On Thu, Oct 18, 2012 at 10:40:53PM +0200, Bjørn Mork wrote: > The USB Communications Device Class "Mobile Broadband Interface Model" > (MBIM) is the USB-IFs alternative to the current chipset/vendor > specific solutions to "Mobile Broadband" device management. The > specification, including the management protocol description, can be > downloaded from http://www.usb.org/developers/devclass_docs#approved > > This driver implementing most MBIM features with the exception of > 32bit NTB and NDP headers. > > An important design principle has been reusing as much as possible of > existing kernel code, in particular the cdc_ncm and cdc_wdm drivers. > The CDC MBIM protocol is based on CDC NCM, and much of the setup and > framing logic is very similar. > > One important addition in MBIM compared to NCM is the new control > protocol embedded in CDC commands. This protocol is comprehensive and > support a number of policy decisions necessary for modern mobile > broadband devices often having multiple radio interfaces. Based on > early comments and the experiences with the qmi_wwan driver, knowledge > of the control protocol has been kept completely out of the driver. > This is userspace material. Like with qmi_wwan, a control protocol > interface is exported to userspace using the cdc_wdm subdriver API, > associating a /dev/cdc-wdmX character device with the network device > for the management application. > > Patch 1 and 2 are independent of the rest and only required to make > test devices with very large buffers work. > > Patch 3 adds new MBIM definitions to the cdc.h header file > > Patches 4 to 9 refactor the cdc_ncm driver to enable reusing common > parts for MBIM. > > Patches 10 and 11 add the new cdc_mbim driver > > Patch 12 prevents cdc_ncm from binding to backward compatible MBIM > devices > > Patches 13 and 14 extend the MBIM driver to support multiplexed > sessions > > The changes to the cdc_ncm driver has been tested and verified to work > with an Ericsson F5521gw device. The new cdc_mbim driver has been > tested with a Huawei E367u-2 device with MBIM firmware, and other > currently undisclosed devices. Very nice work, I don't have any objections to any of this, but I just did a very high-level code review: Acked-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
