OpenCloudOS-Kernel/drivers/acpi/acpica/tbxfload.c

448 lines
14 KiB
C
Raw Normal View History

/******************************************************************************
*
* Module Name: tbxfload - Table load/unload external interfaces
*
*****************************************************************************/
/*
* Copyright (C) 2000 - 2016, Intel Corp.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions, and the following disclaimer,
* without modification.
* 2. Redistributions in binary form must reproduce at minimum a disclaimer
* substantially similar to the "NO WARRANTY" disclaimer below
* ("Disclaimer") and any redistribution must be conditioned upon
* including a substantially similar Disclaimer requirement for further
* binary redistribution.
* 3. Neither the names of the above-listed copyright holders nor the names
* of any contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation.
*
* NO WARRANTY
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* HOLDERS OR CONTRIBUTORS BE LIABLE FOR SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
* IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGES.
*/
#define EXPORT_ACPI_INTERFACES
#include <acpi/acpi.h>
#include "accommon.h"
#include "acnamesp.h"
#include "actables.h"
#include "acevents.h"
#define _COMPONENT ACPI_TABLES
ACPI_MODULE_NAME("tbxfload")
/*******************************************************************************
*
* FUNCTION: acpi_load_tables
*
* PARAMETERS: None
*
* RETURN: Status
*
* DESCRIPTION: Load the ACPI tables from the RSDT/XSDT
*
******************************************************************************/
acpi_status ACPI_INIT_FUNCTION acpi_load_tables(void)
{
acpi_status status;
ACPI_FUNCTION_TRACE(acpi_load_tables);
/*
* Install the default operation region handlers. These are the
* handlers that are defined by the ACPI specification to be
* "always accessible" -- namely, system_memory, system_IO, and
* PCI_Config. This also means that no _REG methods need to be
* run for these address spaces. We need to have these handlers
* installed before any AML code can be executed, especially any
* module-level code (11/2015).
* Note that we allow OSPMs to install their own region handlers
* between acpi_initialize_subsystem() and acpi_load_tables() to use
* their customized default region handlers.
*/
ACPICA / Interpreter: Fix a regression triggered because of wrong Linux ECDT support It is reported that the following commit triggers regressions: Linux commit: efaed9be998b5ae0afb7458e057e5f4402b43fa0 ACPICA commit: 31178590dde82368fdb0f6b0e466b6c0add96c57 Subject: ACPICA: Events: Enhance acpi_ev_execute_reg_method() to ensure no _REG evaluations can happen during OS early boot stages This is because that the ECDT support is not corrected in Linux, and Linux requires to execute _REG for ECDT (though this sounds so wrong), we need to ensure acpi_gbl_namespace_initialized is set before ECDT probing in order for _REG to be executed. Since we have to move "acpi_gbl_namespace_initialized = TRUE" to the initialization step happening before ECDT probing, acpi_load_tables() is the best candidate for now. Thus this patch fixes the regression by doing so. But if the ECDT support is fixed, Linux will not execute _REG for ECDT, and ECDT probing will happen before acpi_load_tables(). At that time, we still want to ensure acpi_gbl_namespace_initialized is set after executing acpi_ns_initialize_objects() (under the condition of acpi_gbl_group_module_level_code = FALSE), this patch also moves acpi_ns_initialize_objects() to acpi_load_tables() accordingly. Since acpi_ns_initialize_objects() doesn't seem to be skippable, this patch also removes ACPI_NO_OBJECT_INIT for the one invoked in acpi_load_tables(). And since the default region handlers should always be installed before loading the tables, this patch also removes useless acpi_gbl_group_module_level_code check accordingly. Reported by Chris Bainbridge, Fixed by Lv Zheng. Fixes: efaed9be998b (ACPICA: Events: Enhance acpi_ev_execute_reg_method() to ensure no _REG evaluations can happen during OS early boot stages) Reported-and-tested-by: Chris Bainbridge <chris.bainbridge@gmail.com> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-03-10 10:54:29 +08:00
status = acpi_ev_install_region_handlers();
if (ACPI_FAILURE(status)) {
