Commit Graph

8 Commits

Author SHA1 Message Date
Bob Moore 50eca3eb89 [ACPI] ACPICA 20050930
Completed a major overhaul of the Resource Manager code -
specifically, optimizations in the area of the AML/internal
resource conversion code. The code has been optimized to
simplify and eliminate duplicated code, CPU stack use has
been decreased by optimizing function parameters and local
variables, and naming conventions across the manager have
been standardized for clarity and ease of maintenance (this
includes function, parameter, variable, and struct/typedef
names.)

All Resource Manager dispatch and information tables have
been moved to a single location for clarity and ease of
maintenance. One new file was created, named "rsinfo.c".

The ACPI return macros (return_ACPI_STATUS, etc.) have
been modified to guarantee that the argument is
not evaluated twice, making them less prone to macro
side-effects. However, since there exists the possibility
of additional stack use if a particular compiler cannot
optimize them (such as in the debug generation case),
the original macros are optionally available.  Note that
some invocations of the return_VALUE macro may now cause
size mismatch warnings; the return_UINT8 and return_UINT32
macros are provided to eliminate these. (From Randy Dunlap)

Implemented a new mechanism to enable debug tracing for
individual control methods. A new external interface,
acpi_debug_trace(), is provided to enable this mechanism. The
intent is to allow the host OS to easily enable and disable
tracing for problematic control methods. This interface
can be easily exposed to a user or debugger interface if
desired. See the file psxface.c for details.

acpi_ut_callocate() will now return a valid pointer if a
length of zero is specified - a length of one is used
and a warning is issued. This matches the behavior of
acpi_ut_allocate().

Signed-off-by: Bob Moore <robert.moore@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
2005-12-10 00:20:25 -05:00
Robert Moore bda663d36b [ACPI] ACPICA 20050916
Fixed a problem within the Resource Manager where
support for the Generic Register descriptor was not fully
implemented.  This descriptor is now fully recognized,
parsed, disassembled, and displayed.

Restructured the Resource Manager code to utilize
table-driven dispatch and lookup, eliminating many of the
large switch() statements.  This reduces overall subsystem
code size and code complexity.  Affects the resource parsing
and construction, disassembly, and debug dump output.

Cleaned up and restructured the debug dump output for all
resource descriptors.  Improved readability of the output
and reduced code size.

Fixed a problem where changes to internal data structures
caused the optional ACPI_MUTEX_DEBUG code to fail
compilation if specified.

Signed-off-by: Robert Moore <Robert.Moore@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
2005-09-21 23:51:39 -04:00
Robert Moore aff8c2777d [ACPI] ACPICA 20050902
Fixed a problem with the internal Owner ID allocation and
deallocation mechanisms for control method execution and
recursive method invocation.  This should eliminate the
OWNER_ID_LIMIT exceptions and "Invalid OwnerId" messages
seen on some systems.  Recursive method invocation depth
is currently limited to 255.  (Alexey Starikovskiy)

http://bugzilla.kernel.org/show_bug.cgi?id=4892

Completely eliminated all vestiges of support for the
"module-level executable code" until this support is
fully implemented and debugged.  This should eliminate the
NO_RETURN_VALUE exceptions seen during table load on some
systems that invoke this support.

http://bugzilla.kernel.org/show_bug.cgi?id=5162

Fixed a problem within the resource manager code where
the transaction flags for a 64-bit address descriptor were
handled incorrectly in the type-specific flag byte.

Consolidated duplicate code within the address descriptor
resource manager code, reducing overall subsystem code size.

Signed-off-by: Robert Moore <Robert.Moore@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
2005-09-03 00:15:11 -04:00
Len Brown 4be44fcd3b [ACPI] Lindent all ACPI files
Signed-off-by: Len Brown <len.brown@intel.com>
2005-08-05 00:45:14 -04:00
Robert Moore f9f4601f33 ACPICA 20050708 from Bob Moore <robert.moore@intel.com>
The use of the CPU stack in the debug version of the
subsystem has been considerably reduced.  Previously, a
debug structure was declared in every function that used
the debug macros.  This structure has been removed in
favor of declaring the individual elements as parameters
to the debug functions.  This reduces the cumulative stack
use during nested execution of ACPI function calls at the
cost of a small increase in the code size of the debug
version of the subsystem.  With assistance from Alexey
Starikovskiy and Len Brown.

Added the ACPI_GET_FUNCTION_NAME macro to enable the
compiler-dependent headers to define a macro that will
return the current function name at runtime (such as
__FUNCTION__ or _func_, etc.) The function name is used
by the debug trace output.  If ACPI_GET_FUNCTION_NAME
is not defined in the compiler-dependent header, the
function name is saved on the CPU stack (one pointer per
function.) This mechanism is used because apparently there
exists no standard ANSI-C defined macro that that returns
the function name.

Alexey Starikovskiy redesigned and reimplemented the
"Owner ID" mechanism used to track namespace objects
created/deleted by ACPI tables and control method
execution.  A bitmap is now used to allocate and free the
IDs, thus solving the wraparound problem present in the
previous implementation.  The size of the namespace node
descriptor was reduced by 2 bytes as a result.

Removed the UINT32_BIT and UINT16_BIT types that were used
for the bitfield flag definitions within the headers for
the predefined ACPI tables.  These have been replaced by
UINT8_BIT in order to increase the code portability of
the subsystem.  If the use of UINT8 remains a problem,
we may be forced to eliminate bitfields entirely because
of a lack of portability.

