Commit 767abe7 ("chardev: forbid 'wait' option with client sockets")
is a bit too strict. Current libvirt always set wait=false, and will
thus fail to add client chardev.
Make the code more permissive, allowing wait=false with client socket
chardevs. Deprecate usage of 'wait' with client sockets.
Fixes: 767abe7f49
Cc: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-id: 20190415163337.2795-1-marcandre.lureau@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Currently any client which can complete the TLS handshake is able to use
a chardev server. The server admin can turn on the 'verify-peer' option
for the x509 creds to require the client to provide a x509
certificate. This means the client will have to acquire a certificate
from the CA before they are permitted to use the chardev server. This is
still a fairly low bar.
This adds a 'tls-authz=OBJECT-ID' option to the socket chardev backend
which takes the ID of a previously added 'QAuthZ' object instance. This
will be used to validate the client's x509 distinguished name. Clients
failing the check will not be permitted to use the chardev server.
For example to setup authorization that only allows connection from a
client whose x509 certificate distinguished name contains 'CN=fred', you
would use:
$QEMU -object tls-creds-x509,id=tls0,dir=/home/berrange/qemutls,\
endpoint=server,verify-peer=yes \
-object authz-simple,id=authz0,identity=CN=laptop.example.com,,\
O=Example Org,,L=London,,ST=London,,C=GB \
-chardev socket,host=127.0.0.1,port=9000,server,\
tls-creds=tls0,tls-authz=authz0 \
...other qemu args...
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
If the socket is connecting or connected, tcp_chr_update_read_handler will
be called but it should not set the NetListener's callbacks again.
Otherwise, tcp_chr_accept is invoked while the socket is in connected
state and you get an assertion failure.
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When the 'reconnect' option is given for a client connection, the
qmp_chardev_open_socket_client method will run an asynchronous
connection attempt. The QIOChannel socket executes this is a single use
background thread, so the connection will succeed immediately (assuming
the server is listening). The chardev, however, won't get the result
from this background thread until the main loop starts running and
processes idle callbacks.
Thus when tcp_chr_wait_connected is run s->ioc will be NULL, but the
state will be TCP_CHARDEV_STATE_CONNECTING, and there may already
be an established connection that will be associated with the chardev
by the pending idle callback. tcp_chr_wait_connected doesn't check the
state, only s->ioc, so attempts to establish another connection
synchronously.
If the server allows multiple connections this is unhelpful but not a
fatal problem as the duplicate connection will get ignored by the
tcp_chr_new_client method when it sees the state is already connected.
If the server only supports a single connection, however, the
tcp_chr_wait_connected method will hang forever because the server will
not accept its synchronous connection attempt until the first connection
is closed.
To deal with this tcp_chr_wait_connected needs to synchronize with the
completion of the background connection task. To do this it needs to
create the QIOTask directly and use the qio_task_wait_thread method.
This will cancel the pending idle callback and directly dispatch the
task completion callback, allowing the connection to be associated
with the chardev. If the background connection failed, it can still
attempt a new synchronous connection.
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-15-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
In the previous commit
commit 1dc8a6695c
Author: Marc-André Lureau <marcandre.lureau@redhat.com>
Date: Tue Aug 16 12:33:32 2016 +0400
char: fix waiting for TLS and telnet connection
the tcp_chr_wait_connected() method was changed to check for a non-NULL
's->ioc' as a sign that there is already a connection present, as
opposed to checking the "connected" flag to supposedly fix handling of
TLS/telnet connections.
The original code would repeatedly call tcp_chr_wait_connected creating
many connections as 'connected' would never become true. The changed
code would still repeatedly call tcp_chr_wait_connected busy waiting
because s->ioc is set but the chardev will never see CHR_EVENT_OPENED.
IOW, the code is still broken with TLS/telnet, but in a different way.
Checking for a non-NULL 's->ioc' does not mean that a CHR_EVENT_OPENED
will be ready for a TLS/telnet connection. These protocols (and the
websocket protocol) all require the main loop to be running in order
to complete the protocol handshake before emitting CHR_EVENT_OPENED.
The tcp_chr_wait_connected() method is only used during early startup
before a main loop is running, so TLS/telnet/websock connections can
never complete initialization.
