sony-vaio-z-series team mailing list archive

[Adam Hill <sidepipeuk@xxxxxxxxx>]

	FWIW, my *guess* would be that it is the ACPI set-up and the way that the 
tables are scanned. There was probably some extension to the protocol which 
newer kernels support but older ones don't, and it is access to this extension 
which makes the BIOS think that the OS is capable of dynamic switching.

	However, if this is the case, it may be that NOT using this extended mode 
of operation will prevent some other feature from working.....

Adam.


On Tuesday 17 August 2010 10:30:44 Fabrizio Giustina wrote:
> Hi Martin,
> I posted this question before but nobody seems to have really
> discovered the change that broke the vga activation on newer kernels.
> 
> An interesting finding is that the 2.6.31-11-rt kernel available in
> the default ubuntu repository (lucid, also maverick I think) does
> work, while probably the stock 2.6.31-11 doesn't (not sure about the
> the last, but everywhere I saw notes about booting a <= 2.30 kernel).
> I had a quick look at the patches in the rt version, trying to find
> out anything related to acpi calls but I couldn't find anything that
> ringed a bell... really, I spent no more than half an hour on this, so
> my investigation is not so meaningful.
> 
> Another thing that may interest you is that in the past there was a
> way to trick old vaios by adding a acpi_osi="!Windows 2006" kernel
> parameter. This was enough to make the bios switch to the static vga
> selection just like for Windows XP. This doesn't work anymore,
> probably due to a different way the OS detection has been implemented
> in newer bioses, but the point is finding out which parameter the bios
> now uses to check for hot switch support (other idea, maybe the
> acpi_osi flag is handled differenctly in newer kernels?)
> 
> This is a post where I could find the most valuable info about the topic:
> http://0xc0dedbad.com/blog/2010/01/30/dual-boot-linux-and-windows-7-on-sony
> -vaio-z/
> 
> HTH
> fabrizio
> 
> On Sat, Aug 14, 2010 at 3:13 PM, Martin <nitram@xxxxxxxxx> wrote:
> > Hi
> > 
> > At some point I remember reading that one person would perhaps try and
> > track down the exact kernel commit which
> > causes the newer (> ~2.30...) kernels not to work properly. Does anyone
> > know if somebody is working on this and would it
> > in theory then be possible to reaply the changes to newer kernels and
> > thus avoid booting an old kernel first ?
> > 
> > I thought about trying to find that commit myself, if no one is doing it
> > and if it's worth tracking down.
> > 
> > -Martin
> > 
> > 
> > _______________________________________________
> > Mailing list: https://launchpad.net/~sony-vaio-z-series
> > Post to     : sony-vaio-z-series@xxxxxxxxxxxxxxxxxxx
> > Unsubscribe : https://launchpad.net/~sony-vaio-z-series
> > More help   : https://help.launchpad.net/ListHelp
> 
> _______________________________________________
> Mailing list: https://launchpad.net/~sony-vaio-z-series
> Post to     : sony-vaio-z-series@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~sony-vaio-z-series
> More help   : https://help.launchpad.net/ListHelp