[PATCH] xen: credit2: enable per cpu runqueue creation

[PATCH] xen: credit2: enable per cpu runqueue creation

[Praveen Kumar]

The patch introduces a new command line option 'cpu' that when used will create
runqueue per logical pCPU. This may be useful for small systems, and also for
development, performance evalution and comparison.

Signed-off-by: Praveen Kumar <kpraveen.lkml@gmail.com>
Reviewed-by: Dario Faggioli <dario.faggioli@citrix.com>

---
 docs/misc/xen-command-line.markdown |  3 ++-
 xen/common/sched_credit2.c          | 15 +++++++++++----
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/docs/misc/xen-command-line.markdown b/docs/misc/xen-command-line.markdown
index 5815d87dab..6e73766574 100644
--- a/docs/misc/xen-command-line.markdown
+++ b/docs/misc/xen-command-line.markdown
@@ -525,7 +525,7 @@ also slow in responding to load changes.
 The default value of `1 sec` is rather long.
 
 ### credit2\_runqueue
-> `= core | socket | node | all`
+> `= cpu | core | socket | node | all`
 
 > Default: `socket`
 
@@ -536,6 +536,7 @@ balancing (for instance, it will deal better with hyperthreading),
 but also more overhead.
 
 Available alternatives, with their meaning, are:
+* `cpu`: one runqueue per each logical pCPUs of the host;
 * `core`: one runqueue per each physical core of the host;
 * `socket`: one runqueue per each physical socket (which often,
             but not always, matches a NUMA node) of the host;
diff --git a/xen/common/sched_credit2.c b/xen/common/sched_credit2.c
index bb1c657e76..ee7b443f9e 100644
--- a/xen/common/sched_credit2.c
+++ b/xen/common/sched_credit2.c
@@ -301,6 +301,9 @@ integer_param("credit2_balance_over", opt_overload_balance_tolerance);
  * want that to happen basing on topology. At the moment, it is possible
  * to choose to arrange runqueues to be:
  *
+ * - per-cpu: meaning that there will be one runqueue per logical cpu. This
+ *            will happen when if the opt_runqueue parameter is set to 'cpu'.
+ *
  * - per-core: meaning that there will be one runqueue per each physical
  *             core of the host. This will happen if the opt_runqueue
  *             parameter is set to 'core';
@@ -322,11 +325,13 @@ integer_param("credit2_balance_over", opt_overload_balance_tolerance);
  * either the same physical core, the same physical socket, the same NUMA
  * node, or just all of them, will be put together to form runqueues.
  */
-#define OPT_RUNQUEUE_CORE   0
-#define OPT_RUNQUEUE_SOCKET 1
-#define OPT_RUNQUEUE_NODE   2
-#define OPT_RUNQUEUE_ALL    3
+#define OPT_RUNQUEUE_CPU    0
+#define OPT_RUNQUEUE_CORE   1
+#define OPT_RUNQUEUE_SOCKET 2
+#define OPT_RUNQUEUE_NODE   3
+#define OPT_RUNQUEUE_ALL    4
 static const char *const opt_runqueue_str[] = {
+    [OPT_RUNQUEUE_CPU] = "cpu",
     [OPT_RUNQUEUE_CORE] = "core",
     [OPT_RUNQUEUE_SOCKET] = "socket",
     [OPT_RUNQUEUE_NODE] = "node",
@@ -682,6 +687,8 @@ cpu_to_runqueue(struct csched2_private *prv, unsigned int cpu)
         BUG_ON(cpu_to_socket(cpu) == XEN_INVALID_SOCKET_ID ||
                cpu_to_socket(peer_cpu) == XEN_INVALID_SOCKET_ID);
 
+        if (opt_runqueue == OPT_RUNQUEUE_CPU)
+            continue;
         if ( opt_runqueue == OPT_RUNQUEUE_ALL ||
              (opt_runqueue == OPT_RUNQUEUE_CORE && same_core(peer_cpu, cpu)) ||
              (opt_runqueue == OPT_RUNQUEUE_SOCKET && same_socket(peer_cpu, cpu)) ||
-- 
2.12.0


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH] xen: credit2: enable per cpu runqueue creation

[George Dunlap]

On 11/04/17 17:15, Praveen Kumar wrote:
> The patch introduces a new command line option 'cpu' that when used will create
> runqueue per logical pCPU. This may be useful for small systems, and also for
> development, performance evalution and comparison.
> 
> Signed-off-by: Praveen Kumar <kpraveen.lkml@gmail.com>
> Reviewed-by: Dario Faggioli <dario.faggioli@citrix.com>

Praveen,

Unfortunately this didn't get checked in before the feature freeze -- I
think I may have dropped the ball on this one.

