russell@ccu1.aukuni.ac.nz (Russell J Fulton;ccc032u) (01/09/91)
I have just applied patches 37-41 and recompiled without appearent problems. But when I try and run the test it fails the last test: chmod +x t/TEST t/base.* t/comp.* t/cmd.* t/io.* t/op.*; \ cd t && (rm -f perl; ln -s ../perl .) && ./perl TEST base.cond.......ok base.if.........ok base.lex........ok base.pat........ok [stuff deleted] op.vec..........ok op.write........ok lib.big.........FAILED on test 0 *** Error code -1 (ignored) Does anybody have any idea what the problem is? I am working on an SGI 4D/240s running Irix 3.3.1. Cheers, Russell. -- Russell Fulton, Computer Center, University of Auckland, New Zealand. <rj_fulton@aukuni.ac.nz>
russell@ccu1.aukuni.ac.nz (Russell J Fulton;ccc032u) (01/09/91)
This should be read as an appendix to my previous posting on this subject. My apologies for not doing all my homework before posting the first time. I have tried running the lib.big test on its own and find that it appears to run perfectly. When run from TEST it says that test 0 failed, but there does not appear to be a test 0. The tests go from 1..128 and it says they are all ok. Now I am thoroughly confused! -- Russell Fulton, Computer Center, University of Auckland, New Zealand. <rj_fulton@aukuni.ac.nz>
lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall) (01/11/91)
In article <SCP.91Jan9201403@grizzly.acl.lanl.gov> scp@acl.lanl.gov writes: : : In article <1991Jan8.205912.3798@ccu1.aukuni.ac.nz> russell@ccu1.aukuni.ac.nz (Russell J Fulton;ccc032u) writes: : : I have just applied patches 37-41 and recompiled without appearent problems. : But when I try and run the test it fails the last test: : : [stuff deleted] : : lib.big.........FAILED on test 0 : *** Error code -1 (ignored) That'll be fixed in 42. : Well I having even more problems than that! : : [on a 3.3.1 380VGX, using the Power C compiler, -O -D_BSD_SIGNALS, : perl 3.0 pl41, usual defaults] : : t/TEST wouldn't even run (problems with "``" evaluations) until I undef'ed : vfork. : : Then it FAILED op.eval test 0, and ran out of memory on op.s, and : FAILED lib.big test 0. As I was using perl's malloc, I reconfiged : to use the system malloc. Now op.s runs until it has consumed all : the memory on may machine, and can't possibly be killed. op.eval : unpredictably fails or goes into never-never land. It's tiresome : using the reset button on the beast! : : I'm still chasing after these, but if anyone's found the cure, I'd sure : like to hear about it! Try not using -O. Larry