all messages for Guix-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
* [bug#53121] [PATCH] gnu: ceres: Update to 2.0.0.
@ 2022-01-07 22:02 Felix Gruber
  2022-01-11 19:10 ` bug#53121: " Ludovic Courtès
  0 siblings, 1 reply; 4+ messages in thread
From: Felix Gruber @ 2022-01-07 22:02 UTC (permalink / raw)
  To: 53121; +Cc: Felix Gruber

* gnu/packages/maths.scm (ceres): Update to 2.0.0.
  [inputs]: Use simplified format.
  (ceres-solver-benchmarks)[phases]: Add schur_eliminator_benchmark.
  Replace autodiff_cost_function_benchmark with new autodiff_benchmarks.
---
 gnu/packages/maths.scm | 41 ++++++++++++++++++++++-------------------
 1 file changed, 22 insertions(+), 19 deletions(-)

diff --git a/gnu/packages/maths.scm b/gnu/packages/maths.scm
index b0944f307b..a8fd8a65fa 100644
--- a/gnu/packages/maths.scm
+++ b/gnu/packages/maths.scm
@@ -2437,7 +2437,7 @@ satisfiability checking (SAT).")
 (define-public ceres
   (package
     (name "ceres-solver")
-    (version "1.14.0")
+    (version "2.0.0")
     (home-page "http://ceres-solver.org/")
     (source (origin
               (method url-fetch)
@@ -2445,7 +2445,7 @@ satisfiability checking (SAT).")
                                   version ".tar.gz"))
               (sha256
                (base32
-                "13lfxy8x58w8vprr0nkbzziaijlh0vvqshgahvcgw0mrqdgh0i27"))))
+                "00vng9vnmdb1qga01m0why90m0041w7bn6kxa2h4m26aflfqla8h"))))
     (build-system cmake-build-system)
     (arguments
      ;; TODO: Build HTML user documentation and install separately.
@@ -2464,11 +2464,11 @@ satisfiability checking (SAT).")
     (propagated-inputs
      (list glog))                           ;for #include <glog/glog.h>
     (inputs
-     `(("eigen" ,eigen)
-       ("blas" ,openblas)
-       ("lapack" ,lapack)
-       ("suitesparse" ,suitesparse)
-       ("gflags" ,gflags)))
+     (list eigen
+           openblas
+           lapack
+           suitesparse
+           gflags))
     (synopsis "C++ library for solving large optimization problems")
     (description
      "Ceres Solver is a C++ library for modeling and solving large,
@@ -2506,21 +2506,24 @@ can solve two kinds of problems:
                                                   "pkg-config" "eigen3"
                                                   "--cflags"))))
 
-                        (define (compile-file file)
-                          (let ((source (string-append file ".cc")))
-                            (format #t "building '~a'...~%" file)
-                            (apply invoke "c++" "-fopenmp" "-O2" "-g" "-DNDEBUG"
-                                   source "-lceres" "-lbenchmark" "-lglog"
-                                   "-pthread"
-                                   "-o" (string-append bin "/" file)
-                                   "-I" ".." flags)))
+                        (define (compile-file top-dir)
+                          (lambda (file)
+                            (let ((source (string-append file ".cc")))
+                              (format #t "building '~a'...~%" file)
+                              (apply invoke "c++" "-fopenmp" "-O2" "-g" "-DNDEBUG"
+                                     source "-lceres" "-lbenchmark" "-lglog"
+                                     "-pthread"
+                                     "-o" (string-append bin "/" file)
+                                     "-I" top-dir flags))))
 
                         (mkdir-p bin)
                         (with-directory-excursion "internal/ceres"
-                          (for-each compile-file
-                                    '("small_blas_gemm_benchmark"
-                                      "small_blas_gemv_benchmark"
-                                      "autodiff_cost_function_benchmark"))))))
+                          (for-each (compile-file "..")
+                                    '("schur_eliminator_benchmark"
+                                      "small_blas_gemm_benchmark"
+                                      "small_blas_gemv_benchmark"))
+                          (with-directory-excursion "autodiff_benchmarks"
+                            ((compile-file "../..") "autodiff_benchmarks"))))))
                   (delete 'check)
                   (delete 'install))))
     (inputs (modify-inputs (package-inputs ceres)
-- 
2.30.2





^ permalink raw reply related	[flat|nested] 4+ messages in thread

* bug#53121: [PATCH] gnu: ceres: Update to 2.0.0.
  2022-01-07 22:02 [bug#53121] [PATCH] gnu: ceres: Update to 2.0.0 Felix Gruber
@ 2022-01-11 19:10 ` Ludovic Courtès
  2022-01-18 12:39   ` [bug#53121] " Felix Gruber
  0 siblings, 1 reply; 4+ messages in thread
From: Ludovic Courtès @ 2022-01-11 19:10 UTC (permalink / raw)
  To: Felix Gruber; +Cc: 53121-done

Hi,

Felix Gruber <felgru@posteo.net> skribis:

> * gnu/packages/maths.scm (ceres): Update to 2.0.0.
>   [inputs]: Use simplified format.
>   (ceres-solver-benchmarks)[phases]: Add schur_eliminator_benchmark.
>   Replace autodiff_cost_function_benchmark with new autodiff_benchmarks.

Applied, thanks!

Since you’re looking at benchmarks, I’d be curious to see how those you
added compare when passing ‘--tune’:

  https://hpc.guix.info/blog/2022/01/tuning-packages-for-a-cpu-micro-architecture/

Ludo’.




^ permalink raw reply	[flat|nested] 4+ messages in thread

* [bug#53121] [PATCH] gnu: ceres: Update to 2.0.0.
  2022-01-11 19:10 ` bug#53121: " Ludovic Courtès
@ 2022-01-18 12:39   ` Felix Gruber
  2022-01-19 10:26     ` Ludovic Courtès
  0 siblings, 1 reply; 4+ messages in thread
From: Felix Gruber @ 2022-01-18 12:39 UTC (permalink / raw)
  To: Ludovic Courtès; +Cc: 53121-done

[-- Attachment #1: Type: text/plain, Size: 1922 bytes --]

Hi Ludo’,

On 1/11/22 20:10, Ludovic Courtès wrote:
> Hi,
> 
> Felix Gruber <felgru@posteo.net> skribis:
> 
>> * gnu/packages/maths.scm (ceres): Update to 2.0.0.
>>    [inputs]: Use simplified format.
>>    (ceres-solver-benchmarks)[phases]: Add schur_eliminator_benchmark.
>>    Replace autodiff_cost_function_benchmark with new autodiff_benchmarks.
> 
> Applied, thanks!
> 
> Since you’re looking at benchmarks, I’d be curious to see how those you
> added compare when passing ‘--tune’:
> 
>    https://hpc.guix.info/blog/2022/01/tuning-packages-for-a-cpu-micro-architecture/

Unfortunately, I'm getting mixed results for the benchmarks. In most 
cases, I got slight (<10%) improvements in runtime, but there are also 
some benchmarks that were worse with the --tune flag. I'm wondering 
whether the compiler flags set by the --tune option are correctly used 
by the custom 'build phase of the ceres-solver-benchmarks package. I 
didn't have the time to look closer into it as I'm currently in the 
middle of moving to another country.

Anyways, I've attached the results of benchmark runs that I've generated 
using guix commit 7f779286df7e8636d901f4734501902cc934a72f once untuned 
and once tuned for broadwell CPUs.
My laptop on which I ran the tests has a Quad Core AMD Ryzen 7 PRO 2700U 
CPU with 2200 MHz.

In the attachments you find
* a script run_benchmarks.sh used to run the benchmarks in tuned and 
untuned guix shells,
* text files ending in `-tuned` or `-untuned` which contain the results 
of those benchmark runs,
* a script compare.sh which calls a Python script compare-results.py to 
generate files ending in `-diff` that contain the relative change 
between untuned and tuned benchmarks (negative time and CPU percentages 
mean the tuned benchmark was faster, while for the number of iterations, 
positive percentages mean the tuned benchmark had run more iterations).

Best regards,
Felix

[-- Attachment #2: run_benchmarks.sh --]
[-- Type: application/x-shellscript, Size: 319 bytes --]

[-- Attachment #3: compare.sh --]
[-- Type: application/x-shellscript, Size: 201 bytes --]

[-- Attachment #4: compare-results.py --]
[-- Type: text/x-python, Size: 1069 bytes --]

#!/usr/bin/env python3

import re
import sys

basename = sys.argv[1]

bench_pattern = re.compile('^(.+?)\s+(\d+\.?\d*) ns\s+(\d+\.?\d*) ns\s+(\d+)')

def parse_file(filename):
    res = []
    with open(filename) as f:
        # ignore header
        f.readline()
        f.readline()
        f.readline()
        while line := f.readline():
            m = bench_pattern.match(line)
            res.append((m.group(1), float(m.group(2)),
                        float(m.group(3)), int(m.group(4))))
    return res

untuned = parse_file(basename + '-untuned')
tuned = parse_file(basename + '-tuned')

print(f'{"Benchmark":50} {"Time":>8}   {"CPU":>8}   {"Iterations":>8}')
for (u_name, u_time, u_cpu, u_iterations), \
        (t_name, t_time, t_cpu, t_iterations) in zip(untuned, tuned):
    assert u_name == t_name
    rel_time = (t_time - u_time) / u_time
    rel_cpu = (t_cpu - u_cpu) / u_cpu
    rel_iterations = (t_iterations - u_iterations) / u_iterations
    print(f'{u_name:50} {rel_time * 100:6.1f} %   {rel_cpu * 100:6.1f} %   {rel_iterations * 100:6.1f} %')

[-- Attachment #5: autodiff_benchmarks-diff --]
[-- Type: text/plain, Size: 3446 bytes --]

Benchmark                                              Time        CPU   Iterations
BM_ConstantAnalytic<1>                                0.0 %      0.0 %     -1.1 %
BM_ConstantAutodiff<1, kNotDynamic>                   0.0 %      0.0 %      0.4 %
BM_ConstantAutodiff<1, kDynamic>                      1.8 %      1.8 %     -3.4 %
BM_ConstantAnalytic<10>                               0.0 %      0.0 %     -0.8 %
BM_ConstantAutodiff<10, kNotDynamic>                  0.0 %      0.0 %      0.2 %
BM_ConstantAutodiff<10, kDynamic>                     6.0 %      6.0 %     -3.6 %
BM_ConstantAnalytic<20>                             -15.3 %    -15.3 %     17.2 %
BM_ConstantAutodiff<20, kNotDynamic>                  0.0 %      0.0 %     -0.3 %
BM_ConstantAutodiff<20, kDynamic>                     3.5 %      3.5 %     -4.5 %
BM_ConstantAnalytic<30>                             -11.6 %    -11.6 %     11.8 %
BM_ConstantAutodiff<30, kNotDynamic>                  0.0 %      0.0 %     -0.8 %
BM_ConstantAutodiff<30, kDynamic>                     2.0 %      2.0 %     -2.1 %
BM_ConstantAnalytic<40>                             -10.9 %    -10.9 %     12.2 %
BM_ConstantAutodiff<40, kNotDynamic>                  0.0 %      0.0 %      0.1 %
BM_ConstantAutodiff<40, kDynamic>                     0.4 %      0.4 %     -2.5 %
BM_ConstantAnalytic<50>                             -10.3 %    -10.3 %     10.4 %
BM_ConstantAutodiff<50, kNotDynamic>                  0.0 %      0.0 %      0.3 %
BM_ConstantAutodiff<50, kDynamic>                   -15.4 %    -15.4 %     18.3 %
BM_ConstantAnalytic<60>                              -4.9 %     -4.9 %      5.3 %
BM_ConstantAutodiff<60, kNotDynamic>                  0.0 %      0.0 %      0.1 %
BM_ConstantAutodiff<60, kDynamic>                   -13.8 %    -13.8 %     15.5 %
BM_Linear1AutoDiff<kNotDynamic>/0                     0.0 %      0.0 %     -0.1 %
BM_Linear1AutoDiff<kNotDynamic>/1                     0.5 %      0.5 %      0.3 %
BM_Linear1AutoDiff<kDynamic>/0                      -26.5 %    -26.5 %     39.1 %
BM_Linear1AutoDiff<kDynamic>/1                        4.4 %      4.4 %     -4.5 %
BM_Linear10AutoDiff<kNotDynamic>/0                  -12.7 %    -12.7 %     18.6 %
BM_Linear10AutoDiff<kNotDynamic>/1                  -17.1 %    -17.1 %     20.6 %
BM_Linear10AutoDiff<kDynamic>/0                      14.7 %     14.7 %    -11.8 %
BM_Linear10AutoDiff<kDynamic>/1                     -17.9 %    -17.9 %     23.7 %
BM_Rat43AutoDiff<kNotDynamic>/0                      -1.4 %     -1.4 %      1.7 %
BM_Rat43AutoDiff<kNotDynamic>/1                      10.2 %     10.2 %     -8.0 %
BM_Rat43AutoDiff<kDynamic>/0                          1.0 %      1.0 %      0.0 %
BM_Rat43AutoDiff<kDynamic>/1                          5.5 %      5.5 %     -4.4 %
BM_SnavelyReprojectionAutoDiff<kNotDynamic>/0        -8.3 %     -8.3 %      9.2 %
BM_SnavelyReprojectionAutoDiff<kNotDynamic>/1        -9.5 %     -9.5 %     10.4 %
BM_SnavelyReprojectionAutoDiff<kDynamic>/0           -7.4 %     -7.4 %      7.5 %
BM_SnavelyReprojectionAutoDiff<kDynamic>/1           16.3 %     16.3 %    -14.9 %
BM_PhotometricAutoDiff<kNotDynamic>/0                -8.3 %     -8.3 %      7.7 %
BM_PhotometricAutoDiff<kNotDynamic>/1                -7.3 %     -7.3 %     12.5 %
BM_PhotometricAutoDiff<kDynamic>/0                   -8.1 %     -8.1 %      7.5 %
BM_PhotometricAutoDiff<kDynamic>/1                    4.9 %      4.9 %     -5.3 %

[-- Attachment #6: autodiff_benchmarks-tuned --]
[-- Type: text/plain, Size: 3916 bytes --]

----------------------------------------------------------------------------------------
Benchmark                                              Time             CPU   Iterations
----------------------------------------------------------------------------------------
BM_ConstantAnalytic<1>                              1.85 ns         1.85 ns    323678102
BM_ConstantAutodiff<1, kNotDynamic>                 1.85 ns         1.85 ns    374996000
BM_ConstantAutodiff<1, kDynamic>                    74.7 ns         74.7 ns      8959632
BM_ConstantAnalytic<10>                             2.11 ns         2.11 ns    326701021
BM_ConstantAutodiff<10, kNotDynamic>                1.85 ns         1.85 ns    374190797
BM_ConstantAutodiff<10, kDynamic>                   78.3 ns         78.3 ns      8561831
BM_ConstantAnalytic<20>                             11.6 ns         11.6 ns     58621429
BM_ConstantAutodiff<20, kNotDynamic>                1.85 ns         1.85 ns    373014934
BM_ConstantAutodiff<20, kDynamic>                   77.3 ns         77.3 ns      8726690
BM_ConstantAnalytic<30>                             14.5 ns         14.5 ns     47338220
BM_ConstantAutodiff<30, kNotDynamic>                1.85 ns         1.85 ns    373188709
BM_ConstantAutodiff<30, kDynamic>                   71.3 ns         71.3 ns      9433456
BM_ConstantAnalytic<40>                             17.2 ns         17.2 ns     40270493
BM_ConstantAutodiff<40, kNotDynamic>                1.85 ns         1.85 ns    374977704
BM_ConstantAutodiff<40, kDynamic>                   79.1 ns         79.1 ns      8684902
BM_ConstantAnalytic<50>                             20.1 ns         20.1 ns     34159714
BM_ConstantAutodiff<50, kNotDynamic>                1.85 ns         1.85 ns    374997749
BM_ConstantAutodiff<50, kDynamic>                   69.7 ns         69.7 ns      9750677
BM_ConstantAnalytic<60>                             23.2 ns         23.2 ns     29953898
BM_ConstantAutodiff<60, kNotDynamic>                1.85 ns         1.85 ns    375215126
BM_ConstantAutodiff<60, kDynamic>                   70.8 ns         70.8 ns      9600159
BM_Linear1AutoDiff<kNotDynamic>/0                   1.85 ns         1.85 ns    373583319
BM_Linear1AutoDiff<kNotDynamic>/1                   2.12 ns         2.12 ns    328191653
BM_Linear1AutoDiff<kDynamic>/0                      4.08 ns         4.08 ns    169692582
BM_Linear1AutoDiff<kDynamic>/1                      77.7 ns         77.7 ns      8669752
BM_Linear10AutoDiff<kNotDynamic>/0                  4.76 ns         4.76 ns    147172382
BM_Linear10AutoDiff<kNotDynamic>/1                  97.0 ns         97.0 ns      7048729
BM_Linear10AutoDiff<kDynamic>/0                     8.48 ns         8.48 ns     80392055
BM_Linear10AutoDiff<kDynamic>/1                      229 ns          229 ns      3059987
BM_Rat43AutoDiff<kNotDynamic>/0                     27.2 ns         27.2 ns     25563169
BM_Rat43AutoDiff<kNotDynamic>/1                      110 ns          110 ns      6250716
BM_Rat43AutoDiff<kDynamic>/0                        29.5 ns         29.5 ns     23548868
BM_Rat43AutoDiff<kDynamic>/1                         174 ns          174 ns      3975393
BM_SnavelyReprojectionAutoDiff<kNotDynamic>/0       37.8 ns         37.8 ns     18508182
BM_SnavelyReprojectionAutoDiff<kNotDynamic>/1        392 ns          392 ns      1780936
BM_SnavelyReprojectionAutoDiff<kDynamic>/0          40.2 ns         40.2 ns     17292729
BM_SnavelyReprojectionAutoDiff<kDynamic>/1           872 ns          872 ns       766258
BM_PhotometricAutoDiff<kNotDynamic>/0                729 ns          729 ns       917210
BM_PhotometricAutoDiff<kNotDynamic>/1               4718 ns         4718 ns       147606
BM_PhotometricAutoDiff<kDynamic>/0                   734 ns          734 ns       916042
BM_PhotometricAutoDiff<kDynamic>/1                  7696 ns         7696 ns        88093

[-- Attachment #7: autodiff_benchmarks-untuned --]
[-- Type: text/plain, Size: 4628 bytes --]

----------------------------------------------------------------------------------------
Benchmark                                              Time             CPU   Iterations
----------------------------------------------------------------------------------------
BM_ConstantAnalytic<1>                              1.85 ns         1.85 ns    327136717
BM_ConstantAutodiff<1, kNotDynamic>                 1.85 ns         1.85 ns    373475525
BM_ConstantAutodiff<1, kDynamic>                    73.4 ns         73.4 ns      9271017
BM_ConstantAnalytic<10>                             2.11 ns         2.11 ns    329171004
BM_ConstantAutodiff<10, kNotDynamic>                1.85 ns         1.85 ns    373310978
BM_ConstantAutodiff<10, kDynamic>                   73.9 ns         73.9 ns      8878461
BM_ConstantAnalytic<20>                             13.7 ns         13.7 ns     49998700
BM_ConstantAutodiff<20, kNotDynamic>                1.85 ns         1.85 ns    374275924
BM_ConstantAutodiff<20, kDynamic>                   74.7 ns         74.7 ns      9134600
BM_ConstantAnalytic<30>                             16.4 ns         16.4 ns     42356437
BM_ConstantAutodiff<30, kNotDynamic>                1.85 ns         1.85 ns    376378806
BM_ConstantAutodiff<30, kDynamic>                   69.9 ns         69.9 ns      9630919
BM_ConstantAnalytic<40>                             19.3 ns         19.3 ns     35900071
BM_ConstantAutodiff<40, kNotDynamic>                1.85 ns         1.85 ns    374628184
BM_ConstantAutodiff<40, kDynamic>                   78.8 ns         78.8 ns      8906277
BM_ConstantAnalytic<50>                             22.4 ns         22.4 ns     30937286
BM_ConstantAutodiff<50, kNotDynamic>                1.85 ns         1.85 ns    373782522
BM_ConstantAutodiff<50, kDynamic>                   82.4 ns         82.4 ns      8240201
BM_ConstantAnalytic<60>                             24.4 ns         24.4 ns     28451093
BM_ConstantAutodiff<60, kNotDynamic>                1.85 ns         1.85 ns    374829748
BM_ConstantAutodiff<60, kDynamic>                   82.1 ns         82.1 ns      8308416
BM_Linear1AutoDiff<kNotDynamic>/0                   1.85 ns         1.85 ns    373901598
BM_Linear1AutoDiff<kNotDynamic>/1                   2.11 ns         2.11 ns    327144528
BM_Linear1AutoDiff<kDynamic>/0                      5.55 ns         5.55 ns    121988031
BM_Linear1AutoDiff<kDynamic>/1                      74.4 ns         74.4 ns      9079372
BM_Linear10AutoDiff<kNotDynamic>/0                  5.45 ns         5.45 ns    124097083
BM_Linear10AutoDiff<kNotDynamic>/1                   117 ns          117 ns      5842397
BM_Linear10AutoDiff<kDynamic>/0                     7.39 ns         7.39 ns     91167201
BM_Linear10AutoDiff<kDynamic>/1                      279 ns          279 ns      2474609
BM_Rat43AutoDiff<kNotDynamic>/0                     27.6 ns         27.6 ns     25139203
BM_Rat43AutoDiff<kNotDynamic>/1                     99.8 ns         99.8 ns      6797236
BM_Rat43AutoDiff<kDynamic>/0                        29.2 ns         29.2 ns     23541878
BM_Rat43AutoDiff<kDynamic>/1                         165 ns          165 ns      4160299
BM_SnavelyReprojectionAutoDiff<kNotDynamic>/0       41.2 ns         41.2 ns     16941339
BM_SnavelyReprojectionAutoDiff<kNotDynamic>/1        433 ns          433 ns      1612921
BM_SnavelyReprojectionAutoDiff<kDynamic>/0          43.4 ns         43.4 ns     16084155
BM_SnavelyReprojectionAutoDiff<kDynamic>/1           750 ns          750 ns       900016
BM_PhotometricAutoDiff<kNotDynamic>/0                795 ns          795 ns       851834
BM_PhotometricAutoDiff<kNotDynamic>/1               5090 ns         5090 ns       131262
BM_PhotometricAutoDiff<kDynamic>/0                   799 ns          799 ns       852333
BM_PhotometricAutoDiff<kDynamic>/1                  7339 ns         7339 ns        93023
BM_RelativePoseAutoDiff<kNotDynamic>/0              62.7 ns         62.7 ns      9234965
BM_RelativePoseAutoDiff<kNotDynamic>/1              1143 ns         1143 ns       599598
BM_RelativePoseAutoDiff<kDynamic>/0                 64.9 ns         64.9 ns     10553484
BM_RelativePoseAutoDiff<kDynamic>/1                 2201 ns         2201 ns       314811
BM_BrdfAutoDiff<kNotDynamic>/0                      61.3 ns         61.3 ns     11172774
BM_BrdfAutoDiff<kNotDynamic>/1                      4040 ns         4040 ns       172634
BM_BrdfAutoDiff<kDynamic>/0                         64.2 ns         64.2 ns     10599325
BM_BrdfAutoDiff<kDynamic>/1                         6773 ns         6773 ns       100289

[-- Attachment #8: schur_eliminator_benchmark-diff --]
[-- Type: text/plain, Size: 1724 bytes --]

Benchmark                                              Time        CPU   Iterations
BM_SchurEliminatorEliminate/10                       -2.1 %     -2.1 %      2.4 %
BM_SchurEliminatorEliminate/64                       -4.4 %     -4.4 %      4.5 %
BM_SchurEliminatorEliminate/512                      -4.0 %     -4.0 %      4.1 %
BM_SchurEliminatorEliminate/4096                     -3.6 %     -3.6 %      4.6 %
BM_SchurEliminatorEliminate/10000                    -3.6 %     -3.6 %      4.1 %
BM_SchurEliminatorForOneFBlockEliminate/10           -7.5 %     -7.5 %      6.9 %
BM_SchurEliminatorForOneFBlockEliminate/64           -7.2 %     -7.2 %      8.1 %
BM_SchurEliminatorForOneFBlockEliminate/512          -6.5 %     -6.5 %      7.0 %
BM_SchurEliminatorForOneFBlockEliminate/4096         -7.4 %     -7.4 %      8.0 %
BM_SchurEliminatorForOneFBlockEliminate/10000        -7.1 %     -7.1 %      7.6 %
BM_SchurEliminatorBackSubstitute/10                  -1.5 %     -1.5 %      0.7 %
BM_SchurEliminatorBackSubstitute/64                  -1.6 %     -1.6 %      1.8 %
BM_SchurEliminatorBackSubstitute/512                 -1.0 %     -1.0 %      0.9 %
BM_SchurEliminatorBackSubstitute/4096                -1.1 %     -1.1 %      1.1 %
BM_SchurEliminatorBackSubstitute/10000               -1.7 %     -1.7 %      1.5 %
BM_SchurEliminatorForOneFBlockBackSubstitute/10      -9.2 %     -9.2 %     10.9 %
BM_SchurEliminatorForOneFBlockBackSubstitute/64      -8.9 %     -8.9 %     10.3 %
BM_SchurEliminatorForOneFBlockBackSubstitute/512    -11.2 %    -11.2 %     11.4 %
BM_SchurEliminatorForOneFBlockBackSubstitute/4096   -10.1 %    -10.1 %     11.9 %
BM_SchurEliminatorForOneFBlockBackSubstitute/10000   23.0 %     23.0 %     -5.9 %

