{
    "mode": "man",
    "parameter": "perltru64",
    "section": "1",
    "url": "https://www.chedong.com/phpMan.php/man/perltru64/1/json",
    "generated": "2026-10-10T09:32:08Z",
    "sections": {
        "NAME": {
            "content": "perltru64 - Perl version 5 on Tru64 (formerly known as Digital UNIX formerly known as DEC\nOSF/1) systems\n",
            "subsections": []
        },
        "DESCRIPTION": {
            "content": "This document describes various features of HP's (formerly Compaq's, formerly Digital's) Unix\noperating system (Tru64) that will affect how Perl version 5 (hereafter just Perl) is\nconfigured, compiled and/or runs.\n",
            "subsections": [
                {
                    "name": "Compiling Perl 5 on Tru64",
                    "content": "The recommended compiler to use in Tru64 is the native C compiler.  The native compiler\nproduces much faster code (the speed difference is noticeable: several dozen percentages) and\nalso more correct code: if you are considering using the GNU C compiler you should use at the\nvery least the release of 2.95.3 since all older gcc releases are known to produce broken\ncode when compiling Perl.  One manifestation of this brokenness is the lib/sdbm test dumping\ncore; another is many of the op/regexp and op/pat, or ext/Storable tests dumping core (the\nexact pattern of failures depending on the GCC release and optimization flags).\n\nBoth the native cc and gcc seem to consume lots of memory when building Perl.  toke.c is a\nknown trouble spot when optimizing: 256 megabytes of data section seems to be enough.\nAnother known trouble spot is the mktables script which builds the Unicode support tables.\nThe default setting of the process data section in Tru64 should be one gigabyte, but some\nsites/setups might have lowered that.  The configuration process of Perl checks for too low\nprocess limits, and lowers the optimization for the toke.c if necessary, and also gives\nadvice on how to raise the process limits (for example: \"ulimit -d 262144\")\n\nAlso, Configure might abort with\n\nBuild a threading Perl? [n]\nConfigure[2437]: Syntax error at line 1 : 'config.sh' is not expected.\n\nThis indicates that Configure is being run with a broken Korn shell (even though you think\nyou are using a Bourne shell by using \"sh Configure\" or \"./Configure\").  The Korn shell bug\nhas been reported to Compaq as of February 1999 but in the meanwhile, the reason ksh is being\nused is that you have the environment variable BINSH set to 'xpg4'.  This causes /bin/sh to\ndelegate its duties to /bin/posix/sh (a ksh).  Unset the environment variable and rerun\nConfigure.\n"
                },
                {
                    "name": "Using Large Files with Perl on Tru64",
                    "content": "In Tru64 Perl is automatically able to use large files, that is, files larger than 2\ngigabytes, there is no need to use the Configure -Duselargefiles option as described in\nINSTALL (though using the option is harmless).\n"
                },
                {
                    "name": "Threaded Perl on Tru64",
                    "content": "If you want to use threads, you should primarily use the Perl 5.8.0 threads model by running\nConfigure with -Duseithreads.\n\nPerl threading is going to work only in Tru64 4.0 and newer releases, older operating\nreleases like 3.2 aren't probably going to work properly with threads.\n\nIn Tru64 V5 (at least V5.1A, V5.1B) you cannot build threaded Perl with gcc because the\nsystem header <pthread.h> explicitly checks for supported C compilers, gcc (at least 3.2.2)\nnot being one of them.  But the system C compiler should work just fine.\n"
                },
                {
                    "name": "Long Doubles on Tru64",
                    "content": "You cannot Configure Perl to use long doubles unless you have at least Tru64 V5.0, the long\ndouble support simply wasn't functional enough before that.  Perl's Configure will override\nattempts to use the long doubles (you can notice this by Configure finding out that the\nmodfl() function does not work as it should).\n\nAt the time of this writing (June 2002), there is a known bug in the Tru64 libc printing of\nlong doubles when not using \"e\" notation.  The values are correct and usable, but you only\nget a limited number of digits displayed unless you force the issue by using \"printf\n\"%.33e\",$num\" or the like.  For Tru64 versions V5.0A through V5.1A, a patch is expected\nsometime after perl 5.8.0 is released.  If your libc has not yet been patched, you'll get a\nwarning from Configure when selecting long doubles.\n"
                },
                {
                    "name": "DB_File tests failing on Tru64",
