Merge back PM tools material for v4.13.

3 files changed, 1500 insertions, 272 deletions

diff --git a/tools/power/x86/x86_energy_perf_policy/Makefile b/tools/power/x86/x86_energy_perf_policy/Makefile
index 971c9ffdcb50..a711eec0c895 100644
--- a/tools/power/x86/x86_energy_perf_policy/Makefile
+++ b/tools/power/x86/x86_energy_perf_policy/Makefile
@@ -1,10 +1,27 @@
-DESTDIR ?=
+CC		= $(CROSS_COMPILE)gcc
+BUILD_OUTPUT    := $(CURDIR)
+PREFIX		:= /usr
+DESTDIR		:=
+
+ifeq ("$(origin O)", "command line")
+	BUILD_OUTPUT := $(O)
+endif
 
 x86_energy_perf_policy : x86_energy_perf_policy.c
+CFLAGS +=	-Wall
+CFLAGS +=	-DMSRHEADER='"../../../../arch/x86/include/asm/msr-index.h"'
+
+%: %.c
+	@mkdir -p $(BUILD_OUTPUT)
+	$(CC) $(CFLAGS) $< -o $(BUILD_OUTPUT)/$@
 
+.PHONY : clean
 clean :
-	rm -f x86_energy_perf_policy
+	@rm -f $(BUILD_OUTPUT)/x86_energy_perf_policy
+
+install : x86_energy_perf_policy
+	install -d  $(DESTDIR)$(PREFIX)/bin
+	install $(BUILD_OUTPUT)/x86_energy_perf_policy $(DESTDIR)$(PREFIX)/bin/x86_energy_perf_policy
+	install -d  $(DESTDIR)$(PREFIX)/share/man/man8
+	install x86_energy_perf_policy.8 $(DESTDIR)$(PREFIX)/share/man/man8
 
-install :
-	install x86_energy_perf_policy ${DESTDIR}/usr/bin/
-	install x86_energy_perf_policy.8 ${DESTDIR}/usr/share/man/man8/
diff --git a/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.8 b/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.8
index 8eaaad648cdb..17db1c3af4d0 100644
--- a/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.8
+++ b/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.8
@@ -1,104 +1,213 @@
-.\"  This page Copyright (C) 2010 Len Brown <len.brown@intel.com>
+.\"  This page Copyright (C) 2010 - 2015 Len Brown <len.brown@intel.com>
 .\"  Distributed under the GPL, Copyleft 1994.
 .TH X86_ENERGY_PERF_POLICY 8
 .SH NAME
-x86_energy_perf_policy \- read or write MSR_IA32_ENERGY_PERF_BIAS
+x86_energy_perf_policy \- Manage Energy vs. Performance Policy via x86 Model Specific Registers
 .SH SYNOPSIS
-.ft B
 .B x86_energy_perf_policy
-.RB [ "\-c cpu" ]
-.RB [ "\-v" ]
-.RB "\-r"
+.RB "[ options ] [ scope ] [field \ value]"
 .br
-.B x86_energy_perf_policy
-.RB [ "\-c cpu" ]
-.RB [ "\-v" ]
-.RB 'performance'
+.RB "scope: \-\-cpu\ cpu-list | \-\-pkg\ pkg-list"
 .br
-.B x86_energy_perf_policy
-.RB [ "\-c cpu" ]
-.RB [ "\-v" ]
-.RB 'normal'
+.RB "cpu-list, pkg-list: # | #,# | #-# | all"
 .br
-.B x86_energy_perf_policy
-.RB [ "\-c cpu" ]
-.RB [ "\-v" ]
-.RB 'powersave'
+.RB "field: \-\-all | \-\-epb | \-\-hwp-epp | \-\-hwp-min | \-\-hwp-max | \-\-hwp-desired"
 .br
-.B x86_energy_perf_policy
-.RB [ "\-c cpu" ]
-.RB [ "\-v" ]
-.RB n
+.RB "other: (\-\-force | \-\-hwp-enable | \-\-turbo-enable)  value)"
 .br
+.RB "value: # | default | performance | balance-performance | balance-power | power"
 .SH DESCRIPTION
 \fBx86_energy_perf_policy\fP
-allows software to convey
-its policy for the relative importance of performance
-versus energy savings to the processor.
+displays and updates energy-performance policy settings specific to
+Intel Architecture Processors.  Settings are accessed via Model Specific Register (MSR)
+updates, no matter if the Linux cpufreq sub-system is enabled or not.
 
-The processor uses this information in model-specific ways
-when it must select trade-offs between performance and
-energy efficiency.
+Policy in MSR_IA32_ENERGY_PERF_BIAS (EPB)
+may affect a wide range of hardware decisions,
+such as how aggressively the hardware enters and exits CPU idle states (C-states)
+and Processor Performance States (P-states).
+This policy hint does not replace explicit OS C-state and P-state selection.
+Rather, it tells the hardware how aggressively to implement those selections.
+Further, it allows the OS to influence energy/performance trade-offs where there
+is no software interface, such as in the opportunistic "turbo-mode" P-state range.
+Note that MSR_IA32_ENERGY_PERF_BIAS is defined per CPU,
+but some implementations
+share a single MSR among all CPUs in each processor package.
+On those systems, a write to EPB on one processor will
+be visible, and will have an effect, on all CPUs
+in the same processor package.
 
