At 1:14 PM +0100 1/16/06, Frans Slothouber wrote:
>One of the questions that configure.com ask is:
>
>  "Will you be sharing your PERL_ROOT with a VAX or Alpha? [n]"
>
>this in theory allows you to build perl on a VAX, Alpha, and IX
>and let them share a common root-directory for the libs and docs.
>It does this by creating a perl executable with a different postfix
>for each type of machine.
>
>Since our shop runs Perl on all these machines it seemed it nice
>idea to try this out.
>
>The problem is that the build fails with:
>
>MCR Sys$Disk:[]miniperl.ixe "-I[.lib]" -"I[-.lib]" [.POD]PODSELECT.PL
>Extracting podselect.com (with variable substitutions)
>Copy/NoConfirm/Log [.pod]podselect.com [.lib.pods]
>%COPY-S-COPIED, DISK$PROSY_USER:[PROSY.FRANS.PERL.POD]PODSELECT.COM;1
>copied to
>DISK$PROSY_USER:[PROSY.FRANS.PERL.LIB.PODS]PODSELECT.COM;1 (6 blocks)
>%DCL-W-ACTIMAGE, error activating image SYS$DISK:[]MINIPERL.EXE
>-CLI-E-IMAGEFNF, image file not found
>$1$DGA22:[PROSY.FRANS.PERL]MINIPERL.EXE;
>%MMS-F-ABORT, For target EXTRA.PODS, CLI returned abort status: %X100388B2.

Yes, there is a bug in the way we generate extra_pods.com from
configure.com.  A manual edit of extra_pods.com should get you going
again, as you perhaps have already figured out.

>This can be fixed and the build can run until completion.
>However, a
>   mms /macro=("IXE=1") test
>then reports a lot of failed tests for the IX and Alpha, as there
>are a number of test scripts that hard-coded the name of the
>perl executable to be perl.exe
>
>
>Maybe it is a good idea for the upcomming release to skip
>the shared root question as it is currently pretty broken.
>(So we don't waste people time by letting them try something
>that does not work).

I'm reluctant to take away the ability to do shared roots just
because we don't have all the pieces working (and possibly never
will).  I would vote more in favor of a documentation update.

Thanks for the report.
-- 
________________________________________
Craig A. Berry
mailto:[EMAIL PROTECTED]

"... getting out of a sonnet is much more
 difficult than getting in."
                 Brad Leithauser

Reply via email to