linux/kernel
Giovanni Gherdovich 1567c3e346 x86, sched: Add support for frequency invariance
Implement arch_scale_freq_capacity() for 'modern' x86. This function
is used by the scheduler to correctly account usage in the face of
DVFS.

The present patch addresses Intel processors specifically and has positive
performance and performance-per-watt implications for the schedutil cpufreq
governor, bringing it closer to, if not on-par with, the powersave governor
from the intel_pstate driver/framework.

Large performance gains are obtained when the machine is lightly loaded and
no regression are observed at saturation. The benchmarks with the largest
gains are kernel compilation, tbench (the networking version of dbench) and
shell-intensive workloads.

1. FREQUENCY INVARIANCE: MOTIVATION
   * Without it, a task looks larger if the CPU runs slower

2. PECULIARITIES OF X86
   * freq invariance accounting requires knowing the ratio freq_curr/freq_max
   2.1 CURRENT FREQUENCY
       * Use delta_APERF / delta_MPERF * freq_base (a.k.a "BusyMHz")
   2.2 MAX FREQUENCY
       * It varies with time (turbo). As an approximation, we set it to a
         constant, i.e. 4-cores turbo frequency.

3. EFFECTS ON THE SCHEDUTIL FREQUENCY GOVERNOR
   * The invariant schedutil's formula has no feedback loop and reacts faster
     to utilization changes

4. KNOWN LIMITATIONS
   * In some cases tasks can't reach max util despite how hard they try

5. PERFORMANCE TESTING
   5.1 MACHINES
       * Skylake, Broadwell, Haswell
   5.2 SETUP
       * baseline Linux v5.2 w/ non-invariant schedutil. Tested freq_max = 1-2-3-4-8-12
         active cores turbo w/ invariant schedutil, and intel_pstate/powersave
   5.3 BENCHMARK RESULTS
       5.3.1 NEUTRAL BENCHMARKS
             * NAS Parallel Benchmark (HPC), hackbench
       5.3.2 NON-NEUTRAL BENCHMARKS
             * tbench (10-30% better), kernbench (10-15% better),
               shell-intensive-scripts (30-50% better)
             * no regressions
       5.3.3 SELECTION OF DETAILED RESULTS
       5.3.4 POWER CONSUMPTION, PERFORMANCE-PER-WATT
             * dbench (5% worse on one machine), kernbench (3% worse),
               tbench (5-10% better), shell-intensive-scripts (10-40% better)

6. MICROARCH'ES ADDRESSED HERE
   * Xeon Core before Scalable Performance processors line (Xeon Gold/Platinum
     etc have different MSRs semantic for querying turbo levels)

7. REFERENCES
   * MMTests performance testing framework, github.com/gormanm/mmtests

 +-------------------------------------------------------------------------+
 | 1. FREQUENCY INVARIANCE: MOTIVATION
 +-------------------------------------------------------------------------+

For example; suppose a CPU has two frequencies: 500 and 1000 Mhz. When
running a task that would consume 1/3rd of a CPU at 1000 MHz, it would
appear to consume 2/3rd (or 66.6%) when running at 500 MHz, giving the
false impression this CPU is almost at capacity, even though it can go
faster [*]. In a nutshell, without frequency scale-invariance tasks look
larger just because the CPU is running slower.

[*] (footnote: this assumes a linear frequency/performance relation; which
everybody knows to be false, but given realities its the best approximation
we can make.)

 +-------------------------------------------------------------------------+
 | 2. PECULIARITIES OF X86
 +-------------------------------------------------------------------------+

Accounting for frequency changes in PELT signals requires the computation of
the ratio freq_curr / freq_max. On x86 neither of those terms is readily
available.

2.1 CURRENT FREQUENCY
====================

Since modern x86 has hardware control over the actual frequency we run
at (because amongst other things, Turbo-Mode), we cannot simply use
the frequency as requested through cpufreq.

Instead we use the APERF/MPERF MSRs to compute the effective frequency
over the recent past. Also, because reading MSRs is expensive, don't
do so every time we need the value, but amortize the cost by doing it
every tick.

2.2 MAX FREQUENCY
=================

Obtaining freq_max is also non-trivial because at any time the hardware can
provide a frequency boost to a selected subset of cores if the package has
enough power to spare (eg: Turbo Boost). This means that the maximum frequency
available to a given core changes with time.

The approach taken in this change is to arbitrarily set freq_max to a constant
value at boot. The value chosen is the "4-cores (4C) turbo frequency" on most
microarchitectures, after evaluating the following candidates:

    * 1-core (1C) turbo frequency (the fastest turbo state available)
    * around base frequency (a.k.a. max P-state)
    * something in between, such as 4C turbo

To interpret these options, consider that this is the denominator in
freq_curr/freq_max, and that ratio will be used to scale PELT signals such as
util_avg and load_avg. A large denominator will undershoot (util_avg looks a
bit smaller than it really is), viceversa with a smaller denominator PELT
signals will tend to overshoot. Given that PELT drives frequency selection
in the schedutil governor, we will have:

    freq_max set to     | effect on DVFS
    --------------------+------------------
    1C turbo            | power efficiency (lower freq choices)
    base freq           | performance (higher util_avg, higher freq requests)
    4C turbo            | a bit of both

4C turbo proves to be a good compromise in a number of benchmarks (see below).

