http://qs321.pair.com?node_id=11120635


in reply to Re^12: XS.c: loadable library and perl binaries are mismatched (got handshake key 0xc100000, needed 0xc180000)
in thread XS.c: loadable library and perl binaries are mismatched (got handshake key 0xc100000, needed 0xc180000)

Yes, indeed. Same mismatch
  • Comment on Re^13: XS.c: loadable library and perl binaries are mismatched (got handshake key 0xc100000, needed 0xc180000)

Replies are listed 'Best First'.
Re^14: XS.c: loadable library and perl binaries are mismatched (got handshake key 0xc100000, needed 0xc180000)
by syphilis (Archbishop) on Aug 12, 2020 at 10:41 UTC
    Yes, indeed. Same mismatch

    I tried to explore possibilities related to gcc-3.4.5 by building perl-5.32.0 on Windows with gcc-3.4.5, but the build failed when trying to compile Time::HiRes.
    Is anyone here (apart from rgren95) in a position to build perl-5.32.0 with gcc-3.4.5, and then see what happens when cpan -I List::MoreUtils::XS is attempted ?
    Is it still to be expected that perl should be buildable with gcc-3.4.5 ? ... or should gcc-3.4.5 be simply dismissed as something that has reached its EOL ?

    rgren95, I'm still interested in looking at that XS.c file that's generated by perl-5.32.0. (Just send it to sisyphus at cpan dot org if you haven't received my /msg notification.)

    Thinking back to bulk88's mention of Config issues, I'm also wondering whether the stuff in the List-MoreUtils-XS-0.428/inc directory could somehow be screwing things up.
    That's something probably best left to someone who knows what-why-and-how the stuff in that directory is doing whatever it does. (ie ... best left to someone that's not me.)

    Cheers,
    Rob

      I tried to explore possibilities related to gcc-3.4.5 by building perl-5.32.0 on Windows with gcc-3.4.5, but the build failed when trying to compile Time::HiRes.

      Hi,

      I encountered the same thing

      Its worth sending a link to this thread to p5p ...

      But

      RHEL4 is a couple of years out of End of Extended Lifecycle Support

      And gcc-3.4.5 is from 2004

      2004

      remember 2004?

      Oh wow :D

        I encountered the same thing

        I got excited there for a moment, thinking that you had encountered the same problem as the OP ;-)
        But, after further consideration, I realized you had probably merely encountered the same problem as I did.

        That problem (ie ours, not the OP's) turns out to be fairly simple to resolve.
        In HiRes.xs at lines 142 and 143 we have:
        /* Visual C++ 2013 and older don't have the timespec structure */ # if defined(MSC_VER) && _MSC_VER < 1900
        Turns out that gcc-3.4.5 also doesn't have the timespec structure, so I just changed line 143 to:
        # if defined(_WIN32)
        which is not a portable solution, but is good enough to get the job done if the version of gcc is 3.4.5.
        The portable solution would be:
        Updated (ie checked and corrected): #if (defined(MSC_VER) && _MSC_VER < 1900) || (defined(__MINGW32__) && +__GNUC__ < 4)
        Anyway, List::MoreUtils::XS-0.428 builds and tests fine on my gcc-3.4.5 built perl-5.32.0.
        At least I now have a functional gcc-3.4.5 build of perl-5.32.0.
        Imagine how useful that will be !! (Not very..)

        BTW, this gcc-3.4.5 build of perl-5.32.0 tested fairly well (IMO).
        The cpan/IO-Compress/t/105oneshot-gzip.t script hangs and has to be slaughtered - but then the same thing happens for me with 64-bit gcc-8.3.0 builds of perl, too.
        Test Summary Report ------------------- porting/pod_rules.t (Wsta +t: 0 Tests: 8 Failed: 1) Failed test: 2 porting/regen.t (Wsta +t: 65280 Tests: 13 Failed: 0) Non-zero exit status: 255 Parse errors: Bad plan. You planned 45 tests but ran 13. ../cpan/IO-Compress/t/105oneshot-gzip.t (Wsta +t: 256 Tests: 239 Failed: 0) Non-zero exit status: 1 Parse errors: Bad plan. You planned 1007 tests but ran 239. ../ext/IPC-Open3/t/IPC-Open3.t (Wsta +t: 0 Tests: 45 Failed: 0) TODO passed: 25 ../ext/XS-APItest/t/call_checker.t (Wsta +t: 256 Tests: 78 Failed: 1) Failed test: 1 Non-zero exit status: 1 Files=2678, Tests=1148685, 2459 wallclock secs (71.23 usr + 4.68 sys += 75.91 CPU) Result: FAIL dmake: Error code 130, while making 'test'

        Cheers,
        Rob