ACPICA / Interpreter: Fix a regression triggered because of wrong Linux ECDT support It is reported that the following commit triggers regressions: Linux commit: efaed9be998b5ae0afb7458e057e5f4402b43fa0 ACPICA commit: 31178590dde82368fdb0f6b0e466b6c0add96c57 Subject: ACPICA: Events: Enhance acpi_ev_execute_reg_method() to ensure no _REG evaluations can happen during OS early boot stages This is because that the ECDT support is not corrected in Linux, and Linux requires to execute _REG for ECDT (though this sounds so wrong), we need to ensure acpi_gbl_namespace_initialized is set before ECDT probing in order for _REG to be executed. Since we have to move "acpi_gbl_namespace_initialized = TRUE" to the initialization step happening before ECDT probing, acpi_load_tables() is the best candidate for now. Thus this patch fixes the regression by doing so. But if the ECDT support is fixed, Linux will not execute _REG for ECDT, and ECDT probing will happen before acpi_load_tables(). At that time, we still want to ensure acpi_gbl_namespace_initialized is set after executing acpi_ns_initialize_objects() (under the condition of acpi_gbl_group_module_level_code = FALSE), this patch also moves acpi_ns_initialize_objects() to acpi_load_tables() accordingly. Since acpi_ns_initialize_objects() doesn't seem to be skippable, this patch also removes ACPI_NO_OBJECT_INIT for the one invoked in acpi_load_tables(). And since the default region handlers should always be installed before loading the tables, this patch also removes useless acpi_gbl_group_module_level_code check accordingly. Reported by Chris Bainbridge, Fixed by Lv Zheng. Fixes: efaed9be998b (ACPICA: Events: Enhance acpi_ev_execute_reg_method() to ensure no _REG evaluations can happen during OS early boot stages) Reported-and-tested-by: Chris Bainbridge <chris.bainbridge@gmail.com> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-03-10 10:54:29 +08:00
ACPI_EXCEPTION((AE_INFO, status,
"During Region initialization"));
return_ACPI_STATUS(status);
}
/* Load the namespace from the tables */
status = acpi_tb_load_namespace();
/* Don't let single failures abort the load */
if (status == AE_CTRL_TERMINATE) {
status = AE_OK;
}
if (ACPI_FAILURE(status)) {
ACPI_EXCEPTION((AE_INFO, status,
"While loading namespace from ACPI tables"));
}
ACPICA: Interpreter: Fix MLC issues by switching to new term_list grammar for table loading ACPICA commit 0e24fb67cde08d7df7671d7d7b183490dc79707e The MLC (Module Level Code) is an ACPICA terminology describing the AML code out of any control method, its support is an indication of the interpreter behavior during the table loading. The original implementation of MLC in ACPICA had several issues: 1. Out of any control method, besides of the object creating opcodes, only the code blocks wrapped by "If/Else/While" opcodes were supported. 2. The supported MLC code blocks were executed after loading the table rather than being executed right in place. ============================================================ The demo of this order issue is as follows: Name (OBJ1, 1) If (CND1 == 1) { Name (OBJ2, 2) } Name (OBJ3, 3) The original MLC support created OBJ2 after OBJ3's creation. ============================================================ Other than these limitations, MLC support in ACPICA looks correct. And supporting this should be easy/natural for ACPICA, but enabling of this was blocked by some ACPICA internal and OSPM specific initialization order issues we've fixed recently. The wrong support started from the following false bug fixing commit: Commit: 7f0c826a437157d2b19662977e9cf3b472cf24a6 Subject: ACPICA: Add support for module-level executable AML code Commit: 9a884ab64a4d092b4c3bf24fd9a30f7fbd4591e7 Subject: ACPICA: Add additional module-level code support ... We can confirm Windows interpreter behavior via reverse engineering means. It can be proven that not only If/Else/While wrapped code blocks, all opcodes can be executed at the module level, including operation region accesses. And it can be proven that the MLC should be executed right in place, not in such a deferred way executed after loading the table. And the above facts indeed reflect the spec words around ACPI definition block tables (DSDT/SSDT/...), the entire table and the Scope object is defined by the AML specification in BNF style as: AMLCode := def_block_header term_list def_scope := scope_op pkg_length name_string term_list The bodies of the scope opening terms (AMLCode/Scope) are all term_list, thus the table loading should be no difference than the control method evaluations as the body of the Method is also defined by the AML specification as term_list: def_method := method_op pkg_length name_string method_flags term_list The only difference is: after evaluating control method, created named objects may be freed due to no reference, while named objects created by the table loading should only be freed after unloading the table. So this patch follows the spec and the de-facto standard behavior, enables the new grammar (term_list) for the table loading. By doing so, beyond the fixes to the above issues, we can see additional differences comparing to the old grammar based table loading: 1. Originally, beyond the scope opening terms (AMLCode/Scope), If/Else/While wrapped code blocks under the scope creating terms (Device/power_resource/Processor/thermal_zone) are also supported as deferred MLC, which violates the spec defined grammar where object_list is enforced. With MLC support improved as non-deferred, the interpreter parses such scope creating terms as term_list rather object_list like the scope opening terms. After probing the Windows behavior and proving that it also parses these terms as term_list, we submitted an ECR (Engineering Change Request) to the ASWG (ACPI Specification Working Group) to clarify this. The ECR is titled as "ASL Grammar Clarification for Executable AML Opcodes" and has been accepted by the ASWG. The new grammar will appear in ACPI specification 6.2. 2. Originally, Buffer/Package/operation_region/create_XXXField/bank_field arguments are evaluated in a deferred way after loading the table. With MLC support improved, they are also parsed right in place during the table loading. This is also Windows compliant and the only difference is the removal of the debugging messages implemented before acpi_ds_execute_arguments(), see Link # [1] for the details. A previous commit should have ensured that acpi_check_address_range() won't regress. Note that enabling this feature may cause regressions due to long term Linux ACPI support on top of the wrong grammar. So this patch also prepares a global option to be used to roll back to the old grammar during the period between a regression is reported and the regression is root-cause-fixed. Lv Zheng. Link: https://bugzilla.kernel.org/show_bug.cgi?id=112911 # [1] Link: https://bugzilla.kernel.org/show_bug.cgi?id=117671 # [1] Link: https://bugzilla.kernel.org/show_bug.cgi?id=153541 # [1] Link: https://github.com/acpica/acpica/issues/122 Link: https://bugs.acpica.org/show_bug.cgi?id=963 Link: https://github.com/acpica/acpica/commit/0e24fb67 Reported-and-tested-by: Chris Bainbridge <chris.bainbridge@gmail.com> Reported-by: Ehsan <dashesy@gmail.com> Reported-and-tested-by: Dutch Guy <lucht_piloot@gmx.net> Tested-by: Mika Westerberg <mika.westerberg@linux.intel.com> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Bob Moore <robert.moore@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-09-07 14:06:54 +08:00
if (acpi_gbl_parse_table_as_term_list
|| !acpi_gbl_group_module_level_code) {
ACPICA / Interpreter: Fix a regression triggered because of wrong Linux ECDT support It is reported that the following commit triggers regressions: Linux commit: efaed9be998b5ae0afb7458e057e5f4402b43fa0 ACPICA commit: 31178590dde82368fdb0f6b0e466b6c0add96c57 Subject: ACPICA: Events: Enhance acpi_ev_execute_reg_method() to ensure no _REG evaluations can happen during OS early boot stages This is because that the ECDT support is not corrected in Linux, and Linux requires to execute _REG for ECDT (though this sounds so wrong), we need to ensure acpi_gbl_namespace_initialized is set before ECDT probing in order for _REG to be executed. Since we have to move "acpi_gbl_namespace_initialized = TRUE" to the initialization step happening before ECDT probing, acpi_load_tables() is the best candidate for now. Thus this patch fixes the regression by doing so. But if the ECDT support is fixed, Linux will not execute _REG for ECDT, and ECDT probing will happen before acpi_load_tables(). At that time, we still want to ensure acpi_gbl_namespace_initialized is set after executing acpi_ns_initialize_objects() (under the condition of acpi_gbl_group_module_level_code = FALSE), this patch also moves acpi_ns_initialize_objects() to acpi_load_tables() accordingly. Since acpi_ns_initialize_objects() doesn't seem to be skippable, this patch also removes ACPI_NO_OBJECT_INIT for the one invoked in acpi_load_tables(). And since the default region handlers should always be installed before loading the tables, this patch also removes useless acpi_gbl_group_module_level_code check accordingly. Reported by Chris Bainbridge, Fixed by Lv Zheng. Fixes: efaed9be998b (ACPICA: Events: Enhance acpi_ev_execute_reg_method() to ensure no _REG evaluations can happen during OS early boot stages) Reported-and-tested-by: Chris Bainbridge <chris.bainbridge@gmail.com> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2016-03-10 10:54:29 +08:00
/*
* Initialize the objects that remain uninitialized. This
* runs the executable AML that may be part of the
* declaration of these objects:
* operation_regions, buffer_fields, Buffers, and Packages.
