svn commit: r359820 - user/pho/stress2/misc
Peter Holm
pho at freebsd.org
Sun Apr 12 18:22:17 UTC 2020
On Sun, Apr 12, 2020 at 12:25:39PM -0400, Mark Johnston wrote:
> On Sun, Apr 12, 2020 at 05:59:20AM +0000, Peter Holm wrote:
> > Author: pho
> > Date: Sun Apr 12 05:59:20 2020
> > New Revision: 359820
> > URL: https://svnweb.freebsd.org/changeset/base/359820
> >
> > Log:
> > Added more syzkaller reproducers.
> >
> > Added:
> > user/pho/stress2/misc/syzkaller5.sh (contents, props changed)
> > user/pho/stress2/misc/syzkaller6.sh (contents, props changed)
> > user/pho/stress2/misc/syzkaller7.sh (contents, props changed)
> > [...]
> > Added: user/pho/stress2/misc/syzkaller7.sh
> > ==============================================================================
> > --- /dev/null 00:00:00 1970 (empty, because file is newly added)
> > +++ user/pho/stress2/misc/syzkaller7.sh Sun Apr 12 05:59:20 2020 (r359820)
> > @@ -0,0 +1,163 @@
> > +#!/bin/sh
> > +
> > +# panic: Duplicate free of 0xfffff800049ad800 from zone
> > +# 0xfffff800041e82c0(mbuf) slab 0xfffff800049adf90(8)
> > +#
> > +# KDB: stack backtrace:
> > +# db_trace_self_wrapper() at db_trace_self_wrapper+0x47/frame
> > +# 0xfffffe0016b2c4a0
> > +# vpanic() at vpanic+0x1e0/frame 0xfffffe0016b2c500
> > +# panic() at panic+0x43/frame 0xfffffe0016b2c560
> > +# uma_dbg_free() at uma_dbg_free+0x246/frame 0xfffffe0016b2c5b0
> > +# uma_zfree_arg() at uma_zfree_arg+0x1aa/frame 0xfffffe0016b2c640
> > +# uipc_ready() at uipc_ready+0x19f/frame 0xfffffe0016b2c690
> > +# sendfile_iodone() at sendfile_iodone+0x342/frame 0xfffffe0016b2c6f0
> > +# vnode_pager_generic_getpages_done_async() at
> > +# vnode_pager_generic_getpages_done_async+0x4a/frame 0xfffffe0016b2c720
> > +# bufdone() at bufdone+0xa1/frame 0xfffffe0016b2c7a0
> > +# g_io_deliver() at g_io_deliver+0x35b/frame 0xfffffe0016b2c800
> > +# g_io_deliver() at g_io_deliver+0x35b/frame 0xfffffe0016b2c860
> > +# g_io_deliver() at g_io_deliver+0x35b/frame 0xfffffe0016b2c8c0
> > +# g_disk_done() at g_disk_done+0x179/frame 0xfffffe0016b2c910
> > +# dadone() at dadone+0x655/frame 0xfffffe0016b2c9a0
> > +# xpt_done_process() at xpt_done_process+0x5b2/frame 0xfffffe0016b2ca00
> > +# xpt_done_td() at xpt_done_td+0x175/frame 0xfffffe0016b2ca60
> > +# fork_exit() at fork_exit+0xb0/frame 0xfffffe0016b2cab0
> > +# fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe0016b2cab0
> > +# --- trap 0, rip = 0, rsp = 0, rbp = 0 ---
> > +
> > +# $FreeBSD$
> > +
> > +# Not reproduced on r359769
>
> There is a bug in old reproducers that causes them to fail.