-This policy hint does not supersede Processor Performance states
-(P-states) or CPU Idle power states (C-states), but allows
-software to have influence where it would otherwise be unable
-to express a preference.
+Hardware P-States (HWP) are effectively an expansion of hardware
+P-state control from the opportunistic turbo-mode P-state range
+to include the entire range of available P-states.
+On Broadwell Xeon, the initial HWP implementation, EBP influenced HWP.
+That influence was removed in subsequent generations,
+where it was moved to the
+Energy_Performance_Preference (EPP) field in
+a pair of dedicated MSRs -- MSR_IA32_HWP_REQUEST and MSR_IA32_HWP_REQUEST_PKG.
 
-For example, this setting may tell the hardware how
-aggressively or conservatively to control frequency
-in the "turbo range" above the explicitly OS-controlled
-P-state frequency range.  It may also tell the hardware
-how aggressively is should enter the OS requested C-states.
+EPP is the most commonly managed knob in HWP mode,
+but MSR_IA32_HWP_REQUEST also allows the user to specify
+minimum-frequency for Quality-of-Service,
+and maximum-frequency for power-capping.
+MSR_IA32_HWP_REQUEST is defined per-CPU.
 
-Support for this feature is indicated by CPUID.06H.ECX.bit3
-per the Intel Architectures Software Developer's Manual.
+MSR_IA32_HWP_REQUEST_PKG has the same capability as MSR_IA32_HWP_REQUEST,
+but it can simultaneously set the default policy for all CPUs within a package.
+A bit in per-CPU MSR_IA32_HWP_REQUEST indicates whether it is
+over-ruled-by or exempt-from MSR_IA32_HWP_REQUEST_PKG.
 
-.SS Options
-\fB-c\fP limits operation to a single CPU.
-The default is to operate on all CPUs.
-Note that MSR_IA32_ENERGY_PERF_BIAS is defined per
-logical processor, but that the initial implementations
-of the MSR were shared among all processors in each package.
-.PP
-\fB-v\fP increases verbosity.  By default
-x86_energy_perf_policy is silent.
-.PP
-\fB-r\fP is for "read-only" mode - the unchanged state
-is read and displayed.
+MSR_HWP_CAPABILITIES shows the default values for the fields
+in MSR_IA32_HWP_REQUEST.  It is displayed when no values
+are being written.
+
+.SS SCOPE OPTIONS
 .PP
-.I performance
-Set a policy where performance is paramount.
-The processor will be unwilling to sacrifice any performance
-for the sake of energy saving. This is the hardware default.
+\fB-c, --cpu\fP Operate on the MSR_IA32_HWP_REQUEST for each CPU in a CPU-list.
+The CPU-list may be comma-separated CPU numbers, with dash for range
+or the string "all".  Eg. '--cpu 1,4,6-8' or '--cpu all'.
+When --cpu is used, \fB--hwp-use-pkg\fP is available, which specifies whether the per-cpu
+MSR_IA32_HWP_REQUEST should be over-ruled by MSR_IA32_HWP_REQUEST_PKG (1),
+or exempt from MSR_IA32_HWP_REQUEST_PKG (0).
+
+\fB-p, --pkg\fP Operate on the MSR_IA32_HWP_REQUEST_PKG for each package in the package-list.
+The list is a string of individual package numbers separated
+by commas, and or ranges of package numbers separated by a dash,
+or the string "all".
+For example '--pkg 1,3' or '--pkg all'
+
+.SS VALUE OPTIONS
 .PP
-.I normal
+.I normal | default
 Set a policy with a normal balance between performance and energy efficiency.
 The processor will tolerate minor performance compromise
 for potentially significant energy savings.
-This reasonable default for most desktops and servers.
+This is a reasonable default for most desktops and servers.
+"default" is a synonym for "normal".
 .PP
-.I powersave
+.I performance
+Set a policy for maximum performance,
+accepting no performance sacrifice for the benefit of energy efficiency.
+.PP
+.I balance-performance
+Set a policy with a high priority on performance,
+but allowing some performance loss to benefit energy efficiency.
+.PP
+.I balance-power
+Set a policy where the performance and power are balanced.
+This is the default.
+.PP
+.I power
 Set a policy where the processor can accept
-a measurable performance hit to maximize energy efficiency.
+a measurable performance impact to maximize energy efficiency.
+
 .PP
-.I n
-Set MSR_IA32_ENERGY_PERF_BIAS to the specified number.
-The range of valid numbers is 0-15, where 0 is maximum
-performance and 15 is maximum energy efficiency.
+The following table shows the mapping from the value strings above to actual MSR values.
+This mapping is defined in the Linux-kernel header, msr-index.h.
 
