When --add was used more than once, overflowed buffers
caused some counters to be stored on top of others,
corrupting the results.  Simplify the code by simply
reserving space for up to 16 added counters per each
cpu, core, package.

Per-cpu added counters were being printed only per-core.
Signed-off-by: NLen Brown <len.brown@intel.com>

The new --add option has replaced the -M, -m, -C, -c options
Eg.

-M 0x10 is now --add msr0x10,raw
-m 0x10 is now --add msr0x10,raw,u32
-C 0x10 is now --add msr0x10,delta
-c 0x10 is now --add msr0x10,delta,u32

The --add option can be repeated to add any number of counters,
while the previous options were limited to adding one of each type.

In addition, the --add option can accept a column label,
and can also display a counter as a percentage of elapsed cycles.

Eg. --add msr0x3fe,core,percent,MY_CC3
Signed-off-by: NLen Brown <len.brown@intel.com>

Create the "--add" parameter.  This can be used to teach an existing
turbostat binary about any number of any type of counter.

turbostat(8) details the syntax for --add.
Signed-off-by: NLen Brown <len.brown@intel.com>

This changes only the TSC frequency decoding line seen with --debug

old: TSC: 1382 MHz (19200000 Hz * 216 / 3 / 1000000)
new: TSC: 1800 MHz (25000000 Hz * 216 / 3 / 1000000)
Signed-off-by: NLen Brown <len.brown@intel.com>

The -M option adds an 18-column item, and the header
needs to be wide enough to keep the header aligned
with the columns.
Signed-off-by: NLen Brown <len.brown@intel.com>

SKX has fewer package C-states than previous generations,
and so the decoding of PKG_CSTATE_LIMIT has changed.

This changes the line ending with pkg-cstate-limit=XXX: pcYYY
Signed-off-by: NLen Brown <len.brown@intel.com>

Original-author: Piotr Luc <piotr.luc@intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

Display if the HWP is enabled in OOB (Out of band) mode.
Signed-off-by: NSrinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

Add Denverton to the group of SandyBridge and later processors,
to let the bclk be recognized as 100MHz rather than 133MHz,
then avoid the wrong value of the frequencies based on it,
including Bzy_MHz, max efficiency freuency, base frequency,
and turbo mode frequencies.
Signed-off-by: NXiaolong Wang <xiaolong.wang@linux.intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

All except for model 1F, a Nehalem, which is currently incorrectly
indentified as a Westmere in that new header.
Signed-off-by: NLen Brown <len.brown@intel.com>

The Denverton CPU RAPL supports package, core, and DRAM domains.
Signed-off-by: NJacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

Denverton is an Atom based micro server which shares the same
Goldmont architecture as Broxton. The available C-states on
Denverton is a subset of Broxton with only C1, C1e, and C6.
Signed-off-by: NJacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

Some CPUs may not have PP0/Core domain power limit MSRs. We
should still allow its domain energy status to be used. This
patch splits PP0/Core RAPL into two separate flags for power
limit and energy status such that energy status can continue
to be reported without power limit.

Without this patch, turbostat will not be able to use the
remaining RAPL features if some PL MSRs are not present.
Signed-off-by: NJacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

When i >= SLM_BCLK_FREQS, the frequency read from the slm_freq_table
is off the end of the array because msr is set to 3 rather than the
actual array index i.  Set i to 3 rather than msr to fix this.
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

The tool uses topo.max_cpu_num to determine number of entries needed for
fd_percpu[] and irqs_per_cpu[]. For example on a system with 4 CPUs
topo.max_cpu_num is 3 so we get too small array for holding per-CPU items.

Fix this to use right number of entries, which is topo.max_cpu_num + 1.
Signed-off-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

Switch to tab-delimited output from fixed-width columns
to make it simpler to import into spreadsheets.

As the fixed width columnns were 8-spaces wide,
the output on the screen should not change.
Signed-off-by: NLen Brown <len.brown@intel.com>

turbostat gives valid results across suspend to idle, aka freeze,
whether invoked in  interval mode, or in command mode.
Indeed, this can be used to measure suspend to idle:

turbostat echo freeze > /sys/power/state

But this does not work across suspend to ACPI S3, because the
processor counters, including the TSC, are reset on resume.
Further, when turbostat detects a problem, it does't forgive
the hardware, and interval mode will print *'s from there on out.

Instead, upon detecting counters going backwards, simply
reset and start over.

