The GPE block initialization code in acpi_ev_initialize_gpe_block()
uses acpi_set_gpe() to make sure that the GPEs with nonzero
runtime counter will remain enabled, but since it already has
a struct acpi_gpe_event_info object for each GPE, it might use
the low-level GPE enabling function, acpi_clear_and_enable_gpe(),
for this purpose.
Rename acpi_clear_and_enable_gpe() to acpi_ev_enable_gpe(),
modify the two existing users of it accordingly and modify
acpi_ev_initialize_gpe_block() to use it instead of acpi_set_gpe()
and to check its return value.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
he commit message for your changes. Lines starting
After the previous patch that introduced acpi_gpe_wakeup() and
modified the ACPI suspend and wakeup code to use it, the third
argument of acpi_{enable|disable}_gpe() and the GPE wakeup
reference counter are not necessary any more. Remove them and
modify all of the users of acpi_{enable|disable}_gpe()
accordingly. Also drop GPE type constants that aren't used
any more. Rafael J. Wysocki.
ACPICA uses reference counters to avoid disabling GPEs too early in
case they have been enabled for many times. This is done separately
for run time and for wakeup, but the wakeup GPE reference counter is
not really necessary, because GPEs are only enabled to wake up the
system at the hardware level by acpi_enter_sleep_state(). Thus it
only is necessary to set the corresponding bits in the wakeup enable
masks of these GPEs' registers right before the system enters a sleep
state. Moreover, the GPE wakeup enable bits can only be set when the
target sleep state of the system is known and they need to be cleared
immediately after wakeup regardless of how many wakeup devices are
associated with a given GPE.
On the basis of the above observations, introduce function
acpi_gpe_wakeup() to be used for setting or clearing the enable bit
corresponding to a given GPE in its enable register's enable_for_wake
mask. Rafael J. Wysocki
ACPICA uses acpi_hw_write_gpe_enable_reg() to re-enable a GPE after
an event signaled by it has been handled. However, this function
writes the entire GPE enable mask to the GPE's enable register which
may not be correct. Namely, if one of the other GPEs in the same
register was previously enabled by acpi_enable_gpe() and subsequently
disabled using acpi_set_gpe(), acpi_hw_write_gpe_enable_reg() will
re-enable it along with the target GPE.
To fix this issue rework acpi_hw_write_gpe_enable_reg() so that it
calls acpi_hw_low_set_gpe() with a special action value,
ACPI_GPE_CONDITIONAL_ENABLE, that will make it only enable the GPE if the
corresponding bit in its register's enable_for_run mask is set.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Patch : https://patchwork.kernel.org/patch/104909/
ACPICA uses acpi_ev_enable_gpe() for enabling GPEs at the low level,
which is incorrect, because this function only enables the GPE if the
corresponding bit in its enable register's enable_for_run mask is set.
This causes acpi_set_gpe() to work incorrectly if used for enabling
GPEs that were not previously enabled with acpi_enable_gpe(). As a
result, among other things, wakeup-only GPEs are never enabled by
acpi_enable_wakeup_device(), so the devices that use them are unable
to wake up the system.
To fix this issue remove acpi_ev_enable_gpe() and its counterpart
acpi_ev_disable_gpe() and replace acpi_hw_low_disable_gpe() with
acpi_hw_low_set_gpe() that will be used instead to manipulate GPE
enable bits at the low level. Make the users of acpi_ev_enable_gpe()
and acpi_ev_disable_gpe() call acpi_hw_low_set_gpe() instead and
make sure that GPE enable masks are only updated by acpi_enable_gpe()
and acpi_disable_gpe() when GPE reference counters change from 0
to 1 and from 1 to 0, respectively.
Rafael Wysocki. Corresponds to:
Patch : https://patchwork.kernel.org/patch/104912/
In quite a few places ACPICA needs to compute a GPE enable mask with
only one bit set, corresponding to a given GPE. Currently, that
computation is always open coded which leads to unnecessary code
duplication. Fix this by introducing a helper function for computing
one-bit GPE enable masks and using it where appropriate.
Rafael Wysocki. corresponds to:
Patch : https://patchwork.kernel.org/patch/104903/
Adds full support for comments within input source files.
Fixed a couple error messages one for unknown table signature,
another for unrecognized source file (not asl or data table source.)
Also, reduced verbosity of the standard templates, but added
option (-vt) to make the output look exactly like the disassembler.
These tables are a bit different from the rest of the tables.
DSDT/SSDT are AML tables, not data tables. FACS/RSDP do not use
the standard ACPI table header.
Previously, the maximum data size passed to the EC operation region
handler was a single byte. Since there are often larger datums that
need to be transfered, and the EC would like to lock these as a single
transaction, this change enables transfers larger than a single byte.
This may require changes to the host-OS Embedded Controller driver
to enable 16/32/64 bit tranfers in addition to 8-bit transfers.
Alexey Starikovskiy, Lin Ming
Remove obsolete AOPOBJ_SINGLE_DATUM. Add AOPOBJ_INVALID for use if
the host OS rejects the address of an operation region (currently
only used by Linux.)
This change enhances the performance of namespace searches and walks
by adding a backpointer to the parent in each namespace node. On large
namespaces, this change can improve overall ACPI performance by up to 9X.
Adding a pointer to each namespace node increases the overall size of
the internal namespace by about 5%, since each namespace entry usually
consists of both a namespace node and an ACPI operand object.
Alexey Starikovskiy.
This feature is optional and is enabled if the BIOS requests any Windows
OSI strings. It can also be enabled by the host OS. Matthew Garrett,
Bob Moore.
On control method exit, only walk the namespace if the method is known
to have created namespace objects outside of its local scope.
Bob Moore, Alexey Starikovskiy.
-dt option will disable the tracking mechanism, which improves performance
considerably. Also restructured the options into -d (disable) and -e (enable) options.
These objects are defined by "Windows Instrumentation", and are not
part of the ACPI spec. Adds compiler support and runtime typechecking
support in the ACPICA core. ACPICA BZ 860.
This change implements the rules for LEN/MIF/MAF/MIN/MAX/GRA that come
from the ACPI specification (Table 6-40). Also added a special
case for IO descriptor where MIN/MAX are both start addresses, not
a start and end address. ACPICA BZ 840.
For all GPEs, including FADT-based and GPE Block Devices, execute
any _PRW methods in the new table, and process any _Lxx/_Exx GPE
methods in the new table. Any runtime GPE that is referred to by
an _Lxx/_Exx method in the new table is immediately enabled.
ACPICA BZ 833. Lin Ming, Bob Moore.
Adds a reference count mechanism to simplify support of shared GPEs that
require multiple device drivers. Several external interfaces have changed.
One external interface has been removed.
Changed:
AcpiEnableGpe
AcpiDisableGpe
AcpiClearGpe
AcpiGetGpeStatus
Removed:
AcpiSetGpeType
New:
AcpiSetGpe
See the ACPICA Programmer Reference for details.
Matthew Garrett
Bob Moore
Rafael Wysocki
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.
This change adds support to detect a DSDT that has been corrupted
and/or replaced from outside the OS (by firmware). This is typically
catastrophic for the system, but has been seen on some machines.
Lin Ming, Bob Moore.
This change will enable debug object output via a global variable,
AcpiGbl_EnableAmlDebugObject. This will help with remote machine
debugging. Also, moved all debug object support code to a new file,
executer/exdebug.c. Entire debug object module can now be configured
out of the ACPICA build if desired. Lin Ming, Bob Moore.
