> > ACPI 5.0 specification introduces mechanisms of enumerating the
> > peripherals connected on the serial buses. It will look like:
> > Device (BTH0)
> > {
> > Name (_HID, "IBMF001") // _HID: Hardware ID
> > Name (_CID, "PNPF001") // _CID: Compatible ID
> > Method (_CRS, 0, NotSerialized) // _CRS: Current Resource Settings
> > {
> > Name (UBUF, ResourceTemplate ()
> > {
> > UartSerialBus (0x0001C200, DataBitsEight, StopBitsOne,
> > 0xC0, LittleEndian, ParityTypeNone,
> FlowControlHardware,
> > 0x0020, 0x0020, "\\_SB.PCI0.UA00",
> > 0x00, ResourceConsumer, ,
> > )
> > }
> > Return (UBUF)
> > }
> > }
> > This patch follows the specification, implementing such UART
> > enumeration machanism by exporting current peripheral HID/CIDs for the
> UART devices.
> > A userspace udev rule may look like:
> > SUBSYSTEM=tty, ATTR{peripheral_type}=="*:IBMF001:*",
> > RUN+="/bin/userprog --devpath=%p"
> > Thanks for the internal composers:
> > Mika Westerberg <mika.westerberg@xxxxxxxxx>
> > Huang Ying <ying.huang@xxxxxxxxx>
> > Zhang Rui <rui.zhang@xxxxxxxxx>
> > Thanks for the internal reviewers:
> > Alan Cox <alan@xxxxxxxxxxxxxxx>
> > Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
> > Heikki Krogerus <heikki.krogerus@xxxxxxxxx>
> > Andriy Shevchenko <andriy.shevchenko@xxxxxxxxx>
> > Fengguang Wu <fengguang.wu@xxxxxxxxx>
>
> Why are these names not on the list below?
I obtained help from them on old revisions.
Mika is the key person for the acpi serial enumeration.
He has implemented the acpi i2c/spi serial enumeration.
> > Signed-off-by: Lv Zheng <lv.zheng@xxxxxxxxx>
> > Signed-off-by: Zhang Rui <rui.zhang@xxxxxxxxx>
> > Acked-by: Huang Ying <ying.huang@xxxxxxxxx>
> > Reviewed-by: Mika Westerberg <mika.westerberg@xxxxxxxxxxxxxxx>
>
>
> > ---
> > drivers/acpi/Kconfig | 6 ++
> > drivers/acpi/Makefile | 1 +
> > drivers/acpi/acpi_uart.c | 135
> ++++++++++++++++++++++++++++++++++++++
> > drivers/acpi/internal.h | 2 +
> > drivers/acpi/scan.c | 8 +--
> > drivers/tty/serial/serial_core.c | 15 +++++
> > include/linux/tty.h | 11 ++++
> > 7 files changed, 174 insertions(+), 4 deletions(-) create mode
> > 100644 drivers/acpi/acpi_uart.c
> >
> > diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig index
> > 92ed969..42233f6 100644
> > --- a/drivers/acpi/Kconfig
> > +++ b/drivers/acpi/Kconfig
> > @@ -186,6 +186,12 @@ config ACPI_I2C
> > help
> > ACPI I2C enumeration support.
> >
> > +config ACPI_UART
> > + def_tristate TTY
> > + depends on TTY
> > + help
> > + ACPI UART enumeration support.
>
> Please provide more information here.
OK.
> As this is primarily an ACPI patch, I need an ack from a ACPI maintainer before I
> can take this.
OK, I'll ask for this.
