The interactions between the ACPI dock driver and the ACPI-based PCI
hotplug (acpiphp) are currently problematic because of ordering
issues during hot-remove operations.
First of all, the current ACPI glue code expects that physical
devices will always be deleted before deleting the companion ACPI
device objects. Otherwise, acpi_unbind_one() will fail with a
warning message printed to the kernel log, for example:
[ 185.026073] usb usb5: Oops, 'acpi_handle' corrupt
[ 185.035150] pci 0000:1b:00.0: Oops, 'acpi_handle' corrupt
[ 185.035515] pci 0000:18:02.0: Oops, 'acpi_handle' corrupt
[ 180.013656] port1: Oops, 'acpi_handle' corrupt
This means, in particular, that struct pci_dev objects have to
be deleted before the struct acpi_device objects they are "glued"
with.
Now, the following happens the during the undocking of an ACPI-based
dock station:
1) hotplug_dock_devices() invokes registered hotplug callbacks to
destroy physical devices associated with the ACPI device objects
depending on the dock station. It calls dd->ops->handler() for
each of those device objects.
2) For PCI devices dd->ops->handler() points to
handle_hotplug_event_func() that queues up a separate work item
to execute _handle_hotplug_event_func() for the given device and
returns immediately. That work item will be executed later.
3) hotplug_dock_devices() calls dock_remove_acpi_device() for each
device depending on the dock station. This runs acpi_bus_trim()
for each of them, which causes the underlying ACPI device object
to be destroyed, but the work items queued up by
handle_hotplug_event_func() haven't been started yet.
4) _handle_hotplug_event_func() queued up in step 2) are executed
and cause the above failure to happen, because the PCI devices
they handle do not have the companion ACPI device objects any
more (those objects have been deleted in step 3).
The possible breakage doesn't end here, though, because
hotplug_dock_devices() may return before at least some of the
_handle_hotplug_event_func() work items spawned by it have a
chance to complete and then undock() will cause _DCK to be
evaluated and that will cause the devices handled by the
_handle_hotplug_event_func() to go away possibly while they are
being accessed.
This means that dd->ops->handler() for PCI devices should not point
to handle_hotplug_event_func(). Instead, it should point to a
function that will do the work of _handle_hotplug_event_func()
synchronously. For this reason, introduce such a function,
hotplug_event_func(), and modity acpiphp_dock_ops to point to
it as the handler.
Unfortunately, however, this is not sufficient, because if the dock
code were not changed further, hotplug_event_func() would now
deadlock with hotplug_dock_devices() that called it, since it would
run unregister_hotplug_dock_device() which in turn would attempt to
acquire the dock station's hp_lock mutex already acquired by
hotplug_dock_devices().
To resolve that deadlock use the observation that
unregister_hotplug_dock_device() won't need to acquire hp_lock
if PCI bridges the devices on the dock station depend on are
prevented from being removed prematurely while the first loop in
hotplug_dock_devices() is in progress.
To make that possible, introduce a mechanism by which the callers of
register_hotplug_dock_device() can provide "init" and "release"
routines that will be executed, respectively, during the addition
and removal of the physical device object associated with the
given ACPI device handle. Make acpiphp use two new functions,
acpiphp_dock_init() and acpiphp_dock_release(), that call
get_bridge() and put_bridge(), respectively, on the acpiphp bridge
holding the given device, for this purpose.
In addition to that, remove the dock station's list of
"hotplug devices" and make the dock code always walk the whole list
of "dependent devices" instead in such a way that the loops in
hotplug_dock_devices() and dock_event() (replacing the loops over
"hotplug devices") will take references to the list entries that
register_hotplug_dock_device() has been called for. That prevents
the "release" routines associated with those entries from being
called while the given entry is being processed and for PCI
devices this means that their bridges won't be removed (by a
concurrent thread) while hotplug_event_func() handling them is
being executed.
This change is based on two earlier patches from Jiang Liu.
