Re: [PATCH]: ACPI : Set 32bit and 64bit waking vector in FCAS table

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Friday, 5 of September 2008, Li, Shaohua wrote:
> 
> >-----Original Message-----
> >From: linux-acpi-owner@vger.kernel.org [mailto:linux-acpi-
> >owner@vger.kernel.org] On Behalf Of Zhao Yakui
> >Sent: Friday, September 05, 2008 9:17 AM
> >To: Matthew Garrett
> >Cc: Rafael J. Wysocki; Zhang, Rui; lenb@kernel.org; linux-
> >acpi@vger.kernel.org; andi@firstfloor.org
> >Subject: Re: [PATCH]: ACPI : Set 32bit and 64bit waking vector in FCAS
> >table
> >
> >On Thu, 2008-09-04 at 13:07 +0100, Matthew Garrett wrote:
> >> On Thu, Sep 04, 2008 at 11:37:51AM +0200, Rafael J. Wysocki wrote:
> >> > > so it seems that the BIOS sets facs->xfirmware_waking_vector during
> >> > > POST, but uses facs->firmware_waking_vector to get back during resume.
> >> >
> >> > So the BIOS is buggy, so let's add a quirk for it.
> >>
> >> Does the machine resume in Windows? If so, do we have any evidence that
> >> Windows has a quirks list to handle this case? If not, then I suspect
> >> that Windows sets both and this is what everyone has tested against.
> >The laptop can be resumed on windows.(XP & Vista). And we don't know
> >whether there exists the quirk list to handler this case on windows.
> >Maybe what you said is right.
> >In fact it is harmless when both 32bit and 64bit waking vector in FACS
> >table are set. When the system is resumed, BIOS will transfer control to
> >the predefined waking vector. As we set the same waking vector, either
> >of them is OK.
> >
> >There exists the difference between 32bit and 64bit waking vector unless
> >the waking address is above 4GB. But in fact the waking address is below
> >1MB on most machines as the waking address needs to be accessed by BIOS.
> >
> >So in most cases the 32bit and 64bit waking vector are the same value.
> >BIOS can transfer control to either of them.
> There was discussion about this issue several months ago (intel's ml), looks
> people forgot to take action after the discussion. The spec owner said 64bit
> vector is used in protected mode. That is if OS sets it, wakeup code is
> called in protected mode by BIOS.

Hm, that's interesting.  Do we have any examples of _working_ systems on which
we set the 64-bit vector?

> So the 64-bit vector shouldn't be used.    

OK

So perhaps let's remove the setting of the 64-bit vector altogether (with an
appropriate comment in the source code) and see if that causes any regressions
to happen.

Thanks,
Rafael
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html