[Qemu-devel] [PATCH v5 1/1] numa: Print warning if no node is assigned to a CPU

[Qemu-devel] [PATCH v5 1/1] numa: Print warning if no node is assigned to a CPU

[Eduardo Habkost]

We need all possible CPUs (including hotplug ones) to be present in the
SRAT when QEMU starts. QEMU already does that correctly today, the only
problem is that when a CPU is omitted from the NUMA configuration, it is
silently assigned to node 0.

Check if all CPUs up to max_cpus are present in the NUMA configuration
and warn about missing CPUs.

Make it just a warning, to allow management software to be updated if
necessary. In the future we may make it a fatal error instead.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
---
v1 -> v2: (no changes)

v2 -> v3:
 * Use enumerate_cpus() and error_report() for error message
 * Simplify logic using bitmap_full()

v3 -> v4:
 * Clarify error message, mention that all CPUs up to
   maxcpus need to be described in NUMA config

v4 -> v5:
 * Commit log update, to make problem description clearer
---
 numa.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/numa.c b/numa.c
index e922265..9b2ba05 100644
--- a/numa.c
+++ b/numa.c
@@ -200,6 +200,16 @@ static void validate_numa_cpus(void)
         bitmap_or(seen_cpus, seen_cpus,
                   numa_info[i].node_cpu, MAX_CPUMASK_BITS);
     }
+
+    if (!bitmap_full(seen_cpus, max_cpus)) {
+        char *msg;
+        bitmap_complement(seen_cpus, seen_cpus, max_cpus);
+        msg = enumerate_cpus(seen_cpus, max_cpus);
+        error_report("warning: CPU(s) not present in any NUMA nodes: %s", msg);
+        error_report("warning: All CPU(s) up to maxcpus should be described "
+                     "in NUMA config");
+        g_free(msg);
+    }
 }
 
 void parse_numa_opts(void)
-- 
2.1.0

Re: [Qemu-devel] [PATCH v5 1/1] numa: Print warning if no node is assigned to a CPU

[Eduardo Habkost]

On Wed, Mar 04, 2015 at 10:52:58AM -0300, Eduardo Habkost wrote:
> We need all possible CPUs (including hotplug ones) to be present in the
> SRAT when QEMU starts. QEMU already does that correctly today, the only
> problem is that when a CPU is omitted from the NUMA configuration, it is
> silently assigned to node 0.
> 
> Check if all CPUs up to max_cpus are present in the NUMA configuration
> and warn about missing CPUs.
> 
> Make it just a warning, to allow management software to be updated if
> necessary. In the future we may make it a fatal error instead.

Michael asked for a command-line example, so I will include this in the commit
log when queueing the patch:

Correct command-line, no warning:

  $ qemu-system-x86_64 -smp 2,maxcpus=4
  $ qemu-system-x86_64 -smp 2,maxcpus=4 -numa node,cpus=0-3

Incomplete command-line, with warnings:

  $ qemu-system-x86_64 -smp 2,maxcpus=4 -numa node,cpus=0
  qemu-system-x86_64: warning: CPU(s) not present in any NUMA nodes: 1 2 3
  qemu-system-x86_64: warning: All CPU(s) up to maxcpus should be described in NUMA config

  $ qemu-system-x86_64 -smp 2,maxcpus=4 -numa node,cpus=0-2
  qemu-system-x86_64: warning: CPU(s) not present in any NUMA nodes: 3
  qemu-system-x86_64: warning: All CPU(s) up to maxcpus should be described in NUMA config

> 
> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
> ---
> v1 -> v2: (no changes)
> 
> v2 -> v3:
>  * Use enumerate_cpus() and error_report() for error message
>  * Simplify logic using bitmap_full()
> 
> v3 -> v4:
>  * Clarify error message, mention that all CPUs up to
>    maxcpus need to be described in NUMA config
> 
> v4 -> v5:
>  * Commit log update, to make problem description clearer
> ---
>  numa.c | 10 ++++++++++
>  1 file changed, 10 insertions(+)
> 
> diff --git a/numa.c b/numa.c
> index e922265..9b2ba05 100644
> --- a/numa.c
> +++ b/numa.c
> @@ -200,6 +200,16 @@ static void validate_numa_cpus(void)
>          bitmap_or(seen_cpus, seen_cpus,
>                    numa_info[i].node_cpu, MAX_CPUMASK_BITS);
>      }
> +
> +    if (!bitmap_full(seen_cpus, max_cpus)) {
> +        char *msg;
> +        bitmap_complement(seen_cpus, seen_cpus, max_cpus);
> +        msg = enumerate_cpus(seen_cpus, max_cpus);
> +        error_report("warning: CPU(s) not present in any NUMA nodes: %s", msg);
> +        error_report("warning: All CPU(s) up to maxcpus should be described "
> +                     "in NUMA config");
> +        g_free(msg);
> +    }
>  }
>  
>  void parse_numa_opts(void)
> -- 
> 2.1.0
> 

