Another point that was brought to my attention and that we overlooked, is the new release versioning.
In the early discussions about the release it was agreed on extending the versioning to 3 numbers: major, minor, patch. Ex. 9.0.0 In the past, if a bug was found after a release, a separate file containing a patch fix was distributed. What we want to do now is to fix the bug and re-release incrementing the patch number, 9.0.1 for example. With this in mind and with the suggestion from Brennan, the branch would be "releases/9.0.0" It has been almost a week since we created the branch, some PRs were contributed in the meantime. I think all those contributions are worthy to get incorporated in the release. So, I suggest to delete the old branch and create a new branch "releases/9.0.0" from master. On Fri, Apr 10, 2020 at 9:56 PM Abdelatif Guettouche <abdelatif.guettou...@gmail.com> wrote: > > I think we still need some organization before creating the PR. Mainly > the driver section, I think it should be divided for all other > subsystems. > > But, please, let's take this conversation to the Release Note thread. > I'm still trying to get everyone's opinion on the branch naming convention. > > On Fri, Apr 10, 2020 at 8:12 PM Adam Feuer <a...@starcat.io> wrote: > > > > Ok, I went through PRs 466-764 for bugfixes, and added a few to the wiki > > page. Many of the improvements were changes to conform to style, or small > > fixes that would not be relevant for release notes. > > > > I think we're done summarizing. Are we ready to create a PR for the Release > > Notes? In other words, get them into the text file form? > > > > cheers > > adam > > > > On Fri, Apr 10, 2020 at 11:58 AM Alan Carvalho de Assis <acas...@gmail.com> > > wrote: > > > > > No problem, I didn't realize that you didn't include the previous > > > commit after reviewing all PRs. > > > > > > I think now we have a better picture about all improvements and > > > bugfixes from 8.2 to 9.0. > > > > > > BR, > > > > > > Alan > > > > > > On 4/10/20, Adam Feuer <a...@starcat.io> wrote: > > > > Ok thanks Alan. I'm sorry I wasn't clear about what we had done. > > > > > > > > -adam > > > > > > > > On Fri, Apr 10, 2020 at 11:27 AM Alan Carvalho de Assis < > > > acas...@gmail.com> > > > > wrote: > > > > > > > >> Hi Adam, > > > >> > > > >> Yesterday I passed all the bugfixes from nuttx-8.2 until nuttx-9.0 > > > >> tag. But I didn't checked for improvements because I thought you, > > > >> Nathan and Abdelatif did it. > > > >> > > > >> I can do it for the improvements from nuttx-8.2 until Nov. 23 2019, no > > > >> problem. > > > >> > > > >> BR, > > > >> > > > >> Alan > > > >> > > > >> On 4/10/20, Adam Feuer <a...@starcat.io> wrote: > > > >> > Alan, > > > >> > > > > >> > Next we need to check for features/improvements and bugfixes between > > > 23 > > > >> > December 2019 (1st PR in github.com/apache/incubator-nuttx) and the > > > 16 > > > >> > November 2019, the date of NuttX release 8.2. The main source of info > > > >> > for > > > >> > this would be commit messages in the current master branch. > > > >> > > > > >> > Would you be willing to look through those? Or figure out a way to > > > >> > divide > > > >> > them up among you, me, Nathan, and Abdelatif? > > > >> > > > > >> > -adam > > > >> > > > > >> > On Fri, Apr 10, 2020 at 10:05 AM Alan Carvalho de Assis < > > > >> acas...@gmail.com> > > > >> > wrote: > > > >> > > > > >> >> Hi guys, > > > >> >> > > > >> >> I finished including the apps/ improvements and bugfixes. > > > >> >> > > > >> >> Is there anything else we need to take care? > > > >> >> > > > >> >> BR, > > > >> >> > > > >> >> Alan > > > >> >> > > > >> >> On 4/10/20, Abdelatif Guettouche <abdelatif.guettou...