It seems the execute implementations have gone out of their way to
produce inconsistent error messages. The unix version explicitly
checks if the file exists before trying to execute. The windows
version checks if it's executable. I don't understand why they
wouldn't just try the execution and check the error code.
# RUN: not llvm-reduce --test=%s.NotAFileInTestingDir %p/Inputs/test-output-format.ll 2>&1 | FileCheck -DFILENAME=%s.NotAFileInTestingDir --strict-whitespace %s
-# CHECK: Error running interesting-ness test: Executable "[[FILENAME]]" doesn't exist!{{$}}
\ No newline at end of file
+# CHECK: Error running interesting-ness test: {{(Executable "[[FILENAME]]" doesn't exist$)?(program not executable$)?}}
\ No newline at end of file