[-- Attachment #9: schur_eliminator_benchmark-tuned --]
[-- Type: text/plain, Size: 2162 bytes --]

---------------------------------------------------------------------------------------------
Benchmark                                                   Time             CPU   Iterations
---------------------------------------------------------------------------------------------
BM_SchurEliminatorEliminate/10                           2519 ns         2519 ns       268829
BM_SchurEliminatorEliminate/64                          14869 ns        14869 ns        46310
BM_SchurEliminatorEliminate/512                        117539 ns       117538 ns         5773
BM_SchurEliminatorEliminate/4096                       948176 ns       948158 ns          725
BM_SchurEliminatorEliminate/10000                     2308337 ns      2308244 ns          303
BM_SchurEliminatorForOneFBlockEliminate/10               1276 ns         1276 ns       533324
BM_SchurEliminatorForOneFBlockEliminate/64               8011 ns         8011 ns        84447
BM_SchurEliminatorForOneFBlockEliminate/512             64582 ns        64582 ns        10339
BM_SchurEliminatorForOneFBlockEliminate/4096           512447 ns       512446 ns         1306
BM_SchurEliminatorForOneFBlockEliminate/10000         1257909 ns      1257886 ns          553
BM_SchurEliminatorBackSubstitute/10                       606 ns          606 ns      1113271
BM_SchurEliminatorBackSubstitute/64                      3699 ns         3699 ns       188719
BM_SchurEliminatorBackSubstitute/512                    29481 ns        29480 ns        23602
BM_SchurEliminatorBackSubstitute/4096                  237311 ns       237307 ns         2939
BM_SchurEliminatorBackSubstitute/10000                 584585 ns       584565 ns         1183
BM_SchurEliminatorForOneFBlockBackSubstitute/10           118 ns          118 ns      5827701
BM_SchurEliminatorForOneFBlockBackSubstitute/64           750 ns          750 ns       902276
BM_SchurEliminatorForOneFBlockBackSubstitute/512         5761 ns         5761 ns       115981
BM_SchurEliminatorForOneFBlockBackSubstitute/4096       48818 ns        48817 ns        13876
BM_SchurEliminatorForOneFBlockBackSubstitute/10000     209774 ns       209766 ns         3890

[-- Attachment #10: schur_eliminator_benchmark-untuned --]
[-- Type: text/plain, Size: 2162 bytes --]

---------------------------------------------------------------------------------------------
Benchmark                                                   Time             CPU   Iterations
---------------------------------------------------------------------------------------------
BM_SchurEliminatorEliminate/10                           2573 ns         2573 ns       262529
BM_SchurEliminatorEliminate/64                          15546 ns        15546 ns        44329
BM_SchurEliminatorEliminate/512                        122405 ns       122402 ns         5548
BM_SchurEliminatorEliminate/4096                       983165 ns       983130 ns          693
BM_SchurEliminatorEliminate/10000                     2393897 ns      2393800 ns          291
BM_SchurEliminatorForOneFBlockEliminate/10               1380 ns         1380 ns       499061
BM_SchurEliminatorForOneFBlockEliminate/64               8633 ns         8632 ns        78098
BM_SchurEliminatorForOneFBlockEliminate/512             69074 ns        69074 ns         9665
BM_SchurEliminatorForOneFBlockEliminate/4096           553412 ns       553392 ns         1209
BM_SchurEliminatorForOneFBlockEliminate/10000         1354469 ns      1354425 ns          514
BM_SchurEliminatorBackSubstitute/10                       615 ns          615 ns      1105884
BM_SchurEliminatorBackSubstitute/64                      3758 ns         3758 ns       185336
BM_SchurEliminatorBackSubstitute/512                    29778 ns        29777 ns        23382
BM_SchurEliminatorBackSubstitute/4096                  239890 ns       239885 ns         2908
BM_SchurEliminatorBackSubstitute/10000                 594449 ns       594438 ns         1165
BM_SchurEliminatorForOneFBlockBackSubstitute/10           130 ns          130 ns      5252703
BM_SchurEliminatorForOneFBlockBackSubstitute/64           823 ns          823 ns       817687
BM_SchurEliminatorForOneFBlockBackSubstitute/512         6485 ns         6485 ns       104099
BM_SchurEliminatorForOneFBlockBackSubstitute/4096       54289 ns        54288 ns        12395
BM_SchurEliminatorForOneFBlockBackSubstitute/10000     170507 ns       170503 ns         4133

[-- Attachment #11: small_blas_gemm_benchmark-diff --]
[-- Type: text/plain, Size: 44856 bytes --]

