zoltan Kiss wrote: > > > >> On 10/11/15 15:08, Grant Likely wrote: >>> On Tue, Nov 10, 2015 at 3:04 PM, Zoltan Kiss wrote: >>> >>> On 10/11/15 12:04, Grant Likely wrote: On Tue, Nov 10, 2015 at 11:08 AM, Maxim Uvarov wrote: > >>

Re: [lng-odp] [API-NEXT PATCH 5/5] validation: implement pktio statistics counters

On Mon, Nov 09, 2015 at 02:21:53PM +0300, Maxim Uvarov wrote: > Signed-off-by: Maxim Uvarov > --- > test/validation/pktio/pktio.c | 130 > ++ > 1 file changed, 130 insertions(+) > > diff --git a/test/validation/pktio/pktio.c b/test/validation/pktio/pktio.

Re: [lng-odp] [API-NEXT PATCH 1/5] linux-generic: sockets: implement pktio statistics counters

On Mon, Nov 09, 2015 at 02:21:49PM +0300, Maxim Uvarov wrote: > Signed-off-by: Maxim Uvarov > --- > platform/linux-generic/Makefile.am | 1 + > .../linux-generic/include/odp_packet_io_internal.h | 9 +++ > platform/linux-generic/odp_packet_io.c | 53 +

Re: [lng-odp] [API-NEXT PATCH 0/5] pktio statistics counters

On Mon, Nov 09, 2015 at 02:21:48PM +0300, Maxim Uvarov wrote: > Initial version of implementation pktio statistics counters. > > ***Did not test netmap, it it should be the same sysfs netdev counters. This doesn't work for netmap, the sysfs counters don't get updated while the interface is in net

Re: [lng-odp] [API-NEXT PATCHv4 8/8] validation: schedule: add chaos test

Bill, I merged that 8/8 patch from v4. All others are from v3 which I merged before. Please let me know if there something changed and I need to apply. Thanks, Maxim. On 11/10/2015 17:47, Bill Fischofer wrote: Add a "chaos" test variant to the scheduler CUnit tests. This test stresses the sc

Re: [lng-odp] [PATCH] linux-generic: pktio: print out the name of pktio used

On Tue, Nov 10, 2015 at 03:54:31PM +, Zoltan Kiss wrote: > Hi, > > On 02/11/15 10:54, Stuart Haslam wrote: > >On Fri, Oct 30, 2015 at 06:38:14PM +, Zoltan Kiss wrote: > >>For debug purposes, otherwise it's not trivial to figure out which pktio was > >>successful. > >> > >>Signed-off-by: Zo

[lng-odp] [PATCH 1/2] linux-generic: pktio: recover from transmit errors

Fix the way transmit errors are handled to avoid getting out of sync between kernel and user space, which causes transmission to hang. Fixes bug https://bugs.linaro.org/show_bug.cgi?id=1890 Signed-off-by: Stuart Haslam --- platform/linux-generic/pktio/socket_mmap.c | 59 +---

[lng-odp] [PATCH 2/2] validation: pktio: test transmit error recovery

After inducing a transmit error send some additional packets to ensure transmit recovers correctly. Signed-off-by: Stuart Haslam --- test/validation/pktio/pktio.c | 19 +++ 1 file changed, 19 insertions(+) diff --git a/test/validation/pktio/pktio.c b/test/validation/pktio/pktio.

Re: [lng-odp] [PATCH] linux-generic: pktio: print out the name of pktio used

Hi, On 02/11/15 10:54, Stuart Haslam wrote: On Fri, Oct 30, 2015 at 06:38:14PM +, Zoltan Kiss wrote: For debug purposes, otherwise it's not trivial to figure out which pktio was successful. Signed-off-by: Zoltan Kiss diff --git a/platform/linux-generic/pktio/socket.c b/platform/linux

Re: [lng-odp] [API-NEXT PATCHv3 8/8] validation: schedule: add chaos test

Feel free to enhance it further :) On Tuesday, November 10, 2015, Ivan Khoronzhuk wrote: > nice test > it good to have same but w/o ordered queues. > It can allow to figure out issues with less step. > > On 10.11.15 06:20, Bill Fischofer wrote: > >> Add a "chaos" test variant to the scheduler CU

[lng-odp] [PATCH 10/10] validation: pktio: add test for start when started and stop when stopped()