> > +
> > config ACPI_PROCESSOR
> > tristate "Processor"
> > select THERMAL
> > diff --git a/drivers/acpi/Makefile b/drivers/acpi/Makefile index
> > 474fcfe..4db5470 100644
> > --- a/drivers/acpi/Makefile
> > +++ b/drivers/acpi/Makefile
> > @@ -72,6 +72,7 @@ obj-$(CONFIG_ACPI_EC_DEBUGFS) += ec_sys.o
> > obj-$(CONFIG_ACPI_CUSTOM_METHOD)+= custom_method.o
> > obj-$(CONFIG_ACPI_BGRT) += bgrt.o
> > obj-$(CONFIG_ACPI_I2C) += acpi_i2c.o
> > +obj-$(CONFIG_ACPI_UART) += acpi_uart.o
> >
> > # processor has its own "processor." module_param namespace
> > processor-y := processor_driver.o processor_throttling.o
> > diff --git a/drivers/acpi/acpi_uart.c b/drivers/acpi/acpi_uart.c new
> > file mode 100644 index 0000000..dd5ac2e
> > --- /dev/null
> > +++ b/drivers/acpi/acpi_uart.c
> > @@ -0,0 +1,135 @@
> > +/*
> > + * acpi_uart.c - ACPI UART enumeration support
> > + *
> > + * Copyright (c) 2013, Intel Corporation
> > + * Author: Lv Zheng <lv.zheng@xxxxxxxxx>
> > + *
> > + * This program is free software; you can redistribute it and/or
> > +modify it
> > + * under the terms of the GNU General Public License as published by
> > +the
> > + * Free Software Foundation; version 2 of the License.
> > + */
> > +
> > +#include <linux/init.h>
> > +#include <linux/tty.h>
> > +#include <linux/module.h>
> > +#include <linux/acpi.h>
> > +
> > +#include "internal.h"
> > +
> > +struct acpi_uart_buf {
> > + char *buf;
> > + size_t size;
> > + int *len;
> > +};
> > +
> > +struct acpi_uart_enum {
> > + void *context;
> > + acpi_status status;
> > +
> > + struct acpi_device *adev;
> > +};
> > +
> > +static acpi_status
> > +acpi_uart_enum_pnpids(struct acpi_device *adev,
> > + struct acpi_resource_uart_serialbus *sb,
> > + void *context);
> > +
> > +static int acpi_uart_enum_resources(struct acpi_resource *ares,
> > + void *context)
> > +{
> > + struct acpi_uart_enum *ctx = context;
> > + struct acpi_resource_uart_serialbus *sb;
> > +
> > + if (ares->type != ACPI_RESOURCE_TYPE_SERIAL_BUS)
> > + return 1;
> > + sb = &ares->data.uart_serial_bus;
> > + if (sb->type != ACPI_RESOURCE_SERIAL_TYPE_UART)
> > + return 1;
> > +
> > + ctx->status = acpi_uart_enum_pnpids(ctx->adev, sb, ctx->context);
> > + return 1;
> > +}
> > +
> > +static acpi_status acpi_uart_enum_devices(acpi_handle handle, u32 level,
> > + void *context,
> > + void **return_value)
> > +{
> > + struct acpi_uart_enum *ctx = context;
> > + struct acpi_device *adev;
> > + struct list_head resource_list;
> > +
> > + if (acpi_bus_get_device(handle, &adev))
> > + return AE_OK;
> > + if (acpi_bus_get_status(adev) || !adev->status.present)
> > + return AE_OK;
> > +
> > + /* Enumerate resources. */
> > + ctx->adev = adev;
> > + ctx->status = AE_OK;
> > + INIT_LIST_HEAD(&resource_list);
> > + acpi_dev_get_resources(adev, &resource_list,
> > + acpi_uart_enum_resources, ctx);
> > + acpi_dev_free_resource_list(&resource_list);
> > +
> > + return ctx->status;
> > +}
> > +
> > +static int acpi_uart_walk_port(struct device *parent,
> > + void *context)
> > +{
> > + acpi_handle handle;
> > + acpi_status status;
> > + struct acpi_uart_enum ctx;
> > +
> > + handle = parent ? ACPI_HANDLE(parent) : NULL;
> > + if (!handle)
> > + return -ENODEV;
> > +
> > + ctx.context = context;
> > +
> > + status = acpi_walk_namespace(ACPI_TYPE_DEVICE, handle, 1,
> > + acpi_uart_enum_devices, NULL,
> > + &ctx, NULL);
> > + if (ACPI_FAILURE(status))
> > + return -EINVAL;
> > +
> > + return 0;
> > +}
> > +
> > +static acpi_status
> > +acpi_uart_enum_pnpids(struct acpi_device *adev,
> > + struct acpi_resource_uart_serialbus *sb,
> > + void *context)
> > +{
> > + int len;
> > + struct acpi_uart_buf *ctx = context;
> > +
> > + len = acpi_device_create_modalias(adev, ctx->buf, ctx->size);
> > + *ctx->len = len;
> > +
> > + return len < 0 ? AE_CTRL_DEPTH : AE_OK; }
> > +
> > +/**
> > + * acpi_uart_get_peripheral_type - get peripheral HID/CIDs
> > + * @dev: the tty class device
> > + * @buf: the string buffer containing HID/CIDs
> > + * @size: the size of the string buffer
> > + *
> > + * Obtain UART peripheral HID/CIDs. Return buffer size on success,
> > + * -errno on failure.
