[dpdk-dev] [PATCH 0/2] Thread termination leak fixes

[dpdk-dev] [PATCH 0/2] Thread termination leak fixes

[David Marchand]

After looking at the control thread API use in recent changes, I noticed
that some thread resources are not properly released.
This series fixes two net drivers.

Additional notes:
- I noticed that event/dlb2 never kills/terminates its service thread.
  It means unplug of this device (if supported) would leak the associated
  thread resources. I leave it to this driver maintainer to comment/fix.

- The vhost library also lets a service thread in the background but since
  there is no way to unitialize this library, there is no leak to fix from
  my pov.

-- 
David Marchand

David Marchand (2):
  net/ark: fix leak on thread termination
  net/ice: fix leak on thread termination

 drivers/net/ark/ark_pktgen.c     | 2 ++
 drivers/net/ice/ice_dcf_parent.c | 2 ++
 2 files changed, 4 insertions(+)

-- 
2.23.0

[dpdk-dev] [PATCH 1/2] net/ark: fix leak on thread termination

[David Marchand]

A terminated pthread should be joined or detached so that its associated
resources are released.

The "ark-delay-pg" thread is just used to delay some task but it is never
joined by the thread that created it.
The easiest solution is to detach the new thread.

Fixes: 727b3fe292bc ("net/ark: integrate PMD")
Cc: stable@dpdk.org

Signed-off-by: David Marchand <david.marchand@redhat.com>
---
 drivers/net/ark/ark_pktgen.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ark/ark_pktgen.c b/drivers/net/ark/ark_pktgen.c
index 28a44f7546..515bfe461c 100644
--- a/drivers/net/ark/ark_pktgen.c
+++ b/drivers/net/ark/ark_pktgen.c
@@ -3,6 +3,7 @@
  */
 
 #include <unistd.h>
+#include <pthread.h>
 
 #include <rte_string_fns.h>
 #include <rte_malloc.h>
@@ -474,6 +475,7 @@ ark_pktgen_delay_start(void *arg)
 	 * perform a blind sleep here to ensure that the external test
 	 * application has time to setup the test before we generate packets
 	 */
+	pthread_detach(pthread_self());
 	usleep(100000);
 	ark_pktgen_run(inst);
 	return NULL;
-- 
2.23.0

Re: [dpdk-dev] [PATCH 1/2] net/ark: fix leak on thread termination

[Ed Czeck]

Thank you.

Acked-by: Ed Czeck <ed.czeck@atomicrules.com>


On Thu, May 6, 2021 at 5:45 AM David Marchand <david.marchand@redhat.com>
wrote:

> A terminated pthread should be joined or detached so that its associated
> resources are released.
>
> The "ark-delay-pg" thread is just used to delay some task but it is never
> joined by the thread that created it.
> The easiest solution is to detach the new thread.
>
> Fixes: 727b3fe292bc ("net/ark: integrate PMD")
> Cc: stable@dpdk.org
>
> Signed-off-by: David Marchand <david.marchand@redhat.com>
> ---
>  drivers/net/ark/ark_pktgen.c | 2 ++
>  1 file changed, 2 insertions(+)
>
> diff --git a/drivers/net/ark/ark_pktgen.c b/drivers/net/ark/ark_pktgen.c
> index 28a44f7546..515bfe461c 100644
> --- a/drivers/net/ark/ark_pktgen.c
> +++ b/drivers/net/ark/ark_pktgen.c
> @@ -3,6 +3,7 @@
>   */
>
>  #include <unistd.h>
> +#include <pthread.h>
>
>  #include <rte_string_fns.h>
>  #include <rte_malloc.h>
> @@ -474,6 +475,7 @@ ark_pktgen_delay_start(void *arg)
>          * perform a blind sleep here to ensure that the external test
>          * application has time to setup the test before we generate
> packets
>          */
> +       pthread_detach(pthread_self());
>         usleep(100000);
>         ark_pktgen_run(inst);
>         return NULL;
> --
> 2.23.0
>
>

[dpdk-dev] [PATCH 2/2] net/ice: fix leak on thread termination

[David Marchand]

A terminated pthread should be joined or detached so that its associated
resources are released.

The "ice-reset-<vf_id>" threads are used to service some reset task in the
background, but they are never joined by the thread that created them.
The easiest solution is to detach new threads.