+.nf
+VALUE STRING      	EPB	EPP
+performance       	0	0
+balance-performance	4	128
+normal, default		6	128
+balance-power	 	8	192
+power       		15	255
+.fi
+.PP
+For MSR_IA32_HWP_REQUEST performance fields
+(--hwp-min, --hwp-max, --hwp-desired), the value option
+is in units of 100 MHz, Eg. 12 signifies 1200 MHz.
+
+.SS FIELD OPTIONS
+\fB-a, --all value-string\fP Sets all EPB and EPP and HWP limit fields to the value associated with
+the value-string.  In addition, enables turbo-mode and HWP-mode, if they were previous disabled.
+Thus "--all normal" will set a system without cpufreq into a well known configuration.
+.PP
+\fB-B, --epb\fP set EPB per-core or per-package.
+See value strings in the table above.
+.PP
+\fB-d, --debug\fP debug increases verbosity.  By default
+x86_energy_perf_policy is silent for updates,
+and verbose for read-only mode.
+.PP
+\fB-P, --hwp-epp\fP set HWP.EPP per-core or per-package.
+See value strings in the table above.
+.PP
+\fB-m, --hwp-min\fP request HWP to not go below the specified core/bus ratio.
+The "default" is the value found in IA32_HWP_CAPABILITIES.min.
+.PP
+\fB-M, --hwp-max\fP request HWP not exceed a the specified core/bus ratio.
+The "default" is the value found in IA32_HWP_CAPABILITIES.max.
+.PP
+\fB-D, --hwp-desired\fP request HWP 'desired' frequency.
+The "normal" setting is 0, which
+corresponds to 'full autonomous' HWP control.
+Non-zero performance values request a specific performance
+level on this processor, specified in multiples of 100 MHz.
+.PP
+\fB-w, --hwp-window\fP specify integer number of microsec
+in the sliding window that HWP uses to maintain average frequency.
+This parameter is meaningful only when the "desired" field above is non-zero.
+Default is 0, allowing the HW to choose.
+.SH OTHER OPTIONS
+.PP
+\fB-f, --force\fP writes the specified values without bounds checking.
+.PP
+\fB-U, --hwp-use-pkg\fP (0 | 1), when used in conjunction with --cpu,
+indicates whether the per-CPU MSR_IA32_HWP_REQUEST should be overruled (1)
+or exempt (0) from per-Package MSR_IA32_HWP_REQUEST_PKG settings.
+The default is exempt.
+.PP
+\fB-H, --hwp-enable\fP enable HardWare-P-state (HWP) mode.  Once enabled, system RESET is required to disable HWP mode.
+.PP
+\fB-t, --turbo-enable\fP enable (1) or disable (0) turbo mode.
+.PP
+\fB-v, --version\fP print version and exit.
+.PP
+If no request to change policy is made,
+the default behavior is to read
+and display the current system state,
+including the default capabilities.
+.SH WARNING
+.PP
+This utility writes directly to Model Specific Registers.
+There is no locking or coordination should this utility
+be used to modify HWP limit fields at the same time that
+intel_pstate's sysfs attributes access the same MSRs.
+.PP
+Note that --hwp-desired and --hwp-window are considered experimental.
+Future versions of Linux reserve the right to access these
+fields internally -- potentially conflicting with user-space access.
+.SH EXAMPLE
+.nf
+# sudo x86_energy_perf_policy
+cpu0: EPB 6
+cpu0: HWP_REQ: min 6 max 35 des 0 epp 128 window 0x0 (0*10^0us) use_pkg 0
+cpu0: HWP_CAP: low 1 eff 8 guar 27 high 35
+cpu1: EPB 6
+cpu1: HWP_REQ: min 6 max 35 des 0 epp 128 window 0x0 (0*10^0us) use_pkg 0
+cpu1: HWP_CAP: low 1 eff 8 guar 27 high 35
+cpu2: EPB 6
+cpu2: HWP_REQ: min 6 max 35 des 0 epp 128 window 0x0 (0*10^0us) use_pkg 0
+cpu2: HWP_CAP: low 1 eff 8 guar 27 high 35
+cpu3: EPB 6
+cpu3: HWP_REQ: min 6 max 35 des 0 epp 128 window 0x0 (0*10^0us) use_pkg 0
+cpu3: HWP_CAP: low 1 eff 8 guar 27 high 35
+.fi
 .SH NOTES
-.B "x86_energy_perf_policy "
+.B "x86_energy_perf_policy"
 runs only as root.
 .SH FILES
 .ta
 .nf
 /dev/cpu/*/msr
 .fi
-
 .SH "SEE ALSO"
+.nf
 msr(4)
+Intel(R) 64 and IA-32 Architectures Software Developer's Manual
+.fi
 .PP
 .SH AUTHORS
 .nf
-Written by Len Brown <len.brown@intel.com>
+Len Brown
diff --git a/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c b/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c
index 40b3e5482f8a..65bbe627a425 100644
--- a/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c
+++ b/tools/power/x86/x86_energy_perf_policy/x86_energy_perf_policy.c
@@ -3,322 +3,1424 @@
  * policy preference bias on recent X86 processors.
  */
 /*
- * Copyright (c) 2010, Intel Corporation.
+ * Copyright (c) 2010 - 2017 Intel Corporation.
  * Len Brown <len.brown@intel.com>
  *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms and conditions of the GNU General Public License,
- * version 2, as published by the Free Software Foundation.
- *
- * This program is distributed in the hope it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program; if not, write to the Free Software Foundation, Inc.,
- * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ * This program is released under GPL v2
  */
 
