perf, x86: Provide means for disabling userspace RDPMC

Allow the disabling of RDPMC via a pmu specific attribute:

  echo 0 > /sys/bus/event_source/devices/cpu/rdpmc

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Arun Sharma <asharma@fb.com>
Link: http://lkml.kernel.org/n/tip-pqeog465zo5hsimtkfz73f27@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>

4 files changed, 64 insertions, 1 deletions

diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 53b569910175..116b040a73a8 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -24,6 +24,7 @@
 #include <linux/slab.h>
 #include <linux/cpu.h>
 #include <linux/bitops.h>
+#include <linux/device.h>
 
 #include <asm/apic.h>
 #include <asm/stacktrace.h>
@@ -1210,7 +1211,8 @@ x86_pmu_notifier(struct notifier_block *self, unsigned long action, void *hcpu)
                 break;
 
         case CPU_STARTING:
-                set_in_cr4(X86_CR4_PCE);
+                if (x86_pmu.attr_rdpmc)
+                        set_in_cr4(X86_CR4_PCE);
                 if (x86_pmu.cpu_starting)
                         x86_pmu.cpu_starting(cpu);
                 break;
@@ -1320,6 +1322,8 @@ static int __init init_hw_perf_events(void)
                 }
         }
 
+        x86_pmu.attr_rdpmc = 1; /* enable userspace RDPMC usage by default */
+
         pr_info("... version:                %d\n",     x86_pmu.version);
         pr_info("... bit width:              %d\n",     x86_pmu.cntval_bits);
         pr_info("... generic registers:      %d\n",     x86_pmu.num_counters);
@@ -1555,10 +1559,59 @@ static int x86_pmu_event_idx(struct perf_event *event)
         return idx + 1;
 }
 
+static ssize_t get_attr_rdpmc(struct device *cdev,
+                              struct device_attribute *attr,
+                              char *buf)
+{
+        return snprintf(buf, 40, "%d\n", x86_pmu.attr_rdpmc);
+}
+
+static void change_rdpmc(void *info)
+{
+        bool enable = !!(unsigned long)info;
+
+        if (enable)
+                set_in_cr4(X86_CR4_PCE);
+        else
+                clear_in_cr4(X86_CR4_PCE);
+}
+
+static ssize_t set_attr_rdpmc(struct device *cdev,
+                              struct device_attribute *attr,
+                              const char *buf, size_t count)
+{
+        unsigned long val = simple_strtoul(buf, NULL, 0);
+
+        if (!!val != !!x86_pmu.attr_rdpmc) {
+                x86_pmu.attr_rdpmc = !!val;
+                smp_call_function(change_rdpmc, (void *)val, 1);
+        }
+
+        return count;
+}
+
+static DEVICE_ATTR(rdpmc, S_IRUSR | S_IWUSR, get_attr_rdpmc, set_attr_rdpmc);
+
+static struct attribute *x86_pmu_attrs[] = {
+        &dev_attr_rdpmc.attr,
+        NULL,
+};
+
+static struct attribute_group x86_pmu_attr_group = {
+        .attrs = x86_pmu_attrs,
+};
+
+static const struct attribute_group *x86_pmu_attr_groups[] = {
+        &x86_pmu_attr_group,
+        NULL,
+};
+
 static struct pmu pmu = {
         .pmu_enable     = x86_pmu_enable,
         .pmu_disable    = x86_pmu_disable,
 
+        .attr_groups    = x86_pmu_attr_groups,
+
         .event_init     = x86_pmu_event_init,
 
         .add            = x86_pmu_add,
diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h
index 8944062f46e2..513d617b93c4 100644
--- a/arch/x86/kernel/cpu/perf_event.h
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -307,6 +307,14 @@ struct x86_pmu {
         struct x86_pmu_quirk *quirks;
         int             perfctr_second_write;
 
+        /*
+         * sysfs attrs
+         */
+        int             attr_rdpmc;
+
+        /*
+         * CPU Hotplug hooks
+         */
         int             (*cpu_prepare)(int cpu);
         void            (*cpu_starting)(int cpu);
         void            (*cpu_dying)(int cpu);
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 02545e6df95b..5311b79fe62c 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -615,6 +615,7 @@ struct pmu {
         struct list_head                entry;
 
         struct device                   *dev;
+        const struct attribute_group    **attr_groups;
         char                            *name;
         int                             type;
 
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 05affc3878ff..dcd4049e92fc 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -5505,6 +5505,7 @@ static int pmu_dev_alloc(struct pmu *pmu)
         if (!pmu->dev)
                 goto out;
 
+        pmu->dev->groups = pmu->attr_groups;
         device_initialize(pmu->dev);
         ret = dev_set_name(pmu->dev, "%s", pmu->name);
         if (ret)