-- 
Eduardo

[Qemu-devel] [PATCH v5 0/1] NUMA: Validate CPU configuration

[Eduardo Habkost]

This adds extra checks to the NUMA code to make sure the CPU configuration is
consistent. This needs to be applied on top of my numa patch queue:

  https://github.com/ehabkost/qemu.git numa

Changes v1 -> v2:
 * (none, v1 was tagged by accident and never sent to qemu-devel)

Changes v2 -> v3:
 * Fix off-by-one error on CPU index check
 * Use GString and error_report() instead of calling fprintf() directly
 * Simplify logic of the CPUs-not-present check

Changes v3 -> v4:
 * Patches 1-3 from v3 are now in the numa queue (URL above)
 * Clarify error message when some CPUs are missing from the config

Changes v4 -> v5:
 * Commit log update, to make problem description clearer

Eduardo Habkost (1):
  numa: Print warning if no node is assigned to a CPU

 numa.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

-- 
2.1.0

[Qemu-devel] [PATCH v5 1/1] numa: Print warning if no node is assigned to a CPU

[Eduardo Habkost]

We need all possible CPUs (including hotplug ones) to be present in the
SRAT when QEMU starts. QEMU already does that correctly today, the only
problem is that when a CPU is omitted from the NUMA configuration, it is
silently assigned to node 0.

Check if all CPUs up to max_cpus are present in the NUMA configuration
and warn about missing CPUs.

Make it just a warning, to allow management software to be updated if
necessary. In the future we may make it a fatal error instead.

Command-line examples:

* Correct, no warning:

  $ qemu-system-x86_64 -smp 2,maxcpus=4
  $ qemu-system-x86_64 -smp 2,maxcpus=4 -numa node,cpus=0-3

* Incomplete, with warnings:

  $ qemu-system-x86_64 -smp 2,maxcpus=4 -numa node,cpus=0
  qemu-system-x86_64: warning: CPU(s) not present in any NUMA nodes: 1 2 3
  qemu-system-x86_64: warning: All CPU(s) up to maxcpus should be described in NUMA config

  $ qemu-system-x86_64 -smp 2,maxcpus=4 -numa node,cpus=0-2
  qemu-system-x86_64: warning: CPU(s) not present in any NUMA nodes: 3
  qemu-system-x86_64: warning: All CPU(s) up to maxcpus should be described in NUMA config

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
---
v1 -> v2: (no changes)

v2 -> v3:
 * Use enumerate_cpus() and error_report() for error message
 * Simplify logic using bitmap_full()

v3 -> v4:
 * Clarify error message, mention that all CPUs up to
   maxcpus need to be described in NUMA config

v4 -> v5:
 * Commit log update, to make problem description clearer
---
 numa.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/numa.c b/numa.c
index fe74e1e..c975fb2 100644
--- a/numa.c
+++ b/numa.c
@@ -200,6 +200,16 @@ static void validate_numa_cpus(void)
         bitmap_or(seen_cpus, seen_cpus,
                   numa_info[i].node_cpu, MAX_CPUMASK_BITS);
     }
+
+    if (!bitmap_full(seen_cpus, max_cpus)) {
+        char *msg;
+        bitmap_complement(seen_cpus, seen_cpus, max_cpus);
+        msg = enumerate_cpus(seen_cpus, max_cpus);
+        error_report("warning: CPU(s) not present in any NUMA nodes: %s", msg);
+        error_report("warning: All CPU(s) up to maxcpus should be described "
+                     "in NUMA config");
+        g_free(msg);
+    }
 }
 
 void parse_numa_opts(MachineClass *mc)
-- 
2.1.0