> The autodetect code in parport_pc is often wrong and can even be
> harmful to parallel ports. It relies partly on undocumented behaviour
> and is poking aggressively at registers (e.g. ZIP Windows drivers
> refuse all ports but standard ones to avoid these sorts of problems, it seems).
I assumed the autodetect code at least started with _some_ knowledge
about the port-type/chip before it goes loose at the registers... if not,
then indeed it may be harmful!?
> This is at 1st insmod (or after reboot):
> Feb 9 22:24:42 m3 kernel: parport1: PC-style at 0x6300 (0x6400) [PCSPP,TRISTATE,COMPAT,ECP]
> after rmmod I get this:
> Feb 9 22:53:28 m3 kernel: PCI parallel port detected: 1409:7168, I/O at 0x6300(0x6400)
> Feb 9 22:53:28 m3 kernel: parport: No ECR <-- custom debug code
> Feb 9 22:53:28 m3 kernel: parport1: PC-style at 0x6300 [PCSPP,TRISTATE,EPP]
I get the same behaviour! After I do rmmod and re-insmod parport_pc it
forgets about COMPAT and ECP and introduces EPP. But still, I cannot
get EPP to work (when using ppdev).
It seems I'm forced to dig into parport_pc anywhay, at least to try
understanding the implementation of autodetection of my card.
Does anyone have datasheets for the SUN1888 chip, or any other chip
compatible with the Timedia 4008 chip? (SUNIX refused to give me their
-- To unsubscribe, send mail to: email@example.com --
-- with the single word "unsubscribe" in the body of the message. --
This archive was generated by hypermail 2b29 : Sat Feb 10 2001 - 16:51:40 EST