Alexey Starikovksiy enhanced the performance of
acpi_ut_update_object_reference.  This is a frequently used
function and this improvement increases the performance
of the entire subsystem.

Alexey Starikovskiy fixed several possible memory leaks
and the inverse - premature object deletion.

Signed-off-by: Len Brown <len.brown@intel.com>
2005-07-14 00:42:23 -04:00
Robert Moore 73459f73e5 ACPICA 20050617-0624 from Bob Moore <robert.moore@intel.com>
ACPICA 20050617:

Moved the object cache operations into the OS interface
layer (OSL) to allow the host OS to handle these operations
if desired (for example, the Linux OSL will invoke the
slab allocator).  This support is optional; the compile
time define ACPI_USE_LOCAL_CACHE may be used to utilize
the original cache code in the ACPI CA core.  The new OSL
interfaces are shown below.  See utalloc.c for an example
implementation, and acpiosxf.h for the exact interface
definitions.  Thanks to Alexey Starikovskiy.
	acpi_os_create_cache
	acpi_os_delete_cache
	acpi_os_purge_cache
	acpi_os_acquire_object
	acpi_os_release_object

Modified the interfaces to acpi_os_acquire_lock and
acpi_os_release_lock to return and restore a flags
parameter.  This fits better with many OS lock models.
Note: the current execution state (interrupt handler
or not) is no longer passed to these interfaces.  If
necessary, the OSL must determine this state by itself, a
simple and fast operation.  Thanks to Alexey Starikovskiy.

Fixed a problem in the ACPI table handling where a valid
XSDT was assumed present if the revision of the RSDP
was 2 or greater.  According to the ACPI specification,
the XSDT is optional in all cases, and the table manager
therefore now checks for both an RSDP >=2 and a valid
XSDT pointer.  Otherwise, the RSDT pointer is used.
Some ACPI 2.0 compliant BIOSs contain only the RSDT.

Fixed an interpreter problem with the Mid() operator in the
case of an input string where the resulting output string
is of zero length.  It now correctly returns a valid,
null terminated string object instead of a string object
with a null pointer.

Fixed a problem with the control method argument handling
to allow a store to an Arg object that already contains an
object of type Device.  The Device object is now correctly
overwritten.  Previously, an error was returned.

ACPICA 20050624:

Modified the new OSL cache interfaces to use ACPI_CACHE_T
as the type for the host-defined cache object.  This allows
the OSL implementation to define and type this object in
any manner desired, simplifying the OSL implementation.
For example, ACPI_CACHE_T is defined as kmem_cache_t for
Linux, and should be defined in the OS-specific header
file for other operating systems as required.

Changed the interface to AcpiOsAcquireObject to directly
return the requested object as the function return (instead
of ACPI_STATUS.) This change was made for performance
reasons, since this is the purpose of the interface in the
first place.  acpi_os_acquire_object is now similar to the
acpi_os_allocate interface.  Thanks to Alexey Starikovskiy.

Modified the initialization sequence in
acpi_initialize_subsystem to call the OSL interface
acpi_osl_initialize first, before any local initialization.
This change was required because the global initialization
now calls OSL interfaces.

Restructured the code base to split some files because
of size and/or because the code logically belonged in a
separate file.  New files are listed below.

  utilities/utcache.c	/* Local cache interfaces */
  utilities/utmutex.c	/* Local mutex support */
  utilities/utstate.c	/* State object support */
  parser/psloop.c	/* Main AML parse loop */

Signed-off-by: Len Brown <len.brown@intel.com>
2005-07-13 23:45:36 -04:00
Robert Moore 44f6c01242 ACPICA 20050408 from Bob Moore
Fixed three cases in the interpreter where an "index"
argument to an ASL function was still (internally) 32
bits instead of the required 64 bits.  This was the Index
argument to the Index, Mid, and Match operators.

The "strupr" function is now permanently local
(acpi_ut_strupr), since this is not a POSIX-defined
function and not present in most kernel-level C
libraries. References to the C library strupr function
have been removed from the headers.

Completed the deployment of static
functions/prototypes. All prototypes with the static
attribute have been moved from the headers to the owning
C file.

ACPICA 20050329 from Bob Moore

An error is now generated if an attempt is made to create
a Buffer Field of length zero (A CreateField with a length
operand of zero.)

The interpreter now issues a warning whenever executable
code at the module level is detected during ACPI table
load. This will give some idea of the prevalence of this
type of code.

Implemented support for references to named objects (other
than control methods) within package objects.

Enhanced package object output for the debug
object. Package objects are now completely dumped, showing
all elements.

Enhanced miscellaneous object output for the debug
object. Any object can now be written to the debug object
(for example, a device object can be written, and the type
of the object will be displayed.)

The "static" qualifier has been added to all local
functions across the core subsystem.

The number of "long" lines (> 80 chars) within the source
has been significantly reduced, by about 1/3.

Cleaned up all header files to ensure that all CA/iASL
functions are prototyped (even static functions) and the
formatting is consistent.

Two new header files have been added, acopcode.h and
acnames.h.

Removed several obsolete functions that were no longer
used.

Signed-off-by: Len Brown <len.brown@intel.com>
2005-07-12 00:08:52 -04:00
Linus Torvalds 1da177e4c3 Linux-2.6.12-rc2
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!
2005-04-16 15:20:36 -07:00