Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 8 Jun 2004 09:50:08 +0800 (CST)
From:      Tai-hwa Liang <avatar@mmlab.cse.yzu.edu.tw>
To:        Jeffrey Katcher <jmkatcher@yahoo.com>
Cc:        freebsd-current@FreeBSD.org
Subject:   Re: Atheros support no longer works on -CURRENT (kldload, possibly ACPI?)
Message-ID:  <0406080934431.98630@www.mmlab.cse.yzu.edu.tw>
In-Reply-To: <20040606181203.45983.qmail@web41111.mail.yahoo.com>
References:  <20040606181203.45983.qmail@web41111.mail.yahoo.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 6 Jun 2004, Jeffrey Katcher wrote:
> Hardware: IBM Thinkpad T40, IBM Atheros A/B card
> Software: -CURRENT as of 6/5/04, but older 5/25 kernel still shows problem

My T40 with IBM Atheros A/B/G miniPCI works well with 5/25 kernel. However,
I have to backout sys/dev/acpica/acpi_cpu.c to 1.38 and
sys/dev/acpica/acpi_pci_link.c to 1.14 such that it would not lockup at
booting after sio0 probing.

>
> Symptoms: I don't use it very frequently, so I kldload if_ath as necessary.
> Previously, it was load, config, and go, now...
>
> I see:
> pci0: Failed to set ACPI power state D3 on (null): AE_BAD_PARAMETER
> before the usual (and correct) ath0 status messages

Did you try the latest source committed by njl@(about 3 hours ago)?
I can't test it unless backing to home; however, they seemed to be able
to eliminate the BAD_PARAMETER error under some circumstances:

	sys/dev/acpica/acpi_pci.c:1.18
	sys/dev/acpica/acpi_powerres.c:1.23

> When I "up" the interface, I see:
> ath0: hardware error; resetting
> in a continuous parade scrolling up the screen

I always put "if_ath_load=YES" in /boot/loader.conf. It works for me since
the first time I purchased this card. Perhaps you would like to give it a try.

> I tried setting hw.ath.debug, but the dump scrolls off faster than I can see
> it.  The system has no serial port to capture from...

Will those lines captured by syslogd?

> This could be an ACPI issue instead, but I don't see anything like this for
> the compiled-in set of drivers.

The last resort: reboot with ACPI disabled. At least this saved me from
troubles(if_em locks up) with recent -CURRENT.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?0406080934431.98630>