 +-------------------------------------------------------------------------+
 | 3. EFFECTS ON THE SCHEDUTIL FREQUENCY GOVERNOR
 +-------------------------------------------------------------------------+

Once an architecture implements a frequency scale-invariant utilization (the
PELT signal util_avg), schedutil switches its frequency selection formula from

    freq_next = 1.25 * freq_curr * util            [non-invariant util signal]

to

    freq_next = 1.25 * freq_max * util             [invariant util signal]

where, in the second formula, freq_max is set to the 1C turbo frequency (max
turbo). The advantage of the second formula, whose usage we unlock with this
patch, is that freq_next doesn't depend on the current frequency in an
iterative fashion, but can jump to any frequency in a single update. This
absence of feedback in the formula makes it quicker to react to utilization
changes and more robust against pathological instabilities.

Compare it to the update formula of intel_pstate/powersave:

    freq_next = 1.25 * freq_max * Busy%

where again freq_max is 1C turbo and Busy% is the percentage of time not spent
idling (calculated with delta_MPERF / delta_TSC); essentially the same as
invariant schedutil, and largely responsible for intel_pstate/powersave good
reputation. The non-invariant schedutil formula is derived from the invariant
one by approximating util_inv with util_raw * freq_curr / freq_max, but this
has limitations.

Testing shows improved performances due to better frequency selections when
the machine is lightly loaded, and essentially no change in behaviour at
saturation / overutilization.

 +-------------------------------------------------------------------------+
 | 4. KNOWN LIMITATIONS
 +-------------------------------------------------------------------------+

It's been shown that it is possible to create pathological scenarios where a
CPU-bound task cannot reach max utilization, if the normalizing factor
freq_max is fixed to a constant value (see [Lelli-2018]).

If freq_max is set to 4C turbo as we do here, one needs to peg at least 5
cores in a package doing some busywork, and observe that none of those task
will ever reach max util (1024) because they're all running at less than the
4C turbo frequency.

While this concern still applies, we believe the performance benefit of
frequency scale-invariant PELT signals outweights the cost of this limitation.

 [Lelli-2018]
 https://lore.kernel.org/lkml/20180517150418.GF22493@localhost.localdomain/

 +-------------------------------------------------------------------------+
 | 5. PERFORMANCE TESTING
 +-------------------------------------------------------------------------+

5.1 MACHINES
============

We tested the patch on three machines, with Skylake, Broadwell and Haswell
CPUs. The details are below, together with the available turbo ratios as
reported by the appropriate MSRs.

* 8x-SKYLAKE-UMA:
  Single socket E3-1240 v5, Skylake 4 cores/8 threads
  Max EFFiciency, BASE frequency and available turbo levels (MHz):

    EFFIC    800 |********
    BASE    3500 |***********************************
    4C      3700 |*************************************
    3C      3800 |**************************************
    2C      3900 |***************************************
    1C      3900 |***************************************

* 80x-BROADWELL-NUMA:
  Two sockets E5-2698 v4, 2x Broadwell 20 cores/40 threads
  Max EFFiciency, BASE frequency and available turbo levels (MHz):

    EFFIC   1200 |************
    BASE    2200 |**********************
    8C      2900 |*****************************
    7C      3000 |******************************
    6C      3100 |*******************************
    5C      3200 |********************************
    4C      3300 |*********************************
    3C      3400 |**********************************
    2C      3600 |************************************
    1C      3600 |************************************

* 48x-HASWELL-NUMA
  Two sockets E5-2670 v3, 2x Haswell 12 cores/24 threads
  Max EFFiciency, BASE frequency and available turbo levels (MHz):

    EFFIC   1200 |************
    BASE    2300 |***********************
    12C     2600 |**************************
    11C     2600 |**************************
    10C     2600 |**************************
    9C      2600 |**************************
    8C      2600 |**************************
    7C      2600 |**************************
    6C      2600 |**************************
    5C      2700 |***************************
    4C      2800 |****************************
    3C      2900 |*****************************
    2C      3100 |*******************************
    1C      3100 |*******************************

5.2 SETUP
=========

* The baseline is Linux v5.2 with schedutil (non-invariant) and the intel_pstate
  driver in passive mode.
* The rationale for choosing the various freq_max values to test have been to
  try all the 1-2-3-4C turbo levels (note that 1C and 2C turbo are identical
  on all machines), plus one more value closer to base_freq but still in the
  turbo range (8C turbo for both 80x-BROADWELL-NUMA and 48x-HASWELL-NUMA).
* In addition we've run all tests with intel_pstate/powersave for comparison.
* The filesystem is always XFS, the userspace is openSUSE Leap 15.1.
* 8x-SKYLAKE-UMA is capable of HWP (Hardware-Managed P-States), so the runs
  with active intel_pstate on this machine use that.

This gives, in terms of combinations tested on each machine:

* 8x-SKYLAKE-UMA
  * Baseline: Linux v5.2, non-invariant schedutil, intel_pstate passive
  * intel_pstate active + powersave + HWP
  * invariant schedutil, freq_max = 1C turbo
  * invariant schedutil, freq_max = 3C turbo
  * invariant schedutil, freq_max = 4C turbo

* both 80x-BROADWELL-NUMA and 48x-HASWELL-NUMA
  * [same as 8x-SKYLAKE-UMA, but no HWP capable]
  * invariant schedutil, freq_max = 8C turbo
    (which on 48x-HASWELL-NUMA is the same as 12C turbo, or "all cores turbo")

5.3 BENCHMARK RESULTS
=====================

5.3.1 NEUTRAL BENCHMARKS
------------------------

Tests that didn't show any measurable difference in performance on any of the
test machines between non-invariant schedutil and our patch are:

* NAS Parallel Benchmarks (NPB) using either MPI or openMP for IPC, any
  computational kernel
* flexible I/O (FIO)
* hackbench (using threads or processes, and using pipes or sockets)

5.3.2 NON-NEUTRAL BENCHMARKS
----------------------------

What follow are summary tables where each benchmark result is given a score.

