Thanks and Lots of Hidraw devices

[jordanpwalsh]

First I wanted to thank @StollD for all the recent commits in January, some of the bug fixes have really increased the stability of the ipts driver, especially resuming out of suspend on a Surface Pro 7.

One issue, however, is the driver seems to create a new hidrawXX device after resuming from suspend and these are piling up after some time. Is there anyway to resolve that?

Thanks again,
Jordan

[StollD]

Could you please post your dmesg after suspending a few times, so that it creates some devices?

[jordanpwalsh]

I attached two dmesg captures - one with type cover and the other without because the first one had so many listed I wanted to make sure that wasn't part of the problem. I believe these lingering devices are causing some kind of other issue as well because the system won't cleanly shut down due to being unable to kill the iptsd processes.

dmesg_no_type_cover.txt
dmesg_with_type_cover.txt

[StollD]

When did you last update? The errors in the log should have been fixed by the earliest commits I made in January.

i.e. this:

ipts 0000:00:16.4-3e8d0870-271a-4208-8eb5-9acb9402ae04: Failed to request flush: -4

would trigger an infinite loop with this commit: 404700a

and should not happen at all with this commit: 862d095

[jordanpwalsh]

I have commit f4dd578 from 20-Jan checked out and built. I suppose it's possible I miffed something with the install, but a little hard to tell with the version being the same. I should have the latest of both ipts driver and iptsd from the other repo, and some of my previous issues did go away. I can bump the version to verify that's what I'm running unless you have other suggestions.

…

On Sun, Jan 22, 2023 at 4:53 PM Dorian Stoll ***@***.***> wrote: When did you last update? The errors in the log should have been fixed by the earliest commits I made in January. i.e. this: ipts 0000:00:16.4-3e8d0870-271a-4208-8eb5-9acb9402ae04: Failed to request flush: -4 would trigger an infinite loop with this commit: 404700a <404700a> and should not happen at all with this commit: 862d095 <862d095> — Reply to this email directly, view it on GitHub <#21 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABDZL6K5KPLADEQIYM623IDWTWT4XANCNFSM6AAAAAAUCT756E> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- Thanks, Jordan

[StollD]

Maybe try to uninstall any versions of the module you have installed via. DKMS etc.

Then reinstall the kernel and finally the module.

Because I believe that running sudo make dkms-install won't actually update the module if it is already installed.

[jordanpwalsh]

Because I believe that running sudo make dkms-install won't actually update the module if it is already installed.

It won't, but it errors out. I ended up removing it from dkms via dkms uninstall as well as manually removing files and rebooting (The module wouldn't reinsert until rebooting).

Let me do that again with a version bump to ensure I'm running the latest. Thanks.