References: https://bugzilla.kernel.org/show_bug.cgi?id=59501
Reported-and-tested-by: Alexander E. Patrakov <patrakov@gmail.com>
Tracked-down-by: Jiang Liu <jiang.liu@huawei.com>
Tested-by: Illya Klymov <xanf@xanf.me>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: Yinghai Lu <yinghai@kernel.org>
Cc: 3.9+ <stable@vger.kernel.org>
Noticed that acpi_pci_bind_root(), which has been deleted,
is left defined in acpi_driver.h. So delete this definition
from the header as well.
Signed-off-by: Toshi Kani <toshi.kani@hp.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Previously, we cached _PRT (PCI routing table, ACPI 5.0 sec 6.2.12)
contents and associated each _PRT entry with a PCI bus number. The bus
number association means dependencies on PCI device enumeration and bus
number assignment, as well as on the PCI/ACPI binding process.
After 4f535093cf ("PCI: Put pci_dev in device tree as early as possible"),
these dependencies caused the IRQ issues reported by Peter:
pci 0000:00:1e.0: PCI bridge to [bus 09] (subtractive decode)
pci 0000:00:1e.0: can't derive routing for PCI INT A
snd_ctxfi 0000:09:02.0: PCI INT A: no GSI - using ISA IRQ 5
irq 18: nobody cared (try booting with the "irqpoll" option)
This patch removes _PRT caching. Instead, we evaluate _PRT as needed
in the pci_enable_device() path. This also removes the dependency on
PCI bus numbers: we can simply look at the _PRT associated with each
bridge as we walk upstream toward the root.
[bhelgaas: changelog]
Reference: https://bugzilla.kernel.org/show_bug.cgi?id=53561
Reported-and-tested-by: Peter Hurley <peter@hurleysoftware.com>
Suggested-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
This effectively reverts 859a3f86ca ("ACPI: simplify
acpi_pci_irq_add_prt() API") and d9efae3688 ("ACPI: simplify
acpi_pci_irq_del_prt() API").
The reason is to disentangle these routines from the struct pci_bus.
We want to be able to add the _PRT before the struct pci_bus
exists, and delete the _PRT after we've removed the pci_bus.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
Recently the ACPI ops structs were constified but the inline version
of register_hotplug_dock_device() was overlooked (see also commit
9c8b04b, June 25 2011). Update the inline function
register_hotplug_dock_device() that is enabled with
CONFIG_ACPI_DOCK=n too. This patch fixes at least the following
compiler warnings:
drivers/ata/libata-acpi.c: In function .ata_acpi_associate.:
drivers/ata/libata-acpi.c:266:11: warning: passing argument 2 of .register_hotplug_dock_device. discards qualifiers from pointer target type
include/acpi/acpi_drivers.h:146:19: note: expected .struct acpi_dock_ops *. but argument is of type .const struct acpi_dock_ops *.
drivers/ata/libata-acpi.c:275:11: warning: passing argument 2 of .register_hotplug_dock_device. discards qualifiers from pointer target type
include/acpi/acpi_drivers.h:146:19: note: expected .struct acpi_dock_ops *. but argument is of type .const struct acpi_dock_ops *.
Cc: stable@vger.kernel.org
Signed-off-by: Len Brown <len.brown@intel.com>
Structs battery_file, acpi_dock_ops, file_operations,
thermal_cooling_device_ops, thermal_zone_device_ops, kernel_param_ops
are not changed in runtime. It is safe to make them const.
register_hotplug_dock_device() was altered to take const "ops" argument
to respect acpi_dock_ops' const notion.
Signed-off-by: Vasiliy Kulikov <segoon@openwall.com>
Acked-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Found by running make namespacecheck on linux-next
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Len Brown <len.brown@intel.com>
The acpi_pci_root structure contains all the individual items (acpi_device,
domain, bus number) we pass to pci_acpi_scan_root(), so just pass the
single acpi_pci_root pointer directly.
This will make it easier to add _CBA support later. For _CBA, we need the
entire downstream bus range, not just the base bus number. We have that in
the acpi_pci_root structure, so passing the pointer makes it available to
the arch-specific code.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Reviewed-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Len Brown <len.brown@intel.com>
On some old IBM workstations and desktop computers, the BIOS presents in the
DSDT an SMBus object that is missing the HID identifier that the i2c-scmi
driver looks for. Modify the ACPI device scan code to insert the missing HID
if it finds an IBM system with such an object.
