- #Logitech mx master 2s linux how to
- #Logitech mx master 2s linux software
- #Logitech mx master 2s linux windows
Parsing itself is pretty easy, I'll just have to figure out how to write it nicely. So the packet combo that does that just switches the wheel to HID++ reporting and is the same as for example when binding it to app switching. Research over, dev time!ĮDIT6: guess what, the spec was right - after sending the packet that should invert the scroll the events are no longer reported on standard HID. Definitely doable in the deamon / kernel driver.
#Logitech mx master 2s linux software
My guess is that the Options software just repeats the scroll events depending on what sensitivity is set. The scroll events are always sent through HID++, but when the wheel is in "horizontal wheel mode" they are also sent through standard HID - this seems different to what I found in the spec, but the document is rather old so it may be inaccurate. Well nevermind, when I have more time to dig into the kernel properly I may do it there, for now this will be userspace pĮDIT5: More and more research (my USB debugging skills got a bit rusty). Maybe my linux kernel knowledge is too limited, but seems like the file mentioned only contains a library to support HID++ that is not actually used anywhere. I'd be happy to provide a PR with such functionality, unless the project maintainers don't want this to get too bloated?ĮDIT3: more research, looks like the linux kernel itself handles HID++ comms: and here it even handles scaled scroll events appropriately: - it's just a question of how to set the wheel scale now.ĮDIT4: Huh, this method seems to not be called anywhere, weird. This, along with advanced button remapping would require a daemon running in the background that would translate HID++ to different events.
#Logitech mx master 2s linux windows
I just got the mouse too, sniffing various USB packets at the moment to figure out what it sends and when - side scroll sensitivity is on my list as well.ĮDIT: figured out inverting, sadly all the packets look the same for sensitivity - unless I find something more it seems like they do it in software somehow.ĮDIT2: Took a look though HID++ spec ( specifically), looks like the thumb-scroll events are sent through that on Windows so the software can catch them and translate to different sensitivities, but requires a flag to be set - otherwise it sends them through standard HID.