Differences between revisions 46 and 48 (spanning 2 versions)
Revision 46 as of 2006-04-14 13:17:47
Size: 5978
Editor: ?smarenka
Comment:
Revision 48 as of 2006-04-14 17:02:53
Size: 5948
Editor: ?smarenka
Comment:
Deletions are marked like this. Additions are marked like this.
Line 17: Line 17:

 * ICE in instantiate_virtual_regs_lossage
  * debian #[http://bugs.debian.org/333536 333536], gnu pr [http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25043 25043] (m68k not primary platform)
  * from 25043: fixed in 4.1, 4.2, fails 4.0.3
  * found in gcc-4.0 4.0.2-7
  * fixed in debian gcc-4.1 4.1-0exp4
  * pkgs: cl-sql_3.5.3-1, ntop_3:3.2-1, plptools_0.14-1, rsplib_1.0.0-1, squashfs_1:2.2r2-1, xfsprogs_2.7.7-1 ksymoops_2.4.11-1, icu_3.4.1-1
Line 44: Line 37:
 * error: unable to find a register to spill in class 'ADDR_REGS'
  * debian #[http://bugs.debian.org/323016 323016]
  * seems like this was fixed, but it just showed up again with ruby1.9 and 4.0.2-7.
  * fixed in gcc-4.1 4.1-0exp4

 * ICE: output_operand: invalid expression as operand:
  * debian #[http://bugs.debian.org/270340 270340], gnu pr 18592 (last updated 20050723)
  * found in gcc-4.0 4.0.2-5
  * pkgs: gnutls11_1.0.16-14, openmsx_0.5.2-4, zynaddsubfx_2.2.1-2.1 log4cxx_0.9.7-4 (w/gcc-4.0) amphetamine_0.8.10-8.1
  * The original test case doesn't fail, so this bug differs from 18592.
  * fixed in gcc-4.1 4.1-0exp4
Line 67: Line 48:
  * test vs. gcc-4.1

 * ICE: segfault
  * debian #[http://bugs.debian.org/343692 343692]
  * found in gcc-4.0 4.0.2-7
  * base-installer_1.45, gst-ffmpeg, asterisk_1:1.0.9.dfsg-5, ghc6, openal, ht_0.9.1-3
  * seems to be a gnu pr [http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23728 23728]
  * are these all the same thing?
Line 95: Line 68:
 * cmake: Illegal instruction cmake
   * debian #[http://bugs.debian.org/360586 360586]
   * found in gcc-4.0 gcc-4.1
   * pkgs affected: vtk, octaviz
   * rebuilding cmake with -O0 allows vtk to build

Line 101: Line 81:
 * vtk_4.4.2-8
   * cmake problems, filed as [http://bugs.debian.org/360586 360586].
     * throwing an instance of 'std::logic_error'
     * Illegal instruction cmake
   * rebuilding cmake with -O0 allows vtk to build
   * test rebuilding cmake with 4.1 shows no diff
Line 111: Line 84:


== fixed in gcc-4.1 ==

 * ICE in instantiate_virtual_regs_lossage
  * debian #[http://bugs.debian.org/333536 333536], gnu pr [http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25043 25043] (m68k not primary platform)
  * from 25043: fixed in 4.1, 4.2, fails 4.0.3
  * found in gcc-4.0 4.0.2-7
  * fixed in debian gcc-4.1 4.1-0exp4
  * pkgs: cl-sql_3.5.3-1, ntop_3:3.2-1, plptools_0.14-1, rsplib_1.0.0-1, squashfs_1:2.2r2-1, xfsprogs_2.7.7-1 ksymoops_2.4.11-1, icu_3.4.1-1

 * error: unable to find a register to spill in class 'ADDR_REGS'
  * debian #[http://bugs.debian.org/323016 323016]
  * seems like this was fixed, but it just showed up again with ruby1.9 and 4.0.2-7.
  * fixed in gcc-4.1 4.1-0exp4