Affected machines: IntelliStation Z20/Z30. Note that the i2c-i801 driver no
longer works on these machines because of ACPI resource conflicts.
Signed-off-by: Darrick J. Wong <djwong@us.ibm.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
The main benefit of using ACPI host bridge window information is that
we can do better resource allocation in systems with multiple host bridges,
e.g., http://bugzilla.kernel.org/show_bug.cgi?id=14183
Sometimes we need _CRS information even if we only have one host bridge,
e.g., https://bugs.launchpad.net/ubuntu/+source/linux/+bug/341681
Most of these systems are relatively new, so this patch turns on
"pci=use_crs" only on machines with a BIOS date of 2008 or newer.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
acpi_get_pci_dev() is better, and all callers have been converted, so
eliminate acpi_get_pci_id().
Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
There is no need to pass a segment/bus tuple to this API, as the callsite
always has a struct pci_bus. We can derive segment/bus from the
struct pci_bus, so let's take this opportunit to simplify the API and
make life easier for the callers.
Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
A PCI domain cannot change as you descend down subordinate buses, which
makes the 'segment' argument to acpi_pci_irq_add_prt() useless.
Change the interface to take a struct pci_bus *, from whence we can derive
the bus number and segment. Reducing the number of arguments makes life
simpler for callers.
Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Now that we can dynamically convert an ACPI CA handle to a
struct pci_dev at runtime, there's no need to statically bind
them during boot.
acpi_pci_bind/unbind are vastly simplified, and are only used
to evaluate _PRT methods on P2P bridges and non-bridge children.
This patch also changes the time-space tradeoff ever so slightly.
Looking up the ACPI-PCI binding is never in the performance path, and by
eliminating this caching, we save 24 bytes for each _ADR device in the
ACPI namespace.
This patch lays further groundwork to eventually eliminate
the acpi_driver_ops.bind callback.
Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Convert an ACPI CA handle to a struct pci_dev.
Performing this lookup dynamically allows us to get rid of the
ACPI-PCI binding code, which:
- eliminates struct acpi_device vs struct pci_dev lifetime issues
- lays more groundwork for eliminating .start from acpi_device_ops
and thus simplifying ACPI drivers
- whacks out a lot of code
This change lays the groundwork for eliminating much of pci_bind.c.
Although pci_root.c may not be the most logical place for this
change, putting it here saves us from having to export acpi_pci_find_root.
Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
acpi_pci_root_add() explicitly assigns device->ops.bind, and later
calls acpi_pci_bind_root(), which also does the same thing.
We don't need to repeat ourselves; removing the explicit assignment
allows us to make acpi_pci_bind() static.
Signed-off-by: Alex Chiang <achiang@hp.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
The ACPI0007 _HID used for processor "Device" objects in the namespace
is not needed outside the processor driver, so move it there. Also, the
#define is only used once, so just remove it and hard-code "ACPI0007".
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
ACPI_PROCESSOR_OBJECT_HID is a synthetic _HID that Linux generates
for "Processor" definitions. Unlike "Device" definitions, "Processor"
definitions do not have a _HID in the namespace, so we generate a
fake _HID. By convention, all these fake _HIDs begin with "LNX".
This does change the user-visible _HID for "Processor" objects --
previously, we used "ACPI_CPU" and this changes that to "LNXCPU",
which starts with "LNX" as do all the other made-up _HIDs. This
change is visible in processor filenames and "hid" files under
/sys/devices/LNXSYSTM:00/.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
This patch adds support for ACPI device driver .notify() methods. If
such a method is present, Linux/ACPI installs a handler for device
notifications (but not for system notifications such as Bus Check,
Device Check, etc). When a device notification occurs, Linux/ACPI
passes it on to the driver's .notify() method.
In most cases, this removes the need for drivers to install their own
handlers for device-specific notifications.
For fixed hardware devices like some power and sleep buttons, there's
no notification value because there's no control method to execute a
Notify opcode. When a fixed hardware device generates an event, we
handle it the same as a regular device notification, except we send
a ACPI_FIXED_HARDWARE_EVENT value. This is outside the normal 0x0-0xff
range used by Notify opcodes.