Interval mode across ACPI S3: (observe TSC going backwards)

root@sharkbay:/home/lenb/turbostat-src# ./turbostat -M 0x10
     CPU Avg_MHz   Busy% Bzy_MHz TSC_MHz           MSR 0x010
       -       1    0.06     858    2294  0x0000000000000000
       0       0    0.06     847    2294  0x0000002a254b98ac
       1       1    0.06     878    2294  0x0000002a254efa3a
       2       1    0.07     843    2294  0x0000002a2551df65
       3       0    0.05     863    2294  0x0000002a2553fea2
turbostat: re-initialized with num_cpus 4
     CPU Avg_MHz   Busy% Bzy_MHz TSC_MHz           MSR 0x010
       -       2    0.20     849    2294  0x0000000000000000
       0       2    0.26     856    2294  0x0000000449abb60d
       1       2    0.20     844    2294  0x0000000449b087ec
       2       2    0.21     850    2294  0x0000000449b35d5d
       3       1    0.12     839    2294  0x0000000449b5fd5a
^C

Command mode across ACPI S3:
root@sharkbay:/home/lenb/turbostat-src# ./turbostat -M 0x10 sleep 10
./turbostat: Counter reset detected
14.196299 sec
Signed-off-by: NLen Brown <len.brown@intel.com>

The RAPL Joules counter is limited in capacity.
Turbostat estimates how soon it can roll-over
based on the max TDP of the processor --
which tells us the maximum increment rate.

eg.
RAPL: 2759 sec. Joule Counter Range, at 95 Watts

So if a sample duration is longer than 2759 seconds on this system,
'**' replace the decimal place in the display to indicate
that the results may be suspect.

But the display had an extra ' ' in this case, throwing off the columns.

Also, the -J "Joules" option appended an extra "time" column
to the display.  While this may be useful, it printed the interval time,
which may not be the accurate time per processor.  Remove this column,
which appeared only when using '-J',
as we plan to add accurate per-cpu interval times in a future commit.
Signed-off-by: NLen Brown <len.brown@intel.com>

Replace MSR_NHM_TURBO_RATIO_LIMIT with MSR_TURBO_RATIO_LIMIT.
Signed-off-by: NSrinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Sometimes the rc6 sysfs counter spontaneously resets,
causing turbostat prints a very large number
as it tries to calcuate % = 100 * (old - new) / interval

When we see (old > new), print ***.**% instead
of a bogus huge number.

Note that this detection is not fool-proof, as the counter
could reset several times and still result in new > old.
Signed-off-by: NLen Brown <len.brown@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

KBL is similar to SKL
Signed-off-by: NLen Brown <len.brown@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

SKX has a lot in common with HSX
Signed-off-by: NLen Brown <len.brown@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Hard-code BXT ART to 19200MHz, so turbostat --debug
can fully enumerate TSC:

CPUID(0x15): eax_crystal: 3 ebx_tsc: 186 ecx_crystal_hz: 0
TSC: 1190 MHz (19200000 Hz * 186 / 3 / 1000000)
Signed-off-by: NLen Brown <len.brown@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Broxton has a lot in common with SKL
Signed-off-by: NLen Brown <len.brown@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

Some processors use the Interrupt Response Time Limit (IRTL) MSR value
to describe the maximum IRQ response time latency for deep
package C-states.  (Though others have the register, but do not use it)
Lets print it out to give insight into the cases where it is used.

IRTL begain in SNB, with PC3/PC6/PC7, and HSW added PC8/PC9/PC10.
Signed-off-by: NLen Brown <len.brown@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

The CPUID.SGX bit was printed, even if --debug was used
Signed-off-by: NLen Brown <len.brown@intel.com>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>

MSR_CONFIG_TDP_NOMINAL:
should print all 8 bits of base_ratio (bit 0:7) 0xFF

MSR_CONFIG_TDP_LEVEL_1:
should print all 15 bits of PKG_MIN_PWR_LVL1 (bit 48:62) 0x7FFF
should print all 15 bits of PKG_MAX_PWR_LVL1 (bit 32:46) 0x7FFF
should print all 8 bits of LVL1_RATIO (bit 16:23) 0xFF
should print all 15 bits of PKG_TDP_LVL1 (bit 0:14) 0x7FFF

And the same modification to MSR_CONFIG_TDP_LEVEL_2.