> Specifically, the call:
>
> syscall(SYS_mmap, 0x20000000, 0x1000000, 3, 0x1012, -1, 0);
>
> should be
>
> syscall(SYS_mmap, 0x20000000, 0x1000000, 3, 0x1012, -1, 0ul);
>
Ah, yes. Thank you.
panic: Duplicate free of 0xfffff8001528fe00 from zone 0xfffff80003b34000(mbuf) slab 0xfffff8001528ffd8(14)
cpuid = 19
time = 1586715583
KDB: stack backtrace:
db_trace_self_wrapper() at db_trace_self_wrapper+0x2b/frame 0xfffffe0126af9780
vpanic() at vpanic+0x182/frame 0xfffffe0126af97d0
panic() at panic+0x43/frame 0xfffffe0126af9830
uma_dbg_free() at uma_dbg_free+0x1f2/frame 0xfffffe0126af9870
uma_zfree_arg() at uma_zfree_arg+0x130/frame 0xfffffe0126af98c0
m_free() at m_free+0x81/frame 0xfffffe0126af98e0
uipc_ready() at uipc_ready+0xf8/frame 0xfffffe0126af9920
sendfile_iodone() at sendfile_iodone+0xb1/frame 0xfffffe0126af9960
vnode_pager_generic_getpages_done_async() at vnode_pager_generic_getpages_done_async+0x3a/frame 0xfffffe0126af9980
bufdone() at bufdone+0x6c/frame 0xfffffe0126af99f0
g_io_deliver() at g_io_deliver+0x25e/frame 0xfffffe0126af9a40
g_io_deliver() at g_io_deliver+0x25e/frame 0xfffffe0126af9a90
g_disk_done() at g_disk_done+0xed/frame 0xfffffe0126af9ad0
dadone() at dadone+0x5e0/frame 0xfffffe0126af9b20
xpt_done_process() at xpt_done_process+0x382/frame 0xfffffe0126af9b60
xpt_done_td() at xpt_done_td+0xf5/frame 0xfffffe0126af9bb0
fork_exit() at fork_exit+0x80/frame 0xfffffe0126af9bf0
fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe0126af9bf0
--- trap 0, rip = 0, rsp = 0, rbp = 0 ---
KDB: enter: panic
[ thread pid 32 tid 100163 ]
Stopped at kdb_enter+0x37: movq $0,0x10cd696(%rip)
db> x/s version
version: FreeBSD 13.0-CURRENT #0 r359769: Fri Apr 10 08:41:12 CEST 2020\012 pho at t2.osted.lan:/usr/src/sys/amd64/compile/PHO\012
db>
- Peter
> There is some discussion of the problem here:
> https://github.com/google/syzkaller/pull/1493
>
> In general, when testing old reproducers, all you need to do is manually
> adjust SYS_mmap calls as above.
>
> > +# Fixed by r359779
> > +
> > +[ `uname -p` = "i386" ] && exit 0
> > +
> > +. ../default.cfg
> > +cat > /tmp/syzkaller7.c <<EOF
> > +// https://syzkaller.appspot.com/bug?id=3a27f6d4656356facc22a9329be797ef1143d0ad
> > +// autogenerated by syzkaller (https://github.com/google/syzkaller)
> > +
> > +#define _GNU_SOURCE
> > +
> > +#include <sys/types.h>
> > +
> > +#include <pwd.h>
> > +#include <signal.h>
> > +#include <stdarg.h>
> > +#include <stdbool.h>
> > +#include <stdint.h>
> > +#include <stdio.h>
> > +#include <stdlib.h>
> > +#include <string.h>
> > +#include <sys/endian.h>
> > +#include <sys/syscall.h>
> > +#include <sys/wait.h>
> > +#include <time.h>
> > +#include <unistd.h>
> > +
> > +static void kill_and_wait(int pid, int* status)
> > +{
> > + kill(pid, SIGKILL);
> > + while (waitpid(-1, status, 0) != pid) {
> > + }
> > +}
> > +