@gmail.com> > > > >> wrote: > > > >> >> >> Speaking about the branch, Brennan is suggesting to have a > > > >> >> >> different > > > >> >> >> naming convention. Instead of nuttx-XX.YY we use releases/XX.YY > > > >> >> >> (the > > > >> >> >> reasons behind this are here: > > > >> >> >> https://github.com/apache/incubator-nuttx/pull/757) > > > >> >> > > > > >> >> > Is everyone okay with changing the branches naming convention? > > > >> >> > I personally do not mind. > > > >> >> > > > > >> >> > For that PR, I asked Brennan on how to proceed with signing the > > > >> >> > tarballs. > > > >> >> > The original flow was: > > > >> >> > 1. Generate the tarballs > > > >> >> > 2. Sign the tarballs and create the hashes. > > > >> >> > 3. Then upload everything (tarballs, hashes, signatures) > > > >> >> > Everything was done locally and then uploaded. > > > >> >> > With the PR757 the release tarballs are generated with a push to > > > the > > > >> >> > release branch. > > > >> >> > Do we download them, sign and then upload the signatures? > > > >> >> > We need at least one way to check the downloads before signing, > > > like > > > >> >> > by creating the hashes with the tarballs. > > > >> >> > > > > >> >> > > > > >> >> > On Fri, Apr 10, 2020 at 4:41 AM Adam Feuer <a...@starcat.io> > > > wrote: > > > >> >> >> > > > >> >> >> Xiang, > > > >> >> >> > > > >> >> >> Ok, thanks. It's great that this doesn't block the release! > > > >> >> >> > > > >> >> >> -adam > > > >> >> >> > > > >> >> >> On Thu, Apr 9, 2020 at 7:49 PM Xiang Xiao > > > >> >> >> <xiaoxiang781...@gmail.com > > > >> > > > > >> >> >> wrote: > > > >> >> >> > > > >> >> >> > This build error happen for several special config only, and > > > >> already > > > >> >> >> > exist > > > >> >> >> > > > > >> >> >> > On Fri, Apr 10, 2020 at 10:38 AM Adam Feuer <a...@starcat.io> > > > >> wrote: > > > >> >> >> > > > > > >> >> >> > > Xiang, > > > >> >> >> > > > > > >> >> >> > > Do the parallel build failures block our release? If so, what > > > >> >> >> > > can > > > >> >> >> > > we > > > >> >> >> > > do > > > >> >> >> > > > > >> >> >> > Shouldn't, because: > > > >> >> >> > 1.The build error already exist for a long time > > > >> >> >> > 2.Only several seldom used configs(all related to > > > >> >> >> > binfmt/module/so) > > > >> >> >> > hit this issue > > > >> >> >> > 3.The single thread build can always finish successfully > > > >> >> >> > > > > >> >> >> > > temporarily to unblock it? Or is there a fix in the works > > > >> >> >> > > that > > > >> >> >> > > will > > > >> >> >> > > be > > > >> >> >> > > ready soon? > > > >> >> >> > > > > > >> >> >> > > > > >> >> >> > Masayuki, Haitao, Yamamoto and I try to fix the problem in the > > > >> >> >> > recent > > > >> >> >> > month, some issue get fixed, but some need more time to > > > >> investigate. > > > >> >> >> > > > > >> >> >> > > -adam > > > >> >> >> > > > > > >> >> >> > > On Thu, Apr 9, 2020 at 7:29 PM Xiang Xiao < > > > >> >> xiaoxiang781...@gmail.com> > > > >> >> >> > wrote: > > > >> >> >> > > > > > >> >> >> > > > At least, the parallel build still fail randomly which > > > >> >> >> > > > block > > > >> our > > > >> >> >> > > > nightly checker. > > > >> >> >> > > > > >> >> >> > > > >> >> >> > > > >> >> >> -- > > > >> >> >> Adam Feuer <a...@starcat.io> > > > >> >> > > > > >> >> > > > >> > > > > >> > > > > >> > -- > > > >> > Adam Feuer <a...@starcat.io> > > > >> > > > > >> > > > > > > > > > > > > -- > > > > Adam Feuer <a...@starcat.io> > > > > > > > > > > > > > -- > > Adam Feuer <a...@starcat.io>
