2018-04-20 03:31:42

by Mark Salter

[permalink] [raw]
Subject: [PATCH] ACPI / scan: Fix regression related to X-Gene UARTs

Commit e361d1f85855 ("ACPI / scan: Fix enumeration for special UART
devices") caused a regression with some X-Gene based platforms (Mustang
and M400) with invalid DSDT. The DSDT makes it appear that the UART
device is also a slave device attached to itself. With the above commit
the UART won't be enumerated by ACPI scan (slave serial devices shouldn't
be). So check for X-Gene UART device and skip slace device check on it.

Signed-off-by: Mark Salter <[email protected]>
---
drivers/acpi/scan.c | 8 ++++++++
1 file changed, 8 insertions(+)

diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
index cc234e6a6297..1dcdd0122862 100644
--- a/drivers/acpi/scan.c
+++ b/drivers/acpi/scan.c
@@ -1551,6 +1551,14 @@ static bool acpi_device_enumeration_by_parent(struct acpi_device *device)
fwnode_property_present(&device->fwnode, "baud")))
return true;

+ /*
+ * Firmware on some arm64 X-Gene platforms will make the UART
+ * device appear as both a UART and a slave of that UART. Just
+ * bail out here for X-Gene UARTs.
+ */
+ if (!strcmp(acpi_device_hid(device), "APMC0D08"))
+ return false;
+
INIT_LIST_HEAD(&resource_list);
acpi_dev_get_resources(device, &resource_list,
acpi_check_serial_bus_slave,
--
2.14.3



2018-04-22 09:35:49

by Rafael J. Wysocki

[permalink] [raw]
Subject: Re: [PATCH] ACPI / scan: Fix regression related to X-Gene UARTs

On Fri, Apr 20, 2018 at 5:29 AM, Mark Salter <[email protected]> wrote:
> Commit e361d1f85855 ("ACPI / scan: Fix enumeration for special UART
> devices") caused a regression with some X-Gene based platforms (Mustang
> and M400) with invalid DSDT.

I'm not convinced that making changes to the core ACPI device
enumeration code in order to cover up for firmware bugs is the right
approach.

> The DSDT makes it appear that the UART
> device is also a slave device attached to itself. With the above commit
> the UART won't be enumerated by ACPI scan (slave serial devices shouldn't
> be). So check for X-Gene UART device and skip slace device check on it.
>
> Signed-off-by: Mark Salter <[email protected]>
> ---
> drivers/acpi/scan.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
> index cc234e6a6297..1dcdd0122862 100644
> --- a/drivers/acpi/scan.c
> +++ b/drivers/acpi/scan.c
> @@ -1551,6 +1551,14 @@ static bool acpi_device_enumeration_by_parent(struct acpi_device *device)
> fwnode_property_present(&device->fwnode, "baud")))
> return true;
>
> + /*
> + * Firmware on some arm64 X-Gene platforms will make the UART
> + * device appear as both a UART and a slave of that UART. Just
> + * bail out here for X-Gene UARTs.
> + */
> + if (!strcmp(acpi_device_hid(device), "APMC0D08"))
> + return false;

Is the device ID never to be used outside of the broken configurations?

Even if that's the plan, how are you going to guarantee that anyway?

> +
> INIT_LIST_HEAD(&resource_list);
> acpi_dev_get_resources(device, &resource_list,
> acpi_check_serial_bus_slave,
> --

2018-04-27 18:26:26

by Mark Salter

[permalink] [raw]
Subject: Re: [PATCH] ACPI / scan: Fix regression related to X-Gene UARTs

On Sun, 2018-04-22 at 11:34 +0200, Rafael J. Wysocki wrote:
> On Fri, Apr 20, 2018 at 5:29 AM, Mark Salter <[email protected]> wrote:
> > Commit e361d1f85855 ("ACPI / scan: Fix enumeration for special UART
> > devices") caused a regression with some X-Gene based platforms (Mustang
> > and M400) with invalid DSDT.
>
> I'm not convinced that making changes to the core ACPI device
> enumeration code in order to cover up for firmware bugs is the right
> approach.

It is unfortunate but the firmware bug predates the change which uncovered
it, so previously working systems no longer work.

>
> > The DSDT makes it appear that the UART
> > device is also a slave device attached to itself. With the above commit
> > the UART won't be enumerated by ACPI scan (slave serial devices shouldn't
> > be). So check for X-Gene UART device and skip slace device check on it.
> >
> > Signed-off-by: Mark Salter <[email protected]>
> > ---
> > drivers/acpi/scan.c | 8 ++++++++
> > 1 file changed, 8 insertions(+)
> >
> > diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c
> > index cc234e6a6297..1dcdd0122862 100644
> > --- a/drivers/acpi/scan.c
> > +++ b/drivers/acpi/scan.c
> > @@ -1551,6 +1551,14 @@ static bool acpi_device_enumeration_by_parent(struct acpi_device *device)
> > fwnode_property_present(&device->fwnode, "baud")))
> > return true;
> >
> > + /*
> > + * Firmware on some arm64 X-Gene platforms will make the UART
> > + * device appear as both a UART and a slave of that UART. Just
> > + * bail out here for X-Gene UARTs.
> > + */
> > + if (!strcmp(acpi_device_hid(device), "APMC0D08"))
> > + return false;
>
> Is the device ID never to be used outside of the broken configurations?
>
> Even if that's the plan, how are you going to guarantee that anyway?

The device ID will always be used for X-Gene UARTs. Whether the DSDT is
broken or not wouldn't matter because the end result would be the same
(the UART being treated as master rather than a serial bus slave).
The broken firmware looks like this:

Device (URT0)
{
Name (_HID, "APMC0D08") // _HID: Hardware ID
...
Name (_CRS, ResourceTemplate () // _CRS: Current Resource Settings
{
Memory32Fixed (ReadWrite,
0x1C020000, // Address Base
0x00000100, // Address Length
)
UartSerialBusV2 (0x0001C200, DataBitsEight, StopBitsOne,
0x00, LittleEndian, ParityTypeNone, FlowControlNone,
0x0010, 0x0010, "URT0",
0x00, ResourceConsumer, , Exclusive,
)
Interrupt (ResourceConsumer, Level, ActiveHigh, Exclusive, ,, )
{
0x0000006D,
}
})
...
}

So "URT0" has a UartSerialBusV2 resource which references itself as the bus master.


>
> > +
> > INIT_LIST_HEAD(&resource_list);
> > acpi_dev_get_resources(device, &resource_list,
> > acpi_check_serial_bus_slave,
> > --