On 01/07/2020 07:50, Markus Armbruster wrote:
> Laurent Vivier <lviv...@redhat.com> writes:
>
>> qemu_set_nonblock() checks that the file descriptor can be used and, if
>> not, crashes QEMU. An assert() is used for that. The use of assert() is
>> used to detect programming error and the coredump will allow to debug
>> the problem.
>>
>> But in the case of the tap device, this assert() can be triggered by
>> a misconfiguration by the user. At startup, it's not a real problem, but it
>> can also happen during the hot-plug of a new device, and here it's a
>> problem because we can crash a perfectly healthy system.
>>
>> For instance:
>> # ip link add link virbr0 name macvtap0 type macvtap mode bridge
>> # ip link set macvtap0 up
>> # TAP=/dev/tap$(ip -o link show macvtap0 | cut -d: -f1)
>> # qemu-system-x86_64 -machine q35 -device
>> pcie-root-port,id=pcie-root-port-0 -monitor stdio 9<> $TAP
>> (qemu) netdev_add type=tap,id=hostnet0,vhost=on,fd=9
>> (qemu) device_add
>> driver=virtio-net-pci,netdev=hostnet0,id=net0,bus=pcie-root-port-0
>> (qemu) device_del net0
>> (qemu) netdev_del hostnet0
>> (qemu) netdev_add type=tap,id=hostnet1,vhost=on,fd=9
>> qemu-system-x86_64: .../util/oslib-posix.c:247: qemu_set_nonblock:
>> Assertion `f != -1' failed.
>> Aborted (core dumped)
>>
>> To avoid that, check the file descriptor is valid before passing it to
>> qemu_set_non_block() for "fd=" and "fds=" parameters.
>>
>> Signed-off-by: Laurent Vivier <lviv...@redhat.com>
>> ---
>> include/qemu/sockets.h | 1 +
>> net/tap.c | 13 +++++++++++++
>> util/oslib-posix.c | 5 +++++
>> util/oslib-win32.c | 6 ++++++
>> 4 files changed, 25 insertions(+)
>>
>> diff --git a/include/qemu/sockets.h b/include/qemu/sockets.h
>> index 57cd049d6edd..5b0c2d77ddad 100644
>> --- a/include/qemu/sockets.h
>> +++ b/include/qemu/sockets.h
>> @@ -17,6 +17,7 @@ int qemu_socket(int domain, int type, int protocol);
>> int qemu_accept(int s, struct sockaddr *addr, socklen_t *addrlen);
>> int socket_set_cork(int fd, int v);
>> int socket_set_nodelay(int fd);
>> +bool qemu_fd_is_valid(int fd);
>> void qemu_set_block(int fd);
>> void qemu_set_nonblock(int fd);
>> int socket_set_fast_reuse(int fd);
>> diff --git a/net/tap.c b/net/tap.c
>> index 6207f61f84ab..f65966aaccd8 100644
>> --- a/net/tap.c
>> +++ b/net/tap.c
>> @@ -795,6 +795,12 @@ int net_init_tap(const Netdev *netdev, const char *name,
>> return -1;
>> }
>>
>> + /* Check if fd is valid */
>> + if (!qemu_fd_is_valid(fd)) {
>> + error_setg(errp, "Invalid file descriptor %d", fd);
>> + return -1;
>> + }
>> +
>> qemu_set_nonblock(fd);
>>
>> vnet_hdr = tap_probe_vnet_hdr(fd);
>> @@ -843,6 +849,13 @@ int net_init_tap(const Netdev *netdev, const char *name,
>> goto free_fail;
>> }
>>
>> + /* Check if fd is valid */
>> + if (!qemu_fd_is_valid(fd)) {
>> + error_setg(errp, "Invalid file descriptor %d", fd);
>> + ret = -1;
>> + goto free_fail;
>> + }
>> +
>> qemu_set_nonblock(fd);
>>
>> if (i == 0) {
>> diff --git a/util/oslib-posix.c b/util/oslib-posix.c
>> index 916f1be2243a..8d5705f598d3 100644
>> --- a/util/oslib-posix.c
>> +++ b/util/oslib-posix.c
>> @@ -244,6 +244,11 @@ void qemu_anon_ram_free(void *ptr, size_t size)
>> qemu_ram_munmap(-1, ptr, size);
>> }
>>
>> +bool qemu_fd_is_valid(int fd)
>> +{
>> + return fcntl(fd, F_GETFL) != -1;
>> +}
>> +
>> void qemu_set_block(int fd)
>> {
>> int f;
>> diff --git a/util/oslib-win32.c b/util/oslib-win32.c
>> index e9b14ab17847..a6be9445cfdb 100644
>> --- a/util/oslib-win32.c
>> +++ b/util/oslib-win32.c
>> @@ -132,6 +132,12 @@ struct tm *localtime_r(const time_t *timep, struct tm
>> *result)
>> }
>> #endif /* CONFIG_LOCALTIME_R */
>>
>> +bool qemu_fd_is_valid(int fd)
>> +{
>> + /* FIXME: how to check if fd is valid? */
>
> Please explain the FIXME's impact in the commit message.
>
>> + return true;
>> +}
>> +
>> void qemu_set_block(int fd)
>> {
>> unsigned long opt = 0;
>
> This patch is okay as it is. But I'd solve the problem differently,
> avoiding the FIXME: have qemu_try_set_block() return -errno on failure,
> have qemu_set_block() wrap around it and assert it succeeds. Then in
> net_init_tap():
>
> ret = qemu_try_set_block(fd);
> if (ret < 0) {
> error_setg_errno(errp, -ret, "Can't use file descriptor %d",
> fd);
> }
>
> When @fd is bad (say -1), @ret is set to EBADF, and the error message
> looks like
>
> Can't use file descriptor -1: Bad file descriptor
>
> You should of course test the error message to see whether it makes
> sense even with a complex command line. If it doesn't, having
> it mention @name could perhaps help.
>
Yes, it's a good idea. I'm going to do that.
Thanks,
Laurent
