From: "Michael R. Head" <burner@suppressingfire.org>
To: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: Apparent problems with interrupt latency with PCI2PCI bridges (register_serial(): autoconfig failed)
Date: Thu, 05 Jan 2006 11:59:14 -0500 [thread overview]
Message-ID: <1136480354.9411.26.camel@phoenix> (raw)
In-Reply-To: <1136398726.31979.29.camel@phoenix>
[-- Attachment #1: Type: text/plain, Size: 2934 bytes --]
In case anyone else ends up with this problem, here's how it's
resolving...
I've been told that this is a result of PCI interface controller on the
card (a PLX PCI9050) having trouble with PCI bridges. All of the cards
that I have been having trouble with use this interface controller, so
that seems to explain my problems.
On Wed, 2006-01-04 at 13:18 -0500, Michael R. Head wrote:
> Hello, I'm having some trouble with a number of PCI cards and several
> PCI bridges. In order to reduce the problem, I'm going to ask about 1
> specific card and one specific bridge.
>
> The hardware I'm working with is a P4 2.4Ghz machine in a 4U rackmount
> chassis. On board are 3 PCI slots connected to the main PCI bus and 9
> PCI slots connected to a bridge to the main PCI bus.
>
> The software I'm running is an up to date RHEL3 (kernel 2.4.21-37.EL),
> but I experience the same problems when I boot off an Ubuntu Breezy
> LiveCD (which uses 2.6.12).
>
> I have an 8 port RS232 card using 16550A UARTs (Sealevel Systems Inc
> Eight Port RS-232 Interface). When I plug it into the main PCI bus (#2),
> it works perfectly and I get ttyS4-11 working automatically and
> properly.
>
> However, when I put the card in a PCI slot on bus #3 (which is on the
> other side of a Pericom Semiconductor PCI to PCI Bridge), I see this
> during boot:
>
> $ dmesg | grep -2 serial
> ttyS3 at 0x02e8 (irq = 3) is 16550A
> register_serial(): autoconfig failed
> Real Time Clock Driver v1.10e
>
>
> When I plug it into a slot on bus #2, I get this from setserial:
> $ setserial -g /dev/ttyS4
> /dev/ttyS4, UART: 16550A, Port 0xd200, IRQ: 2
>
> When I plut the card into a slot on bus #3, setserial tells me this (by
> default);
> $ setserial -g /dev/ttyS4
> /dev/ttyS4, UART: unknown, Port: 0xc100, IRQ: 2
>
> If I attempt to reconfigure ttyS4 to have the settings from above, I get
> the dreaded "ttyS4: LSR safety check engaged!" and from then on, any
> access to /dev/ttyS4 results in "/dev/ttyS4: No such device".
>
>
>
> Now, I also have some other PCI cards that are having trouble with
> living on the other side of a PCI bridge, but their drivers are
> available separately from the kernel. I figure that if I can get the
> serial card working, then hopefully the rest will work, too.
>
> I also have an external 4U rackmount chassis containing 12 additional
> PCI slots that are connected via 4 Intel 21142 PCI-to-PCI bridges, and
> when I connect the serial card into any of those busses, I get the same
> problem.
>
>
> I've tried fiddling with all sorts of kernel PCI options at boot and
> fiddling with various setpci commands. Nothing seems to help. Any
> suggestions or information would be greatly appreciated.
>
> thanks,
> mike
>
--
Michael R. Head <burner@suppressingfire.org>
GPG: http://www.suppressingfire.org/~burner/gpg.key.txt [0x4C9DA1D0]
[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]
prev parent reply other threads:[~2006-01-05 17:00 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-01-04 18:18 Apparent problems with interrupt latency with PCI2PCI bridges (register_serial(): autoconfig failed) Michael R. Head
2006-01-05 16:59 ` Michael R. Head [this message]
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=1136480354.9411.26.camel@phoenix \
--to=burner@suppressingfire.org \
--cc=linux-kernel@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.