Subject: Re: clamav 0.72 segfault
Hi,
On Jun 10, 2004, at 4:05 PM, Kevin Horton wrote:
I get a repeatable segfault when clamscan 0.72 tries to scan a certain large Word file.
sudo clamscan /Users/kwh/Documents/Flying/RV/Avionics:Panel/gyros/gyros.doc
zsh: 23956 segmentation fault sudo clamscan /Users/kwh/Documents/Flying/RV/Avionics:Panel/gyros/gyros.doc
I wasn't sure whether it was a zsh or a clamav segfault, so I tried it from bash - still segfaults.
clamav 0.71 handles the same file just fine.
I'm running Mac OS X 10.3.4, with clamav installed via the Fink project. http://fink.sourceforge.net/
I can confirm the problem (using tcsh for that matter).
Here the crashlog from Mac OS X 10.3.4:
Command: clamscan Path: /sw/bin/clamscan Version: ??? (???) PID: 27690 Thread: 0
Exception: EXC_BAD_ACCESS (0x0001) Codes: KERN_INVALID_ADDRESS (0x0001) at 0xb9008200
Thread 0 Crashed:
0 <<00000000>> 0xffff8a48 __memcpy + 0x2a8
1 clamscan 0x0001323c ole2_read_block + 0xe8 (ole2_extract.c:283)
2 clamscan 0x000133d4 ole2_get_next_xbat_block + 0xcc (ole2_extract.c:339)
3 clamscan 0x00013d80 handler_writefile + 0x294 (ole2_extract.c:627)
4 clamscan 0x0001396c ole2_walk_property_tree + 0x234 (ole2_extract.c:516)
5 clamscan 0x000139c8 ole2_walk_property_tree + 0x290 (ole2_extract.c:524)
6 clamscan 0x000139c8 ole2_walk_property_tree + 0x290 (ole2_extract.c:524)
7 clamscan 0x000139e8 ole2_walk_property_tree + 0x2b0 (ole2_extract.c:526)
8 clamscan 0x0001421c cli_ole2_extract + 0x228 (ole2_extract.c:776)
9 clamscan 0x0000e368 cli_scanole2 + 0x98 (scanners.c:893)
10 clamscan 0x0000e7c0 cli_magic_scandesc + 0x2c0 (scanners.c:1026)
11 clamscan 0x0000e8b0 cl_scandesc + 0x1c (scanners.c:1058)
12 clamscan 0x00006c08 checkfile + 0x80 (manager.c:832)
13 clamscan 0x0000600c scanfile + 0x6a8 (manager.c:513)
14 clamscan 0x0000768c treewalk + 0x158 (treewalk.c:72)
15 clamscan 0x00007664 treewalk + 0x130 (treewalk.c:69)
16 clamscan 0x00007664 treewalk + 0x130 (treewalk.c:69)
17 clamscan 0x00007664 treewalk + 0x130 (treewalk.c:69)
18 clamscan 0x00007664 treewalk + 0x130 (treewalk.c:69)
19 clamscan 0x000057f0 scanmanager + 0x808 (manager.c:311)
20 clamscan 0x00004250 clamscan + 0x390 (clamscan.c:149)
21 clamscan 0x00004888 main + 0x190 (options.c:149)
22 clamscan 0x00001d7c _start + 0x188 (crt.c:267)
23 dyld 0x8fe1a558 _dyld_start + 0x64
and here the stack trace from gdb:
#0 0xffff8a48 in __memcpy ()
#1 0x0001323c in __gmpz_rrandomb (x=0x9, rstate=0xbfffd940, nbits=4294967295) at rrandomb.c:45
#2 0x000133d4 in __gmpz_scan0 (u=0xbfffd940, starting_bit=3103818240) at scan0.c:47
#3 0x00013d80 in __gmpz_setbit (d=0xbfffd940, bit_index=9) at setbit.c:28
#4 0x0001396c in digit_value_in_base (c=-1241513472, base=-1) at /usr/include/ctype.h:208
#5 0x000139c8 in digit_value_in_base (c=-1241513472, base=-1) at set_str.c:41
#6 0x000139c8 in digit_value_in_base (c=-1241513472, base=-1) at set_str.c:41
#7 0x000139e8 in digit_value_in_base (c=-1241513472, base=-1) at set_str.c:41
#8 0x0001421c in __gmpz_sqrt (root=0xbfffd940, op=0x200) at sqrt.c:83
#9 0x0000e368 in cli_scanole2 () at ../gmp.h:1983
#10 0x0000e7c0 in __gmpz_init_set_si (dest=0xb6000200, val=-1) at iset_si.c:38
#11 0x0000e8b0 in cl_scandesc () at iset_ui.c:32
#12 0x00006c08 in __gmpf_ceil_or_floor (r=0xbfffd940, u=0xb9008200, dir=512) at ceilfloor.c:55
#13 0x0000600c in __gmpf_mul_2exp (r=0xbfffe820, u=0xb9008200, exp=9) at mul_2exp.c:87
#14 0x0000768c in treewalk () at aors.h:120
#15 0x00007664 in treewalk () at aors.h:119
#16 0x00007664 in treewalk () at aors.h:119
#17 0x00007664 in treewalk () at aors.h:119
#18 0x00007664 in treewalk () at aors.h:119
#19 0x000057f0 in __gmpf_div (r=0x13aec, u=0xb9008200, v=0x200) at div.c:99
#20 0x00004250 in __gmpf_sub (r=0x9, u=0x0, v=0x200) at sub.c:192
#21 0x00004888 in main () at sub.c:384
I rebuilt gmp as this stack trace points t gmp, but no avail.
Cheers, Remi
------------------------------------------------------- This SF.Net email is sponsored by the new InstallShield X.
From Windows to Linux, servers to mobile, InstallShield X is the
one installation-authoring solution that does it all. Learn more and evaluate today! http://www.installshield.com/Dev2Dev/0504 _______________________________________________ Clamav-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/clamav-users