+#define _GNU_SOURCE
+#include MSRHEADER
 #include <stdio.h>
 #include <unistd.h>
 #include <sys/types.h>
+#include <sched.h>
 #include <sys/stat.h>
 #include <sys/resource.h>
+#include <getopt.h>
+#include <err.h>
 #include <fcntl.h>
 #include <signal.h>
 #include <sys/time.h>
+#include <limits.h>
 #include <stdlib.h>
 #include <string.h>
+#include <cpuid.h>
+#include <errno.h>
+
+#define	OPTARG_NORMAL			(INT_MAX - 1)
+#define	OPTARG_POWER			(INT_MAX - 2)
+#define	OPTARG_BALANCE_POWER		(INT_MAX - 3)
+#define	OPTARG_BALANCE_PERFORMANCE	(INT_MAX - 4)
+#define	OPTARG_PERFORMANCE		(INT_MAX - 5)
+
+struct msr_hwp_cap {
+	unsigned char highest;
+	unsigned char guaranteed;
+	unsigned char efficient;
+	unsigned char lowest;
+};
 
-unsigned int verbose;		/* set with -v */
-unsigned int read_only;		/* set with -r */
+struct msr_hwp_request {
+	unsigned char hwp_min;
+	unsigned char hwp_max;
+	unsigned char hwp_desired;
+	unsigned char hwp_epp;
+	unsigned int hwp_window;
+	unsigned char hwp_use_pkg;
+} req_update;
+
+unsigned int debug;
+unsigned int verbose;
+unsigned int force;
 char *progname;
-unsigned long long new_bias;
-int cpu = -1;
+int base_cpu;
+unsigned char update_epb;
+unsigned long long new_epb;
+unsigned char turbo_is_enabled;
+unsigned char update_turbo;
+unsigned char turbo_update_value;
+unsigned char update_hwp_epp;
+unsigned char update_hwp_min;
+unsigned char update_hwp_max;
+unsigned char update_hwp_desired;
+unsigned char update_hwp_window;
+unsigned char update_hwp_use_pkg;
+unsigned char update_hwp_enable;
+#define hwp_update_enabled() (update_hwp_enable | update_hwp_epp | update_hwp_max | update_hwp_min | update_hwp_desired | update_hwp_window | update_hwp_use_pkg)
+int max_cpu_num;
+int max_pkg_num;
+#define MAX_PACKAGES 64
+unsigned int first_cpu_in_pkg[MAX_PACKAGES];
+unsigned long long pkg_present_set;
+unsigned long long pkg_selected_set;
+cpu_set_t *cpu_present_set;
+cpu_set_t *cpu_selected_set;
+int genuine_intel;
+
+size_t cpu_setsize;
+
+char *proc_stat = "/proc/stat";
+
+unsigned int has_epb;	/* MSR_IA32_ENERGY_PERF_BIAS */
+unsigned int has_hwp;	/* IA32_PM_ENABLE, IA32_HWP_CAPABILITIES */
+			/* IA32_HWP_REQUEST, IA32_HWP_STATUS */
+unsigned int has_hwp_notify;		/* IA32_HWP_INTERRUPT */
+unsigned int has_hwp_activity_window;	/* IA32_HWP_REQUEST[bits 41:32] */
+unsigned int has_hwp_epp;	/* IA32_HWP_REQUEST[bits 31:24] */
+unsigned int has_hwp_request_pkg;	/* IA32_HWP_REQUEST_PKG */
+
+unsigned int bdx_highest_ratio;
 
 /*
- * Usage:
- *
- * -c cpu: limit action to a single CPU (default is all CPUs)
- * -v: verbose output (can invoke more than once)
- * -r: read-only, don't change any settings
- *
- *  performance
- *	Performance is paramount.
- *	Unwilling to sacrifice any performance
- *	for the sake of energy saving. (hardware default)
- *
- *  normal
- *	Can tolerate minor performance compromise
- *	for potentially significant energy savings.
- *	(reasonable default for most desktops and servers)
- *
- *  powersave
- *	Can tolerate significant performance hit
- *	to maximize energy savings.
- *
- * n
- *	a numerical value to write to the underlying MSR.
+ * maintain compatibility with original implementation, but don't document it:
  */
 void usage(void)
 {
-	printf("%s: [-c cpu] [-v] "
-		"(-r | 'performance' | 'normal' | 'powersave' | n)\n",
-		progname);
+	fprintf(stderr, "%s [options] [scope][field value]\n", progname);
+	fprintf(stderr, "scope: --cpu cpu-list [--hwp-use-pkg #] | --pkg pkg-list\n");
+	fprintf(stderr, "field: --all | --epb | --hwp-epp | --hwp-min | --hwp-max | --hwp-desired\n");
+	fprintf(stderr, "other: --hwp-enable | --turbo-enable (0 | 1) | --help | --force\n");
+	fprintf(stderr,
+		"value: ( # | \"normal\" | \"performance\" | \"balance-performance\" | \"balance-power\"| \"power\")\n");
+	fprintf(stderr, "--hwp-window usec\n");
+
+	fprintf(stderr, "Specify only Energy Performance BIAS (legacy usage):\n");
+	fprintf(stderr, "%s: [-c cpu] [-v] (-r | policy-value )\n", progname);
+
 	exit(1);
 }
 
