Re: [ACPI-sppt] supported machines

Re: [ACPI-sppt] supported machines

[Sebastian Henschel]

[-- Attachment #1: Type: text/plain, Size: 2039 bytes --]

hello derek..

i think, i'll put this onto acpi-devel, too.

* Derek Broughton <dbroughton-F3m4/jXepmf3fQ9qLvQP4Q@public.gmane.org> [2003-06-13 09:38 +0200]:
> From: "Emma Jane Hogbin" <emmajane-MHIYrZpDPrNWk0Htik3J/w@public.gmane.org>
> 
> > On Thu, Jun 12, 2003 at 11:42:57AM -0300, Derek Broughton wrote:
> > > So everybody who's reading, please take the time to check those pages and if
> > > your BIOS/system is not listed please add what you can.
> >
> > Do you mean here?
> > http://acpi.sourceforge.net/wiki/index.php/InstallationReports
> >
> 
> I actually meant the page at http://acpi.sourceforge.net/dsdt/index.php, but I
> hadn't realized that they actually want the DSDT asl included.  I think that's a
> mistake.  
> In the first place, it probably isn't legal to post manufacturer's
> disassembled DSDTs (it should really only include patches), and 

well, if any manufacturer reads this, please raise your voice. :)
only including patches whould also have a technical benefit: the file
sizes would become smaller.
so if you found some more supporters, i would update the pages, files and
database as good as i can. the downside is that i cannot generate a diff
from the DSTDs where there is no "original" table provided.

> in the second
> that prevents using the list as a good source of the information you're looking
> for with lookups for manufacturer and model.  It should exist somewhere, and I
> thought this was the place, but I guess not.

hmm... it really is not thought of the place to "have a list where you
can see how good ACPI works with which laptop". the focus is on offering
fixed DSDTs for people not able to do that themselves. though it is true
that you can use some data provided there to compile such a list...

> So, yes, update the wiki InstallationReports page if you can. :-)

agreed, you should use that, meanwhile...

cheers,
 sebastian
-- 
::: sebastian henschel
::: kodeaffe
::: lynx -source http://www.kodeaffe.de/shensche.pub | gpg --import

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [ACPI] Re: [ACPI-sppt] supported machines

[Karol Kozimor]

Thus wrote Sebastian Henschel:
> well, if any manufacturer reads this, please raise your voice. :)
> only including patches whould also have a technical benefit: the file
> sizes would become smaller.
> so if you found some more supporters, i would update the pages, files and
> database as good as i can. the downside is that i cannot generate a diff
> from the DSTDs where there is no "original" table provided.

On the other hand, having the DSDT's in one place is very handy and helps
in driver development. At least, we used the database frequently with
Julien, and handling the development without it would be much less
convenient. Does file size (we're talking kilobytes of gzipped data) really
matter anyway?
Best regards,

-- 
Karol 'sziwan' Kozimor
sziwan-DETuoxkZsSqrDJvtcaxF/A@public.gmane.org


-------------------------------------------------------
This SF.NET email is sponsored by: eBay
Great deals on office technology -- on eBay now! Click here:
http://adfarm.mediaplex.com/ad/ck/711-11697-6916-5

Re: [ACPI] Re: [ACPI-sppt] supported machines

[Sebastian Henschel]

[-- Attachment #1: Type: text/plain, Size: 2210 bytes --]


hi karol..

* Karol Kozimor <sziwan-DETuoxkZsSqrDJvtcaxF/A@public.gmane.org> [2003-06-16 09:06 +0200]:
> Thus wrote Sebastian Henschel:
> > well, if any manufacturer reads this, please raise your voice. :)
> > only including patches whould also have a technical benefit: the file
> > sizes would become smaller.
> > so if you found some more supporters, i would update the pages, files and
> > database as good as i can. the downside is that i cannot generate a diff
> > from the DSTDs where there is no "original" table provided.
> 
> On the other hand, having the DSDT's in one place is very handy and helps
> in driver development. At least, we used the database frequently with
> Julien, and handling the development without it would be much less
> convenient.

you mean you looked into the DSDTs of machines you did not have at hand?
valid point, i think.

another problem i found with diffs is the following:
we have no "real" source code of the table. what we have is some binary
file where we run a tool for disassembly on. _most_ people use iasl, but
do they all do? and apart from that, perhaps different versions of the same
tool produce different ASL code. so it is quite possible that a patch
can fail, which can be a trivial... or something more sophisticated.

> Does file size (we're talking kilobytes of gzipped data) really
> matter anyway?

well, one archive of DSDTS is about 590k (40k DB) at the moment. this
surely does not matter. but due to constraints of sf.net i had to
install a little backup scheme which accumulates that amount of data for
each day. i usually download the backups once or twice a week and delete
them on sf.net afterwards. since we have 100MB available on sf.net, it
should not be a problem for now. but the backup scheme must be slowed
down when i am away from internet for a couple of weeks. so no, it does
not matter that much, just a little.

ok, i suggest to put the discussion back on acpi-support only, since no
big discussion has arisen on acpi-devel, but they know about it. :)

cheers,
 sebastian
-- 
::: sebastian henschel
::: kodeaffe
::: lynx -source http://www.kodeaffe.de/shensche.pub | gpg --import

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]