linux-ide.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Akkana Peck <akkana@shallowsky.com>
To: linux-ide@vger.kernel.org
Subject: Problem with PCMCIA IDE
Date: Sun, 18 Nov 2007 12:23:09 -0800	[thread overview]
Message-ID: <20071118202309.GE15355@shallowsky.com> (raw)

I've been having problems getting PCMCIA IDE to work on my Vaio
laptop on recent kernels (I've been working most with 2.6.23.1 and
.8). I've been hesitant to report it as a kernel bug because it
may very well be a problem with my .config parameters -- the CDROM
worked in distro kernels back around 2.6.17 and .18, and self-built
kernels even earlier than that, but the configuration has changed
quite a bit since then so it's hard to compare directly and figure
out if I'm missing something.

First, is this a reasonable place to report problems with PCMCIA,
or does it belong on a PCMCIA list?

The machine is a Vaio SR17 and I'm hotplugging a PCMCIA CDROM drive
(this is a subnotebook, so the drive is external).  I don't think
it's CD specific (the CONFIG_BLK_DEV_IDECD code), because /proc/ide
only shows ide0, not ide2 like I'm used to seeing in earlier kernels
when the CD drive was connected.

PCMCIA/Cardbus itself works fine (e.g. for network cards).
There's no sata (this is an old machine).

Depending on how I set kernel options, I can get the kernel to lock
up when the PCMCIA CDROM drive is plugged in, or not. When it locks,
it stays locked up (no keyboard or X events), until I eject the
PCMCIA plug, then after a few seconds things are back to normal.
Either way, it never notices the CD drive or creates an ide2 bus.

In the cases where it locks up, dmesg says:
BUG: soft lockup detected on CPU#0!
 [<c01213b7>] update_process_times+0x27/0x70
 [<c0130e7d>] tick_nohz_handler+0x5d/0xf0
 [<c01069ab>] timer_interrupt+0xb/0x20
 [<c013c795>] handle_IRQ_event+0x25/0x60
 [<c013dd14>] handle_level_irq+0x44/0xa0
 [<c0105d72>] do_IRQ+0x42/0x80
 [<c01041af>] common_interrupt+0x23/0x28
 [<c011dbcd>] __do_softirq+0x2d/0x90
 [<c011dc56>] do_softirq+0x26/0x30
 [<c011df1a>] irq_exit+0x5a/0x60
 [<c0105d77>] do_IRQ+0x47/0x80
 [<c01041af>] common_interrupt+0x23/0x28
 [<c02eb163>] ide_register_hw_with_fixup+0x33/0x1e0
 [<c02fc3d6>] idecs_register+0x76/0x80
[ various more backtrace, then ]
 =======================
Probing IDE interface ide2...

then that last message keeps repeating until I eject.

Since it's obviously complaining about IRQs, I tried turning on
CONFIG_IDEPCI_SHARE_IRQ=y (I also turned off CONFIG_BLK_DEV_IDEACPI
to see if that made any difference) and that stopped the
"BUG: soft lockup", but I still got the loop over "Probing IDE
interface ide2..." and it still doesn't see the CDROM or ide2.

I've put the full .config files at http://shallowsky.com/tmp/configs/
so as not to clutter this list.

Can anybody offer any tips for figuring out whether this is a
config error on my part, or a real kernel bug in IDECS?
Or send me to a more appropriate forum, if this isn't it?

Thanks in advance!

	...Akkana

             reply	other threads:[~2007-11-18 20:52 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-11-18 20:23 Akkana Peck [this message]
2007-11-18 22:37 ` Problem with PCMCIA IDE Alan Cox
2007-11-19  2:58   ` Akkana Peck
2007-11-22 22:14 ` Bartlomiej Zolnierkiewicz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20071118202309.GE15355@shallowsky.com \
    --to=akkana@shallowsky.com \
    --cc=linux-ide@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).