Loading ...
Sorry, an error occurred while loading the content.

vim 6.1.310 not compiling

Expand Messages
  • Neil Bird
    I got the nirvani.org vim6.1-130 SRC RPM, but it doesn t rpmbuild on my RH8.0 box. Any thoughts? term.c:129: conflicting types for `ospeed
    Message 1 of 8 , Feb 3 5:08 AM
    • 0 Attachment
      I got the nirvani.org vim6.1-130 SRC RPM, but it doesn't rpmbuild on
      my RH8.0 box. Any thoughts?


      term.c:129: conflicting types for `ospeed'
      /usr/include/termcap.h:30: previous declaration of `ospeed'


      termcap.h: [libtermcap-devel-2.0.8-31]
      #ifdef __linux__
      extern speed_t ospeed; <--- here
      #else
      extern short ospeed;
      #endif

      I *think* 'speed_t' gets defined to 'unsigned int'.


      term.c:
      # ifndef HAVE_OSPEED
      # ifdef OSPEED_EXTERN
      extern short ospeed;
      # else
      short ospeed; <--- here
      # endif
      # endif

      --
      [neil@fnx ~]# rm -f .signature
      [neil@fnx ~]# ls -l .signature
      ls: .signature: No such file or directory
      [neil@fnx ~]# exit
    • Bram Moolenaar
      ... There is a configure check for ospeed. Check the output of configure for whether termcap.h contains ospeed . Why doesn t it find the declaration for
      Message 2 of 8 , Feb 3 11:28 AM
      • 0 Attachment
        Neil Bird wrote:

        > I got the nirvani.org vim6.1-130 SRC RPM, but it doesn't rpmbuild on
        > my RH8.0 box. Any thoughts?
        >
        >
        > term.c:129: conflicting types for `ospeed'
        > /usr/include/termcap.h:30: previous declaration of `ospeed'
        >
        >
        > termcap.h: [libtermcap-devel-2.0.8-31]
        > #ifdef __linux__
        > extern speed_t ospeed; <--- here
        > #else
        > extern short ospeed;
        > #endif
        >
        > I *think* 'speed_t' gets defined to 'unsigned int'.
        >
        >
        > term.c:
        > # ifndef HAVE_OSPEED
        > # ifdef OSPEED_EXTERN
        > extern short ospeed;
        > # else
        > short ospeed; <--- here
        > # endif
        > # endif

        There is a configure check for ospeed. Check the output of configure
        for "whether termcap.h contains ospeed". Why doesn't it find the
        declaration for ospeed in termcap.h? Check auto/config.log.

        --
        hundred-and-one symptoms of being an internet addict:
        101. U can read htis w/o ny porblm and cant figur eout Y its evn listd.

        /// Bram Moolenaar -- Bram@... -- http://www.Moolenaar.net \\\
        /// Creator of Vim - Vi IMproved -- http://www.Vim.org \\\
        \\\ Project leader for A-A-P -- http://www.A-A-P.org ///
        \\\ Help AIDS victims, buy at Amazon -- http://ICCF.nl/click1.html ///
      • Neil Bird
        Around about 03/02/2003 19:28, Bram Moolenaar typed ... ... Thanks; it tells me that the conftest utils. is being compiled -lelf but that filas as I don t
        Message 3 of 8 , Feb 6 1:42 AM
        • 0 Attachment
          Around about 03/02/2003 19:28, Bram Moolenaar typed ...
          > There is a configure check for ospeed. Check the output of configure
          > for "whether termcap.h contains ospeed". Why doesn't it find the
          > declaration for ospeed in termcap.h? Check auto/config.log.

          Thanks; it tells me that the conftest utils. is being compiled
          '-lelf' but that filas as I don't have libelf-devel installed (and
          that's not listed as an RPM BuildRequires).

          Not sure yet whether I should need libelf-devel or whether it's an
          error of some kind pulling in -lelf - but it's compiling now that I've
          installed it!

          --
          [neil@fnx ~]# rm -f .signature
          [neil@fnx ~]# ls -l .signature
          ls: .signature: No such file or directory
          [neil@fnx ~]# exit
        • Bram Moolenaar
          ... No, you don t need to install an extra library. The whole idea of configure is that it figures out what you have and adjusts to that. I suspect a check
          Message 4 of 8 , Feb 6 2:37 AM
          • 0 Attachment
            Neil Bird wrote:

            > Around about 03/02/2003 19:28, Bram Moolenaar typed ...
            > > There is a configure check for ospeed. Check the output of configure
            > > for "whether termcap.h contains ospeed". Why doesn't it find the
            > > declaration for ospeed in termcap.h? Check auto/config.log.
            >
            > Thanks; it tells me that the conftest utils. is being compiled
            > '-lelf' but that filas as I don't have libelf-devel installed (and
            > that's not listed as an RPM BuildRequires).
            >
            > Not sure yet whether I should need libelf-devel or whether it's an
            > error of some kind pulling in -lelf - but it's compiling now that I've
            > installed it!

            No, you don't need to install an extra library. The whole idea of
            configure is that it figures out what you have and adjusts to that.

            I suspect a check before the "ospeed" check has added "-lelf" even
            though you don't have it. Perhaps the Perl or Python check? Hopefully
            you can see that in the config.log.

            --
            hundred-and-one symptoms of being an internet addict:
            147. You finally give up smoking...because it made the monitor dirty.

