vulkan

What is the correct way of using a Vulkan VkImage as a CUDA cuArray? I've been trying to follow some examples, however I get a CUDA_ERROR_INVALID_VALUE on a call to cuExternalMemoryGetMappedMipmappedArray() To provide the information in an ordered way. I'm using CUDA 10.1 Base code comes from https://github.com/SaschaWillems/Vulkan , in particular I'm using the 01 - Vulkan Gears demo, enriched with the saveScreenshot method 09 - Capturing screenshots Instead of saving the snapshot image to a file, I'll be sending the snapshot image into CUDA as a CUarray. I've enabled the following instance

In OpenGL you can linearize a depth value like so: float linearize_depth(float d,float zNear,float zFar) { float z_n = 2.0 * d - 1.0; return 2.0 * zNear * zFar / (zFar + zNear - z_n * (zFar - zNear)); } (Source: https://stackoverflow.com/a/6657284/10011415 ) However, Vulkan handles depth values somewhat differently ( https://matthewwellings.com/blog/the-new-vulkan-coordinate-system/ ). I don't quite understand the math behind it, what changes would I have to make to the function to linearize a depth value with Vulkan? The important difference between OpenGL and Vulkan here is that the

From the man page for vkEnumerateDeviceExtensionProperties , vkEnumerateDeviceExtensionProperties retrieves properties for extensions on a physical device whose handle is given in physicalDevice. To determine the extensions implemented by a layer set pLayerName to point to the layer’s name and any returned extensions are implemented by that layer. Setting pLayerName to NULL will return the available non-layer extensions. pPropertyCount must be set to the size of the VkExtensionProperties array pointed to by pProperties. The pProperties should point to an array of VkExtensionProperties to be

Does Forcing "High" DPM Performance Help Out Your AMDGPU Performance? Written by Michael Larabel in Radeon on 7 June 2018 at 06:04 AM EDT. 14 Comments A premium patron recently asked about testing the open-source Radeon driver performance when testing the forced "high" dynamic power management state rather than the default "auto" mode. Here are some benchmarks. This is about forcing /sys/class/drm/card0/device/power_dpm_force_performance_level to high rather than auto, to ensure the GPU is bound to its highest performance state rather than dynamically changing performance states based upon

I am trying to do two simple subpasses, where the second one has a dependency on the first one. //subpass 1 VkAttachmentReference colorReferences = { 0, VK_IMAGE_LAYOUT_COLOR_ATTACHMENT_OPTIMAL }; VkSubpassDescription subpass1 = {}; subpass1.pipelineBindPoint = VK_PIPELINE_BIND_POINT_GRAPHICS; subpass1.pColorAttachments = &colorReferences; subpass1.colorAttachmentCount = 1; //subpass 2 VkAttachmentReference inputRefernce = { 0, VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL }; VkSubpassDescription subpass2 = {}; subpass2.pipelineBindPoint = VK_PIPELINE_BIND_POINT_GRAPHICS; subpass2

At Vulkan specs 1.0.9 (pg. 180), we have the following: typedef struct VkMemoryHeap { VkDeviceSize size; VkMemoryHeapFlags flags; } VkMemoryHeap; and this description: • size is the total memory size in bytes in the heap. • flags is a bitmask of attribute flags for the heap. The bits specified in flags are: typedef enum VkMemoryHeapFlagBits { VK_MEMORY_HEAP_DEVICE_LOCAL_BIT = 0x00000001, } VkMemoryHeapFlagBits; But when I query VkPhysicalDeviceMemoryProperties I've got flags with zero values. My code matches the output from Vulkan SDK vkjson_info.exe tool, which output a JSON file with