From r.phillips@mailbox.uq.edu.au

I've been trying to install perl 5.005_03 on a DEC Alpha 3000 400 which is
running DEC OSF/1 V2.0 Worksystem Software (Rev. 240) and encountered these
problems.

When I ran

  make test

after executing

  sh Configure
and make

one of the 190 tests failed​:

op/ord.............FAILED at test 0

and ../perl harness gives

op/ord..............dubious
  Test returned status 0 (wstat 139, 0x8b)
  test program seems to have generated a core

and the summary for harness says

Failed Test Status Wstat Total Fail Failed List of failed

op/ord.t 0 139 ?? ?? % ??
6 tests skipped, plus 14 subtests skipped.
Failed 1/190 test scripts, 99.47% okay. 0/6500 subtests failed, 100.00% okay.

By trial and error I've found that the chr 193 in the last line of ord.t
causes a core dump. In face any value over 127 in a chr instruction makes
that happen. Since I expect that will never use a value greater that 127
I've gone ahead and installed perl as is, but it would have been nice if
the core dump hadn't happened.

If you want or need any more details please let me know.

Ray Phillips

From @jhi

Ray Phillips writes​:

Dear PerlBug​:

I've been trying to install perl 5.005_03 on a DEC Alpha 3000 400 which is
running DEC OSF/1 V2.0 Worksystem Software (Rev. 240) and encountered these

Wow. That's quite, errr, vintage release. (I do not think even any
3.*-release is any more supported by Compaq.)

Failed 1/190 test scripts, 99.47% okay. 0/6500 subtests failed, 100.00% okay.

By trial and error I've found that the chr 193 in the last line of ord.t
causes a core dump. In face any value over 127 in a chr instruction makes

Off-hand, sounds like a C optimizer bug. This bug does not plague
newer DEC OSF/1 releases.

that happen. Since I expect that will never use a value greater that 127
I've gone ahead and installed perl as is, but it would have been nice if
the core dump hadn't happened.

If you want or need any more details please let me know.

Ray Phillips

--
$jhi++; # http​://www.iki.fi/jhi/
  # There is this special biologist word we use for 'stable'.
  # It is 'dead'. -- Jack Cohen