[Fw_Os_Forum] possible UEFI vs ACPI documentation disagreement/improvement suggestions

Rudolf Marek r.marek at assembler.cz
Thu Jun 30 17:01:58 EDT 2016


Hi all,

No idea if this is correct channel. I'm looking for some kind of contact of
responsible entity of ACPI specification. I noticed following:

ACPI specs 6.1

1) Please mark in the "5.2.5.3 RSDP Structure" in a more clear in a way that it
is understandable where the ACPI 1.0 RSDP ends. The ACPI 1.0 ends at offset 20,
and contains no length field. This seems to be never ending problem. It looks
like OVMF has also some issue (I need to investigate that)

Even ACPICA guys had it wrong some time ago
http://bugs.acpica.org/show_bug.cgi?id=1097

2) The ACPI and UEFI specs seems to contradict of when it is safe to retrieve
the ACPI RSDP from UEFI

ACPI spec says:

5.2.5.2 Finding the RSDP on UEFI Enabled Systems

The OS loader must retrieve the pointer to the RSDP structure from the EFI
System Table *before* assuming platform control via the EFI ExitBootServices
interface. See the UEFI Specification for more information.

However UEFI 2.6 says:

4.3 EFI System Table
...

After an operating system has taken control of the platform with a call to
ExitBootServices(), only the Hdr, FirmwareVendor, FirmwareRevision,
RuntimeServices, NumberOfTableEntries, and ConfigurationTable fields are
valid.

So, is the RSDP valid after ExitBootServices or not?

3)  The ACPI specs does not mention this GUID:

#define ACPI_TABLE_GUID \
 {0xeb9d2d30,0x2d88,0x11d3,\
 {0x9a,0x16,0x00,0x90,0x27,0x3f,0xc1,0x4d}}

It looks there are some duplicate names for same ACPI GUIDs, I did not noticed
this at first. Perhaps the UEFI specs can make it more clear?

4) I failed to find the contact inside the document for this kind of
questions/remarks

Thanks
Rudolf


More information about the Fw_os_forum mailing list