#1504: config/auto/arch.pm: Uninitialized value warning on Darwin due to change
in order of configuration steps
-----------------------+----------------------------------------------------
Reporter: jkeenan | Owner:
Type: bug | Status: new
Priority: normal | Milestone:
Component: configure | Version: 2.1.0
Severity: medium | Keywords: configure byteorder arch
Lang: | Patch:
Platform: darwin |
-----------------------+----------------------------------------------------
Comment(by coke):
Replying to [comment:2 jkeenan]:
> Replying to [comment:1 dukeleto]:
> > I can verify that this happens on darwin-x86 at r44753 and causes
'parrot --version' to report the wrong arch:
> >
> > {{{ This is Parrot version 2.1.0-devel built for ppc-darwin. }}}
> >
>
> So here's what I propose:
>
> 1. Coke: Can you present the rationale for moving two config steps
earlier in the process? (Note: It's probably the right thing, but
perhaps we should have moved `auto::byteorder` as well.)
Sure - init::optimize now has a dependency on the config flag 'cpuarch',
so the steps that calculate this information need to come before
init:optimize; so we moved those 2, but didn't follow the full chain of
dependencies. (This is one of the reasons why I want to eventually move
all this stuff to a mini makefile, since then we can easily declare our
step dependencies and the let the tool figure out what order to run things
in.)
> 2. dukeleto: Can you do a test where you move `auto::byteorder`
immediately before `auto::arch`, then reconfigure, build, test, reporting
any anomalies?
>
> 3. kid51: Will look at adjacent config steps and see if anything would
be harmed by taking the action proposed in (2).
>
> Thank you very much.
>
> kid51
--
Ticket URL: <https://trac.parrot.org/parrot/ticket/1504#comment:3>
Parrot <https://trac.parrot.org/parrot/>
Parrot Development
_______________________________________________
parrot-tickets mailing list
[email protected]
http://lists.parrot.org/mailman/listinfo/parrot-tickets
