diff options
author | Robert Moore <robert.moore@intel.com> | 2005-05-26 00:00:00 -0400 |
---|---|---|
committer | Len Brown <len.brown@intel.com> | 2005-07-13 16:46:34 -0400 |
commit | 88ac00f5a841dcfc5c682000f4a6add0add8caac (patch) | |
tree | 80dcff8323c6c79e8706f42eb4b7c3884bc03152 /drivers/acpi/utilities | |
parent | 6f42ccf2fc50ecee8ea170040627f268430c1648 (diff) | |
download | kernel_samsung_smdk4412-88ac00f5a841dcfc5c682000f4a6add0add8caac.zip kernel_samsung_smdk4412-88ac00f5a841dcfc5c682000f4a6add0add8caac.tar.gz kernel_samsung_smdk4412-88ac00f5a841dcfc5c682000f4a6add0add8caac.tar.bz2 |
ACPICA 20050526 from Bob Moore <robert.moore@intel.com>
Implemented support to execute Type 1 and Type 2 AML
opcodes appearing at the module level (not within a control
method.) These opcodes are executed exactly once at the
time the table is loaded. This type of code was legal up
until the release of ACPI 2.0B (2002) and is now supported
within ACPI CA in order to provide backwards compatibility
with earlier BIOS implementations. This eliminates the
"Encountered executable code at module level" warning that
was previously generated upon detection of such code.
Fixed a problem in the interpreter where an AE_NOT_FOUND
exception could inadvertently be generated during the
lookup of namespace objects in the second pass parse of
ACPI tables and control methods. It appears that this
problem could occur during the resolution of forward
references to namespace objects.
Added the ACPI_MUTEX_DEBUG #ifdef to the
acpi_ut_release_mutex function, corresponding to the same
the deadlock detection debug code to be compiled out in
the normal case, improving mutex performance (and overall
subsystem performance) considerably. As suggested by
Alexey Starikovskiy.
Implemented a handful of miscellaneous fixes for possible
memory leaks on error conditions and error handling
control paths. These fixes were suggested by FreeBSD and
the Coverity Prevent source code analysis tool.
Added a check for a null RSDT pointer in
acpi_get_firmware_table (tbxfroot.c) to prevent a fault
in this error case.
Signed-off-by Len Brown <len.brown@intel.com>
Diffstat (limited to 'drivers/acpi/utilities')
-rw-r--r-- | drivers/acpi/utilities/utmisc.c | 38 |
1 files changed, 22 insertions, 16 deletions
diff --git a/drivers/acpi/utilities/utmisc.c b/drivers/acpi/utilities/utmisc.c index f6de4ed..bb65877 100644 --- a/drivers/acpi/utilities/utmisc.c +++ b/drivers/acpi/utilities/utmisc.c @@ -787,7 +787,6 @@ acpi_ut_release_mutex ( acpi_mutex_handle mutex_id) { acpi_status status; - u32 i; u32 this_thread_id; @@ -814,25 +813,32 @@ acpi_ut_release_mutex ( return (AE_NOT_ACQUIRED); } - /* - * Deadlock prevention. Check if this thread owns any mutexes of value - * greater than this one. If so, the thread has violated the mutex - * ordering rule. This indicates a coding error somewhere in - * the ACPI subsystem code. - */ - for (i = mutex_id; i < MAX_MUTEX; i++) { - if (acpi_gbl_mutex_info[i].owner_id == this_thread_id) { - if (i == mutex_id) { - continue; - } +#ifdef ACPI_MUTEX_DEBUG + { + u32 i; + /* + * Mutex debug code, for internal debugging only. + * + * Deadlock prevention. Check if this thread owns any mutexes of value + * greater than this one. If so, the thread has violated the mutex + * ordering rule. This indicates a coding error somewhere in + * the ACPI subsystem code. + */ + for (i = mutex_id; i < MAX_MUTEX; i++) { + if (acpi_gbl_mutex_info[i].owner_id == this_thread_id) { + if (i == mutex_id) { + continue; + } - ACPI_DEBUG_PRINT ((ACPI_DB_ERROR, - "Invalid release order: owns [%s], releasing [%s]\n", - acpi_ut_get_mutex_name (i), acpi_ut_get_mutex_name (mutex_id))); + ACPI_DEBUG_PRINT ((ACPI_DB_ERROR, + "Invalid release order: owns [%s], releasing [%s]\n", + acpi_ut_get_mutex_name (i), acpi_ut_get_mutex_name (mutex_id))); - return (AE_RELEASE_DEADLOCK); + return (AE_RELEASE_DEADLOCK); + } } } +#endif /* Mark unlocked FIRST */ |