> > +static void sleep_ms(uint64_t ms)
> > +{
> > + usleep(ms * 1000);
> > +}
> > +
> > +static uint64_t current_time_ms(void)
> > +{
> > + struct timespec ts;
> > + if (clock_gettime(CLOCK_MONOTONIC, &ts))
> > + exit(1);
> > + return (uint64_t)ts.tv_sec * 1000 + (uint64_t)ts.tv_nsec / 1000000;
> > +}
> > +
> > +static void execute_one(void);
> > +
> > +#define WAIT_FLAGS 0
> > +
> > +static void loop(void)
> > +{
> > + int iter;
> > + for (iter = 0;; iter++) {
> > + int pid = fork();
> > + if (pid < 0)
> > + exit(1);
> > + if (pid == 0) {
> > + execute_one();
> > + exit(0);
> > + }
> > + int status = 0;
> > + uint64_t start = current_time_ms();
> > + for (;;) {
> > + if (waitpid(-1, &status, WNOHANG | WAIT_FLAGS) == pid)
> > + break;
> > + sleep_ms(1);
> > + if (current_time_ms() - start < 5 * 1000)
> > + continue;
> > + kill_and_wait(pid, &status);
> > + break;
> > + }
> > + }
> > +}
> > +
> > +uint64_t r[5] = {0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff,
> > + 0xffffffffffffffff, 0xffffffffffffffff};
> > +
> > +void execute_one(void)
> > +{
> > + long res = 0;
> > + memcpy((void*)0x20001180, "./file0\000", 8);
> > + res = syscall(SYS_open, 0x20001180, 0x8240, 0);
> > + if (res != -1)
> > + r[0] = res;
> > + res = syscall(SYS_socketpair, 1, 1, 0, 0x20000100);
> > + if (res != -1) {
> > + r[1] = *(uint32_t*)0x20000100;
> > + r[2] = *(uint32_t*)0x20000104;
> > + }
> > + memcpy((void*)0x20000480, "./file0\000", 8);
> > + res = syscall(SYS_open, 0x20000480, 0x80000000000206, 0);
> > + if (res != -1)
> > + r[3] = res;
> > + res = syscall(SYS_dup, r[3]);
> > + if (res != -1)
> > + r[4] = res;
> > + *(uint64_t*)0x20000100 = 0x200002c0;
> > + memcpy((void*)0x200002c0, "\xdd", 1);
> > + *(uint64_t*)0x20000108 = 1;
> > + syscall(SYS_pwritev, r[4], 0x20000100, 1, 0);
> > + *(uint64_t*)0x20002e80 = 0x20000540;
> > + memcpy((void*)0x20000540, "\x7f", 1);
> > + *(uint64_t*)0x20002e88 = 1;
> > + syscall(SYS_pwritev, r[3], 0x20002e80, 1, 0xbf24);
> > + memcpy((void*)0x200004c0,
> > + "\x89\x88\xaa\x4a\xc3\x95\x23\x77\x54\xee\x66\xf3\x8d\xa4\xae\xf3\x47"
> > + "\x6d\x78\xb7\x1f\xe6\x0d\xb7\x4a\x9f\xb9\xc9\x99\x91\x6c\x98",
> > + 32);
> > + syscall(SYS_setsockopt, r[2], 0, 2, 0x200004c0, 0x20);
> > + syscall(SYS_fcntl, r[4], 4, 0x10044);
> > + syscall(SYS_read, r[4], 0x20000000, 0x6d999);
> > + syscall(SYS_sendfile, r[0], r[1], 0, 2);
> > +}
> > +int main(void)
> > +{
> > + syscall(SYS_mmap, 0x20000000, 0x1000000, 3, 0x1012, -1, 0);
> > + loop();
> > + return 0;
> > +}
> > +EOF
> > +mycc -o /tmp/syzkaller7 -Wall -Wextra -O2 /tmp/syzkaller7.c -lpthread ||
> > + exit 1
> > +
> > +(cd /tmp; ./syzkaller7) &
> > +sleep 60
> > +pkill -9 syzkaller7
> > +wait
> > +
> > +rm -f /tmp/syzkaller7 /tmp/syzkaller7.c /tmp/syzkaller7.core /tmp/file0
> > +exit 0
More information about the svn-src-user
mailing list