Mac version soon?

[xpertxp]

Will there be a Mac OS silicon version?

Thanks

[akdev1l]

I tried to build this for aarch64 and I was able to get a binary, however it looks like it crashes with the following backtrace:

Thread 0 Crashed:
0   libsystem_kernel.dylib        	       0x18bf7e5b0 __pthread_kill + 8
1   libsystem_pthread.dylib       	       0x18bfb8888 pthread_kill + 296
2   libsystem_c.dylib             	       0x18bebe808 abort + 124
3   libc++abi.dylib               	       0x18bf6c858 __abort_message + 132
4   libc++abi.dylib               	       0x18bf5b4bc demangling_terminate_handler() + 280
5   libobjc.A.dylib               	       0x18bb90474 _objc_terminate() + 172
6   libc++abi.dylib               	       0x18bf6bc2c std::__terminate(void (*)()) + 16
7   libc++abi.dylib               	       0x18bf6f394 __cxxabiv1::failed_throw(__cxxabiv1::__cxa_exception*) + 88
8   libc++abi.dylib               	       0x18bf6f33c __cxa_throw + 92
9   Starship                      	       0x100c5ed3c std::__1::__throw_bad_variant_access[abi:ne200100]() + 56
10  Starship                      	       0x100e94154 auto&& std::__1::__generic_get[abi:ne200100]<1ul, std::__1::variant<prism::Void, int, float, prism::MTDArray<bool>, prism::MTDArray<int>, prism::MTDArray<float>, prism::GeneratedRange, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>, prism::ForContext, unsigned long (*)(), prism::Opaque>&>(std::__1::variant<prism::Void, int, float, prism::MTDArray<bool>, prism::MTDArray<int>, prism::MTDArray<float>, prism::GeneratedRange, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>, prism::ForContext, unsigned long (*)(), prism::Opaque>&) + 36
11  Starship                      	       0x100e94124 std::__1::variant_alternative<1ul, std::__1::variant<prism::Void, int, float, prism::MTDArray<bool>, prism::MTDArray<int>, prism::MTDArray<float>, prism::GeneratedRange, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>, prism::ForContext, unsigned long (*)(), prism::Opaque>>::type& std::__1::get[abi:ne200100]<1ul, prism::Void, int, float, prism::MTDArray<bool>, prism::MTDArray<int>, prism::MTDArray<float>, prism::GeneratedRange, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>, prism::ForContext, unsigned long (*)(), prism::Opaque>(std::__1::variant<prism::Void, int, float, prism::MTDArray<bool>, prism::MTDArray<int>, prism::MTDArray<float>, prism::GeneratedRange, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>, prism::ForContext, unsigned long (*)(), prism::Opaque>&) + 24
12  Starship                      	       0x100e91358 int& std::__1::get[abi:ne200100]<int, prism::Void, int, float, prism::MTDArray<bool>, prism::MTDArray<int>, prism::MTDArray<float>, prism::GeneratedRange, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>, prism::ForContext, unsigned long (*)(), prism::Opaque>(std::__1::variant<prism::Void, int, float, prism::MTDArray<bool>, prism::MTDArray<int>, prism::MTDArray<float>, prism::GeneratedRange, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>, prism::ForContext, unsigned long (*)(), prism::Opaque>&) + 24
13  Starship                      	       0x100e92340 update_raw_floats(std::__1::variant<prism::Void, int, float, prism::MTDArray<bool>, prism::MTDArray<int>, prism::MTDArray<float>, prism::GeneratedRange, std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>, prism::ForContext, unsigned long (*)(), prism::Opaque>*) + 36
14  Starship                      	       0x10122f29c invoke + 148
15  Starship                      	       0x101204950 prism::Processor::evaluate(std::__1::shared_ptr<prism::ast::ASTNode> const&) + 12200
16  Starship                      	       0x10120b01c prism::Processor::evaluate_node(std::__1::shared_ptr<std::__1::vector<std::__1::shared_ptr<prism::Node>, std::__1::allocator<std::__1::shared_ptr<prism::Node>>>>&) + 312
17  Starship                      	       0x10120c968 prism::Processor::process() + 124
18  Starship                      	       0x100e936cc gfx_metal_build_shader(std::__1::basic_string<char, std::__1::char_traits<char>, std::__1::allocator<char>>&, unsigned long&, CCFeatures const&, bool) + 3488
19  Starship                      	       0x100e70904 Fast::GfxRenderingAPIMetal::CreateAndLoadNewShader(unsigned long long, unsigned int) + 276
20  Starship                      	       0x100eb1424 Fast::Interpreter::LookupOrCreateShaderProgram(unsigned long long, unsigned long long) + 132
21  Starship                      	       0x100eb9c00 Fast::Interpreter::GfxSpTri1(unsigned char, unsigned char, unsigned char, bool) + 3992
22  Starship                      	       0x100ebce3c Fast::Interpreter::GfxDrawRectangle(int, int, int, int) + 1164
23  Starship                      	       0x100ebd858 Fast::Interpreter::GfxDpFillRectangle(int, int, int, int) + 464
24  Starship                      	       0x100ec42c4 Fast::gfx_fill_wide_rect_handler_custom(Fast::F3DGfx**) + 240
25  Starship                      	       0x100ec5acc Fast::gfx_step() + 180
26  Starship                      	       0x100ec5810 Fast::Interpreter::Run(Gfx*, std::__1::unordered_map<MtxF*, MtxF, std::__1::hash<MtxF*>, std::__1::equal_to<MtxF*>, std::__1::allocator<std::__1::pair<MtxF* const, MtxF>>> const&) + 792
27  Starship                      	       0x100e6caac Fast::Fast3dWindow::DrawAndRunGraphicsCommands(Gfx*, std::__1::unordered_map<MtxF*, MtxF, std::__1::hash<MtxF*>, std::__1::equal_to<MtxF*>, std::__1::allocator<std::__1::pair<MtxF* const, MtxF>>> const&) + 260
28  Starship                      	       0x100bd4830 GameEngine::RunCommands(Gfx*, std::__1::vector<std::__1::unordered_map<MtxF*, MtxF, std::__1::hash<MtxF*>, std::__1::equal_to<MtxF*>, std::__1::allocator<std::__1::pair<MtxF* const, MtxF>>>, std::__1::allocator<std::__1::unordered_map<MtxF*, MtxF, std::__1::hash<MtxF*>, std::__1::equal_to<MtxF*>, std::__1::allocator<std::__1::pair<MtxF* const, MtxF>>>>> const&) + 348
29  Starship                      	       0x100bd4fac GameEngine::ProcessGfxCommands(Gfx*) + 788
30  Starship                      	       0x100c097c4 Graphics_PushFrame + 24
31  Starship                      	       0x100cc2cd8 Graphics_ThreadUpdate + 600
32  Starship                      	       0x100c097dc push_frame() + 12
33  Starship                      	       0x100c09848 main + 72
34  dyld                          	       0x18bbf9d54 start + 7184