*/
status = acpi_ns_initialize_objects();
if (ACPI_FAILURE(status)) {
return_ACPI_STATUS(status);
}
}
acpi_gbl_namespace_initialized = TRUE;
return_ACPI_STATUS(status);
}
ACPI_EXPORT_SYMBOL_INIT(acpi_load_tables)
/*******************************************************************************
*
* FUNCTION: acpi_tb_load_namespace
*
* PARAMETERS: None
*
* RETURN: Status
*
* DESCRIPTION: Load the namespace from the DSDT and all SSDTs/PSDTs found in
* the RSDT/XSDT.
*
******************************************************************************/
acpi_status acpi_tb_load_namespace(void)
{
acpi_status status;
u32 i;
struct acpi_table_header *new_dsdt;
struct acpi_table_desc *table;
u32 tables_loaded = 0;
u32 tables_failed = 0;
ACPI_FUNCTION_TRACE(tb_load_namespace);
(void)acpi_ut_acquire_mutex(ACPI_MTX_TABLES);
/*
* Load the namespace. The DSDT is required, but any SSDT and
* PSDT tables are optional. Verify the DSDT.
*/
table = &acpi_gbl_root_table_list.tables[acpi_gbl_dsdt_index];
if (!acpi_gbl_root_table_list.current_table_count ||
!ACPI_COMPARE_NAME(table->signature.ascii, ACPI_SIG_DSDT) ||
ACPI_FAILURE(acpi_tb_validate_table(table))) {
status = AE_NO_ACPI_TABLES;
goto unlock_and_exit;
}
/*
* Save the DSDT pointer for simple access. This is the mapped memory
* address. We must take care here because the address of the .Tables
* array can change dynamically as tables are loaded at run-time. Note:
ACPICA: Tables: Clean up split INSTALLED/VALIDATED table state logics. This patch is mainly a naming cleanup to clarify hidden logics, no functional changes. acpi_initialize_tables() is used by Linux to install table addresses for early boot steps. During this stage, table addresses are mapped by early_ioremap() mechanism which is different from the runtime IO mappings. Thus it is not safe for ACPICA to keep mapped pointers in struct acpi_table_desc structure during this stage. In order to support this in ACPICA, table states are divided into 1. "INSTALLED" (where struct acpi_table_desc.Pointer is always NULL) and 2. "VALIDATED" (where struct acpi_table_desc.Pointer is always not NULL). During acpi_initialize_tables(), table state are ensured to be "INSTALLED" but not "VALIDATED". This logic is ensured by the original code in very ambigious way. For example, currently acpi_tb_delete_table() is invoked in some place to perform an uninstallation while it is invoked in other place to perform an invalidation. They happen to work just because no one enters the penalty where the 2 behaviours are not equivalent. The naming cleanups are made in this patch: A. For installation and validation: There is code setting struct acpi_table_desc.Pointer first and delete it immediately to keep the descriptor's state as "INSTALLED" during the installation. This patch implements this in more direct way. After applying it, struct acpi_table_desc.Pointer will never be set in acpi_tb_install_table() and acpi_tb_override_table() as they are the only functions invoked during acpi_initialize_tables(). This is achieved by: 1. Rename acpi_tb_verify_table() to acpi_tb_validate_table() to clarify this change. 2. Rename acpi_tb_table_override() to acpi_tb_override_table() to keep nameing consistencies as other APIs (verb. Table). 3. Stops setting struct acpi_table_desc.Pointer in acpi_tb_install_table() and acpi_tb_table_override(). 4. Introduce acpi_tb_acquire_table() to acquire the table pointer that is not maintained in the struct acpi_table_desc of the global root table list and rewrite acpi_tb_validate_table() using this new function to reduce redundancies. 