2010-08-09 14:23:25

by Dmitry Baryshkov

[permalink] [raw]
Subject: [PATCH 1/2] tty: add tty_struct->dev pointer to corresponding device instance

Some device drivers (mostly tty line disciplines) would like to have way
know a struct device instance corresponding to passed tty_struct. Add
a struct device pointer to struct tty_struct and populate it during
initialize_tty_struct().

Signed-off-by: Dmitry Eremin-Solenikov <[email protected]>
Cc: Alan Cox <[email protected]>
---
drivers/char/tty_io.c | 17 +++++++++++++++++
include/linux/tty.h | 1 +
2 files changed, 18 insertions(+), 0 deletions(-)

diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index d71f0fc..2469682 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -179,6 +179,8 @@ struct tty_struct *alloc_tty_struct(void)

void free_tty_struct(struct tty_struct *tty)
{
+ if (tty->dev)
+ put_device(tty->dev);
kfree(tty->write_buf);
tty_buffer_free_all(tty);
kfree(tty);
@@ -2737,6 +2739,20 @@ void do_SAK(struct tty_struct *tty)

EXPORT_SYMBOL(do_SAK);

+static int dev_match_devt(struct device *dev, void *data)
+{
+ dev_t *devt = data;
+ return dev->devt == *devt;
+}
+
+/* Must put_device() after it's unused! */
+static struct device *tty_get_device(struct tty_struct *tty)
+{
+ dev_t devt = tty_devnum(tty);
+ return class_find_device(tty_class, NULL, &devt, dev_match_devt);
+}
+
+
/**
* initialize_tty_struct
* @tty: tty to initialize
@@ -2777,6 +2793,7 @@ void initialize_tty_struct(struct tty_struct *tty,
tty->ops = driver->ops;
tty->index = idx;
tty_line_name(driver, idx, tty->name);
+ tty->dev = tty_get_device(tty);
}

/**
diff --git a/include/linux/tty.h b/include/linux/tty.h
index 931078b..d64db5a 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -254,6 +254,7 @@ struct tty_operations;
struct tty_struct {
int magic;
struct kref kref;
+ struct device *dev;
struct tty_driver *driver;
const struct tty_operations *ops;
int index;
--
1.7.1


2010-08-09 14:23:31

by Dmitry Baryshkov

[permalink] [raw]
Subject: [PATCH 2/2] serport: place serport serio device correctly in the device tree

Make serport serio device to be a child of corresponding tty device
instead of just hanging at /sys/devices/serioX.

Signed-off-by: Dmitry Eremin-Solenikov <[email protected]>
Cc: Dmitry Torokhov <[email protected]>
---
drivers/input/serio/serport.c | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/input/serio/serport.c b/drivers/input/serio/serport.c
index 6d34511..6e362de 100644
--- a/drivers/input/serio/serport.c
+++ b/drivers/input/serio/serport.c
@@ -165,6 +165,7 @@ static ssize_t serport_ldisc_read(struct tty_struct * tty, struct file * file, u
serio->open = serport_serio_open;
serio->close = serport_serio_close;
serio->port_data = serport;
+ serio->dev.parent = tty->dev;

serio_register_port(serport->serio);
printk(KERN_INFO "serio: Serial port %s\n", tty_name(tty, name));
--
1.7.1

2010-08-09 14:53:51

by Alan

[permalink] [raw]
Subject: Re: [PATCH 1/2] tty: add tty_struct->dev pointer to corresponding device instance

On Mon, 9 Aug 2010 18:22:49 +0400
Dmitry Eremin-Solenikov <[email protected]> wrote:

> Some device drivers (mostly tty line disciplines) would like to have way
> know a struct device instance corresponding to passed tty_struct. Add
> a struct device pointer to struct tty_struct and populate it during
> initialize_tty_struct().

Acked-by: Alan Cox <[email protected]>

2010-08-11 03:27:54

by Dmitry Torokhov

[permalink] [raw]
Subject: Re: [PATCH 2/2] serport: place serport serio device correctly in the device tree

On Mon, Aug 09, 2010 at 06:22:50PM +0400, Dmitry Eremin-Solenikov wrote:
> Make serport serio device to be a child of corresponding tty device
> instead of just hanging at /sys/devices/serioX.
>
> Signed-off-by: Dmitry Eremin-Solenikov <[email protected]>
> Cc: Dmitry Torokhov <[email protected]>

Looks good to me.

Acked-by: Dmitry Torokhov <[email protected]>

> ---
> drivers/input/serio/serport.c | 1 +
> 1 files changed, 1 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/input/serio/serport.c b/drivers/input/serio/serport.c
> index 6d34511..6e362de 100644
> --- a/drivers/input/serio/serport.c
> +++ b/drivers/input/serio/serport.c
> @@ -165,6 +165,7 @@ static ssize_t serport_ldisc_read(struct tty_struct * tty, struct file * file, u
> serio->open = serport_serio_open;
> serio->close = serport_serio_close;
> serio->port_data = serport;
> + serio->dev.parent = tty->dev;
>
> serio_register_port(serport->serio);
> printk(KERN_INFO "serio: Serial port %s\n", tty_name(tty, name));
> --
> 1.7.1
>

--
Dmitry

2010-08-16 10:57:30

by Dmitry Baryshkov

[permalink] [raw]
Subject: Re: [PATCH 1/2] tty: add tty_struct->dev pointer to corresponding device instance

Hello,

On Mon, Aug 9, 2010 at 7:08 PM, Alan Cox <[email protected]> wrote:
> On Mon, ?9 Aug 2010 18:22:49 +0400
> Dmitry Eremin-Solenikov <[email protected]> wrote:
>
>> Some device drivers (mostly tty line disciplines) would like to have way
>> know a struct device instance corresponding to passed tty_struct. Add
>> a struct device pointer to struct tty_struct and populate it during
>> initialize_tty_struct().
>
> Acked-by: Alan Cox <[email protected]>
>

Greg, what about these two patches? Can I expect for them to be pulled
into your patch queue for 2.6.37?

--
With best wishes
Dmitry

2010-08-16 16:25:14

by Greg KH

[permalink] [raw]
Subject: Re: [PATCH 1/2] tty: add tty_struct->dev pointer to corresponding device instance

On Mon, Aug 16, 2010 at 02:57:28PM +0400, Dmitry Eremin-Solenikov wrote:
> Hello,
>
> On Mon, Aug 9, 2010 at 7:08 PM, Alan Cox <[email protected]> wrote:
> > On Mon, ?9 Aug 2010 18:22:49 +0400
> > Dmitry Eremin-Solenikov <[email protected]> wrote:
> >
> >> Some device drivers (mostly tty line disciplines) would like to have way
> >> know a struct device instance corresponding to passed tty_struct. Add
> >> a struct device pointer to struct tty_struct and populate it during
> >> initialize_tty_struct().
> >
> > Acked-by: Alan Cox <[email protected]>
> >
>
> Greg, what about these two patches? Can I expect for them to be pulled
> into your patch queue for 2.6.37?

Yes, I will queue them up to my -next branch soon and will go to Linus
for .37.

thanks,

greg k-h