[distcc] distcc 2.14 bugs in lzo code(?)

Arkadiusz Miskiewicz arekm at pld-linux.org
Sun Jun 13 12:19:57 GMT 2004


On Sunday 13 of June 2004 14:07, Arkadiusz Miskiewicz wrote:

> ==25454== Invalid read of size 1
> ==25454==    at 0x3C114B16: _dl_close (in /lib/libc-2.3.3.so)
> ==25454==    by 0x3C11583A: (within /lib/libc-2.3.3.so)
> ==25454==    by 0x3C00AA35: _dl_catch_error (in /lib/ld-2.3.3.so)
> ==25454==    by 0x3C115764: (within /lib/libc-2.3.3.so)
> ==25454==  Address 0x1EB is not stack'd, malloc'd or (recently) free'd
> ==25454==
> ==25454== Process terminating with default action of signal 11 (SIGSEGV)
> ==25454==  Access not within mapped region at address 0x1EB
> ==25454==    at 0x3C114B16: _dl_close (in /lib/libc-2.3.3.so)
> ==25454==    by 0x3C11583A: (within /lib/libc-2.3.3.so)
> ==25454==    by 0x3C00AA35: _dl_catch_error (in /lib/ld-2.3.3.so)
> ==25454==    by 0x3C115764: (within /lib/libc-2.3.3.so)
> ==25454==
> ==25454== ERROR SUMMARY: 165 errors from 45 contexts (suppressed: 13 from
> 1) ==25454== malloc/free: in use at exit: 3818 bytes in 107 blocks.
> ==25454== malloc/free: 263 allocs, 156 frees, 412103 bytes allocated.
> ==25454== For counts of detected errors, rerun with: -v
> ==25454== searching for pointers to 107 not-freed blocks.
> ==25454== checked 1708080 bytes.
>
> off-by-one somewhere?

FYI this one was from distccd because both - distcc as client and distccd are 
segfaulting for me. DISTCC_MMAP=0 cures both. distccd is easier to debug 
using valgrind.

-- 
Arkadiusz Miśkiewicz     CS at FoE, Wroclaw University of Technology
arekm.pld-linux.org, 1024/3DB19BBD, JID: arekm.jabber.org, PLD/Linux



More information about the distcc mailing list