5. Replace the table pointer using the overridden table pointer in acpi_tb_add_table(). As acpi_tb_add_table() is not invoked during early boot stage, tables returned from this functions should be "VALIDATED". As acpi_tb_override_table() is modified by this patch to return a "INSTALLED" but not "VALIDATED" descriptor, to keep acpi_tb_add_table() unchanged, struct acpi_table_desc.Pointer is filled in acpi_tb_add_table(). B. For invalidation and uninstallation: The original code invalidate table by invoking acpi_tb_delete_table() here and there, but actually this function should only be used to uninstall tables. This can work just because its invocations are equivalent to invalidation in some cases. This patch splits acpi_tb_delete_table() into acpi_tb_invalidate_table() and acpi_tb_uninstall_table() and cleans up the hidden logic using the new APIs. This is achieved by: 1. Rename acpi_tb_delete_table() to acpi_tb_uninstall_table() as it is mainly called before resetting struct acpi_table_desc.Address. Thus the table descriptor is in "not INSTALLED" state. This patch enforces this by setting struct acpi_table_desc.Address to NULL in this function. 2. Introduce acpi_tb_invalidate_table() to be the reversal of acpi_tb_validate_table() and invoke it in acpi_tb_uninstall_table(). 3. Introduce acpi_tb_release_table() to release the table pointer that is not maintained in acpi_gbl_root_table_list and rewrite acpi_tb_invalidate_table() using this new function to reduce redundancies. After cleaning up, the maintainability of the internal APIs are also improved: 1. acpi_tb_acquire_table: Acquire struct acpi_table_header according to ACPI_TABLE_ORIGIN_xxx flags. 2. acpi_tb_release_table: Release struct acpi_table_header according to ACPI_TABLE_ORIGIN_xxx flags. 3. acpi_tb_install_table: Make struct acpi_table_desc.Address not NULL according to ACPI_TABLE_ORIGIN_xxx flags. 4. acpi_tb_uninstall_table: Make struct acpi_table_desc.Address NULL according to ACPI_TABLE_ORIGIN_xxx flags. 5. acpi_tb_validate_table: Make struct acpi_table_desc.Pointer not NULL according to ACPI_TABLE_ORIGIN_xxx flags. 6. acpi_tb_invalidate_table: Make struct acpi_table_desc.Pointer NULL according to ACPI_TABLE_ORIGIN_xxx flags. 7. acpi_tb_override_table: Replace struct acpi_table_desc.Address and struct acpi_table_desc.Flags. It only happens in "INSTALLED" state. The patch has been unit tested in acpi_exec by: 1. Initializing; 2. Executing exc_tbl ASLTS tests; 3. Executing "Load" command. So that all original acpi_tb_install_table() and acpi_tb_override_table() invocations are covered. Known Issues: 1. Cleanup acpi_tb_add_table() to Kill Code Redundancies Current implementation in acpi_tb_add_table() is not very clean, further patch can rewrite acpi_tb_add_table() with ordered acpi_tb_install_table(), acpi_tb_override_table() and acpi_tb_validate_table(). It is not done in this patch so that it is easy for the reviewers to understand the changes in this patch. Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Bob Moore <robert.moore@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-04-04 12:38:42 +08:00
* .Pointer field is not validated until after call to acpi_tb_validate_table.
*/
acpi_gbl_DSDT = table->pointer;
/*
* Optionally copy the entire DSDT to local memory (instead of simply
* mapping it.) There are some BIOSs that corrupt or replace the original
* DSDT, creating the need for this option. Default is FALSE, do not copy
* the DSDT.
*/
if (acpi_gbl_copy_dsdt_locally) {
new_dsdt = acpi_tb_copy_dsdt(acpi_gbl_dsdt_index);
if (new_dsdt) {
acpi_gbl_DSDT = new_dsdt;
}
}
/*
* Save the original DSDT header for detection of table corruption
* and/or replacement of the DSDT from outside the OS.