* A tilde (~) means a neutral result, i.e. no difference from baseline.
* Scores are computed with the ratio result_new / result_baseline, so a tilde
  means a score of 1.00.
* The results in the score ratio are the geometric means of results running
  the benchmark with different parameters (eg: for kernbench: using 1, 2, 4,
  ... number of processes; for pgbench: varying the number of clients, and so
  on).
* The first three tables show higher-is-better kind of tests (i.e. measured in
  operations/second), the subsequent three show lower-is-better kind of tests
  (i.e. the workload is fixed and we measure elapsed time, think kernbench).
* "gitsource" is a name we made up for the test consisting in running the
  entire unit tests suite of the Git SCM and measuring how long it takes. We
  take it as a typical example of shell-intensive serialized workload.
* In the "I_PSTATE" column we have the results for intel_pstate/powersave. Other
  columns show invariant schedutil for different values of freq_max. 4C turbo
  is circled as it's the value we've chosen for the final implementation.

80x-BROADWELL-NUMA (comparison ratio; higher is better)
                                         +------+
                 I_PSTATE   1C     3C    | 4C   |  8C
pgbench-ro           1.14   ~      ~     | 1.11 |  1.14
pgbench-rw           ~      ~      ~     | ~    |  ~
netperf-udp          1.06   ~      1.06  | 1.05 |  1.07
netperf-tcp          ~      1.03   ~     | 1.01 |  1.02
tbench4              1.57   1.18   1.22  | 1.30 |  1.56
                                         +------+

8x-SKYLAKE-UMA (comparison ratio; higher is better)
                                         +------+
             I_PSTATE/HWP   1C     3C    | 4C   |
pgbench-ro           ~      ~      ~     | ~    |
pgbench-rw           ~      ~      ~     | ~    |
netperf-udp          ~      ~      ~     | ~    |
netperf-tcp          ~      ~      ~     | ~    |
tbench4              1.30   1.14   1.14  | 1.16 |
                                         +------+

48x-HASWELL-NUMA (comparison ratio; higher is better)
                                         +------+
                 I_PSTATE   1C     3C    | 4C   |  12C
pgbench-ro           1.15   ~      ~     | 1.06 |  1.16
pgbench-rw           ~      ~      ~     | ~    |  ~
netperf-udp          1.05   0.97   1.04  | 1.04 |  1.02
netperf-tcp          0.96   1.01   1.01  | 1.01 |  1.01
tbench4              1.50   1.05   1.13  | 1.13 |  1.25
                                         +------+

In the table above we see that active intel_pstate is slightly better than our
4C-turbo patch (both in reference to the baseline non-invariant schedutil) on
read-only pgbench and much better on tbench. Both cases are notable in which
it shows that lowering our freq_max (to 8C-turbo and 12C-turbo on
80x-BROADWELL-NUMA and 48x-HASWELL-NUMA respectively) helps invariant
schedutil to get closer.

If we ignore active intel_pstate and focus on the comparison with baseline
alone, there are several instances of double-digit performance improvement.

80x-BROADWELL-NUMA (comparison ratio; lower is better)
                                         +------+
                 I_PSTATE   1C     3C    | 4C   |  8C
dbench4              1.23   0.95   0.95  | 0.95 |  0.95
kernbench            0.93   0.83   0.83  | 0.83 |  0.82
gitsource            0.98   0.49   0.49  | 0.49 |  0.48
                                         +------+

8x-SKYLAKE-UMA (comparison ratio; lower is better)
                                         +------+
             I_PSTATE/HWP   1C     3C    | 4C   |
dbench4              ~      ~      ~     | ~    |
kernbench            ~      ~      ~     | ~    |
gitsource            0.92   0.55   0.55  | 0.55 |
                                         +------+

48x-HASWELL-NUMA (comparison ratio; lower is better)
                                         +------+
                 I_PSTATE   1C     3C    | 4C   |  8C
dbench4              ~      ~      ~     | ~    |  ~
kernbench            0.94   0.90   0.89  | 0.90 |  0.90
gitsource            0.97   0.69   0.69  | 0.69 |  0.69
                                         +------+

dbench is not very remarkable here, unless we notice how poorly active
intel_pstate is performing on 80x-BROADWELL-NUMA: 23% regression versus
non-invariant schedutil. We repeated that run getting consistent results. Out
of scope for the patch at hand, but deserving future investigation. Other than
that, we previously ran this campaign with Linux v5.0 and saw the patch doing
better on dbench a the time. We haven't checked closely and can only speculate
at this point.

On the NUMA boxes kernbench gets 10-15% improvements on average; we'll see in
the detailed tables that the gains concentrate on low process counts (lightly
loaded machines).

