Copy stable/9 to releng/9.3 as part of the 9.3-RELEASE cycle.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

MFC: r251186

Fix a long standing logic bug introduced in r167814. The code was added to
get RSDP from loader(8) hint via kenv(2) but the bug nullified the new code
and we always fell back to the previous method, i. e., sysctlbyname(3).

Copy head to stable/9 as part of 9.0-RELEASE release cycle.

Approved by: re (implicit)

Merge ACPICA 20100702.

- Use the headers from ACPI-CA to define various constants and structures
for table layouts, etc. rather than homerolling our own structures and
constants in acpidump.h.
- Verify the extended checksum on the RSDP.
- Handle new ACPI 3.0 fields in MADT including X2APIC entries and
UIDs for local SAPICs.
- Add handling for new ACPI 3.0 flags in the FADT.

Reviewed by: jkim
MFC after: 1 month

Fix acpidump(8) on ia64. Revision 1.13 introduced an uninitialized
variable bug that's hidden by the precense of the hint_acpi_0_rsdp
hint on 386 and amd64. There's never a need for such hint on ia64.

Approved by: re (kensmith)

Catch up with ACPI-CA 20070320 import.

Unify the start/size parameters for the RSDP search area. Don't bother
trying to exclude the top end of the range since it should hurt to overlap
by 4 bytes in the off-chance the RSDP signature appears incorrectly at the
very top of our search space.

Use the correct location of the EBDA for searching for the RSDP.
The EBDA is the 1 KB area addressed by the 16 bit pointer at 0x40E.

Pointed out by: robert.moore AT intel.com

If the revision is 0, don't trust the length in the RSDP. Instead assume
an ACPI version 1.0 length.

Read in the entire RSDP but only run the standard checksum over the
version 1 header. Add comments to explain what we're doing here better.

Reported by: Alex Vasylenko <lxv@omut.org>

Instead of scanning the entire lower 1 MB of RAM, only scan locations
where the RSD PTR can actually occur. According to section 5.2.2
of the ACPI spec, we only consider two regions for the base address:

1. EBDA (0x0 - 0x3FF)
2. High memory (0xE0000 - 0xFFFFF)

I don't know whether this fixes any actual problems but is more correct.

Minor style issues.

Modify acpidump to use iasl(8) as the backend for disassembling AML.
Also clean up the output of dumped tables. Update the man page for the
new usage. Make WARNS=6 clean.

o Use sysctl machdep.acpi_root to get the physical address of the
RSDP. Scan the first 1MB on i386 if the sysctl fails,
o Extend struct ACPIrsdp with the ACPI 2.0 fields which involves
changing a prior reserved field into the ACPI revision,
o Only calculate the RSDP checksum on the first 20 bytes to remain
compatible with ACPI 1.0 tables; we don't check the extended
checksum covering the whole table,
o Use the length field in the RSDP to map the RSDP into the address
space so that we don't have to know about future extensions here.

Fix UserTerms disassembling. Now that dumped ASL can be compiled with
ports/devel/acpitools (iasl).
- Merge AML parser to build ACPI namespace
- Comment header info. out so that ASL compiler ignore them
- Fix DSDT header size to be discarded when DSDT file is specified
for input (acpidump and amldb)
- Write DSDT header as well into DSDT file for output
- Fix some trivial typo (Concatenate and SizeOf)
- Remove DEBUG_FLAGS from Makefile (acpidump and amldb)

Remove dependence on old acpi driver from acpi userland tools.

- acpiconf Replace include files from old acpi driver to acpica driver.
New sleep type `4b' had been added (S4BIOS) for `-s' option.
Of course this has no effect because driver doesn't
support it for now :-)

- acpidump All needed structs in sys/dev/acpi/*.h had been merged
into local header file. No changes on its usage.

Update include paths for new header locations.

Submitted by: iwasaki

This commit was generated by cvs2svn to compensate for changes in r65285,
which included commits to RCS files with non-trunk default branches.

import acpidump(8) from ACPI For FreeBSD project.

Obtained from: ACPI For FreeBSD project