it seems the game is crashing on libultraship update_raw_floats, specifically on these lines:

prism::ContextTypes* update_raw_floats(prism::ContextTypes* num) {
    MTL::VertexFormat format;
    int *size = std::get_if<int>(num);

    // I added this part for debugging, this condition is hit and the game crashes
    if(size == nullptr) {
      spdlog::info("we failed to get size from num: {}", num->index());
      exit(0);
    }

    switch (*size) {
        case 4:
            format = MTL::VertexFormatFloat4;
            break;
        case 3:
            format = MTL::VertexFormatFloat3;
            break;
        case 2:
            format = MTL::VertexFormatFloat2;
            break;
        case 1:
            format = MTL::VertexFormatFloat;
            break;
    }
    vertex_descriptor->attributes()->object(vertex_index)->setFormat(format);
    vertex_descriptor->attributes()->object(vertex_index)->setBufferIndex(0);
    vertex_descriptor->attributes()->object(vertex_index++)->setOffset(raw_numFloats * sizeof(float));
    raw_numFloats += *size;

    return nullptr;
}

from further looking into it I theorize that the prism-processor is somehow messing up after preprocessing the shaders and this is leading to an empty/invalid variant

however I have no idea what any of this means, just posting here in case it is useful

additionally, I tried building the prism-processor by itself and that seems to work - however I am not sure how to test it fully and 1 out of the 3 provided files in that repo crashes

[mkoterski]

I have forked a version for Intel Macs that I am currently testing. Happy for anyone willing to test it on other macs.
Starship MacAlfa

[akdev1l]

@mkoterski I am happy to try this but please enable issues on that repo so I could report my results over there and avoid polluting this issue

[niekvlessert]

With a little Codex I managed to build it. Seems to work fine, just segfaults when going full screen mode.

[mkoterski]

@mkoterski I am happy to try this but please enable issues on that repo so I could report my results over there and avoid polluting this issue

FYI: Had someone confirm it to be working on Apple Silicon.

[niekvlessert]

I use Imac M1. It didn't crash when I changed to fullscreen just now! Basically all Codex did was adding the target to CMakeLists.txt. Should be easy to update the action so it builds that one too. I have a DMG now if anyone wants it...