Benchmark                                              Time        CPU   Iterations
BM_MatrixMatrixMultiplyDynamic/1/1/1                 -1.1 %     -1.1 %      6.0 %
BM_MatrixMatrixMultiplyDynamic/1/1/2                  2.4 %      2.4 %     -3.3 %
BM_MatrixMatrixMultiplyDynamic/1/1/3                  0.1 %      0.1 %      0.7 %
BM_MatrixMatrixMultiplyDynamic/1/1/4                  0.0 %      0.0 %      2.0 %
BM_MatrixMatrixMultiplyDynamic/1/1/8                  0.7 %      0.7 %      0.6 %
BM_MatrixMatrixMultiplyDynamic/1/1/12                 1.7 %      1.7 %     -1.4 %
BM_MatrixMatrixMultiplyDynamic/1/1/15                 0.5 %      0.5 %     -0.1 %
BM_MatrixMatrixMultiplyDynamic/1/2/1                 -0.6 %     -0.6 %      2.7 %
BM_MatrixMatrixMultiplyDynamic/1/2/2                  1.5 %      1.5 %     -0.4 %
BM_MatrixMatrixMultiplyDynamic/1/2/3                 -0.2 %     -0.2 %      1.0 %
BM_MatrixMatrixMultiplyDynamic/1/2/4                 -1.7 %     -1.7 %      2.9 %
BM_MatrixMatrixMultiplyDynamic/1/2/8                  0.6 %      0.6 %     -0.1 %
BM_MatrixMatrixMultiplyDynamic/1/2/12                 1.0 %      1.0 %     -1.0 %
BM_MatrixMatrixMultiplyDynamic/1/2/15                 1.1 %      1.1 %     -1.1 %
BM_MatrixMatrixMultiplyDynamic/1/3/1                  2.5 %      2.5 %     -0.9 %
BM_MatrixMatrixMultiplyDynamic/1/3/2                  1.0 %      1.0 %      0.8 %
BM_MatrixMatrixMultiplyDynamic/1/3/3                 -0.9 %     -0.9 %      1.2 %
BM_MatrixMatrixMultiplyDynamic/1/3/4                  0.0 %      0.0 %      0.1 %
BM_MatrixMatrixMultiplyDynamic/1/3/8                  0.5 %      0.5 %     -1.0 %
BM_MatrixMatrixMultiplyDynamic/1/3/12                 2.5 %      2.5 %     -2.7 %
BM_MatrixMatrixMultiplyDynamic/1/3/15                 2.3 %      2.3 %     -2.2 %
BM_MatrixMatrixMultiplyDynamic/1/4/1                 -3.8 %     -3.8 %      5.3 %
BM_MatrixMatrixMultiplyDynamic/1/4/2                 -0.9 %     -0.9 %      5.2 %
BM_MatrixMatrixMultiplyDynamic/1/4/3                  5.7 %      5.7 %     -4.7 %
BM_MatrixMatrixMultiplyDynamic/1/4/4                  0.7 %      0.7 %      0.5 %
BM_MatrixMatrixMultiplyDynamic/1/4/8                  1.9 %      1.9 %      0.3 %
BM_MatrixMatrixMultiplyDynamic/1/4/12                 2.6 %      2.6 %     -2.6 %
BM_MatrixMatrixMultiplyDynamic/1/4/15                 2.3 %      2.3 %     -2.6 %
BM_MatrixMatrixMultiplyDynamic/1/8/1                 10.0 %     10.0 %     -7.5 %
BM_MatrixMatrixMultiplyDynamic/1/8/2                 -0.8 %     -0.8 %      0.7 %
BM_MatrixMatrixMultiplyDynamic/1/8/3                  1.9 %      1.9 %     -2.0 %
BM_MatrixMatrixMultiplyDynamic/1/8/4                  1.2 %      1.2 %     -1.2 %
BM_MatrixMatrixMultiplyDynamic/1/8/8                  2.1 %      2.1 %     -3.3 %
BM_MatrixMatrixMultiplyDynamic/1/8/12                 3.1 %      3.1 %     -2.0 %
BM_MatrixMatrixMultiplyDynamic/1/8/15                -1.0 %     -1.0 %      7.9 %
BM_MatrixMatrixMultiplyDynamic/1/12/1                 1.7 %      1.7 %     -1.9 %
BM_MatrixMatrixMultiplyDynamic/1/12/2                 2.8 %      2.8 %     -1.0 %
BM_MatrixMatrixMultiplyDynamic/1/12/3                 5.0 %      5.0 %     -5.3 %
BM_MatrixMatrixMultiplyDynamic/1/12/4                 3.5 %      3.5 %     -3.4 %
BM_MatrixMatrixMultiplyDynamic/1/12/8                 2.0 %      2.0 %     -2.0 %
BM_MatrixMatrixMultiplyDynamic/1/12/12                2.7 %      2.7 %      3.5 %
BM_MatrixMatrixMultiplyDynamic/1/12/15               -1.6 %     -1.6 %      4.6 %
BM_MatrixMatrixMultiplyDynamic/1/15/1                 3.8 %      3.8 %     -3.4 %
BM_MatrixMatrixMultiplyDynamic/1/15/2                 3.2 %      3.2 %      2.1 %
BM_MatrixMatrixMultiplyDynamic/1/15/3                -3.8 %     -3.8 %      3.8 %
BM_MatrixMatrixMultiplyDynamic/1/15/4                 1.3 %      1.3 %     -0.4 %
BM_MatrixMatrixMultiplyDynamic/1/15/8                 1.2 %      1.2 %     -1.3 %
BM_MatrixMatrixMultiplyDynamic/1/15/12                0.2 %      0.2 %      4.6 %
BM_MatrixMatrixMultiplyDynamic/1/15/15               -1.2 %     -1.2 %      4.3 %
BM_MatrixMatrixMultiplyDynamic/2/1/1                 -0.7 %     -0.7 %      0.3 %
BM_MatrixMatrixMultiplyDynamic/2/1/2                  2.1 %      2.1 %      0.8 %
BM_MatrixMatrixMultiplyDynamic/2/1/3                 -0.9 %     -0.9 %      1.3 %
BM_MatrixMatrixMultiplyDynamic/2/1/4                  0.0 %      0.0 %      0.2 %
BM_MatrixMatrixMultiplyDynamic/2/1/8                  2.0 %      2.0 %     -1.7 %
BM_MatrixMatrixMultiplyDynamic/2/1/12                 1.5 %      1.5 %     -1.9 %
BM_MatrixMatrixMultiplyDynamic/2/1/15                 0.0 %      0.0 %     -0.5 %
BM_MatrixMatrixMultiplyDynamic/2/2/1                  1.3 %      1.3 %      0.9 %
BM_MatrixMatrixMultiplyDynamic/2/2/2                 -2.7 %     -2.7 %      3.4 %
BM_MatrixMatrixMultiplyDynamic/2/2/3                  3.1 %      3.1 %     -2.5 %
BM_MatrixMatrixMultiplyDynamic/2/2/4                  0.0 %      0.0 %      0.1 %
BM_MatrixMatrixMultiplyDynamic/2/2/8                  3.1 %      3.1 %     -3.8 %
BM_MatrixMatrixMultiplyDynamic/2/2/12                 3.3 %      3.3 %     -3.8 %
BM_MatrixMatrixMultiplyDynamic/2/2/15                 2.1 %      2.1 %     -2.3 %
BM_MatrixMatrixMultiplyDynamic/2/3/1                  1.7 %      1.7 %     -1.8 %
BM_MatrixMatrixMultiplyDynamic/2/3/2                  0.0 %      0.0 %      0.4 %
BM_MatrixMatrixMultiplyDynamic/2/3/3                  1.3 %      1.3 %     -1.9 %
BM_MatrixMatrixMultiplyDynamic/2/3/4                  1.2 %      1.2 %     -2.0 %
BM_MatrixMatrixMultiplyDynamic/2/3/8                  1.9 %      1.9 %     -1.9 %
BM_MatrixMatrixMultiplyDynamic/2/3/12                 3.1 %      3.1 %     -3.2 %
BM_MatrixMatrixMultiplyDynamic/2/3/15                -0.6 %     -0.6 %      8.1 %
BM_MatrixMatrixMultiplyDynamic/2/4/1                  0.6 %      0.6 %     -0.5 %
BM_MatrixMatrixMultiplyDynamic/2/4/2                  0.8 %      0.8 %      2.3 %
BM_MatrixMatrixMultiplyDynamic/2/4/3                 -1.2 %     -1.2 %      0.8 %
BM_MatrixMatrixMultiplyDynamic/2/4/4                  0.0 %      0.0 %     -0.3 %
BM_MatrixMatrixMultiplyDynamic/2/4/8                  0.3 %      0.3 %     -1.6 %
BM_MatrixMatrixMultiplyDynamic/2/4/12                 2.6 %      2.6 %     -2.6 %
BM_MatrixMatrixMultiplyDynamic/2/4/15                -0.9 %     -0.9 %      6.6 %
BM_MatrixMatrixMultiplyDynamic/2/8/1                  3.6 %      3.6 %     -2.9 %
BM_MatrixMatrixMultiplyDynamic/2/8/2                  4.3 %      4.3 %     -1.4 %
BM_MatrixMatrixMultiplyDynamic/2/8/3                  0.4 %      0.0 %      0.1 %
BM_MatrixMatrixMultiplyDynamic/2/8/4                  0.8 %      0.8 %     -1.0 %
BM_MatrixMatrixMultiplyDynamic/2/8/8                  1.9 %      1.9 %     -2.2 %
BM_MatrixMatrixMultiplyDynamic/2/8/12               -11.1 %    -11.1 %      4.7 %
BM_MatrixMatrixMultiplyDynamic/2/8/15                -5.2 %     -5.2 %     44.5 %
BM_MatrixMatrixMultiplyDynamic/2/12/1                 6.6 %      6.6 %     -5.1 %
BM_MatrixMatrixMultiplyDynamic/2/12/2                 2.8 %      2.8 %     -1.2 %
BM_MatrixMatrixMultiplyDynamic/2/12/3                 5.6 %      5.6 %     -5.8 %
BM_MatrixMatrixMultiplyDynamic/2/12/4                -5.4 %     -5.4 %      3.5 %
BM_MatrixMatrixMultiplyDynamic/2/12/8                -1.4 %     -1.4 %      5.8 %
BM_MatrixMatrixMultiplyDynamic/2/12/12               -6.6 %     -6.6 %      6.1 %
BM_MatrixMatrixMultiplyDynamic/2/12/15               -1.8 %     -1.8 %      2.7 %
BM_MatrixMatrixMultiplyDynamic/2/15/1                13.2 %     13.2 %    -11.6 %
BM_MatrixMatrixMultiplyDynamic/2/15/2                 3.9 %      3.9 %     -1.7 %
BM_MatrixMatrixMultiplyDynamic/2/15/3                 9.3 %      9.3 %     -8.8 %
BM_MatrixMatrixMultiplyDynamic/2/15/4                -1.8 %     -1.8 %      2.6 %
BM_MatrixMatrixMultiplyDynamic/2/15/8               -14.3 %    -14.3 %     17.7 %
BM_MatrixMatrixMultiplyDynamic/2/15/12              -12.7 %    -12.7 %      5.1 %
BM_MatrixMatrixMultiplyDynamic/2/15/15               -2.1 %     -2.1 %     14.6 %
BM_MatrixMatrixMultiplyDynamic/3/1/1                  0.4 %      0.4 %      0.0 %
BM_MatrixMatrixMultiplyDynamic/3/1/2                 10.4 %     10.4 %     -7.5 %
BM_MatrixMatrixMultiplyDynamic/3/1/3                 -1.5 %     -1.5 %      2.4 %
BM_MatrixMatrixMultiplyDynamic/3/1/4                  0.6 %      0.6 %     -0.1 %
BM_MatrixMatrixMultiplyDynamic/3/1/8                  3.3 %      3.3 %     -3.2 %
BM_MatrixMatrixMultiplyDynamic/3/1/12                 3.6 %      3.6 %     -3.5 %
BM_MatrixMatrixMultiplyDynamic/3/1/15                -0.7 %     -0.7 %      0.7 %
BM_MatrixMatrixMultiplyDynamic/3/2/1                 -0.9 %     -0.9 %      1.3 %
BM_MatrixMatrixMultiplyDynamic/3/2/2                  0.0 %      0.0 %      0.4 %
BM_MatrixMatrixMultiplyDynamic/3/2/3                 -1.2 %     -1.2 %      1.3 %
BM_MatrixMatrixMultiplyDynamic/3/2/4                  0.0 %      0.0 %     -0.5 %
BM_MatrixMatrixMultiplyDynamic/3/2/8                  2.1 %      2.1 %     -2.2 %
BM_MatrixMatrixMultiplyDynamic/3/2/12                 2.5 %      2.5 %     -2.8 %
BM_MatrixMatrixMultiplyDynamic/3/2/15                -1.1 %     -1.1 %      7.8 %
BM_MatrixMatrixMultiplyDynamic/3/3/1                 -3.4 %     -3.4 %      3.1 %
BM_MatrixMatrixMultiplyDynamic/3/3/2                  1.5 %      1.5 %     -0.2 %
BM_MatrixMatrixMultiplyDynamic/3/3/3                 -7.7 %     -7.7 %      7.0 %
BM_MatrixMatrixMultiplyDynamic/3/3/4                  0.0 %      0.0 %     -1.0 %
BM_MatrixMatrixMultiplyDynamic/3/3/8                -12.1 %    -12.1 %     -1.1 %
BM_MatrixMatrixMultiplyDynamic/3/3/12                -1.2 %     -1.2 %     13.3 %
BM_MatrixMatrixMultiplyDynamic/3/3/15                 0.2 %      0.2 %      2.3 %
BM_MatrixMatrixMultiplyDynamic/3/4/1                  0.8 %      0.8 %     -2.6 %
BM_MatrixMatrixMultiplyDynamic/3/4/2                  4.1 %      4.1 %     -2.9 %
BM_MatrixMatrixMultiplyDynamic/3/4/3                 -3.9 %     -3.9 %      3.5 %
BM_MatrixMatrixMultiplyDynamic/3/4/4                  0.8 %      0.8 %     -1.5 %
BM_MatrixMatrixMultiplyDynamic/3/4/8                 -3.3 %     -3.3 %     -2.7 %
BM_MatrixMatrixMultiplyDynamic/3/4/12                -2.2 %     -2.2 %     33.1 %
BM_MatrixMatrixMultiplyDynamic/3/4/15                 1.3 %      1.3 %     -0.4 %
BM_MatrixMatrixMultiplyDynamic/3/8/1                  2.4 %      2.4 %     -3.1 %
BM_MatrixMatrixMultiplyDynamic/3/8/2                 -4.9 %     -4.9 %      5.3 %
BM_MatrixMatrixMultiplyDynamic/3/8/3                  2.8 %      2.8 %     -3.5 %
BM_MatrixMatrixMultiplyDynamic/3/8/4                 -2.6 %     -2.6 %     -0.6 %
BM_MatrixMatrixMultiplyDynamic/3/8/8                 -1.0 %     -1.0 %      3.7 %
BM_MatrixMatrixMultiplyDynamic/3/8/12                -2.4 %     -2.4 %      3.0 %
BM_MatrixMatrixMultiplyDynamic/3/8/15                -0.9 %     -0.9 %      1.9 %
BM_MatrixMatrixMultiplyDynamic/3/12/1                -8.7 %     -8.7 %      8.1 %
BM_MatrixMatrixMultiplyDynamic/3/12/2                -9.1 %     -9.1 %     -2.4 %
BM_MatrixMatrixMultiplyDynamic/3/12/3                -3.9 %     -3.9 %     -6.1 %
BM_MatrixMatrixMultiplyDynamic/3/12/4                -4.8 %     -4.8 %     34.1 %
BM_MatrixMatrixMultiplyDynamic/3/12/8               -13.5 %    -13.5 %      7.7 %
BM_MatrixMatrixMultiplyDynamic/3/12/12               -5.0 %     -5.0 %      5.0 %
BM_MatrixMatrixMultiplyDynamic/3/12/15               -3.1 %     -3.1 %      2.4 %
BM_MatrixMatrixMultiplyDynamic/3/15/1                16.2 %     16.2 %    -14.0 %
BM_MatrixMatrixMultiplyDynamic/3/15/2                -4.6 %     -4.6 %     -6.7 %
BM_MatrixMatrixMultiplyDynamic/3/15/3                12.5 %     12.5 %     -7.7 %
BM_MatrixMatrixMultiplyDynamic/3/15/4                -3.9 %     -3.9 %     10.1 %
BM_MatrixMatrixMultiplyDynamic/3/15/8                -3.7 %     -3.7 %      4.6 %
BM_MatrixMatrixMultiplyDynamic/3/15/12               -3.4 %     -3.4 %      3.3 %
BM_MatrixMatrixMultiplyDynamic/3/15/15               -1.0 %     -1.0 %     -0.3 %
BM_MatrixMatrixMultiplyDynamic/4/1/1                 -1.4 %     -1.4 %      1.7 %
BM_MatrixMatrixMultiplyDynamic/4/1/2                 -2.4 %     -2.4 %      2.9 %
BM_MatrixMatrixMultiplyDynamic/4/1/3                 -0.6 %     -0.6 %      0.5 %
BM_MatrixMatrixMultiplyDynamic/4/1/4                  1.7 %      1.7 %     -2.0 %
BM_MatrixMatrixMultiplyDynamic/4/1/8                -21.0 %    -21.0 %     -3.6 %
BM_MatrixMatrixMultiplyDynamic/4/1/12               -12.4 %    -12.4 %     -4.0 %
BM_MatrixMatrixMultiplyDynamic/4/1/15                 0.4 %      0.4 %      2.7 %
BM_MatrixMatrixMultiplyDynamic/4/2/1                  5.2 %      5.2 %     -5.5 %
BM_MatrixMatrixMultiplyDynamic/4/2/2                  0.7 %      0.7 %      0.6 %
BM_MatrixMatrixMultiplyDynamic/4/2/3                 -2.5 %     -2.5 %      2.6 %
BM_MatrixMatrixMultiplyDynamic/4/2/4                  2.4 %      2.4 %     -2.4 %
BM_MatrixMatrixMultiplyDynamic/4/2/8                -19.2 %    -19.2 %     -1.8 %
BM_MatrixMatrixMultiplyDynamic/4/2/12               -26.2 %    -26.2 %      1.9 %
BM_MatrixMatrixMultiplyDynamic/4/2/15                -1.3 %     -1.3 %     31.1 %
BM_MatrixMatrixMultiplyDynamic/4/3/1                -28.2 %    -28.2 %     65.9 %
BM_MatrixMatrixMultiplyDynamic/4/3/2                  2.6 %      2.6 %     -2.9 %
BM_MatrixMatrixMultiplyDynamic/4/3/3                 -2.1 %     -2.1 %      1.9 %
BM_MatrixMatrixMultiplyDynamic/4/3/4                -29.2 %    -29.2 %     -0.3 %
BM_MatrixMatrixMultiplyDynamic/4/3/8                 -4.2 %     -4.2 %      9.0 %
BM_MatrixMatrixMultiplyDynamic/4/3/12                -2.7 %     -2.7 %     16.5 %
BM_MatrixMatrixMultiplyDynamic/4/3/15                -5.7 %     -5.7 %      7.4 %
BM_MatrixMatrixMultiplyDynamic/4/4/1                  1.4 %      1.4 %      2.1 %
BM_MatrixMatrixMultiplyDynamic/4/4/2                  2.3 %      2.3 %     -2.6 %
BM_MatrixMatrixMultiplyDynamic/4/4/3                 -0.4 %     -0.4 %     -0.5 %
BM_MatrixMatrixMultiplyDynamic/4/4/4                  1.0 %      1.0 %     -1.0 %
BM_MatrixMatrixMultiplyDynamic/4/4/8                 -3.6 %     -3.6 %      2.9 %
BM_MatrixMatrixMultiplyDynamic/4/4/12                 1.5 %      1.5 %     -0.7 %
BM_MatrixMatrixMultiplyDynamic/4/4/15                 0.1 %      0.1 %     -0.6 %
BM_MatrixMatrixMultiplyDynamic/4/8/1                -10.4 %    -10.4 %     10.5 %
BM_MatrixMatrixMultiplyDynamic/4/8/2                -14.2 %    -14.2 %      0.3 %
BM_MatrixMatrixMultiplyDynamic/4/8/3                 -9.8 %     -9.8 %     -4.1 %
BM_MatrixMatrixMultiplyDynamic/4/8/4                -11.0 %    -11.0 %     -0.2 %
BM_MatrixMatrixMultiplyDynamic/4/8/8                -14.3 %    -14.3 %      5.0 %
BM_MatrixMatrixMultiplyDynamic/4/8/12                -2.7 %     -2.7 %     37.9 %
BM_MatrixMatrixMultiplyDynamic/4/8/15               -13.0 %    -13.0 %      1.6 %
BM_MatrixMatrixMultiplyDynamic/4/12/1                -1.4 %     -1.4 %      1.2 %
BM_MatrixMatrixMultiplyDynamic/4/12/2                 6.3 %      6.3 %    -18.5 %
BM_MatrixMatrixMultiplyDynamic/4/12/3                -2.1 %     -2.1 %     27.0 %
BM_MatrixMatrixMultiplyDynamic/4/12/4                -2.9 %     -2.9 %      8.2 %
BM_MatrixMatrixMultiplyDynamic/4/12/8               -18.8 %    -18.8 %     20.1 %
BM_MatrixMatrixMultiplyDynamic/4/12/12              -20.9 %    -20.9 %      6.1 %
BM_MatrixMatrixMultiplyDynamic/4/12/15               -5.0 %     -5.0 %      4.0 %
BM_MatrixMatrixMultiplyDynamic/4/15/1                -4.1 %     -4.1 %     12.3 %
BM_MatrixMatrixMultiplyDynamic/4/15/2                 9.0 %      9.0 %     -8.4 %
BM_MatrixMatrixMultiplyDynamic/4/15/3                 1.3 %      1.3 %     22.0 %
BM_MatrixMatrixMultiplyDynamic/4/15/4                -4.5 %     -4.5 %      5.6 %
BM_MatrixMatrixMultiplyDynamic/4/15/8               -12.7 %    -12.7 %      8.7 %
BM_MatrixMatrixMultiplyDynamic/4/15/12              -11.0 %    -11.0 %      7.5 %
BM_MatrixMatrixMultiplyDynamic/4/15/15               -2.8 %     -2.8 %     17.6 %
BM_MatrixMatrixMultiplyDynamic/6/1/1                  0.0 %      0.0 %     -2.0 %
BM_MatrixMatrixMultiplyDynamic/6/1/2                  0.7 %      0.7 %     -2.5 %
BM_MatrixMatrixMultiplyDynamic/6/1/3                 -3.8 %     -3.8 %      3.8 %
BM_MatrixMatrixMultiplyDynamic/6/1/4                  1.3 %      1.3 %     -1.5 %
BM_MatrixMatrixMultiplyDynamic/6/1/8                  2.6 %      2.6 %     -2.7 %
BM_MatrixMatrixMultiplyDynamic/6/1/12                -8.0 %     -8.0 %      2.0 %
BM_MatrixMatrixMultiplyDynamic/6/1/15                17.3 %     17.3 %    -14.3 %
BM_MatrixMatrixMultiplyDynamic/6/2/1                 -3.5 %     -3.5 %     18.9 %
BM_MatrixMatrixMultiplyDynamic/6/2/2                  0.6 %      0.6 %     -1.3 %
BM_MatrixMatrixMultiplyDynamic/6/2/3                 -3.7 %     -3.7 %      2.6 %
BM_MatrixMatrixMultiplyDynamic/6/2/4                  0.7 %      1.1 %     -1.4 %
BM_MatrixMatrixMultiplyDynamic/6/2/8                 -7.6 %     -7.6 %      4.0 %
BM_MatrixMatrixMultiplyDynamic/6/2/12                -5.7 %     -5.7 %      5.0 %
BM_MatrixMatrixMultiplyDynamic/6/2/15                -7.2 %     -7.2 %      3.4 %
BM_MatrixMatrixMultiplyDynamic/6/3/1                  0.7 %      0.7 %     -0.8 %
BM_MatrixMatrixMultiplyDynamic/6/3/2                 -6.5 %     -6.5 %      7.1 %
BM_MatrixMatrixMultiplyDynamic/6/3/3                 -0.3 %     -0.3 %     -0.4 %
BM_MatrixMatrixMultiplyDynamic/6/3/4                -16.0 %    -16.0 %      1.0 %
BM_MatrixMatrixMultiplyDynamic/6/3/8                 -0.2 %     -0.2 %     23.4 %
BM_MatrixMatrixMultiplyDynamic/6/3/12                -1.8 %     -1.8 %      2.2 %
BM_MatrixMatrixMultiplyDynamic/6/3/15                 0.0 %      0.0 %      2.0 %
BM_MatrixMatrixMultiplyDynamic/6/4/1                -11.2 %    -11.2 %      9.5 %
BM_MatrixMatrixMultiplyDynamic/6/4/2                -24.3 %    -24.3 %      7.4 %
BM_MatrixMatrixMultiplyDynamic/6/4/3                  0.8 %      0.8 %     -2.1 %
BM_MatrixMatrixMultiplyDynamic/6/4/4                  1.6 %      1.6 %      5.3 %
BM_MatrixMatrixMultiplyDynamic/6/4/8                  1.9 %      1.9 %      0.1 %
BM_MatrixMatrixMultiplyDynamic/6/4/12                 1.0 %      1.0 %     -0.4 %
BM_MatrixMatrixMultiplyDynamic/6/4/15                 0.7 %      0.7 %     -1.2 %
BM_MatrixMatrixMultiplyDynamic/6/8/1                  5.6 %      5.6 %     -5.2 %
BM_MatrixMatrixMultiplyDynamic/6/8/2                -13.7 %    -13.7 %     -0.6 %
BM_MatrixMatrixMultiplyDynamic/6/8/3                  2.9 %      2.9 %     -2.4 %
BM_MatrixMatrixMultiplyDynamic/6/8/4                 -2.4 %     -2.4 %      4.4 %
BM_MatrixMatrixMultiplyDynamic/6/8/8                 -3.6 %     -3.6 %     -4.9 %
BM_MatrixMatrixMultiplyDynamic/6/8/12                -4.3 %     -4.3 %      5.0 %
BM_MatrixMatrixMultiplyDynamic/6/8/15                -5.4 %     -5.4 %      6.8 %
BM_MatrixMatrixMultiplyDynamic/6/12/1                 6.3 %      6.3 %    -11.7 %
BM_MatrixMatrixMultiplyDynamic/6/12/2               -15.4 %    -15.4 %     -1.2 %
BM_MatrixMatrixMultiplyDynamic/6/12/3                 8.6 %      8.6 %     18.8 %
BM_MatrixMatrixMultiplyDynamic/6/12/4               -25.0 %    -25.0 %      8.4 %
BM_MatrixMatrixMultiplyDynamic/6/12/8               -14.5 %    -14.5 %      4.8 %
BM_MatrixMatrixMultiplyDynamic/6/12/12               -8.5 %     -8.5 %      6.8 %
BM_MatrixMatrixMultiplyDynamic/6/12/15               -3.1 %     -3.1 %      2.2 %
BM_MatrixMatrixMultiplyDynamic/6/15/1                15.0 %     15.0 %    -15.3 %
BM_MatrixMatrixMultiplyDynamic/6/15/2               -24.3 %    -24.3 %     -2.4 %
BM_MatrixMatrixMultiplyDynamic/6/15/3                14.2 %     14.2 %    -11.8 %
BM_MatrixMatrixMultiplyDynamic/6/15/4                -8.4 %     -8.4 %      8.8 %
BM_MatrixMatrixMultiplyDynamic/6/15/8                -5.3 %     -5.3 %      4.1 %
BM_MatrixMatrixMultiplyDynamic/6/15/12               -3.9 %     -3.9 %      3.3 %
BM_MatrixMatrixMultiplyDynamic/6/15/15                0.0 %      0.0 %     -0.0 %
BM_MatrixMatrixMultiplyDynamic/8/1/1                 -1.5 %     -1.5 %     -0.5 %
BM_MatrixMatrixMultiplyDynamic/8/1/2                  3.1 %      3.1 %     -3.1 %
BM_MatrixMatrixMultiplyDynamic/8/1/3                 -3.5 %     -3.5 %      4.5 %
BM_MatrixMatrixMultiplyDynamic/8/1/4                -20.4 %    -20.4 %     -0.8 %
BM_MatrixMatrixMultiplyDynamic/8/1/8                 -5.6 %     -5.6 %      0.4 %
BM_MatrixMatrixMultiplyDynamic/8/1/12                 1.5 %      1.5 %     -1.1 %
BM_MatrixMatrixMultiplyDynamic/8/1/15                -1.0 %     -1.0 %     15.0 %
BM_MatrixMatrixMultiplyDynamic/8/2/1                 -5.4 %     -5.4 %      6.3 %
BM_MatrixMatrixMultiplyDynamic/8/2/2                -21.8 %    -21.8 %      8.4 %
BM_MatrixMatrixMultiplyDynamic/8/2/3                 -0.9 %     -0.9 %      1.3 %
BM_MatrixMatrixMultiplyDynamic/8/2/4                -16.6 %    -16.6 %     -1.1 %
BM_MatrixMatrixMultiplyDynamic/8/2/8                -27.5 %    -27.5 %      7.0 %
BM_MatrixMatrixMultiplyDynamic/8/2/12                 0.1 %      0.1 %     17.1 %
BM_MatrixMatrixMultiplyDynamic/8/2/15                -7.2 %     -7.2 %      3.6 %
BM_MatrixMatrixMultiplyDynamic/8/3/1                 -1.9 %     -1.9 %      2.3 %
BM_MatrixMatrixMultiplyDynamic/8/3/2                -30.4 %    -30.4 %      7.1 %
BM_MatrixMatrixMultiplyDynamic/8/3/3                  0.5 %      0.5 %     -0.5 %
BM_MatrixMatrixMultiplyDynamic/8/3/4                -20.5 %    -20.5 %     33.2 %
BM_MatrixMatrixMultiplyDynamic/8/3/8                 -9.0 %     -9.0 %     10.2 %
BM_MatrixMatrixMultiplyDynamic/8/3/12                -6.4 %     -6.4 %      8.3 %
BM_MatrixMatrixMultiplyDynamic/8/3/15                -2.5 %     -2.5 %      2.1 %
BM_MatrixMatrixMultiplyDynamic/8/4/1                 -0.8 %     -0.8 %     -0.7 %
BM_MatrixMatrixMultiplyDynamic/8/4/2                 -4.5 %     -4.5 %     -0.7 %
BM_MatrixMatrixMultiplyDynamic/8/4/3                  1.7 %      1.7 %     -3.5 %
BM_MatrixMatrixMultiplyDynamic/8/4/4                  0.2 %      0.2 %     -0.4 %
BM_MatrixMatrixMultiplyDynamic/8/4/8                 -3.1 %     -3.1 %      5.6 %
BM_MatrixMatrixMultiplyDynamic/8/4/12                -2.2 %     -2.2 %      1.3 %
BM_MatrixMatrixMultiplyDynamic/8/4/15                -2.7 %     -2.7 %      3.5 %
BM_MatrixMatrixMultiplyDynamic/8/8/1                  0.0 %      0.0 %     -0.3 %
BM_MatrixMatrixMultiplyDynamic/8/8/2                 -0.4 %     -0.4 %      1.1 %
BM_MatrixMatrixMultiplyDynamic/8/8/3                 -0.3 %     -0.3 %     -0.6 %
BM_MatrixMatrixMultiplyDynamic/8/8/4                 -7.7 %     -7.7 %      9.2 %
BM_MatrixMatrixMultiplyDynamic/8/8/8                 -7.4 %     -7.4 %      7.9 %
BM_MatrixMatrixMultiplyDynamic/8/8/12                -8.7 %     -8.7 %      9.0 %
BM_MatrixMatrixMultiplyDynamic/8/8/15                -4.8 %     -4.8 %      5.0 %
BM_MatrixMatrixMultiplyDynamic/8/12/1                 9.5 %      9.5 %     -9.0 %
