Implemented support for FADT revision 2.  This was an
interim table (between ACPI 1.0 and ACPI 2.0) that adds
support for the FADT reset register.

Implemented optional support to allow uninitialized LocalX
and ArgX variables in a control method.  The variables
are initialized to an Integer object with a value
of zero.  This support is enabled by setting the
acpi_gbl_enable_interpreter_slack flag to TRUE,
which is default unless booted with "acpi=strict".

Implemented support for Integer objects for the SizeOf
operator.  Either 4 or 8 is returned, depending on the
current integer size (32-bit or 64-bit, depending on the
parent table revision).

Fixed a problem in the implementation of the SizeOf and
ObjectType operators where the operand was resolved to
a value too early, causing incorrect return values for
some objects.

Fixed some possible memory leaks during exceptional conditions.

Signed-off-by: Len Brown <len.brown@intel.com>

Note: ACPI CA is currently undergoing an in-depth and
complete formal evaluation to test/verify the following
areas. Other suggestions are welcome. This will result in
an increase in the frequency of releases and the number
of bug fixes in the next few months.

- Functional tests for all ASL/AML operators
- All implicit/explicit type conversions
- Bit fields and operation regions
- 64-bit math support and 32-bit-only "truncated" math support
- Exceptional conditions, both compiler and interpreter
- Dynamic object deletion and memory leaks
- ACPI 3.0 support when implemented
- External interfaces to the ACPI subsystem

Fixed two alignment issues on 64-bit platforms -
within debug statements in acpi_ev_gpe_detect and
acpi_ev_create_gpe_block. Removed references to the
Address field within the non-aligned ACPI generic address
structure.

Fixed a problem in the Increment and Decrement operators
where incorrect operand resolution could result in the
inadvertent modification of the original integer when the
integer is passed into another method as an argument and
the arg is then incremented/decremented.

Fixed a problem in the FromBCD operator where the upper
32-bits of a 64-bit BCD number were truncated during
conversion.

Fixed a problem in the ToDecimal operator where the length
of the resulting string could be set incorrectly too long
if the input operand was a Buffer object.

Fixed a problem in the Logical operators (LLess,
etc.) where a NULL byte (0) within a buffer would
prematurely terminate a compare between buffer objects.

Added a check for string overflow (>200 characters as per
the ACPI specification) during the Concatenate operator
with two string operands.

Signed-off-by: Len Brown <len.brown@intel.com>

Implemented support for the ACPI 3.0 Timer operator. This
ASL function implements a 64-bit timer with 100 nanosecond
granularity.

Defined a new OSL interface, acpi_os_get_timer. This
interface is used to implement the ACPI 3.0 Timer
operator. This allows the host OS to implement the timer
with the best clock available.  Also, it keeps the core
subsystem out of the clock handling business, since the
host OS (usually) performs this function.

Fixed an alignment issue on 64-bit platforms. The
hw_low_level_read/write() functions use a 64-bit address
which is part of the packed ACPI Generic Address
Structure. Since the structure is non-aligned, the
alignment macros are now used to extract the address to
a local variable before use.

Fixed a problem where the ToInteger operator assumed all
input strings were hexadecimal. The operator now handles
both decimal strings and hex strings (prefixed with "0x").

Fixed a problem where the string length in the string
object created as a result of the internal ConvertToString
procedure could be incorrect.  This potentially affected
all implicit conversions and also the ToDecimalString and
ToHexString operators.

Fixed two problems in the ToString operator. If the
length parameter was zero, an incorrect string object was
created and the value of the input length parameter was
inadvertently changed from zero to Ones.

Fixed a problem where the optional ResourceSource string
in the ExtendedIRQ resource macro was ignored.

Simplified the interfaces to the internal division
functions, reducing code size and complexity.

Signed-off-by: Len Brown <len.brown@intel.com>

Added a new OSL interface, acpi_os_get_timer.  This
interface implements a 64-bit monotonic timer in 100
nanosecond units.

Implemented support for the ACPI 3.0 Timer operator.
This 64-bit timer utilizes the timer provided by the
acpi_os_get_timer interface.

Signed-off-by: Len Brown <len.brown@intel.com>

Signed-off-by: Len Brown <len.brown@intel.com>

Fixed a problem with the implementation of the LNot()
operator where "Ones" was not returned for the TRUE
case. Changed the code to return Ones instead of (!Arg)
which was usually 1. This change affects iASL constant
folding for this operator also.

Fixed a problem in acpi_ut_initialize_buffer where an
existing buffer was not initialized properly -- Now zero
the entire buffer in this case where the buffer already
exists.

Changed the interface to acpi_os_sleep from (UINT32
Seconds, UINT32 Milliseconds) to simply (ACPI_INTEGER
Milliseconds). This simplifies all related code
considerably. This requires changes/updates to all OS
interface layers (OSLs.)

Implemented a new external interface,
acpi_install_exception_handler, to allow a system exception
handler to be installed. This handler is invoked upon
any run-time exception that occurs during control method
execution.

Added support for the DSDT in acpi_tb_find_table. This
allows the DataTableRegion() operator to access the local
copy of the DSDT.

