cvs commit: src/usr.sbin/acpi/acpidump acpi.c acpidump.c acpidump.h

[Nate Lawson]

John Baldwin wrote:

> On Tuesday 05 October 2004 01:50 pm, Nate Lawson wrote:
> 
>>Takanori Watanabe wrote:
>>
>>>In message <200410050218.i952IrBK008982 at repoman.freebsd.org>, Nate Lawson
>>>wrote
>>>
>>>
>>>>njl         2004-10-05 02:18:53 UTC
>>>>
>>>>FreeBSD src repository
>>>>
>>>>Modified files:
>>>>  usr.sbin/acpi/acpidump acpi.c acpidump.c acpidump.h
>>>>Log:
>>>>Add the -s flag to make dumping SSDTs optional (disabled by default).
>>>>Since we can only override the DSDT, a custom ASL dumped previously that
>>>>contained SSDTs would result in lots of multiple definition errors.
>>>>
>>>>A longer-term fix involves adding the ability to override SSDTs to
>>>>ACPI-CA.
>>>
>>>iasl(8) may have to be able to compile multiple AML codes from single
>>>ASL data, instead of merging SSDT into DSDT in acpidump(8).
>>
>>I think the most straightforward fix is for ACPI-CA to not load SSDT
>>tables when the user is overriding the DSDT.  If you treat overridden
>>SSDTs separately, you go down the path of having to specify _which_
>>table you want to override, which implies you need strict ordering of
>>tables in both ASL and AML and it just gets too complicated.  Some HP
>>Itanium machines have 8 to 10 SSDTs.
> 
> 
> I agree.  It makes the user's lives a lot easier if they procedure remains:
> 
> - acpidump > foo.asl
> - hack on foo.asl to fix it
> - compile foo.asl to foo.dsdt
> - load foo.dsdt in the loader
> 
> and have it all just work.  Actually, you know, we could make this work on our 
> own without requiring ACPI-CA to change.  You just need to hack 
> AcpiOsTableOverride() in OsdTable.c to return a pointer to an empty table for 
> "SSDT" tables when the DSDT is overridden.  In fact, I think that is probably 
> where this behavior decision really belongs.

Good idea, I'll look into this.  If feasible, I'll commit this and 
revert the acpidump change.

-- 
Nate