                    "content": "The DBFile tests (db-btree.t, db-hash.t, db-recno.t) may fail you have installed a newer\nversion of Berkeley DB into the system and the -I and -L compiler and linker flags introduce\nversion conflicts with the DB 1.85 headers and libraries that came with the Tru64.  For\nexample, mixing a DB v2 library with the DB v1 headers is a bad idea.  Watch out for\nConfigure options -Dlocincpth and -Dloclibpth, and check your /usr/local/include and\n/usr/local/lib since they are included by default.\n\nThe second option is to explicitly instruct Configure to detect the newer Berkeley DB\ninstallation, by supplying the right directories with \"-Dlocincpth=/some/include\" and\n\"-Dloclibpth=/some/lib\" and before running \"make test\" setting your LDLIBRARYPATH to\n/some/lib.\n\nThe third option is to work around the problem by disabling the DBFile completely when build\nPerl by specifying -Uidb to Configure, and then using the BerkeleyDB module from CPAN\ninstead of DBFile.  The BerkeleyDB works with Berkeley DB versions 2.* or greater.\n\nThe Berkeley DB 4.1.25 has been tested with Tru64 V5.1A and found to work.  The latest\nBerkeley DB can be found from <http://www.sleepycat.com>.\n"
                },
                {
                    "name": "64-bit Perl on Tru64",
                    "content": "In Tru64 Perl's integers are automatically 64-bit wide, there is no need to use the Configure\n-Duse64bitint option as described in INSTALL.  Similarly, there is no need for -Duse64bitall\nsince pointers are automatically 64-bit wide.\n"
                },
                {
                    "name": "Warnings about floating-point overflow when compiling Perl on Tru64",
                    "content": "When compiling Perl in Tru64 you may (depending on the compiler release) see two warnings\nlike this\n\ncc: Warning: numeric.c, line 104: In this statement, floating-point\noverflow occurs in evaluating the expression \"1.8e308\". (floatoverfl)\nreturn HUGEVAL;\n-----------^\n\nand when compiling the POSIX extension\n\ncc: Warning: const-c.inc, line 2007: In this statement, floating-point\noverflow occurs in evaluating the expression \"1.8e308\". (floatoverfl)\nreturn HUGEVAL;\n-------------------^\n\nThe exact line numbers may vary between Perl releases.  The warnings are benign and can be\nignored: in later C compiler releases the warnings should be gone.\n\nWhen the file ppsys.c is being compiled you may (depending on the operating system release)\nsee an additional compiler flag being used: \"-DNOEFFONLYOK\".  This is normal and refers to\na feature that is relevant only if you use the \"filetest\" pragma.  In older releases of the\noperating system the feature was broken and the NOEFFONLYOK instructs Perl not to use the\nfeature.\n"
                }
            ]
        },
        "Testing Perl on Tru64": {
            "content": "During \"make test\" the \"comp\"/\"cpp\" will be skipped because on Tru64 it cannot be tested\nbefore Perl has been installed.  The test refers to the use of the \"-P\" option of Perl.\n\next/ODBMFile/odbm Test Failing With Static Builds\nThe ext/ODBMFile/odbm is known to fail with static builds (Configure -Uusedl) due to a known\nbug in Tru64's static libdbm library.  The good news is that you very probably don't need to\never use the ODBMFile extension since more advanced NDBMFile works fine, not to mention the\neven more advanced DBFile.\n",
            "subsections": []
        },
        "Perl Fails Because Of Unresolved Symbol sockatmark": {
            "content": "If you get an error like\n\nCan't load '.../OSF1/lib/perl5/5.8.0/alpha-decosf/auto/IO/IO.so' for module IO: Unresolved symbol in .../lib/perl5/5.8.0/alpha-decosf/auto/IO/IO.so: sockatmark at .../lib/perl5/5.8.0/alpha-decosf/XSLoader.pm line 75.\n\nyou need to either recompile your Perl in Tru64 4.0D or upgrade your Tru64 4.0D to at least\n4.0F: the sockatmark() system call was added in Tru64 4.0F, and the IO extension refers that\nsymbol.\n\nreadcurobjinfo: bad file magic number\nYou may be mixing the Tru64 cc/ar/ld with the GNU gcc/ar/ld.  That may work, but sometimes it\ndoesn't (your gcc or GNU utils may have been compiled for an incompatible OS release).\n\nTry 'which ld' and 'which ld' (or try 'ar --version' and 'ld --version', which work only for\nthe GNU tools, and will announce themselves to be such), and adjust your PATH so that you are\nconsistently using either the native tools or the GNU tools.  After fixing your PATH, you\nshould do 'make distclean' and start all the way from running the Configure since you may\nhave quite a confused situation.\n",
            "subsections": []
        },
        "AUTHOR": {
            "content": "Jarkko Hietaniemi <jhi@iki.fi>\n\nperl v5.38.2                                 2026-08-18                                 PERLTRU64(1)",
            "subsections": []
        }
    },
    "summary": "perltru64 - Perl version 5 on Tru64 (formerly known as Digital UNIX formerly known as DEC OSF/1) systems",
    "flags": [],
    "examples": [],
    "see_also": []
}