BM_MatrixMatrixMultiplyDynamic/8/12/2                 3.7 %      3.7 %     -1.6 %
BM_MatrixMatrixMultiplyDynamic/8/12/3                 5.6 %      5.6 %     -6.6 %
BM_MatrixMatrixMultiplyDynamic/8/12/4               -10.7 %    -10.7 %     13.0 %
BM_MatrixMatrixMultiplyDynamic/8/12/8               -11.5 %    -11.5 %     12.4 %
BM_MatrixMatrixMultiplyDynamic/8/12/12              -11.7 %    -11.4 %     12.7 %
BM_MatrixMatrixMultiplyDynamic/8/12/15               -6.2 %     -6.2 %      7.1 %
BM_MatrixMatrixMultiplyDynamic/8/15/1                11.7 %     11.7 %    -11.1 %
BM_MatrixMatrixMultiplyDynamic/8/15/2                 5.1 %      5.1 %     -3.7 %
BM_MatrixMatrixMultiplyDynamic/8/15/3                 8.0 %      8.0 %     -7.9 %
BM_MatrixMatrixMultiplyDynamic/8/15/4               -12.8 %    -12.8 %     13.7 %
BM_MatrixMatrixMultiplyDynamic/8/15/8                -9.6 %     -9.6 %     10.2 %
BM_MatrixMatrixMultiplyDynamic/8/15/12               -9.3 %     -9.3 %      9.5 %
BM_MatrixMatrixMultiplyDynamic/8/15/15               -3.5 %     -3.5 %      2.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/1         1.6 %      1.6 %     -0.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/2        -5.8 %     -5.8 %      9.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/3         1.3 %      1.3 %      1.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/4         0.2 %      0.2 %      2.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/6         0.0 %      0.0 %      1.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/8         0.0 %      0.0 %     -0.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/1         0.0 %      0.0 %      2.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/2         4.3 %      4.3 %      0.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/3         0.0 %      0.0 %      3.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/4        -1.9 %     -1.9 %      1.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/6         2.3 %      2.3 %     -2.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/8         0.4 %      0.4 %     -0.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/1         2.2 %      2.0 %     -3.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/2         0.6 %      0.6 %     -0.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/3         0.7 %      0.7 %     -0.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/4        -1.1 %     -1.1 %     -0.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/6         3.1 %      3.1 %     -3.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/8        -1.0 %     -1.0 %     -0.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/1         3.7 %      3.7 %      7.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/2         0.7 %      0.7 %     -0.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/3         2.4 %      2.4 %     -2.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/4        -0.9 %     -0.9 %      0.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/6         1.4 %      1.4 %     -0.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/8        -1.2 %     -1.2 %     -0.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/1         5.6 %      5.6 %     -6.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/2         3.3 %      3.3 %     -3.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/3         2.2 %      2.2 %     -2.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/4        -6.3 %     -6.3 %      6.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/6        -3.8 %     -3.8 %      5.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/8        -4.9 %     -4.9 %      5.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/1        2.1 %      2.1 %     -2.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/2       -1.7 %     -1.7 %      1.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/3       -0.5 %     -0.5 %      0.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/4       -7.1 %     -7.1 %      7.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/6       -3.3 %     -3.3 %      4.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/8       -0.6 %     -0.6 %     -0.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/1        8.4 %      8.4 %     -5.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/2       -2.5 %     -2.5 %      2.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/3       -2.3 %     -2.3 %      2.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/4       -5.7 %     -5.7 %      6.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/6       -2.8 %     -2.8 %      2.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/8       -6.8 %     -6.8 %      7.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/1        -5.2 %     -5.2 %      5.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/2        -5.1 %     -5.1 %      3.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/3        -4.1 %     -4.1 %      4.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/4         0.9 %      0.9 %      1.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/6         2.0 %      2.0 %     -1.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/8        -1.6 %     -1.6 %      1.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/1         2.6 %      2.6 %     -2.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/2        -1.3 %     -1.2 %      0.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/3         2.1 %      2.1 %      0.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/4        -0.6 %     -0.6 %     -1.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/6        -3.2 %     -3.2 %      2.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/8        -4.9 %     -4.9 %      4.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/1         3.0 %      3.0 %     -5.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/2        -0.7 %      0.0 %     -0.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/3         1.2 %      1.2 %     -1.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/4         0.5 %      0.5 %      0.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/6        -0.7 %     -0.7 %      1.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/8        -2.8 %     -2.8 %      2.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/1        -3.2 %     -3.2 %      2.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/2        -3.1 %     -3.1 %      0.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/3         1.4 %      1.4 %     -1.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/4        -3.1 %     -3.1 %      3.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/6        -3.8 %     -3.8 %      4.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/8        -5.6 %     -5.6 %      7.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/1         2.3 %      2.3 %     -2.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/2        -2.6 %     -2.6 %      3.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/3        -2.3 %     -2.3 %      3.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/4        -9.2 %     -9.2 %      9.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/6        -7.8 %     -7.8 %      6.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/8        -7.9 %     -7.9 %     -0.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/1        5.5 %      5.5 %     -5.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/2       -6.6 %     -6.6 %      6.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/3       -5.2 %     -5.2 %      6.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/4       -8.4 %     -8.4 %      9.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/6       -6.9 %     -6.9 %     -4.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/8       -8.9 %     -8.9 %      3.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/1        5.0 %      5.0 %     -4.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/2       -7.4 %     -7.4 %      7.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/3       -4.2 %     -4.2 %      3.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/4       -7.9 %     -7.9 %     -2.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/6       -6.3 %     -6.3 %     -4.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/8       -9.6 %     -9.6 %      5.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/1         2.3 %      2.3 %     -6.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/2        -0.8 %     -0.8 %     -0.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/3        -3.5 %     -3.5 %      3.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/4        -1.4 %     -1.4 %      1.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/6         1.8 %      1.8 %     -2.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/8        -3.2 %     -3.2 %      3.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/1         7.6 %      7.6 %     -6.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/2        -9.6 %     -9.6 %      9.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/3        -1.8 %     -1.8 %      2.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/4        -0.5 %     -0.5 %      0.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/6        -3.4 %     -3.4 %      3.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/8        -2.5 %     -2.5 %      2.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/1        -3.9 %     -3.9 %      4.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/2        -1.9 %     -1.9 %      1.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/3        -2.8 %     -3.2 %      3.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/4        -1.2 %     -1.2 %      0.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/6        -4.1 %     -4.1 %      4.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/8        -5.3 %     -5.3 %      5.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/1         2.1 %      2.1 %     -2.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/2         0.6 %      0.6 %     -0.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/3         1.2 %      1.2 %     -1.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/4        -7.0 %     -7.0 %      7.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/6        -6.3 %     -6.3 %      6.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/8        -7.7 %     -7.7 %     12.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/1         8.8 %      8.8 %     -7.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/2        -5.3 %     -5.3 %      5.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/3        -2.3 %     -2.3 %    -21.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/4       -10.9 %    -10.9 %     15.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/6         6.1 %      6.1 %    -13.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/8        -2.4 %     -2.4 %     -3.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/1        6.1 %      6.1 %     -5.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/2       -8.5 %     -8.5 %      8.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/3        3.5 %      3.5 %     -2.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/4       -8.7 %     -8.7 %     14.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/6       -7.9 %     -7.9 %      9.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/8      -12.6 %    -12.6 %     13.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/1        6.4 %      6.4 %     -5.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/2      -10.1 %    -10.1 %     13.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/3       -3.6 %     -3.6 %      4.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/4       -9.8 %     -9.8 %     12.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/6       -8.3 %     -8.3 %      8.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/8      -11.8 %    -11.8 %     13.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/1         3.8 %      3.8 %     -5.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/2         0.4 %      0.4 %     -2.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/3        -0.7 %     -0.7 %     -0.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/4         0.6 %      0.6 %     -0.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/6         1.1 %      1.1 %     -1.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/8        -2.5 %     -2.5 %      1.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/1        -0.1 %     -0.1 %      0.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/2         4.8 %      4.8 %     -4.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/3        -0.5 %     -0.5 %     -0.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/4        -1.9 %     -1.9 %      1.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/6        -2.1 %     -2.1 %      1.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/8        -4.0 %     -4.0 %      4.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/1         2.8 %      2.8 %     -3.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/2         1.2 %      1.2 %     -1.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/3         0.4 %      0.4 %     -0.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/4        -1.9 %     -1.9 %      1.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/6        -2.7 %     -2.7 %      2.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/8        -4.4 %     -4.4 %      4.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/1         1.2 %      1.2 %     -2.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/2         1.0 %      1.0 %     -3.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/3        -0.3 %     -0.3 %      0.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/4        -3.2 %     -3.2 %      3.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/6        -3.3 %     -3.3 %      3.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/8        -3.9 %     -3.9 %      1.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/1         1.9 %      1.9 %     -3.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/2        -7.4 %     -7.4 %      7.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/3        -1.2 %     -1.2 %      1.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/4        -5.0 %     -5.0 %      4.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/6        -6.2 %     -6.2 %      6.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/8        -4.4 %     -4.4 %      5.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/1        5.7 %      5.7 %     -5.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/2      -10.1 %    -10.1 %      8.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/3       -2.6 %     -2.6 %      1.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/4       -5.0 %     -5.0 %      5.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/6       -7.4 %     -7.4 %      7.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/8       -5.6 %     -5.6 %      5.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/1       10.3 %     10.3 %    -10.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/2       -8.8 %     -8.8 %     12.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/3       -1.2 %     -1.2 %      1.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/4       -5.5 %     -5.5 %      5.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/6       -7.4 %     -7.4 %      7.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/8       -5.6 %     -5.6 %      6.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/1        10.7 %     10.7 %     -9.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/2        -6.5 %     -6.5 %      5.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/3         6.4 %      6.4 %     -3.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/4        -2.2 %     -2.2 %      2.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/6        -4.4 %     -3.9 %      4.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/8        -8.7 %     -8.7 %      3.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/1         1.4 %      1.4 %      7.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/2         0.0 %     -0.6 %     -0.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/3        -3.6 %     -3.6 %      3.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/4        -3.9 %     -3.9 %      4.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/6        -3.1 %     -3.1 %      3.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/8        -5.1 %     -5.1 %      4.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/1         4.3 %      5.0 %     -4.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/2        -1.0 %     -1.0 %      5.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/3         0.0 %      0.0 %     -0.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/4        -5.1 %     -5.1 %      5.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/6        -4.4 %     -4.4 %      3.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/8        -5.1 %     -5.1 %      6.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/1         7.1 %      7.1 %     -6.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/2        -8.9 %     -8.9 %      9.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/3        -3.8 %     -3.8 %      6.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/4        -5.0 %     -5.0 %      5.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/6        -5.7 %     -5.7 %      5.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/8        -6.0 %     -6.0 %      6.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/1         9.0 %      9.0 %     -7.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/2       -13.6 %    -13.6 %     13.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/3         0.4 %      0.4 %      0.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/4        -6.2 %     -6.2 %      6.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/6        -6.2 %     -6.2 %      6.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/8        -6.8 %     -6.8 %      6.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/1       14.8 %     14.8 %    -12.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/2       -7.0 %     -7.0 %      7.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/3        0.0 %      0.0 %     -1.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/4       -5.6 %     -5.6 %      6.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/6       -5.7 %     -5.7 %      6.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/8       -6.2 %     -6.2 %      6.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/1       19.9 %     19.9 %    -15.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/2       -6.5 %     -6.5 %      6.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/3        3.8 %      3.8 %     -3.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/4       -6.5 %     -6.5 %      7.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/6       -6.2 %     -6.2 %      6.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/8       -6.4 %     -6.4 %      6.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/1        12.3 %     12.3 %    -10.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/2         2.1 %      2.1 %     -1.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/3         6.9 %      6.9 %     -6.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/4         1.0 %      1.0 %     -0.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/6        -3.0 %     -3.0 %      3.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/8        -0.6 %     -0.6 %      0.3 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/1         7.3 %      7.3 %     -6.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/2         0.0 %      0.0 %     -0.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/3         3.0 %      3.0 %     -3.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/4        -2.5 %     -2.5 %      2.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/6        -1.7 %     -1.7 %      1.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/8        -2.4 %     -2.4 %      2.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/1        16.1 %     16.1 %    -13.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/2       -10.2 %    -10.2 %     10.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/3        -0.3 %     -0.3 %      0.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/4        -1.4 %     -1.4 %      1.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/6        -3.4 %     -3.4 %      4.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/8        -2.1 %     -2.1 %      2.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/1         9.7 %      9.7 %     -8.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/2        -1.3 %     -1.3 %      0.9 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/3         4.9 %      4.9 %      0.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/4        -2.4 %     -2.4 %      2.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/6        -1.3 %     -1.3 %      1.7 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/8        -2.7 %     -2.7 %      4.0 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/1        26.1 %     26.1 %    -23.5 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/2        -2.9 %     -2.9 %      2.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/3         9.7 %      9.7 %     -7.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/4        -5.1 %     -5.1 %      4.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/6        -3.8 %     -3.8 %      2.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/8        -3.9 %     -3.9 %      1.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/1       25.3 %     25.3 %    -21.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/2       -4.4 %     -4.4 %      3.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/3        7.9 %      7.9 %     -8.8 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/4       -4.3 %     -4.3 %      4.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/6       -3.2 %     -3.2 %      3.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/8       -2.2 %     -2.2 %      3.1 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/1       23.4 %     23.4 %    -20.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/2       -4.9 %     -4.8 %      3.4 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/3        8.9 %      8.9 %     -7.6 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/4       -4.0 %     -4.0 %      4.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/6       -4.2 %     -4.2 %      3.2 %
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/8       -4.1 %     -4.1 %      4.7 %

