[elephant-devel] Updated version of last Postmodern patch bundle

[Robert L. Read]

Dear Leslie and Ian Alex and Henrik,
	I can see now that I did not apply Leslie's updated patch.  I attempted
to rollback and reapply the correct patch.  Unfortunately, I was unable
to get to a "green" state under any system.  I find darcs much harder to
work with than CVS (however, I had a great deal of experience with CVS.)

	So, unfortunately, Leslie, I cannot re-apply our patch.  If you have
working code, please compute another patch directly against the CURRENT
tip of the main development branch in the repository.  I will then apply
this and see if I can get to green.

	Furthermore, in attempt to solve this problem, I restored my local copy
to the head of the main repository, and am green only under BDB.

	I'm afraid things have gotten a little out of control, although I think
we are using proper discipline and committing things only when we are
green on the local systems that we can test on.  However, I would like
to get the main repository back to a "known working" state for my
systems (SBCL, BDB, Postgres, Postmodern, X86 architecture.)
	
	I think the best way to do this is for Leslie to get green and send me
a diff; if I still have problems after that I may have to look into
Ian's checkins.  As a last resort we can rollback patches until we get
to something working...but that is a last resort.

	If I can't get things worked out after Leslie sends me the patch, then
I may have to ask you guys to test against the head and send me your
results on your systems.

	Ian wants to make a release in a few weeks, and we really need to get
back to a solid state ahead of that.



On Sun, 2008-03-16 at 12:23 +0100, Leslie P. Polzer wrote:
> Dear Robert,
> 
> I posted an updated version of the patch shortly after, and it seems
> you applied the obsolete one. Can you check?
> 
> 
> > 	I did personally have some problems...perhaps because Bordeaux Threads
> > is not claiming to work on x86_64 architectures, darn it!
> 
> Hrm, I guess it should just be a thin compatibility layer for SBCL.
> Does SBCL officially support threading on ia64?
> 
> 
> > 	So the upshot is that I am green under BDB and postmodern with your
> > changes, except for your test itself, which fails for me on sbcl 1.0.13
> > on an x86_64, apparently more because of the threading than the
> > reapoing.
> 
> I will look into this when we have sorted out the patch problem.
> 
>   Leslie
> 
> 
> 
> _______________________________________________
> elephant-devel site list
> elephant-devel at common-lisp.net
> http://common-lisp.net/mailman/listinfo/elephant-devel