Fedora Account System
Red Hat Associate
Red Hat Customer
today freshly installed distro: Fedora-Cloud-Base-AmazonEC2.x86_64-Rawhide-20251124.0 # rpm -qa | grep vulkan vulkan-tools-1.4.321.0-5.fc44.x86_64 vulkan-validation-layers-1.4.321.0-4.fc44.x86_64 vulkan-loader-1.4.321.0-4.fc44.x86_64 mesa-vulkan-drivers-25.3.0-3.fc44.x86_64 # rpm -qa | grep mesa mesa-filesystem-25.3.0-3.fc44.x86_64 mesa-vulkan-drivers-25.3.0-3.fc44.x86_64 # uname -a Linux ip-172-31-27-138.us-east-2.compute.internal 6.18.0-0.rc6.251121gfd95357fd8c6.55.fc44.x86_64 #1 SMP PREEMPT_DYNAMIC Fri Nov 21 18:23:30 UTC 2025 x86_64 GNU/Linux Reproducible: Always Steps to Reproduce: 1.# vkcube --validate Selected WSI platform: xlib WARNING : GENERAL - Message Id Number: 0 | Message Id Name: Loader Message terminator_CreateInstance: Received return code -3 from call to vkCreateInstance in ICD /usr/lib64/libvulkan_dzn.so. Skipping this driver. Objects - 1 Object[0] - VK_OBJECT_TYPE_INSTANCE, Handle 0x5638816a2eb0 Segmentation fault (core dumped) vkcube --validate rpm -qf /usr/lib64/libvulkan_dzn.so mesa-vulkan-drivers-25.3.0-3.fc44.x86_64 Actual Results: Segmentation fault (core dumped) vkcube --validate Expected Results: no segfault
it also segfault with physical machine , with internal graphics: ❯ vulkaninfo | grep -E "GPU id|deviceName|vendorName" WARNING: [Loader Message] Code 0 : terminator_CreateInstance: Received return code -3 from call to vkCreateInstance in ICD /usr/lib64/libvulkan_dzn.so. Skipping this driver. GPU id = 0 (Intel(R) Iris(R) Xe Graphics (TGL GT2)) GPU id = 1 (llvmpipe (LLVM 21.1.5, 256 bits)) GPU id = 0 (Intel(R) Iris(R) Xe Graphics (TGL GT2)) GPU id = 1 (llvmpipe (LLVM 21.1.5, 256 bits)) GPU id = 0 (Intel(R) Iris(R) Xe Graphics (TGL GT2)) GPU id = 1 (llvmpipe (LLVM 21.1.5, 256 bits)) GPU id : 0 (Intel(R) Iris(R) Xe Graphics (TGL GT2)) [VK_KHR_xcb_surface, VK_KHR_xlib_surface]: GPU id : 1 (llvmpipe (LLVM 21.1.5, 256 bits)) [VK_KHR_xcb_surface, VK_KHR_xlib_surface]: GPU id : 0 (Intel(R) Iris(R) Xe Graphics (TGL GT2)) [VK_KHR_wayland_surface]: GPU id : 1 (llvmpipe (LLVM 21.1.5, 256 bits)) [VK_KHR_wayland_surface]: deviceName = Intel(R) Iris(R) Xe Graphics (TGL GT2) deviceName = llvmpipe (LLVM 21.1.5, 256 bits) ❯ echo $DISPLAY :0
❯ vkcube Selected WSI platform: wayland Selected GPU 0: Intel(R) Iris(R) Xe Graphics (TGL GT2), type: IntegratedGpu ^C ❯ vkcube --validate Selected WSI platform: wayland WARNING : GENERAL - Message Id Number: 0 | Message Id Name: Loader Message terminator_CreateInstance: Received return code -3 from call to vkCreateInstance in ICD /usr/lib64/libvulkan_dzn.so. Skipping this driver. Objects - 1 Object[0] - VK_OBJECT_TYPE_INSTANCE, Handle 0x55922f8f7420 [1] 67747 segmentation fault (core dumped) vkcube --validate
FYI, after the fix in mesa: https://bodhi.fedoraproject.org/updates/FEDORA-2025-2f4ba7cd17 it stop to segfault at my physical machine but its still segfault at cloud AWS ... [root@ip-172-31-29-21 ~]# dmidecode | grep -C1 -i amaz BIOS Information Vendor: Amazon EC2 Version: 1.0 -- System Information Manufacturer: Amazon EC2 Product Name: c6a.large <snip /> # vkcube --validate Selected WSI platform: xlib WARNING : GENERAL - Message Id Number: 0 | Message Id Name: Loader Message terminator_CreateInstance: Received return code -3 from call to vkCreateInstance in ICD /usr/lib64/libvulkan_dzn.so. Skipping this driver. Objects - 1 Object[0] - VK_OBJECT_TYPE_INSTANCE, Handle 0x560591384eb0 Segmentation fault (core dumped) vkcube --validate => It’s not really a priority bug, since it’s unlikely to be used in a cloud; still, segfault is segfault