 * ICE: output_operand: invalid expression as operand:
  * debian #[http://bugs.debian.org/270340 270340], gnu pr 18592 (last updated 20050723)
  * found in gcc-4.0 4.0.2-5
  * pkgs: gnutls11_1.0.16-14, openmsx_0.5.2-4, zynaddsubfx_2.2.1-2.1 log4cxx_0.9.7-4 (w/gcc-4.0) amphetamine_0.8.10-8.1
  * The original test case doesn't fail, so this bug differs from 18592.
  * fixed in gcc-4.1 4.1-0exp4

 * ICE: segfault
  * debian #[http://bugs.debian.org/343692 343692]
  * found in gcc-4.0 4.0.2-7
  * base-installer_1.45, gst-ffmpeg, asterisk_1:1.0.9.dfsg-5, ghc6, openal, ht_0.9.1-3
  * seems to be a gnu pr [http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23728 23728]
  * are these all the same thing?
  * fixed in gcc-4.1 4.1.0-1

This page tracks current issues with ["GCC"] for ?DebianM68kPorting. User support is at debian-68k@lists.debian.org. Buildd admins can be found at m68k-build@nocrew.org.

m68k usertags

gcc-4.0 bugs

upstream

not upstream yet

  • ICE reload_cse_simplify_operands
    • debian #[http://bugs.debian.org/340293 340293], gnu pr 24912 (last updated 20051121)

    • nmh, openmsx_0.6.0-1
    • from 24912: fixed in trunk? (gnu pr is closed)
    • found in gcc-4.0 4.0.2-7, found in 4.1.0-1
    • The original test case doesn't fail, so this bug differs from 24912.
  • fftw3: ICE in reload
  • boost ICE
  • gij-4.0: dies with sigabrt
  • Assembler messages: Error: value out of range
    • debian #[http://bugs.debian.org/359281 359281]

    • found in gcc-4.0 4.0.2-3, gcc-4.1 4.1-0exp9
    • pkgs: torcs_1.2.4-1, grace_1:5.1.18-4 ([http://bugs.debian.org/338433 338433]), rscheme_0.7.3.4.b1-2.1

    • works fine with latest binutils and gcc-3.3, but not gcc-4.0.
    • torcs & grace work with -O1 (both gcc-4.1 & gcc-4.0)

    • rscheme works with gcc-4.1 -O2 & gcc-4.0 -O1

  • cmake: Illegal instruction cmake
    • debian #[http://bugs.debian.org/360586 360586]

    • found in gcc-4.0 gcc-4.1
    • pkgs affected: vtk, octaviz
    • rebuilding cmake with -O0 allows vtk to build

no gcc bug filed yet

  • segfaulting exe's (like csound?)
    • found in gcc-4.0 4.0.2-9
    • pkgs: crystalspace_0.99-20060125-1 drscheme_1:301-12 uim_1:1.0.0-1.1? pnet asymptote emacs-snapshot libgc
  • pike7.2_7.2.580-4: floating point exception
    • retry with modern gcc-4.0
    • already gone from unstable, so don't waste time on it.

fixed in gcc-4.1

  • ICE in instantiate_virtual_regs_lossage
  • error: unable to find a register to spill in class 'ADDR_REGS'
    • debian #[http://bugs.debian.org/323016 323016]

    • seems like this was fixed, but it just showed up again with ruby1.9 and 4.0.2-7.
    • fixed in gcc-4.1 4.1-0exp4
  • ICE: output_operand: invalid expression as operand:
    • debian #[http://bugs.debian.org/270340 270340], gnu pr 18592 (last updated 20050723)

    • found in gcc-4.0 4.0.2-5
    • pkgs: gnutls11_1.0.16-14, openmsx_0.5.2-4, zynaddsubfx_2.2.1-2.1 log4cxx_0.9.7-4 (w/gcc-4.0) amphetamine_0.8.10-8.1
    • The original test case doesn't fail, so this bug differs from 18592.
    • fixed in gcc-4.1 4.1-0exp4
  • ICE: segfault

binNMUs

gcc-3.4

  • fseries_221.10065-1: ICE in verify_initial_elim_offsets
    • found in g77-3.4 3.4.5-2
    • fixed in gfortran-4.0 4.0.2-10
    • working around with -O1