*/
memcpy(&acpi_gbl_original_dsdt_header, acpi_gbl_DSDT,
sizeof(struct acpi_table_header));
/* Load and parse tables */
(void)acpi_ut_release_mutex(ACPI_MTX_TABLES);
status = acpi_ns_load_table(acpi_gbl_dsdt_index, acpi_gbl_root_node);
(void)acpi_ut_acquire_mutex(ACPI_MTX_TABLES);
if (ACPI_FAILURE(status)) {
ACPI_EXCEPTION((AE_INFO, status, "[DSDT] table load failed"));
tables_failed++;
} else {
tables_loaded++;
}
/* Load any SSDT or PSDT tables. Note: Loop leaves tables locked */
for (i = 0; i < acpi_gbl_root_table_list.current_table_count; ++i) {
table = &acpi_gbl_root_table_list.tables[i];
ACPICA: Tables: Enable both 32-bit and 64-bit FACS ACPICA commit f7b86f35416e3d1f71c3d816ff5075ddd33ed486 The following commit is reported to have broken s2ram on some platforms: Commit: 0249ed2444d65d65fc3f3f64f398f1ad0b7e54cd ACPICA: Add option to favor 32-bit FADT addresses. The platform reports 2 FACS tables (which is not allowed by ACPI specification) and the new 32-bit address favor rule forces OSPMs to use the FACS table reported via FADT's X_FIRMWARE_CTRL field. The root cause of the reported bug might be one of the followings: 1. BIOS may favor the 64-bit firmware waking vector address when the version of the FACS is greater than 0 and Linux currently only supports resuming from the real mode, so the 64-bit firmware waking vector has never been set and might be invalid to BIOS while the commit enables higher version FACS. 2. BIOS may favor the FACS reported via the "FIRMWARE_CTRL" field in the FADT while the commit doesn't set the firmware waking vector address of the FACS reported by "FIRMWARE_CTRL", it only sets the firware waking vector address of the FACS reported by "X_FIRMWARE_CTRL". This patch excludes the cases that can trigger the bugs caused by the root cause 2. There is no handshaking mechanism can be used by OSPM to tell BIOS which FACS is currently used. Thus the FACS reported by "FIRMWARE_CTRL" may still be used by BIOS and the 0 value of the 32-bit firmware waking vector might trigger such failure. This patch tries to favor 32bit FACS address in another way where both the FACS reported by "FIRMWARE_CTRL" and the FACS reported by "X_FIRMWARE_CTRL" are loaded so that further commit can set firmware waking vector in the both tables to ensure we can exclude the cases that trigger the bugs caused by the root cause 2. The exclusion is split into 2 commits as this commit is also useful for dumping more ACPI tables, it won't get reverted when such exclusion is no longer necessary. Lv Zheng. Link: https://bugzilla.kernel.org/show_bug.cgi?id=74021 Link: https://github.com/acpica/acpica/commit/f7b86f35 Cc: 3.14.1+ <stable@vger.kernel.org> # 3.14.1+ Reported-and-tested-by: Oswald Buddenhagen <ossi@kde.org> Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Bob Moore <robert.moore@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2015-07-01 14:43:11 +08:00
if (!acpi_gbl_root_table_list.tables[i].address ||
(!ACPI_COMPARE_NAME(table->signature.ascii, ACPI_SIG_SSDT)
&& !ACPI_COMPARE_NAME(table->signature.ascii,
ACPI_SIG_PSDT)
&& !ACPI_COMPARE_NAME(table->signature.ascii,
ACPI_SIG_OSDT))
|| ACPI_FAILURE(acpi_tb_validate_table(table))) {
continue;
}
/* Ignore errors while loading tables, get as many as possible */
(void)acpi_ut_release_mutex(ACPI_MTX_TABLES);
status = acpi_ns_load_table(i, acpi_gbl_root_node);
(void)acpi_ut_acquire_mutex(ACPI_MTX_TABLES);
if (ACPI_FAILURE(status)) {
ACPI_EXCEPTION((AE_INFO, status,
"(%4.4s:%8.8s) while loading table",
table->signature.ascii,
table->pointer->oem_table_id));
tables_failed++;
ACPI_DEBUG_PRINT_RAW((ACPI_DB_INIT,
"Table [%4.4s:%8.8s] (id FF) - Table namespace load failed\n\n",
table->signature.ascii,
table->pointer->oem_table_id));
} else {
tables_loaded++;
}
}
if (!tables_failed) {
ACPI_INFO(("%u ACPI AML tables successfully acquired and loaded\n", tables_loaded));
} else {
ACPI_ERROR((AE_INFO,
"%u table load failures, %u successful",
tables_failed, tables_loaded));
/* Indicate at least one failure */
status = AE_CTRL_TERMINATE;
}
unlock_and_exit:
(void)acpi_ut_release_mutex(ACPI_MTX_TABLES);
return_ACPI_STATUS(status);
}
/*******************************************************************************
*
* FUNCTION: acpi_install_table
*
* PARAMETERS: address - Address of the ACPI table to be installed.
* physical - Whether the address is a physical table
* address or not
*
* RETURN: Status
*
* DESCRIPTION: Dynamically install an ACPI table.
* Note: This function should only be invoked after
* acpi_initialize_tables() and before acpi_load_tables().
