Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1592166 > unrolled thread
| Started by | Dmitry Vyukov <dvyukov@google.com> |
|---|---|
| First post | 2017-03-03 18:40 +0100 |
| Last post | 2017-03-07 19:30 +0100 |
| Articles | 2 on this page of 22 — 3 participants |
Back to article view | Back to linux.kernel
net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-03 18:40 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-03 20:20 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone David Ahern <dsa@cumulusnetworks.com> - 2017-03-03 20:50 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-04 20:00 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-04 20:10 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Eric Dumazet <eric.dumazet@gmail.com> - 2017-03-04 21:20 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-05 12:00 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone David Ahern <dsa@cumulusnetworks.com> - 2017-03-06 18:40 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-06 20:00 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone David Ahern <dsa@cumulusnetworks.com> - 2017-03-07 01:00 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-07 09:50 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-07 11:30 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-07 19:20 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone David Ahern <dsa@cumulusnetworks.com> - 2017-03-07 19:50 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-07 20:20 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-07 20:50 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-07 21:10 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-08 13:30 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone David Ahern <dsa@cumulusnetworks.com> - 2017-03-08 05:50 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone Dmitry Vyukov <dvyukov@google.com> - 2017-03-07 18:50 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone David Ahern <dsa@cumulusnetworks.com> - 2017-03-07 18:50 +0100
Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone David Ahern <dsa@cumulusnetworks.com> - 2017-03-07 19:30 +0100
Page 2 of 2 — ← Prev page 1 [2]
| From | David Ahern <dsa@cumulusnetworks.com> |
|---|---|
| Date | 2017-03-07 18:50 +0100 |
| Subject | Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone |
| Message-ID | <tirpf-2s0-3@gated-at.bofh.it> |
| In reply to | #1593992 |
On 3/7/17 1:43 AM, Dmitry Vyukov wrote: > This is on c1ae3cfa0e89fa1a7ecc4c99031f5e9ae99d9201. No other kernel > output from your patch (pr_err). Is the below supposed to be from the same qemu instance at the time of the crash? cpu1 and cpu2 are both supposedly doing a route insert? > > ------------[ cut here ]------------ > WARNING: CPU: 1 PID: 30179 at net/ipv6/ip6_fib.c:158 > rt6_rcu_free+0x61/0x70 net/ipv6/ip6_fib.c:158 > Kernel panic - not syncing: panic_on_warn set ... > > CPU: 1 PID: 30179 Comm: syz-executor3 Not tainted 4.11.0-rc1+ #310 > Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011 > Call Trace: > __dump_stack lib/dump_stack.c:16 [inline] > dump_stack+0x2fb/0x3fd lib/dump_stack.c:52 > panic+0x20f/0x426 kernel/panic.c:180 > __warn+0x1c4/0x1e0 kernel/panic.c:541 > warn_slowpath_null+0x2c/0x40 kernel/panic.c:584 > rt6_rcu_free+0x61/0x70 net/ipv6/ip6_fib.c:158 > rt6_release+0x1ee/0x290 net/ipv6/ip6_fib.c:189 > fib6_add_rt2node net/ipv6/ip6_fib.c:922 [inline] > fib6_add+0x1d51/0x3290 net/ipv6/ip6_fib.c:1081 > __ip6_ins_rt+0x60/0x80 net/ipv6/route.c:948 > ip6_route_add+0x1a7/0x310 net/ipv6/route.c:2130 > inet6_rtm_newroute+0x191/0x1b0 net/ipv6/route.c:3294 > rtnetlink_rcv_msg+0x609/0x860 net/core/rtnetlink.c:4104 > netlink_rcv_skb+0x2ab/0x390 net/netlink/af_netlink.c:2298 > rtnetlink_rcv+0x2a/0x40 net/core/rtnetlink.c:4110 > netlink_unicast_kernel net/netlink/af_netlink.c:1231 [inline] > netlink_unicast+0x525/0x730 net/netlink/af_netlink.c:1257 > netlink_sendmsg+0xab3/0xe70 net/netlink/af_netlink.c:1803 > sock_sendmsg_nosec net/socket.c:633 [inline] > sock_sendmsg+0xca/0x110 net/socket.c:643 > sock_write_iter+0x326/0x600 net/socket.c:846 > call_write_iter include/linux/fs.h:1733 [inline] > do_iter_readv_writev fs/read_write.c:696 [inline] > __do_readv_writev+0xbbc/0x10a0 fs/read_write.c:862 > do_readv_writev+0x13f/0x200 fs/read_write.c:894 > vfs_writev+0x87/0xc0 fs/read_write.c:921 > do_writev+0x110/0x2c0 fs/read_write.c:954 > SYSC_writev fs/read_write.c:1027 [inline] > SyS_writev+0x27/0x30 fs/read_write.c:1024 > entry_SYSCALL_64_fastpath+0x1f/0xc2 > RIP: 0033:0x4458d9 > RSP: 002b:00007f31fcf33b58 EFLAGS: 00000292 ORIG_RAX: 0000000000000014 > RAX: ffffffffffffffda RBX: 0000000000000005 RCX: 00000000004458d9 > RDX: 0000000000000001 RSI: 00000000207cd000 RDI: 0000000000000005 > RBP: 00000000006e30c0 R08: 0000000000000000 R09: 0000000000000000 > R10: 0000000000000000 R11: 0000000000000292 R12: 0000000000708000 > R13: 0000000020fad000 R14: 0000000000001000 R15: 0000000000000003 > > > > ------------[ cut here ]------------ > WARNING: CPU: 2 PID: 31175 at net/ipv6/ip6_fib.c:158 > rt6_rcu_free+0x61/0x70 net/ipv6/ip6_fib.c:158 > Kernel panic - not syncing: panic_on_warn set ... > > CPU: 2 PID: 31175 Comm: syz-executor1 Not tainted 4.11.0-rc1+ #310 > Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011 > Call Trace: > __dump_stack lib/dump_stack.c:16 [inline] > dump_stack+0x2fb/0x3fd lib/dump_stack.c:52 > panic+0x20f/0x426 kernel/panic.c:180 > __warn+0x1c4/0x1e0 kernel/panic.c:541 > warn_slowpath_null+0x2c/0x40 kernel/panic.c:584 > rt6_rcu_free+0x61/0x70 net/ipv6/ip6_fib.c:158 > rt6_release+0x1ee/0x290 net/ipv6/ip6_fib.c:189 > fib6_add_rt2node net/ipv6/ip6_fib.c:922 [inline] > fib6_add+0x1d51/0x3290 net/ipv6/ip6_fib.c:1081 > kvm_vm_ioctl_deassign_device: device hasn't been assigned before, so > cannot be deassigned > __ip6_ins_rt+0x60/0x80 net/ipv6/route.c:948 > ip6_route_add+0x1a7/0x310 net/ipv6/route.c:2130 > inet6_rtm_newroute+0x191/0x1b0 net/ipv6/route.c:3294 > rtnetlink_rcv_msg+0x609/0x860 net/core/rtnetlink.c:4104 > netlink_rcv_skb+0x2ab/0x390 net/netlink/af_netlink.c:2298 > rtnetlink_rcv+0x2a/0x40 net/core/rtnetlink.c:4110 > netlink_unicast_kernel net/netlink/af_netlink.c:1231 [inline] > netlink_unicast+0x525/0x730 net/netlink/af_netlink.c:1257 > netlink_sendmsg+0xab3/0xe70 net/netlink/af_netlink.c:1803 > sock_sendmsg_nosec net/socket.c:633 [inline] > sock_sendmsg+0xca/0x110 net/socket.c:643 > sock_write_iter+0x326/0x600 net/socket.c:846 > call_write_iter include/linux/fs.h:1733 [inline] > do_iter_readv_writev fs/read_write.c:696 [inline] > __do_readv_writev+0xbbc/0x10a0 fs/read_write.c:862 > do_readv_writev+0x13f/0x200 fs/read_write.c:894 > vfs_writev+0x87/0xc0 fs/read_write.c:921 > do_writev+0x110/0x2c0 fs/read_write.c:954 > SYSC_writev fs/read_write.c:1027 [inline] > SyS_writev+0x27/0x30 fs/read_write.c:1024 > entry_SYSCALL_64_fastpath+0x1f/0xc2 > RIP: 0033:0x4458d9 > RSP: 002b:00007f1639006b58 EFLAGS: 00000292 ORIG_RAX: 0000000000000014 > RAX: ffffffffffffffda RBX: 0000000000000019 RCX: 00000000004458d9 > RDX: 0000000000000001 RSI: 00000000207cd000 RDI: 0000000000000019 > RBP: 00000000006e30c0 R08: 0000000000000000 R09: 0000000000000000 > R10: 0000000000000000 R11: 0000000000000292 R12: 0000000000708000 > R13: 0000000000000010 R14: 0000000000000003 R15: 0000000000000000 >
[toc] | [prev] | [next] | [standalone]
| From | David Ahern <dsa@cumulusnetworks.com> |
|---|---|
| Date | 2017-03-07 19:30 +0100 |
| Subject | Re: net: heap out-of-bounds in fib6_clean_node/rt6_fill_node/fib6_age/fib6_prune_clone |
| Message-ID | <tis1Y-2YE-9@gated-at.bofh.it> |
| In reply to | #1593992 |
On 3/7/17 1:43 AM, Dmitry Vyukov wrote:
> This is on c1ae3cfa0e89fa1a7ecc4c99031f5e9ae99d9201. No other kernel
> output from your patch (pr_err).
>
> ------------[ cut here ]------------
> WARNING: CPU: 1 PID: 30179 at net/ipv6/ip6_fib.c:158
> rt6_rcu_free+0x61/0x70 net/ipv6/ip6_fib.c:158
> Kernel panic - not syncing: panic_on_warn set ...
you have panic_on_warn set ...
>
> CPU: 1 PID: 30179 Comm: syz-executor3 Not tainted 4.11.0-rc1+ #310
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
> Call Trace:
> __dump_stack lib/dump_stack.c:16 [inline]
> dump_stack+0x2fb/0x3fd lib/dump_stack.c:52
> panic+0x20f/0x426 kernel/panic.c:180
> __warn+0x1c4/0x1e0 kernel/panic.c:541
> warn_slowpath_null+0x2c/0x40 kernel/panic.c:584
> rt6_rcu_free+0x61/0x70 net/ipv6/ip6_fib.c:158
and this is my WARN_ON in rt6_rcu_free which is showing an additional
change is needed
> rt6_release+0x1ee/0x290 net/ipv6/ip6_fib.c:189
> fib6_add_rt2node net/ipv6/ip6_fib.c:922 [inline]
in fib6_add_rt2node for the route replace path (whitespace damaged on
the copy-paste):
@@ -916,6 +919,7 @@ static int fib6_add_rt2node(struct fib6_node *fn,
struct rt6_info *rt,
}
nsiblings = iter->rt6i_nsiblings;
fib6_purge_rt(iter, fn, info->nl_net);
+ iter->dst.flags &= ~DST_IN_FIB;
rt6_release(iter);
if (nsiblings) {
@@ -926,6 +930,7 @@ static int fib6_add_rt2node(struct fib6_node *fn,
struct rt6_info *rt,
if (rt6_qualify_for_ecmp(iter)) {
*ins = iter->dst.rt6_next;
fib6_purge_rt(iter, fn,
info->nl_net);
+ iter->dst.flags &= ~DST_IN_FIB;
rt6_release(iter);
nsiblings--;
} else {
[toc] | [prev] | [standalone]
Page 2 of 2 — ← Prev page 1 [2]
Back to top | Article view | linux.kernel
csiph-web