So I’m comparing known info of properly resetting USB controllers and listings of lspci -vv for FLReset+. Oddly, some USB controllers that support FLReset+ end up not resetting properly, but some other variants of ls-iommu.sh with reset functionality also printed show different results that don’t quite match what FLReset+ indicates.
FLReset+ is very important for VGA controllers as it’s been absent in Vega and Radeon VII and RDNA1. But it’s weird that the setting that’s supposed to give you an idea of how VGA controllers reset doesn’t also give clear indications of reset capabilities of USB controllers.
It’s weird because I see Linus using Asmedia ASM1142 controllers in ASUS USB 3.1 cards with supposedly proper reset, but the FLReset- flag is present instead of FLReset+, yet some controllers like Etrontech controllers with the + flag don’t reset properly…
Really wish there was a clear way in terms of a PCI function to tell if something can reset properly that ISN’T a GPU. Right now we have to go off of scattered reports with no real formal list of proper reset compatible controllers.
I know my Fresco Logic and VIA controllers had issues, but again, there’s no correlation between that and FLReset+, and what makes it even more confusing is that the ASM1142 doesn’t show up with FLReset+ and still works… My Renesas one is the same thing.
Is this post I made 2 years ago the only proper reference people have? If so, why are people flocking to the ASmedia controllers when they don’t have as extensive testing? Linus may have the ASM1142 working, but it doesn’t mean their entire lineup is guaranteed to work…