All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v2] kdbus: do not append the same connection to the queue twice
@ 2015-04-14  8:24 Lukasz Skalski
  2015-06-04  5:51 ` Greg KH
  2015-06-22 22:53 ` Greg KH
  0 siblings, 2 replies; 5+ messages in thread
From: Lukasz Skalski @ 2015-04-14  8:24 UTC (permalink / raw)
  To: gregkh, daniel, dh.herrmann, tixxdz; +Cc: sergei, linux-kernel, Lukasz Skalski

As it was discussed on systemd ML [1], the same connection should be
queued up only once for a given well-known name.

[1] http://lists.freedesktop.org/archives/systemd-devel/2015-April/030494.html

Signed-off-by: Lukasz Skalski <l.skalski@samsung.com>

diff --git a/ipc/kdbus/names.c b/ipc/kdbus/names.c
index 657008e..df99e4d 100644
--- a/ipc/kdbus/names.c
+++ b/ipc/kdbus/names.c
@@ -353,10 +353,24 @@ int kdbus_name_acquire(struct kdbus_name_registry *reg,
 	} else if (flags & KDBUS_NAME_QUEUE) {
 		/* add to waiting-queue of the name */
 
-		ret = kdbus_name_pending_new(e, conn, flags);
-		if (ret >= 0)
-			/* tell the caller that we queued it */
-			rflags |= KDBUS_NAME_IN_QUEUE;
+		struct kdbus_name_pending *p;
+		bool in_queue = false;
+
+		list_for_each_entry(p, &e->queue, name_entry) {
+			if (p->conn == conn) {
+				/* connection is already queued */
+				rflags |= KDBUS_NAME_IN_QUEUE;
+				in_queue = true;
+				break;
+			}
+		}
+
+		if (!in_queue) {
+			ret = kdbus_name_pending_new(e, conn, flags);
+			if (ret >= 0)
+				/* tell the caller that we queued it */
+				rflags |= KDBUS_NAME_IN_QUEUE;
+		}
 	} else {
 		/* the name is busy, return a failure */
 		ret = -EEXIST;
-- 
1.9.3


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH v2] kdbus: do not append the same connection to the queue twice
  2015-04-14  8:24 [PATCH v2] kdbus: do not append the same connection to the queue twice Lukasz Skalski
@ 2015-06-04  5:51 ` Greg KH
  2015-06-04  6:19   ` David Herrmann
  2015-06-22 22:53 ` Greg KH
  1 sibling, 1 reply; 5+ messages in thread
From: Greg KH @ 2015-06-04  5:51 UTC (permalink / raw)
  To: Lukasz Skalski; +Cc: daniel, dh.herrmann, tixxdz, sergei, linux-kernel

On Tue, Apr 14, 2015 at 10:24:53AM +0200, Lukasz Skalski wrote:
> As it was discussed on systemd ML [1], the same connection should be
> queued up only once for a given well-known name.
> 
> [1] http://lists.freedesktop.org/archives/systemd-devel/2015-April/030494.html
> 
> Signed-off-by: Lukasz Skalski <l.skalski@samsung.com>
> 
> diff --git a/ipc/kdbus/names.c b/ipc/kdbus/names.c
> index 657008e..df99e4d 100644
> --- a/ipc/kdbus/names.c
> +++ b/ipc/kdbus/names.c
> @@ -353,10 +353,24 @@ int kdbus_name_acquire(struct kdbus_name_registry *reg,
>  	} else if (flags & KDBUS_NAME_QUEUE) {
>  		/* add to waiting-queue of the name */
>  
> -		ret = kdbus_name_pending_new(e, conn, flags);
> -		if (ret >= 0)
> -			/* tell the caller that we queued it */
> -			rflags |= KDBUS_NAME_IN_QUEUE;
> +		struct kdbus_name_pending *p;
> +		bool in_queue = false;
> +
> +		list_for_each_entry(p, &e->queue, name_entry) {
> +			if (p->conn == conn) {
> +				/* connection is already queued */
> +				rflags |= KDBUS_NAME_IN_QUEUE;
> +				in_queue = true;
> +				break;
> +			}
> +		}
> +
> +		if (!in_queue) {
> +			ret = kdbus_name_pending_new(e, conn, flags);
> +			if (ret >= 0)
> +				/* tell the caller that we queued it */
> +				rflags |= KDBUS_NAME_IN_QUEUE;
> +		}
>  	} else {
>  		/* the name is busy, return a failure */
>  		ret = -EEXIST;

