pkgsrc-Bugs archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: pkg/60059: math/gnumeric 10.0_2025Q4 segfaults on RPi4





El 11/3/26 a las 14:10, Ramiro Aceves escribió:


El 11/3/26 a las 11:05, Ramiro Aceves via gnats escribió:
The following reply was made to PR pkg/60059; it has been noted by GNATS.

From: Ramiro Aceves <ea1abz%gmail.com@localhost>
To: gnats-bugs%netbsd.org@localhost, pkg-manager%netbsd.org@localhost, gnats-admin%netbsd.org@localhost,
  pkgsrc-bugs%netbsd.org@localhost
Cc:
Subject: Re: pkg/60059: math/gnumeric 10.0_2025Q4 segfaults on RPi4
Date: Wed, 11 Mar 2026 11:03:20 +0100

  El 11/3/26 a las 11:00, Ramiro Aceves via gnats escribió:
  > The following reply was made to PR pkg/60059; it has been noted by GNATS.
  >
  > From: Ramiro Aceves <ea1abz%gmail.com@localhost>
  > To: gnats-bugs%netbsd.org@localhost, pkg-manager%netbsd.org@localhost, gnats-admin%netbsd.org@localhost,
  >   pkgsrc-bugs%netbsd.org@localhost
  > Cc:
  > Subject: Re: pkg/60059: math/gnumeric 10.0_2025Q4 segfaults on RPi4
  > Date: Wed, 11 Mar 2026 10:57:24 +0100
  >
  >   >   Hello Dave,
  >   >
  >   >   Thanks for the information. It compiles fine now!
  >   >
  >   >   => Automatic manual page handling
  >   >   => Generating post-install file lists
  >   >   => Checking file-check results for gnumeric-1.12.60
  >   >   => Creating binary package /usr/pkgsrc/math/gnumeric112/work/.packages/gnumeric-1.12.60.tgz
  >   >   ===> Building binary package for gnumeric-1.12.60
  >   >   => Creating binary package /usr/pkgsrc/packages/All/gnumeric-1.12.60.tgz
  >   >   ===> Installing binary package of gnumeric-1.12.60
  >   >         1450,92 real      2217,79 user       433,18 sys
  >   >   netbsd-raspa4#
  >   >
  >   >   But it segfaults the same:
  >   >
  >   >   netbsd-raspa4$ gnumeric
  >   >
  >   >   (gnumeric:28722): GVFS-RemoteVolumeMonitor-WARNING **: 22:12:15.345: cannot open directory /usr/pkg/share/gvfs/remote-volume-monitors: Falló al abrir
  >   >   la carpeta «/usr/pkg/share/gvfs/remote-volume-monitors»: No such file or directory
  >   >   [1]   Segmentation fault      gnumeric
  >   >   netbsd-raspa4$
  >   >
  >   >   In amd64 it works fine.
  >   >
  >   >   Regards.
  >   >   Ramiro.
  >   >
  >
  >   Hi, I have compiled math/gnumeric112 and misc/goffice0.10 with debugging symbols just in case it gives some clues to the programming experts.
  >   Regards.
  >
  >
  >   netbsd-raspa4$ gdb /usr/pkg/bin/gnumeric
  >   GNU gdb (GDB) 11.0.50.20200914-git
  >   Copyright (C) 2020 Free Software Foundation, Inc.
  >   License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
  >   This is free software: you are free to change and redistribute it.
  >   There is NO WARRANTY, to the extent permitted by law.
  >   Type "show copying" and "show warranty" for details.
  >   This GDB was configured as "aarch64--netbsd".
  >   Type "show configuration" for configuration details.
  >   For bug reporting instructions, please see:
  >   <https://www.gnu.org/software/gdb/bugs/>.
  >   Find the GDB manual and other documentation resources online at:
  >        <http://www.gnu.org/software/gdb/documentation/>.
  >
  >   For help, type "help".
  >   Type "apropos word" to search for commands related to "word"...
  >   Reading symbols from /usr/pkg/bin/gnumeric...
  >   (gdb) run
  >   Starting program: /usr/pkg/bin/gnumeric
  >   [New LWP 18863 of process 2372]
  >   [New LWP 12759 of process 2372]
  >   [New LWP 11072 of process 2372]
  >   [New LWP 6829 of process 2372]
  >
  >   (gnumeric:2372): GVFS-RemoteVolumeMonitor-WARNING **: 10:49:34.530: cannot open directory /usr/pkg/share/gvfs/remote-volume-monitors: Falló al abrir la carpeta
  >   «/usr/pkg/share/gvfs/remote-volume-monitors»: No such file or directory
  >   [New LWP 15933 of process 2372]
  >
  >   Thread 1 "" received signal SIGSEGV, Segmentation fault.
  >   0x0000f9130f4ca13c in fmt_fp (f=f@entry=0xf913089f6840, y=inf, y@entry=0, w=w@entry=1, p=<optimized out>,
  >        p@entry=0, fl=fl@entry=536870918, t=t@entry=70) at math/go-dtoa.c:248
  >   248            *z = y;
  >   (gdb) bt
  >   #0  0x0000f9130f4ca13c in fmt_fp (f=f@entry=0xf913089f6840, y=inf, y@entry=0, w=w@entry=1,
  >        p=<optimized out>, p@entry=0, fl=fl@entry=536870918, t=t@entry=70) at math/go-dtoa.c:248
  >   #1  0x0000f9130f4cb420 in fmt_shortest (fltyp=<optimized out>, t=70, fl=536870918, d=0xffffffe3d3f0,
  >        dst=0xf913089f6840) at math/go-dtoa.c:592
  >   #2  go_dtoa (dst=0xf913089f6840, fmt=<optimized out>) at math/go-dtoa.c:700
  >   #3  0x0000000000000000 in ?? ()
  >   Backtrace stopped: previous frame identical to this frame (corrupt stack?)
  >   (gdb)
  >
  240        if (p<0) p=6;
  241
  242        if (y) y *= 0x1p28, e2-=28;
  243
  244        if (e2<0) a=r=z=big;
  (gdb) list
  245        else a=r=z=big+sizeof(big)/sizeof(*big) - LDBL_MANT_DIG - 1;
  246
  247        do {
  248            *z = y;
  249            y = 1000000000*(y-*z++);
  250        } while (y);
  251
  252        while (e2>0) {
  253            uint32_t carry=0;
  254            int sh=MIN(29,e2);
  (gdb) list
  255            for (d=z-1; d>=a; d--) {
  256                uint64_t x = ((uint64_t)*d<<sh)+carry;
  257                *d = x % 1000000000;
  258                carry = x / 1000000000;
  259            }
  260            if (carry) *--a = carry;
  261            while (z>a && !z[-1]) z--;
  262            e2-=sh;
  263        }
  264        while (e2<0) {
  (gdb)

Hi,

Changing from "long double" to "double" inside

fmt-fp function in the file



/usr/pkgsrc/misc/goffice0.10/work/goffice-0.10.60/goffice/math
fixes de crash.

I do not know how to properly fix it only for aarch64 and if it is a proper fix.

Regards.



Sorry, the file is /usr/pkgsrc/misc/goffice0.10/work/goffice-0.10.60/goffice/math/go-dtoa.c


> Hi,
>
> Changing from "long double" to "double" inside
>
> fmt-fp function in the file
>
>
>
> /usr/pkgsrc/misc/goffice0.10/work/goffice-0.10.60/goffice/math
> fixes de crash.
>
> I do not know how to properly fix it only for aarch64 and if it is a proper fix.
>
> Regards.
>
>


Home | Main Index | Thread Index | Old Index