diff --git a/profiles/latency-performance/tuned.conf b/profiles/latency-performance/tuned.conf

index 054c554..4cd69a1 100644

--- a/profiles/latency-performance/tuned.conf

+++ b/profiles/latency-performance/tuned.conf

@@ -6,7 +6,7 @@

 summary=Optimize for deterministic performance at the cost of increased power consumption

 [cpu]

-force_latency=1

+force_latency=cstate.id:1|1

 governor=performance

 energy_perf_bias=performance

 min_perf_pct=100

diff --git a/profiles/sap-hana/tuned.conf b/profiles/sap-hana/tuned.conf

index 9afac82..d74250b 100644

--- a/profiles/sap-hana/tuned.conf

+++ b/profiles/sap-hana/tuned.conf

@@ -7,7 +7,7 @@ summary=Optimize for SAP HANA

 include=throughput-performance

 [cpu]

-force_latency=70

+force_latency=cstate.id:3|70

 [vm]

 transparent_hugepages=never

diff --git a/profiles/virtual-host/tuned.conf b/profiles/virtual-host/tuned.conf

index ca493d6..8acb92d 100644

--- a/profiles/virtual-host/tuned.conf

+++ b/profiles/virtual-host/tuned.conf

@@ -18,4 +18,4 @@ kernel.sched_migration_cost_ns = 5000000

 [cpu]

 # Setting C3 state sleep mode/power savings

-force_latency=70

+force_latency=cstate.id:3|70

diff --git a/tuned/plugins/plugin_cpu.py b/tuned/plugins/plugin_cpu.py

index 77e9448..04c7b19 100644

--- a/tuned/plugins/plugin_cpu.py

+++ b/tuned/plugins/plugin_cpu.py

@@ -12,8 +12,7 @@ import procfs

 log = tuned.logs.get()

-# TODO: force_latency -> command

-#       intel_pstate

+cpuidle_states_path = "/sys/devices/system/cpu/cpu0/cpuidle"

 class CPULatencyPlugin(base.Plugin):

 	"""

@@ -210,13 +209,74 @@ class CPULatencyPlugin(base.Plugin):

 	def _instance_unapply_dynamic(self, instance, device):

 		pass

+	def _str2int(self, s):

+		try:

+			return int(s)

+		except (ValueError, TypeError):

+			return None

+

+	def _read_cstates_latency(self):

+		self.cstates_latency = {}

+		for d in os.listdir(cpuidle_states_path):

+			cstate_path = cpuidle_states_path + "/%s/" % d

+			name = self._cmd.read_file(cstate_path + "name", err_ret = None, no_error = True)

+			latency = self._cmd.read_file(cstate_path + "latency", err_ret = None, no_error = True)

+			if name is not None and latency is not None:

+				latency = self._str2int(latency)

+				if latency is not None:

+					self.cstates_latency[name.strip()] = latency

+

+	def _get_latency_by_cstate_name(self, name):

+		log.debug("getting latency for cstate with name '%s'" % name)

+		if self.cstates_latency is None:

+			log.debug("reading cstates latency table")

+			self._read_cstates_latency()

+		latency = self.cstates_latency.get(name, None)

+		log.debug("cstate name mapped to latency: %s" % str(latency))

+		return latency

+

+	def _get_latency_by_cstate_id(self, lid):

+		log.debug("getting latency for cstate with ID '%s'" % str(lid))

+		lid = self._str2int(lid)

+		if lid is None:

+			log.debug("cstate ID is invalid")

+			return None

+		latency_path = cpuidle_states_path + "/%s/latency" % ("state%d" % lid)

+		latency = self._str2int(self._cmd.read_file(latency_path, err_ret = None, no_error = True))

+		log.debug("cstate ID mapped to latency: %s" % str(latency))

+		return latency

+

+	def _parse_latency(self, latency):

+		self.cstates_latency = None

+		latencies = str(latency).split("|")

+		log.debug("parsing latency")

+		for latency in latencies:

+			try:

+				latency = int(latency)

+				log.debug("parsed directly specified latency value: %d" % latency)

+			except ValueError:

+				if latency[0:10] == "cstate.id:":

+					latency = self._get_latency_by_cstate_id(latency[10:])

+				elif latency[0:12] == "cstate.name:":

+					latency = self._get_latency_by_cstate_name(latency[12:])

+				else:

+					latency = None

+					log.debug("invalid latency specified: '%s'" % str(latency))

+			if latency is not None:

+				break

+		return latency

+

 	def _set_latency(self, latency):

-		latency = int(latency)

-		if self._has_pm_qos and self._latency != latency:

-			log.info("setting new cpu latency %d" % latency)

-			latency_bin = struct.pack("i", latency)

-			os.write(self._cpu_latency_fd, latency_bin)

-			self._latency = latency

+		latency = self._parse_latency(latency)

+		if self._has_pm_qos:

+			if latency is None:

+				log.error("unable to evaluate latency value (probably wrong settings in the 'cpu' section of current profile), disabling PM QoS")

+				self._has_pm_qos = False

+			elif self._latency != latency:

+				log.info("setting new cpu latency %d" % latency)

+				latency_bin = struct.pack("i", latency)

+				os.write(self._cpu_latency_fd, latency_bin)

+				self._latency = latency

 	def _get_available_governors(self, device):

 		return self._cmd.read_file("/sys/devices/system/cpu/%s/cpufreq/scaling_available_governors" % device).strip().split()