The test we call "gitsource" (running the git unit test suite, a long-running
single-threaded shell script) appears rather spectacular in this table (gains
of 30-50% depending on the machine). It is to be noted, however, that
gitsource has no adjustable parameters (such as the number of jobs in
kernbench, which we average over in order to get a single-number summary
score) and is exactly the kind of low-parallelism workload that benefits the
most from this patch. When looking at the detailed tables of kernbench or
tbench4, at low process or client counts one can see similar numbers.

5.3.3 SELECTION OF DETAILED RESULTS
-----------------------------------

Machine            : 48x-HASWELL-NUMA
Benchmark          : tbench4 (i.e. dbench4 over the network, actually loopback)
Varying parameter  : number of clients
Unit               : MB/sec (higher is better)

                   5.2.0 vanilla (BASELINE)               5.2.0 intel_pstate                   5.2.0 1C-turbo
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Hmean  1        126.73  +- 0.31% (        )      315.91  +- 0.66% ( 149.28%)      125.03  +- 0.76% (  -1.34%)
Hmean  2        258.04  +- 0.62% (        )      614.16  +- 0.51% ( 138.01%)      269.58  +- 1.45% (   4.47%)
Hmean  4        514.30  +- 0.67% (        )     1146.58  +- 0.54% ( 122.94%)      533.84  +- 1.99% (   3.80%)
Hmean  8       1111.38  +- 2.52% (        )     2159.78  +- 0.38% (  94.33%)     1359.92  +- 1.56% (  22.36%)
Hmean  16      2286.47  +- 1.36% (        )     3338.29  +- 0.21% (  46.00%)     2720.20  +- 0.52% (  18.97%)
Hmean  32      4704.84  +- 0.35% (        )     4759.03  +- 0.43% (   1.15%)     4774.48  +- 0.30% (   1.48%)
Hmean  64      7578.04  +- 0.27% (        )     7533.70  +- 0.43% (  -0.59%)     7462.17  +- 0.65% (  -1.53%)
Hmean  128     6998.52  +- 0.16% (        )     6987.59  +- 0.12% (  -0.16%)     6909.17  +- 0.14% (  -1.28%)
Hmean  192     6901.35  +- 0.25% (        )     6913.16  +- 0.10% (   0.17%)     6855.47  +- 0.21% (  -0.66%)

                             5.2.0 3C-turbo                   5.2.0 4C-turbo                  5.2.0 12C-turbo
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Hmean  1        128.43  +- 0.28% (   1.34%)      130.64  +- 3.81% (   3.09%)      153.71  +- 5.89% (  21.30%)
Hmean  2        311.70  +- 6.15% (  20.79%)      281.66  +- 3.40% (   9.15%)      305.08  +- 5.70% (  18.23%)
Hmean  4        641.98  +- 2.32% (  24.83%)      623.88  +- 5.28% (  21.31%)      906.84  +- 4.65% (  76.32%)
Hmean  8       1633.31  +- 1.56% (  46.96%)     1714.16  +- 0.93% (  54.24%)     2095.74  +- 0.47% (  88.57%)
Hmean  16      3047.24  +- 0.42% (  33.27%)     3155.02  +- 0.30% (  37.99%)     3634.58  +- 0.15% (  58.96%)
Hmean  32      4734.31  +- 0.60% (   0.63%)     4804.38  +- 0.23% (   2.12%)     4674.62  +- 0.27% (  -0.64%)
Hmean  64      7699.74  +- 0.35% (   1.61%)     7499.72  +- 0.34% (  -1.03%)     7659.03  +- 0.25% (   1.07%)
Hmean  128     6935.18  +- 0.15% (  -0.91%)     6942.54  +- 0.10% (  -0.80%)     7004.85  +- 0.12% (   0.09%)
Hmean  192     6901.62  +- 0.12% (   0.00%)     6856.93  +- 0.10% (  -0.64%)     6978.74  +- 0.10% (   1.12%)

This is one of the cases where the patch still can't surpass active
intel_pstate, not even when freq_max is as low as 12C-turbo. Otherwise, gains are
visible up to 16 clients and the saturated scenario is the same as baseline.

The scores in the summary table from the previous sections are ratios of
geometric means of the results over different clients, as seen in this table.

Machine            : 80x-BROADWELL-NUMA
Benchmark          : kernbench (kernel compilation)
Varying parameter  : number of jobs
Unit               : seconds (lower is better)

                   5.2.0 vanilla (BASELINE)               5.2.0 intel_pstate                   5.2.0 1C-turbo
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Amean  2        379.68  +- 0.06% (        )      330.20  +- 0.43% (  13.03%)      285.93  +- 0.07% (  24.69%)
Amean  4        200.15  +- 0.24% (        )      175.89  +- 0.22% (  12.12%)      153.78  +- 0.25% (  23.17%)
Amean  8        106.20  +- 0.31% (        )       95.54  +- 0.23% (  10.03%)       86.74  +- 0.10% (  18.32%)
Amean  16        56.96  +- 1.31% (        )       53.25  +- 1.22% (   6.50%)       48.34  +- 1.73% (  15.13%)
Amean  32        34.80  +- 2.46% (        )       33.81  +- 0.77% (   2.83%)       30.28  +- 1.59% (  12.99%)
Amean  64        26.11  +- 1.63% (        )       25.04  +- 1.07% (   4.10%)       22.41  +- 2.37% (  14.16%)
Amean  128       24.80  +- 1.36% (        )       23.57  +- 1.23% (   4.93%)       21.44  +- 1.37% (  13.55%)
Amean  160       24.85  +- 0.56% (        )       23.85  +- 1.17% (   4.06%)       21.25  +- 1.12% (  14.49%)

