[Bug 2052929]

[Cvs-commit]

The master branch has been updated by Jakub Jelinek <jakub at gcc.gnu.org>:

https://gcc.gnu.org/g:8455d6f6cd43b7b143ab9ee19437452fceba9cc9

commit r14-9769-g8455d6f6cd43b7b143ab9ee19437452fceba9cc9
Author: Jakub Jelinek <jakub at redhat.com>
Date:   Wed Apr 3 10:02:35 2024 +0200

    libquadmath: Don't assume the storage for __float128 arguments is aligned [PR114533]
    
    With the register_printf_type/register_printf_modifier/register_printf_specifier
    APIs the C library is just told the size of the argument and is provided with
    a callback to fetch the argument from va_list using va_arg into C library provided
    memory.  The C library isn't told what alignment requirement it has, but we were
    using direct load of a __float128 value from that memory which assumes
    __alignof (__float128) alignment.
    
    The following patch fixes that by using memcpy instead.
    
    I haven't been able to reproduce an actual crash, tried
     #include <quadmath.h>
     #include <stdlib.h>
     #include <stdio.h>
    
    int main ()
    {
      __float128 r;
      int prec = 20;
      int width = 46;
      char buf[128];
    
      r = 2.0q;
      r = sqrtq (r);
      int n = quadmath_snprintf (buf, sizeof buf, "%+-#*.20Qe", width, r);
      if ((size_t) n < sizeof buf)
        printf ("%s\n", buf);
        /* Prints: +1.41421356237309504880e+00 */
      quadmath_snprintf (buf, sizeof buf, "%Qa", r);
      if ((size_t) n < sizeof buf)
        printf ("%s\n", buf);
        /* Prints: 0x1.6a09e667f3bcc908b2fb1366ea96p+0 */
      n = quadmath_snprintf (NULL, 0, "%+-#46.*Qe", prec, r);
      if (n > -1)
        {
          char *str = malloc (n + 1);
          if (str)
            {
              quadmath_snprintf (str, n + 1, "%+-#46.*Qe", prec, r);
              printf ("%s\n", str);
              /* Prints: +1.41421356237309504880e+00 */
            }
          free (str);
        }
      printf ("%+-#*.20Qe\n", width, r);
      printf ("%Qa\n", r);
      printf ("%+-#46.*Qe\n", prec, r);
      printf ("%d %Qe %d %Qe %d %Qe\n", 1, r, 2, r, 3, r);
      return 0;
    }
    In any case, I think memcpy for loading from it is right.
    
    2024-04-03  Simon Chopin  <simon.chopin at canonical.com>
                Jakub Jelinek  <jakub at redhat.com>
    
            PR libquadmath/114533
            * printf/printf_fp.c (__quadmath_printf_fp): Use memcpy to copy
            __float128 out of args.
            * printf/printf_fphex.c (__quadmath_printf_fphex): Likewise.
    
    Signed-off-by: Simon Chopin <simon.chopin at canonical.com>

-- 
You received this bug notification because you are a member of Ubuntu
Foundations Bugs, which is subscribed to glibc in Ubuntu.
https://bugs.launchpad.net/bugs/2052929

Title:
  failed autopkgtests for evolver vs glibc 2.39 on amd64

Status in gcc:
  In Progress
Status in evolver package in Ubuntu:
  New
Status in gcc-13 package in Ubuntu:
  Fix Released
Status in gcc-14 package in Ubuntu:
  Fix Released
Status in glibc package in Ubuntu:
  Invalid
Status in evolver package in Debian:
  New

Bug description:
  The autopkgests for evolver fail when run against glibc 2.39 on amd64
  with a segfault:

  3537s autopkgtest [09:43:29]: test command6: [-----------------------
  3537s Surface Evolver Version 2.70a (Debian 2.70+ds-8build1), August 27, 2013, 64-bit.
  3537s Compiled for float128, 33 digits precision.
  3537s Built with Geomview support.
  3537s 
  3537s Enter command: 
  3537s Enter command: // Typical evolution to sphere
  3537s Enter command: gogo := { g 5; r; g 5; hessian; r; g 5; hessian; }
  3537s Enter command: 
  3537s Enter command: // Evolution to very high accuracy, using higher-order Lagrange elements.
  3537s Enter command: // To be run on original datafile.
  3537s Enter command: gogo2 := { g 5; r; g 5; hessian; r; g 5; hessian;
  3537s more>            lagrange 2; g 5; hessian; 
  3537s more>            lagrange 4; g 5; hessian; 
  3537s more>            lagrange 6; g 5; hessian; 
  3537s more>            ideal_rad := (3*body[1].volume/4/pi)^(1/3);
  3537s more>            printf "Area error: %g\n",total_area - 4*pi*ideal_rad^2;
  3537s more>            printf "Vertex radius spread: %g\n",
  3537s more>              max(vertex,sqrt((x-.5)^2+(y-.5)^2+(z-.5)^2))
  3537s more>                - min(vertex,sqrt((x-.5)^2+(y-.5)^2+(z-.5)^2));
  3537s more> }
  3537s Enter command: g 5; v; r ; g 10; v;
  3537s bash: line 1:  1012 Done                    echo "g 5; v; r ; g 10; v;"
  3537s       1013 Segmentation fault      (core dumped) | evolver-nox-q cube

To manage notifications about this bug go to:
https://bugs.launchpad.net/gcc/+bug/2052929/+subscriptions