internal void
PushRect(rectangle RenderRegion)
{
    render_entry *Entry = Entries + EntryCount;
    Entry->RenderRegion = RenderRegion;
    __atomic_add_fetch(&EntryCount, 1, __ATOMIC_ACQ_REL);
    SDL_SemPost(Semaphore);
}

internal void
AVX2_RenderLayer(transform_info TransformInfo, pixel_buffer *Buffer, rectangle RenderRegion);
internal void
Fallback_RenderLayer(transform_info TransformInfo, pixel_buffer *Buffer, rectangle RenderRegion);

internal bool32
CheckQueue(render_queue RenderInfo, uint16 Index)
{
    bool32 Result = 0;
    uint32 OriginalEntry = NextEntryToDo;
    if (NextEntryToDo < EntryCount)
        {
            if (__atomic_compare_exchange_n(&NextEntryToDo, &OriginalEntry, NextEntryToDo + 1, true, __ATOMIC_RELEASE, __ATOMIC_ACQUIRE)) {
                render_entry *Entry = Entries + OriginalEntry;
                Assert(Entry->RenderRegion.Max.x != 0);
                for (int16 i = 0; i < RenderInfo.State->NumberOfLayersToRender; i++) {
                    int16 Idx = RenderInfo.State->LayersToRender[i];
#if ARM
                    Fallback_RenderLayer(RenderInfo.File->Layer[Idx]->TransformInfo, RenderInfo.CompBuffer, Entry->RenderRegion);
                    // RenderLayerNeon(RenderInfo.File->Layer[Idx], RenderInfo.CompBuffer, Entry->RenderRegion);
#else
                    // printf("(RENDERING) Thread %i, region X%i Y%i\n", Index, Entry->RenderRegion.Min.x/240, Entry->RenderRegion.Min.y/135);
                    if (AVXEnabled)
                        AVX2_RenderLayer(RenderInfo.File->Layer[Idx]->TransformInfo, RenderInfo.CompBuffer, Entry->RenderRegion);
                    else
                        Fallback_RenderLayer(RenderInfo.File->Layer[Idx]->TransformInfo, RenderInfo.CompBuffer, Entry->RenderRegion);
#endif
                }

                // printf("(FINISHED) Thread %i, region X%i Y%i\n", Index, Entry->RenderRegion.Min.x/240, Entry->RenderRegion.Min.y/135);
                __atomic_add_fetch(&CompletedJobs, 1, __ATOMIC_ACQ_REL);
                Result = 1;
            }
        }
    return Result;
}

internal int
TestThread(void *ptr)
{
    thread_info *ThreadInfo = (thread_info *)ptr;
    uint16 Index = ThreadInfo->Index;
    for(;;)
    {
        if (!CheckQueue(*ThreadInfo->RenderInfo, Index))
        {
            SDL_SemWait(Semaphore);
        }
    }
}