[2.6.11.7 / CLPS711x/SkyMinder] module integration issue: keyboard driver _still_ not working after port from 2.4.27

From: Luke Kenneth Casson Leighton (lkcl_at_lkcl.net)
Date: 04/21/05

  • Next message: Stan Bubrouski: "Re: [Gelato-technical] Re: Serious performance degradation on a RAID with kernel 2.6.10-bk7 and later"
    Date:	Thu, 21 Apr 2005 19:28:45 +0100
    To: linux-kernel@vger.kernel.org, Linux ARM Kernel list <linux-arm-kernel@lists.arm.linux.org.uk>
    
    

    hi,

    after some thought, i decided on friday to port all of the 2.4.27
    code over to 2.6.11.7.

    bearing in mind that reading from /dev/input/evdev0 worked fine
    on 2.4.27 - and now also works fine under 2.6.11.7 - the exact
    same problem occurs on 2.6.11.7 as occurred under 2.4.27 -
    hanging of the linux kernel whilst providing absolutely zero
    crash/debugging information responses whatsoever.

    this device does _not_ have and cannot have a standard PC keyboard
    attached to it.

    in order to be able to debug what is going on, i have enabled
    a dummy/virtual serial console, all is well so far.

    in order to test the screen, i have written, enabled, tested,
    confirmed as reasonably working, a framebuffer driver (which
    i would like to make the console framebuffer - eventually -
    when the serial console is disabled and no longer needed -
    so i am enabling Framebuffer Console support AS WELL as serial
    console support)

    now i load the keyboard event module... splat - absolutely no response:
    complete lock-up.

    okay, let's try that again.

    reboot.

    install the keyboard event module.

    install evdev module.

    run a program (multilog) which is a bit like tail -f except it outputs
    to a logging file rather than to stdout on /dev/input/evdev0 and press a
    few buttons.

    do a hexdump -C on the output from multilog - GREAT!

    oh...keeyyy...

    so, individually, the components work fine.

    put them together, and something goes badly wrong.

    has _anyone_ else tried running two console drivers at the
    same time?

    should it be expected to work?

    any assistance / guidance greatly appreciated.

    l.

    ----- Forwarded message from Luke Kenneth Casson Leighton <lkcl@lkcl.net> -----

    From: Luke Kenneth Casson Leighton <lkcl@lkcl.net>
    To: linux-kernel@vger.kernel.org
    Subject: [2.4.27 arm skyminder] writing new keyboard handler - help!

    hi, please respond cc to me because i am not on the lkml.

    for a new arm linux embedded device called a skyminder, i'm responsible
    for getting all the drivers working. ha ha.

    the success of this device presents the linux community with
    an opportunity to own a linux-based mobile phone (even if it's
    not a very small phone - 8cm x 10cm x 1cm) - and it has a GPS
    module in it, as well.

    i'm endeavouring to adapt various bits of code to create a
    keyboard driver. they've adopted 2.4.27 and are too far down
    the line to move to 2.6 - yet.

    i particularly want to avoid - if i can - compiling this keyboard
    driver under development into the kernel (even though it's the primary
    keyboard) because downloading 600k over a serial link into flash ram
    isn't a) funny b) a good idea c) slows development time down.

    with that in mind, so far, i have:

    - cut/paste pc_keyb.c just like everyone else has (in celps_keyb.c,
      c711x_keyb.c, dummy_keyb.c etc.) to create k_translate,
      k_unexpected_up, k_setkeycode and k_getkeycode routines.

    - cut/paste usbkbd.c and adapted it to successfully call
      input_report_key on a key press and key release.

    then, on installation of module input, keybdev and sky_buttons,
    i happily get debug messages indicating key presses (keycode 31
    indicating 's') ... but no actual key events appear down my serial
    console.

    so, my question is: does anyone know off the top of their heads what i
    may have missed out that causes the keybdev event handler to _not_
    actually stuff keys out?

    am i... like... missing something really obvious, given that
    the console has been set to "serial"?

    where should i look to, to find the keys being outputted, if they're
    going anywhere?

    help, help, gloop.

    cheers,

    l.

    -- 
    --
    <a href="http://lkcl.net">http://lkcl.net>
    --
    ----- End forwarded message -----
    -- 
    --
    <a href="http://lkcl.net">http://lkcl.net>
    --
    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at  http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at  http://www.tux.org/lkml/
    

  • Next message: Stan Bubrouski: "Re: [Gelato-technical] Re: Serious performance degradation on a RAID with kernel 2.6.10-bk7 and later"