kvm: Add experimental support for a perf-based execution timer

Add support for using the CPU cycle counter instead of a normal POSIX
timer to generate timed exits to gem5. This should, in theory, provide
better resolution when requesting timer signals.

The perf-based timer requires a fairly recent kernel since it requires
a working PERF_EVENT_IOC_PERIOD ioctl. This ioctl has existed in the
kernel for a long time, but it used to be completely broken due to an
inverted match when the kernel copied things from user
space. Additionally, the ioctl does not change the sample period
correctly on all kernel versions which implement it. It is currently
only known to work reliably on kernel version 3.7 and above on ARM.

src/cpu/kvm/timer.cc

@@ -110,3 +110,37 @@ PosixKvmTimer::calcResolution()
 
     return resolution;
 }
+
+
+PerfKvmTimer::PerfKvmTimer(PerfKvmCounter &ctr,
+                           int signo, float hostFactor, Tick hostFreq)
+    : BaseKvmTimer(signo, hostFactor, hostFreq),
+      hwOverflow(ctr)
+{
+    hwOverflow.enableSignals(signo);
+}
+
+PerfKvmTimer::~PerfKvmTimer()
+{
+}
+
+void
+PerfKvmTimer::arm(Tick ticks)
+{
+    hwOverflow.period(hostCycles(ticks));
+    hwOverflow.refresh(1);
+}
+
+void
+PerfKvmTimer::disarm()
+{
+    hwOverflow.stop();
+}
+
+Tick
+PerfKvmTimer::calcResolution()
+{
+    // This is a bit arbitrary, but in practice, we can't really do
+    // anything useful in less than ~1000 anyway.
+    return ticksFromHostCycles(1000);
+}