[PATCH] board enable for DFI AD77

[Tadas Slotkus]

Hello,

Thank you people on the IRC for helping with assembler :)

Happy Easter!

[Tadas Slotkus]

Hi,

updated patch for current tree rev.

[Tadas Slotkus]

Sorry, forgot this:
Signed-off-by: Tadas Slotkus

Could someone review?

[Stefan Tauner]

On Sun, 22 May 2011 19:27:49 +0300
Tadas Slotkus  wrote:
> Could someone review?

hey

the url should link to the manufacturer's site if possible e.g.
http://dfics.dfi.com.tw/Product/xx_product_spec_details_r_us.jsp?PRODUCT_ID=...

you have used the pciids of the generic VT8377 host bridge and the
audio controller. the audio controller is the only device with a DFI
subsystem id, but it can probably be disabled in the bios and then it
does not match...
since we can (and you patch does) use dmi matching anyway i think we
should use (for example) the isa bridge instead of the audio controller.
also we could mark it as tested if we match ^AD77$ instead of ^AD77.

any comments?

[Uwe Hermann]

Hi,

On Mon, May 23, 2011 at 07:12:40AM +0200, Stefan Tauner wrote:
> On Sun, 22 May 2011 19:27:49 +0300
> Tadas Slotkus  wrote:
> > Could someone review?
>
> hey
>
> the url should link to the manufacturer's site if possible e.g.
> http://dfics.dfi.com.tw/Product/xx_product_spec_details_r_us.jsp?PRODUCT_ID=...
>
> you have used the pciids of the generic VT8377 host bridge and the
> audio controller. the audio controller is the only device with a DFI
> subsystem id, but it can probably be disabled in the bios and then it
> does not match...
> since we can (and you patch does) use dmi matching anyway i think we
> should use (for example) the isa bridge instead of the audio controller.
> also we could mark it as tested if we match ^AD77$ instead of ^AD77.

A few more comments:

Tadas, could you please re-run your lspci as "lspci -xxxnnnvvv" as root,
the current one is missing the full info.

Also, did you try a write/erase operation on the board/chip? Without a
successful write operation we cannot mark the board/chip as supported,
reading alone is not enough unfortunately.

Please post write logs (-w -VV) and the lspci mentioned above, if possible.

Thanks, Uwe.

[Stefan Tauner]

On Thu, 23 Jun 2011 17:53:24 +0200
Uwe Hermann  wrote:
> Hi,
>
> On Mon, May 23, 2011 at 07:12:40AM +0200, Stefan Tauner wrote:
> > On Sun, 22 May 2011 19:27:49 +0300
> > Tadas Slotkus  wrote:
> > > Could someone review?
> >
> > hey
> >
> > the url should link to the manufacturer's site if possible e.g.
> > http://dfics.dfi.com.tw/Product/xx_product_spec_details_r_us.jsp?PRODUCT_ID=...
> >
> > you have used the pciids of the generic VT8377 host bridge and the
> > audio controller. the audio controller is the only device with a DFI
> > subsystem id, but it can probably be disabled in the bios and then it
> > does not match...
> > since we can (and you patch does) use dmi matching anyway i think we
> > should use (for example) the isa bridge instead of the audio controller.
> > also we could mark it as tested if we match ^AD77$ instead of ^AD77.
>
> A few more comments:
>
> Tadas, could you please re-run your lspci as "lspci -xxxnnnvvv" as root,
> the current one is missing the full info.
>
> Also, did you try a write/erase operation on the board/chip? Without a
> successful write operation we cannot mark the board/chip as supported,
> reading alone is not enough unfortunately.
>
> Please post write logs (-w -VV) and the lspci mentioned above, if possible.

I have committed a refined version in r1581.
It uses the generic PCI IDs and therefore requires manual activation
via -p internal:mainboard=…