                             5.2.0 3C-turbo                   5.2.0 4C-turbo                   5.2.0 8C-turbo
- - - - - - - -  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Amean  2        284.08  +- 0.13% (  25.18%)      283.96  +- 0.51% (  25.21%)      285.05  +- 0.21% (  24.92%)
Amean  4        153.18  +- 0.22% (  23.47%)      154.70  +- 1.64% (  22.71%)      153.64  +- 0.30% (  23.24%)
Amean  8         87.06  +- 0.28% (  18.02%)       86.77  +- 0.46% (  18.29%)       86.78  +- 0.22% (  18.28%)
Amean  16        48.03  +- 0.93% (  15.68%)       47.75  +- 1.99% (  16.17%)       47.52  +- 1.61% (  16.57%)
Amean  32        30.23  +- 1.20% (  13.14%)       30.08  +- 1.67% (  13.57%)       30.07  +- 1.67% (  13.60%)
Amean  64        22.59  +- 2.02% (  13.50%)       22.63  +- 0.81% (  13.32%)       22.42  +- 0.76% (  14.12%)
Amean  128       21.37  +- 0.67% (  13.82%)       21.31  +- 1.15% (  14.07%)       21.17  +- 1.93% (  14.63%)
Amean  160       21.68  +- 0.57% (  12.76%)       21.18  +- 1.74% (  14.77%)       21.22  +- 1.00% (  14.61%)

The patch outperform active intel_pstate (and baseline) by a considerable
margin; the summary table from the previous section says 4C turbo and active
intel_pstate are 0.83 and 0.93 against baseline respectively, so 4C turbo is
0.83/0.93=0.89 against intel_pstate (~10% better on average). There is no
noticeable difference with regard to the value of freq_max.

Machine            : 8x-SKYLAKE-UMA
Benchmark          : gitsource (time to run the git unit test suite)
Varying parameter  : none
Unit               : seconds (lower is better)

                            5.2.0 vanilla           5.2.0 intel_pstate/hwp         5.2.0 1C-turbo
- - - - - - - -  - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Amean         858.85  +- 1.16% (        )      791.94  +- 0.21% (   7.79%)      474.95 (  44.70%)

                           5.2.0 3C-turbo                   5.2.0 4C-turbo
- - - - - - - -  - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Amean         475.26  +- 0.20% (  44.66%)      474.34  +- 0.13% (  44.77%)

In this test, which is of interest as representing shell-intensive
(i.e. fork-intensive) serialized workloads, invariant schedutil outperforms
intel_pstate/powersave by a whopping 40% margin.

5.3.4 POWER CONSUMPTION, PERFORMANCE-PER-WATT
---------------------------------------------

The following table shows average power consumption in watt for each
benchmark. Data comes from turbostat (package average), which in turn is read
from the RAPL interface on CPUs. We know the patch affects CPU frequencies so
it's reasonable to ignore other power consumers (such as memory or I/O). Also,
we don't have a power meter available in the lab so RAPL is the best we have.

turbostat sampled average power every 10 seconds for the entire duration of
each benchmark. We took all those values and averaged them (i.e. with don't
have detail on a per-parameter granularity, only on whole benchmarks).

80x-BROADWELL-NUMA (power consumption, watts)
                                                    +--------+
               BASELINE I_PSTATE       1C       3C  |     4C |      8C
pgbench-ro       130.01   142.77   131.11   132.45  | 134.65 |  136.84
pgbench-rw        68.30    60.83    71.45    71.70  |  71.65 |   72.54
dbench4           90.25    59.06   101.43    99.89  | 101.10 |  102.94
netperf-udp       65.70    69.81    66.02    68.03  |  68.27 |   68.95
netperf-tcp       88.08    87.96    88.97    88.89  |  88.85 |   88.20
tbench4          142.32   176.73   153.02   163.91  | 165.58 |  176.07
kernbench         92.94   101.95   114.91   115.47  | 115.52 |  115.10
gitsource         40.92    41.87    75.14    75.20  |  75.40 |   75.70
                                                    +--------+
8x-SKYLAKE-UMA (power consumption, watts)
                                                    +--------+
              BASELINE I_PSTATE/HWP    1C       3C  |     4C |
pgbench-ro        46.49    46.68    46.56    46.59  |  46.52 |
pgbench-rw        29.34    31.38    30.98    31.00  |  31.00 |
dbench4           27.28    27.37    27.49    27.41  |  27.38 |
netperf-udp       22.33    22.41    22.36    22.35  |  22.36 |
netperf-tcp       27.29    27.29    27.30    27.31  |  27.33 |
tbench4           41.13    45.61    43.10    43.33  |  43.56 |
kernbench         42.56    42.63    43.01    43.01  |  43.01 |
gitsource         13.32    13.69    17.33    17.30  |  17.35 |
                                                    +--------+
48x-HASWELL-NUMA (power consumption, watts)
                                                    +--------+
               BASELINE I_PSTATE       1C       3C  |     4C |     12C
pgbench-ro       128.84   136.04   129.87   132.43  | 132.30 |  134.86
pgbench-rw        37.68    37.92    37.17    37.74  |  37.73 |   37.31
dbench4           28.56    28.73    28.60    28.73  |  28.70 |   28.79
netperf-udp       56.70    60.44    56.79    57.42  |  57.54 |   57.52
netperf-tcp       75.49    75.27    75.87    76.02  |  76.01 |   75.95
tbench4          115.44   139.51   119.53   123.07  | 123.97 |  130.22
kernbench         83.23    91.55    95.58    95.69  |  95.72 |   96.04
gitsource         36.79    36.99    39.99    40.34  |  40.35 |   40.23
                                                    +--------+

A lower power consumption isn't necessarily better, it depends on what is done
with that energy. Here are tables with the ratio of performance-per-watt on
each machine and benchmark. Higher is always better; a tilde (~) means a
neutral ratio (i.e. 1.00).