David, did you miss this patch in your last round of kdbus updates?

thanks,

greg k-h

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH v2] kdbus: do not append the same connection to the queue twice
  2015-06-04  5:51 ` Greg KH
@ 2015-06-04  6:19   ` David Herrmann
  0 siblings, 0 replies; 5+ messages in thread
From: David Herrmann @ 2015-06-04  6:19 UTC (permalink / raw)
  To: Greg KH
  Cc: Lukasz Skalski, Daniel Mack, Djalal Harouni, Sergei Zviagintsev,
	linux-kernel

Hi

On Thu, Jun 4, 2015 at 7:51 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> On Tue, Apr 14, 2015 at 10:24:53AM +0200, Lukasz Skalski wrote:
>> As it was discussed on systemd ML [1], the same connection should be
>> queued up only once for a given well-known name.
>>
>> [1] http://lists.freedesktop.org/archives/systemd-devel/2015-April/030494.html
>>
>> Signed-off-by: Lukasz Skalski <l.skalski@samsung.com>
>>
>> diff --git a/ipc/kdbus/names.c b/ipc/kdbus/names.c
>> index 657008e..df99e4d 100644
>> --- a/ipc/kdbus/names.c
>> +++ b/ipc/kdbus/names.c
>> @@ -353,10 +353,24 @@ int kdbus_name_acquire(struct kdbus_name_registry *reg,
>>       } else if (flags & KDBUS_NAME_QUEUE) {
>>               /* add to waiting-queue of the name */
>>
>> -             ret = kdbus_name_pending_new(e, conn, flags);
>> -             if (ret >= 0)
>> -                     /* tell the caller that we queued it */
>> -                     rflags |= KDBUS_NAME_IN_QUEUE;
>> +             struct kdbus_name_pending *p;
>> +             bool in_queue = false;
>> +
>> +             list_for_each_entry(p, &e->queue, name_entry) {
>> +                     if (p->conn == conn) {
>> +                             /* connection is already queued */
>> +                             rflags |= KDBUS_NAME_IN_QUEUE;
>> +                             in_queue = true;
>> +                             break;
>> +                     }
>> +             }
>> +
>> +             if (!in_queue) {
>> +                     ret = kdbus_name_pending_new(e, conn, flags);
>> +                     if (ret >= 0)
>> +                             /* tell the caller that we queued it */
>> +                             rflags |= KDBUS_NAME_IN_QUEUE;
>> +             }
>>       } else {
>>               /* the name is busy, return a failure */
>>               ret = -EEXIST;
>
> David, did you miss this patch in your last round of kdbus updates?

I already talked to Lukasz off-list. There are some more changes
required, they'll be included in round #2.

Thanks
David

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH v2] kdbus: do not append the same connection to the queue twice
  2015-04-14  8:24 [PATCH v2] kdbus: do not append the same connection to the queue twice Lukasz Skalski
  2015-06-04  5:51 ` Greg KH
@ 2015-06-22 22:53 ` Greg KH
  2015-08-04 16:43   ` David Herrmann
  1 sibling, 1 reply; 5+ messages in thread
From: Greg KH @ 2015-06-22 22:53 UTC (permalink / raw)
  To: Lukasz Skalski; +Cc: daniel, dh.herrmann, tixxdz, sergei, linux-kernel

On Tue, Apr 14, 2015 at 10:24:53AM +0200, Lukasz Skalski wrote:
> As it was discussed on systemd ML [1], the same connection should be
> queued up only once for a given well-known name.
> 
> [1] http://lists.freedesktop.org/archives/systemd-devel/2015-April/030494.html
> 
> Signed-off-by: Lukasz Skalski <l.skalski@samsung.com>
> 
> diff --git a/ipc/kdbus/names.c b/ipc/kdbus/names.c
> index 657008e..df99e4d 100644
> --- a/ipc/kdbus/names.c
> +++ b/ipc/kdbus/names.c
> @@ -353,10 +353,24 @@ int kdbus_name_acquire(struct kdbus_name_registry *reg,
>  	} else if (flags & KDBUS_NAME_QUEUE) {
>  		/* add to waiting-queue of the name */
>  
> -		ret = kdbus_name_pending_new(e, conn, flags);
> -		if (ret >= 0)
> -			/* tell the caller that we queued it */
> -			rflags |= KDBUS_NAME_IN_QUEUE;
> +		struct kdbus_name_pending *p;
> +		bool in_queue = false;
> +
> +		list_for_each_entry(p, &e->queue, name_entry) {
> +			if (p->conn == conn) {
> +				/* connection is already queued */
> +				rflags |= KDBUS_NAME_IN_QUEUE;
> +				in_queue = true;
> +				break;
> +			}
> +		}
> +
> +		if (!in_queue) {
> +			ret = kdbus_name_pending_new(e, conn, flags);
> +			if (ret >= 0)
> +				/* tell the caller that we queued it */
> +				rflags |= KDBUS_NAME_IN_QUEUE;
> +		}
>  	} else {
>  		/* the name is busy, return a failure */
>  		ret = -EEXIST;
> -- 
> 1.9.3