MSR_TURBO_ACTIVATION_RATIO:
should print all 8 bits of MAX_NON_TURBO_RATIO (bit 0:7) 0xFF
Signed-off-by: NChen Yu <yu.c.chen@intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

MSR_NHM_SNB_PKG_CST_CFG_CTL: 0x1e008008 (...pkg-cstate-limit=0: unlimited)
should print as
MSR_NHM_SNB_PKG_CST_CFG_CTL: 0x1e008008 (...pkg-cstate-limit=8: unlimited)
Signed-off-by: NLen Brown <len.brown@intel.com>

turbostat already checks whether calling each cpuid leavf is legal,
and it doesn't look at the function return value,
so call the simpler gcc intrinsic __cpuid() instead of __get_cpuid().

syntax only, no functional change
Signed-off-by: NLen Brown <len.brown@intel.com>

SGX presence is related to a SKL power workaround,
so lets show when that is enabled.
Signed-off-by: NLen Brown <len.brown@intel.com>

The accuracy of Bzy_Mhz and Busy% depend on reading
the TSC, APERF, and MPERF close together in time.

When there is a very short measurement interval,
or a large system is profoundly idle, the changes
in APERF and MPERF may be very small.
They can be small enough that an expensive interrupt
between reading APERF and MPERF can cause the APERF/MPERF
ratio to become inaccurate, resulting in invalid
calculation and display of Bzy_MHz.

A dummy APERF read of APERF makes this problem
much more rare.  Apparently this 1st systemn call
after exiting a long stretch of idle is when we
typically see expensive timer interrupts that cause
large jitter.

For the cases that dummy APERF read fails to prevent,
we compare the latency of the APERF and MPERF reads.
If they differ by more than 2x, we re-issue them.
Signed-off-by: NLen Brown <len.brown@intel.com>

The column "GFX%c6" show the percentage of time the GPU
is in the "render C6" state, rc6.  Deep package C-states on several
systems depend on the GPU being in RC6.

This information comes from the counter
/sys/class/drm/card0/power/rc6_residency_ms,
as read before and after the measurement interval.
Signed-off-by: NLen Brown <len.brown@intel.com>

Under the column "GFXMHz", show a snapshot of this attribute:
/sys/class/graphics/fb0/device/drm/card0/gt_cur_freq_mhz

This is an instantaneous snapshot of what sysfs presents
at the end of the measurement interval.  turbostat does
not average or otherwise perform any math on this value.
Signed-off-by: NLen Brown <len.brown@intel.com>

The new IRQ column shows how many interrupts have occurred on each CPU
during the measurement inteval.  This information comes from
the difference between /proc/interrupts shapshots made before
and after the measurement interval.

The first row, the system summary, shows the sum of the IRQS
for all CPUs during that interval.
Signed-off-by: NLen Brown <len.brown@intel.com>

skip the open(2)/close(2) on each msr read
by keeping the /dev/cpu/*/msr files open.

The remaining read(2) is generally far fewer cycles
than the removed open(2) system call.
Signed-off-by: NLen Brown <len.brown@intel.com>

Signed-off-by: NLen Brown <len.brown@intel.com>

By default...

Turbostat --debug gconfiguration info goes to stderr.

In FORK mode, turbostat statistics go to stderr.

In PERIODIC mode, turbostat statistics go to stdout.

These defaults do not change, but an option "--out file"
will send all output above only to the specified file.
Signed-off-by: NLen Brown <len.brown@intel.com>

some tools processing turbostat output
have difficulty with items that begin with %...
Reported-by: NJacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

Following changes have been made:
- changed MSR_NHM_TURBO_RATIO_LIMIT to MSR_TURBO_RATIO_LIMIT in debug print
  for consistency with Developer Manual
- updated definition of bitfields in MSR_TURBO_RATIO_LIMIT and appropriate
  parsing code
- added x200 to list of architectures that do not support Nahlem compatible
  definition of MSR_TURBO_RATIO_LIMIT register (x200 has the register but
  bits definition is custom)
- fixed typo in code that parses MSR_TURBO_RATIO_LIMIT
  (logical instead of bitwise operator)
- changed MSR_TURBO_RATIO_LIMIT parsing algorithm so the print out had the
  same order as implementations for other platforms
Signed-off-by: NHubert Chrzaniuk <hubert.chrzaniuk@intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

x200 does not enable any way to programmatically obtain bus clock
speed. Bclk for the architecture has a fixed value of 100 MHz.
At the same time x200 cannot be included in has_snb_msrs since
it does not support C7 idle state.

prior to this patch, MHz values reported on this chip
were erroneously calculated using bclk of 133MHz,
causing MHz values to be reported 33% higher than actual.
Signed-off-by: NHubert Chrzaniuk <hubert.chrzaniuk@intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>