Also checks that interface are stopped by default Signed-off-by: Nicolas Morey-Chaisemartin --- test/validation/pktio/pktio.c | 7 ++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/test/validation/pktio/pktio.c b/test/validation/pktio/pktio.c index e5458c4..8d41a2f 100644 ---

[lng-odp] [PATCH 07/10] validation: classification: stronger checks to avoid SEGV on pktio failure

Signed-off-by: Nicolas Morey-Chaisemartin --- .../classification/odp_classification_test_pmr.c | 14 +++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/test/validation/classification/odp_classification_test_pmr.c b/test/validation/classification/odp_classific

[lng-odp] [PATCH 08/10] linux-generic: pktio: check for pktio_start when started and pktio_stop when stopped

Signed-off-by: Nicolas Morey-Chaisemartin --- platform/linux-generic/odp_packet_io.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/platform/linux-generic/odp_packet_io.c b/platform/linux-generic/odp_packet_io.c index 4d574ad..9d5c971 100644 --- a/platform/linux-generic/odp_packet_io

[lng-odp] [PATCH 06/10] validation: classification: start pktio after setting inq and stop it before removing it

Signed-off-by: Nicolas Morey-Chaisemartin --- test/validation/classification/odp_classification_common.c | 2 ++ test/validation/classification/odp_classification_test_pmr.c | 11 +-- 2 files changed, 7 insertions(+), 6 deletions(-) diff --git a/test/validation/classification/odp_clas

[lng-odp] [PATCH 09/10] linux-generic: pktio: configuration functions check that interface is stopped

Signed-off-by: Nicolas Morey-Chaisemartin --- platform/linux-generic/odp_packet_io.c | 12 1 file changed, 12 insertions(+) diff --git a/platform/linux-generic/odp_packet_io.c b/platform/linux-generic/odp_packet_io.c index 9d5c971..1e9d08c 100644 --- a/platform/linux-generic/odp_pa

[lng-odp] [PATCH 05/10] validation: pktio: remove unneeded stop as interface is stopped after open()

Signed-off-by: Nicolas Morey-Chaisemartin --- test/validation/pktio/pktio.c | 4 1 file changed, 4 deletions(-) diff --git a/test/validation/pktio/pktio.c b/test/validation/pktio/pktio.c index 81f03db..e5458c4 100644 --- a/test/validation/pktio/pktio.c +++ b/test/validation/pktio/pktio.c @@

[lng-odp] [PATCH 03/10] validation: pktio: add tests for rrecv() on WONLY, and send on RONLY pktios

Signed-off-by: Nicolas Morey-Chaisemartin --- test/validation/pktio/pktio.c | 67 +++ 1 file changed, 67 insertions(+) diff --git a/test/validation/pktio/pktio.c b/test/validation/pktio/pktio.c index b52da06..d39251c 100644 --- a/test/validation/pktio/pkti

[lng-odp] [PATCH 04/10] validation: pktio: stop interfaces before removing the default inq

Signed-off-by: Nicolas Morey-Chaisemartin --- test/validation/pktio/pktio.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/test/validation/pktio/pktio.c b/test/validation/pktio/pktio.c index d39251c..81f03db 100644 --- a/test/validation/pktio/pktio.c +++ b/test/validation/pktio/pktio.c @@

[lng-odp] [PATCH 02/10] validation: pktio: add customizable out mode for pktios

Signed-off-by: Nicolas Morey-Chaisemartin --- test/validation/pktio/pktio.c | 35 +++ 1 file changed, 23 insertions(+), 12 deletions(-) diff --git a/test/validation/pktio/pktio.c b/test/validation/pktio/pktio.c index a2da47b..b52da06 100644 --- a/test/validation/p

[lng-odp] [PATCH 01/10] linux-generic: pktio: check interface mode is compatible before receiving or sending

Signed-off-by: Nicolas Morey-Chaisemartin --- platform/linux-generic/odp_packet_io.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/platform/linux-generic/odp_packet_io.c b/platform/linux-generic/odp_packet_io.c index c3bc6a0..4d574ad 100644 --- a/platform/linux-generi

[lng-odp] [PATCH 00/10] Pktio checks

This series add several tests for pktios and fixes the issue they raised. The main features checked are: * RONLY pktio cannot be sent to * WONLY pktio cannot be read from * stopped pktio cannot be stopped * started pktio cannot be started * started pktio cannot be configured Nicolas Morey-Chaisema