Making this work would require changing tcp_chr_wait_connected to run
a main loop. This is quite complex since we must not allow GSource's
that other parts of QEMU have registered to run yet. The current callers
of tcp_chr_wait_connected do not require use of the TLS/telnet/websocket
protocols, so the simplest option is to just forbid this combination
completely for now.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-14-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
If establishing a client connection fails, the tcp_chr_wait_connected
method should sleep for the reconnect timeout and then retry the
attempt. This ensures the callers don't immediately abort with an
error when the initial connection fails.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-13-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
The socket connection state is indicated via the 'bool connected' field
in the SocketChardev struct. This variable is somewhat misleading
though, as it is only set to true once the connection has completed all
required handshakes (eg for TLS, telnet or websockets). IOW there is a
period of time in which the socket is connected, but the "connected"
flag is still false.
The socket chardev really has three states that it can be in,
disconnected, connecting and connected and those should be tracked
explicitly.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-12-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
In qmp_chardev_open_socket the code for connecting client chardevs is
split across two conditionals far apart with some server chardev code in
the middle. Split up the method so that code for client connection setup
is separate from code for server connection setup.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-11-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
The tcp_chr_wait_connected method can deal with either server or client
chardevs, but some callers only care about one of these possibilities.
The tcp_chr_wait_connected method will also need some refactoring to
reliably deal with its primary goal of allowing a device frontend to
wait for an established connection, which will interfere with other
callers.
Split it into two methods, one responsible for server initiated
connections, the other responsible for client initiated connections.
In doing this split the tcp_char_connect_async() method is renamed
to become consistent with naming of the new methods.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-10-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
The 'sioc' variable in qmp_chardev_open_socket was unused since
commit 3e7d4d20d3
Author: Peter Xu <peterx@redhat.com>
Date: Tue Mar 6 13:33:17 2018 +0800
chardev: use chardev's gcontext for async connect
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-9-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Now that all validation is separated off into a separate method,
we can directly populate the ChardevSocket struct from the
QemuOpts values, avoiding many local variables.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-7-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
The 'wait'/'nowait' parameter is used to tell server sockets whether to
block until a client is accepted during initialization. Client chardevs
have always silently ignored this option. Various tests were mistakenly
passing this option for their client chardevs.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-6-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
The 'reconnect' option is used to give the sleep time, in seconds,
before a client socket attempts to re-establish a connection to the
server. It does not make sense to set this for server sockets, as they
will always accept a new client connection immediately after the
previous one went away.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-5-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
The TLS creds option is not valid with certain address types. The user
config was only checked for errors when parsing legacy QemuOpts, thus
the user could pass unsupported values via QMP.
Pull all code for validating options out into a new method
qmp_chardev_validate_socket, that is called from the main
qmp_chardev_open_socket method. This adds a missing check for rejecting
TLS creds with the vsock address type.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190211182442.8542-4-berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
New option "websocket" added to allow using WebSocket protocol for
chardev socket backend.
Example:
-chardev socket,websocket,server,id=...
Signed-off-by: Julia Suvorova <jusual@mail.ru>
Message-Id: <20181018223501.21683-3-jusual@mail.ru>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Upcoming websocket support requires additional parameters in function
headers that are already overloaded. This patch replaces the bunch of
parameters with a single structure pointer.
Signed-off-by: Julia Suvorova <jusual@mail.ru>
Message-Id: <20181018223501.21683-2-jusual@mail.ru>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
A chardev socket created with the 'fd=' argument is not going to
handle reconnection properly by recycling the same fd (or not in a
supported way). Let's forbid this case.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
qemu_chr_parse_socket() fills all ChardevSocket fields, but that
doesn't reflect correctly the arguments given with the options / on
the command line. "reconnect" takes a number as argument, and the
default value is 0, which doesn't help to identify the missing
option. The other arguments have default values that are less
problematic, leave them set by default for now.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
A socket chardev may not have associated address (when adding client
fd manually for example). But on disconnect, updating socket filename
expects an address and may lead to this crash:
Thread 1 "qemu-system-x86" received signal SIGSEGV, Segmentation fault.