80x-BROADWELL-NUMA (performance-per-watt ratios; higher is better)
                                     +------+
             I_PSTATE     1C     3C  |   4C |    8C
pgbench-ro       1.04   1.06   0.94  | 1.07 |  1.08
pgbench-rw       1.10   0.97   0.96  | 0.96 |  0.97
dbench4          1.24   0.94   0.95  | 0.94 |  0.92
netperf-udp      ~      1.02   1.02  | ~    |  1.02
netperf-tcp      ~      1.02   ~     | ~    |  1.02
tbench4          1.26   1.10   1.06  | 1.12 |  1.26
kernbench        0.98   0.97   0.97  | 0.97 |  0.98
gitsource        ~      1.11   1.11  | 1.11 |  1.13
                                     +------+

8x-SKYLAKE-UMA (performance-per-watt ratios; higher is better)
                                     +------+
         I_PSTATE/HWP     1C     3C  |   4C |
pgbench-ro       ~      ~      ~     | ~    |
pgbench-rw       0.95   0.97   0.96  | 0.96 |
dbench4          ~      ~      ~     | ~    |
netperf-udp      ~      ~      ~     | ~    |
netperf-tcp      ~      ~      ~     | ~    |
tbench4          1.17   1.09   1.08  | 1.10 |
kernbench        ~      ~      ~     | ~    |
gitsource        1.06   1.40   1.40  | 1.40 |
                                     +------+

48x-HASWELL-NUMA  (performance-per-watt ratios; higher is better)
                                     +------+
             I_PSTATE     1C     3C  |   4C |   12C
pgbench-ro       1.09   ~      1.09  | 1.03 |  1.11
pgbench-rw       ~      0.86   ~     | ~    |  0.86
dbench4          ~      1.02   1.02  | 1.02 |  ~
netperf-udp      ~      0.97   1.03  | 1.02 |  ~
netperf-tcp      0.96   ~      ~     | ~    |  ~
tbench4          1.24   ~      1.06  | 1.05 |  1.11
kernbench        0.97   0.97   0.98  | 0.97 |  0.96
gitsource        1.03   1.33   1.32  | 1.32 |  1.33
                                     +------+

These results are overall pleasing: in plenty of cases we observe
performance-per-watt improvements. The few regressions (read/write pgbench and
dbench on the Broadwell machine) are of small magnitude. kernbench loses a few
percentage points (it has a 10-15% performance improvement, but apparently the
increase in power consumption is larger than that). tbench4 and gitsource, which
benefit the most from the patch, keep a positive score in this table which is
a welcome surprise; that suggests that in those particular workloads the
non-invariant schedutil (and active intel_pstate, too) makes some rather
suboptimal frequency selections.

+-------------------------------------------------------------------------+
| 6. MICROARCH'ES ADDRESSED HERE
+-------------------------------------------------------------------------+

The patch addresses Xeon Core processors that use MSR_PLATFORM_INFO and
MSR_TURBO_RATIO_LIMIT to advertise their base frequency and turbo frequencies
respectively. This excludes the recent Xeon Scalable Performance processors
line (Xeon Gold, Platinum etc) whose MSRs have to be parsed differently.

Subsequent patches will address:

* Xeon Scalable Performance processors and Atom Goldmont/Goldmont Plus
* Xeon Phi (Knights Landing, Knights Mill)
* Atom Silvermont

+-------------------------------------------------------------------------+
| 7. REFERENCES
+-------------------------------------------------------------------------+

Tests have been run with the help of the MMTests performance testing
framework, see github.com/gormanm/mmtests. The configuration file names for
the benchmark used are:

    db-pgbench-timed-ro-small-xfs
    db-pgbench-timed-rw-small-xfs
    io-dbench4-async-xfs
    network-netperf-unbound
    network-tbench
    scheduler-unbound
    workload-kerndevel-xfs
    workload-shellscripts-xfs
    hpc-nas-c-class-mpi-full-xfs
    hpc-nas-c-class-omp-full

All those benchmarks are generally available on the web:

pgbench: https://www.postgresql.org/docs/10/pgbench.html
netperf: https://hewlettpackard.github.io/netperf/
dbench/tbench: https://dbench.samba.org/
gitsource: git unit test suite, github.com/git/git
NAS Parallel Benchmarks: https://www.nas.nasa.gov/publications/npb.html
hackbench: https://people.redhat.com/mingo/cfs-scheduler/tools/hackbench.c

Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Giovanni Gherdovich <ggherdovich@suse.cz>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Doug Smythies <dsmythies@telus.net>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Link: https://lkml.kernel.org/r/20200122151617.531-2-ggherdovich@suse.cz
2020-01-28 21:36:59 +01:00
..
bpf Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2019-12-22 09:54:33 -08:00
cgroup Merge branch 'for-5.5' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup 2019-11-25 19:23:46 -08:00
configs
debug kdb: Tweak escape handling for vi users 2019-10-28 12:08:29 +00:00
dma lib/genalloc.c: rename addr_in_gen_pool to gen_pool_has_addr 2019-12-04 19:44:13 -08:00
events perf/core: Add SRCU annotation for pmus list walk 2019-12-17 13:32:46 +01:00
gcov um: Enable CONFIG_CONSTRUCTORS 2019-09-15 21:37:13 +02:00
irq irqchip updates for Linux 5.5 2019-11-20 14:16:34 +01:00
livepatch New tracing features: 2019-11-27 11:42:01 -08:00
locking Revert "locking/mutex: Complain upon mutex API misuse in IRQ contexts" 2019-12-11 00:27:43 +01:00
power Additional power management updates for 5.5-rc1 2019-12-04 10:48:09 -08:00
printk Merge branch 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2019-12-03 09:29:50 -08:00
rcu Merge branches 'doc.2019.10.29a', 'fixes.2019.10.30a', 'nohz.2019.10.28a', 'replace.2019.10.30a', 'torture.2019.10.05a' and 'lkmm.2019.10.05a' into HEAD 2019-10-30 08:47:13 -07:00
sched x86, sched: Add support for frequency invariance 2020-01-28 21:36:59 +01:00
time Merge branch 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2019-12-03 12:20:25 -08:00
trace tracing: Fix endianness bug in histogram trigger 2019-12-21 16:08:59 -05:00
.gitignore Provide in-kernel headers to make extending kernel easier 2019-04-29 16:48:03 +02:00
Kconfig.freezer treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
Kconfig.hz treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
Kconfig.locks sched/rt, locking: Use CONFIG_PREEMPTION 2019-12-08 14:37:36 +01:00
Kconfig.preempt sched/Kconfig: Fix spelling mistake in user-visible help text 2019-11-12 11:35:32 +01:00
Makefile Kbuild updates for v5.5 2019-12-02 17:35:04 -08:00
acct.c acct_on(): don't mess with freeze protection 2019-04-04 21:04:13 -04:00
async.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 441 2019-06-05 17:37:17 +02:00
audit.c audit: remove redundant condition check in kauditd_thread() 2019-10-25 11:48:14 -04:00
audit.h audit/stable-5.3 PR 20190702 2019-07-08 18:55:42 -07:00
audit_fsnotify.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 157 2019-05-30 11:26:37 -07:00
audit_tree.c fsnotify: switch send_to_group() and ->handle_event to const struct qstr * 2019-04-26 13:51:03 -04:00
audit_watch.c audit_get_nd(): don't unlock parent too early 2019-11-10 11:56:55 -05:00
auditfilter.c audit/stable-5.3 PR 20190702 2019-07-08 18:55:42 -07:00
auditsc.c Revert "bpf: Emit audit messages upon successful prog load and unload" 2019-11-23 09:56:02 -08:00
backtracetest.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 441 2019-06-05 17:37:17 +02:00
bounds.c
capability.c
compat.c y2038: itimer: compat handling to itimer.c 2019-11-15 14:38:30 +01:00
configs.c kernel/configs: Replace GPL boilerplate code with SPDX identifier 2019-07-30 18:34:15 +02:00
context_tracking.c context_tracking: Rename context_tracking_is_enabled() => context_tracking_enabled() 2019-10-29 10:01:12 +01:00
cpu.c cpu/hotplug, stop_machine: Fix stop_machine vs hotplug order 2019-12-17 13:32:50 +01:00
cpu_pm.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 282 2019-06-05 17:36:37 +02:00
crash_core.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 230 2019-06-19 17:09:06 +02:00
crash_dump.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
cred.c Merge branch 'access-creds' 2019-07-25 08:36:29 -07:00
delayacct.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 25 2019-05-21 11:52:39 +02:00
dma.c
elfcore.c kernel/elfcore.c: include proper prototypes 2019-09-25 17:51:39 -07:00
exec_domain.c
exit.c Pipework for general notification queue 2019-11-30 14:12:13 -08:00
extable.c bpf: Add support for BTF pointers to x86 JIT 2019-10-17 16:44:36 +02:00
fail_function.c fail_function: no need to check return value of debugfs_create functions 2019-06-03 15:49:06 +02:00
fork.c Merge branch 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2019-12-03 12:20:25 -08:00
freezer.c Revert "libata, freezer: avoid block device removal while system is frozen" 2019-10-06 09:11:37 -06:00
futex.c futex: Prevent exit livelock 2019-11-20 09:40:38 +01:00
gen_kheaders.sh kheaders: explain why include/config/autoconf.h is excluded from md5sum 2019-11-11 20:10:01 +09:00
groups.c
hung_task.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
iomem.c mm/nvdimm: add is_ioremap_addr and use that to check ioremap address 2019-07-12 11:05:40 -07:00
irq_work.c irq_work: Fix IRQ_WORK_BUSY bit clearing 2019-11-15 10:48:37 +01:00
jump_label.c jump_label: Don't warn on __exit jump entries 2019-08-29 15:10:10 +01:00