Fixes: 3b3757bda3c3 ("net/ice: get VF hardware index in DCF")
Cc: stable@dpdk.org

Signed-off-by: David Marchand <david.marchand@redhat.com>
---
 drivers/net/ice/ice_dcf_parent.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ice/ice_dcf_parent.c b/drivers/net/ice/ice_dcf_parent.c
index c8e433239b..1d7aa8bc87 100644
--- a/drivers/net/ice/ice_dcf_parent.c
+++ b/drivers/net/ice/ice_dcf_parent.c
@@ -121,6 +121,8 @@ ice_dcf_vsi_update_service_handler(void *param)
 	struct ice_dcf_hw *hw = reset_param->dcf_hw;
 	struct ice_dcf_adapter *adapter;
 
+	pthread_detach(pthread_self());
+
 	rte_delay_us(ICE_DCF_VSI_UPDATE_SERVICE_INTERVAL);
 
 	rte_spinlock_lock(&vsi_update_lock);
-- 
2.23.0

Re: [dpdk-dev] [PATCH 2/2] net/ice: fix leak on thread termination

[Wang, Haiyue]

> -----Original Message-----
> From: David Marchand <david.marchand@redhat.com>
> Sent: Thursday, May 6, 2021 17:45
> To: dev@dpdk.org
> Cc: stable@dpdk.org; Yang, Qiming <qiming.yang@intel.com>; Zhang, Qi Z <qi.z.zhang@intel.com>; Wang,
> Haiyue <haiyue.wang@intel.com>
> Subject: [PATCH 2/2] net/ice: fix leak on thread termination
> 
> A terminated pthread should be joined or detached so that its associated
> resources are released.
> 
> The "ice-reset-<vf_id>" threads are used to service some reset task in the
> background, but they are never joined by the thread that created them.
> The easiest solution is to detach new threads.
> 
> Fixes: 3b3757bda3c3 ("net/ice: get VF hardware index in DCF")
> Cc: stable@dpdk.org
> 
> Signed-off-by: David Marchand <david.marchand@redhat.com>
> ---
>  drivers/net/ice/ice_dcf_parent.c | 2 ++
>  1 file changed, 2 insertions(+)
> 

Got it, thanks!

Acked-by: Haiyue Wang <haiyue.wang@intel.com>

> --
> 2.23.0

Re: [dpdk-dev] [dpdk-stable] [PATCH 2/2] net/ice: fix leak on thread termination

[David Marchand]

On Thu, May 6, 2021 at 11:45 AM David Marchand
<david.marchand@redhat.com> wrote:
>
> A terminated pthread should be joined or detached so that its associated
> resources are released.
>
> The "ice-reset-<vf_id>" threads are used to service some reset task in the
> background, but they are never joined by the thread that created them.
> The easiest solution is to detach new threads.

Not so "easy" for Windows.

I think I'll simply #ifndef WINDOWS my addition.
Maybe the leak does not exist on Windows? but if it does, the
situation won't change by doing nothing on Windows.

If there is strong objection, I'll need some help to add a
pthread_detach() wrapper in windows EAL.
From my quick read at the API, I'd say I need to find the current
thread handle (via OpenThread) then CloseHandle it.
But does it work from "pthread_self()" ?


Since a new API thread is in preparation, the "thread detaching" from
the pthread API is something to consider.
I could not find it in Narcissa series.


-- 
David Marchand

Re: [dpdk-dev] [dpdk-stable] [PATCH 2/2] net/ice: fix leak on thread termination

[Dmitry Kozlyuk]

2021-05-07 10:13 (UTC+0200), David Marchand:
> On Thu, May 6, 2021 at 11:45 AM David Marchand
> <david.marchand@redhat.com> wrote:
> >
> > A terminated pthread should be joined or detached so that its associated
> > resources are released.
> >
> > The "ice-reset-<vf_id>" threads are used to service some reset task in the
> > background, but they are never joined by the thread that created them.
> > The easiest solution is to detach new threads.  
> 
> Not so "easy" for Windows.
> 
> I think I'll simply #ifndef WINDOWS my addition.
> Maybe the leak does not exist on Windows? but if it does, the
> situation won't change by doing nothing on Windows.
>
> If there is strong objection, I'll need some help to add a
> pthread_detach() wrapper in windows EAL.
> From my quick read at the API, I'd say I need to find the current
> thread handle (via OpenThread) then CloseHandle it.
> But does it work from "pthread_self()" ?
> 
> 
> Since a new API thread is in preparation, the "thread detaching" from
> the pthread API is something to consider.
> I could not find it in Narcissa series.

