Thread (2 messages) 2 messages, 2 authors, 2018-08-03

Re: [PATCH 2/2] HID: redragon Add additional verification to rdesc modification quirk

From: Robert Munteanu <hidden>
Date: 2018-08-03 13:22:03
Also in: lkml

On Sat, 2018-07-28 at 15:53 -0400, John S Gruber wrote:
There are many devices using the vendor_id 0c45 and device_id of 760b
combination. Also the two bytes 0x81 0x00 aren't rare for a report
description. For these reasons the report description being altered
by the quirk should be verified more completely

If I'm understanding this correctly, I believe for an array field the
report_size should be greater or equal to
ceil(log2(usage_maximum - usage_minimum + 1)). That's 3 bits for
these 8
shift keys, 0xe0-0xe7. Therefore the incorrect report description
can't
be valid for any device.

Check the actual count of the rdesc and compare the entire field
description to reduce the chance of patching the wrong thing by
inadvertence.

Signed-off-by: John S Gruber <redacted>
I tested this on 4.17.11 with 85455dd906d5 and cbe7e3ad0eab from Jiri's
for-4.19/upstream tree and it works just fine.

Feel free to add

  Acked-By: Robert Munteanu [off-list ref]

Thanks for looking into this.

Robert
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help