kallsyms.c kallsyms: Don't let kallsyms_lookup_size_offset() fail on retrieving the first symbol 2019-08-27 16:19:56 +01:00
kcmp.c
kcov.c kcov: remote coverage support 2019-12-04 19:44:14 -08:00
kexec.c kexec_load: Disable at runtime if the kernel is locked down 2019-08-19 21:54:15 -07:00
kexec_core.c kexec: bail out upon SIGKILL when allocating memory. 2019-09-25 17:51:40 -07:00
kexec_elf.c kexec_elf: support 32 bit ELF files 2019-09-06 23:58:44 +02:00
kexec_file.c kexec: Fix pointer-to-int-cast warnings 2019-11-01 21:42:58 +01:00
kexec_internal.h
kheaders.c kheaders: Move from proc to sysfs 2019-05-24 20:16:01 +02:00
kmod.c
kprobes.c Tracing updates: 2019-09-20 11:19:48 -07:00
ksysfs.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 170 2019-05-30 11:26:39 -07:00
kthread.c kthread: make __kthread_queue_delayed_work static 2019-10-16 09:20:58 -07:00
latencytop.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 441 2019-06-05 17:37:17 +02:00
module-internal.h treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 36 2019-05-24 17:27:11 +02:00
module.c This contains 3 changes: 2019-12-11 12:22:38 -08:00
module_signature.c MODSIGN: Export module signature definitions 2019-08-05 18:39:56 -04:00
module_signing.c MODSIGN: Export module signature definitions 2019-08-05 18:39:56 -04:00
notifier.c kernel/notifier.c: remove blocking_notifier_chain_cond_register() 2019-12-04 19:44:12 -08:00
nsproxy.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 441 2019-06-05 17:37:17 +02:00
padata.c padata: remove cpu_index from the parallel_queue 2019-09-13 21:15:41 +10:00
panic.c locking/refcount: Remove unused 'refcount_error_report()' function 2019-11-25 09:15:42 +01:00
params.c lockdown: Lock down module params that specify hardware parameters (eg. ioport) 2019-08-19 21:54:16 -07:00
pid.c fork: extend clone3() to support setting a PID 2019-11-15 23:49:22 +01:00
pid_namespace.c fork: extend clone3() to support setting a PID 2019-11-15 23:49:22 +01:00
profile.c kernel/profile.c: use cpumask_available to check for NULL cpumask 2019-12-04 19:44:12 -08:00
ptrace.c ptrace: add PTRACE_GET_SYSCALL_INFO request 2019-07-16 19:23:24 -07:00
range.c
reboot.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
relay.c Merge branch 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2019-03-12 13:27:20 -07:00
resource.c mm/memory_hotplug.c: use PFN_UP / PFN_DOWN in walk_system_ram_range() 2019-09-24 15:54:09 -07:00
rseq.c signal: Remove task parameter from force_sig 2019-05-27 09:36:28 -05:00
seccomp.c seccomp: add SECCOMP_USER_NOTIF_FLAG_CONTINUE 2019-10-10 14:45:51 -07:00
signal.c cgroup: freezer: call cgroup_enter_frozen() with preemption disabled in ptrace_stop() 2019-10-11 08:39:57 -07:00
smp.c smp: Warn on function calls from softirq context 2019-07-20 11:27:16 +02:00
smpboot.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
smpboot.h
softirq.c Merge branch 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2019-07-08 11:01:13 -07:00
stackleak.c
stacktrace.c stacktrace: Get rid of unneeded '!!' pattern 2019-11-11 10:30:59 +01:00
stop_machine.c stop_machine: Make stop_cpus() static 2020-01-17 10:19:21 +01:00
sys.c kernel/sys.c: avoid copying possible padding bytes in copy_to_user 2019-12-04 19:44:12 -08:00
sys_ni.c y2038: allow disabling time32 system calls 2019-11-15 14:38:30 +01:00
sysctl-test.c kernel/sysctl-test: Add null pointer test for sysctl.c:proc_dointvec() 2019-09-30 17:35:01 -06:00
sysctl.c kernel: sysctl: make drop_caches write-only 2019-12-01 12:59:07 -08:00
sysctl_binary.c sysctl: Remove the sysctl system call 2019-11-26 13:03:56 -06:00
task_work.c
taskstats.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 157 2019-05-30 11:26:37 -07:00
test_kprobes.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 25 2019-05-21 11:52:39 +02:00
torture.c torture: Remove exporting of internal functions 2019-08-01 14:30:22 -07:00
tracepoint.c The main changes in this release include: 2019-07-18 11:51:00 -07:00
tsacct.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 157 2019-05-30 11:26:37 -07:00
ucount.c proc/sysctl: add shared variables for range check 2019-07-18 17:08:07 -07:00
uid16.c
uid16.h
umh.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
up.c smp: Remove smp_call_function() and on_each_cpu() return values 2019-06-23 14:26:26 +02:00
user-return-notifier.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
user.c Keyrings namespacing 2019-07-08 19:36:47 -07:00
user_namespace.c Keyrings namespacing 2019-07-08 19:36:47 -07:00
utsname.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 441 2019-06-05 17:37:17 +02:00
utsname_sysctl.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 441 2019-06-05 17:37:17 +02:00
watchdog.c watchdog: Remove soft_lockup_hrtimer_cnt and related code 2020-01-17 10:19:19 +01:00
watchdog_hld.c kernel/watchdog_hld.c: hard lockup message should end with a newline 2019-04-19 09:46:05 -07:00
workqueue.c Linux 5.5-rc3 2019-12-25 10:41:37 +01:00
workqueue_internal.h sched/core, workqueues: Distangle worker accounting from rq lock 2019-04-16 16:55:15 +02:00