Huawei ME909s-120

[Bjørn Mork]

Markus Gothe <nietzsche at lysator.liu.se> writes:

> Have you guys tried this Huawei embedded hi-silcon module?
>
> Seems like the MBIM stack is broken on it; can anyone verify this please?

Do you have more details on how it is broken?  Could it be this issue,
observed on an E3372:
https://git.kernel.org/cgit/linux/kernel/git/davem/net.git/commit/drivers/net/usb/cdc_mbim.c?id=f8c0cfa5eca902d388c0b57c7ca29a1ff2e6d8c6

For now, you'll just have to patch the driver with a device specific
entry to test.  But eventually, this patch should make it easier to
test and verify the issue:
https://git.kernel.org/cgit/linux/kernel/git/davem/net-next.git/commit/drivers/net/usb/cdc_ncm.c?id=404814af69d4732276319b90886b81fb2884ae1b

I believe we should reimplement the NDP creation code in a way that will
make this behaviour the default.  Collecting skbs in a list (or
several for multiplexed sessions?), and create the NDPs on the fly when
preparing to transmit, seems logical.

If the problem extends to the m.2/miniPCIe modules, then I guess there
are a number of affected laptop vendor device IDs out there.  Tracking
them all is going to be hard. We don't want to do that.  Better just
make the driver behave like the Windows driver.


Bjørn