-#define MSR_IA32_ENERGY_PERF_BIAS	0x000001b0
+/*
+ * If bdx_highest_ratio is set,
+ * then we must translate between MSR format and simple ratio
+ * used on the cmdline.
+ */
+int ratio_2_msr_perf(int ratio)
+{
+	int msr_perf;
+
+	if (!bdx_highest_ratio)
+		return ratio;
+
+	msr_perf = ratio * 255 / bdx_highest_ratio;
+
+	if (debug)
+		fprintf(stderr, "%d = ratio_to_msr_perf(%d)\n", msr_perf, ratio);
+
+	return msr_perf;
+}
+int msr_perf_2_ratio(int msr_perf)
+{
+	int ratio;
+	double d;
+
+	if (!bdx_highest_ratio)
+		return msr_perf;
+
+	d = (double)msr_perf * (double) bdx_highest_ratio / 255.0;
+	d = d + 0.5;	/* round */
+	ratio = (int)d;
+
+	if (debug)
+		fprintf(stderr, "%d = msr_perf_ratio(%d) {%f}\n", ratio, msr_perf, d);
+
+	return ratio;
+}
+int parse_cmdline_epb(int i)
+{
+	if (!has_epb)
+		errx(1, "EPB not enabled on this platform");
+
+	update_epb = 1;
+
+	switch (i) {
+	case OPTARG_POWER:
+		return ENERGY_PERF_BIAS_POWERSAVE;
+	case OPTARG_BALANCE_POWER:
+		return ENERGY_PERF_BIAS_BALANCE_POWERSAVE;
+	case OPTARG_NORMAL:
+		return ENERGY_PERF_BIAS_NORMAL;
+	case OPTARG_BALANCE_PERFORMANCE:
+		return ENERGY_PERF_BIAS_BALANCE_PERFORMANCE;
+	case OPTARG_PERFORMANCE:
+		return ENERGY_PERF_BIAS_PERFORMANCE;
+	}
+	if (i < 0 || i > ENERGY_PERF_BIAS_POWERSAVE)
+		errx(1, "--epb must be from 0 to 15");
+	return i;
+}
+
+#define HWP_CAP_LOWEST 0
+#define HWP_CAP_HIGHEST 255
+
+/*
+ * "performance" changes hwp_min to cap.highest
+ * All others leave it at cap.lowest
+ */
+int parse_cmdline_hwp_min(int i)
+{
+	update_hwp_min = 1;
+
+	switch (i) {
+	case OPTARG_POWER:
+	case OPTARG_BALANCE_POWER:
+	case OPTARG_NORMAL:
+	case OPTARG_BALANCE_PERFORMANCE:
+		return HWP_CAP_LOWEST;
+	case OPTARG_PERFORMANCE:
+		return HWP_CAP_HIGHEST;
+	}
+	return i;
+}
+/*
+ * "power" changes hwp_max to cap.lowest
+ * All others leave it at cap.highest
+ */
+int parse_cmdline_hwp_max(int i)
+{
+	update_hwp_max = 1;
+
+	switch (i) {
+	case OPTARG_POWER:
+		return HWP_CAP_LOWEST;
+	case OPTARG_NORMAL:
+	case OPTARG_BALANCE_POWER:
+	case OPTARG_BALANCE_PERFORMANCE:
+	case OPTARG_PERFORMANCE:
+		return HWP_CAP_HIGHEST;
+	}
+	return i;
+}
+/*
+ * for --hwp-des, all strings leave it in autonomous mode
+ * If you want to change it, you need to explicitly pick a value
+ */
+int parse_cmdline_hwp_desired(int i)
+{
+	update_hwp_desired = 1;
+
+	switch (i) {
+	case OPTARG_POWER:
+	case OPTARG_BALANCE_POWER:
+	case OPTARG_BALANCE_PERFORMANCE:
+	case OPTARG_NORMAL:
+	case OPTARG_PERFORMANCE:
+		return 0;	/* autonomous */
+	}
+	return i;
+}
+
+int parse_cmdline_hwp_window(int i)
+{
+	unsigned int exponent;
+
+	update_hwp_window = 1;
+
+	switch (i) {
+	case OPTARG_POWER:
+	case OPTARG_BALANCE_POWER:
+	case OPTARG_NORMAL:
+	case OPTARG_BALANCE_PERFORMANCE:
+	case OPTARG_PERFORMANCE:
+		return 0;
+	}
+	if (i < 0 || i > 1270000000) {
+		fprintf(stderr, "--hwp-window: 0 for auto; 1 - 1270000000 usec for window duration\n");
+		usage();
+	}
+	for (exponent = 0; ; ++exponent) {
+		if (debug)
+			printf("%d 10^%d\n", i, exponent);
+
+		if (i <= 127)
+			break;
+
+		i = i / 10;
+	}
+	if (debug)
+		fprintf(stderr, "%d*10^%d: 0x%x\n", i, exponent, (exponent << 7) | i);
+
+	return (exponent << 7) | i;
+}
+int parse_cmdline_hwp_epp(int i)
+{
+	update_hwp_epp = 1;
+
+	switch (i) {
+	case OPTARG_POWER:
+		return HWP_EPP_POWERSAVE;
+	case OPTARG_BALANCE_POWER:
+		return HWP_EPP_BALANCE_POWERSAVE;
+	case OPTARG_NORMAL:
+	case OPTARG_BALANCE_PERFORMANCE:
+		return HWP_EPP_BALANCE_PERFORMANCE;
+	case OPTARG_PERFORMANCE:
+		return HWP_EPP_PERFORMANCE;
+	}
+	if (i < 0 || i > 0xff) {
+		fprintf(stderr, "--hwp-epp must be from 0 to 0xff\n");
+		usage();
+	}
+	return i;
+}
+int parse_cmdline_turbo(int i)
+{
+	update_turbo = 1;
+
+	switch (i) {
+	case OPTARG_POWER:
+		return 0;
+	case OPTARG_NORMAL:
+	case OPTARG_BALANCE_POWER:
+	case OPTARG_BALANCE_PERFORMANCE:
+	case OPTARG_PERFORMANCE:
+		return 1;
+	}
+	if (i < 0 || i > 1) {
+		fprintf(stderr, "--turbo-enable: 1 to enable, 0 to disable\n");
+		usage();
+	}
+	return i;
+}
+
+int parse_optarg_string(char *s)
+{
+	int i;
+	char *endptr;
+
+	if (!strncmp(s, "default", 7))
+		return OPTARG_NORMAL;
+
+	if (!strncmp(s, "normal", 6))
+		return OPTARG_NORMAL;
+
+	if (!strncmp(s, "power", 9))
+		return OPTARG_POWER;
+
+	if (!strncmp(s, "balance-power", 17))
+		return OPTARG_BALANCE_POWER;
+
+	if (!strncmp(s, "balance-performance", 19))
+		return OPTARG_BALANCE_PERFORMANCE;
+
+	if (!strncmp(s, "performance", 11))
+		return OPTARG_PERFORMANCE;
+
+	i = strtol(s, &endptr, 0);
+	if (s == endptr) {
+		fprintf(stderr, "no digits in \"%s\"\n", s);
+		usage();
+	}
+	if (i == LONG_MIN || i == LONG_MAX)
+		errx(-1, "%s", s);
+
+	if (i > 0xFF)
+		errx(-1, "%d (0x%x) must be < 256", i, i);
+
+	if (i < 0)
+		errx(-1, "%d (0x%x) must be >= 0", i, i);
+	return i;
+}
+
+void parse_cmdline_all(char *s)
+{
+	force++;
+	update_hwp_enable = 1;
+	req_update.hwp_min = parse_cmdline_hwp_min(parse_optarg_string(s));
+	req_update.hwp_max = parse_cmdline_hwp_max(parse_optarg_string(s));
+	req_update.hwp_epp = parse_cmdline_hwp_epp(parse_optarg_string(s));
+	if (has_epb)
+		new_epb = parse_cmdline_epb(parse_optarg_string(s));
+	turbo_update_value = parse_cmdline_turbo(parse_optarg_string(s));
+	req_update.hwp_desired = parse_cmdline_hwp_desired(parse_optarg_string(s));
+	req_update.hwp_window = parse_cmdline_hwp_window(parse_optarg_string(s));
+}
+
+void validate_cpu_selected_set(void)
+{
+	int cpu;
+
+	if (CPU_COUNT_S(cpu_setsize, cpu_selected_set) == 0)
+		errx(0, "no CPUs requested");
+
+	for (cpu = 0; cpu <= max_cpu_num; ++cpu) {
+		if (CPU_ISSET_S(cpu, cpu_setsize, cpu_selected_set))
+			if (!CPU_ISSET_S(cpu, cpu_setsize, cpu_present_set))
+				errx(1, "Requested cpu% is not present", cpu);
+	}
+}
+
+void parse_cmdline_cpu(char *s)
+{
+	char *startp, *endp;
+	int cpu = 0;
+
+	if (pkg_selected_set) {
+		usage();
+		errx(1, "--cpu | --pkg");
+	}
+	cpu_selected_set = CPU_ALLOC((max_cpu_num + 1));
+	if (cpu_selected_set == NULL)
+		err(1, "cpu_selected_set");
+	CPU_ZERO_S(cpu_setsize, cpu_selected_set);
+
+	for (startp = s; startp && *startp;) {
+
+		if (*startp == ',') {
+			startp++;
+			continue;
+		}
+
+		if (*startp == '-') {
+			int end_cpu;
 
