[Bug 205928] Ipfw segmentation fault with nat command

[bugzilla-noreply at freebsd.org]

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=205928

            Bug ID: 205928
           Summary: Ipfw segmentation fault with nat command
           Product: Base System
           Version: 10.1-RELEASE
          Hardware: amd64
                OS: Any
            Status: New
          Severity: Affects Only Me
          Priority: ---
         Component: bin
          Assignee: freebsd-bugs at FreeBSD.org
          Reporter: mk at steepath.eu
                CC: freebsd-amd64 at FreeBSD.org
                CC: freebsd-amd64 at FreeBSD.org

I was testing nat with ipfw, and during my test (reading example file), a
segmentation fault was found after executing one example command.

Here the bug:
----------------------------------------------------------------------
$ ipfw add nat
Segmentation fault
$ echo $?
139
----------------------------------------------------------------------

Here a small investigation:
----------------------------------------------------------------------
# cd /usr/src/sbin/ipfw
# make DEBUG_FLAGS=-g
$ gdb ./ipfw
(gdb) run add nat
Program received signal SIGSEGV, Segmentation fault.
0x0000000800b85fbb in strlen () from /lib/libc.so.7
(gdb) i r
rax            0x1b     27
rbx            0x6183f0 6390768
rcx            0x0      0
rdx            0x0      0
rsi            0x0      0
rdi            0x0      0
rbp            0x7fffffffd2e0   0x7fffffffd2e0
rsp            0x7fffffffd2e0   0x7fffffffd2e0
r8             0xfffff80235122920       -8786612704992
r9             0x0      0
r10            0x0      0
r11            0x246    582
r12            0x801406050      34380734544
r13            0x801406058      34380734552
r14            0x0      0
r15            0x3      3
rip            0x800b85fbb      0x800b85fbb <strlen+11>
eflags         0x10246  66118
cs             0x43     67
ss             0x3b     59
ds             0x0      0
es             0x0      0
fs             0x0      0
gs             0x0      0
(gdb) bt
#0  0x0000000800b85fbb in strlen () from /lib/libc.so.7
#1  0x0000000000405c43 in ipfw_add (av=0x801406058) at ipfw2.c:527
#2  0x0000000000410182 in ipfw_main (oldac=<value optimized out>,
       oldav=<value optimized out>)
    at main.c:407
    #3  0x000000000040f55b in main (ac=3, av=0x7fffffffe688) at
       main.c:620
----------------------------------------------------------------------

after adding some c code into main.c:
----------------------------------------------------------------------
$ svnlite diff ipfw2.c
Index: ipfw2.c
===================================================================
--- ipfw2.c     (revision 291749)
+++ ipfw2.c     (working copy)
@@ -522,6 +522,7 @@
 int
 _substrcmp(const char *str1, const char* str2)
 {
+       printf("str1: %s, str2: %s\n", str1, str2);

        if (strncmp(str1, str2, strlen(str1)) != 0)
                return 1;
@@ -2865,6 +2866,7 @@
         * Some things that need to go out of order (prob, action etc.)
         * go into actbuf[].
         */
+       printf("in\n");
        static uint32_t rulebuf[255], actbuf[255], cmdbuf[255];
        int rblen, ablen, cblen;
----------------------------------------------------------------------

This code return this:
----------------------------------------------------------------------
str1: add, str2: queue
str1: add, str2: flowset
str1: add, str2: sched
str1: add, str2: add
before: 
in
str1: (null), str2: global
Segmentation fault
----------------------------------------------------------------------

So, ipfw try to compare a null-pointer to a const char* in strncmp() function.
I don't know if this segmentation fault is due to this last function, or linked
to another issue.

If you have any idea. ;)

-- 
You are receiving this mail because:
You are the assignee for the bug.