Re: [lng-odp] [API-NEXT PATCHv3 8/8] validation: schedule: add chaos test

nice test it good to have same but w/o ordered queues. It can allow to figure out issues with less step. On 10.11.15 06:20, Bill Fischofer wrote: Add a "chaos" test variant to the scheduler CUnit tests. This test stresses the scheduler by circulating events among parallel, atomic, and ordered qu

[lng-odp] [PATCH 2/3 v5] linux-generic: odp_timer: warn if tick is late

If tick is late then application should be warned about this. It means that actual timer resolution is worse than expected. The default timer resolution is set 10ms, that is equal to jiffy on most systems. The default resolution is set bearing in mind that on a CPU runs maximum two threads that ide

[lng-odp] [PATCH 3/3 v3] linux-generic: cpumask: warn that CPU0 is used by control and worker thread

By default all control threads on CPU0, and odp_cpumask_default_control returns it. It shouldn't overlap with worker cpumask, but for some cases it's correct to use potential processing capability, so better leave this choice to application and only draw attention to it when cpumask for worker thre

[lng-odp] [PATCH 1/3 v3] example: timer: warn if timeout less than resolution

The period shouldn't be less than timer resolution. If it's less there is no guarantee about timeout accuracy, it can be executed right after setting it or after up to timer resolution timeout or even be set late, in which case application will be stopped with error. So it's better to warn user tha

[lng-odp] [PATCH 0/3 v2] add warnings to improve timer usage

This series is intended to help user to set timer in frames that OS allows him. The timer resolution shouldn't be better than system allows, minimum timeout shouldn't be more than resolution, and warn application that CPU0, that is used for tick notification handling, can have impact on worker thre

[lng-odp] [API-NEXT PATCHv4 5/6] api: crypto: add AES128-CBC encrypt/decrypt methods

Signed-off-by: Nicolas Morey-Chaisemartin --- include/odp/api/crypto.h | 2 + .../linux-generic/include/odp_crypto_internal.h| 4 + platform/linux-generic/odp_crypto.c| 89 ++ 3 files changed, 95 insertions(+) diff --git a/inclu

[lng-odp] [API-NEXT PATCHv4 4/6] example: ipsec: add support for HMAC-SHA-256-128

Signed-off-by: Nicolas Morey-Chaisemartin --- example/ipsec/odp_ipsec.c| 2 +- example/ipsec/odp_ipsec_misc.h | 8 ++-- example/ipsec/odp_ipsec_sa_db.c | 4 example/ipsec/odp_ipsec_stream.c | 3 ++- 4 files changed, 13 insertions(+), 4 deletions(-) diff --git a/example/ipsec/

[lng-odp] [API-NEXT PATCHv4 6/6] validation: crypto: add test for AES128 CBC

Test reference vectors from RFC3602 Signed-off-by: Nicolas Morey-Chaisemartin --- test/validation/crypto/crypto.h | 4 + test/validation/crypto/odp_crypto_test_inp.c | 135 ++- test/validation/crypto/test_vectors.h| 67 + test/validatio

[lng-odp] [API-NEXT PATCHv4 3/6] validation: crypto: add test for HMAC-SHA-256-128

Signed-off-by: Nicolas Morey-Chaisemartin --- test/validation/crypto/crypto.h | 1 + test/validation/crypto/odp_crypto_test_inp.c | 37 +++ test/validation/crypto/test_vectors.h| 44 test/validation/crypto/test_vectors_len.h

[lng-odp] [API-NEXT PATCHv4 1/6] api: crypto: move enums from platform types to odp and rename to fit the API format

Signed-off-by: Nicolas Morey-Chaisemartin --- example/ipsec/odp_ipsec_cache.c| 2 +- example/ipsec/odp_ipsec_cache.h| 4 +- example/ipsec/odp_ipsec_misc.h | 4 +- include/odp/api/crypto.h | 140 ++-

[lng-odp] [API-NEXT PATCHv4 2/6] api: crypto: add HMAC-SHA-256-128 support

Signed-off-by: Nicolas Morey-Chaisemartin --- include/odp/api/crypto.h | 2 + .../linux-generic/include/odp_crypto_internal.h| 4 + platform/linux-generic/odp_crypto.c| 89 ++ 3 files changed, 95 insertions(+) diff --git a/inclu

[lng-odp] [API-NEXT PATCHv4 0/6] Add HMAC-SHA-256-128 and AES128 CBC support

The first patch is a cleanup suggested by Petri. All the crypto enums are moved from linux-generic back to ODP API, and renamed (odp__t) Following patches add support for the new HMAC function, then a validation test and finally support in odp-ipsec And then same things for AES128 v4: * add mis

Re: [lng-odp] Runtime inlining

On 10/11/15 15:08, Grant Likely wrote: On Tue, Nov 10, 2015 at 3:04 PM, Zoltan Kiss wrote: On 10/11/15 12:04, Grant Likely wrote: On Tue, Nov 10, 2015 at 11:08 AM, Maxim Uvarov wrote: On 10 November 2015 at 13:41, Zoltan Kiss wrote: On 10/11/15 07:39, Maxim Uvarov wrote: And it lo

[lng-odp] [Bug 1879] Ordered queues occasionally not restoring order properly

https://bugs.linaro.org/show_bug.cgi?id=1879 --- Comment #23 from Bill Fischofer --- v4 of patch series (part 8) http://patches.opendataplane.org/patch/3702/ fixes issues Maxim identified with the enhanced validation test. Parts 1-7 already merged. -- You are receiving this mail because: You a

Re: [lng-odp] [API-NEXT PATCHv3 0/8] Ordered Queue Bug Fixes

v4 is changed to use PRIu64. It also uses the same sort of conversion casting that is used in strongtypes.h for coercing indexes to pointers and vice versa, so it should be clean. On Tue, Nov 10, 2015 at 8:57 AM, Maxim Uvarov wrote: > On 11/10/2015 17:49, Bill Fischofer wrote: > >> Thanks, Maxi

Re: [lng-odp] Runtime inlining

As I said in the call last week, the problem is wider than that. ODP specifies a lot of types but not their sizes, a lot of enums/defines (things like ODP_PKTIO_INVALID) but not their value either. For our port a lot of those values were changed for performance/implementation reason. So I'm not

Re: [lng-odp] Runtime inlining

On Tue, Nov 10, 2015 at 3:04 PM, Zoltan Kiss wrote: > > > On 10/11/15 12:04, Grant Likely wrote: >> >> On Tue, Nov 10, 2015 at 11:08 AM, Maxim Uvarov >> wrote: >>> >>> On 10 November 2015 at 13:41, Zoltan Kiss wrote: On 10/11/15 07:39, Maxim Uvarov wrote: > > And it looks like

[lng-odp] [Bug 1449] odp_timer_test core dump

https://bugs.linaro.org/show_bug.cgi?id=1449 --- Comment #13 from Ivan Khoronzhuk --- Sent series that combine previously mentioned updated patches: [lng-odp] [PATCH 0/3] add warnings to improve timer usage https://lists.linaro.org/pipermail/lng-odp/2015-November/017160.html -- You are receivin

Re: [lng-odp] Runtime inlining

On 10/11/15 12:04, Grant Likely wrote: On Tue, Nov 10, 2015 at 11:08 AM, Maxim Uvarov wrote: On 10 November 2015 at 13:41, Zoltan Kiss wrote: On 10/11/15 07:39, Maxim Uvarov wrote: And it looks like it's problem in OVS, not in ODP. I.e. OVS should allow to use library functions for fast p

Re: [lng-odp] [PATCH v3] linux-generic: odp_timer: warn if tick is lost

On 10 November 2015 at 15:50, Ivan Khoronzhuk wrote: > > > On 10.11.15 16:43, Ola Liljedahl wrote: > >> >> >> On 10 November 2015 at 13:24, Ivan Khoronzhuk > > wrote: >> >> >> >> On 10.11.15 13:37, Ivan Khoronzhuk wrote: >> >> >> >> On 10.11.15 13:11

Re: [lng-odp] [PATCH 2/3 v4] linux-generic: odp_timer: warn if tick is late

On 10.11.15 16:47, Ola Liljedahl wrote: On 10 November 2015 at 15:40, Ivan Khoronzhuk mailto:ivan.khoronz...@linaro.org>> wrote: If tick is late then application should be warned about this. It means that actual timer resolution is worse than expected. The default timer resolutio

Re: [lng-odp] [API-NEXT PATCHv3 0/8] Ordered Queue Bug Fixes

On 11/10/2015 17:49, Bill Fischofer wrote: Thanks, Maxim. I sent a v4 that should correct those issues. If it doesn't, I'd take parts 1-7 that fix the actual bugs and I'll tweak the enhanced test case further separately. Bill, I already merged 1-7. In test you need to use PRIu64 instead o