> > + */
> > +int acpi_uart_get_peripheral_type(struct device *dev,
> > + char *buf, size_t size)
> > +{
> > + int ret;
> > + int len = 0;
> > + struct acpi_uart_buf ctx = { buf, size, &len };
> > +
> > + ret = acpi_uart_walk_port(dev->parent, &ctx);
> > + if (ret < 0)
> > + return ret;
> > +
> > + return len;
> > +}
> > +EXPORT_SYMBOL_GPL(acpi_uart_get_peripheral_type);
> > diff --git a/drivers/acpi/internal.h b/drivers/acpi/internal.h index
> > 3c94a73..30d7440 100644
> > --- a/drivers/acpi/internal.h
> > +++ b/drivers/acpi/internal.h
> > @@ -61,6 +61,8 @@ void acpi_init_device_object(struct acpi_device *device,
> acpi_handle handle,
> > int type, unsigned long long sta); void
> > acpi_device_add_finalize(struct acpi_device *device); void
> > acpi_free_ids(struct acpi_device *device);
> > +int acpi_device_create_modalias(struct acpi_device *acpi_dev,
> > + char *modalias, int size);
> >
> > /* --------------------------------------------------------------------------
> > Power Resource diff --git
> > a/drivers/acpi/scan.c b/drivers/acpi/scan.c index 5e7e991..0d7d49f
> > 100644
> > --- a/drivers/acpi/scan.c
> > +++ b/drivers/acpi/scan.c
> > @@ -68,8 +68,8 @@ int acpi_scan_add_handler(struct acpi_scan_handler
> *handler)
> > * e.g. on a device with hid:IBM0001 and cid:ACPI0001 you get:
> > * char *modalias: "acpi:IBM0001:ACPI0001"
> > */
> > -static int create_modalias(struct acpi_device *acpi_dev, char *modalias,
> > - int size)
> > +int acpi_device_create_modalias(struct acpi_device *acpi_dev,
> > + char *modalias, int size)
> > {
> > int len;
> > int count;
> > @@ -99,7 +99,7 @@ acpi_device_modalias_show(struct device *dev, struct
> device_attribute *attr, cha
> > int len;
> >
> > /* Device has no HID and no CID or string is >1024 */
> > - len = create_modalias(acpi_dev, buf, 1024);
> > + len = acpi_device_create_modalias(acpi_dev, buf, 1024);
> > if (len <= 0)
> > return 0;
> > buf[len++] = '\n';
> > @@ -567,7 +567,7 @@ static int acpi_device_uevent(struct device *dev,
> > struct kobj_uevent_env *env)
> >
> > if (add_uevent_var(env, "MODALIAS="))
> > return -ENOMEM;
> > - len = create_modalias(acpi_dev, &env->buf[env->buflen - 1],
> > + len = acpi_device_create_modalias(acpi_dev, &env->buf[env->buflen -
> > +1],
> > sizeof(env->buf) - env->buflen);
> > if (len >= (sizeof(env->buf) - env->buflen))
> > return -ENOMEM;
> > diff --git a/drivers/tty/serial/serial_core.c
> > b/drivers/tty/serial/serial_core.c
> > index a400002..17e104e 100644
> > --- a/drivers/tty/serial/serial_core.c
> > +++ b/drivers/tty/serial/serial_core.c
> > @@ -2515,6 +2515,19 @@ static ssize_t
> uart_get_attr_iomem_reg_shift(struct device *dev,
> > return snprintf(buf, PAGE_SIZE, "%d\n", tmp.iomem_reg_shift); }
> >
> > +static ssize_t uart_get_attr_peripheral_type(struct device *dev,
> > + struct device_attribute *attr, char *buf) {
> > + int len;
> > +
> > + len = acpi_uart_get_peripheral_type(dev, buf, 1024);
> > + if (len > 0) {
> > + buf[len++] = '\n';
> > + return len;
> > + }
> > + return snprintf(buf, PAGE_SIZE, "uart\n");
>
> What is the "uart" part for? Isn't that kind of obvious?
