Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 19 Dec 2006 02:46:46 +0800
From:      Stepan Zastupov <redchrom@gmail.com>
To:        Sam Leffler <sam@errno.com>
Cc:        Stepan Zastupov <redchrom@gmail.com>, freebsd-current@freebsd.org, Nate Lawson <nate@root.org>
Subject:   Re: cbb hangs during suspend if ath card active
Message-ID:  <20061218184646.GA7183@stepan.ispsystem.net>
In-Reply-To: <4585F7BB.9010704@errno.com>
References:  <20061214234849.GA1062@stepan.ispsystem.net> <4584708B.3020902@root.org> <20061217214408.GA902@stepan.ispsystem.net> <4585BAF3.1010504@root.org> <4585F7BB.9010704@errno.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Dec 17, 2006 at 06:06:51PM -0800, Sam Leffler wrote:
> Nate Lawson wrote:
> > Stepan Zastupov wrote:
> >> On Sat, Dec 16, 2006 at 02:17:47PM -0800, Nate Lawson wrote:
> >>> Stepan Zastupov wrote:
> >>>> Does anyone still work on this problem?
> >>>> (http://lists.freebsd.org/pipermail/freebsd-current/2006-July/064550.html)
> >>>>
> >>>> I incure the same problems. If need testing I can help with it.
> >>> Just comment out this line in ath_stop() in if_ath.c:
> >>>     // ath_hal_setpower(sc->sc_ah, HAL_PM_FULL_SLEEP);
> > 
> >> It is already commented out in current and doesn't help. Card wont work
> >> even if it inserted after suspend and resume.
> >>
> > 
> > That's something different then.  I was talking about the suspend
> > process hanging if the card is inserted and active.  But ejecting it
> > would allow the process to continue and after resume, re-inserting it
> > worked fine.
> > 
> 
> What Nate is referring to is that on many ath parts if you "power down"
> the MAC then any pci references to registers outside the PCI clock
> domain will hang the bus.  Not putting the chip in "full sleep" is a
> hack as it means it'll continue to draw full power after the interface
> is marked down.  Something is causing the driver to be entered and
> registers accessed when not intended/expected.  I haven't been able to
> make this happen on any of my laptops (most have ACPI problems s.t.
> suspend/resume doesn't work right) so haven't had a test case to
> debug--and noone with the problem has been able to provide the needed info.
Ok what info you need exactly? I can provide only the messages after
some systls verbose tuned.

-- 
Best regards, Stepan Zastupov aka RedChrom
ISPSystem



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20061218184646.GA7183>