[lng-odp] [API-NEXT PATCHv4 7/8] linux-generic: schedule: allow order to be ignored for internal use

When a schedulable queue transitions from QUEUE_STATUS_NOTSCHED to QUEUE_STATUS_SCHED, the scheduler makes it ready by adding it to one of the scheduler's internal queues. This enqueue operation should not participate in any current ordered context to avoid potential deadlock. This patch resolves

Re: [lng-odp] Runtime inlining

On 6 November 2015 at 15:48, Zoltan Kiss wrote: > Hi, > > We have a packaging/linking/optimization problem at LNG, I hope you guys > can give us some advice on that. (Cc'ing ODP list in case someone want to > add something) > We have OpenDataPlane (ODP), an API stretching between userspace > appl

Re: [lng-odp] [API-NEXT PATCHv3 0/8] Ordered Queue Bug Fixes

Thanks, Maxim. I sent a v4 that should correct those issues. If it doesn't, I'd take parts 1-7 that fix the actual bugs and I'll tweak the enhanced test case further separately. On Tue, Nov 10, 2015 at 1:01 AM, Maxim Uvarov wrote: > On 11/10/2015 07:20, Bill Fischofer wrote: > >> Changes for

Re: [lng-odp] [PATCH v3] linux-generic: odp_timer: warn if tick is lost

On 10.11.15 16:43, Ola Liljedahl wrote: On 10 November 2015 at 13:24, Ivan Khoronzhuk mailto:ivan.khoronz...@linaro.org>> wrote: On 10.11.15 13:37, Ivan Khoronzhuk wrote: On 10.11.15 13:11, Maxim Uvarov wrote: On 11/10/2015 14:01, Ivan Khoronzhuk wrote:

[lng-odp] [API-NEXT PATCHv4 8/8] validation: schedule: add chaos test

Add a "chaos" test variant to the scheduler CUnit tests. This test stresses the scheduler by circulating events among parallel, atomic, and ordered queues to verify that the scheduler can handle arbitrary looping paths without deadlock. Suggested-by: Carl Wallen Signed-off-by: Bill Fischofer ---

[lng-odp] [API-NEXT PATCHv4 6/8] linux-generic: queue: streamline and correct release_order() routine

Resolve the corner case of releasing order for an order that still has events on the reorder queue. This also allows the reorder_complete() routine to be streamlined. This patch resolves Bug https://bugs.linaro.org/show_bug.cgi?id=1879 Signed-off-by: Bill Fischofer --- .../linux-generic/include

[lng-odp] [API-NEXT PATCHv4 4/8] linux-generic: queue: add ordered_queue_enq() routine - part 2

Restructure queue_enq() to streamline it while using the new ordered_queue_enq() routine to handle ordered queues. This change is made in two parts to make the diffs easier to follow. Signed-off-by: Bill Fischofer --- platform/linux-generic/odp_queue.c | 134 -

[lng-odp] [API-NEXT PATCHv4 3/8] linux-generic: queue: add ordered_queue_enq() routine - part 1

Add the new ordered_queue_enq() internal routine. This is done in two parts to make the diffs easier to follow. Part 1 adds the new routine while Part 2 replaces queue_enq() to use it. Signed-off-by: Bill Fischofer --- .../linux-generic/include/odp_queue_internal.h | 2 + platform/linux-ge

[lng-odp] [API-NEXT PATCHv4 5/8] linux-generic: queue: streamline reorder_deq() routine

Signed-off-by: Bill Fischofer --- .../linux-generic/include/odp_queue_internal.h | 16 ++--- platform/linux-generic/odp_queue.c | 27 ++ 2 files changed, 19 insertions(+), 24 deletions(-) diff --git a/platform/linux-generic/include/odp_queue_intern

[lng-odp] [API-NEXT PATCHv4 2/8] linux-generic: queue: add utility functions for restructure

Signed-off-by: Bill Fischofer --- .../linux-generic/include/odp_queue_internal.h | 44 ++ platform/linux-generic/odp_queue.c | 7 2 files changed, 51 insertions(+) diff --git a/platform/linux-generic/include/odp_queue_internal.h b/platform/linux-gen