[-- Attachment #12: small_blas_gemm_benchmark-tuned --]
[-- Type: text/plain, Size: 49410 bytes --]

-----------------------------------------------------------------------------------------
Benchmark                                               Time             CPU   Iterations
-----------------------------------------------------------------------------------------
BM_MatrixMatrixMultiplyDynamic/1/1/1                 6.45 ns         6.45 ns     82223351
BM_MatrixMatrixMultiplyDynamic/1/1/2                 7.12 ns         7.12 ns     93312431
BM_MatrixMatrixMultiplyDynamic/1/1/3                 7.56 ns         7.56 ns     90131089
BM_MatrixMatrixMultiplyDynamic/1/1/4                 9.14 ns         9.14 ns     74827210
BM_MatrixMatrixMultiplyDynamic/1/1/8                 13.9 ns         13.9 ns     49289642
BM_MatrixMatrixMultiplyDynamic/1/1/12                17.9 ns         17.9 ns     38344924
BM_MatrixMatrixMultiplyDynamic/1/1/15                21.8 ns         21.8 ns     31736186
BM_MatrixMatrixMultiplyDynamic/1/2/1                 6.69 ns         6.69 ns    101983240
BM_MatrixMatrixMultiplyDynamic/1/2/2                 6.89 ns         6.89 ns     97647728
BM_MatrixMatrixMultiplyDynamic/1/2/3                 8.88 ns         8.88 ns     76250928
BM_MatrixMatrixMultiplyDynamic/1/2/4                 11.4 ns         11.4 ns     59589937
BM_MatrixMatrixMultiplyDynamic/1/2/8                 16.3 ns         16.3 ns     42098988
BM_MatrixMatrixMultiplyDynamic/1/2/12                21.1 ns         21.1 ns     32667825
BM_MatrixMatrixMultiplyDynamic/1/2/15                26.5 ns         26.5 ns     26178656
BM_MatrixMatrixMultiplyDynamic/1/3/1                 6.90 ns         6.90 ns     98483394
BM_MatrixMatrixMultiplyDynamic/1/3/2                 7.20 ns         7.20 ns     94898902
BM_MatrixMatrixMultiplyDynamic/1/3/3                 11.3 ns         11.3 ns     60153723
BM_MatrixMatrixMultiplyDynamic/1/3/4                 13.2 ns         13.2 ns     51539256
BM_MatrixMatrixMultiplyDynamic/1/3/8                 18.9 ns         18.9 ns     36377040
BM_MatrixMatrixMultiplyDynamic/1/3/12                24.6 ns         24.6 ns     28076900
BM_MatrixMatrixMultiplyDynamic/1/3/15                30.7 ns         30.7 ns     22557853
BM_MatrixMatrixMultiplyDynamic/1/4/1                 6.81 ns         6.81 ns     98989035
BM_MatrixMatrixMultiplyDynamic/1/4/2                 7.43 ns         7.43 ns     93247062
BM_MatrixMatrixMultiplyDynamic/1/4/3                 13.0 ns         13.0 ns     52470139
BM_MatrixMatrixMultiplyDynamic/1/4/4                 14.0 ns         14.0 ns     48701458
BM_MatrixMatrixMultiplyDynamic/1/4/8                 21.0 ns         21.0 ns     33304530
BM_MatrixMatrixMultiplyDynamic/1/4/12                28.0 ns         28.0 ns     24691398
BM_MatrixMatrixMultiplyDynamic/1/4/15                35.4 ns         35.4 ns     19587065
BM_MatrixMatrixMultiplyDynamic/1/8/1                 9.57 ns         9.57 ns     70634203
BM_MatrixMatrixMultiplyDynamic/1/8/2                 12.4 ns         12.4 ns     54606015
BM_MatrixMatrixMultiplyDynamic/1/8/3                 16.5 ns         16.5 ns     41449078
BM_MatrixMatrixMultiplyDynamic/1/8/4                 17.4 ns         17.4 ns     39671851
BM_MatrixMatrixMultiplyDynamic/1/8/8                 28.6 ns         28.6 ns     24035643
BM_MatrixMatrixMultiplyDynamic/1/8/12                39.3 ns         39.3 ns     17713900
BM_MatrixMatrixMultiplyDynamic/1/8/15                49.1 ns         49.1 ns     14157729
BM_MatrixMatrixMultiplyDynamic/1/12/1                12.3 ns         12.3 ns     55013937
BM_MatrixMatrixMultiplyDynamic/1/12/2                14.5 ns         14.5 ns     47666179
BM_MatrixMatrixMultiplyDynamic/1/12/3                21.1 ns         21.1 ns     32517622
BM_MatrixMatrixMultiplyDynamic/1/12/4                20.5 ns         20.5 ns     33669028
BM_MatrixMatrixMultiplyDynamic/1/12/8                35.3 ns         35.3 ns     19688778
BM_MatrixMatrixMultiplyDynamic/1/12/12               49.5 ns         49.5 ns     14076036
BM_MatrixMatrixMultiplyDynamic/1/12/15               66.7 ns         66.7 ns     10271371
BM_MatrixMatrixMultiplyDynamic/1/15/1                13.8 ns         13.8 ns     49207089
BM_MatrixMatrixMultiplyDynamic/1/15/2                16.0 ns         16.0 ns     44189152
BM_MatrixMatrixMultiplyDynamic/1/15/3                25.2 ns         25.2 ns     27383135
BM_MatrixMatrixMultiplyDynamic/1/15/4                24.0 ns         24.0 ns     28839690
BM_MatrixMatrixMultiplyDynamic/1/15/8                42.1 ns         42.1 ns     16494047
BM_MatrixMatrixMultiplyDynamic/1/15/12               59.9 ns         59.9 ns     11565491
BM_MatrixMatrixMultiplyDynamic/1/15/15               82.3 ns         82.3 ns      8281880
BM_MatrixMatrixMultiplyDynamic/2/1/1                 7.22 ns         7.22 ns     92044649
BM_MatrixMatrixMultiplyDynamic/2/1/2                 7.22 ns         7.22 ns     95984971
BM_MatrixMatrixMultiplyDynamic/2/1/3                 11.5 ns         11.5 ns     59313385
BM_MatrixMatrixMultiplyDynamic/2/1/4                 13.4 ns         13.4 ns     50946816
BM_MatrixMatrixMultiplyDynamic/2/1/8                 20.8 ns         20.8 ns     33141112
BM_MatrixMatrixMultiplyDynamic/2/1/12                27.1 ns         27.1 ns     25569836
BM_MatrixMatrixMultiplyDynamic/2/1/15                32.8 ns         32.8 ns     21111491
BM_MatrixMatrixMultiplyDynamic/2/2/1                 7.18 ns         7.18 ns     95705826
BM_MatrixMatrixMultiplyDynamic/2/2/2                 9.00 ns         9.00 ns     75577192
BM_MatrixMatrixMultiplyDynamic/2/2/3                 13.4 ns         13.4 ns     51079487
BM_MatrixMatrixMultiplyDynamic/2/2/4                 15.0 ns         15.0 ns     45568571
BM_MatrixMatrixMultiplyDynamic/2/2/8                 23.1 ns         23.1 ns     29825714
BM_MatrixMatrixMultiplyDynamic/2/2/12                31.2 ns         31.2 ns     22186417
BM_MatrixMatrixMultiplyDynamic/2/2/15                39.8 ns         39.8 ns     17490888
BM_MatrixMatrixMultiplyDynamic/2/3/1                 8.00 ns         8.00 ns     84020572
BM_MatrixMatrixMultiplyDynamic/2/3/2                 12.0 ns         12.0 ns     56994359
BM_MatrixMatrixMultiplyDynamic/2/3/3                 15.2 ns         15.2 ns     44822673
BM_MatrixMatrixMultiplyDynamic/2/3/4                 17.4 ns         17.4 ns     39230012
BM_MatrixMatrixMultiplyDynamic/2/3/8                 26.9 ns         26.9 ns     25715578
BM_MatrixMatrixMultiplyDynamic/2/3/12                36.9 ns         36.9 ns     18835922
BM_MatrixMatrixMultiplyDynamic/2/3/15                46.6 ns         46.6 ns     14924909
BM_MatrixMatrixMultiplyDynamic/2/4/1                 9.33 ns         9.33 ns     71851370
BM_MatrixMatrixMultiplyDynamic/2/4/2                 12.9 ns         12.9 ns     53066790
BM_MatrixMatrixMultiplyDynamic/2/4/3                 16.8 ns         16.8 ns     40731856
BM_MatrixMatrixMultiplyDynamic/2/4/4                 18.7 ns         18.7 ns     36714024
BM_MatrixMatrixMultiplyDynamic/2/4/8                 30.5 ns         30.5 ns     22666127
BM_MatrixMatrixMultiplyDynamic/2/4/12                43.1 ns         43.1 ns     16153163
BM_MatrixMatrixMultiplyDynamic/2/4/15                54.6 ns         54.6 ns     12729557
BM_MatrixMatrixMultiplyDynamic/2/8/1                 14.2 ns         14.2 ns     47983135
BM_MatrixMatrixMultiplyDynamic/2/8/2                 16.9 ns         16.9 ns     41660384
BM_MatrixMatrixMultiplyDynamic/2/8/3                 27.2 ns         27.1 ns     25614640
BM_MatrixMatrixMultiplyDynamic/2/8/4                 24.9 ns         24.9 ns     27586376
BM_MatrixMatrixMultiplyDynamic/2/8/8                 43.7 ns         43.7 ns     15934340
BM_MatrixMatrixMultiplyDynamic/2/8/12                62.6 ns         62.6 ns     11133054
BM_MatrixMatrixMultiplyDynamic/2/8/15                83.4 ns         83.4 ns      8249225
BM_MatrixMatrixMultiplyDynamic/2/12/1                17.8 ns         17.8 ns     38386056
BM_MatrixMatrixMultiplyDynamic/2/12/2                21.7 ns         21.7 ns     32243674
BM_MatrixMatrixMultiplyDynamic/2/12/3                37.7 ns         37.7 ns     18407533
BM_MatrixMatrixMultiplyDynamic/2/12/4                31.4 ns         31.4 ns     22071770
BM_MatrixMatrixMultiplyDynamic/2/12/8                56.8 ns         56.8 ns     12238771
BM_MatrixMatrixMultiplyDynamic/2/12/12               81.1 ns         81.1 ns      8534412
BM_MatrixMatrixMultiplyDynamic/2/12/15                112 ns          112 ns      6078611
BM_MatrixMatrixMultiplyDynamic/2/15/1                21.5 ns         21.5 ns     31990273
BM_MatrixMatrixMultiplyDynamic/2/15/2                26.7 ns         26.7 ns     26336839
BM_MatrixMatrixMultiplyDynamic/2/15/3                46.8 ns         46.8 ns     14862338
BM_MatrixMatrixMultiplyDynamic/2/15/4                37.7 ns         37.7 ns     18540709
BM_MatrixMatrixMultiplyDynamic/2/15/8                69.5 ns         69.5 ns     10058726
BM_MatrixMatrixMultiplyDynamic/2/15/12                103 ns          103 ns      6854752
BM_MatrixMatrixMultiplyDynamic/2/15/15                139 ns          139 ns      4841183
BM_MatrixMatrixMultiplyDynamic/3/1/1                 7.20 ns         7.20 ns     93672891
BM_MatrixMatrixMultiplyDynamic/3/1/2                 10.5 ns         10.5 ns     64389821
BM_MatrixMatrixMultiplyDynamic/3/1/3                 13.4 ns         13.4 ns     51219093
BM_MatrixMatrixMultiplyDynamic/3/1/4                 16.0 ns         16.0 ns     42799646
BM_MatrixMatrixMultiplyDynamic/3/1/8                 25.3 ns         25.3 ns     27355717
BM_MatrixMatrixMultiplyDynamic/3/1/12                34.3 ns         34.3 ns     20204911
BM_MatrixMatrixMultiplyDynamic/3/1/15                42.7 ns         42.7 ns     16301255
BM_MatrixMatrixMultiplyDynamic/3/2/1                 8.66 ns         8.66 ns     78395630
BM_MatrixMatrixMultiplyDynamic/3/2/2                 12.4 ns         12.4 ns     54994369
BM_MatrixMatrixMultiplyDynamic/3/2/3                 16.1 ns         16.1 ns     42509724
BM_MatrixMatrixMultiplyDynamic/3/2/4                 18.1 ns         18.1 ns     37907353
BM_MatrixMatrixMultiplyDynamic/3/2/8                 29.5 ns         29.5 ns     23521531
BM_MatrixMatrixMultiplyDynamic/3/2/12                41.3 ns         41.3 ns     16810776
BM_MatrixMatrixMultiplyDynamic/3/2/15                52.6 ns         52.6 ns     13094800
BM_MatrixMatrixMultiplyDynamic/3/3/1                 11.3 ns         11.3 ns     59670224
BM_MatrixMatrixMultiplyDynamic/3/3/2                 13.8 ns         13.8 ns     49850228
BM_MatrixMatrixMultiplyDynamic/3/3/3                 19.2 ns         19.2 ns     35909837
BM_MatrixMatrixMultiplyDynamic/3/3/4                 21.1 ns         21.1 ns     32564013
BM_MatrixMatrixMultiplyDynamic/3/3/8                 34.9 ns         34.9 ns     19953888
BM_MatrixMatrixMultiplyDynamic/3/3/12                49.4 ns         49.4 ns     14085250
BM_MatrixMatrixMultiplyDynamic/3/3/15                64.0 ns         64.0 ns     10828021
BM_MatrixMatrixMultiplyDynamic/3/4/1                 12.8 ns         12.8 ns     52615453
BM_MatrixMatrixMultiplyDynamic/3/4/2                 15.4 ns         15.4 ns     44665894
BM_MatrixMatrixMultiplyDynamic/3/4/3                 22.1 ns         22.1 ns     31248613
BM_MatrixMatrixMultiplyDynamic/3/4/4                 24.0 ns         24.0 ns     28751733
BM_MatrixMatrixMultiplyDynamic/3/4/8                 40.5 ns         40.5 ns     17161869
BM_MatrixMatrixMultiplyDynamic/3/4/12                57.9 ns         57.9 ns     11945676
BM_MatrixMatrixMultiplyDynamic/3/4/15                75.7 ns         75.7 ns      9096102
BM_MatrixMatrixMultiplyDynamic/3/8/1                 17.0 ns         17.0 ns     40042645
BM_MatrixMatrixMultiplyDynamic/3/8/2                 21.2 ns         21.2 ns     32645572
BM_MatrixMatrixMultiplyDynamic/3/8/3                 36.3 ns         36.3 ns     19063299
BM_MatrixMatrixMultiplyDynamic/3/8/4                 33.3 ns         33.3 ns     20826240
BM_MatrixMatrixMultiplyDynamic/3/8/8                 59.7 ns         59.7 ns     11686443
BM_MatrixMatrixMultiplyDynamic/3/8/12                85.9 ns         85.9 ns      7986812
BM_MatrixMatrixMultiplyDynamic/3/8/15                 115 ns          115 ns      5975312
BM_MatrixMatrixMultiplyDynamic/3/12/1                22.1 ns         22.1 ns     31006601
BM_MatrixMatrixMultiplyDynamic/3/12/2                31.9 ns         31.9 ns     21979519
BM_MatrixMatrixMultiplyDynamic/3/12/3                51.8 ns         51.8 ns     13302077
BM_MatrixMatrixMultiplyDynamic/3/12/4                42.1 ns         42.1 ns     16549113
BM_MatrixMatrixMultiplyDynamic/3/12/8                77.9 ns         77.9 ns      8868931
BM_MatrixMatrixMultiplyDynamic/3/12/12                113 ns          113 ns      6083736
BM_MatrixMatrixMultiplyDynamic/3/12/15                156 ns          156 ns      4384979
BM_MatrixMatrixMultiplyDynamic/3/15/1                34.5 ns         34.5 ns     20144552
BM_MatrixMatrixMultiplyDynamic/3/15/2                39.6 ns         39.6 ns     17821983
BM_MatrixMatrixMultiplyDynamic/3/15/3                65.7 ns         65.7 ns     10421513
BM_MatrixMatrixMultiplyDynamic/3/15/4                51.3 ns         51.3 ns     13700802
BM_MatrixMatrixMultiplyDynamic/3/15/8                96.1 ns         96.1 ns      7138779
BM_MatrixMatrixMultiplyDynamic/3/15/12                141 ns          141 ns      4863465
BM_MatrixMatrixMultiplyDynamic/3/15/15                199 ns          199 ns      3420659
BM_MatrixMatrixMultiplyDynamic/4/1/1                 8.18 ns         8.18 ns     82940281
BM_MatrixMatrixMultiplyDynamic/4/1/2                 12.1 ns         12.1 ns     56258082
BM_MatrixMatrixMultiplyDynamic/4/1/3                 15.5 ns         15.5 ns     44327548
BM_MatrixMatrixMultiplyDynamic/4/1/4                 18.3 ns         18.3 ns     37624360
BM_MatrixMatrixMultiplyDynamic/4/1/8                 29.7 ns         29.7 ns     23309470
BM_MatrixMatrixMultiplyDynamic/4/1/12                41.6 ns         41.6 ns     16730841
BM_MatrixMatrixMultiplyDynamic/4/1/15                53.3 ns         53.3 ns     12907673
BM_MatrixMatrixMultiplyDynamic/4/2/1                 12.1 ns         12.1 ns     55994380
BM_MatrixMatrixMultiplyDynamic/4/2/2                 14.1 ns         14.1 ns     48548044
BM_MatrixMatrixMultiplyDynamic/4/2/3                 19.8 ns         19.8 ns     34989651
BM_MatrixMatrixMultiplyDynamic/4/2/4                 21.5 ns         21.5 ns     32325291
BM_MatrixMatrixMultiplyDynamic/4/2/8                 35.7 ns         35.7 ns     19464709
BM_MatrixMatrixMultiplyDynamic/4/2/12                51.3 ns         51.3 ns     13612821
BM_MatrixMatrixMultiplyDynamic/4/2/15                66.9 ns         66.9 ns     10307144
BM_MatrixMatrixMultiplyDynamic/4/3/1                 13.0 ns         13.0 ns     52146501
BM_MatrixMatrixMultiplyDynamic/4/3/2                 15.8 ns         15.8 ns     43302079
BM_MatrixMatrixMultiplyDynamic/4/3/3                 23.4 ns         23.4 ns     29470661
BM_MatrixMatrixMultiplyDynamic/4/3/4                 25.0 ns         25.0 ns     27637505
BM_MatrixMatrixMultiplyDynamic/4/3/8                 43.1 ns         43.1 ns     16162082
BM_MatrixMatrixMultiplyDynamic/4/3/12                67.4 ns         67.4 ns     10310399
BM_MatrixMatrixMultiplyDynamic/4/3/15                88.4 ns         88.4 ns      7952017
BM_MatrixMatrixMultiplyDynamic/4/4/1                 14.4 ns         14.4 ns     47004594
BM_MatrixMatrixMultiplyDynamic/4/4/2                 17.8 ns         17.8 ns     38587095
BM_MatrixMatrixMultiplyDynamic/4/4/3                 27.7 ns         27.7 ns     24998834
BM_MatrixMatrixMultiplyDynamic/4/4/4                 29.1 ns         29.1 ns     23786532
BM_MatrixMatrixMultiplyDynamic/4/4/8                 50.7 ns         50.7 ns     13787723
BM_MatrixMatrixMultiplyDynamic/4/4/12                73.0 ns         73.0 ns      9480354
BM_MatrixMatrixMultiplyDynamic/4/4/15                95.4 ns         95.4 ns      7188015
BM_MatrixMatrixMultiplyDynamic/4/8/1                 20.6 ns         20.6 ns     33208445
BM_MatrixMatrixMultiplyDynamic/4/8/2                 28.4 ns         28.4 ns     24437731
BM_MatrixMatrixMultiplyDynamic/4/8/3                 45.3 ns         45.3 ns     15352393
BM_MatrixMatrixMultiplyDynamic/4/8/4                 41.1 ns         41.1 ns     16914879
BM_MatrixMatrixMultiplyDynamic/4/8/8                 75.2 ns         75.2 ns      9220643
BM_MatrixMatrixMultiplyDynamic/4/8/12                 110 ns          110 ns      6284794
BM_MatrixMatrixMultiplyDynamic/4/8/15                 147 ns          147 ns      4674935
BM_MatrixMatrixMultiplyDynamic/4/12/1                34.0 ns         34.0 ns     20559912
BM_MatrixMatrixMultiplyDynamic/4/12/2                47.5 ns         47.5 ns     14984055
BM_MatrixMatrixMultiplyDynamic/4/12/3                70.6 ns         70.6 ns      9683405
BM_MatrixMatrixMultiplyDynamic/4/12/4                53.3 ns         53.3 ns     13095001
BM_MatrixMatrixMultiplyDynamic/4/12/8                99.1 ns         99.1 ns      6928330
BM_MatrixMatrixMultiplyDynamic/4/12/12                144 ns          144 ns      4777467
BM_MatrixMatrixMultiplyDynamic/4/12/15                209 ns          209 ns      3294391
BM_MatrixMatrixMultiplyDynamic/4/15/1                46.2 ns         46.2 ns     15035935
BM_MatrixMatrixMultiplyDynamic/4/15/2                56.9 ns         56.9 ns     12263866
BM_MatrixMatrixMultiplyDynamic/4/15/3                95.1 ns         95.1 ns      7066702
BM_MatrixMatrixMultiplyDynamic/4/15/4                65.3 ns         65.3 ns     10613159
BM_MatrixMatrixMultiplyDynamic/4/15/8                 124 ns          124 ns      5506525
BM_MatrixMatrixMultiplyDynamic/4/15/12                186 ns          186 ns      3687494
BM_MatrixMatrixMultiplyDynamic/4/15/15                279 ns          279 ns      2468174
BM_MatrixMatrixMultiplyDynamic/6/1/1                 11.7 ns         11.7 ns     56611908
BM_MatrixMatrixMultiplyDynamic/6/1/2                 14.4 ns         14.4 ns     46822824
BM_MatrixMatrixMultiplyDynamic/6/1/3                 20.3 ns         20.3 ns     33939929
BM_MatrixMatrixMultiplyDynamic/6/1/4                 23.0 ns         23.0 ns     29994780
BM_MatrixMatrixMultiplyDynamic/6/1/8                 39.7 ns         39.7 ns     17541782
BM_MatrixMatrixMultiplyDynamic/6/1/12                57.4 ns         57.4 ns     11994706
BM_MatrixMatrixMultiplyDynamic/6/1/15                86.3 ns         86.3 ns      7899054
BM_MatrixMatrixMultiplyDynamic/6/2/1                 13.6 ns         13.6 ns     50058821
BM_MatrixMatrixMultiplyDynamic/6/2/2                 17.1 ns         17.1 ns     40098489
BM_MatrixMatrixMultiplyDynamic/6/2/3                 25.8 ns         25.8 ns     26838989
BM_MatrixMatrixMultiplyDynamic/6/2/4                 27.9 ns         27.9 ns     24780674
BM_MatrixMatrixMultiplyDynamic/6/2/8                 48.8 ns         48.8 ns     14298927
BM_MatrixMatrixMultiplyDynamic/6/2/12                78.3 ns         78.3 ns      8721980
BM_MatrixMatrixMultiplyDynamic/6/2/15                 103 ns          103 ns      6558187
BM_MatrixMatrixMultiplyDynamic/6/3/1                 15.4 ns         15.4 ns     44652868
BM_MatrixMatrixMultiplyDynamic/6/3/2                 20.0 ns         20.0 ns     34731508
BM_MatrixMatrixMultiplyDynamic/6/3/3                 31.7 ns         31.7 ns     21893731
BM_MatrixMatrixMultiplyDynamic/6/3/4                 33.1 ns         33.1 ns     20944511
BM_MatrixMatrixMultiplyDynamic/6/3/8                 60.1 ns         60.1 ns     11488557
BM_MatrixMatrixMultiplyDynamic/6/3/12                92.6 ns         92.6 ns      7313115
BM_MatrixMatrixMultiplyDynamic/6/3/15                 123 ns          123 ns      5580422
BM_MatrixMatrixMultiplyDynamic/6/4/1                 17.5 ns         17.5 ns     38231104
BM_MatrixMatrixMultiplyDynamic/6/4/2                 23.1 ns         23.1 ns     29999766
BM_MatrixMatrixMultiplyDynamic/6/4/3                 38.0 ns         38.0 ns     18187053
BM_MatrixMatrixMultiplyDynamic/6/4/4                 38.8 ns         38.8 ns     17910982
BM_MatrixMatrixMultiplyDynamic/6/4/8                 71.3 ns         71.3 ns      9679133
BM_MatrixMatrixMultiplyDynamic/6/4/12                 103 ns          103 ns      6617488
BM_MatrixMatrixMultiplyDynamic/6/4/15                 136 ns          136 ns      5026823
BM_MatrixMatrixMultiplyDynamic/6/8/1                 37.7 ns         37.7 ns     18512154
BM_MatrixMatrixMultiplyDynamic/6/8/2                 46.5 ns         46.5 ns     15112069
BM_MatrixMatrixMultiplyDynamic/6/8/3                 78.0 ns         78.0 ns      8552896
BM_MatrixMatrixMultiplyDynamic/6/8/4                 57.6 ns         57.6 ns     11791472
BM_MatrixMatrixMultiplyDynamic/6/8/8                  107 ns          107 ns      5861971
BM_MatrixMatrixMultiplyDynamic/6/8/12                 156 ns          156 ns      4444903
BM_MatrixMatrixMultiplyDynamic/6/8/15                 226 ns          226 ns      3060036
BM_MatrixMatrixMultiplyDynamic/6/12/1                47.2 ns         47.2 ns     13859211
BM_MatrixMatrixMultiplyDynamic/6/12/2                61.9 ns         61.9 ns     11256171
BM_MatrixMatrixMultiplyDynamic/6/12/3                 101 ns          101 ns      6627850
BM_MatrixMatrixMultiplyDynamic/6/12/4                75.0 ns         75.0 ns      9181133
BM_MatrixMatrixMultiplyDynamic/6/12/8                 142 ns          142 ns      4813444
BM_MatrixMatrixMultiplyDynamic/6/12/12                216 ns          216 ns      3171975
BM_MatrixMatrixMultiplyDynamic/6/12/15                310 ns          310 ns      2229126
BM_MatrixMatrixMultiplyDynamic/6/15/1                63.0 ns         63.0 ns     10499469
BM_MatrixMatrixMultiplyDynamic/6/15/2                74.2 ns         74.2 ns      9432996
BM_MatrixMatrixMultiplyDynamic/6/15/3                 129 ns          129 ns      5266838
BM_MatrixMatrixMultiplyDynamic/6/15/4                94.3 ns         94.3 ns      7311880
BM_MatrixMatrixMultiplyDynamic/6/15/8                 180 ns          180 ns      3806890
BM_MatrixMatrixMultiplyDynamic/6/15/12                272 ns          272 ns      2546953
BM_MatrixMatrixMultiplyDynamic/6/15/15                411 ns          411 ns      1689732
BM_MatrixMatrixMultiplyDynamic/8/1/1                 12.9 ns         12.9 ns     51830025
BM_MatrixMatrixMultiplyDynamic/8/1/2                 16.5 ns         16.5 ns     41665189
BM_MatrixMatrixMultiplyDynamic/8/1/3                 24.6 ns         24.6 ns     28325592
BM_MatrixMatrixMultiplyDynamic/8/1/4                 28.1 ns         28.1 ns     24650096
BM_MatrixMatrixMultiplyDynamic/8/1/8                 50.4 ns         50.4 ns     13817859
BM_MatrixMatrixMultiplyDynamic/8/1/12                82.2 ns         82.2 ns      8201453
BM_MatrixMatrixMultiplyDynamic/8/1/15                 104 ns          104 ns      6522454
BM_MatrixMatrixMultiplyDynamic/8/2/1                 15.7 ns         15.7 ns     43640752
BM_MatrixMatrixMultiplyDynamic/8/2/2                 20.5 ns         20.5 ns     33788580
BM_MatrixMatrixMultiplyDynamic/8/2/3                 32.3 ns         32.3 ns     21521559
BM_MatrixMatrixMultiplyDynamic/8/2/4                 34.6 ns         34.6 ns     20052916
BM_MatrixMatrixMultiplyDynamic/8/2/8                 62.1 ns         62.1 ns     11172158
BM_MatrixMatrixMultiplyDynamic/8/2/12                99.6 ns         99.6 ns      6864508
BM_MatrixMatrixMultiplyDynamic/8/2/15                 128 ns          128 ns      5281273
BM_MatrixMatrixMultiplyDynamic/8/3/1                 20.3 ns         20.3 ns     33656313
BM_MatrixMatrixMultiplyDynamic/8/3/2                 24.5 ns         24.5 ns     28351536
BM_MatrixMatrixMultiplyDynamic/8/3/3                 40.2 ns         40.2 ns     17333360
BM_MatrixMatrixMultiplyDynamic/8/3/4                 41.5 ns         41.5 ns     16781376
BM_MatrixMatrixMultiplyDynamic/8/3/8                 82.9 ns         82.9 ns      8268678
BM_MatrixMatrixMultiplyDynamic/8/3/12                 117 ns          117 ns      5898498
BM_MatrixMatrixMultiplyDynamic/8/3/15                 153 ns          153 ns      4469791
BM_MatrixMatrixMultiplyDynamic/8/4/1                 23.5 ns         23.5 ns     29028221
BM_MatrixMatrixMultiplyDynamic/8/4/2                 29.4 ns         29.4 ns     22518396
BM_MatrixMatrixMultiplyDynamic/8/4/3                 49.2 ns         49.2 ns     13285640
BM_MatrixMatrixMultiplyDynamic/8/4/4                 49.6 ns         49.6 ns     13705255
BM_MatrixMatrixMultiplyDynamic/8/4/8                 91.3 ns         91.3 ns      7544354
BM_MatrixMatrixMultiplyDynamic/8/4/12                 133 ns          133 ns      5123374
BM_MatrixMatrixMultiplyDynamic/8/4/15                 177 ns          177 ns      3887869
BM_MatrixMatrixMultiplyDynamic/8/8/1                 42.0 ns         42.0 ns     16557438
BM_MatrixMatrixMultiplyDynamic/8/8/2                 54.7 ns         54.7 ns     12589135
BM_MatrixMatrixMultiplyDynamic/8/8/3                 89.8 ns         89.8 ns      7495177
BM_MatrixMatrixMultiplyDynamic/8/8/4                 73.4 ns         73.4 ns      9423179
BM_MatrixMatrixMultiplyDynamic/8/8/8                  137 ns          137 ns      5007100
BM_MatrixMatrixMultiplyDynamic/8/8/12                 211 ns          211 ns      3272333
BM_MatrixMatrixMultiplyDynamic/8/8/15                 298 ns          298 ns      2340973
BM_MatrixMatrixMultiplyDynamic/8/12/1                63.2 ns         63.2 ns     10660455
BM_MatrixMatrixMultiplyDynamic/8/12/2                78.1 ns         78.1 ns      9004579
BM_MatrixMatrixMultiplyDynamic/8/12/3                 131 ns          131 ns      5181790
BM_MatrixMatrixMultiplyDynamic/8/12/4                97.3 ns         97.3 ns      7180566
BM_MatrixMatrixMultiplyDynamic/8/12/8                 185 ns          185 ns      3710101
BM_MatrixMatrixMultiplyDynamic/8/12/12                280 ns          280 ns      2480219
BM_MatrixMatrixMultiplyDynamic/8/12/15                411 ns          411 ns      1695882
BM_MatrixMatrixMultiplyDynamic/8/15/1                76.6 ns         76.6 ns      8791057
BM_MatrixMatrixMultiplyDynamic/8/15/2                95.4 ns         95.4 ns      7319786
BM_MatrixMatrixMultiplyDynamic/8/15/3                 162 ns          162 ns      4196734
BM_MatrixMatrixMultiplyDynamic/8/15/4                 123 ns          123 ns      5619524
BM_MatrixMatrixMultiplyDynamic/8/15/8                 246 ns          246 ns      2819942
BM_MatrixMatrixMultiplyDynamic/8/15/12                359 ns          359 ns      1927596
BM_MatrixMatrixMultiplyDynamic/8/15/15                550 ns          550 ns      1170409
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/1        6.95 ns         6.95 ns     97944251
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/2        6.70 ns         6.70 ns    102333634
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/3        7.95 ns         7.95 ns     86769099
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/4        9.81 ns         9.81 ns     69515607
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/6        12.6 ns         12.6 ns     54000002
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/8        16.3 ns         16.3 ns     41907245
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/1        6.86 ns         6.86 ns     99244308
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/2        7.79 ns         7.79 ns     88351789
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/3        11.6 ns         11.6 ns     58986305
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/4        15.7 ns         15.7 ns     43754279
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/6        18.1 ns         18.1 ns     38290884
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/8        23.2 ns         23.2 ns     29547768
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/1        7.56 ns         7.55 ns     90233090
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/2        9.67 ns         9.67 ns     70983046
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/3        14.9 ns         14.9 ns     45783291
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/4        18.7 ns         18.7 ns     36663627
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/6        23.6 ns         23.6 ns     29323312
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/8        29.4 ns         29.4 ns     23604290
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/1        8.80 ns         8.80 ns     78162974
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/2        14.0 ns         14.0 ns     49052535
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/3        17.1 ns         17.1 ns     40475321
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/4        21.2 ns         21.2 ns     32383493
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/6        28.0 ns         28.0 ns     24973084
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/8        33.9 ns         33.9 ns     20145940
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/1        15.0 ns         15.0 ns     45313101
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/2        18.7 ns         18.7 ns     36730534
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/3        27.3 ns         27.3 ns     25428848
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/4        31.3 ns         31.3 ns     22167018
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/6        42.9 ns         42.9 ns     16238796
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/8        54.8 ns         54.8 ns     12699785
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/1       19.1 ns         19.1 ns     35878758
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/2       23.8 ns         23.8 ns     28966216
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/3       36.6 ns         36.6 ns     18972698
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/4       40.7 ns         40.7 ns     17083117
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/6       59.4 ns         59.4 ns     11706597
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/8       85.1 ns         85.1 ns      8065655
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/1       23.2 ns         23.2 ns     30176638
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/2       27.5 ns         27.5 ns     25224609
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/3       42.9 ns         42.9 ns     16199926
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/4       54.7 ns         54.7 ns     12693647
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/6       77.5 ns         77.5 ns      8857769
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/8        109 ns          109 ns      6295589
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/1        6.79 ns         6.79 ns     98403167
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/2        6.76 ns         6.76 ns    101268757
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/3        9.29 ns         9.29 ns     73859167
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/4        11.2 ns         11.2 ns     61899296
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/6        15.1 ns         15.1 ns     45738844
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/8        18.1 ns         18.1 ns     37691000
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/1        7.60 ns         7.60 ns     90547658
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/2        9.20 ns         9.20 ns     74556472
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/3        14.8 ns         14.8 ns     45961294
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/4        18.0 ns         18.0 ns     37340585
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/6        20.9 ns         20.9 ns     32932495
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/8        27.1 ns         27.1 ns     25375364
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/1        9.32 ns         9.32 ns     70337142
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/2        14.1 ns         14.1 ns     48386112
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/3        17.5 ns         17.5 ns     38939403
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/4        21.6 ns         21.6 ns     31758983
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/6        28.4 ns         28.4 ns     24379466
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/8        34.7 ns         34.7 ns     20004027
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/1        12.1 ns         12.1 ns     54877812
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/2        15.6 ns         15.6 ns     43226415
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/3        22.0 ns         22.0 ns     31171968
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/4        24.6 ns         24.6 ns     28112734
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/6        33.2 ns         33.2 ns     20874032
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/8        40.7 ns         40.7 ns     17059661
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/1        17.7 ns         17.7 ns     38730947
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/2        22.6 ns         22.6 ns     30585850
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/3        34.7 ns         34.7 ns     20179027
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/4        37.6 ns         37.6 ns     18320379
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/6        54.0 ns         54.0 ns     12761526
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/8        68.0 ns         68.0 ns      9252632
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/1       24.9 ns         24.9 ns     27723705
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/2       29.5 ns         29.5 ns     23485915
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/3       47.3 ns         47.3 ns     14747081
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/4       56.6 ns         56.6 ns     12161682
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/6       80.6 ns         80.6 ns      7606498
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/8        112 ns          112 ns      5786018
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/1       29.6 ns         29.6 ns     23401238
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/2       34.9 ns         34.9 ns     19898222
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/3       57.6 ns         57.6 ns     11967139
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/4       67.2 ns         67.2 ns      9195940
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/6       96.5 ns         96.5 ns      6386739
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/8        132 ns          132 ns      4908544
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/1        6.80 ns         6.80 ns     95269644
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/2        7.31 ns         7.31 ns     93572305
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/3        10.9 ns         10.9 ns     62796512
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/4        13.6 ns         13.6 ns     50366589
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/6        16.6 ns         16.6 ns     41461387
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/8        20.9 ns         20.9 ns     32988538
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/1        8.89 ns         8.89 ns     77405335
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/2        12.2 ns         12.2 ns     55688748
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/3        16.5 ns         16.5 ns     41588347
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/4        20.0 ns         20.0 ns     34262113
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/6        25.6 ns         25.6 ns     27108632
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/8        31.2 ns         31.2 ns     22234895
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/1        12.2 ns         12.2 ns     55984837
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/2        15.7 ns         15.7 ns     43604121
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/3        21.2 ns         21.1 ns     32637924
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/4        24.3 ns         24.3 ns     28271186
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/6        32.5 ns         32.5 ns     21365093
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/8        39.6 ns         39.6 ns     17586994
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/1        14.9 ns         14.9 ns     45627689
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/2        17.7 ns         17.7 ns     38710353
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/3        26.3 ns         26.3 ns     26199276
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/4        27.8 ns         27.8 ns     24871100
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/6        38.6 ns         38.6 ns     17974729
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/8        47.7 ns         47.7 ns     14579073
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/1        23.6 ns         23.6 ns     29226020
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/2        26.8 ns         26.8 ns     25757629
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/3        43.1 ns         43.1 ns     12478175
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/4        44.3 ns         44.3 ns     15701009
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/6        76.4 ns         76.4 ns      8211420
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/8        95.1 ns         95.1 ns      6744097
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/1       31.1 ns         31.1 ns     22351780
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/2       35.7 ns         35.7 ns     19433735
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/3       64.4 ns         64.4 ns     10571611
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/4       68.5 ns         68.5 ns     10109708
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/6       98.5 ns         98.5 ns      6986006
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/8        132 ns          132 ns      5178418
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/1       41.8 ns         41.8 ns     16544517
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/2       46.5 ns         46.5 ns     14982602
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/3       80.5 ns         80.5 ns      8505082
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/4       80.8 ns         80.8 ns      8513556
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/6        122 ns          122 ns      5618990
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/8        157 ns          157 ns      4377283
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/1        7.18 ns         7.18 ns     90419790
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/2        8.06 ns         8.06 ns     83278080
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/3        13.4 ns         13.4 ns     50331474
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/4        16.5 ns         16.5 ns     41970060
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/6        18.5 ns         18.5 ns     37369338
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/8        23.1 ns         23.1 ns     29629424
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/1        9.95 ns         9.95 ns     68058577
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/2        15.2 ns         15.2 ns     45057584
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/3        18.4 ns         18.4 ns     37399242
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/4        20.6 ns         20.6 ns     33340372
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/6        28.2 ns         28.2 ns     24354441
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/8        33.3 ns         33.3 ns     20833525
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/1        14.5 ns         14.5 ns     46623718
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/2        17.3 ns         17.3 ns     39399512
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/3        25.4 ns         25.4 ns     27178322
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/4        25.9 ns         25.9 ns     26693667
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/6        36.2 ns         36.2 ns     19141932
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/8        43.3 ns         43.3 ns     16056433
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/1        16.3 ns         16.3 ns     41336464
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/2        20.1 ns         20.1 ns     33967004
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/3        30.3 ns         30.3 ns     22906368
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/4        30.7 ns         30.7 ns     22567875
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/6        44.1 ns         44.1 ns     15854939
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/8        54.4 ns         54.4 ns     12419343
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/1        26.8 ns         26.8 ns     25774711
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/2        31.4 ns         31.4 ns     22059894
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/3        51.3 ns         51.3 ns     13547664
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/4        51.8 ns         51.8 ns     13310627
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/6        76.8 ns         76.8 ns      8980359
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/8        95.6 ns         95.6 ns      7187784
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/1       42.4 ns         42.4 ns     16411831
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/2       46.5 ns         46.5 ns     14797483
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/3       81.7 ns         81.7 ns      8281901
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/4       72.7 ns         72.7 ns      9498668
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/6        112 ns          112 ns      6101056
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/8        136 ns          136 ns      5041675
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/1       51.6 ns         51.6 ns     13348281
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/2       55.8 ns         55.8 ns     12576918
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/3       98.8 ns         98.8 ns      6923539
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/4       88.3 ns         88.3 ns      7786781
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/6        137 ns          137 ns      4988411
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/8        167 ns          167 ns      4110647
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/1        8.80 ns         8.80 ns     76679339
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/2        9.63 ns         9.63 ns     71202135
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/3        16.6 ns         16.6 ns     42237489
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/4        18.0 ns         18.0 ns     38245368
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/6        21.9 ns         21.9 ns     31735468
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/8        27.2 ns         27.2 ns     25131332
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/1        14.3 ns         14.3 ns     47876916
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/2        16.8 ns         16.7 ns     40985497
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/3        24.1 ns         24.1 ns     28642726
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/4        24.6 ns         24.6 ns     28083952
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/6        34.5 ns         34.5 ns     20099529
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/8        41.1 ns         41.1 ns     16864583
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/1        16.9 ns         16.9 ns     40396242
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/2        20.3 ns         20.3 ns     34244975
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/3        31.9 ns         31.9 ns     21659919
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/4        31.6 ns         31.6 ns     22044077
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/6        45.2 ns         45.2 ns     15278046
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/8        55.3 ns         55.3 ns     12527544
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/1        19.7 ns         19.7 ns     34655596
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/2        23.6 ns         23.6 ns     29266737
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/3        38.4 ns         38.4 ns     18098726
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/4        38.3 ns         38.3 ns     18132975
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/6        56.4 ns         56.4 ns     12312499
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/8        69.5 ns         69.5 ns      9916055
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/1        43.7 ns         43.7 ns     16021953
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/2        45.8 ns         45.8 ns     14921558
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/3        81.2 ns         81.2 ns      8419855
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/4        66.6 ns         66.6 ns     10430415
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/6         106 ns          106 ns      6399689
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/8         124 ns          124 ns      5516701
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/1       58.3 ns         58.3 ns     11769478
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/2       62.3 ns         62.3 ns     11173948
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/3        111 ns          111 ns      6114993
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/4       95.3 ns         95.3 ns      7249849
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/6        150 ns          150 ns      4592495
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/8        180 ns          180 ns      3835025
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/1       72.4 ns         72.4 ns      9449677
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/2       74.6 ns         74.6 ns      9295873
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/3        135 ns          135 ns      5014872
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/4        115 ns          115 ns      5983284
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/6        181 ns          181 ns      3804195
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/8        220 ns          220 ns      3133954
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/1        10.3 ns         10.3 ns     65516041
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/2        14.3 ns         14.3 ns     48038530
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/3        18.5 ns         18.5 ns     37285962
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/4        19.5 ns         19.5 ns     35320533
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/6        26.2 ns         26.2 ns     26399382
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/8        32.0 ns         32.0 ns     21626608
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/1        16.2 ns         16.2 ns     42362372
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/2        18.6 ns         18.6 ns     37002693
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/3        30.5 ns         30.5 ns     22686089
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/4        27.7 ns         27.7 ns     24985359
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/6        40.2 ns         40.2 ns     17261721
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/8        48.2 ns         48.2 ns     14429753
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/1        20.9 ns         20.9 ns     32765164
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/2        22.9 ns         22.9 ns     30112782
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/3        38.8 ns         38.8 ns     17927630
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/4        36.1 ns         36.1 ns     19217285
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/6        54.1 ns         54.1 ns     12783404
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/8        65.0 ns         65.0 ns     10628227
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/1        27.1 ns         27.1 ns     25489296
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/2        31.1 ns         31.1 ns     22301686
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/3        51.3 ns         51.3 ns     13533863
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/4        44.5 ns         44.5 ns     15619771
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/6        69.6 ns         69.6 ns      9912062
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/8        81.4 ns         81.4 ns      8471918
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/1        57.5 ns         57.5 ns     11663366
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/2        57.0 ns         57.0 ns     12040560
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/3         106 ns          106 ns      6477245
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/4        78.3 ns         78.3 ns      8826020
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/6         127 ns          127 ns      5362050
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/8         147 ns          147 ns      4667107
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/1       81.6 ns         81.6 ns      8207016
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/2       78.2 ns         78.2 ns      8831132
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/3        150 ns          150 ns      4546166
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/4        112 ns          112 ns      6132743
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/6        182 ns          182 ns      3779329
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/8        224 ns          224 ns      3122466
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/1       99.1 ns         99.1 ns      6792122
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/2       93.7 ns         93.7 ns      7312416
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/3        184 ns          184 ns      3752006
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/4        144 ns          144 ns      4767925
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/6        228 ns          228 ns      3007858
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/8        284 ns          284 ns      2449847

[-- Attachment #13: small_blas_gemm_benchmark-untuned --]
[-- Type: text/plain, Size: 49410 bytes --]