0x0000555555d8c70c in SocketAddress_to_str (prefix=0x555556043062 "disconnected:", addr=0x0, is_listen=false, is_telnet=false) at /home/elmarco/src/qq/chardev/char-socket.c:388
388 switch (addr->type) {
(gdb) bt
#0 0x0000555555d8c70c in SocketAddress_to_str (prefix=0x555556043062 "disconnected:", addr=0x0, is_listen=false, is_telnet=false) at /home/elmarco/src/qq/chardev/char-socket.c:388
#1 0x0000555555d8c8aa in update_disconnected_filename (s=0x555556b1ed00) at /home/elmarco/src/qq/chardev/char-socket.c:419
#2 0x0000555555d8c959 in tcp_chr_disconnect (chr=0x555556b1ed00) at /home/elmarco/src/qq/chardev/char-socket.c:438
#3 0x0000555555d8cba1 in tcp_chr_hup (channel=0x555556b75690, cond=G_IO_HUP, opaque=0x555556b1ed00) at /home/elmarco/src/qq/chardev/char-socket.c:482
#4 0x0000555555da596e in qio_channel_fd_source_dispatch (source=0x555556bb68b0, callback=0x555555d8cb58 <tcp_chr_hup>, user_data=0x555556b1ed00) at /home/elmarco/src/qq/io/channel-watch.c:84
Replace filename with a generic "disconnected:socket" in this case.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
So far, tcp_chr_update_read_handler() only updated the read
handler. Let's also update the hup handler.
Factorize the code while at it. (note that s->ioc != NULL when
s->connected)
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20180817135224.22971-4-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This reverts commit 25679e5d58.
This commit broke "reconnect socket" chardev that are created after
"machine_done": they no longer try to connect. It broke also
vhost-user-test that uses chardev while there is no "machine_done"
event.
The goal of this patch was to move the "connect" source to the
frontend context. chr->gcontext is set with
qemu_chr_fe_set_handlers(). But there is no guarantee that it will be
called, so we can't delay connection until then: the chardev should
still attempt to connect during open(). qemu_chr_fe_set_handlers() is
eventually called later and will update the context.
Unless there is a good reason to not use initially the default
context, I think we should revert to the previous state to fix the
regressions.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20180817135224.22971-3-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This reverts commit 99f2f54174.
See next commit reverting 25679e5d58 as
well for rationale.
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20180817135224.22971-2-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In the tcp_chr_write function, we checked errno,
but errno was not reset before a read or write operation.
Therefore, this check of errno's actions is often
incorrect after EAGAIN has occurred.
we need check errno together with ret < 0.
Signed-off-by: xinhua.Cao <caoxinhua@huawei.com>
Message-Id: <20180704033642.15996-1-caoxinhua@huawei.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Fixes: 9fc53a10f8
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If we see EAGAIN, no data was sent over the socket, so we still have to
retry sending of msgfds next time.
Signed-off-by: linzhecheng <linzhecheng@huawei.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
This trips Coverity, which believes the subsequent qio_channel_create_watch
can dereference a NULL pointer. In reality, tcp_chr_connect's callers
all have s->ioc properly initialized, since they are all rooted at
tcp_chr_new_client.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
TLS handshake may create background GSource tasks, while we won't know
the correct GMainContext until the whole chardev (including frontend)
inited. Let's postpone the initial TLS handshake until machine done.
