Peter Willendrup
Peter Willendrup
I have installed myself a fresh macOS 11.13 High Sierra VM, and loaded up a McStas 2.5. On this system I have  I...
It does however seem that my compiler is slightly different still from what @mads-bertelsen initially reported: ``` Apple LLVM version 10.0.0 (clang-1000.11.45.5) Target: x86_64-apple-darwin17.7.0 Thread model: posix InstalledDir: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin ```...
Ah! Now reproduced without updating Xcode - with: ``` mcstas n_component.txt clang -lm n_component.txt.c ./a.out ``` The default McStas CFLAGS include compiler optimisation to level -O2 which is what I...
OK, now also reproduced on Big Sur for the 2.x releases, 3.0 does not behave this way:  - and the moment one uses...
@mads-bertelsen agreed, this is a bit surprising. But maybe (by sheer luck) -O2 clears out e.g. some unused vars/arrays that are part of the equation? I tried running on Linux...
Or maybe some Valgrind could hint in the direction of some issues? On the other hand, that is mostly for dynamic arrays etc., right? (Which maybe are not really the...
Note to self: MIT license has to be explicit for the *conic*.h files
share-code, components and (a first) test instrument added to mcstas and mccode-3
The code now offloads to GPU in mccode-3 and uses the McStas RNG + standard reflectivity function. The components currently forward the `R0,Qc,m,alpha,W` parameters but not the reflectivity-file parametrisation.
Postpone -> after 3.1