I'll queue this one up to be merged once the Xen 4.10 development window
opens.

 -George


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH] xen: credit2: enable per cpu runqueue creation

[Dario Faggioli]

[-- Attachment #1.1: Type: text/plain, Size: 4363 bytes --]

On Tue, 2017-04-11 at 21:45 +0530, Praveen Kumar wrote:
> The patch introduces a new command line option 'cpu' that when used
> will create
> runqueue per logical pCPU. This may be useful for small systems, and
> also for
> development, performance evalution and comparison.
> 
> Signed-off-by: Praveen Kumar <kpraveen.lkml@gmail.com>
> Reviewed-by: Dario Faggioli <dario.faggioli@citrix.com>
> 
Hey Geoge,

I don't see this patch in staging, nor I think you've commented on it.

IIRC, it was sent very close to feature freeze... So, is it possible
that it fell through some crack? :-)

Any thoughts about it? If not, what about applying? :-D

Thanks and Regards,
Dario

> ---
>  docs/misc/xen-command-line.markdown |  3 ++-
>  xen/common/sched_credit2.c          | 15 +++++++++++----
>  2 files changed, 13 insertions(+), 5 deletions(-)
> 
> diff --git a/docs/misc/xen-command-line.markdown b/docs/misc/xen-
> command-line.markdown
> index 5815d87dab..6e73766574 100644
> --- a/docs/misc/xen-command-line.markdown
> +++ b/docs/misc/xen-command-line.markdown
> @@ -525,7 +525,7 @@ also slow in responding to load changes.
>  The default value of `1 sec` is rather long.
>  
>  ### credit2\_runqueue
> -> `= core | socket | node | all`
> +> `= cpu | core | socket | node | all`
>  
>  > Default: `socket`
>  
> @@ -536,6 +536,7 @@ balancing (for instance, it will deal better with
> hyperthreading),
>  but also more overhead.
>  
>  Available alternatives, with their meaning, are:
> +* `cpu`: one runqueue per each logical pCPUs of the host;
>  * `core`: one runqueue per each physical core of the host;
>  * `socket`: one runqueue per each physical socket (which often,
>              but not always, matches a NUMA node) of the host;
> diff --git a/xen/common/sched_credit2.c b/xen/common/sched_credit2.c
> index bb1c657e76..ee7b443f9e 100644
> --- a/xen/common/sched_credit2.c
> +++ b/xen/common/sched_credit2.c
> @@ -301,6 +301,9 @@ integer_param("credit2_balance_over",
> opt_overload_balance_tolerance);
>   * want that to happen basing on topology. At the moment, it is
> possible
>   * to choose to arrange runqueues to be:
>   *
> + * - per-cpu: meaning that there will be one runqueue per logical
> cpu. This
> + *            will happen when if the opt_runqueue parameter is set
> to 'cpu'.
> + *
>   * - per-core: meaning that there will be one runqueue per each
> physical
>   *             core of the host. This will happen if the
> opt_runqueue
>   *             parameter is set to 'core';
> @@ -322,11 +325,13 @@ integer_param("credit2_balance_over",
> opt_overload_balance_tolerance);
>   * either the same physical core, the same physical socket, the same
> NUMA
>   * node, or just all of them, will be put together to form
> runqueues.
>   */
> -#define OPT_RUNQUEUE_CORE   0
> -#define OPT_RUNQUEUE_SOCKET 1
> -#define OPT_RUNQUEUE_NODE   2
> -#define OPT_RUNQUEUE_ALL    3
> +#define OPT_RUNQUEUE_CPU    0
> +#define OPT_RUNQUEUE_CORE   1
> +#define OPT_RUNQUEUE_SOCKET 2
> +#define OPT_RUNQUEUE_NODE   3
> +#define OPT_RUNQUEUE_ALL    4
>  static const char *const opt_runqueue_str[] = {
> +    [OPT_RUNQUEUE_CPU] = "cpu",
>      [OPT_RUNQUEUE_CORE] = "core",
>      [OPT_RUNQUEUE_SOCKET] = "socket",
>      [OPT_RUNQUEUE_NODE] = "node",
> @@ -682,6 +687,8 @@ cpu_to_runqueue(struct csched2_private *prv,
> unsigned int cpu)
>          BUG_ON(cpu_to_socket(cpu) == XEN_INVALID_SOCKET_ID ||
>                 cpu_to_socket(peer_cpu) == XEN_INVALID_SOCKET_ID);
>  
> +        if (opt_runqueue == OPT_RUNQUEUE_CPU)
> +            continue;
>          if ( opt_runqueue == OPT_RUNQUEUE_ALL ||
>               (opt_runqueue == OPT_RUNQUEUE_CORE &&
> same_core(peer_cpu, cpu)) ||
>               (opt_runqueue == OPT_RUNQUEUE_SOCKET &&
> same_socket(peer_cpu, cpu)) ||
-- 
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)

