- 25 Nov, 2013 1 commit
-
-
Lv Zheng authored
Header file <acpi/acpi.h> contains environemnt settings and architecture specific implementation that should be included before any other ACPICA headers in order to keep a consistent build environment for ACPICA users. The following internal ACPICA header files should be included from <acpi/acpi.h> and should not be included by other kernel files: <acpi/acpiosxf.h> <acpi/acpixf.h> Clean up incorrect inclusions of these files from non-ACPICA source files. [rjw: Subject and changelog] Signed-off-by:
Lv Zheng <lv.zheng@intel.com> Signed-off-by:
Rafael J. Wysocki <rafael.j.wysocki@intel.com>
-
- 21 Nov, 2013 6 commits
-
-
Bob Moore authored
Version 20131115. Signed-off-by:
Bob Moore <robert.moore@intel.com> Signed-off-by:
-
Bob Moore authored
This fix deletes any and all objects that have been attached to the root node (via acpi_attach_data). Reported by Tomasz Nowicki. ACPICA BZ 1026. Signed-off-by:
-
Tomasz Nowicki authored
This fix updates namespace node deletion to delete the entire list of attached objects (attached via acpi_attach_data) instead of just one of the attached items. ACPICA BZ 1024. Tomasz Nowicki (tomasz.nowicki@linaro.org). Signed-off-by:
Tomasz Nowicki <tomasz.nowicki@linaro.org> Signed-off-by:
-
Lv Zheng authored
The loop terminates on a NULL resource pointer, which can never happen since the loop simply increments a valid resource pointer. This fix changes the loop to terminate on an end-of-buffer condition. Problem can be seen by callers to AcpiSetCurrentResources with an invalid or corrupted resource descriptor; or a resource descriptor without an END_TAG descriptor. (refined by Bob Moore) Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
Lv Zheng authored
This patch adds protection around ACPI_CHECKSUM_ABORT so that ACPI user space test utilities can re-define it for their own purposes (currently used by ASLTS build environment). This patch doesn't affect Linux kernel behavior. Lv Zheng. Signed-off-by:
-
Bob Moore authored
The nesting level is really only useful during a single-thread execution. Therefore, only enable this output for the AcpiExec utility. Also, only emit the thread ID when executing under AcpiExec. (Context switches are still detected and a message is emitted.) ACPICA BZ 972. Signed-off-by:
-
- 31 Oct, 2013 25 commits
-
-
Lv Zheng authored
This patch adds __init to the ACPICA documented initializers: acpi_initialize_tables() acpi_initialize_subsystem() acpi_load_tables() acpi_enable_subsystem() acpi_initialize_objects() and to acpi_reallocate_root_table(), acpi_find_root_pointer() which are also meant to be called only during initialization. This patch adds __init to the ACPICA documented finalizer: acpi_terminate() as this finalizer is only called in __init function now. This change helps to reduce source code differences between ACPICA upstream and Linux. [rjw: Changelog] Signed-off-by:
-
Lv Zheng authored
Add an asmlinkage wrapper around acpi_enter_sleep_state() to prevent an empty stub from being called by assmebly code for ACPI_REDUCED_HARDWARE set. As arch/x86/kernel/acpi/wakeup_xx.S is only compiled when CONFIG_ACPI=y and there are no users of ACPI_HARDWARE_REDUCED, currently this is in fact not a real issue, but a cleanup to reduce source code differences between Linux and ACPICA upstream. [rjw: Changelog] Signed-off-by:
-
Lv Zheng authored
This patch updates header files used by acpidump to reduce the source code differences between Linux and ACPICA upstream. This patch does not affect the generation of the Linux kernel binary. Signed-off-by:
-
Lv Zheng authored
This patch updates architecture specific environment settings to reduce source differences between Linux and ACPICA upstream. This patch does not affect the generation of the Linux kernel binary. Signed-off-by:
-
Lv Zheng authored
Function acpi_ut_delete_caches() is meant to be used to implement ACPI_ALLOCATE()/ACPI_FREE() mechanisms in ACPICA, so it should call acpi_os_free() rather than ACPI_FREE(). Linux is not affected by this issue as it uses kmem_cache instead of the ACPICA local cache, but the change helps to reduce source code differences between Linux and ACPICA upstream. [rjw: Changelog] Signed-off-by:
-
Lv Zheng authored
Reversion commit 7cb7f45c (Revert "ACPICA: Remove obsolete acpi_os_validate_address interface") restored a piece of dead code related to an obselete AOPOBJ_INVALID flag. Remove that code again. This helps to reduce source code differences between Linux and ACPICA upstream. [rjw: Changelog] Signed-off-by:
-
Lv Zheng authored
This patch removes unnecessary memset() invocations where kmem_cache_zalloc() is used for allocating memory. It also helps to reduce source code differences between ACPICA upstream and Linux. [rjw: Changelog] Signed-off-by:
-
Lv Zheng authored
ACPI_FREE() should always be used to free objects allocated with ACPI_ALLOCATE_ZEROED(), so make acpi_ut_create_package_object() do that. This patch ports a fix from ACPICA upstream to Linux and helps to reduce source code differences between them. [rjw: Changelog] Signed-off-by:
-
Lv Zheng authored
The object length returned by acpi_ut_get_simple_object_size() should be rounded up to the closest word boundary. This patch ports a fix from ACPICA upstream to Linux. [rjw: Changelog] Signed-off-by:
-
Lv Zheng authored
