We already have a QOM interface for this so let's use it.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Right now, DeviceInfo acts as the class for qdev.  In order to switch to a
proper ObjectClass derivative, we need to ween all of the callers off of
interacting directly with the info pointer.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Start maintaining endpoint state at USBDevice level.  Add USBEndpoint
struct and some helper functions to deal with it.  For now it contains
the endpoint type only.  Moved over some bits from usb-linux.c
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Commit 6daf194d got rid of them, but Hans and Gerd added some more
lately.  Tracked down with this Coccinelle semantic patch:

@r@
    expression fmt;
    position p;
@@
    error_report(fmt, ...)@p
@script:python@
    fmt << r.fmt;
    p << r.p;
@@
if "\\n" in str(fmt):
    print "%s:%s:%s:%s" % (p[0].file, p[0].line, p[0].column, fmt)
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>

Commit f462141f introduced clean up code
when usb_qdev_init() fails.  Unfortunately it calls .handle_destroy()
when .init() was never invoked or failed.  This can lead to crashes when
.handle_destroy() tries to clean up things that were never initialized.

This patch is careful to undo only those steps that completed along the
usb_qdev_init() code path.  It's not as pretty as the unified error
handling in f462141f but it's necessary.
Signed-off-by: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

qdev doesn't call the ->exit callback on ->init failures, so we have to
take care ourself that we cleanup property on errors.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Use qdev_init() instead of qdev_init_nofail(), usb device initialization
can fail, most common case being port and device speed mismatch.  Handle
failures correctly and pass up NULL pointers then.

Also fixup usb_create_simple() callers (only one was buggy) to properly
check for NULL pointers before referncing the usb_create_simple() return
value.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

This patch makes qemu assign a port when creating the device, not when
attaching it.  For most usb devices this isn't a noticable difference
because they are in attached state all the time.

The change affects usb-host devices which live in detached state while
the real device is unplugged from the host.  They have a fixed port
assigned all the time now instead of getting grabbing one on attach and
releasing it at detach, i.e. they stop floating around at the usb bus.

The change also allows to simplify usb-hub.  It doesn't need the
handle_attach() callback any more to configure the downstream ports.
This can be done at device initialitation time now.  The changed
initialization order (first grab upstream port, then register downstream
ports) also fixes some icky corner cases.  For example it is not possible
any more to plug the hub into one of its own downstream ports.

The usb host adapters must care too.  USBPort->dev being non-NULL
doesn't imply any more the device is in attached state.  The host
adapters must additionally check the USBPort->dev->attached flag.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

qemu_malloc/qemu_free no longer exist after this commit.
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NJes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Note this fixes 2 things in one go, first of all the device_destroy bus
op should be a device_detach bus op, as pending async packets from the
device should be cancelled on detach not on destroy.

Secondly having this as a bus op won't work with companion controllers, since
then there will be 1 bus driven by the ehci controller and thus 1 set of bus
ops, but the device being detached may be downstream of a handed over port.
Making the detach of a downstream device a port op allows the ehci controller
to forward this to the companion controller port for handed over ports.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

This is a preparation patch for adding support for USB companion controllers.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Cleanup / preparation patch for companion controller support. Note that
as a "side-effect" this patch also fixes the milkymist-softusb controller
not having a port_location set for its ports.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

This causes an "Error: tried to detach unattached usb device " to be printed,
this can happen when deleting ie a usb host qdev, which did not
get attached (because a device matching the filter never got plugged in).
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

This patch adds USBBusOps struct with (for now) only a single callback
which is called when a device is about to be destroyed.  The USB Host
adapters are implementing this callback and use it to cancel any async
requests which might be in flight before the device actually goes away.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Avoid this warning from OpenBSD linker:
  LINK  i386-softmmu/qemu
../usb-bus.o(.text+0x27c): In function `usb_get_fw_dev_path':
/src/qemu/hw/usb-bus.c:294: warning: sprintf() is often misused,
please use snprintf()
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Yes, seriously.  There is no migration support at all for usb devices.
They loose state, especially the device address, and stop responding
because of that.  Oops.

Luckily there is so much broken usb hardware out there that the guest
usually just kicks the device hard (via port reset and
reinitialization), then continues without a hitch.  So we got away with
that in a surprising high number of cases.

The arrival of remote wakeup (which enables autosuspend support) changes
that picture though.  The usb devices also forget that it they are
supposed to wakeup, so they don't do that.  The host also doesn't notice
the device stopped working in case it suspended the device and thus
expects it waking up instead of polling it.  Result is that your mouse
is dead.

Lets start fixing that.  Add a vmstate struct for USBDevice.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

It isn't needed any more.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

This patch rewrites the firmware path code to use the physical port
location tracking just added to the qemu usb core.  It also fixes the
port numbering to start with "1" in the firmware path.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

This allows to explictily set the physical port where you want to
plug the usb device.  Example:

  -device usb-tablet,bus=usb.0,port=2

With explicit port addressing qemu can and will not automagically add
USB Hubs.  This means that:

  (a) You can plug two devices of your choice into the two uhci
      root ports.
  (b) If you want plug in more that two devices you have to care
      about adding a hub yourself.

Plugging a hub works this way:

  -device usb-hub,bus=usb.0,port=1

Use this to add a device to the hub:

  -device usb-tablet,bus=usb.0,port=1.1
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Add a path string to USBPort.  Add usb_port_location() function to set
the physical location of the usb port.  Update all drivers implementing
usb ports to call it.  Update the monitor commands to print it.  Wind it
up in qdev.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Add a field to usb ports indicating the speed(s) they are
able to handle.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Create USBPortOps struct, move the attach function to that struct.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

This patch allows to set usb descriptor strings per device instance.
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Signed-off-by: NGleb Natapov <gleb@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Ports on root hub will have NULL here. This is needed to reconstruct
path from device to its root hub to build device path.
Signed-off-by: NGleb Natapov <gleb@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Value stored is never read.
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

After commit 702f3e0f, the params is
nerver NULL. It should check *params instead of params to determine
whether the params is empty.
Signed-off-by: NTeLeMan <geleman@gmail.com>
Signed-off-by: NAurelien Jarno <aurelien@aurel32.net>

Many usbdevice_init implementors assume params is non-NULL.
Signed-off-by: NJan Kiszka <jan.kiszka@web.de>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

error_report() terminates the message with a newline.  Strip it it
from its arguments.

This fixes a few error messages lacking a newline:
net_handle_fd_param()'s "No file descriptor named %s found", and
tap_open()'s "vnet_hdr=1 requested, but no kernel support for
IFF_VNET_HDR available" (all three versions).

There's one place that passes arguments without newlines
intentionally: load_vmstate().  Fix it up.

If -usbdevice is used on a machine with no USB busses, usb_create
will fail and return NULL.  Patch below handles this failure gracefully
rather than crashing when we try to init the device.
Signed-off-by: NPaul Brook <paul@codesourcery.com>

Device names with whitespace require quoting in the shell and in the
monitor.  Some of the offenders are also overly long.  Some have a
more convenient alias, some don't.

The place for verbose device names is DeviceInfo member desc.  The
name should be short & sweet.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Using the qdev name for the product description makes for inconvenient
qdev names.

Put the product description in new USBDeviceInfo member product_desc.
Make usb_qdev_init() use it.  No user or guest visible change, since
the value is still the same.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

It's not a device name, it's the USB product description string.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>