David, what happened with this, should it be applied?

thanks,

greg k-h
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
Please read the FAQ at  http://www.tux.org/lkml/

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH v2] kdbus: do not append the same connection to the queue twice
  2015-06-22 22:53 ` Greg KH
@ 2015-08-04 16:43   ` David Herrmann
  0 siblings, 0 replies; 5+ messages in thread
From: David Herrmann @ 2015-08-04 16:43 UTC (permalink / raw)
  To: Greg KH; +Cc: Lukasz Skalski, Daniel Mack, Djalal Harouni, linux-kernel

Hi Lukasz & Greg

On Tue, Jun 23, 2015 at 12:53 AM, Greg KH <gregkh@linuxfoundation.org> wrote:
> On Tue, Apr 14, 2015 at 10:24:53AM +0200, Lukasz Skalski wrote:
>> As it was discussed on systemd ML [1], the same connection should be
>> queued up only once for a given well-known name.
>>
>> [1] http://lists.freedesktop.org/archives/systemd-devel/2015-April/030494.html
>>
>> Signed-off-by: Lukasz Skalski <l.skalski@samsung.com>
>>
>> diff --git a/ipc/kdbus/names.c b/ipc/kdbus/names.c
>> index 657008e..df99e4d 100644
>> --- a/ipc/kdbus/names.c
>> +++ b/ipc/kdbus/names.c
>> @@ -353,10 +353,24 @@ int kdbus_name_acquire(struct kdbus_name_registry *reg,
>>       } else if (flags & KDBUS_NAME_QUEUE) {
>>               /* add to waiting-queue of the name */
>>
>> -             ret = kdbus_name_pending_new(e, conn, flags);
>> -             if (ret >= 0)
>> -                     /* tell the caller that we queued it */
>> -                     rflags |= KDBUS_NAME_IN_QUEUE;
>> +             struct kdbus_name_pending *p;
>> +             bool in_queue = false;
>> +
>> +             list_for_each_entry(p, &e->queue, name_entry) {
>> +                     if (p->conn == conn) {
>> +                             /* connection is already queued */
>> +                             rflags |= KDBUS_NAME_IN_QUEUE;
>> +                             in_queue = true;
>> +                             break;
>> +                     }
>> +             }
>> +
>> +             if (!in_queue) {
>> +                     ret = kdbus_name_pending_new(e, conn, flags);
>> +                     if (ret >= 0)
>> +                             /* tell the caller that we queued it */
>> +                             rflags |= KDBUS_NAME_IN_QUEUE;
>> +             }
>>       } else {
>>               /* the name is busy, return a failure */
>>               ret = -EEXIST;
>> --
>> 1.9.3
>
> David, what happened with this, should it be applied?

Sorry for the delay. I finally came around restructuring
name-acquisition so we follow the dbus-space (including the change
from Lukasz here). It's the last thing I had on my TODO list (so also
the last non-trivial patch that is pending).

See here: https://github.com/systemd/kdbus/commit/95a96676295cd633f66822659ec248398259a8fc

I will send it to LKML later/tomorrow.

Thanks, and sorry for the delay!
David

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2015-08-04 16:43 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-04-14  8:24 [PATCH v2] kdbus: do not append the same connection to the queue twice Lukasz Skalski
2015-06-04  5:51 ` Greg KH
2015-06-04  6:19   ` David Herrmann
2015-06-22 22:53 ` Greg KH
2015-08-04 16:43   ` David Herrmann

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.