Several drivers install their own handlers for system Bus Check and
Device Check notifications so they can support hot-plug. This patch
doesn't affect that usage.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Reviewed-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
A number of things that shouldn't be exposed outside the ACPI core
were declared in include/acpi/acpi_drivers.h, where anybody can
see them. This patch moves those declarations to a new "internal.h"
inside drivers/acpi.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
/sys/module/acpi/parameters/debug_layers used to contain only the
debug layers defined by the ACPI CA. This patch adds the additional
layer definitions for ACPI drivers.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Some of the component definitions that were previous scattered around
the drivers conflict with each other. That doesn't hurt anything
except that setting one bit in the debug_layer mask would turn on
debugging in two different modules. This patch fixes the conflicts.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Move all the component definitions for drivers to a single shared place,
include/acpi/acpi_drivers.h.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Declaring processors in ACPI namespace can be done using either a
"Processor" definition or a "Device" definition (see section 8.4 -
Declaring Processors; "Advanced Configuration and Power Interface
Specification", Revision 3.0b). Currently the two processor
declaration types are conflated.
This patch disambiguates the processor declaration's definition type
enabling subsequent code to behave uniquely based explicitly on the
declaration's type.
Signed-off-by: Myron Stowe <myron.stowe@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Remove CONFIG_ACPI_EC. It was always set the same as CONFIG_ACPI,
and it had no menu label, so there was no way to set it to anything
other than "y".
Per section 6.5.4 of the ACPI 3.0b specification,
OSPM must make Embedded Controller operation regions, accessed
via the Embedded Controllers described in ECDT, available before
executing any control method.
The ECDT table is optional, but if it is present, the above text
means that the EC it describes is a required part of the ACPI
subsystem, so CONFIG_ACPI_EC=n wouldn't make sense.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Acked-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Len Brown <len.brown@intel.com>
Remove CONFIG_ACPI_POWER. It was always set the same as CONFIG_ACPI,
and it had no menu label, so there was no way to set it to anything
other than "y".
The interfaces under CONFIG_ACPI_POWER (acpi_device_sleep_wake(),
acpi_power_transition(), etc) are called unconditionally from the
ACPI core, so we already depend on it always being present.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Maybe the incorrect power state is returned on the bogus bios, which
is different with the real power state. For example: the bios returns D0
state and the real power state is D3. OS expects to set the device to D0
state. In such case if OS uses the power state returned by the BIOS and
checks the device power state very strictly in power transition, the device
can't be transited to the correct power state.
So the boot option of "acpi.power_nocheck=1" is added to avoid checking
the device power in the course of device power transition.
http://bugzilla.kernel.org/show_bug.cgi?id=8049http://bugzilla.kernel.org/show_bug.cgi?id=11000
Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Li Shaohua <shaohua.li@intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
when there is no ECDT table and no _INI object for EC device, it will be
enabled before scanning ACPI device. But it is too late after the following
the commit is merged.
>commit 7752d5cfe3
> Author: Robert Hancock <hancockr@shaw.ca>
> Date: Fri Feb 15 01:27:20 2008 -0800
>x86: validate against acpi motherboard resources
After the above commit is merged, OS will check whether MCFG area is
reserved in ACPI motherboard resources by calling the function of
acpi_get_devices when there exists MCFG table. In the acpi_get_devices the _STA
object will be evaluated to check the status of the ACPI device. On some broken
BIOS the MYEC object of EC device is initialized as one, which indicates that
EC operation region is already accessible before enabling EC device.So on these
broken BIOS the EC operation region will be accessed in course of evaluating
the _STA object before enabling EC device, which causes that OS will print the
following warning messages:
>ACPI Error (evregion-0315): No handler for Region [EC__] (ffff88007f8145e8)
[EmbeddedControl] [20080609]
>ACPI Error (exfldio-0290): Region EmbeddedControl(3) has no handler [20080321]
>ACPI Error (psparse-0530): Method parse/execution failed [\_SB_.PCI0.SBRG.
EC__.BAT1._STA] (Node ffff81013fc17a00), AE_NOT_EXIST
>ACPI Error (uteval-0233): Method execution failed [\_SB_.PCI0.SBRG.EC__.BAT1.
