1. Forum
  2. Usenet
  3. LINUX.DEBIAN.PORTS.HPPA

  • of possible interest to hppa people

    From David Bremner@21:1/5 to All on Wed Apr 19 16:30:01 2023
    I'm not sure of the history of this bug, but it seems not to have
    attracted much intersted from porters in the last decade or so

    https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=559639

    It may be hopeless, but if someone with a particular interest in hppa
    wants to tackle it, please follow up on the bug. I am nearly certain
    that the underlying issue is different now, since racket has changed a
    lot in the meantime.

    d

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From John David Anglin@21:1/5 to All on Wed Apr 19 18:30:01 2023
    The current version of racket fails to build because of a typo:

    gcc -Wdate-time -D_FORTIFY_SOURCE=2 -Ibc -Ibc/rktio -I/<<PKGBUILDDIR>>/src/rktio -I/<<PKGBUILDDIR>>/src/bc/include
    -I/<<PKGBUILDDIR>>/src/version -I/<<PKGBUILDDIR>>/src/bc/src -Ibc/phase2 -g -O2 -ffile-prefix-map=/<<PKGBUILDDIR>>=. -Wformat
    -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -DUSE_SENORA_GC -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -o bc/src/setjmpup.o -c
    /<<PKGBUILDDIR>>/src/bc/src/setjmpup.c
    /<<PKGBUILDDIR>>/src/bc/src/setjmpup.c: In function ‘scheme_uncopy_stack’: /<<PKGBUILDDIR>>/src/bc/src/setjmpup.c:337:30: error: ‘struct Scheme_Cont’ has no member named ‘buf’
      337 |                     + c->cont->buf.stack_size)
          |                              ^~ compile failed

    I believe "buf" should be "buf_ptr".

    Dave
    --

    John David Anglin dave.anglin@bell.net

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Helge Deller@21:1/5 to David Bremner on Wed Apr 19 18:40:01 2023
    Hi David,

    On 4/19/23 16:16, David Bremner wrote:
    I'm not sure of the history of this bug, but it seems not to have
    attracted much intersted from porters in the last decade or so

    https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=559639

    It may be hopeless, but if someone with a particular interest in hppa
    wants to tackle it, please follow up on the bug. I am nearly certain
    that the underlying issue is different now, since racket has changed a
    lot in the meantime.

    Thanks a lot for pointing to the bug!

    It seems racket fails to build on various (secondary) platforms like
    ia64, m68k, sh4 and x32 too: https://buildd.debian.org/status/package.php?p=racket&suite=sid

    I haven't looked myself into the issue yet, but racket seems to be some kind
    of programming language, which might not be easy to port. I'm not sure!
    It doesn't seem to be used by many people either, as popcon sees "only" ~ 450 installations:
    https://qa.debian.org/popcon.php?package=racket

    So, I wonder if there would be anyone using it on hppa at all, and thus if
    it makes sense to put efforts in fixing racket for hppa (and some other 2nd arches)?
    But even that said, maybe someone else is interested to check?

    Helge

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From John David Anglin@21:1/5 to John David Anglin on Wed Apr 19 20:00:01 2023
    On 2023-04-19 12:27 p.m., John David Anglin wrote:
    The current version of racket fails to build because of a typo:

    gcc -Wdate-time -D_FORTIFY_SOURCE=2 -Ibc -Ibc/rktio -I/<<PKGBUILDDIR>>/src/rktio -I/<<PKGBUILDDIR>>/src/bc/include
    -I/<<PKGBUILDDIR>>/src/version -I/<<PKGBUILDDIR>>/src/bc/src -Ibc/phase2 -g -O2 -ffile-prefix-map=/<<PKGBUILDDIR>>=. -Wformat
    -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 -DUSE_SENORA_GC -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -o bc/src/setjmpup.o -c
    /<<PKGBUILDDIR>>/src/bc/src/setjmpup.c /<<PKGBUILDDIR>>/src/bc/src/setjmpup.c: In function ‘scheme_uncopy_stack’:
    /<<PKGBUILDDIR>>/src/bc/src/setjmpup.c:337:30: error: ‘struct Scheme_Cont’ has no member named ‘buf’
      337 |                     + c->cont->buf.stack_size)
          |                              ^~ compile failed
    Things go a bit further with the following change:

    dave@mx3210:~/debian/racket/racket-8.7+dfsg1/src/bc/src$ cat setjmpup.c.d
    --- setjmpup.c.save     2022-11-15 20:49:35.000000000 +0000
    +++ setjmpup.c  2023-04-19 17:26:34.316615331 +0000
    @@ -334,7 +334,7 @@
         if (c->cont) {
     #ifdef STACK_GROWS_UP
           top_delta = (((uintptr_t)c->cont->buf_ptr->buf.stack_from -                   + c->cont->buf.stack_size) +                   + (uintptr_t)c->stack_size)
                       - (uintptr_t)c->stack_from);
     #else
           bottom_delta = ((uintptr_t)c->stack_from

    bc/racketcgc1 -O info'@'compiler/cm -l- setup --boot /home/dave/debian/racket/racket-8.7+dfsg1/src/setup-go.rkt bc/compiled ignored bc/ignored.d
    /home/dave/debian/racket/racket-8.7+dfsg1/src/bc/src/compile-startup.rkt bc/phase2/cstartup.inc bc/cstartup.zo
    /home/dave/debian/racket/racket-8.7+dfsg1/src/bc/src/startup.inc /home/dave/debian/racket/racket-8.7+dfsg1/src/version/racket_version.h
    failed
     in build-one
     in loop
     in module->hash
    make[1]: *** [Makefile:18: all] Error