All,

(please keep me on CC:, I'm not on the list. thx)

I think I may have painted myself into a corner, and am seeking advice 
(even help ;-).

In order to benefit from the recent putback of ofmt.c et. al to Nevada in 
our project gate (which was based on build 107), I created a clone of both 
the project gate (ilb-merge-clone) and (on-swan) Nevada (nv-merge-clone), 
and then did the following in ilb-merge-clone, in approx. this order:

- reparent to nv-merge-clone
- hg pull -u
- hg merge (there wasn't much to do there, we have almost no overlap with 
existing files)
- hg commit
- hg reparent to our opensolaris gate
- hg recommit -m "nv_111 merge"
- hg push (to the opensolaris ilb gate)

(I'm not 100% sure at which stage I did the re-parenting)

I created webrevs vs. ilb and nevada, as well as build on sparc and x86 and 
ran rudimentary tests on a machine bfu'd with these bits.

We've had two (minor) changes pushed into the ilb gate since.

the problem (in case it's not yet obvious): in a fresh clone of the osol. 
ilb gate, when I do a webrev vs. nevada, I see *tons* of files. ofmt.c, 
which is present in the workspace, is shown as "new" file in the webrev - 
basically, it looks like hg is at a completely different rev than Nevada. I 
suspect the recommit may have been a mistake here ...

If anybody has a suggestion how I can get our gate into a sane state (ie 
where hg agrees with the code and with me :-) without recreating it, I'd be 
more than grateful to hear/read it.

TIA
Michael
-- 
Michael Schuster        http://blogs.sun.com/recursion
Recursion, n.: see 'Recursion'

Reply via email to