[lmi] Benchmarking: gcc-8 beats gcc-10 soundly?

[Greg Chicares]

It looks like gcc-10 gives us slower lmi binaries. Picking
the third '--selftest' scenario as an index of performance
(results in microseconds--less is better):

     gcc-10   gcc-8  ratio
     ------   -----  -----
     102659   84947   1.21  32-bit
      50121   37410   1.34  64-bit

The fourth scenario is even worse:

      33250   20654   1.61  32-bit
      24616   13009   1.89  64-bit

Comparing to the benchmarks here:
  
https://openbenchmarking.org/result/1912179-HU-COREI759632&obr_sgm=y&obr_vb=y&obr_sgm=y&obr_swl=y&obr_vb=y
lmi fares worse than the worst phoronix scenarios, i.e.,
"libgav1" and "function objects".

Vadim--Does this seem so astonishing that it can't be
true? These results are all observed on the same machine.
The only real difference I can think of is that one is a
  new debian bullseye within centos within debian buster chroot
while the other is an
  old debian bullseye within debian buster chroot [my old gcc-8]
both of which identify themselves as:
  Debian GNU/Linux bullseye/sid
But I've never noticed any penalty before for nested chroots.

The gcc-8 timings above are from "README.branch.patch"
on branch valyuta/002 (pushed to savannah). [Although
the last sentence mentions a branch, all timings in
this email are for 'master'.] Here are the commands
that produced the gcc-10 timings:

/opt/lmi/src/lmi[0]$i686-w64-mingw32-gcc --version
i686-w64-mingw32-gcc (GCC) 10-win32 20200525 [...]
/opt/lmi/src/lmi[0]$x86_64-w64-mingw32-gcc --version
x86_64-w64-mingw32-gcc (GCC) 10-win32 20200525 [...]

/opt/lmi/src/lmi[0]$git switch master
Switched to branch 'master'
Your branch is up to date with 'origin/master'.

/opt/lmi/src/lmi[0]$env |grep LMI_
LMI_COMPILER=gcc
LMI_TRIPLET=i686-w64-mingw32

/opt/lmi/src/lmi[0]$make $coefficiency --output-sync=recurse install 
check_physical_closure 2>&1 | tee eraseme | less -SN
/opt/lmi/src/lmi[0]$wine /opt/lmi/bin/lmi_cli_shared.exe --accept 
--data_path=/opt/lmi/data --selftest
Test speed:
  naic, no solve      : 6.965e-02 s mean;      66490 us least of  15 runs
  naic, specamt solve : 1.114e-01 s mean;     110625 us least of   9 runs
  naic, ee prem solve : 1.030e-01 s mean;     102659 us least of  10 runs
  finra, no solve     : 3.460e-02 s mean;      33250 us least of  29 runs
  finra, specamt solve: 7.337e-02 s mean;      72938 us least of  14 runs
  finra, ee prem solve: 6.886e-02 s mean;      68593 us least of  15 runs

/opt/lmi/src/lmi[0]$LMI_COMPILER=gcc ; LMI_TRIPLET=x86_64-w64-mingw32 ; . 
/opt/lmi/src/lmi/set_toolchain.sh
/opt/lmi/src/lmi[0]$env |grep LMI_
LMI_COMPILER=gcc
LMI_TRIPLET=x86_64-w64-mingw32

/opt/lmi/src/lmi[0]$make $coefficiency --output-sync=recurse install 
check_physical_closure 2>&1 | tee eraseme | less -SN
/opt/lmi/src/lmi[0]$wine /opt/lmi/bin/lmi_cli_shared.exe --accept 
--data_path=/opt/lmi/data --selftest     
Test speed:
  naic, no solve      : 3.696e-02 s mean;      36789 us least of  28 runs
  naic, specamt solve : 5.340e-02 s mean;      53208 us least of  19 runs
  naic, ee prem solve : 5.033e-02 s mean;      50121 us least of  20 runs
  finra, no solve     : 2.483e-02 s mean;      24616 us least of  41 runs
  finra, specamt solve: 3.974e-02 s mean;      39548 us least of  26 runs
  finra, ee prem solve: 3.785e-02 s mean;      37684 us least of  27 runs