We can remove the port type hint udev tags for the wwan subsystem, as
that logic is now incorporated in the port type hint processing logic
in the daemon itself.
This makes it clearer to know what exact hints are being used, as the
logic is in a single place and it has proper logging of all possible
cases.
The udev rules expect matches based on shell globbing pattern, not
based on regex. This means that "[0-9]*" doesn't mean "zero or more
digits", it actually means "one digit and zero or more other
characters".
So, avoid the regex like match, even if it worked for us, because it
didn't mean what we thought it meant, plus, it breaks our custom udev
rules processing logic, in which we convert the input patterns into
real regex expressions.
See https://bbs.archlinux.org/viewtopic.php?pid=1415916#p1415916
While at it, also make sure that the wildcard matching is done
exclusively for prefix and suffix matches, never in the middle of
a string, or our custom rules parser would not match correctly.
For now WWAN subsystem support has only been validated for PCI/MHI
based devices and extra patches for USB based devices (qmi_wwan,
cdc-mbim...) may be required, so do not consider WWAN ports being
on USB bus.
Signed-off-by: Loic Poulain <loic.poulain@linaro.org>
The new Linux WWAN subsystem exposes the various WWAN control ports
as character devices. WWAN port device naming follows the following
convention: wwanXpYP, where X is the wwan device index, Y the port
index and P the port type/protocol.
Signed-off-by: Loic Poulain <loic.poulain@linaro.org>
Back in Linux < 3.6 days, the cdc-wdm ports exposed by the QMI driver
were flagged as owned by the 'usb' subsystem. That changed in 3.6 when
the subsystem was renamed to 'usbmisc':
https://mail.gnome.org/archives/networkmanager-list/2012-June/msg00125.html
This patch removes all monitoring of the 'usb' subsystem completely,
which is anyway a valid subsystem but for which we shouldn't need any
special handling. Right now, with newer kernels, we were using that
monitoring exclusively to get notified of full USB device remove
events, which is really not required as we already process the port
removals one by one.
We simplify the logic everywhere that attempted to match either the
'usb' or 'usbmisc' subsystems, and we no longer require the explicit
checks for the port name being named 'cdc-wdm[0-9]*' in the code, as
that is already taken care of by the ID_MM_CANDIDATE udev tag rule.
When a new USB device is hotplugged, e.g. a USB<->RS232 converter that
exposes a single ttyUSB0, these udev events happen:
add /devices/pci0000:00/0000:00:14.0/usb2/2-1 (usb/usb-device)
add /devices/pci0000:00/0000:00:14.0/usb2/2-1/2-1:1.0 (usb/usb-interface)
add /devices/pci0000:00/0000:00:14.0/usb2/2-1/2-1:1.0/ttyUSB0 (usb-serial)
add /devices/pci0000:00/0000:00:14.0/usb2/2-1/2-1:1.0/ttyUSB0/tty/ttyUSB0 (tty)
bind /devices/pci0000:00/0000:00:14.0/usb2/2-1/2-1:1.0/ttyUSB0 (usb-serial)
bind /devices/pci0000:00/0000:00:14.0/usb2/2-1/2-1:1.0 (usb/usb-interface)
bind /devices/pci0000:00/0000:00:14.0/usb2/2-1 (usb/usb-device)
Our udev rules in MM only added tags in the 'add' events, and it looks
like the only ones 'persistent' after this sequence are those of the
last event happening on the specific path.
This meant that all TTY subsystem rules (e.g. ID_MM_CANDIDATE) would
be stored for later check (e.g. if ModemManager is started after these
rules have been applied), which was ok. "udevadm info -p ..." would
show these tags correctly always.
But this also meant that the 'bind' udev event happening for the USB
device didn't get any of our device-specific tags, and so we would be
missing them (e.g. ID_MM_DEVICE_MANUAL_SCAN_ONLY) if MM is started
after the last event has happened. "udevadm info -p ..." would
not show these tags.
Modify all our rules to also run at the 'bind' events.
See, for context:
https://github.com/systemd/systemd/issues/8221
Bluetooth serial devices are weird. They begin life being bound with
RFCOMMCREATEDEV ioctl() and then move around the sysfs tree when they are
connected and disconnected. The connection is estabilished upon the first
open() and torn down upon last close(), their first user virtually being
"owner" of the connection. We don't want to be that process, we're only
interested in actually connected modems. However, currently we have no
knowledge of that and therefore we connect and disconnect multiple times while
probing.
This patch marks unconnected RFCOMM devices as uninteresting to us.
The actual connection and disconnection will be handled by NetworkManager.
We'll try to cope with getting devices being reported in either 'usb' or
'usbmisc', trying to avoid the need of checking kernel version during runtime.
The kernel appears to add the device first without any parents,
then to move it to the correct place in the hierarchy, with its
immediate parent being the device's HCI controller. So we need
to capture the 'move' event too since that's when the rfcomm
device is finally usable.
Ignore devices that aren't completely configured by udev yet. If
ModemManager is started in parallel with udev, explicitly requesting
devices may return devices for which not all udev rules have yet been
applied (a bug in udev/gudev). Since we often need those rules to match
the device to a specific ModemManager driver, we need to ensure that all
rules have been processed before handling a device.
Do this by adding an item to the environment of each device that MM
might possibly be interested in, and ignoring devices that don't
have that. When the device is fully processed by udev, MM will get
an 'add' event and the device will have all rules applied.
Aleksander Morgado
Aleksander Morgado
Stephan Gerhold
Loic Poulain
Lubomir Rintel
Aleksander Morgado
Dan Williams