This patch ports new counters and statistics interface, already implemented in ACPICA upstream, to Linux. That helps to reduce source code differences between Linux and ACPICA upstream. [rjw: Changelog] Signed-off-by:
-
Lv Zheng authored
This patch updates DMAR table header definitions as such enhancement has been made in ACPICA upstream already. It ports that change to the Linux source to reduce source code differences between Linux and ACPICA upstream. Build test done on x86-64 machine with the following configs enabled: CONFIG_DMAR_TABLE CONFIG_IRQ_REMAP CONFIG_INTEL_IOMMU This patch does not affect the generation of the Linux kernel binary. [rjw: Changelog] Signed-off-by:
-
Lv Zheng authored
This patch updates RSDP table header definitions as such enhancement has been made in ACPICA upstream already. It ports that change to the Linux source to reduce source code differences between Linux and ACPICA upstream. This patch does not affect the generation of the Linux kernel binary. [rjw: Changelog] Signed-off-by:
-
Lv Zheng authored
This patch merges namespace dump code updates from ACPICA upstream to reduce the source code differences between Linux and ACPICA upstream. No functional changes as currently nsdumpdv.c is not used by Linux. This patch does not affect the generation of the Linux kernel binary. Signed-off-by:
-
Lv Zheng authored
The ANOBJ_IS_EXTERNAL flag is only used by an ACPICA utilities - iASL. No functional change for the Linux kernel should results from applying this patch, but it helps to reduce source code differences between the kernel and ACPICA upstream. [rjw: Subject and changelog] Signed-off-by:
-
Lv Zheng authored
This patch adds code that is already in ACPICA upstream, so applying it can help to reduce the source code differences between Linux and ACPICA upstream. This code fragment is only useful for ACPICA utilities and no functional changes to the Linux kernel should result from it. [rjw: Changelog] Signed-off-by:
-
Bob Moore authored
Version 20130927. Signed-off-by:
-
Lv Zheng authored
The new ACPICA OSL override mechanism is used to solve these issues for the Linux OSL: 1. Linux can implement OSL using a macro. 2. Linux can implement OSL using an inlined function. 3. Linux can leave OSL not implemented for __KERNEL__ undefined code fragments. 4. Linux can add sparse declarators (__iomem) to OSL. 5. Linux can add memory tuning declarators (__init/__exit) to OSL. This patch also moves Linux specific OSL to aclinux.h which has not been maintained in the ACPICA code base. Lv Zheng. Known issue: From ACPICA's perspective, actypes.h should be included after inclusion of acenv.h. But currently in Linux, aclinux.h included by acenv.h has included actypes.h to find ACPICA types for inline functions. This is a known and existing issue and currently there is no real problem caused by this issue for Linux kernel build. Thus this issue is not covered by this cleanup commit. Signed-off-by: -
Lv Zheng authored
This change enables the host OS to redefine OSL prototypes found in the acpiosxf.h file. This allows the host OS to implement OSL interfaces with a macro or inlined function. Further, it allows the host OS to add any additional required modifiers such as __iomem, __init, __exit, etc., as necessary on a per-interface basis. Enables maximum flexibility for the OSL interfaces. Lv Zheng. Signed-off-by:
-
Bob Moore authored
Surround definition of this with a #ifndef so that the kernel can define it elsewhere if desired. Signed-off-by:
-
Lv Zheng authored
During the automatic translation of the upstream ACPICA source code into Linux kernel source code some extra white spaces are added by the "indent" program at the beginning of each line which is an invocation of a macro and there is no ";" at the end of the line. For this reason, a new mode has been added to the translation scripts to remove the extra spaces inserted before invoking such macros and add an empty line between the invocations of such macros (like the other function declarations). This new mode is executed after executing "indent" during the Linux release process. Consequently, some existing ACPICA source code in the Linux kernel tree needs to be adjusted to allow the new scripts to work correctly. The affected macros and files are: 1. ACPI_HW_DEPENDENT_RETURN (acpixf.h/acdebug.h/acevents.h): This macro is used as a wrapper for hardware dependent APIs to offer a stub when the reduced hardware is configured during compilation. 2. ACPI_EXPORT_SYMBOL (utglobal.c): This macro is used by Linux to export symbols to be found by Linux modules. All such invocations are well formatted except those exported as global variables. This can help to reduce the source code differences between Linux and upstream ACPICA, and also help to automate the release process. No functional or binary generation changes should result from it. Lv Zheng. [rjw: Changelog] Signed-off-by: -
Lv Zheng authored
The following build error: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ CC arch/x86/kernel/setup.o In file included from include/acpi/acpi.h:64:0, from include/linux/iscsi_ibft.h:24, from arch/x86/kernel/setup.c:43: include/acpi/acpixf.h:543:1: error: expected ',' or ';' before '{' token include/acpi/acpixf.h:540:1: warning: 'acpi_error' declared 'static' but never defined [-Wunused-function] make[2]: *** [arch/x86/kernel/setup.o] Error 1 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ can be triggerred by the following stub function (if implemented): ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ static inline void ACPI_INTERNAL_VAR_XFACE acpi_error(const char *module_name, u32 line_number, const char *format, ...) ACPI_PRINTF_LIKE(3) { } ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ This patch changes the position of ACPI_PRINTF_LIKE(x) to follow the style of __printf(x, x+1) used in Linux to prevent such issues from happening. Lv Zheng. Signed-off-by: -
