Hi, This patch and similar ones are actually fixing bugs. https://gerrit.iotivity.org/gerrit/gitweb?p=iotivity.git;a=commitdiff;h=b13839cceaae97d7d23b54586da1ad7ac85ab2b6
When you build an application which only uses the C++ API, you have to link against liboc.so, liboctbstack.so, liboc_logger.so and depending on the system libpthread.so without this patch, with this patch you only have to link against liboc.so and the others are pulled in automatically. These dependencies are added to the lib. The last patch only adds the check to the build system. Hauke -----Original Message----- From: Keane, Erich [mailto:[email protected]] Sent: Freitag, 24. Juli 2015 20:20 To: hunje.yeon at samsung.com Cc: iotivity-dev at lists.iotivity.org; jonc at osg.samsung.com; Mehrtens Hauke (LQKG BU AE PS SW) Subject: Re: [dev] Jenkins tests done with dependencies None of these patches were moved to master. They are build-improvement things, not bug-fixes, thus by my understanding of the rules aren't eligible. On Fri, 2015-07-24 at 18:15 +0000, ??? wrote: > Hi All, > > > > I want to ask one thing about this patches (already merged to master) > > Are those patches applied to 0.9.2-DEV branch? > > If not, it could affect cherry-picking to 0.9.2-DEV. > > > > Regards > > Hun-je. > > > > ------- Original Message ------- > > Sender : Jon A. Cruz<jonc at osg.samsung.com> > > Date : 2015-07-25 01:28 (GMT+09:00) > > Title : Re: [dev] Jenkins tests done with dependencies > > > > I'd like to highlight one problem here that people need to be aware > of. > > Aside from the general issues that Hauke pointed out, getting patches > committed out of order breaks a major feature of git. That is the use > of git bisect to track down issues. > > It is very important to try to always keep changes in proper order so > that the build at any given revision is not broken. This is in > contrast with use of other source control people might be used to > where a breaking fix can be followed up with it's correction as a > common practice. > > So again, please stay aware of "git bisect" and it's behavior. > > > On 07/24/2015 03:09 AM, Hauke.Mehrtens at lantiq.com wrote: > > Hi, > > > > Now also to the mailing list. > > > > You applied my patch > https://gerrit.iotivity.org/gerrit/#/c/1802/ which makes the build > process fail if shared libs are having any unresolved symbols. These > unresolved symbols are an error in IoTivity. I found some places where > dependencies to other shared libs were missing and fixed them in the > patches before that patch, so I made a pull request which contained > more than one patch. When Jenkins tested them he tested the patch and > all the patches it depends on this, so it was tested successfully > because I fixed all the places where shared libs with unresolved > symbols were built in patches before the patches added the warnings. > > Now this patch which make the build process check for these > unresolved symbols was cherry-picked and not all the other patches > before were applied. This now broke the build of upstream IoTivity. > > > > To fix the immediate problem you should apply these patches: > > https://gerrit.iotivity.org/gerrit/1795 > > https://gerrit.iotivity.org/gerrit/1796 > > https://gerrit.iotivity.org/gerrit/1799 > > https://gerrit.iotivity.org/gerrit/1800 > > https://gerrit.iotivity.org/gerrit/1825 > > > > I think we should not make big patches (Gerrit marks the size > already red) and them we will get pull requests with more than one > patch which cannot be applied in an arbitrary order. > > When applying patches you should take more care about their > dependencies, Gerrit shows a list of related changes which are the > other patches which are in this patchset. > > > > Hauke Mehrtens > > Lantiq Beteiligungs-GmbH & Co.KG > > > > office: Lilienthalstra?e 15, 85579 Neubiberg Germany > > phone: +49 (89) 89899 - 7179 > > e-mail: hauke.mehrtens at lantiq.com > > > > > > > > > > _______________________________________________ > > iotivity-dev mailing list > > iotivity-dev at lists.iotivity.org > > https://lists.iotivity.org/mailman/listinfo/iotivity-dev > > > > _______________________________________________ > iotivity-dev mailing list > iotivity-dev at lists.iotivity.org > https://lists.iotivity.org/mailman/listinfo/iotivity-dev
