Please read this email in full if you work on nanojit. You're likely to
lose work in the near future if you do not.
Today is our "nanojit merge completion" flag day. The merge is done; we
finished at 7pm pacific tonight, revision 9df70d3d88ea in
nanojit-central. This corresponds to the following two revisions in
tamarin and tracemonkey:
- There is a new mechanism being added to the tamarin and tracemonkey
repositories for importing batches of changes from nanojit-central.
The tracemonkey variant has landed already, the tamarin one will
- You really, really need to not make changes directly to the nanojit
subdirectory of tamarin or tracemonkey. If you do, they will be
clobbered (un-done) during the next automated import from
nanojit-central that touches the same file. So seriously, don't.
Great way to cause regressions.
- In order to learn how to operate the import machinery, come by the
#nanojit IRC channel on irc.mozilla.org, or look at the
http://developer.mozilla.org/en/NanojitMerge wiki page, which will
(shortly, probably late tonight or early tomorrow) contain
directions on how to do so.
- The new workflow *will* be more complex than the old one. In
particular it will require making multiple commits, and for the
first while at least, testing against multiple trees. Nonetheless I
believe it will be less pain than shutting everyone down to do these
giant merges, and each team will get faster access to one another's
fixes this way.
- The tinderbox is currently only showing a few platforms, and the
builders are all (at present) sitting on my desk. I'll be pushing
the scripts to run them up to a user repo; if you want to bring more
nanojit-central builders online, email me.
Thanks for your patience with all this. Sorry for the shouting. I just
don't want to lose anyone's work.