[lng-odp] [API-NEXT PATCHv4 1/8] linux-generic: schedule: move ordered lock routines to odp_schedule.c

Move the odp_schedule_order_lock() and odp_schedule_order_unlock() routines from odp_queue.c to odp_schedule.c Signed-off-by: Bill Fischofer --- platform/linux-generic/odp_queue.c| 31 platform/linux-generic/odp_schedule.c | 45 ++- 2

[lng-odp] [API-NEXT PATCHv4 0/8] Ordered Queue Bug Fixes

Changes for v4 -- Handle schedule test conversions properly for operation on 32-bit systems -- Schedule test must use atomics for accessing global counts Changes for v3 -- Resolves deadlock issue identified by Carl Wallen -- Adds a "chaos" test to the scheduler CUnit suite to verify the above Cha

Re: [lng-odp] [PATCH 2/3 v4] linux-generic: odp_timer: warn if tick is late

On 10 November 2015 at 15:40, Ivan Khoronzhuk wrote: > If tick is late then application should be warned about this. > It means that actual timer resolution is worse than expected. > The default timer resolution is set 10ms, that is equal to jiffy > on most systems. The default resolution is set

Re: [lng-odp] [PATCH v3] linux-generic: odp_timer: warn if tick is lost

On 10 November 2015 at 13:24, Ivan Khoronzhuk wrote: > > > On 10.11.15 13:37, Ivan Khoronzhuk wrote: > >> >> >> On 10.11.15 13:11, Maxim Uvarov wrote: >> >>> On 11/10/2015 14:01, Ivan Khoronzhuk wrote: >>> If tick is lost then application should be warned about this. It means that actua

[lng-odp] [PATCH 3/3 v2] linux-generic: cpumask: warn that CPU0 is used by control and worker thread

By default all control threads on CPU0, and odp_cpumask_default_control returns it. It shouldn't overlap with worker cpumask, but for some cases it's correct to use potential processing capability, so better leave this choice to application and only draw attention to it when cpumask for worker thre

[lng-odp] [PATCH 1/3 v2] example: timer: warn if timeout less than resolution

The period shouldn't be less than timer resolution. If it's less there is no guarantee about timeout accuracy, it can be executed right after setting it or after up to timer resolution timeout or even be set late, in which case application will be stopped with error. So it's better to warn user tha

[lng-odp] [PATCH 2/3 v4] linux-generic: odp_timer: warn if tick is late

If tick is late then application should be warned about this. It means that actual timer resolution is worse than expected. The default timer resolution is set 10ms, that is equal to jiffy on most systems. The default resolution is set bearing in mind that on a CPU runs maximum two threads that ide

[lng-odp] [PATCH 0/3] add warnings to improve timer usage

This series is intended to help user to set timer in frames that OS allows him. The timer resolution shouldn't be better than system allows, minimum timeout shouldn't be more than resolution, and warn application that CPU0, that is used for tick notification handling, can have impact on worker thre

Re: [lng-odp] [PATCH v3] linux-generic: odp_timer: warn if tick is lost

On 10.11.15 13:37, Ivan Khoronzhuk wrote: On 10.11.15 13:11, Maxim Uvarov wrote: On 11/10/2015 14:01, Ivan Khoronzhuk wrote: If tick is lost then application should be warned about this. It means that actual timer resolution is worse than expected. The default timer resolution is set 10ms,

Re: [lng-odp] [PATCH 1/2] linux-generic: odp_cpumask.c: assert on ODP_CPUMASK_SIZE

Merged, Maxim. On 11/06/2015 21:34, Mike Holmes wrote: On 29 October 2015 at 10:18, Anders Roxell > wrote: The API definition of a CPU mask must not be incompatible with the platform it is implemented on Signed-off-by: Anders Roxell mailto:anders

Re: [lng-odp] Runtime inlining

On 10/11/15 11:08, Maxim Uvarov wrote: On 10 November 2015 at 13:41, Zoltan Kiss mailto:zoltan.k...@linaro.org>> wrote: On 10/11/15 07:39, Maxim Uvarov wrote: JIT like lua might also not work because you need to rewrite OVS to support it. I don't think that it will be

Re: [lng-odp] Runtime inlining

On Tue, Nov 10, 2015 at 11:08 AM, Maxim Uvarov wrote: > On 10 November 2015 at 13:41, Zoltan Kiss wrote: >> On 10/11/15 07:39, Maxim Uvarov wrote: >>> And it looks like it's problem in OVS, not in ODP. I.e. OVS should allow >>> to use library functions for fast path (where inlines are critical).