            /// Bram Moolenaar -- Bram@... -- http://www.Moolenaar.net \\\
            /// Creator of Vim - Vi IMproved -- http://www.Vim.org \\\
            \\\ Project leader for A-A-P -- http://www.A-A-P.org ///
            \\\ Help AIDS victims, buy at Amazon -- http://ICCF.nl/click1.html ///
          • Neil Bird
            Around about 06/02/2003 10:37, Bram Moolenaar typed ... ... I ll have another look to see why it s picked up the -lelf ; but AFAIK, checks *in* configure
            Message 5 of 8 , Feb 6 3:23 AM
            • 0 Attachment
              Around about 06/02/2003 10:37, Bram Moolenaar typed ...
              > I suspect a check before the "ospeed" check has added "-lelf" even
              > though you don't have it. Perhaps the Perl or Python check? Hopefully
              > you can see that in the config.log.

              I'll have another look to see why it's picked up the '-lelf'; but
              AFAIK, checks *in* configure don't affect the compiler flags ($CFLAGS)
              that it uses when compiling the multitude of conftest.c files that it
              uses to /perform/ the tests.

              But as I said, I'll have a look.

              --
              [neil@fnx ~]# rm -f .signature
              [neil@fnx ~]# ls -l .signature
              ls: .signature: No such file or directory
              [neil@fnx ~]# exit
            • Bram Moolenaar
              ... While configure progresses the flags used for compiling are extended with the results of previous checks. This is required for some tests to work
              Message 6 of 8 , Feb 6 4:03 AM
              • 0 Attachment
                Neil Bird wrote:

                > Around about 06/02/2003 10:37, Bram Moolenaar typed ...
                > > I suspect a check before the "ospeed" check has added "-lelf" even
                > > though you don't have it. Perhaps the Perl or Python check? Hopefully
                > > you can see that in the config.log.
                >
                > I'll have another look to see why it's picked up the '-lelf'; but
                > AFAIK, checks *in* configure don't affect the compiler flags ($CFLAGS)
                > that it uses when compiling the multitude of conftest.c files that it
                > uses to /perform/ the tests.

                While configure progresses the flags used for compiling are extended
                with the results of previous checks. This is required for some tests to
                work properly. It may break other tests, which is probably what happens
                in this case.

                --
                hundred-and-one symptoms of being an internet addict:
                150. You find yourself counting emoticons to get to sleep.

                /// Bram Moolenaar -- Bram@... -- http://www.Moolenaar.net \\\
                /// Creator of Vim - Vi IMproved -- http://www.Vim.org \\\
                \\\ Project leader for A-A-P -- http://www.A-A-P.org ///
                \\\ Help AIDS victims, buy at Amazon -- http://ICCF.nl/click1.html ///
              • Neil Bird
                Around about 06/02/2003 10:37, Bram Moolenaar typed ... ... Had a look last night, and it s going to be tricky to find - it s in an early part of configure
                Message 7 of 8 , Feb 7 12:46 AM
                • 0 Attachment
                  Around about 06/02/2003 10:37, Bram Moolenaar typed ...
                  > I suspect a check before the "ospeed" check has added "-lelf" even
                  > though you don't have it. Perhaps the Perl or Python check? Hopefully
                  > you can see that in the config.log.

                  Had a look last night, and it's going to be tricky to find - it's in
                  an early part of configure that doesn't log the commands/tests it's
                  issuing so I'm not sure what's doing it.

                  In amongst a whole slew of early low-level tests, there's a test for
                  elf.h (that's in glibc) and for 'main in libelf' (not sure of the
                  outcome of that).

                  After this slew, all [conftest] gcc commands have had '-lelf' put
                  into them.


                  [redirecting to vim-dev]

                  --
                  [neil@fnx ~]# rm -f .signature
                  [neil@fnx ~]# ls -l .signature
                  ls: .signature: No such file or directory
                  [neil@fnx ~]# exit
                • Bram Moolenaar
                  ... What I can see in the configure script is that -lelf is added to $LIBS and linking is attempted. What possibly happens is that linking works OK, but
                  Message 8 of 8 , Feb 7 1:04 PM
                  • 0 Attachment
                    Neil Bird wrote:

                    > Around about 06/02/2003 10:37, Bram Moolenaar typed ...
                    > > I suspect a check before the "ospeed" check has added "-lelf" even
                    > > though you don't have it. Perhaps the Perl or Python check? Hopefully
                    > > you can see that in the config.log.
                    >
                    > Had a look last night, and it's going to be tricky to find - it's in
                    > an early part of configure that doesn't log the commands/tests it's
                    > issuing so I'm not sure what's doing it.
                    >
                    > In amongst a whole slew of early low-level tests, there's a test for
                    > elf.h (that's in glibc) and for 'main in libelf' (not sure of the
                    > outcome of that).
                    >
                    > After this slew, all [conftest] gcc commands have had '-lelf' put
                    > into them.

                    What I can see in the configure script is that "-lelf" is added to $LIBS
                    and linking is attempted. What possibly happens is that linking works
                    OK, but executing the resulting program doesn't work. But that isn't
                    tested.

                    Can you verify this by linking a small program with -lelf and then
                    running it?

                    --
                    hundred-and-one symptoms of being an internet addict:
                    168. You have your own domain name.

                    /// Bram Moolenaar -- Bram@... -- http://www.Moolenaar.net \\\
                    /// Creator of Vim - Vi IMproved -- http://www.Vim.org \\\
                    \\\ Project leader for A-A-P -- http://www.A-A-P.org ///
                    \\\ Help AIDS victims, buy at Amazon -- http://ICCF.nl/click1.html ///
                  Your message has been successfully submitted and would be delivered to recipients shortly.