*
******************************************************************************/
acpi_status ACPI_INIT_FUNCTION
acpi_install_table(acpi_physical_address address, u8 physical)
{
acpi_status status;
u8 flags;
u32 table_index;
ACPI_FUNCTION_TRACE(acpi_install_table);
if (physical) {
flags = ACPI_TABLE_ORIGIN_INTERNAL_PHYSICAL;
} else {
flags = ACPI_TABLE_ORIGIN_EXTERNAL_VIRTUAL;
}
status = acpi_tb_install_standard_table(address, flags,
FALSE, FALSE, &table_index);
return_ACPI_STATUS(status);
}
ACPI_EXPORT_SYMBOL_INIT(acpi_install_table)
/*******************************************************************************
*
* FUNCTION: acpi_load_table
*
* PARAMETERS: table - Pointer to a buffer containing the ACPI
* table to be loaded.
*
* RETURN: Status
*
* DESCRIPTION: Dynamically load an ACPI table from the caller's buffer. Must
* be a valid ACPI table with a valid ACPI table header.
* Note1: Mainly intended to support hotplug addition of SSDTs.
* Note2: Does not copy the incoming table. User is responsible
* to ensure that the table is not deleted or unmapped.
*
******************************************************************************/
acpi_status acpi_load_table(struct acpi_table_header *table)
{
acpi_status status;
u32 table_index;
ACPI_FUNCTION_TRACE(acpi_load_table);
/* Parameter validation */
if (!table) {
return_ACPI_STATUS(AE_BAD_PARAMETER);
}
/* Install the table and load it into the namespace */
ACPI_INFO(("Host-directed Dynamic ACPI Table Load:"));
ACPICA: Tables: Fix table checksums verification before installation. The original table handling code does not always verify checksums before installing a table, this is because code to achieve this must be implemented here and there in the redundant code blocks. There are two stages during table initialization: 1. "INSTALLED" after acpi_tb_install_table() and acpi_tb_override_table(), struct acpi_table_desc.Pointer is ensured to be NULL. This can be safely used during OSPM's early boot stage. 2. "VALIDATED" after acpi_tb_validate_table(), struct acpi_table_desc.Pointer is ensured to be not NULL. This must not be used during OSPM's early boot stage. This patch changes acpi_tb_add_table() into an early boot safe API to reduce code redundancies by changing the table state that is returned by this function from "VALIDATED" to "INSTALLED". Then the table verification code can be done in a single place. Originally, the acpi_tb_add_table() can only be used by dynamic table loadings that are executed after early boot stage, it cannot be used by static table loadings that are executed in early boot stage as: 1. The address of the table is a virtual address either maintained by OSPMs who call acpi_load_table() or by ACPICA whenever "Load" or "LoadTable" opcodes are executed, while during early boot stage, physical address of the table should be used for table loading. 2. The API will ensure the state of the loaded table to be "VALIDATED" while during early boot stage, tables maintained by root table list should be kept as "INSTALLED". To achieve this: 1. Rename acpi_tb_install_table() to acpi_tb_install_fixed_table() as it only applies to DSDT/FACS installation. Rename acpi_tb_add_table() to acpi_tb_install_non_fixed_table() as it will be applied to the installation of the rest kinds of tables. 2. Introduce acpi_tb_install_table(), acpi_tb_install_and_override_table to collect redudant code where their invocations actually have slight differences. 1. acpi_tb_install_table() is used to fill an struct acpi_table_desc where the table length is known to the caller. 2. acpi_tb_install_and_override_table() is used to perform necessary overriding before installation. 3. Change a parameter of acpi_tb_install_non_fixed_table() from struct acpi_table_desc to acpi_physical_address to allow it to be invoked by static table loadings. Also cleanup acpi_ex_load_op() and acpi_load_table() to accomodate to the parameter change. 4. Invoke acpi_tb_install_non_fixed_table() for all table loadings other than DSDT/FACS in acpi_tb_parse_root_table() to improve code maintainability (logics are collected in the single function). Also delete useless code from acpi_tb_parse_root_table(). 5. Remove all acpi_tb_validate_table() from acpi_tb_install_non_fixed_table() and acpi_tb_install_fixed_table() so that the table descriptor is kept in the state of "INSTALLED" but not "VALIDATED" after returning from these functions. 