"David A. Desrosiers" <desrod at gnu-designs.com> writes: > On Tue, 2007-01-30 at 21:53 +0000, Joe Wells wrote: >> >> > (I suspect maybe one of the kernel modules configures the adapter >> > only at module load-time, which is triggered by the insertion of >> > the adapter, and of course the adapter will lose any >> > configuration when it is removed and therefore unpowered.) > >> I no longer suspect this at all. > > Well, that particular Keyspan device also has firmware sent to it at > module load time, its not just a "dumb" device, there's an actual > firmware blob in the Keyspan module that is required to make it > function. Agreed. I now think the uploading of the firmware to my Keyspan USA-19 is most likely happening properly. > I have the clear'ish one with the green tail, Same here. > and it works great on this end (though, I never had to fiddle with > the serial rates). I'm not sure what the cause of my problems has been. I might have a bad Keyspan USA-19 adapter (although I have not seen any complaints about this model on the web other than my problems), a bad USB port (on a Sony Vaio VGN-TX770P), or it might be something to do with my Linux kernel (2.6.17 with some patches). Anyway, my problems seem to be fixed by unplugging the adapter and unloading the kernel device driver modules between every device opening. If I can only figure out how to do this completely by software (without actually physically unplugging the adapter), then I think I will be happy. > Its quite old though, at least a few years. Yes, I think I got this one 3 or 4 years ago. -- Joe
More information about the pilot-link-general mailing list