_STA] (Node ffff81013fc17a00), AE_NOT_EXIST
Although the above warning message is harmless, it looks confusing.
So it is necessary to enable EC device as early as possible.Maybe it is
appropriate to enable it immediately after ACPI full initialization.
http://bugzilla.kernel.org/show_bug.cgi?id=11255http://bugzilla.kernel.org/show_bug.cgi?id=11374http://bugzilla.kernel.org/show_bug.cgi?id=11660
Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Acked-by: Alexey Starikovskiy <astarikovskiy@suse.de>
Signed-off-by: Len Brown <len.brown@intel.com>
dock's uevent reported itself, not ata. It might be difficult to find an
ata device just according to a dock. This patch introduces docking ops
for each device in a dock. when docking, dock driver can send device
specific uevent. This should help dock station too (not just bay)
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Acked-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Len Brown <len.brown@intel.com>
The currect ACPI code attempts to execute _PSW at three different
places and in one of them only it tries to execute _DSW before _PSW,
which is inconsistent with the other two cases.
Move the execution of _DSW and _PSW into a separate function called
acpi_device_sleep_wake() and call it wherever appropriate instead of
executing _DSW and/or _PSW directly.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
References: https://bugzilla.novell.com/show_bug.cgi?id=302482
Signed-off-by: Thomas Renninger <trenn@novell.com>
Signed-off-by: Kay Sievers <kasievers@novell.com>
Signed-off-by: Frank Seidel <fseidel@suse.de>
Signed-off-by: Len Brown <len.brown@intel.com>
Restore the 2.6.22 CONFIG_ACPI_SLEEP build option, but now shadowing the
new CONFIG_PM_SLEEP option.
Signed-off-by: Len Brown <len.brown@intel.com>
[ Modified to work with the PM config setup changes. ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Introduce CONFIG_SUSPEND representing the ability to enter system sleep
states, such as the ACPI S3 state, and allow the user to choose SUSPEND
and HIBERNATION independently of each other.
Make HOTPLUG_CPU be selected automatically if SUSPEND or HIBERNATION has
been chosen and the kernel is intended for SMP systems.
Also, introduce CONFIG_PM_SLEEP which is automatically selected if
CONFIG_SUSPEND or CONFIG_HIBERNATION is set and use it to select the
code needed for both suspend and hibernation.
The top-level power management headers and the ACPI code related to
suspend and hibernation are modified to use the new definitions (the
changes in drivers/acpi/sleep/main.c are, mostly, moving code to reduce
the number of ifdefs).
There are many other files in which CONFIG_PM can be replaced with
CONFIG_PM_SLEEP or even with CONFIG_SUSPEND, but they can be updated in
the future.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
As it was a synonym for (CONFIG_ACPI && CONFIG_X86),
the ifdefs for it were more clutter than they were worth.
For ia64, just add a few stubs in anticipation of future
S3 or S4 support.
Signed-off-by: Len Brown <len.brown@intel.com>
Define standardized HIDs - Rename current acpi_device_id to acpica_device_id
Signed-off-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Len Brown <len.brown@intel.com>
late_initcall() is too late for acpi_sleep_init().
Call it directly from acpi_init code.
http://bugzilla.kernel.org/show_bug.cgi?id=7887
Signed-off-by: Alexey Starikovskiy <alexey.y.starikovskiy@linux.intel.com>
Signed-off-by: Vladimir Lebedev <vladimir.p.lebedev@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
drivers/acpi/bay.c: In function 'bay_add':
drivers/acpi/bay.c:310: warning: statement with no effect
Fix it by rewriting those macros in C. Much nicer.
Cc: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Len Brown <len.brown@intel.com>
Rafael J. Wysocki
Toshi Kani
Yinghai Lu
Bjorn Helgaas
Bart Van Assche
Vasiliy Kulikov
Stephen Hemminger
Bjorn Helgaas
Darrick J. Wong
Len Brown
Alexander Chiang
Myron Stowe
Zhao Yakui
Shaohua Li
Rafael J. Wysocki
Adrian Bunk
Frank Seidel
Alexey Starikovskiy
Thomas Renninger
Alexey Starikovskiy
Andrew Morton