-----------------------------------------------------------------------------------------
Benchmark                                               Time             CPU   Iterations
-----------------------------------------------------------------------------------------
BM_MatrixMatrixMultiplyDynamic/1/1/1                 6.52 ns         6.52 ns     77579800
BM_MatrixMatrixMultiplyDynamic/1/1/2                 6.95 ns         6.95 ns     96454301
BM_MatrixMatrixMultiplyDynamic/1/1/3                 7.55 ns         7.55 ns     89537904
BM_MatrixMatrixMultiplyDynamic/1/1/4                 9.14 ns         9.14 ns     73395018
BM_MatrixMatrixMultiplyDynamic/1/1/8                 13.8 ns         13.8 ns     48998543
BM_MatrixMatrixMultiplyDynamic/1/1/12                17.6 ns         17.6 ns     38874919
BM_MatrixMatrixMultiplyDynamic/1/1/15                21.7 ns         21.7 ns     31755617
BM_MatrixMatrixMultiplyDynamic/1/2/1                 6.73 ns         6.73 ns     99309819
BM_MatrixMatrixMultiplyDynamic/1/2/2                 6.79 ns         6.79 ns     98038242
BM_MatrixMatrixMultiplyDynamic/1/2/3                 8.90 ns         8.90 ns     75513643
BM_MatrixMatrixMultiplyDynamic/1/2/4                 11.6 ns         11.6 ns     57928640
BM_MatrixMatrixMultiplyDynamic/1/2/8                 16.2 ns         16.2 ns     42161047
BM_MatrixMatrixMultiplyDynamic/1/2/12                20.9 ns         20.9 ns     33003426
BM_MatrixMatrixMultiplyDynamic/1/2/15                26.2 ns         26.2 ns     26467014
BM_MatrixMatrixMultiplyDynamic/1/3/1                 6.73 ns         6.73 ns     99376293
BM_MatrixMatrixMultiplyDynamic/1/3/2                 7.13 ns         7.13 ns     94155066
BM_MatrixMatrixMultiplyDynamic/1/3/3                 11.4 ns         11.4 ns     59468509
BM_MatrixMatrixMultiplyDynamic/1/3/4                 13.2 ns         13.2 ns     51477793
BM_MatrixMatrixMultiplyDynamic/1/3/8                 18.8 ns         18.8 ns     36758086
BM_MatrixMatrixMultiplyDynamic/1/3/12                24.0 ns         24.0 ns     28841627
BM_MatrixMatrixMultiplyDynamic/1/3/15                30.0 ns         30.0 ns     23069967
BM_MatrixMatrixMultiplyDynamic/1/4/1                 7.08 ns         7.08 ns     94038732
BM_MatrixMatrixMultiplyDynamic/1/4/2                 7.50 ns         7.50 ns     88636348
BM_MatrixMatrixMultiplyDynamic/1/4/3                 12.3 ns         12.3 ns     55035085
BM_MatrixMatrixMultiplyDynamic/1/4/4                 13.9 ns         13.9 ns     48455985
BM_MatrixMatrixMultiplyDynamic/1/4/8                 20.6 ns         20.6 ns     33204140
BM_MatrixMatrixMultiplyDynamic/1/4/12                27.3 ns         27.3 ns     25341521
BM_MatrixMatrixMultiplyDynamic/1/4/15                34.6 ns         34.6 ns     20111794
BM_MatrixMatrixMultiplyDynamic/1/8/1                 8.70 ns         8.70 ns     76389460
BM_MatrixMatrixMultiplyDynamic/1/8/2                 12.5 ns         12.5 ns     54221218
BM_MatrixMatrixMultiplyDynamic/1/8/3                 16.2 ns         16.2 ns     42296491
BM_MatrixMatrixMultiplyDynamic/1/8/4                 17.2 ns         17.2 ns     40170012
BM_MatrixMatrixMultiplyDynamic/1/8/8                 28.0 ns         28.0 ns     24861896
BM_MatrixMatrixMultiplyDynamic/1/8/12                38.1 ns         38.1 ns     18081888
BM_MatrixMatrixMultiplyDynamic/1/8/15                49.6 ns         49.6 ns     13121803
BM_MatrixMatrixMultiplyDynamic/1/12/1                12.1 ns         12.1 ns     56097928
BM_MatrixMatrixMultiplyDynamic/1/12/2                14.1 ns         14.1 ns     48159669
BM_MatrixMatrixMultiplyDynamic/1/12/3                20.1 ns         20.1 ns     34330832
BM_MatrixMatrixMultiplyDynamic/1/12/4                19.8 ns         19.8 ns     34836922
BM_MatrixMatrixMultiplyDynamic/1/12/8                34.6 ns         34.6 ns     20093798
BM_MatrixMatrixMultiplyDynamic/1/12/12               48.2 ns         48.2 ns     13597673
BM_MatrixMatrixMultiplyDynamic/1/12/15               67.8 ns         67.8 ns      9822689
BM_MatrixMatrixMultiplyDynamic/1/15/1                13.3 ns         13.3 ns     50942537
BM_MatrixMatrixMultiplyDynamic/1/15/2                15.5 ns         15.5 ns     43292275
BM_MatrixMatrixMultiplyDynamic/1/15/3                26.2 ns         26.2 ns     26387538
BM_MatrixMatrixMultiplyDynamic/1/15/4                23.7 ns         23.7 ns     28968768
BM_MatrixMatrixMultiplyDynamic/1/15/8                41.6 ns         41.6 ns     16717011
BM_MatrixMatrixMultiplyDynamic/1/15/12               59.8 ns         59.8 ns     11054292
BM_MatrixMatrixMultiplyDynamic/1/15/15               83.3 ns         83.3 ns      7941332
BM_MatrixMatrixMultiplyDynamic/2/1/1                 7.27 ns         7.27 ns     91770453
BM_MatrixMatrixMultiplyDynamic/2/1/2                 7.07 ns         7.07 ns     95240205
BM_MatrixMatrixMultiplyDynamic/2/1/3                 11.6 ns         11.6 ns     58552950
BM_MatrixMatrixMultiplyDynamic/2/1/4                 13.4 ns         13.4 ns     50827882
BM_MatrixMatrixMultiplyDynamic/2/1/8                 20.4 ns         20.4 ns     33730100
BM_MatrixMatrixMultiplyDynamic/2/1/12                26.7 ns         26.7 ns     26059845
BM_MatrixMatrixMultiplyDynamic/2/1/15                32.8 ns         32.8 ns     21216727
BM_MatrixMatrixMultiplyDynamic/2/2/1                 7.09 ns         7.09 ns     94872047
BM_MatrixMatrixMultiplyDynamic/2/2/2                 9.25 ns         9.25 ns     73083745
BM_MatrixMatrixMultiplyDynamic/2/2/3                 13.0 ns         13.0 ns     52366562
BM_MatrixMatrixMultiplyDynamic/2/2/4                 15.0 ns         15.0 ns     45517727
BM_MatrixMatrixMultiplyDynamic/2/2/8                 22.4 ns         22.4 ns     31009008
BM_MatrixMatrixMultiplyDynamic/2/2/12                30.2 ns         30.2 ns     23056368
BM_MatrixMatrixMultiplyDynamic/2/2/15                39.0 ns         39.0 ns     17901293
BM_MatrixMatrixMultiplyDynamic/2/3/1                 7.87 ns         7.87 ns     85554922
BM_MatrixMatrixMultiplyDynamic/2/3/2                 12.0 ns         12.0 ns     56781659
BM_MatrixMatrixMultiplyDynamic/2/3/3                 15.0 ns         15.0 ns     45705924
BM_MatrixMatrixMultiplyDynamic/2/3/4                 17.2 ns         17.2 ns     40012632
BM_MatrixMatrixMultiplyDynamic/2/3/8                 26.4 ns         26.4 ns     26204560
BM_MatrixMatrixMultiplyDynamic/2/3/12                35.8 ns         35.8 ns     19456831
BM_MatrixMatrixMultiplyDynamic/2/3/15                46.9 ns         46.9 ns     13802439
BM_MatrixMatrixMultiplyDynamic/2/4/1                 9.27 ns         9.27 ns     72237368
BM_MatrixMatrixMultiplyDynamic/2/4/2                 12.8 ns         12.8 ns     51851311
BM_MatrixMatrixMultiplyDynamic/2/4/3                 17.0 ns         17.0 ns     40396698
BM_MatrixMatrixMultiplyDynamic/2/4/4                 18.7 ns         18.7 ns     36832593
BM_MatrixMatrixMultiplyDynamic/2/4/8                 30.4 ns         30.4 ns     23028688
BM_MatrixMatrixMultiplyDynamic/2/4/12                42.0 ns         42.0 ns     16576455
BM_MatrixMatrixMultiplyDynamic/2/4/15                55.1 ns         55.1 ns     11937131
BM_MatrixMatrixMultiplyDynamic/2/8/1                 13.7 ns         13.7 ns     49415772
BM_MatrixMatrixMultiplyDynamic/2/8/2                 16.2 ns         16.2 ns     42247281
BM_MatrixMatrixMultiplyDynamic/2/8/3                 27.1 ns         27.1 ns     25596937
BM_MatrixMatrixMultiplyDynamic/2/8/4                 24.7 ns         24.7 ns     27878002
BM_MatrixMatrixMultiplyDynamic/2/8/8                 42.9 ns         42.9 ns     16292842
BM_MatrixMatrixMultiplyDynamic/2/8/12                70.4 ns         70.4 ns     10628409
BM_MatrixMatrixMultiplyDynamic/2/8/15                88.0 ns         88.0 ns      5709431
BM_MatrixMatrixMultiplyDynamic/2/12/1                16.7 ns         16.7 ns     40452283
BM_MatrixMatrixMultiplyDynamic/2/12/2                21.1 ns         21.1 ns     32628888
BM_MatrixMatrixMultiplyDynamic/2/12/3                35.7 ns         35.7 ns     19538060
BM_MatrixMatrixMultiplyDynamic/2/12/4                33.2 ns         33.2 ns     21332505
BM_MatrixMatrixMultiplyDynamic/2/12/8                57.6 ns         57.6 ns     11570197
BM_MatrixMatrixMultiplyDynamic/2/12/12               86.8 ns         86.8 ns      8043090
BM_MatrixMatrixMultiplyDynamic/2/12/15                114 ns          114 ns      5921225
BM_MatrixMatrixMultiplyDynamic/2/15/1                19.0 ns         19.0 ns     36171156
BM_MatrixMatrixMultiplyDynamic/2/15/2                25.7 ns         25.7 ns     26780312
BM_MatrixMatrixMultiplyDynamic/2/15/3                42.8 ns         42.8 ns     16292405
BM_MatrixMatrixMultiplyDynamic/2/15/4                38.4 ns         38.4 ns     18067585
BM_MatrixMatrixMultiplyDynamic/2/15/8                81.1 ns         81.1 ns      8544478
BM_MatrixMatrixMultiplyDynamic/2/15/12                118 ns          118 ns      6520273
BM_MatrixMatrixMultiplyDynamic/2/15/15                142 ns          142 ns      4224830
BM_MatrixMatrixMultiplyDynamic/3/1/1                 7.17 ns         7.17 ns     93627778
BM_MatrixMatrixMultiplyDynamic/3/1/2                 9.51 ns         9.51 ns     69594412
BM_MatrixMatrixMultiplyDynamic/3/1/3                 13.6 ns         13.6 ns     50036777
BM_MatrixMatrixMultiplyDynamic/3/1/4                 15.9 ns         15.9 ns     42842938
BM_MatrixMatrixMultiplyDynamic/3/1/8                 24.5 ns         24.5 ns     28248215
BM_MatrixMatrixMultiplyDynamic/3/1/12                33.1 ns         33.1 ns     20944554
BM_MatrixMatrixMultiplyDynamic/3/1/15                43.0 ns         43.0 ns     16193384
BM_MatrixMatrixMultiplyDynamic/3/2/1                 8.74 ns         8.74 ns     77386826
BM_MatrixMatrixMultiplyDynamic/3/2/2                 12.4 ns         12.4 ns     54784050
BM_MatrixMatrixMultiplyDynamic/3/2/3                 16.3 ns         16.3 ns     41984348
BM_MatrixMatrixMultiplyDynamic/3/2/4                 18.1 ns         18.1 ns     38092381
BM_MatrixMatrixMultiplyDynamic/3/2/8                 28.9 ns         28.9 ns     24053039
BM_MatrixMatrixMultiplyDynamic/3/2/12                40.3 ns         40.3 ns     17299318
BM_MatrixMatrixMultiplyDynamic/3/2/15                53.2 ns         53.2 ns     12142647
BM_MatrixMatrixMultiplyDynamic/3/3/1                 11.7 ns         11.7 ns     57864890
BM_MatrixMatrixMultiplyDynamic/3/3/2                 13.6 ns         13.6 ns     49968280
BM_MatrixMatrixMultiplyDynamic/3/3/3                 20.8 ns         20.8 ns     33550408
BM_MatrixMatrixMultiplyDynamic/3/3/4                 21.1 ns         21.1 ns     32880019
BM_MatrixMatrixMultiplyDynamic/3/3/8                 39.7 ns         39.7 ns     20167368
BM_MatrixMatrixMultiplyDynamic/3/3/12                50.0 ns         50.0 ns     12434931
BM_MatrixMatrixMultiplyDynamic/3/3/15                63.9 ns         63.9 ns     10585258
BM_MatrixMatrixMultiplyDynamic/3/4/1                 12.7 ns         12.7 ns     54042761
BM_MatrixMatrixMultiplyDynamic/3/4/2                 14.8 ns         14.8 ns     45991494
BM_MatrixMatrixMultiplyDynamic/3/4/3                 23.0 ns         23.0 ns     30197823
BM_MatrixMatrixMultiplyDynamic/3/4/4                 23.8 ns         23.8 ns     29201594
BM_MatrixMatrixMultiplyDynamic/3/4/8                 41.9 ns         41.9 ns     17646306
BM_MatrixMatrixMultiplyDynamic/3/4/12                59.2 ns         59.2 ns      8975877
BM_MatrixMatrixMultiplyDynamic/3/4/15                74.7 ns         74.7 ns      9136570
BM_MatrixMatrixMultiplyDynamic/3/8/1                 16.6 ns         16.6 ns     41320173
BM_MatrixMatrixMultiplyDynamic/3/8/2                 22.3 ns         22.3 ns     30992513
BM_MatrixMatrixMultiplyDynamic/3/8/3                 35.3 ns         35.3 ns     19764144
BM_MatrixMatrixMultiplyDynamic/3/8/4                 34.2 ns         34.2 ns     20942510
BM_MatrixMatrixMultiplyDynamic/3/8/8                 60.3 ns         60.3 ns     11271646
BM_MatrixMatrixMultiplyDynamic/3/8/12                88.0 ns         88.0 ns      7757092
BM_MatrixMatrixMultiplyDynamic/3/8/15                 116 ns          116 ns      5865586
BM_MatrixMatrixMultiplyDynamic/3/12/1                24.2 ns         24.2 ns     28688045
BM_MatrixMatrixMultiplyDynamic/3/12/2                35.1 ns         35.1 ns     22512212
BM_MatrixMatrixMultiplyDynamic/3/12/3                53.9 ns         53.9 ns     14171426
BM_MatrixMatrixMultiplyDynamic/3/12/4                44.2 ns         44.2 ns     12343147
BM_MatrixMatrixMultiplyDynamic/3/12/8                90.1 ns         90.1 ns      8231228
BM_MatrixMatrixMultiplyDynamic/3/12/12                119 ns          119 ns      5795295
BM_MatrixMatrixMultiplyDynamic/3/12/15                161 ns          161 ns      4280989
BM_MatrixMatrixMultiplyDynamic/3/15/1                29.7 ns         29.7 ns     23415122
BM_MatrixMatrixMultiplyDynamic/3/15/2                41.5 ns         41.5 ns     19093740
BM_MatrixMatrixMultiplyDynamic/3/15/3                58.4 ns         58.4 ns     11289064
BM_MatrixMatrixMultiplyDynamic/3/15/4                53.4 ns         53.4 ns     12440887
BM_MatrixMatrixMultiplyDynamic/3/15/8                99.8 ns         99.8 ns      6821749
BM_MatrixMatrixMultiplyDynamic/3/15/12                146 ns          146 ns      4708773
BM_MatrixMatrixMultiplyDynamic/3/15/15                201 ns          201 ns      3431935
BM_MatrixMatrixMultiplyDynamic/4/1/1                 8.30 ns         8.30 ns     81548998
BM_MatrixMatrixMultiplyDynamic/4/1/2                 12.4 ns         12.4 ns     54659897
BM_MatrixMatrixMultiplyDynamic/4/1/3                 15.6 ns         15.6 ns     44104946
BM_MatrixMatrixMultiplyDynamic/4/1/4                 18.0 ns         18.0 ns     38381756
BM_MatrixMatrixMultiplyDynamic/4/1/8                 37.6 ns         37.6 ns     24188228
BM_MatrixMatrixMultiplyDynamic/4/1/12                47.5 ns         47.5 ns     17426437
BM_MatrixMatrixMultiplyDynamic/4/1/15                53.1 ns         53.1 ns     12567797
BM_MatrixMatrixMultiplyDynamic/4/2/1                 11.5 ns         11.5 ns     59238093
BM_MatrixMatrixMultiplyDynamic/4/2/2                 14.0 ns         14.0 ns     48248988
BM_MatrixMatrixMultiplyDynamic/4/2/3                 20.3 ns         20.3 ns     34101701
BM_MatrixMatrixMultiplyDynamic/4/2/4                 21.0 ns         21.0 ns     33127051
BM_MatrixMatrixMultiplyDynamic/4/2/8                 44.2 ns         44.2 ns     19828521
BM_MatrixMatrixMultiplyDynamic/4/2/12                69.5 ns         69.5 ns     13364218
BM_MatrixMatrixMultiplyDynamic/4/2/15                67.8 ns         67.8 ns      7859215
BM_MatrixMatrixMultiplyDynamic/4/3/1                 18.1 ns         18.1 ns     31440452
BM_MatrixMatrixMultiplyDynamic/4/3/2                 15.4 ns         15.4 ns     44613256
BM_MatrixMatrixMultiplyDynamic/4/3/3                 23.9 ns         23.9 ns     28917406
BM_MatrixMatrixMultiplyDynamic/4/3/4                 35.3 ns         35.3 ns     27720907
BM_MatrixMatrixMultiplyDynamic/4/3/8                 45.0 ns         45.0 ns     14833460
BM_MatrixMatrixMultiplyDynamic/4/3/12                69.3 ns         69.3 ns      8849115
BM_MatrixMatrixMultiplyDynamic/4/3/15                93.7 ns         93.7 ns      7401437
BM_MatrixMatrixMultiplyDynamic/4/4/1                 14.2 ns         14.2 ns     46045357
BM_MatrixMatrixMultiplyDynamic/4/4/2                 17.4 ns         17.4 ns     39616025
BM_MatrixMatrixMultiplyDynamic/4/4/3                 27.8 ns         27.8 ns     25133955
BM_MatrixMatrixMultiplyDynamic/4/4/4                 28.8 ns         28.8 ns     24027263
BM_MatrixMatrixMultiplyDynamic/4/4/8                 52.6 ns         52.6 ns     13393081
BM_MatrixMatrixMultiplyDynamic/4/4/12                71.9 ns         71.9 ns      9547656
BM_MatrixMatrixMultiplyDynamic/4/4/15                95.3 ns         95.3 ns      7233668
BM_MatrixMatrixMultiplyDynamic/4/8/1                 23.0 ns         23.0 ns     30039355
BM_MatrixMatrixMultiplyDynamic/4/8/2                 33.1 ns         33.1 ns     24365085
BM_MatrixMatrixMultiplyDynamic/4/8/3                 50.2 ns         50.2 ns     16004721
BM_MatrixMatrixMultiplyDynamic/4/8/4                 46.2 ns         46.2 ns     16954724
BM_MatrixMatrixMultiplyDynamic/4/8/8                 87.7 ns         87.7 ns      8778136
BM_MatrixMatrixMultiplyDynamic/4/8/12                 113 ns          113 ns      4555954
BM_MatrixMatrixMultiplyDynamic/4/8/15                 169 ns          169 ns      4601825
BM_MatrixMatrixMultiplyDynamic/4/12/1                34.5 ns         34.5 ns     20306769
BM_MatrixMatrixMultiplyDynamic/4/12/2                44.7 ns         44.7 ns     18396436
BM_MatrixMatrixMultiplyDynamic/4/12/3                72.1 ns         72.1 ns      7626176
BM_MatrixMatrixMultiplyDynamic/4/12/4                54.9 ns         54.9 ns     12099632
BM_MatrixMatrixMultiplyDynamic/4/12/8                 122 ns          122 ns      5770378
BM_MatrixMatrixMultiplyDynamic/4/12/12                182 ns          182 ns      4503140
BM_MatrixMatrixMultiplyDynamic/4/12/15                220 ns          220 ns      3168270
BM_MatrixMatrixMultiplyDynamic/4/15/1                48.2 ns         48.2 ns     13387369
BM_MatrixMatrixMultiplyDynamic/4/15/2                52.2 ns         52.2 ns     13386694
BM_MatrixMatrixMultiplyDynamic/4/15/3                93.9 ns         93.9 ns      5793800
BM_MatrixMatrixMultiplyDynamic/4/15/4                68.4 ns         68.4 ns     10050005
BM_MatrixMatrixMultiplyDynamic/4/15/8                 142 ns          142 ns      5065196
BM_MatrixMatrixMultiplyDynamic/4/15/12                209 ns          209 ns      3429016
BM_MatrixMatrixMultiplyDynamic/4/15/15                287 ns          287 ns      2099382
BM_MatrixMatrixMultiplyDynamic/6/1/1                 11.7 ns         11.7 ns     57789212
BM_MatrixMatrixMultiplyDynamic/6/1/2                 14.3 ns         14.3 ns     48007649
BM_MatrixMatrixMultiplyDynamic/6/1/3                 21.1 ns         21.1 ns     32683435
BM_MatrixMatrixMultiplyDynamic/6/1/4                 22.7 ns         22.7 ns     30451855
BM_MatrixMatrixMultiplyDynamic/6/1/8                 38.7 ns         38.7 ns     18027327
BM_MatrixMatrixMultiplyDynamic/6/1/12                62.4 ns         62.4 ns     11763644
BM_MatrixMatrixMultiplyDynamic/6/1/15                73.6 ns         73.6 ns      9213477
BM_MatrixMatrixMultiplyDynamic/6/2/1                 14.1 ns         14.1 ns     42103836
BM_MatrixMatrixMultiplyDynamic/6/2/2                 17.0 ns         17.0 ns     40609238
BM_MatrixMatrixMultiplyDynamic/6/2/3                 26.8 ns         26.8 ns     26165787
BM_MatrixMatrixMultiplyDynamic/6/2/4                 27.7 ns         27.6 ns     25131925
BM_MatrixMatrixMultiplyDynamic/6/2/8                 52.8 ns         52.8 ns     13754099
BM_MatrixMatrixMultiplyDynamic/6/2/12                83.0 ns         83.0 ns      8304289
BM_MatrixMatrixMultiplyDynamic/6/2/15                 111 ns          111 ns      6340029
BM_MatrixMatrixMultiplyDynamic/6/3/1                 15.3 ns         15.3 ns     45022295
BM_MatrixMatrixMultiplyDynamic/6/3/2                 21.4 ns         21.4 ns     32417297
BM_MatrixMatrixMultiplyDynamic/6/3/3                 31.8 ns         31.8 ns     21985627
BM_MatrixMatrixMultiplyDynamic/6/3/4                 39.4 ns         39.4 ns     20745398
BM_MatrixMatrixMultiplyDynamic/6/3/8                 60.2 ns         60.2 ns      9307703
BM_MatrixMatrixMultiplyDynamic/6/3/12                94.3 ns         94.3 ns      7159177
BM_MatrixMatrixMultiplyDynamic/6/3/15                 123 ns          123 ns      5472598
BM_MatrixMatrixMultiplyDynamic/6/4/1                 19.7 ns         19.7 ns     34920145
BM_MatrixMatrixMultiplyDynamic/6/4/2                 30.5 ns         30.5 ns     27920442
BM_MatrixMatrixMultiplyDynamic/6/4/3                 37.7 ns         37.7 ns     18572765
BM_MatrixMatrixMultiplyDynamic/6/4/4                 38.2 ns         38.2 ns     17015454
BM_MatrixMatrixMultiplyDynamic/6/4/8                 70.0 ns         70.0 ns      9665778
BM_MatrixMatrixMultiplyDynamic/6/4/12                 102 ns          102 ns      6640984
BM_MatrixMatrixMultiplyDynamic/6/4/15                 135 ns          135 ns      5088338
BM_MatrixMatrixMultiplyDynamic/6/8/1                 35.7 ns         35.7 ns     19531452
BM_MatrixMatrixMultiplyDynamic/6/8/2                 53.9 ns         53.9 ns     15207441
BM_MatrixMatrixMultiplyDynamic/6/8/3                 75.8 ns         75.8 ns      8759833
BM_MatrixMatrixMultiplyDynamic/6/8/4                 59.0 ns         59.0 ns     11289757
BM_MatrixMatrixMultiplyDynamic/6/8/8                  111 ns          111 ns      6161367
BM_MatrixMatrixMultiplyDynamic/6/8/12                 163 ns          163 ns      4235139
BM_MatrixMatrixMultiplyDynamic/6/8/15                 239 ns          239 ns      2864068
BM_MatrixMatrixMultiplyDynamic/6/12/1                44.4 ns         44.4 ns     15698744
BM_MatrixMatrixMultiplyDynamic/6/12/2                73.2 ns         73.2 ns     11391829
BM_MatrixMatrixMultiplyDynamic/6/12/3                93.0 ns         93.0 ns      5577095
BM_MatrixMatrixMultiplyDynamic/6/12/4                 100 ns          100 ns      8470820
BM_MatrixMatrixMultiplyDynamic/6/12/8                 166 ns          166 ns      4591061
BM_MatrixMatrixMultiplyDynamic/6/12/12                236 ns          236 ns      2970607
BM_MatrixMatrixMultiplyDynamic/6/12/15                320 ns          320 ns      2180277
BM_MatrixMatrixMultiplyDynamic/6/15/1                54.8 ns         54.8 ns     12398156
BM_MatrixMatrixMultiplyDynamic/6/15/2                98.0 ns         98.0 ns      9668181
BM_MatrixMatrixMultiplyDynamic/6/15/3                 113 ns          113 ns      5972119
BM_MatrixMatrixMultiplyDynamic/6/15/4                 103 ns          103 ns      6719041
BM_MatrixMatrixMultiplyDynamic/6/15/8                 190 ns          190 ns      3656465
BM_MatrixMatrixMultiplyDynamic/6/15/12                283 ns          283 ns      2465259
BM_MatrixMatrixMultiplyDynamic/6/15/15                411 ns          411 ns      1690361
BM_MatrixMatrixMultiplyDynamic/8/1/1                 13.1 ns         13.1 ns     52105613
BM_MatrixMatrixMultiplyDynamic/8/1/2                 16.0 ns         16.0 ns     42980722
BM_MatrixMatrixMultiplyDynamic/8/1/3                 25.5 ns         25.5 ns     27109841
BM_MatrixMatrixMultiplyDynamic/8/1/4                 35.3 ns         35.3 ns     24848065
BM_MatrixMatrixMultiplyDynamic/8/1/8                 53.4 ns         53.4 ns     13765427
BM_MatrixMatrixMultiplyDynamic/8/1/12                81.0 ns         81.0 ns      8295249
BM_MatrixMatrixMultiplyDynamic/8/1/15                 105 ns          105 ns      5672826
BM_MatrixMatrixMultiplyDynamic/8/2/1                 16.6 ns         16.6 ns     41041443
BM_MatrixMatrixMultiplyDynamic/8/2/2                 26.2 ns         26.2 ns     31172046
BM_MatrixMatrixMultiplyDynamic/8/2/3                 32.6 ns         32.6 ns     21251126
BM_MatrixMatrixMultiplyDynamic/8/2/4                 41.5 ns         41.5 ns     20267288
BM_MatrixMatrixMultiplyDynamic/8/2/8                 85.6 ns         85.6 ns     10437381
BM_MatrixMatrixMultiplyDynamic/8/2/12                99.5 ns         99.5 ns      5861390
BM_MatrixMatrixMultiplyDynamic/8/2/15                 138 ns          138 ns      5095575
BM_MatrixMatrixMultiplyDynamic/8/3/1                 20.7 ns         20.7 ns     32892786
BM_MatrixMatrixMultiplyDynamic/8/3/2                 35.2 ns         35.2 ns     26462010
BM_MatrixMatrixMultiplyDynamic/8/3/3                 40.0 ns         40.0 ns     17426490
BM_MatrixMatrixMultiplyDynamic/8/3/4                 52.2 ns         52.2 ns     12603206
BM_MatrixMatrixMultiplyDynamic/8/3/8                 91.1 ns         91.1 ns      7501683
BM_MatrixMatrixMultiplyDynamic/8/3/12                 125 ns          125 ns      5446391
BM_MatrixMatrixMultiplyDynamic/8/3/15                 157 ns          157 ns      4376004
BM_MatrixMatrixMultiplyDynamic/8/4/1                 23.7 ns         23.7 ns     29232595
BM_MatrixMatrixMultiplyDynamic/8/4/2                 30.8 ns         30.8 ns     22680362
BM_MatrixMatrixMultiplyDynamic/8/4/3                 48.4 ns         48.4 ns     13773067
BM_MatrixMatrixMultiplyDynamic/8/4/4                 49.5 ns         49.5 ns     13757774
BM_MatrixMatrixMultiplyDynamic/8/4/8                 94.2 ns         94.2 ns      7145537
BM_MatrixMatrixMultiplyDynamic/8/4/12                 136 ns          136 ns      5059549
BM_MatrixMatrixMultiplyDynamic/8/4/15                 182 ns          182 ns      3758157
BM_MatrixMatrixMultiplyDynamic/8/8/1                 42.0 ns         42.0 ns     16612227
BM_MatrixMatrixMultiplyDynamic/8/8/2                 54.9 ns         54.9 ns     12457943
BM_MatrixMatrixMultiplyDynamic/8/8/3                 90.1 ns         90.1 ns      7537350
BM_MatrixMatrixMultiplyDynamic/8/8/4                 79.5 ns         79.5 ns      8626128
BM_MatrixMatrixMultiplyDynamic/8/8/8                  148 ns          148 ns      4640227
BM_MatrixMatrixMultiplyDynamic/8/8/12                 231 ns          231 ns      3001649
BM_MatrixMatrixMultiplyDynamic/8/8/15                 313 ns          313 ns      2228978
BM_MatrixMatrixMultiplyDynamic/8/12/1                57.7 ns         57.7 ns     11720168
BM_MatrixMatrixMultiplyDynamic/8/12/2                75.3 ns         75.3 ns      9148284
BM_MatrixMatrixMultiplyDynamic/8/12/3                 124 ns          124 ns      5549894