For dynamically created tcp chardev, we don't postpone that by checking
the init_machine_done variable.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
[peterx: add missing include line, do unit test]
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20180308140714.28906-1-peterx@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
* SCSI fix to pass maximum transfer size (Daniel Barboza)
* chardev fixes and improved iothread support (Daniel Berrangé, Peter)
* checkpatch tweak (Eric)
* make help tweak (Marc-André)
* make more PCI NICs available with -net or -nic (myself)
* change default q35 NIC to e1000e (myself)
* SCSI support for NDOB bit (myself)
* membarrier system call support (myself)
* SuperIO refactoring (Philippe)
* miscellaneous cleanups and fixes (Thomas)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iQEcBAABAgAGBQJapqaMAAoJEL/70l94x66DQoUH/Rvg+a8giz/SrEA4P8D3Cb2z
4GNbNUUoy4oU0ltD5IAMskMwpOsvl1batE0D+pKIlfO9NV4+Cj2kpgo0p9TxoYqM
VCby3wRtx27zb5nVytC6M++iIKXmeEMqXmFw61I6umddNPSl4IR3hiHEE0DM+7dV
UPIOvJeEiazyQaw3Iw+ZctNn8dDBKc/+6oxP9xRcYTaZ6hB4G9RZkqGNNSLcJkk7
R0UotdjzIZhyWMOkjIwlpTF4sWv8gsYUV4bPYKMYho5B0Obda2dBM3I1kpA8yDa/
xZ5lheOaAVBZvM5aMIcaQPa65MO9hLyXFmhMOgyfpJhLBBz6Qpa4OLLI6DeTN+0=
=UAgA
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging
* Record-replay lockstep execution, log dumper and fixes (Alex, Pavel)
* SCSI fix to pass maximum transfer size (Daniel Barboza)
* chardev fixes and improved iothread support (Daniel Berrangé, Peter)
* checkpatch tweak (Eric)
* make help tweak (Marc-André)
* make more PCI NICs available with -net or -nic (myself)
* change default q35 NIC to e1000e (myself)
* SCSI support for NDOB bit (myself)
* membarrier system call support (myself)
* SuperIO refactoring (Philippe)
* miscellaneous cleanups and fixes (Thomas)
# gpg: Signature made Mon 12 Mar 2018 16:10:52 GMT
# gpg: using RSA key BFFBD25F78C7AE83
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>"
# gpg: aka "Paolo Bonzini <pbonzini@redhat.com>"
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4 E2F7 7E15 100C CD36 69B1
# Subkey fingerprint: F133 3857 4B66 2389 866C 7682 BFFB D25F 78C7 AE83
* remotes/bonzini/tags/for-upstream: (69 commits)
tcg: fix cpu_io_recompile
replay: update documentation
replay: save vmstate of the asynchronous events
replay: don't process async events when warping the clock
scripts/replay-dump.py: replay log dumper
replay: avoid recursive call of checkpoints
replay: check return values of fwrite
replay: push replay_mutex_lock up the call tree
replay: don't destroy mutex at exit
replay: make locking visible outside replay code
replay/replay-internal.c: track holding of replay_lock
replay/replay.c: bump REPLAY_VERSION again
replay: save prior value of the host clock
replay: added replay log format description
replay: fix save/load vm for non-empty queue
replay: fixed replay_enable_events
replay: fix processing async events
cpu-exec: fix exception_index handling
hw/i386/pc: Factor out the superio code
hw/alpha/dp264: Use the TYPE_SMC37C669_SUPERIO
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
# Conflicts:
# default-configs/i386-softmmu.mak
# default-configs/x86_64-softmmu.mak
When starting QEMU management apps will usually setup a monitor socket, and
then open it immediately after startup. If not using QEMU's own -daemonize
arg, this process can be troublesome to handle correctly. The mgmt app will
need to repeatedly call connect() until it succeeds, because it does not
know when QEMU has created the listener socket. If can't retry connect()
forever though, because an error might have caused QEMU to exit before it
even creates the monitor.
The obvious way to fix this kind of problem is to just pass in a pre-opened
socket file descriptor for the QEMU monitor to listen on. The management
app can now immediately call connect() just once. If connect() fails it
knows that QEMU has exited with an error.
The SocketAddress(Legacy) structs allow for FD passing via the monitor, and
now via inherited file descriptors from the process that spawned QEMU. The
final missing piece is adding a 'fd' parameter in the socket chardev
options.
This allows both HMP usage, pass any FD number with SCM_RIGHTS, then
running HMP commands:
getfd myfd
chardev-add socket,fd=myfd
Note that numeric FDs cannot be referenced directly in HMP, only named FDs.
And also CLI usage, by leak FD 3 from parent by clearing O_CLOEXEC, then
spawning QEMU with
-chardev socket,fd=3,id=mon
-mon chardev=mon,mode=control
Note that named FDs cannot be referenced in CLI args, only numeric FDs.
We do not wire this up in the legacy chardev syntax, so you cannot use FD
passing with '-qmp', you must use the modern '-mon' + '-chardev' pair.
When passing pre-opened FDs there is a restriction on use of TLS encryption.
It can be used on a server socket chardev, but cannot be used for a client
socket chardev. This is because when validating a server's certificate, the
client needs to have a hostname available to match against the certificate
identity.
