[PATCH 1/6] kvm_stat: Only consider online cpus

From: Michael Ellerman <mpe@ellerman.id.au>
To: qemu-devel@nongnu.org
Cc: <kvm@vger.kernel.org>,
	alfs@linux.vnet.ibm.com, jfrei@linux.vnet.ibm.com, agraf@suse.de,
	jan.kiszka@siemens.com, pbonzini@redhat.com, aik@ozlabs.ru
Subject: [PATCH 1/6] kvm_stat: Only consider online cpus
Date: Tue, 17 Jun 2014 17:54:30 +1000	[thread overview]
Message-ID: <1402991675-24905-1-git-send-email-mpe@ellerman.id.au> (raw)

In kvm_stat we grovel through /sys to find out how many cpus are in the
system. However if a cpu is offline it will still be present in /sys,
and the perf_event_open() will fail.

Modify the logic to only return online cpus. We need to be careful on
systems which don't support cpu hotplug, the online file will not be
present at all.

Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
---
 scripts/kvm/kvm_stat | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)

diff --git a/scripts/kvm/kvm_stat b/scripts/kvm/kvm_stat
index d7e97e7..2a788bc 100755
--- a/scripts/kvm/kvm_stat
+++ b/scripts/kvm/kvm_stat
@@ -311,18 +311,30 @@ class TracepointProvider(object):
         self.select(fields)
     def fields(self):
         return self._fields
+
+    def _online_cpus(self):
+        l = []
+        pattern = r'cpu([0-9]+)'
+        basedir = '/sys/devices/system/cpu'
+        for entry in os.listdir(basedir):
+            match = re.match(pattern, entry)
+            if not match:
+                continue
+            path = os.path.join(basedir, entry, 'online')
+            if os.path.exists(path) and open(path).read().strip() != '1':
+                continue
+            l.append(int(match.group(1)))
+        return l
+
     def _setup(self, _fields):
         self._fields = _fields
-        cpure = r'cpu([0-9]+)'
-        self.cpus = [int(re.match(cpure, x).group(1))
-                     for x in os.listdir('/sys/devices/system/cpu')
-                     if re.match(cpure, x)]
+        cpus = self._online_cpus()
         import resource
-        nfiles = len(self.cpus) * 1000
+        nfiles = len(cpus) * 1000
         resource.setrlimit(resource.RLIMIT_NOFILE, (nfiles, nfiles))
         events = []
         self.group_leaders = []
-        for cpu in self.cpus:
+        for cpu in cpus:
             group = Group(cpu)
             for name in _fields:
                 tracepoint = name
-- 
1.9.1