There is no leak.
On Windows, pthread_t is just a number, not a handle.
pthread_detach can be implemented as a no-op and added to new API.
See also: https://devblogs.microsoft.com/oldnewthing/20100827-00/?p=13023

[dpdk-dev] [PATCH v2 0/2] Thread termination leak fixes

[David Marchand]

After looking at the control thread API use in recent changes, I noticed
that some thread resources are not properly released.
This series fixes two net drivers.

Additional notes:
- I noticed that event/dlb2 never kills/terminates its service thread.
  It means unplug of this device (if supported) would leak the associated
  thread resources. I leave it to this driver maintainer to comment/fix.

- The vhost library also lets a service thread in the background but since
  there is no way to unitialize this library, there is no leak to fix from
  my pov.

-- 
David Marchand

Changes since v1:
- fixed build for net/ice on Windows

David Marchand (2):
  net/ark: fix leak on thread termination
  net/ice: fix leak on thread termination

 drivers/net/ark/ark_pktgen.c      | 2 ++
 drivers/net/ice/ice_dcf_parent.c  | 2 ++
 lib/eal/windows/include/pthread.h | 6 ++++++
 3 files changed, 10 insertions(+)

-- 
2.23.0

[dpdk-dev] [PATCH v2 1/2] net/ark: fix leak on thread termination

[David Marchand]

A terminated pthread should be joined or detached so that its associated
resources are released.

The "ark-delay-pg" thread is just used to delay some task but it is never
joined by the thread that created it.
The easiest solution is to detach the new thread.

Fixes: 727b3fe292bc ("net/ark: integrate PMD")
Cc: stable@dpdk.org

Signed-off-by: David Marchand <david.marchand@redhat.com>
---
 drivers/net/ark/ark_pktgen.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ark/ark_pktgen.c b/drivers/net/ark/ark_pktgen.c
index 28a44f7546..515bfe461c 100644
--- a/drivers/net/ark/ark_pktgen.c
+++ b/drivers/net/ark/ark_pktgen.c
@@ -3,6 +3,7 @@
  */
 
 #include <unistd.h>
+#include <pthread.h>
 
 #include <rte_string_fns.h>
 #include <rte_malloc.h>
@@ -474,6 +475,7 @@ ark_pktgen_delay_start(void *arg)
 	 * perform a blind sleep here to ensure that the external test
 	 * application has time to setup the test before we generate packets
 	 */
+	pthread_detach(pthread_self());
 	usleep(100000);
 	ark_pktgen_run(inst);
 	return NULL;
-- 
2.23.0

[dpdk-dev] [PATCH v2 2/2] net/ice: fix leak on thread termination

[David Marchand]

A terminated pthread should be joined or detached so that its associated
resources are released.

The "ice-reset-<vf_id>" threads are used to service some reset task in the
background, but they are never joined by the thread that created them.
The easiest solution is to detach new threads.

The Windows EAL did not provide a pthread_detach wrapper but there is no
resource to release for Windows threads, so add an empty wrapper.

Fixes: 3b3757bda3c3 ("net/ice: get VF hardware index in DCF")
Cc: stable@dpdk.org

Signed-off-by: David Marchand <david.marchand@redhat.com>
Acked-by: Haiyue Wang <haiyue.wang@intel.com>
---
Changes since v1:
- fixed build for net/ice on Windows

---
 drivers/net/ice/ice_dcf_parent.c  | 2 ++
 lib/eal/windows/include/pthread.h | 6 ++++++
 2 files changed, 8 insertions(+)

diff --git a/drivers/net/ice/ice_dcf_parent.c b/drivers/net/ice/ice_dcf_parent.c
index c8e433239b..1d7aa8bc87 100644
--- a/drivers/net/ice/ice_dcf_parent.c
+++ b/drivers/net/ice/ice_dcf_parent.c
@@ -121,6 +121,8 @@ ice_dcf_vsi_update_service_handler(void *param)
 	struct ice_dcf_hw *hw = reset_param->dcf_hw;
 	struct ice_dcf_adapter *adapter;
 