An illustrative example of usage is:
#!/usr/bin/perl
use IO::Socket::UNIX;
use Fcntl;
unlink "/tmp/qmp";
my $srv = IO::Socket::UNIX->new(
Type => SOCK_STREAM(),
Local => "/tmp/qmp",
Listen => 1,
);
my $flags = fcntl $srv, F_GETFD, 0;
fcntl $srv, F_SETFD, $flags & ~FD_CLOEXEC;
my $fd = $srv->fileno();
exec "qemu-system-x86_64", \
"-chardev", "socket,fd=$fd,server,nowait,id=mon", \
"-mon", "chardev=mon,mode=control";
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
To prepare for handling more address types, refactor the parsing of
socket address information to make it more robust and extensible.
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Even if common tn3270 implementations do not support TLS, it is trivial to
have them proxied over a proxy like stunnel which adds TLS at the sockets
layer. We should thus not silently skip tn3270 protocol initialization
when TLS is enabled.
Reviewed-by: Eric Blake <eblake@redhat.com>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Now qio_channel_tls_handshake() is ready to receive the context. Let
socket chardev use it, then the TLS handshake of chardev will always be
with the chardev's context.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20180306053320.15401-9-peterx@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This patch allows the socket chardev async connection be setup with
non-default gcontext. We do it by postponing the setup to machine done,
since until then we can know which context we should run the async
operation on.
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20180306053320.15401-8-peterx@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Generalize the function to create the async QIO task connection. Also,
fix the context pointer to use the chardev's gcontext.
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20180306053320.15401-7-peterx@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It was originally created by qio_channel_add_watch() so it's always
assigning the task to main context. Now we use the new API called
qio_channel_add_watch_source() so that we get the GSource handle rather
than the tag ID.
Meanwhile, caching the gsource and TCPChardevTelnetInit (which holds the
handshake data) in SocketChardev.telnet_source so that we can also do
dynamic context switch when update read handlers.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20180306053320.15401-5-peterx@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
TCP chardevs can be using QIO network listeners working in the
background when in listening mode. However the network listeners are
always running in main context. This can race with chardevs that are
running in non-main contexts.
To solve this, we need to re-setup the net listeners in
tcp_chr_update_read_handler() with the newly cached gcontext.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20180306053320.15401-4-peterx@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When this commit was applied
commit 9894dc0cdc
Author: Daniel P. Berrange <berrange@redhat.com>
Date: Tue Jan 19 11:14:29 2016 +0000
char: convert from GIOChannel to QIOChannel
The tcp_chr_recv() function was changed to return QIO_CHANNEL_ERR_BLOCK
which corresonds to -2. As such the handling for EAGAIN was able to be
removed from tcp_chr_read(). Unfortunately in a later commit:
commit b6572b4f97
Author: Marc-André Lureau <marcandre.lureau@redhat.com>
Date: Fri Mar 11 18:55:24 2016 +0100
char: translate from QIOChannel error to errno
The tcp_chr_recv() function was changed back to return -1, with errno
set to EAGAIN, without also re-addding support for this to tcp_chr_read()
Reported-by: Aleksey Kuleshov <rndfax@yandex.ru>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20180222121351.26191-1-berrange@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Need to free TCPChardevTelnetInit when session established.
Since at it, switch to use G_SOURCE_* macros.
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20180301084438.13594-2-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A new parameter "context" is added to qio_channel_tls_handshake() is to
allow the TLS to be run on a non-default context. Still, no functional
change.
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
We have worked on qio_task_run_in_thread() already. Further, let
all the qio channel APIs use that context.
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
In my "build everything" tree, a change to the types in
qapi-schema.json triggers a recompile of about 4800 out of 5100
objects.
The previous commit split up qmp-commands.h, qmp-event.h, qmp-visit.h,
qapi-types.h. Each of these headers still includes all its shards.
Reduce compile time by including just the shards we actually need.
To illustrate the benefits: adding a type to qapi/migration.json now
recompiles some 2300 instead of 4800 objects. The next commit will
improve it further.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20180211093607.27351-24-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
[eblake: rebase to master]
Signed-off-by: Eric Blake <eblake@redhat.com>
qemu-common.h includes qemu/option.h, but most places that include the
former don't actually need the latter. Drop the include, and add it
to the places that actually need it.
While there, drop superfluous includes of both headers, and
separate #include from file comment with a blank line.
This cleanup makes the number of objects depending on qemu/option.h
drop from 4545 (out of 4743) to 284 in my "build everything" tree.