Lv Zheng authored
This changes can reduce source code differences between Linux and ACPICA upstream to help improving the release automation. The side effect of applying this patch in Linux is: 1. Some ACPICA initialization/termination APIs are no longer exported in Linux, these include: acpi_load_tables acpi_initialize_subsystem acpi_enable_subsystem acpi_initialize_objects acpi_terminate 2. This patch does not affect the following APIs as they are currently not marked with ACPI_EXPORT_SYMBOL in Linux: acpi_reallocate_root_table acpi_initialize_tables Such functions should not be exported as they are internal to ACPI subsystem in Linux, and will only be invoked inside of ACPI subsystem's initialization routines marked with __init and termination routines marked with __exit. While on other OSPMs, such functions may still need to be exported. Thus this patch adds the configurability for ACPICA, so that it leaves OSPMs to determine if the __init/__exit marked functions should be exported or not. Lv Zheng. Signed-off-by: -
Lv Zheng authored
It is reported by kernel build test systems that all ACPICA source files in the kernel tree have incorrect label indentation. This patch changes default indent option used in the release process to fix this bug. Lv Zheng. Signed-off-by:
-
Bob Moore authored
Add a comment to clarify reason for using ACPI_FREE_BUFFER directly instead of ACPI_FREE. In addition to that, change one instance in which ACPI_FREE_BUFFER() should be used instead of ACPI_FREE(). [rjw: Subject and changelog] Signed-off-by:
-
Lv Zheng authored
For Linux, there are no functional changes/binary generation differences introduced by this patch. This change adds a new macro to all files that contain external ACPICA interfaces. It can be detected and used by the host (via the host-specific header) for any special processing required for such modules. Lv Zheng. Signed-off-by:
-
- 30 Oct, 2013 8 commits
-
-
Bob Moore authored
The ACPI spec requires the reset register width to be 8, so we now hardcode it and ignore the FADT value. This provides/maintains compatibility with other ACPI implementations that have allowed BIOS code with bad register width values to go unnoticed. Matthew Garett, Bob Moore, Lv Zheng. Signed-off-by:
-
Lv Zheng authored
This change fixes potential memory leaks in the error paths of the GPE handling code. Lv Zheng. Signed-off-by:
-
Bob Moore authored
Sort the method names in acnames.h. Signed-off-by:
-
Lv Zheng authored
In the common case, the ACPI_ALLOCATE and related macros now resolve directly to their respective acpi_os* OSL interfaces. Two options: 1) The ACPI_ALLOCATE_ZEROED macro defaults to a simple local implementation by default, unless overridden by the USE_NATIVE_ALLOCATE_ZEROED define. 2) For ACPI execution simulation environment (AcpiExec) which is not shipped with the Linux kernel, the macros can optionally be resolved to the local interfaces that track each allocation (used to immediately detect memory leaks). Signed-off-by:
-
Bob Moore authored
This change adds and deploys "safe" versions of strcpy and strcat that ensure that the target buffer does not overflow. These safe functions are only helpful for processing user input and command lines. For most ACPICA code however, the required buffer length is precisely calculated before buffer allocation, so the use of these functions is unnecessary. ACPICA BZ 1043. This change only applies to the ACPICA utilities and the debugger, none of which are not shipped with the kernel yet, so the kernel's behavior remains unchanged after it. References: https://bugs.acpica.org/show_bug.cgi?id=1043Signed-off-by:
-
Bob Moore authored
This fix repairs a version of a macro that is used for the hardware reduced case only. It adds a return statement to the macro definition so that the translation into the Linux kernel source will not completely delete the second line of the macro because it thinks that it is an empty block. It actually clarifies the use of the macro anyway. Reported-by:
-
Bob Moore authored
The original upstream ACPICA change added full history and limited line editing to the debugger: This change adds full history and limited line editing for Unix and Linux systems. It does not use readline() because of portability issues. Instead it uses the POSIX termio interface to put the terminal in raw input mode so that the various special keys can be trapped (such as up-arrow and down-arrow for history support). Since the debugger is not shipped in the kernel, it only is necessary to update one header file to keep the kernel source in sync with the upstream. [rjw: Changelog] Signed-off-by:
-
Bob Moore authored
Mostly for consistency. ACPICA BZ 1042. Actually, currently no one is experiencing problem without this check as the obj_handle is guaranteed to be valid. References: https://bugs.acpica.org/show_bug.cgi?id=1042Reported-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
