[PATCH] Drop W83627THF, it's the same device as W83627THG

[Uwe Hermann]

See patch.

Uwe.

[PATCH] Drop W83627THF, it's the same device as W83627THG

[Stefan Reinauer]

Should we rename the part to w83627 then?

Stefan

Sent from my mobile phone

On 16.11.2010, at 11:47, Uwe Hermann  wrote:
> See patch.
> Uwe.
>
> http://hermann-uwe.de     | http://sigrok.org
> http://randomprojects.org | http://unmaintained-free-software.org
> <v4_drop_duplicated_w83627thf.patch>
>
> coreboot mailing list: [e-mail]
> http://www.coreboot.org/mailman/listinfo/coreboot

[Uwe Hermann]

On Tue, Nov 16, 2010 at 12:06:23PM -0800, Stefan Reinauer wrote:
> Should we rename the part to w83627 then?

That's too generic, many other Winbond Super I/Os start with that prefix:

w83627dhg
w83627ehg
w83627hf
w83627thf
w83627thg
w83627uhg

(and probably some others)

We could do w83627th, or w83627thf_g or so, but I guess it's also fine
to just keep one of the two names. I'll post a follow-up patch which
adds code comments like "This code supports W83627THF/THG" etc. to
all Super I/Os anyway to document this a bit better.

Uwe.

[Stefan Reinauer]

* Uwe Hermann  [101116 21:46]:
> On Tue, Nov 16, 2010 at 12:06:23PM -0800, Stefan Reinauer wrote:
> > Should we rename the part to w83627 then?
>
> That's too generic, many other Winbond Super I/Os start with that prefix:
>
> w83627dhg
> w83627ehg
> w83627hf
> w83627thf
> w83627thg
> w83627uhg
>
> (and probably some others)

How different are they? Is there any knowlegde about what the letters
mean? ie. do dhg, ehg and hf share some "h" feature while thf and thg
share "t"? Or is it more or less random, or even big internal changes
like for the ICH southbridges?
> We could do w83627th, or w83627thf_g or so, but I guess it's also fine
> to just keep one of the two names.

I don't particularly like the underscore, but w83627th sounds good.
> I'll post a follow-up patch which
> adds code comments like "This code supports W83627THF/THG" etc. to
> all Super I/Os anyway to document this a bit better.

Please remember to add the change to our Flag Days page:
http://www.coreboot.org/Flag_Days

[Uwe Hermann]

On Tue, Nov 16, 2010 at 10:05:47PM +0100, Stefan Reinauer wrote:
> Uwe Hermann  [101116 21:46]:
> > On Tue, Nov 16, 2010 at 12:06:23PM -0800, Stefan Reinauer wrote:
> > > Should we rename the part to w83627 then?
> >
> > That's too generic, many other Winbond Super I/Os start with that prefix:
> >
> > w83627dhg
> > w83627ehg
> > w83627hf
> > w83627thf
> > w83627thg
> > w83627uhg
> >
> > (and probably some others)
>
> How different are they? Is there any knowlegde about what the letters
> mean? ie. do dhg, ehg and hf share some "h" feature while thf and thg
> share "t"? Or is it more or less random, or even big internal changes
> like for the ICH southbridges?

I'd say more or less random. Only the F/G suffix usually means Pb-free
or not, the rest doesn't directly translate to "similar enough" I think.

Now, it would be possible to do one common driver for multiple (probably
not all, but many) Winbond Super I/Os, see smscsuperio for an
(admittedly incomplete and a bit convoluted) example, but I had a feeling
you normally do not like such unified drivers all that much ;)

I'd like to drop the definately useless W83627THF first though, if there is
enough interest I can also try to do a more unified driver later.
> > We could do w83627th, or w83627thf_g or so, but I guess it's also fine
> > to just keep one of the two names.
>
> I don't particularly like the underscore, but w83627th sounds good.

OK, I'll make another patch to rename _all_ superios which have
multiple irrelevant suffixes accordingly.
> > I'll post a follow-up patch which
> > adds code comments like "This code supports W83627THF/THG" etc. to
> > all Super I/Os anyway to document this a bit better.
>
> Please remember to add the change to our Flag Days page:
> http://www.coreboot.org/Flag_Days

Yup, will do after the commit.

Uwe.

[PATCH] Drop W83627THF, it's the same device as W83627THG

[Peter Stuge]

Uwe Hermann wrote:
> Drop W83627THF, it's the same device as W83627THG.
>
> The only difference is that the "G" version is in a Pb-free package, which
> is not relevant from a programmer's view.
>
> We keep W83627THG (and drop W83627THF) because:
>
> The W83627THF had a CIR device / LDN which doesn't actually exist.
>
> The W83627THF had no GPIO2, GPIO3 LDNs (were commented out).
>
> The W83627THF didn't use the PNP_MSC0/1 which is needed/used by boards.
>
> This also fixes an issue on MSI MS7135's devicetree.cb:
>
> device pnp 4e.6 off end           # XXX keep allocator happy
>
> The line above can be (and is) removed, as it was only needed due to the
> incorrect CIR LDN in the W83627THF.
>
> In the iwill/dk8x target: Drop incorrect LDNs 4 and 6, add 0xb.
>
> Signed-off-by: Uwe Hermann

Acked-by: Peter Stuge

[Uwe Hermann]

On Tue, Nov 16, 2010 at 11:51:08PM +0100, Peter Stuge wrote:
> Acked-by: Peter Stuge

Thanks, r6080.

Uwe.