+	pthread_detach(pthread_self());
+
 	rte_delay_us(ICE_DCF_VSI_UPDATE_SERVICE_INTERVAL);
 
 	rte_spinlock_lock(&vsi_update_lock);
diff --git a/lib/eal/windows/include/pthread.h b/lib/eal/windows/include/pthread.h
index 1939b0121c..27fd2cca52 100644
--- a/lib/eal/windows/include/pthread.h
+++ b/lib/eal/windows/include/pthread.h
@@ -143,6 +143,12 @@ pthread_create(void *threadid, const void *threadattr, void *threadfunc,
 	return ((hThread != NULL) ? 0 : E_FAIL);
 }
 
+static inline int
+pthread_detach(__rte_unused pthread_t thread)
+{
+	return 0;
+}
+
 static inline int
 pthread_join(__rte_unused pthread_t thread,
 	__rte_unused void **value_ptr)
-- 
2.23.0

Re: [dpdk-dev] [PATCH v2 2/2] net/ice: fix leak on thread termination

[Dmitry Kozlyuk]

2021-05-11 13:33 (UTC+0200), David Marchand:
> A terminated pthread should be joined or detached so that its associated
> resources are released.
> 
> The "ice-reset-<vf_id>" threads are used to service some reset task in the
> background, but they are never joined by the thread that created them.
> The easiest solution is to detach new threads.
> 
> The Windows EAL did not provide a pthread_detach wrapper but there is no
> resource to release for Windows threads, so add an empty wrapper.
> 
> Fixes: 3b3757bda3c3 ("net/ice: get VF hardware index in DCF")
> Cc: stable@dpdk.org
> 
> Signed-off-by: David Marchand <david.marchand@redhat.com>
> Acked-by: Haiyue Wang <haiyue.wang@intel.com>
> ---
> Changes since v1:
> - fixed build for net/ice on Windows
> 
> ---
>  drivers/net/ice/ice_dcf_parent.c  | 2 ++
>  lib/eal/windows/include/pthread.h | 6 ++++++
>  2 files changed, 8 insertions(+)
> 
> diff --git a/drivers/net/ice/ice_dcf_parent.c b/drivers/net/ice/ice_dcf_parent.c
> index c8e433239b..1d7aa8bc87 100644
> --- a/drivers/net/ice/ice_dcf_parent.c
> +++ b/drivers/net/ice/ice_dcf_parent.c
> @@ -121,6 +121,8 @@ ice_dcf_vsi_update_service_handler(void *param)
>  	struct ice_dcf_hw *hw = reset_param->dcf_hw;
>  	struct ice_dcf_adapter *adapter;
>  
> +	pthread_detach(pthread_self());
> +
>  	rte_delay_us(ICE_DCF_VSI_UPDATE_SERVICE_INTERVAL);
>  
>  	rte_spinlock_lock(&vsi_update_lock);
> diff --git a/lib/eal/windows/include/pthread.h b/lib/eal/windows/include/pthread.h
> index 1939b0121c..27fd2cca52 100644
> --- a/lib/eal/windows/include/pthread.h
> +++ b/lib/eal/windows/include/pthread.h
> @@ -143,6 +143,12 @@ pthread_create(void *threadid, const void *threadattr, void *threadfunc,
>  	return ((hThread != NULL) ? 0 : E_FAIL);
>  }
>  
> +static inline int
> +pthread_detach(__rte_unused pthread_t thread)
> +{
> +	return 0;
> +}
> +
>  static inline int
>  pthread_join(__rte_unused pthread_t thread,
>  	__rte_unused void **value_ptr)

For Windows part,
Acked-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>

Re: [dpdk-dev] [PATCH v2 0/2] Thread termination leak fixes

[Ferruh Yigit]

On 5/11/2021 12:33 PM, David Marchand wrote:
> After looking at the control thread API use in recent changes, I noticed
> that some thread resources are not properly released.
> This series fixes two net drivers.
> 

Series applied to dpdk-next-net/main, thanks.

> Additional notes:
> - I noticed that event/dlb2 never kills/terminates its service thread.
>   It means unplug of this device (if supported) would leak the associated
>   thread resources. I leave it to this driver maintainer to comment/fix.
> 
> - The vhost library also lets a service thread in the background but since
>   there is no way to unitialize this library, there is no leak to fix from
>   my pov.
>