Re: [lng-odp] [PATCH 1/2] linux-generic: odp_cpumask.c: assert on ODP_CPUMASK_SIZE

Merged, Maxim. On 11/06/2015 21:34, Mike Holmes wrote: On 29 October 2015 at 10:18, Anders Roxell > wrote: The API definition of a CPU mask must not be incompatible with the platform it is implemented on Signed-off-by: Anders Roxell mailto:anders

Re: [lng-odp] [PATCH v3] linux-generic: odp_timer: warn if tick is lost

On 10.11.15 13:11, Maxim Uvarov wrote: On 11/10/2015 14:01, Ivan Khoronzhuk wrote: If tick is lost then application should be warned about this. It means that actual timer resolution is worse than expected. The default timer resolution is set 10ms, that is equal to jiffy on most systems. The d

Re: [lng-odp] [PATCH] doc: improve asciidoc presentation

Merged, Maxim. On 11/09/2015 03:31, Bill Fischofer wrote: Shouldn't these be tagged --subject-prefix="NEXT PATCH" to make it clear which branch it's intended for? Other than that, looks fine. Bill, we agreed to not send patches to next branch. It's used as staging for new api and I will ta

Re: [lng-odp] [PATCH v3] linux-generic: odp_timer: warn if tick is lost

On 11/10/2015 14:01, Ivan Khoronzhuk wrote: If tick is lost then application should be warned about this. It means that actual timer resolution is worse than expected. The default timer resolution is set 10ms, that is equal to jiffy on most systems. The default resolution is set bearing in mind t

Re: [lng-odp] Runtime inlining

On 10 November 2015 at 13:41, Zoltan Kiss wrote: > > > On 10/11/15 07:39, Maxim Uvarov wrote: > >> JIT like lua might also not work because you need to rewrite OVS to >> support it. I don't think that it will be accepted. >> >> And it looks like it's problem in OVS, not in ODP. I.e. OVS should al

[lng-odp] [PATCH v3] linux-generic: odp_timer: warn if tick is lost

If tick is lost then application should be warned about this. It means that actual timer resolution is worse than expected. The default timer resolution is set 10ms, that is equal to jiffy on most systems. The default resolution is set bearing in mind that on a CPU runs maximum two threads that ide

Re: [lng-odp] [PATCH v2] linux-generic: odp_timer: warn if tick is lost

ignore it. On 10.11.15 12:58, Ivan Khoronzhuk wrote: If tick is lost then application should be warned about this. It means that actual timer resolution is worse than expected. The default timer resolution is set 10ms, that is equal to jiffy on most systems. The default resolution is set bearing

[lng-odp] [PATCH v2] linux-generic: odp_timer: warn if tick is lost

If tick is lost then application should be warned about this. It means that actual timer resolution is worse than expected. The default timer resolution is set 10ms, that is equal to jiffy on most systems. The default resolution is set bearing in mind that on a CPU runs maximum two threads that ide

Re: [lng-odp] Runtime inlining

On 10/11/15 07:39, Maxim Uvarov wrote: JIT like lua might also not work because you need to rewrite OVS to support it. I don't think that it will be accepted. And it looks like it's problem in OVS, not in ODP. I.e. OVS should allow to use library functions for fast path (where inlines are cri

Re: [lng-odp] [API-NEXT PATCH v7 0/6] api: time: unbind CPU cycles from time API

Merged, Maxim. On 11/05/2015 18:33, Ivan Khoronzhuk wrote: This seres is intended to unbind time API names from CPU "cycles". Also remove usage of word "cycle" from appropriate places as it's no more valid. This patch series only change API names and adds calls required for that and doesn't add

Re: [lng-odp] [API-NEXT PATCH v8 3/6] api: time: unbind CPU cycles from time API

On 11/10/2015 12:17, Ivan Khoronzhuk wrote: On 10.11.15 11:08, Savolainen, Petri (Nokia - FI/Espoo) wrote: -Original Message- From: EXT Ivan Khoronzhuk [mailto:ivan.khoronz...@linaro.org] Sent: Tuesday, November 10, 2015 10:59 AM To: Savolainen, Petri (Nokia - FI/Espoo); Maxim Uvaro

  1   2   >