search

VoodooSMBus / VoodooRMI

Synaptic Trackpad driver over SMBus/I2C for macOS
GNU General Public License v2.0
233 stars 19 forks source link

Delayed Packets #84

Open 1Revenger1 opened 3 years ago

1Revenger1 commented 3 years ago

While using the trackpad, some people have come across an issue where the trackpad will randomly stop working for a second or two, before macOS acts upon whatever actions occurred during that time. This only occurs on some devices.

The only time this has happened to me is when using ACPIBatteryManager.kext, and I rarely get it after sleeping under Catalina. This may be related to some scheduling issues and VoodooInput, or other weird interaction between VoodooInput and VoodooRMI? I've seen people with these issues have similar issues with Acidanthera's VoodooPS2 while are fine with Rehabman's VoodooPS2. I have no clue why this would happen though - and don't have a good way to figure out why this happens.

Related Issues:

gimdh commented 3 years ago

I think you mean this.

Basically entire ACPI system halts whenever ACPIBatterManager polls battery information.

1Revenger1 commented 3 years ago

I've seen it on laptops not using ACPIBatteryManager as well - I'm not entirely sure what it could be. The trackpad on my device feels like my MT2 in terms of smoothness/mapping/etc.

gimdh commented 3 years ago

I think SMCBatteryManager might also not be able to avoid spin wait. VoodooI2C suffers similar issue AFAIK. Have you ever seen a laptop with this issue even without any battery manager kext?

//Edit: BTW, mine also works great. Another laptop had this issue with ACPIBatteryManager and VoodooI2C, but either of removing sleep() and using SMCBatteryManager instead solved problem.

FreeJHack commented 3 years ago

One or two seconds is a long time, however, I had a similar issue and I found the fix. It was related to ACPIBatteryManager kext and some newer ACPI. Take a look at this, maybe can help...