BM_MatrixMatrixMultiplyDynamic/8/12/4                 109 ns          109 ns      6352232
BM_MatrixMatrixMultiplyDynamic/8/12/8                 209 ns          209 ns      3301403
BM_MatrixMatrixMultiplyDynamic/8/12/12                317 ns          316 ns      2200568
BM_MatrixMatrixMultiplyDynamic/8/12/15                438 ns          438 ns      1583464
BM_MatrixMatrixMultiplyDynamic/8/15/1                68.6 ns         68.6 ns      9894031
BM_MatrixMatrixMultiplyDynamic/8/15/2                90.8 ns         90.8 ns      7601643
BM_MatrixMatrixMultiplyDynamic/8/15/3                 150 ns          150 ns      4555631
BM_MatrixMatrixMultiplyDynamic/8/15/4                 141 ns          141 ns      4942516
BM_MatrixMatrixMultiplyDynamic/8/15/8                 272 ns          272 ns      2558484
BM_MatrixMatrixMultiplyDynamic/8/15/12                396 ns          396 ns      1760796
BM_MatrixMatrixMultiplyDynamic/8/15/15                570 ns          570 ns      1146004
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/1        6.84 ns         6.84 ns     98547958
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/2        7.11 ns         7.11 ns     93927153
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/3        7.85 ns         7.85 ns     85225647
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/4        9.79 ns         9.79 ns     67805328
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/6        12.6 ns         12.6 ns     53149343
BM_MatrixTransposeMatrixMultiplyDynamic/1/1/8        16.3 ns         16.3 ns     41942596
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/1        6.86 ns         6.86 ns     96735465
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/2        7.47 ns         7.47 ns     87624654
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/3        11.6 ns         11.6 ns     56843502
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/4        16.0 ns         16.0 ns     43035913
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/6        17.7 ns         17.7 ns     39149010
BM_MatrixTransposeMatrixMultiplyDynamic/1/2/8        23.1 ns         23.1 ns     29791880
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/1        7.40 ns         7.40 ns     93367721
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/2        9.61 ns         9.61 ns     71268437
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/3        14.8 ns         14.8 ns     46121275
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/4        18.9 ns         18.9 ns     36708490
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/6        22.9 ns         22.9 ns     30398218
BM_MatrixTransposeMatrixMultiplyDynamic/1/3/8        29.7 ns         29.7 ns     23673701
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/1        8.49 ns         8.49 ns     72825954
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/2        13.9 ns         13.9 ns     49161483
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/3        16.7 ns         16.7 ns     41655760
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/4        21.4 ns         21.4 ns     32318609
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/6        27.6 ns         27.6 ns     25002035
BM_MatrixTransposeMatrixMultiplyDynamic/1/4/8        34.3 ns         34.3 ns     20233637
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/1        14.2 ns         14.2 ns     48273331
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/2        18.1 ns         18.1 ns     38042543
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/3        26.7 ns         26.7 ns     25949603
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/4        33.4 ns         33.4 ns     20781402
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/6        44.6 ns         44.6 ns     15348767
BM_MatrixTransposeMatrixMultiplyDynamic/1/8/8        57.6 ns         57.6 ns     11994459
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/1       18.7 ns         18.7 ns     36612733
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/2       24.2 ns         24.2 ns     28581682
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/3       36.8 ns         36.8 ns     18892898
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/4       43.8 ns         43.8 ns     15934330
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/6       61.4 ns         61.4 ns     11235622
BM_MatrixTransposeMatrixMultiplyDynamic/1/12/8       85.6 ns         85.6 ns      8069732
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/1       21.4 ns         21.4 ns     32068237
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/2       28.2 ns         28.2 ns     24581807
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/3       43.9 ns         43.9 ns     15873880
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/4       58.0 ns         58.0 ns     11915793
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/6       79.7 ns         79.7 ns      8680514
BM_MatrixTransposeMatrixMultiplyDynamic/1/15/8        117 ns          117 ns      5835651
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/1        7.16 ns         7.16 ns     93599921
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/2        7.12 ns         7.12 ns     98348648
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/3        9.69 ns         9.69 ns     70419071
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/4        11.1 ns         11.1 ns     61262688
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/6        14.8 ns         14.8 ns     46291573
BM_MatrixTransposeMatrixMultiplyDynamic/2/1/8        18.4 ns         18.4 ns     37295805
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/1        7.41 ns         7.41 ns     92848745
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/2        9.32 ns         9.31 ns     74278767
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/3        14.5 ns         14.5 ns     45845185
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/4        18.1 ns         18.1 ns     37878239
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/6        21.6 ns         21.6 ns     32182534
BM_MatrixTransposeMatrixMultiplyDynamic/2/2/8        28.5 ns         28.5 ns     24326440
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/1        9.05 ns         9.05 ns     74719669
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/2        14.2 ns         14.1 ns     48539899
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/3        17.3 ns         17.3 ns     39658801
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/4        21.5 ns         21.5 ns     31757766
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/6        28.6 ns         28.6 ns     24113353
BM_MatrixTransposeMatrixMultiplyDynamic/2/3/8        35.7 ns         35.7 ns     19481663
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/1        12.5 ns         12.5 ns     53813528
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/2        16.1 ns         16.1 ns     42825042
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/3        21.7 ns         21.7 ns     31531208
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/4        25.4 ns         25.4 ns     27164495
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/6        34.5 ns         34.5 ns     20048867
BM_MatrixTransposeMatrixMultiplyDynamic/2/4/8        43.1 ns         43.1 ns     15893218
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/1        17.3 ns         17.3 ns     39576084
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/2        23.2 ns         23.2 ns     29690631
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/3        35.5 ns         35.5 ns     19500269
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/4        41.4 ns         41.4 ns     16806140
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/6        58.6 ns         58.6 ns     11975769
BM_MatrixTransposeMatrixMultiplyDynamic/2/8/8        73.8 ns         73.8 ns      9299104
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/1       23.6 ns         23.6 ns     29298120
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/2       31.6 ns         31.6 ns     21995383
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/3       49.9 ns         49.9 ns     13890407
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/4       61.8 ns         61.8 ns     11138012
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/6       86.6 ns         86.6 ns      7943918
BM_MatrixTransposeMatrixMultiplyDynamic/2/12/8        123 ns          123 ns      5567801
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/1       28.2 ns         28.2 ns     24486586
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/2       37.7 ns         37.7 ns     18454560
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/3       60.1 ns         60.1 ns     11521740
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/4       73.0 ns         73.0 ns      9463341
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/6        103 ns          103 ns      6668050
BM_MatrixTransposeMatrixMultiplyDynamic/2/15/8        146 ns          146 ns      4676853
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/1        6.65 ns         6.65 ns    101750837
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/2        7.37 ns         7.37 ns     94257192
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/3        11.3 ns         11.3 ns     60808983
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/4        13.8 ns         13.8 ns     49512058
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/6        16.3 ns         16.3 ns     42370092
BM_MatrixTransposeMatrixMultiplyDynamic/3/1/8        21.6 ns         21.6 ns     31833205
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/1        8.26 ns         8.26 ns     83109103
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/2        13.5 ns         13.5 ns     50654194
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/3        16.8 ns         16.8 ns     40667572
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/4        20.1 ns         20.1 ns     34159571
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/6        26.5 ns         26.5 ns     26141674
BM_MatrixTransposeMatrixMultiplyDynamic/3/2/8        32.0 ns         32.0 ns     21664876
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/1        12.7 ns         12.7 ns     53602363
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/2        16.0 ns         16.0 ns     42886216
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/3        21.8 ns         21.8 ns     31587759
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/4        24.6 ns         24.6 ns     28089028
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/6        33.9 ns         33.9 ns     20442092
BM_MatrixTransposeMatrixMultiplyDynamic/3/3/8        41.8 ns         41.8 ns     16613382
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/1        14.6 ns         14.6 ns     46722434
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/2        17.6 ns         17.6 ns     39062720
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/3        26.0 ns         26.0 ns     26551288
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/4        29.9 ns         29.9 ns     23168965
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/6        41.2 ns         41.2 ns     16925473
BM_MatrixTransposeMatrixMultiplyDynamic/3/4/8        51.7 ns         51.7 ns     12910180
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/1        21.7 ns         21.7 ns     31597326
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/2        28.3 ns         28.3 ns     24334122
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/3        44.1 ns         44.1 ns     15845861
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/4        49.7 ns         49.7 ns     13571712
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/6        72.0 ns         72.0 ns      9455799
BM_MatrixTransposeMatrixMultiplyDynamic/3/8/8        97.4 ns         97.4 ns      7001118
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/1       29.3 ns         29.3 ns     23570144
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/2       39.0 ns         39.0 ns     17852225
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/3       62.2 ns         62.2 ns     10856238
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/4       75.0 ns         75.0 ns      8821073
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/6        107 ns          107 ns      6381917
BM_MatrixTransposeMatrixMultiplyDynamic/3/12/8        151 ns          151 ns      4546579
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/1       39.3 ns         39.3 ns     17567309
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/2       51.7 ns         51.7 ns     13221511
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/3       83.5 ns         83.5 ns      8146927
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/4       89.6 ns         89.6 ns      7585598
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/6        133 ns          133 ns      5159159
BM_MatrixTransposeMatrixMultiplyDynamic/3/15/8        178 ns          178 ns      3849984
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/1        6.92 ns         6.92 ns     95871098
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/2        8.03 ns         8.03 ns     84984494
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/3        13.5 ns         13.5 ns     50689926
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/4        16.4 ns         16.4 ns     42066627
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/6        18.3 ns         18.3 ns     37922486
BM_MatrixTransposeMatrixMultiplyDynamic/4/1/8        23.7 ns         23.7 ns     29161957
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/1        9.96 ns         9.96 ns     67872256
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/2        14.5 ns         14.5 ns     47022714
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/3        18.5 ns         18.5 ns     37738658
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/4        21.0 ns         21.0 ns     32821953
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/6        28.8 ns         28.8 ns     24102530
BM_MatrixTransposeMatrixMultiplyDynamic/4/2/8        34.7 ns         34.7 ns     19957110
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/1        14.1 ns         14.1 ns     48095907
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/2        17.1 ns         17.1 ns     40084974
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/3        25.3 ns         25.3 ns     27374300
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/4        26.4 ns         26.4 ns     26222623
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/6        37.2 ns         37.2 ns     18697269
BM_MatrixTransposeMatrixMultiplyDynamic/4/3/8        45.3 ns         45.3 ns     15369099
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/1        16.1 ns         16.1 ns     42512592
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/2        19.9 ns         19.9 ns     35050144
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/3        30.4 ns         30.4 ns     22730866
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/4        31.7 ns         31.7 ns     21860789
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/6        45.6 ns         45.6 ns     15256230
BM_MatrixTransposeMatrixMultiplyDynamic/4/4/8        56.6 ns         56.6 ns     12235547
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/1        26.3 ns         26.3 ns     26758100
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/2        33.9 ns         33.9 ns     20560696
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/3        51.9 ns         51.9 ns     13326339
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/4        54.5 ns         54.5 ns     12683695
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/6        81.9 ns         81.9 ns      8418540
BM_MatrixTransposeMatrixMultiplyDynamic/4/8/8         100 ns          100 ns      6836722
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/1       40.1 ns         40.1 ns     17356727
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/2       51.7 ns         51.7 ns     13582616
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/3       83.9 ns         83.9 ns      8183380
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/4       76.5 ns         76.5 ns      9041249
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/6        121 ns          121 ns      5668697
BM_MatrixTransposeMatrixMultiplyDynamic/4/12/8        144 ns          144 ns      4774784
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/1       46.8 ns         46.8 ns     14867075
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/2       61.2 ns         61.2 ns     11232147
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/3        100 ns          100 ns      6838609
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/4       93.4 ns         93.4 ns      7373474
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/6        148 ns          148 ns      4645822
BM_MatrixTransposeMatrixMultiplyDynamic/4/15/8        177 ns          177 ns      3869284
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/1        7.95 ns         7.95 ns     84762446
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/2        10.3 ns         10.3 ns     67498234
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/3        15.6 ns         15.6 ns     43871824
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/4        18.4 ns         18.4 ns     37355667
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/6        22.9 ns         22.8 ns     30496316
BM_MatrixTransposeMatrixMultiplyDynamic/6/1/8        29.8 ns         29.8 ns     24261698
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/1        14.1 ns         14.1 ns     44432050
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/2        16.8 ns         16.8 ns     41223474
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/3        25.0 ns         25.0 ns     27767778
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/4        25.6 ns         25.6 ns     26941006
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/6        35.6 ns         35.6 ns     19501196
BM_MatrixTransposeMatrixMultiplyDynamic/6/2/8        43.3 ns         43.3 ns     16093244
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/1        16.2 ns         16.1 ns     42352553
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/2        20.5 ns         20.5 ns     32486365
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/3        31.9 ns         31.9 ns     21726529
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/4        33.3 ns         33.3 ns     20956958
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/6        47.3 ns         47.3 ns     14698077
BM_MatrixTransposeMatrixMultiplyDynamic/6/3/8        58.3 ns         58.3 ns     11800586
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/1        18.4 ns         18.4 ns     37239428
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/2        25.9 ns         25.9 ns     26795067
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/3        39.9 ns         39.9 ns     17020628
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/4        40.3 ns         40.3 ns     17235370
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/6        59.8 ns         59.8 ns     11633860
BM_MatrixTransposeMatrixMultiplyDynamic/6/4/8        73.9 ns         73.9 ns      9317366
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/1        40.1 ns         40.1 ns     17363468
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/2        53.0 ns         53.0 ns     13130133
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/3        80.9 ns         80.9 ns      8407136
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/4        71.0 ns         71.0 ns      9756823
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/6         113 ns          113 ns      6007946
BM_MatrixTransposeMatrixMultiplyDynamic/6/8/8         133 ns          133 ns      5183210
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/1       50.8 ns         50.8 ns     13395284
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/2       67.0 ns         67.0 ns     10439863
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/3        111 ns          111 ns      6211108
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/4        101 ns          101 ns      6798646
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/6        159 ns          159 ns      4297178
BM_MatrixTransposeMatrixMultiplyDynamic/6/12/8        192 ns          192 ns      3602507
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/1       60.4 ns         60.4 ns     11212885
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/2       79.8 ns         79.8 ns      8736355
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/3        130 ns          130 ns      5213548
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/4        123 ns          123 ns      5582600
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/6        193 ns          193 ns      3578296
BM_MatrixTransposeMatrixMultiplyDynamic/6/15/8        235 ns          235 ns      2937013
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/1        9.17 ns         9.17 ns     73134052
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/2        14.0 ns         14.0 ns     48667849
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/3        17.3 ns         17.3 ns     39761098
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/4        19.3 ns         19.3 ns     35567020
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/6        27.0 ns         27.0 ns     25625384
BM_MatrixTransposeMatrixMultiplyDynamic/8/1/8        32.2 ns         32.2 ns     21570142
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/1        15.1 ns         15.1 ns     45303514
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/2        18.6 ns         18.6 ns     37322471
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/3        29.6 ns         29.6 ns     23415483
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/4        28.4 ns         28.4 ns     24458051
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/6        40.9 ns         40.9 ns     17024035
BM_MatrixTransposeMatrixMultiplyDynamic/8/2/8        49.4 ns         49.4 ns     14133068
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/1        18.0 ns         18.0 ns     37951528
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/2        25.5 ns         25.5 ns     27155372
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/3        38.9 ns         38.9 ns     17861019
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/4        36.6 ns         36.6 ns     18932495
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/6        56.0 ns         56.0 ns     12273509
BM_MatrixTransposeMatrixMultiplyDynamic/8/3/8        66.4 ns         66.4 ns     10327587
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/1        24.7 ns         24.7 ns     27896582
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/2        31.5 ns         31.5 ns     22104946
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/3        48.9 ns         48.9 ns     13525414
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/4        45.6 ns         45.6 ns     15252423
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/6        70.5 ns         70.5 ns      9746886
BM_MatrixTransposeMatrixMultiplyDynamic/8/4/8        83.7 ns         83.7 ns      8145416
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/1        45.6 ns         45.6 ns     15244971
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/2        58.7 ns         58.7 ns     11783597
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/3        96.6 ns         96.6 ns      7028760
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/4        82.5 ns         82.5 ns      8422218
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/6         132 ns          132 ns      5234277
BM_MatrixTransposeMatrixMultiplyDynamic/8/8/8         153 ns          153 ns      4585471
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/1       65.1 ns         65.1 ns     10474283
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/2       81.8 ns         81.8 ns      8520237
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/3        139 ns          139 ns      4984273
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/4        117 ns          117 ns      5872776
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/6        188 ns          188 ns      3660630
BM_MatrixTransposeMatrixMultiplyDynamic/8/12/8        229 ns          229 ns      3028433
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/1       80.3 ns         80.3 ns      8514280
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/2       98.5 ns         98.4 ns      7074805
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/3        169 ns          169 ns      4059933
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/4        150 ns          150 ns      4574776
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/6        238 ns          238 ns      2914228
BM_MatrixTransposeMatrixMultiplyDynamic/8/15/8        296 ns          296 ns      2340432