It is the default "peripheral_type" content.
Before sending next revision out, let me ask a question here:
Shall I leave this empty as my previous revisions?
> > +}
>
> You are creating a new sysfs file, you must create a new Documentation/ABI
> entry as well.
OK.
> > +
> > static DEVICE_ATTR(type, S_IRUSR | S_IRGRP, uart_get_attr_type,
> > NULL); static DEVICE_ATTR(line, S_IRUSR | S_IRGRP,
> > uart_get_attr_line, NULL); static DEVICE_ATTR(port, S_IRUSR |
> > S_IRGRP, uart_get_attr_port, NULL); @@ -2528,6 +2541,7 @@ static
> > DEVICE_ATTR(custom_divisor, S_IRUSR | S_IRGRP,
> > uart_get_attr_custom_divis static DEVICE_ATTR(io_type, S_IRUSR |
> > S_IRGRP, uart_get_attr_io_type, NULL); static DEVICE_ATTR(iomem_base,
> > S_IRUSR | S_IRGRP, uart_get_attr_iomem_base, NULL); static
> > DEVICE_ATTR(iomem_reg_shift, S_IRUSR | S_IRGRP,
> > uart_get_attr_iomem_reg_shift, NULL);
> > +static DEVICE_ATTR(peripheral_type, S_IRUSR | S_IRGRP,
> > +uart_get_attr_peripheral_type, NULL);
> >
> > static struct attribute *tty_dev_attrs[] = {
> > &dev_attr_type.attr,
> > @@ -2543,6 +2557,7 @@ static struct attribute *tty_dev_attrs[] = {
> > &dev_attr_io_type.attr,
> > &dev_attr_iomem_base.attr,
> > &dev_attr_iomem_reg_shift.attr,
> > + &dev_attr_peripheral_type.attr,
> > NULL,
> > };
> >
> > diff --git a/include/linux/tty.h b/include/linux/tty.h index
> > 367a9df..723fdd01 100644
> > --- a/include/linux/tty.h
> > +++ b/include/linux/tty.h
> > @@ -692,4 +692,15 @@ do { \
> > } while (0)
> >
> >
> > +#ifdef CONFIG_ACPI_UART
> > +int acpi_uart_get_peripheral_type(struct device *dev,
> > + char *buf, size_t size);
> > +#else
> > +static inline int acpi_uart_get_peripheral_type(struct device *dev,
> > + char *buf, size_t size)
> > +{
> > + return -ENODEV;
> > +}
> > +#endif
>
> I see you never tested this code without the configuration option enabled :(
>
> {sigh}
Yes, I rely too much on an internal test system as it can do this for me automatically.
OK, I'll take care of this next time.
I've a plan and a TODO to offer a complete test framework for this as I've done 10 years ago.
Hope you will like it.
Thanks and best regards
-Lv
--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