6. Introduce temporary struct acpi_table_desc (new_table_desc/old_table_desc) into the functions to indicate a table descriptor that is not maintained by acpi_gbl_root_table_list. Introduce acpi_tb_acquire_temporal_table() and acpi_tb_release_temporal_table() to handle the use cases of such temporal tables. They are only used for verified installation. 7. Introduce acpi_tb_verify_table() to validate table and verify table checksum, also remove table checksum verification from acpi_tb_validate_table(). Invoke acpi_tb_validate_table() in the functions that will convert a table into "LOADED" state or invoke it from acpi_get_table_XXX() APIs. Invoke acpi_tb_verify_table() on temporary struct acpi_table_desc(s) that are going to be "INSTALLED". 8. Change acpi_tb_override_table() logic so that a temporary struct acpi_table_desc will be overridden before installtion, this makes code simpler. After applying the patch, tables are always installed after being overridden and the table checksums are always verified before installation. Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Bob Moore <robert.moore@intel.com> [rjw: Subject] Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
2014-04-04 12:38:57 +08:00
status =
acpi_tb_install_and_load_table(table, ACPI_PTR_TO_PHYSADDR(table),
ACPI_TABLE_ORIGIN_EXTERNAL_VIRTUAL,
FALSE, &table_index);
return_ACPI_STATUS(status);
}
ACPI_EXPORT_SYMBOL(acpi_load_table)
/*******************************************************************************
*
* FUNCTION: acpi_unload_parent_table
*
* PARAMETERS: object - Handle to any namespace object owned by
* the table to be unloaded
*
* RETURN: Status
*
* DESCRIPTION: Via any namespace object within an SSDT or OEMx table, unloads
* the table and deletes all namespace objects associated with
* that table. Unloading of the DSDT is not allowed.
* Note: Mainly intended to support hotplug removal of SSDTs.
*
******************************************************************************/
acpi_status acpi_unload_parent_table(acpi_handle object)
{
struct acpi_namespace_node *node =
ACPI_CAST_PTR(struct acpi_namespace_node, object);
acpi_status status = AE_NOT_EXIST;
acpi_owner_id owner_id;
u32 i;
ACPI_FUNCTION_TRACE(acpi_unload_parent_table);
/* Parameter validation */
if (!object) {
return_ACPI_STATUS(AE_BAD_PARAMETER);
}
/*
* The node owner_id is currently the same as the parent table ID.
* However, this could change in the future.
*/
owner_id = node->owner_id;
if (!owner_id) {
/* owner_id==0 means DSDT is the owner. DSDT cannot be unloaded */
return_ACPI_STATUS(AE_TYPE);
}
/* Must acquire the table lock during this operation */
status = acpi_ut_acquire_mutex(ACPI_MTX_TABLES);
if (ACPI_FAILURE(status)) {
return_ACPI_STATUS(status);
}
/* Find the table in the global table list */
for (i = 0; i < acpi_gbl_root_table_list.current_table_count; i++) {
if (owner_id != acpi_gbl_root_table_list.tables[i].owner_id) {
continue;
}
/*
* Allow unload of SSDT and OEMx tables only. Do not allow unload
* of the DSDT. No other types of tables should get here, since
* only these types can contain AML and thus are the only types
* that can create namespace objects.
*/
if (ACPI_COMPARE_NAME
(acpi_gbl_root_table_list.tables[i].signature.ascii,
ACPI_SIG_DSDT)) {
status = AE_TYPE;
break;
}
/* Ensure the table is actually loaded */
(void)acpi_ut_release_mutex(ACPI_MTX_TABLES);
if (!acpi_tb_is_table_loaded(i)) {
status = AE_NOT_EXIST;
(void)acpi_ut_acquire_mutex(ACPI_MTX_TABLES);
break;
}
/* Invoke table handler if present */
if (acpi_gbl_table_handler) {
(void)acpi_gbl_table_handler(ACPI_TABLE_EVENT_UNLOAD,
acpi_gbl_root_table_list.
tables[i].pointer,
acpi_gbl_table_handler_context);
}
/*
* Delete all namespace objects owned by this table. Note that
* these objects can appear anywhere in the namespace by virtue
* of the AML "Scope" operator. Thus, we need to track ownership
* by an ID, not simply a position within the hierarchy.
*/
status = acpi_tb_delete_namespace_by_owner(i);
if (ACPI_FAILURE(status)) {
break;
}
status = acpi_tb_release_owner_id(i);
acpi_tb_set_table_loaded_flag(i, FALSE);
(void)acpi_ut_acquire_mutex(ACPI_MTX_TABLES);
break;
}
(void)acpi_ut_release_mutex(ACPI_MTX_TABLES);
return_ACPI_STATUS(status);
}
ACPI_EXPORT_SYMBOL(acpi_unload_parent_table)