Signed-off-by: Len Brown <len.brown@intel.com>

Implemented support for implicit object conversion in
the non-numeric logical operators (LEqual, LGreater,
LGreaterEqual, LLess, LLessEqual, and LNotEqual.) Any
combination of Integers/Strings/Buffers may now be used;
the second operand is implicitly converted on the fly to
match the type of the first operand.  For example:

LEqual (Source1, Source2)

Source1 and Source2 must each evaluate to an integer, a
string, or a buffer. The data type of Source1 dictates the
required type of Source2.  Source2 is implicitly converted
if necessary to match the type of Source1.

Updated and corrected the behavior of the string
conversion support.  The rules concerning conversion of
buffers to strings (according to the ACPI specification)
are as follows:

ToDecimalString - explicit byte-wise conversion of buffer
to string of decimal values (0-255) separated by commas.

ToHexString - explicit byte-wise conversion of buffer to
string of hex values (0-FF) separated by commas.

ToString - explicit byte-wise conversion of buffer to
string.  Byte-by-byte copy with no transform except NULL
terminated.  Any other implicit buffer-to-string conversion

byte-wise conversion of buffer to string of hex values
(0-FF) separated by spaces.

Fixed a problem in acpi_ns_get_pathname_length where the
returned length was one byte too short in the case of a
node in the root scope.  This could cause a fault during
debug output.

CONFIG_ACPI_BLACKLIST_YEAR=2001 for old behaviour

new cmdline options: "acpi_dbg_layer=",  "acpi_dbg_level="
and /proc/acpi/debug_layer, debug_level now describe levels

http://bugzilla.kernel.org/show_bug.cgi?id=2398

This will enable drivers to work around BIOS deficiencies,
while still allowing the drivers to be more picky with "acpi=strict"
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>

Signed-off-by: Takayoshi Kochi <t-kochi@bq.jp.nec.com>

Signed-off-by: Jesse Barnes <jbarnes@sgi.com>

These workarounds are disabled if "acpi=strict"

Designed and implemented support within the AML interpreter
for the so-called implicit return.  This support returns
the result of the last ASL operation within a control
method, in the absence of an explicit Return() operator.
A few machines depend on this behavior, even though it
is not explicitly supported by the ASL language.  It is
optional support that can be enabled at runtime via the
acpi_gbl_enable_interpreter_slack flag.

Removed support for the PCI_Config address space from the
internal low level hardware interfaces (acpi_hw_low_level_read
and acpi_hw_low_level_write).  This support was not used
internally, and would not work correctly anyway because
the PCI bus number and segment number were not supported.
There are separate interfaces for PCI configuration space
access because of the unique interface.
acpica-unix-20040816.patch

AE_CODE_AML_MAX fix from Bjorn Helgaas

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>

into intel.com:/home/lenb/src/linux-acpi-test-2.6.8

into intel.com:/home/lenb/src/linux-acpi-test-2.6.8

Signed-off-by: Jesse Barnes <jbarnes@sgi.com>

into intel.com:/home/lenb/src/linux-acpi-test-2.6.8

into intel.com:/home/lenb/src/linux-acpi-test-2.6.8

Fixes two common boot failures due to buggy SMM BIOS code

SMP boot crash if SMI_CMD=ACPI written from CPU1
http://bugzilla.kernel.org/show_bug.cgi?id=2941

laptop crash due to LAPIC timer before SMI_CMD=ACPI
http://bugzilla.kernel.org/show_bug.cgi?id=1269

Fix stupid thinkos in the fcntl f_op removal code.

into intel.com:/home/lenb/src/linux-acpi-test-2.6.8

Update lasi 82596 driver:
 - Fix boot messages (Helge Deller)
 - Whitespace and __FILE__ usage (Joel Soete)
 - 64-bit compile warning fixes (James Bottomley)

This trivial fix allows mounting MO-drives readwrite.

Acked by Jens.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

 - __PAGE_OFFSET is 0x10000000 (Randolph Chung)
 - PA8800 support (Grant Grundler)
 - debuglocks (Thibaut Varene)
 - PDC chassis disabling (Thibaut Varene)
 - Distinguish between Dinos in request_irq (Thibaut Varene)
 - Document interrupt registers (Randolph Chung)
 - Revamp CONFIG_DISCONTIGMEM support (Randolph Chung)
 - Remove STI console warning and special casing (Randolph Chung)
 - n4000 defconfig (Randolph Chung)
 - iosapic fixes (Bjorn Helgaas)
 - Fix a bug in entry.S where pa_dbit_lock was being trashed (Randolph Chung)
 - SMP support (Randolph Chung, Grant Grundler, James Bottomley)
 - Clear the pte in the fault handler (Joel Soete)
 - Change _exit prototype (Carlos O'Donell)
 - Better unwinding support (Randolph Chung)
 - GCC 3.4 fixes (Carlos O'Donell, Randolph Chung)

The newly introduced ->fcntl file_operation is badly thought out,
not to mention undocumented.  This patch replaces it with two better
defined operations -- check_flags and dir_notify.  Any other fcntl()s
that filesystems are interested in can have their own properly typed
f_op method when they need it.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

into ppc970.osdl.org:/home/torvalds/v2.6/linux