[PARPORT] parallel port detection woes

From: James Hsu (james@phasespace.com)
Date: Sat Oct 20 2001 - 10:29:00 EDT

  • Next message: shun: "[PARPORT] parport0"

    I've been trying different configurations for the better part of a day
    now, and I
    can't quite get parport_pc to recognize ECP mode. If I switch to EPP mode in
    BIOS, parport_pc will recognize it, but unfortunately vmware really wants
    ECP to work.

    So, with ECP enabled, I get this:

    Oct 20 04:29:14 dualbox kernel: Winbond Super-IO detection, now testing
    ports 3F0,370,250,4E,2E ...
    Oct 20 04:29:14 dualbox kernel: Winbond chip at EFER=0x2e key=0x87
    devid=52 devrev=17 oldid=ff type=83627
    Oct 20 04:29:14 dualbox kernel: Winbond LPT Config: cr_30=01 60,61=0378
    70=07 74=03, f0=3a
    Oct 20 04:29:14 dualbox kernel: Winbond LPT Config: active=yes,
    io=0x0378 irq=7, dma=3
    Oct 20 04:29:14 dualbox kernel: Winbond LPT Config: irqtype=pulsed low,
    high-Z, ECP fifo threshold=7
    Oct 20 04:29:14 dualbox kernel: Winbond LPT Config: Port mode=ECP
    Oct 20 04:29:14 dualbox kernel: SMSC Super-IO detection, now testing
    Ports 2F0, 370 ...
    Oct 20 04:29:14 dualbox kernel: parport0: PC-style at 0x378 [PCSPP]
    Oct 20 04:29:14 dualbox kernel: parport0: irq 7 detected
    Oct 20 04:29:14 dualbox kernel: parport0: cpp_daisy: aa5500ff(38)
    Oct 20 04:29:14 dualbox kernel: parport0: assign_addrs: aa5500ff(38)
    Oct 20 04:29:14 dualbox kernel: parport0: cpp_daisy: aa5500ff(38)
    Oct 20 04:29:14 dualbox kernel: parport0: assign_addrs: aa5500ff(38)

    This is a snippet of the kern.log. Its running on a 2.4.12, SMP machine. The
    motherboard is a tyan S2460, and I've already noted a couple BIOS
    pecularities,
    so I'm sorta inclined to believe that it might be part of the problem.

    I traced through the parport_pc.c file, and I've noticed that the
    parport_ECR_present(...) doesn't detect the "ECR". It tends to fail on the
    second test { if ((inb (ECONTROL (pb)) & 0x3 ) != 0x1) }, but i'm not
    even sure if it is supposed to go here.

    Does anyone have any insight on this issue?

    James Hsu

    -- To unsubscribe, send mail to: linux-parport-request@torque.net --
    -- with the single word "unsubscribe" in the body of the message. --



    This archive was generated by hypermail 2b29 : Sat Oct 20 2001 - 10:47:49 EDT