-#define	BIAS_PERFORMANCE		0
-#define BIAS_BALANCE			6
-#define	BIAS_POWERSAVE			15
+			startp++;
+			end_cpu = strtol(startp, &endp, 10);
+			if (startp == endp)
+				continue;
+
+			while (cpu <= end_cpu) {
+				if (cpu > max_cpu_num)
+					errx(1, "Requested cpu%d exceeds max cpu%d", cpu, max_cpu_num);
+				CPU_SET_S(cpu, cpu_setsize, cpu_selected_set);
+				cpu++;
+			}
+			startp = endp;
+			continue;
+		}
+
+		if (strncmp(startp, "all", 3) == 0) {
+			for (cpu = 0; cpu <= max_cpu_num; cpu += 1) {
+				if (CPU_ISSET_S(cpu, cpu_setsize, cpu_present_set))
+					CPU_SET_S(cpu, cpu_setsize, cpu_selected_set);
+			}
+			startp += 3;
+			if (*startp == 0)
+				break;
+		}
+		/* "--cpu even" is not documented */
+		if (strncmp(startp, "even", 4) == 0) {
+			for (cpu = 0; cpu <= max_cpu_num; cpu += 2) {
+				if (CPU_ISSET_S(cpu, cpu_setsize, cpu_present_set))
+					CPU_SET_S(cpu, cpu_setsize, cpu_selected_set);
+			}
+			startp += 4;
+			if (*startp == 0)
+				break;
+		}
+
+		/* "--cpu odd" is not documented */
+		if (strncmp(startp, "odd", 3) == 0) {
+			for (cpu = 1; cpu <= max_cpu_num; cpu += 2) {
+				if (CPU_ISSET_S(cpu, cpu_setsize, cpu_present_set))
+					CPU_SET_S(cpu, cpu_setsize, cpu_selected_set);
+			}
+			startp += 3;
+			if (*startp == 0)
+				break;
+		}
+
+		cpu = strtol(startp, &endp, 10);
+		if (startp == endp)
+			errx(1, "--cpu cpu-set: confused by '%s'", startp);
+		if (cpu > max_cpu_num)
+			errx(1, "Requested cpu%d exceeds max cpu%d", cpu, max_cpu_num);
+		CPU_SET_S(cpu, cpu_setsize, cpu_selected_set);
+		startp = endp;
+	}
+
+	validate_cpu_selected_set();
+
+}
+
+void parse_cmdline_pkg(char *s)
+{
+	char *startp, *endp;
+	int pkg = 0;
+
+	if (cpu_selected_set) {
+		usage();
+		errx(1, "--pkg | --cpu");
+	}
+	pkg_selected_set = 0;
+
+	for (startp = s; startp && *startp;) {
+
+		if (*startp == ',') {
+			startp++;
+			continue;
+		}
+
+		if (*startp == '-') {
+			int end_pkg;
+
+			startp++;
+			end_pkg = strtol(startp, &endp, 10);
+			if (startp == endp)
+				continue;
+
+			while (pkg <= end_pkg) {
+				if (pkg > max_pkg_num)
+					errx(1, "Requested pkg%d exceeds max pkg%d", pkg, max_pkg_num);
+				pkg_selected_set |= 1 << pkg;
+				pkg++;
+			}
+			startp = endp;
+			continue;
+		}
+
+		if (strncmp(startp, "all", 3) == 0) {
+			pkg_selected_set = pkg_present_set;
+			return;
+		}
+
+		pkg = strtol(startp, &endp, 10);
+		if (pkg > max_pkg_num)
+			errx(1, "Requested pkg%d Exceeds max pkg%d", pkg, max_pkg_num);
+		pkg_selected_set |= 1 << pkg;
+		startp = endp;
+	}
+}
+
+void for_packages(unsigned long long pkg_set, int (func)(int))
+{
+	int pkg_num;
+
+	for (pkg_num = 0; pkg_num <= max_pkg_num; ++pkg_num) {
+		if (pkg_set & (1UL << pkg_num))
+			func(pkg_num);
+	}
+}
+
+void print_version(void)
+{
+	printf("x86_energy_perf_policy 17.05.11 (C) Len Brown <len.brown@intel.com>\n");
+}
 