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20180201111846.21846-20-armbru@redhat.com>
[Semantic conflict with commit bdd6a90a9e in block/nvme.c resolved]
The following behavior was observed for QEMU configured by libvirt
to use guest agent as usual for the guests without virtio-serial
driver (Windows or the guest remaining in BIOS stage).
In QEMU on first connect to listen character device socket
the listen socket is removed from poll just after the accept().
virtio_serial_guest_ready() returns 0 and the descriptor
of the connected Unix socket is removed from poll and it will
not be present in poll() until the guest will initialize the driver
and change the state of the serial to "guest connected".
In libvirt connect() to guest agent is performed on restart and
is run under VM state lock. Connect() is blocking and can
wait forever.
In this case libvirt can not perform ANY operation on that VM.
The bug can be easily reproduced this way:
Terminal 1:
qemu-system-x86_64 -m 512 -device pci-serial,chardev=serial1 -chardev socket,id=serial1,path=/tmp/console.sock,server,nowait
(virtio-serial and isa-serial also fit)
Terminal 2:
minicom -D unix\#/tmp/console.sock
(type something and press enter)
C-a x (to exit)
Do 3 times:
minicom -D unix\#/tmp/console.sock
C-a x
It needs 4 connections, because the first one is accepted by QEMU, then two are queued by
the kernel, and the 4th blocks.
The problem is that QEMU doesn't add a read watcher after succesful read
until the guest device wants to acquire recieved data, so
I propose to install a separate pullhup watcher regardless of
whether the device waits for data or not.
Signed-off-by: Klim Kireev <klim.kireev@virtuozzo.com>
Message-Id: <20180125135129.9305-1-klim.kireev@virtuozzo.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It's a replacement of g_timeout_add[_seconds]() for chardevs. Chardevs
now can have dedicated gcontext, we should always bind chardev tasks
onto those gcontext rather than the default main context. Since there
are quite a few of g_timeout_add[_seconds]() callers, a new function
qemu_chr_timeout_add_ms() is introduced.
One thing to mention is that, terminal3270 is still always running on
main gcontext. However let's convert that as well since it's still part
of chardev codes and in case one day we'll miss that when we move it out
of main gcontext too.
Also, convert all the timers from GSource tags into GSource pointers.
Gsource tag IDs and g_source_remove()s can only work with default
gcontext, while now these GSources can logically be attached to other
contexts. So let's use explicit g_source_destroy() plus another
g_source_unref() to remove a timer.
Note: when in the timer handler, we don't need the g_source_destroy()
any more since that'll be done automatically if the timer handler
returns false (and that's what all the current handlers do).
Yet another note: in pty_chr_rearm_timer() we take special care for
ms=1000. This patch merged the two cases into one.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20180104141835.17987-4-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Instead of creating a QIOChannelSocket directly for the chardev
server socket, use a QIONetListener. This provides the ability
to listen on multiple sockets at the same time, so enables
full support for IPv4/IPv6 dual stack.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Message-Id: <20171218135417.28301-2-berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The tcp_chr_free_connection & tcp_chr_disconnect methods both
skip all of their cleanup work unless the 's->connected' flag
is set. This flag is set when the incoming client connection
is ready to use. Crucially this is *after* the TLS handshake
has been completed. So if the TLS handshake fails and we try
to cleanup the failed client, all the cleanup is skipped as
's->connected' is still false.
The only important thing that should be skipped in this case
is sending of the CHR_EVENT_CLOSED, because we never got as
far as sending the corresponding CHR_EVENT_OPENED. Every other
bit of cleanup can be robust against being called even when
s->connected is false.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Message-Id: <20171005155057.7664-1-berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We had a per-chardev cache for context, then we don't need this
parameter to be passed in every time when chr_update_read_handler()
called. As long as we are calling chr_update_read_handler() using
qemu_chr_be_update_read_handlers() we'll be fine.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1505975754-21555-5-git-send-email-peterx@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It was only passed in by chr_update_read_handlers(). However when
reconnect, we'll lose that context information. So if a chardev was
running on another context (rather than the default context, the NULL
pointer), it'll switch back to the default context if reconnection
happens. But, it should really stick to the old context.
Convert all the callers of io_add_watch_poll() to use the internally
cached gcontext. Then the context should be able to survive even after
reconnections.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1505975754-21555-4-git-send-email-peterx@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>