Vicente Bolea

@pnorbert this is just adding the same comment in about every test that do `MPI_Thread_init(...MPI_THREAD_MULTIPLE)`. I will merge this soon without review since it does not change any functional code.

@hyoklee it is near impossible to debug a test error with `--test root` since it simply internally calls `make test`, unfortunately it does not allows you to change the flags...

@lwan86 this final error seems to be related to the locales of your system. See this answer (https://stackoverflow.com/questions/13385780/i-am-not-able-to-compile-with-mpi-compiler-with-c). On the other hand I will solely export the fortram compiler since...

@ax3l It seems that some variables are not being passing down to the install tests. In the particular case of macOS, the CMake variables such as ' https://github.com/conda-forge/adios2-feedstock/blob/2f5d76640398af1af5298b4cd300706679177e4d/recipe/cross-osx.cmake': ``` set(CMAKE_FIND_ROOT_PATH_MODE_LIBRARY...

@JasonRuonanWang I cannot seem to reproduce this, could you retry this?

@LebedevRI thanks for the prompt reply. I just went through the thread, great read. And I agree with your initial idea of using tdist. However, it still does not answer...

@LebedevRI Here is where I am confused, we set the line as red (failure) when the large pvalue (results are same) and green when results are different. It is counter-intuitive...

> Do you believe the documentation is backwards to what we do? > If there is a sufficient repetition count of the benchmarks, the tool can do a [U Test](https://en.wikipedia.org/wiki/Mann%E2%80%93Whitney_U_test),...

@LebedevRI probably at before the last sentence at https://github.com/google/benchmark/blob/main/docs/tools.md

> Patches welcomed? Is this a question or an statement?