 void cmdline(int argc, char **argv)
 {
 	int opt;
+	int option_index = 0;
+
+	static struct option long_options[] = {
+		{"all",		required_argument,	0, 'a'},
+		{"cpu",		required_argument,	0, 'c'},
+		{"pkg",		required_argument,	0, 'p'},
+		{"debug",	no_argument,		0, 'd'},
+		{"hwp-desired",	required_argument,	0, 'D'},
+		{"epb",	required_argument,	0, 'B'},
+		{"force",	no_argument,	0, 'f'},
+		{"hwp-enable",	no_argument,	0, 'e'},
+		{"help",	no_argument,	0, 'h'},
+		{"hwp-epp",	required_argument,	0, 'P'},
+		{"hwp-min",	required_argument,	0, 'm'},
+		{"hwp-max",	required_argument,	0, 'M'},
+		{"read",	no_argument,		0, 'r'},
+		{"turbo-enable",	required_argument,	0, 't'},
+		{"hwp-use-pkg",	required_argument,	0, 'u'},
+		{"version",	no_argument,		0, 'v'},
+		{"hwp-window",	required_argument,	0, 'w'},
+		{0,		0,			0, 0 }
+	};
 
 	progname = argv[0];
 
-	while ((opt = getopt(argc, argv, "+rvc:")) != -1) {
+	while ((opt = getopt_long_only(argc, argv, "+a:c:dD:E:e:f:m:M:rt:u:vw",
+				long_options, &option_index)) != -1) {
 		switch (opt) {
+		case 'a':
+			parse_cmdline_all(optarg);
+			break;
+		case 'B':
+			new_epb = parse_cmdline_epb(parse_optarg_string(optarg));
+			break;
 		case 'c':
-			cpu = atoi(optarg);
+			parse_cmdline_cpu(optarg);
+			break;
+		case 'e':
+			update_hwp_enable = 1;
+			break;
+		case 'h':
+			usage();
+			break;
+		case 'd':
+			debug++;
+			verbose++;
+			break;
+		case 'f':
+			force++;
+			break;
+		case 'D':
+			req_update.hwp_desired = parse_cmdline_hwp_desired(parse_optarg_string(optarg));
+			break;
+		case 'm':
+			req_update.hwp_min = parse_cmdline_hwp_min(parse_optarg_string(optarg));
+			break;
+		case 'M':
+			req_update.hwp_max = parse_cmdline_hwp_max(parse_optarg_string(optarg));
+			break;
+		case 'p':
+			parse_cmdline_pkg(optarg);
+			break;
+		case 'P':
+			req_update.hwp_epp = parse_cmdline_hwp_epp(parse_optarg_string(optarg));
 			break;
 		case 'r':
-			read_only = 1;
+			/* v1 used -r to specify read-only mode, now the default */
+			break;
+		case 't':
+			turbo_update_value = parse_cmdline_turbo(parse_optarg_string(optarg));
+			break;
+		case 'u':
+			update_hwp_use_pkg++;
+			if (atoi(optarg) == 0)
+				req_update.hwp_use_pkg = 0;
+			else
+				req_update.hwp_use_pkg = 1;
 			break;
 		case 'v':
-			verbose++;
+			print_version();
+			exit(0);
+			break;
+		case 'w':
+			req_update.hwp_window = parse_cmdline_hwp_window(parse_optarg_string(optarg));
 			break;
 		default:
 			usage();
 		}
 	}
-	/* if -r, then should be no additional optind */
-	if (read_only && (argc > optind))
-		usage();
-
 	/*
-	 * if no -r , then must be one additional optind
+	 * v1 allowed "performance"|"normal"|"power" with no policy specifier
+	 * to update BIAS.  Continue to support that, even though no longer documented.
 	 */
-	if (!read_only) {
+	if (argc == optind + 1)
+		new_epb = parse_cmdline_epb(parse_optarg_string(argv[optind]));
 