[-- Attachment #14: small_blas_gemv_benchmark-diff --]
[-- Type: text/plain, Size: 6972 bytes --]

Benchmark                                              Time        CPU   Iterations
BM_MatrixVectorMultiply/1/1                           5.8 %      5.8 %      3.3 %
BM_MatrixVectorMultiply/1/2                          11.8 %     11.8 %     -9.8 %
BM_MatrixVectorMultiply/1/3                           3.1 %      3.1 %     -2.0 %
BM_MatrixVectorMultiply/1/4                          -1.2 %     -1.2 %    -12.0 %
BM_MatrixVectorMultiply/1/8                           3.4 %      3.4 %     -4.5 %
BM_MatrixVectorMultiply/1/12                          1.7 %      1.7 %     -2.1 %
BM_MatrixVectorMultiply/1/15                          0.0 %      0.0 %     -0.4 %
BM_MatrixVectorMultiply/2/1                          -0.4 %     -0.4 %     -3.0 %
BM_MatrixVectorMultiply/2/2                          -5.0 %     -5.2 %      3.3 %
BM_MatrixVectorMultiply/2/3                          -4.8 %     -4.8 %      1.6 %
BM_MatrixVectorMultiply/2/4                          -5.4 %     -5.4 %      2.8 %
BM_MatrixVectorMultiply/2/8                          -2.4 %     -2.4 %      2.7 %
BM_MatrixVectorMultiply/2/12                         -2.1 %     -2.1 %      1.8 %
BM_MatrixVectorMultiply/2/15                         -0.6 %     -0.6 %      0.3 %
BM_MatrixVectorMultiply/3/1                          -3.6 %     -3.5 %      5.0 %
BM_MatrixVectorMultiply/3/2                           8.2 %      8.2 %     -8.9 %
BM_MatrixVectorMultiply/3/3                         -10.2 %    -10.2 %     12.2 %
BM_MatrixVectorMultiply/3/4                          -2.5 %     -2.5 %      1.7 %
BM_MatrixVectorMultiply/3/8                           0.0 %      0.0 %      0.6 %
BM_MatrixVectorMultiply/3/12                          1.6 %      1.6 %     -1.6 %
BM_MatrixVectorMultiply/3/15                         -1.7 %     -1.7 %      1.4 %
BM_MatrixVectorMultiply/4/1                           1.2 %      1.2 %     -0.8 %
BM_MatrixVectorMultiply/4/2                          -6.3 %     -6.3 %      6.9 %
BM_MatrixVectorMultiply/4/3                          12.8 %     12.8 %    -12.2 %
BM_MatrixVectorMultiply/4/4                          -1.6 %     -1.6 %      2.3 %
BM_MatrixVectorMultiply/4/8                          -1.3 %     -1.3 %      1.8 %
BM_MatrixVectorMultiply/4/12                         -1.5 %     -1.5 %      2.0 %
BM_MatrixVectorMultiply/4/15                         -4.3 %     -4.3 %      6.2 %
BM_MatrixVectorMultiply/6/1                          -2.5 %     -2.5 %      3.7 %
BM_MatrixVectorMultiply/6/2                          -0.8 %     -0.8 %      0.5 %
BM_MatrixVectorMultiply/6/3                           0.7 %      0.7 %      0.6 %
BM_MatrixVectorMultiply/6/4                          -2.0 %     -2.0 %      1.4 %
BM_MatrixVectorMultiply/6/8                          -1.8 %     -1.8 %      1.2 %
BM_MatrixVectorMultiply/6/12                        -15.2 %    -15.2 %     17.7 %
BM_MatrixVectorMultiply/6/15                        -19.6 %    -19.6 %     20.6 %
BM_MatrixVectorMultiply/8/1                           0.0 %      0.0 %      0.4 %
BM_MatrixVectorMultiply/8/2                           0.7 %      0.7 %     -0.7 %
BM_MatrixVectorMultiply/8/3                          -4.5 %     -4.5 %      4.7 %
BM_MatrixVectorMultiply/8/4                          -1.2 %     -1.2 %      1.8 %
BM_MatrixVectorMultiply/8/8                          -2.5 %     -2.5 %      2.0 %
BM_MatrixVectorMultiply/8/12                         -6.4 %     -6.4 %      7.7 %
BM_MatrixVectorMultiply/8/15                         -9.3 %     -9.3 %     10.6 %
BM_MatrixTransposeVectorMultiply/1/1                 -6.9 %     -6.9 %      6.0 %
BM_MatrixTransposeVectorMultiply/1/2                 -1.1 %     -1.1 %      2.7 %
BM_MatrixTransposeVectorMultiply/1/3                  5.7 %      5.7 %     -4.8 %
BM_MatrixTransposeVectorMultiply/1/4                  0.5 %      0.5 %     -2.3 %
BM_MatrixTransposeVectorMultiply/1/8                  0.0 %      0.0 %      0.2 %
BM_MatrixTransposeVectorMultiply/1/12                 0.0 %      0.0 %      0.3 %
BM_MatrixTransposeVectorMultiply/1/15                -1.3 %     -1.3 %      1.5 %
BM_MatrixTransposeVectorMultiply/2/1                  1.2 %      1.2 %     -0.9 %
BM_MatrixTransposeVectorMultiply/2/2                 10.1 %     10.1 %    -10.2 %
BM_MatrixTransposeVectorMultiply/2/3                  1.5 %      1.5 %     -4.2 %
BM_MatrixTransposeVectorMultiply/2/4                 -1.4 %     -1.4 %      1.0 %
BM_MatrixTransposeVectorMultiply/2/8                 -0.8 %     -0.8 %      1.3 %
BM_MatrixTransposeVectorMultiply/2/12                 1.9 %      1.9 %     -1.8 %
BM_MatrixTransposeVectorMultiply/2/15                -7.1 %     -7.1 %      8.0 %
BM_MatrixTransposeVectorMultiply/3/1                  1.5 %      1.5 %     -1.9 %
BM_MatrixTransposeVectorMultiply/3/2                 -5.7 %     -5.6 %      4.2 %
BM_MatrixTransposeVectorMultiply/3/3                 -2.1 %     -2.1 %      1.7 %
BM_MatrixTransposeVectorMultiply/3/4                  2.0 %      2.0 %     -1.7 %
BM_MatrixTransposeVectorMultiply/3/8                 -1.4 %     -1.4 %      1.6 %
BM_MatrixTransposeVectorMultiply/3/12                -2.2 %     -2.2 %      2.3 %
BM_MatrixTransposeVectorMultiply/3/15               -12.3 %    -12.3 %     18.0 %
BM_MatrixTransposeVectorMultiply/4/1                 -1.5 %     -1.5 %      0.8 %
BM_MatrixTransposeVectorMultiply/4/2                 -8.2 %     -8.2 %      9.3 %
BM_MatrixTransposeVectorMultiply/4/3                  0.9 %      0.9 %     -0.6 %
BM_MatrixTransposeVectorMultiply/4/4                 -3.3 %     -3.3 %      4.2 %
BM_MatrixTransposeVectorMultiply/4/8                 -2.4 %     -2.4 %      3.0 %
BM_MatrixTransposeVectorMultiply/4/12                -2.4 %     -2.4 %      3.8 %
BM_MatrixTransposeVectorMultiply/4/15                -9.4 %     -9.4 %     11.6 %
BM_MatrixTransposeVectorMultiply/6/1                  1.6 %      1.6 %     -2.8 %
BM_MatrixTransposeVectorMultiply/6/2                 -1.8 %     -1.8 %      2.5 %
BM_MatrixTransposeVectorMultiply/6/3                 -0.7 %     -0.7 %      1.0 %
BM_MatrixTransposeVectorMultiply/6/4                  0.7 %      0.7 %     -0.3 %
BM_MatrixTransposeVectorMultiply/6/8                 -1.0 %     -1.0 %      1.2 %
BM_MatrixTransposeVectorMultiply/6/12                -4.4 %     -4.4 %      4.1 %
BM_MatrixTransposeVectorMultiply/6/15               -10.8 %    -10.8 %     12.5 %
BM_MatrixTransposeVectorMultiply/8/1                  1.1 %      1.1 %     -1.7 %
BM_MatrixTransposeVectorMultiply/8/2                 -3.2 %     -3.2 %      3.5 %
BM_MatrixTransposeVectorMultiply/8/3                 -0.7 %     -0.7 %      1.0 %
BM_MatrixTransposeVectorMultiply/8/4                 -0.7 %     -0.7 %      0.1 %
BM_MatrixTransposeVectorMultiply/8/8                 -1.3 %     -1.3 %      0.7 %
BM_MatrixTransposeVectorMultiply/8/12                -5.7 %     -5.7 %      7.9 %
BM_MatrixTransposeVectorMultiply/8/15                11.6 %     11.6 %    -34.4 %

[-- Attachment #15: small_blas_gemv_benchmark-tuned --]
[-- Type: text/plain, Size: 7047 bytes --]

--------------------------------------------------------------------------------
Benchmark                                      Time             CPU   Iterations
--------------------------------------------------------------------------------
BM_MatrixVectorMultiply/1/1                 9.12 ns         9.12 ns     92925630
BM_MatrixVectorMultiply/1/2                 6.54 ns         6.54 ns    101637206
BM_MatrixVectorMultiply/1/3                 6.65 ns         6.65 ns    101375991
BM_MatrixVectorMultiply/1/4                 6.44 ns         6.44 ns     90400426
BM_MatrixVectorMultiply/1/8                 7.37 ns         7.37 ns     90705905
BM_MatrixVectorMultiply/1/12                11.9 ns         11.9 ns     56883700
BM_MatrixVectorMultiply/1/15                12.7 ns         12.7 ns     53579315
BM_MatrixVectorMultiply/2/1                 6.66 ns         6.66 ns     99249936
BM_MatrixVectorMultiply/2/2                 6.60 ns         6.59 ns    101068476
BM_MatrixVectorMultiply/2/3                 6.92 ns         6.92 ns     97081578
BM_MatrixVectorMultiply/2/4                 6.25 ns         6.25 ns    108396767
BM_MatrixVectorMultiply/2/8                 12.0 ns         12.0 ns     56675174
BM_MatrixVectorMultiply/2/12                13.8 ns         13.8 ns     49537513
BM_MatrixVectorMultiply/2/15                15.3 ns         15.3 ns     44545580
BM_MatrixVectorMultiply/3/1                 6.91 ns         6.91 ns     98564055
BM_MatrixVectorMultiply/3/2                 7.65 ns         7.65 ns     89352471
BM_MatrixVectorMultiply/3/3                 7.99 ns         7.99 ns     85849057
BM_MatrixVectorMultiply/3/4                 11.7 ns         11.7 ns     58146713
BM_MatrixVectorMultiply/3/8                 15.0 ns         15.0 ns     45791876
BM_MatrixVectorMultiply/3/12                19.1 ns         19.1 ns     35725402
BM_MatrixVectorMultiply/3/15                23.5 ns         23.5 ns     29307973
BM_MatrixVectorMultiply/4/1                 7.31 ns         7.31 ns     92852521
BM_MatrixVectorMultiply/4/2                 7.19 ns         7.19 ns     96217600
BM_MatrixVectorMultiply/4/3                 10.6 ns         10.6 ns     63654395
BM_MatrixVectorMultiply/4/4                 12.1 ns         12.1 ns     56767011
BM_MatrixVectorMultiply/4/8                 15.5 ns         15.5 ns     44399714
BM_MatrixVectorMultiply/4/12                19.2 ns         19.2 ns     35878150
BM_MatrixVectorMultiply/4/15                22.2 ns         22.2 ns     31104148
BM_MatrixVectorMultiply/6/1                 7.93 ns         7.93 ns     86048156
BM_MatrixVectorMultiply/6/2                 12.0 ns         12.0 ns     56748306
BM_MatrixVectorMultiply/6/3                 13.6 ns         13.6 ns     50775685
BM_MatrixVectorMultiply/6/4                 14.8 ns         14.8 ns     46182567
BM_MatrixVectorMultiply/6/8                 21.6 ns         21.6 ns     32084929
BM_MatrixVectorMultiply/6/12                27.4 ns         27.4 ns     25128036
BM_MatrixVectorMultiply/6/15                31.9 ns         31.9 ns     21750514
BM_MatrixVectorMultiply/8/1                 11.7 ns         11.7 ns     57656634
BM_MatrixVectorMultiply/8/2                 13.5 ns         13.5 ns     50855999
BM_MatrixVectorMultiply/8/3                 14.7 ns         14.7 ns     46511976
BM_MatrixVectorMultiply/8/4                 16.1 ns         16.1 ns     42358995
BM_MatrixVectorMultiply/8/8                 23.5 ns         23.5 ns     29298978
BM_MatrixVectorMultiply/8/12                30.8 ns         30.8 ns     22470550
BM_MatrixVectorMultiply/8/15                36.2 ns         36.2 ns     19253927
BM_MatrixTransposeVectorMultiply/1/1        7.54 ns         7.54 ns     88868173
BM_MatrixTransposeVectorMultiply/1/2        6.93 ns         6.93 ns     97536428
BM_MatrixTransposeVectorMultiply/1/3        7.22 ns         7.22 ns     92829699
BM_MatrixTransposeVectorMultiply/1/4        6.56 ns         6.56 ns    102015165
BM_MatrixTransposeVectorMultiply/1/8        11.2 ns         11.2 ns     60973842
BM_MatrixTransposeVectorMultiply/1/12       13.3 ns         13.3 ns     51509976
BM_MatrixTransposeVectorMultiply/1/15       15.1 ns         15.1 ns     45394002
BM_MatrixTransposeVectorMultiply/2/1        6.65 ns         6.65 ns    101908239
BM_MatrixTransposeVectorMultiply/2/2        6.77 ns         6.77 ns     99096130
BM_MatrixTransposeVectorMultiply/2/3        7.63 ns         7.63 ns     87992619
BM_MatrixTransposeVectorMultiply/2/4        7.04 ns         7.04 ns     97423520
BM_MatrixTransposeVectorMultiply/2/8        12.6 ns         12.6 ns     54249770
BM_MatrixTransposeVectorMultiply/2/12       15.9 ns         15.9 ns     43068768
BM_MatrixTransposeVectorMultiply/2/15       18.2 ns         18.2 ns     37852292
BM_MatrixTransposeVectorMultiply/3/1        6.20 ns         6.20 ns    107691519
BM_MatrixTransposeVectorMultiply/3/2        6.92 ns         6.92 ns     99210487
BM_MatrixTransposeVectorMultiply/3/3        8.97 ns         8.97 ns     75914900
BM_MatrixTransposeVectorMultiply/3/4        10.4 ns         10.4 ns     65456106
BM_MatrixTransposeVectorMultiply/3/8        14.4 ns         14.4 ns     47805949
BM_MatrixTransposeVectorMultiply/3/12       17.7 ns         17.7 ns     39029777
BM_MatrixTransposeVectorMultiply/3/15       21.4 ns         21.4 ns     32168470
BM_MatrixTransposeVectorMultiply/4/1        7.07 ns         7.07 ns     94821144
BM_MatrixTransposeVectorMultiply/4/2        6.71 ns         6.71 ns    103919623
BM_MatrixTransposeVectorMultiply/4/3        11.7 ns         11.7 ns     58048173
BM_MatrixTransposeVectorMultiply/4/4        11.7 ns         11.7 ns     58781880
BM_MatrixTransposeVectorMultiply/4/8        16.0 ns         16.0 ns     43027790
BM_MatrixTransposeVectorMultiply/4/12       20.3 ns         20.3 ns     33884938
BM_MatrixTransposeVectorMultiply/4/15       25.9 ns         25.9 ns     26742685
BM_MatrixTransposeVectorMultiply/6/1        6.89 ns         6.89 ns     97270079
BM_MatrixTransposeVectorMultiply/6/2        9.08 ns         9.08 ns     75334071
BM_MatrixTransposeVectorMultiply/6/3        13.4 ns         13.4 ns     51036563
BM_MatrixTransposeVectorMultiply/6/4        13.6 ns         13.6 ns     50318744
BM_MatrixTransposeVectorMultiply/6/8        19.7 ns         19.7 ns     35105747
BM_MatrixTransposeVectorMultiply/6/12       26.0 ns         26.0 ns     26587912
BM_MatrixTransposeVectorMultiply/6/15       33.0 ns         33.0 ns     21095033
BM_MatrixTransposeVectorMultiply/8/1        7.69 ns         7.69 ns     86590464
BM_MatrixTransposeVectorMultiply/8/2        12.2 ns         12.2 ns     55865294
BM_MatrixTransposeVectorMultiply/8/3        14.8 ns         14.8 ns     46463099
BM_MatrixTransposeVectorMultiply/8/4        14.9 ns         14.9 ns     45516102
BM_MatrixTransposeVectorMultiply/8/8        23.4 ns         23.4 ns     29429471
BM_MatrixTransposeVectorMultiply/8/12       31.7 ns         31.7 ns     22293660
BM_MatrixTransposeVectorMultiply/8/15       50.9 ns         50.9 ns     10000000

[-- Attachment #16: small_blas_gemv_benchmark-untuned --]
[-- Type: text/plain, Size: 7047 bytes --]

--------------------------------------------------------------------------------
Benchmark                                      Time             CPU   Iterations
--------------------------------------------------------------------------------
BM_MatrixVectorMultiply/1/1                 8.62 ns         8.62 ns     89948658
BM_MatrixVectorMultiply/1/2                 5.85 ns         5.85 ns    112691480
BM_MatrixVectorMultiply/1/3                 6.45 ns         6.45 ns    103471625
BM_MatrixVectorMultiply/1/4                 6.52 ns         6.52 ns    102738581
BM_MatrixVectorMultiply/1/8                 7.13 ns         7.13 ns     94938191
BM_MatrixVectorMultiply/1/12                11.7 ns         11.7 ns     58124147
BM_MatrixVectorMultiply/1/15                12.7 ns         12.7 ns     53792008
BM_MatrixVectorMultiply/2/1                 6.69 ns         6.69 ns    102357695
BM_MatrixVectorMultiply/2/2                 6.95 ns         6.95 ns     97836014
BM_MatrixVectorMultiply/2/3                 7.27 ns         7.27 ns     95547591
BM_MatrixVectorMultiply/2/4                 6.61 ns         6.61 ns    105486275
BM_MatrixVectorMultiply/2/8                 12.3 ns         12.3 ns     55181473
BM_MatrixVectorMultiply/2/12                14.1 ns         14.1 ns     48685219
BM_MatrixVectorMultiply/2/15                15.4 ns         15.4 ns     44428862
BM_MatrixVectorMultiply/3/1                 7.17 ns         7.16 ns     93878590
BM_MatrixVectorMultiply/3/2                 7.07 ns         7.07 ns     98093589
BM_MatrixVectorMultiply/3/3                 8.90 ns         8.90 ns     76536309
BM_MatrixVectorMultiply/3/4                 12.0 ns         12.0 ns     57186993
BM_MatrixVectorMultiply/3/8                 15.0 ns         15.0 ns     45521769
BM_MatrixVectorMultiply/3/12                18.8 ns         18.8 ns     36292397
BM_MatrixVectorMultiply/3/15                23.9 ns         23.9 ns     28910412
BM_MatrixVectorMultiply/4/1                 7.22 ns         7.22 ns     93571728
BM_MatrixVectorMultiply/4/2                 7.67 ns         7.67 ns     90031857
BM_MatrixVectorMultiply/4/3                 9.40 ns         9.40 ns     72519946
BM_MatrixVectorMultiply/4/4                 12.3 ns         12.3 ns     55480795
BM_MatrixVectorMultiply/4/8                 15.7 ns         15.7 ns     43606420
BM_MatrixVectorMultiply/4/12                19.5 ns         19.5 ns     35167444
BM_MatrixVectorMultiply/4/15                23.2 ns         23.2 ns     29294134
BM_MatrixVectorMultiply/6/1                 8.13 ns         8.13 ns     82941916
BM_MatrixVectorMultiply/6/2                 12.1 ns         12.1 ns     56472023
BM_MatrixVectorMultiply/6/3                 13.5 ns         13.5 ns     50450079
BM_MatrixVectorMultiply/6/4                 15.1 ns         15.1 ns     45553440
BM_MatrixVectorMultiply/6/8                 22.0 ns         22.0 ns     31705708
BM_MatrixVectorMultiply/6/12                32.3 ns         32.3 ns     21357555
BM_MatrixVectorMultiply/6/15                39.7 ns         39.7 ns     18035575
BM_MatrixVectorMultiply/8/1                 11.7 ns         11.7 ns     57437148
BM_MatrixVectorMultiply/8/2                 13.4 ns         13.4 ns     51218500
BM_MatrixVectorMultiply/8/3                 15.4 ns         15.4 ns     44427668
BM_MatrixVectorMultiply/8/4                 16.3 ns         16.3 ns     41592264
BM_MatrixVectorMultiply/8/8                 24.1 ns         24.1 ns     28713260
BM_MatrixVectorMultiply/8/12                32.9 ns         32.9 ns     20854799
BM_MatrixVectorMultiply/8/15                39.9 ns         39.9 ns     17415005
BM_MatrixTransposeVectorMultiply/1/1        8.10 ns         8.10 ns     83855577
BM_MatrixTransposeVectorMultiply/1/2        7.01 ns         7.01 ns     95009714
BM_MatrixTransposeVectorMultiply/1/3        6.83 ns         6.83 ns     97523522
BM_MatrixTransposeVectorMultiply/1/4        6.53 ns         6.53 ns    104425807
BM_MatrixTransposeVectorMultiply/1/8        11.2 ns         11.2 ns     60837585
BM_MatrixTransposeVectorMultiply/1/12       13.3 ns         13.3 ns     51352263
BM_MatrixTransposeVectorMultiply/1/15       15.3 ns         15.3 ns     44707928
BM_MatrixTransposeVectorMultiply/2/1        6.57 ns         6.57 ns    102832073
BM_MatrixTransposeVectorMultiply/2/2        6.15 ns         6.15 ns    110364516
BM_MatrixTransposeVectorMultiply/2/3        7.52 ns         7.52 ns     91844921
BM_MatrixTransposeVectorMultiply/2/4        7.14 ns         7.14 ns     96425429
BM_MatrixTransposeVectorMultiply/2/8        12.7 ns         12.7 ns     53572754
BM_MatrixTransposeVectorMultiply/2/12       15.6 ns         15.6 ns     43880286
BM_MatrixTransposeVectorMultiply/2/15       19.6 ns         19.6 ns     35057393
BM_MatrixTransposeVectorMultiply/3/1        6.11 ns         6.11 ns    109824234
BM_MatrixTransposeVectorMultiply/3/2        7.34 ns         7.33 ns     95225678
BM_MatrixTransposeVectorMultiply/3/3        9.16 ns         9.16 ns     74666520
BM_MatrixTransposeVectorMultiply/3/4        10.2 ns         10.2 ns     66580926
BM_MatrixTransposeVectorMultiply/3/8        14.6 ns         14.6 ns     47048162
BM_MatrixTransposeVectorMultiply/3/12       18.1 ns         18.1 ns     38157417
BM_MatrixTransposeVectorMultiply/3/15       24.4 ns         24.4 ns     27272734
BM_MatrixTransposeVectorMultiply/4/1        7.18 ns         7.18 ns     94058873
BM_MatrixTransposeVectorMultiply/4/2        7.31 ns         7.31 ns     95039943
BM_MatrixTransposeVectorMultiply/4/3        11.6 ns         11.6 ns     58401243
BM_MatrixTransposeVectorMultiply/4/4        12.1 ns         12.1 ns     56414612
BM_MatrixTransposeVectorMultiply/4/8        16.4 ns         16.4 ns     41789101
BM_MatrixTransposeVectorMultiply/4/12       20.8 ns         20.8 ns     32636074
BM_MatrixTransposeVectorMultiply/4/15       28.6 ns         28.6 ns     23969087
BM_MatrixTransposeVectorMultiply/6/1        6.78 ns         6.78 ns    100088196
BM_MatrixTransposeVectorMultiply/6/2        9.25 ns         9.25 ns     73515383
BM_MatrixTransposeVectorMultiply/6/3        13.5 ns         13.5 ns     50522373
BM_MatrixTransposeVectorMultiply/6/4        13.5 ns         13.5 ns     50463809
BM_MatrixTransposeVectorMultiply/6/8        19.9 ns         19.9 ns     34705192
BM_MatrixTransposeVectorMultiply/6/12       27.2 ns         27.2 ns     25534377
BM_MatrixTransposeVectorMultiply/6/15       37.0 ns         37.0 ns     18744927
BM_MatrixTransposeVectorMultiply/8/1        7.61 ns         7.61 ns     88069473
BM_MatrixTransposeVectorMultiply/8/2        12.6 ns         12.6 ns     53966293
BM_MatrixTransposeVectorMultiply/8/3        14.9 ns         14.9 ns     45996393
BM_MatrixTransposeVectorMultiply/8/4        15.0 ns         15.0 ns     45481312
BM_MatrixTransposeVectorMultiply/8/8        23.7 ns         23.7 ns     29237859
BM_MatrixTransposeVectorMultiply/8/12       33.6 ns         33.6 ns     20651875
BM_MatrixTransposeVectorMultiply/8/15       45.6 ns         45.6 ns     15235896

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [bug#53121] [PATCH] gnu: ceres: Update to 2.0.0.
  2022-01-18 12:39   ` [bug#53121] " Felix Gruber
@ 2022-01-19 10:26     ` Ludovic Courtès
  0 siblings, 0 replies; 4+ messages in thread
From: Ludovic Courtès @ 2022-01-19 10:26 UTC (permalink / raw)
  To: Felix Gruber; +Cc: 53121-done

Hi,

Felix Gruber <felgru@posteo.net> skribis:

> Unfortunately, I'm getting mixed results for the benchmarks. In most
> cases, I got slight (<10%) improvements in runtime, but there are also 
> some benchmarks that were worse with the --tune flag. I'm wondering
> whether the compiler flags set by the --tune option are correctly used 
> by the custom 'build phase of the ceres-solver-benchmarks package. I
> didn't have the time to look closer into it as I'm currently in the 
> middle of moving to another country.

OK.

> Anyways, I've attached the results of benchmark runs that I've
> generated using guix commit 7f779286df7e8636d901f4734501902cc934a72f
> once untuned and once tuned for broadwell CPUs.
> My laptop on which I ran the tests has a Quad Core AMD Ryzen 7 PRO
> 2700U CPU with 2200 MHz.

Could it be that ‘znver3’ or something works better on those CPUs?

> In the attachments you find
> * a script run_benchmarks.sh used to run the benchmarks in tuned and
>   untuned guix shells,
> * text files ending in `-tuned` or `-untuned` which contain the
>   results of those benchmark runs,
> * a script compare.sh which calls a Python script compare-results.py
>   to generate files ending in `-diff` that contain the relative change 
> between untuned and tuned benchmarks (negative time and CPU
> percentages mean the tuned benchmark was faster, while for the number
> of iterations, positive percentages mean the tuned benchmark had run
> more iterations).

Interesting, thanks for taking the time to run these benchmarks.

It’s hard to draw conclusions.  I wonder how noisy these measurements
are and whether the differences we’re seeing are significant.  Food for
thoughts!

Ludo’.




^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2022-01-19 10:38 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2022-01-07 22:02 [bug#53121] [PATCH] gnu: ceres: Update to 2.0.0 Felix Gruber
2022-01-11 19:10 ` bug#53121: " Ludovic Courtès
2022-01-18 12:39   ` [bug#53121] " Felix Gruber
2022-01-19 10:26     ` Ludovic Courtès

Code repositories for project(s) associated with this external index

	https://git.savannah.gnu.org/cgit/guix.git

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.