On Sun, Jun 29, 2008 at 09:10:00AM +0200, Mike Hommey wrote:
> Package: libtrash
> Version: 2.4-1
> Severity: important
> 
> Interestingly, it doesn't crash under gdb, but fortunately, it does
> under valgrind.
> 
> Here is the stacktrace:
> ==17769== Jump to the invalid address stated on the next line
> ==17769==    at 0x0: ???
> ==17769==    by 0x40235DD: (within /usr/lib/libtrash/libtrash.so.2.4)
> ==17769==    by 0x4023C05: fopen (in /usr/lib/libtrash/libtrash.so.2.4)
> ==17769==    by 0x67944DA: internal_setpwent (compat-pwd.c:244)
> ==17769==    by 0x6794FB9: _nss_compat_getpwuid_r (compat-pwd.c:1114)
> ==17769==    by 0x54ED301: getpwuid_r@@GLIBC_2.2.5 (getXXbyYY_r.c:226)
> ==17769==    by 0x54ECBCE: getpwuid (getXXbyYY.c:116)
> ==17769==    by 0x40B673: (within /bin/ls)
> ==17769==    by 0x4031A4: (within /bin/ls)
> ==17769==    by 0x403DE0: (within /bin/ls)
> ==17769==    by 0x4062C6: (within /bin/ls)
> ==17769==    by 0x40737F: (within /bin/ls)
> ==17769==  Address 0x0 is not stack'd, malloc'd or (recently) free'd

Here is the culprit:
real_fopen = dlvsym(RTLD_NEXT, "fopen", "GLIBC_2.1");

There is no such version of the symbol:
$ objdump -T /lib/libc.so.6 | grep ' fopen$'
0000000000064840 g    DF .text  000000000000000a  GLIBC_2.2.5 fopen

Mike



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Reply via email to