2
1
Fork 0
mirror of https://github.com/yuzu-emu/yuzu.git synced 2024-07-04 23:31:19 +01:00

vk_stream_buffer: Fix initializing Vulkan with NVIDIA on Linux

The previous fix only partially solved the issue, as only certain GPUs that needed 9 or less MiB subtracted would work (i.e. GTX 980 Ti, GT 730). This takes from DXVK's example to divide `heap_size` by 2 to determine `allocable_size`. Additionally tested on my Quadro K4200, which previously required setting it to 12 to boot.
This commit is contained in:
lat9nq 2020-09-25 17:42:59 -04:00
parent 4d4afc1502
commit ca26fd0f42

View file

@ -120,7 +120,8 @@ void VKStreamBuffer::CreateBuffers(VkBufferUsageFlags usage) {
// Substract from the preferred heap size some bytes to avoid getting out of memory. // Substract from the preferred heap size some bytes to avoid getting out of memory.
const VkDeviceSize heap_size = memory_properties.memoryHeaps[preferred_heap].size; const VkDeviceSize heap_size = memory_properties.memoryHeaps[preferred_heap].size;
const VkDeviceSize allocable_size = heap_size - 9 * 1024 * 1024; // As per DXVK's example, using `heap_size / 2`
const VkDeviceSize allocable_size = heap_size / 2;
buffer = device.GetLogical().CreateBuffer({ buffer = device.GetLogical().CreateBuffer({
.sType = VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO, .sType = VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO,
.pNext = nullptr, .pNext = nullptr,