On Fri, Jun 3, 2016 at 1:18 AM, Renato Golin via llvm-dev < llvm-...@lists.llvm.org> wrote:
> The issues that were raised: > > * Co-dependent patches already break buildbots, but the sequential ID > helps us identify and ignore. They will continue to break, even if we > use git sub-modules, so that doesn't change much, but it will be > harder to spot the issue. Server side hooks may help, as well as > sub-modules. > So as I understand it, there will be a master repository with it's own commit history. Each commit in this repo should be automatically generated as a consistent snapshot of all sub-projects after they have passed build-bot smoke tests. No matter how this master repo is maintained, I'm fairly certain we can make sure that this doesn't become an issue in practice. > > - won't replace SVN's monotonic IDs, but do we *really* need them? > Sub-modules have a bad fame, I gather, but people in the thread > reported success on using it to build validation and release > infrastructure as well as doing bisects, checking out code, etc. We > probably need some documentation on how to do these things, as well as > some scripts to help people work out the dependencies (or use them). > > This master repo could replicate svn's ID's. I mean if the commits are created by a bot / script, each commit message could automatically include `git rev-list --count`+1. Note however that each subproject might have more than one commit in their history between the commits in the master repo. For example; write a new feature as a series of patches against both llvm & clang open a pull request somebody pushes some "approval" button (in Phabricator?) both patch series are rebased onto llvm & clang's TOT a single commit is created in the master repo, capturing how the llvm & clang submodules jumped from the commit before my patches, to the last commit in each series.
_______________________________________________ lldb-dev mailing list lldb-dev@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-dev