-		if (argc != optind + 1) {
-			printf("must supply -r or policy param\n");
-			usage();
-			}
+	if (argc > optind + 1) {
+		fprintf(stderr, "stray parameter '%s'\n", argv[optind + 1]);
+		usage();
+	}
+}
 
-		if (!strcmp("performance", argv[optind])) {
-			new_bias = BIAS_PERFORMANCE;
-		} else if (!strcmp("normal", argv[optind])) {
-			new_bias = BIAS_BALANCE;
-		} else if (!strcmp("powersave", argv[optind])) {
-			new_bias = BIAS_POWERSAVE;
-		} else {
-			char *endptr;
-
-			new_bias = strtoull(argv[optind], &endptr, 0);
-			if (endptr == argv[optind] ||
-				new_bias > BIAS_POWERSAVE) {
-					fprintf(stderr, "invalid value: %s\n",
-						argv[optind]);
-				usage();
-			}
-		}
+
+int get_msr(int cpu, int offset, unsigned long long *msr)
+{
+	int retval;
+	char pathname[32];
+	int fd;
+
+	sprintf(pathname, "/dev/cpu/%d/msr", cpu);
+	fd = open(pathname, O_RDONLY);
+	if (fd < 0)
+		err(-1, "%s open failed, try chown or chmod +r /dev/cpu/*/msr, or run as root", pathname);
+
+	retval = pread(fd, msr, sizeof(*msr), offset);
+	if (retval != sizeof(*msr))
+		err(-1, "%s offset 0x%llx read failed", pathname, (unsigned long long)offset);
+
+	if (debug > 1)
+		fprintf(stderr, "get_msr(cpu%d, 0x%X, 0x%llX)\n", cpu, offset, *msr);
+
+	close(fd);
+	return 0;
+}
+
+int put_msr(int cpu, int offset, unsigned long long new_msr)
+{
+	char pathname[32];
+	int retval;
+	int fd;
+
+	sprintf(pathname, "/dev/cpu/%d/msr", cpu);
+	fd = open(pathname, O_RDWR);
+	if (fd < 0)
+		err(-1, "%s open failed, try chown or chmod +r /dev/cpu/*/msr, or run as root", pathname);
+
+	retval = pwrite(fd, &new_msr, sizeof(new_msr), offset);
+	if (retval != sizeof(new_msr))
+		err(-2, "pwrite(cpu%d, offset 0x%x, 0x%llx) = %d", cpu, offset, new_msr, retval);
+
+	close(fd);
+
+	if (debug > 1)
+		fprintf(stderr, "put_msr(cpu%d, 0x%X, 0x%llX)\n", cpu, offset, new_msr);
+
+	return 0;
+}
+
+void print_hwp_cap(int cpu, struct msr_hwp_cap *cap, char *str)
+{
+	if (cpu != -1)
+		printf("cpu%d: ", cpu);
+
+	printf("HWP_CAP: low %d eff %d guar %d high %d\n",
+		cap->lowest, cap->efficient, cap->guaranteed, cap->highest);
+}
+void read_hwp_cap(int cpu, struct msr_hwp_cap *cap, unsigned int msr_offset)
+{
+	unsigned long long msr;
+
+	get_msr(cpu, msr_offset, &msr);
+
+	cap->highest = msr_perf_2_ratio(HWP_HIGHEST_PERF(msr));
+	cap->guaranteed = msr_perf_2_ratio(HWP_GUARANTEED_PERF(msr));
+	cap->efficient = msr_perf_2_ratio(HWP_MOSTEFFICIENT_PERF(msr));
+	cap->lowest = msr_perf_2_ratio(HWP_LOWEST_PERF(msr));
+}
+
+void print_hwp_request(int cpu, struct msr_hwp_request *h, char *str)
+{
+	if (cpu != -1)
+		printf("cpu%d: ", cpu);
+
+	if (str)
+		printf("%s", str);
+
+	printf("HWP_REQ: min %d max %d des %d epp %d window 0x%x (%d*10^%dus) use_pkg %d\n",
+		h->hwp_min, h->hwp_max, h->hwp_desired, h->hwp_epp,
+		h->hwp_window, h->hwp_window & 0x7F, (h->hwp_window >> 7) & 0x7, h->hwp_use_pkg);
+}
+void print_hwp_request_pkg(int pkg, struct msr_hwp_r