[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

[-- Attachment #2: Type: text/plain, Size: 127 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Re: [PATCH] xen: credit2: enable per cpu runqueue creation

[George Dunlap]

On 04/11/2017 05:15 PM, Praveen Kumar wrote:
> The patch introduces a new command line option 'cpu' that when used will create
> runqueue per logical pCPU. This may be useful for small systems, and also for
> development, performance evalution and comparison.
> 
> Signed-off-by: Praveen Kumar <kpraveen.lkml@gmail.com>
> Reviewed-by: Dario Faggioli <dario.faggioli@citrix.com>

OK, I've pushed this.  Sorry again for missing the 4.9 window.

 -George

> 
> ---
>  docs/misc/xen-command-line.markdown |  3 ++-
>  xen/common/sched_credit2.c          | 15 +++++++++++----
>  2 files changed, 13 insertions(+), 5 deletions(-)
> 
> diff --git a/docs/misc/xen-command-line.markdown b/docs/misc/xen-command-line.markdown
> index 5815d87dab..6e73766574 100644
> --- a/docs/misc/xen-command-line.markdown
> +++ b/docs/misc/xen-command-line.markdown
> @@ -525,7 +525,7 @@ also slow in responding to load changes.
>  The default value of `1 sec` is rather long.
>  
>  ### credit2\_runqueue
> -> `= core | socket | node | all`
> +> `= cpu | core | socket | node | all`
>  
>  > Default: `socket`
>  
> @@ -536,6 +536,7 @@ balancing (for instance, it will deal better with hyperthreading),
>  but also more overhead.
>  
>  Available alternatives, with their meaning, are:
> +* `cpu`: one runqueue per each logical pCPUs of the host;
>  * `core`: one runqueue per each physical core of the host;
>  * `socket`: one runqueue per each physical socket (which often,
>              but not always, matches a NUMA node) of the host;
> diff --git a/xen/common/sched_credit2.c b/xen/common/sched_credit2.c
> index bb1c657e76..ee7b443f9e 100644
> --- a/xen/common/sched_credit2.c
> +++ b/xen/common/sched_credit2.c
> @@ -301,6 +301,9 @@ integer_param("credit2_balance_over", opt_overload_balance_tolerance);
>   * want that to happen basing on topology. At the moment, it is possible
>   * to choose to arrange runqueues to be:
>   *
> + * - per-cpu: meaning that there will be one runqueue per logical cpu. This
> + *            will happen when if the opt_runqueue parameter is set to 'cpu'.
> + *
>   * - per-core: meaning that there will be one runqueue per each physical
>   *             core of the host. This will happen if the opt_runqueue
>   *             parameter is set to 'core';
> @@ -322,11 +325,13 @@ integer_param("credit2_balance_over", opt_overload_balance_tolerance);
>   * either the same physical core, the same physical socket, the same NUMA
>   * node, or just all of them, will be put together to form runqueues.
>   */
> -#define OPT_RUNQUEUE_CORE   0
> -#define OPT_RUNQUEUE_SOCKET 1
> -#define OPT_RUNQUEUE_NODE   2
> -#define OPT_RUNQUEUE_ALL    3
> +#define OPT_RUNQUEUE_CPU    0
> +#define OPT_RUNQUEUE_CORE   1
> +#define OPT_RUNQUEUE_SOCKET 2
> +#define OPT_RUNQUEUE_NODE   3
> +#define OPT_RUNQUEUE_ALL    4
>  static const char *const opt_runqueue_str[] = {
> +    [OPT_RUNQUEUE_CPU] = "cpu",
>      [OPT_RUNQUEUE_CORE] = "core",
>      [OPT_RUNQUEUE_SOCKET] = "socket",
>      [OPT_RUNQUEUE_NODE] = "node",
> @@ -682,6 +687,8 @@ cpu_to_runqueue(struct csched2_private *prv, unsigned int cpu)
>          BUG_ON(cpu_to_socket(cpu) == XEN_INVALID_SOCKET_ID ||
>                 cpu_to_socket(peer_cpu) == XEN_INVALID_SOCKET_ID);
>  
> +        if (opt_runqueue == OPT_RUNQUEUE_CPU)
> +            continue;
>          if ( opt_runqueue == OPT_RUNQUEUE_ALL ||
>               (opt_runqueue == OPT_RUNQUEUE_CORE && same_core(peer_cpu, cpu)) ||
>               (opt_runqueue == OPT_RUNQUEUE_SOCKET && same_socket(peer_cpu, cpu)) ||
> 


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel