On 30 April 2015 at 09:32, Maxim Uvarov <[email protected]> wrote:
> On 04/30/2015 14:51, Mike Holmes wrote: > > >> On 30 April 2015 at 06:55, Savolainen, Petri (Nokia - FI/Espoo) < >> [email protected] <mailto:[email protected]>> wrote: >> >> Hi, >> >> This time it is appropriate to bump the ODP_VERSION_API_MINOR, >> since "api: time: force time defines as ULL to avoid computation" >> actually changed the API signature visible to the application. But >> I question if previous increments to *API version* were necessary. >> For example, here are listed all API changes between APIs v1.0.2 >> and v1.0.3 ... >> >> git diff v1.0.2..v1.0.3 -- include/odp >> >> diff --git a/include/odp/api/version.h b/include/odp/api/version.h >> index ae1cf0d..3338559 100644 >> --- a/include/odp/api/version.h >> +++ b/include/odp/api/version.h >> @@ -46,7 +46,7 @@ extern "C" { >> * to the API. For an API with common generation and major >> version, but with >> * different minor numbers the two versions are backward compatible. >> */ >> -#define ODP_VERSION_API_MINOR 2 >> +#define ODP_VERSION_API_MINOR 3 >> >> /** >> * Returns ODP API version string >> >> >> >> git diff v1.0.2..v1.0.3 -- platform/linux-generic/include/odp >> >> >> >> .. absolutely nothing, but still we have another API version out >> there. ODP release/validation suite/linux-generic implementation >> version can be combined into one number, but it should be >> different from the ODP API version number. Today the API version >> should be actually v1.0.1. >> >> >> Completely agree should have been bumping ODP_VERSION_IMPL_STR - not >> sure how we did not pick up on that prior. The docs dont describe that well >> either. >> > > > Because we also update validation test suite, which also needs it's number > increased somewhere and we have only API_MINOR for that. > Good point. Since tests should reflect the API version we are back in a tangle unless we introduce a test version number that normally equals API number but increments minor independently. We keep coming wound to these issues becasue the API, tests and reference implementation at all in the same repo IMHO. Would it be better to use the git repo command and a manifest that gathers the correct parts, the api, implementation and tests so that the developers local directory looks approximately as it does now, but it is contained in separate repos ? Yes I know this will mean some synchronization when an API, implementation and tests all need to be updated at once. However in many cases we can independently improve tests and implementation. The same rules apply though in that a new api does not get out of api-next until all these parts are present in api-next and pass CI. > Maxim. > > >> >> -Petri >> >> >> >> > -----Original Message----- >> > From: lng-odp [mailto:[email protected] >> <mailto:[email protected]>] On Behalf Of ext >> > Maxim Uvarov >> > Sent: Wednesday, April 29, 2015 8:26 PM >> > To: [email protected] <mailto:[email protected]> >> > Subject: [lng-odp] [PATCH] update version number from v1.0.3 to >> v1.0.4 >> > >> > Signed-off-by: Maxim Uvarov <[email protected] >> <mailto:[email protected]>> >> >> > --- >> > debian/changelog | 27 +++++++++++++++++++++++++++ >> > include/odp/api/version.h | 2 +- >> > 2 files changed, 28 insertions(+), 1 deletion(-) >> > >> > diff --git a/debian/changelog b/debian/changelog >> > index 2087289..c6c0ec8 100644 >> > --- a/debian/changelog >> > +++ b/debian/changelog >> > @@ -1,3 +1,30 @@ >> > +opendataplane (1.0.4-1) unstable; urgency=low >> > + >> > + * test: >> > + * validation: >> > + - synchronizers: fix init of custom barrier atomic >> > + - pktio: don't pass invalid handles >> > + - use PRIu32 instead of %u to print uint32_t variables >> > + - thread: add odp_thread sunnyday tests >> > + - classification: add missing init of atomic variable >> > + - api_test: remove odp_shm_test >> > + * performance: >> > + - pktio: fix check for pktio_tx >> > + - pktio: test return value of odp_pktio_close() >> > + - pktio: add throughput test >> > + * example: >> > + - ODP classifier example >> > + * bugs: >> > + - example: timer: clean up prescheduled events >> > + * general: >> > + - api: time: force time defines as ULL to avoid >> computation >> > overflows on 32bits systems >> > + - linux-generic: packet_io: init l2 and l3 cos table >> spinlocks >> > + - linux-generic: pool: initialize atomics during >> > odp_pool_init_global >> > + - linux-generic: pool: fix initialization of >> odp_buffer_hdr >> > ref_count >> > + - doc: api_guide_lines: add guidance on ODP_DEPRECATED >> > + >> > + -- Maxim Uvarov <[email protected] >> <mailto:[email protected]>> Wed, 29 Apr 2015 20:21:25 >> > +0300 >> > + >> > opendataplane (1.0.3-1) unstable; urgency=low >> > >> > * test: >> > diff --git a/include/odp/api/version.h b/include/odp/api/version.h >> > index 3338559..027095d 100644 >> > --- a/include/odp/api/version.h >> > +++ b/include/odp/api/version.h >> > @@ -46,7 +46,7 @@ extern "C" { >> > * to the API. For an API with common generation and major >> version, but >> > with >> > * different minor numbers the two versions are backward >> compatible. >> > */ >> > -#define ODP_VERSION_API_MINOR 3 >> > +#define ODP_VERSION_API_MINOR 4 >> > >> > /** >> > * Returns ODP API version string >> > -- >> > 1.9.1 >> > >> > _______________________________________________ >> > lng-odp mailing list >> > [email protected] <mailto:[email protected]> >> > https://lists.linaro.org/mailman/listinfo/lng-odp >> _______________________________________________ >> lng-odp mailing list >> [email protected] <mailto:[email protected]> >> https://lists.linaro.org/mailman/listinfo/lng-odp >> >> >> >> >> -- >> Mike Holmes >> Technical Manager - Linaro Networking Group >> Linaro.org <http://www.linaro.org/>***│ *Open source software for ARM >> SoCs >> >> > -- Mike Holmes Technical Manager - Linaro Networking Group Linaro.org <http://www.linaro.org/> *│ *Open source software for ARM SoCs
_______________________________________________ lng-odp mailing list [email protected] https://lists.linaro.org/mailman/listinfo/lng-odp
