Description of problem: Fresh install od fedora 28 beta, launching vkmark simple fails. Version-Release number of selected component: vkmark-2017.08-0.3.20180123git68b6f23.fc28 Additional info: reporter: libreport-2.9.4 backtrace_rating: 4 cmdline: vkmark crash_function: vk::Instance::enumeratePhysicalDevices<std::allocator<vk::PhysicalDevice> > executable: /usr/bin/vkmark journald_cursor: s=c7b26c21c6c344c892de1f9a779dda66;i=52d1;b=36365bcc8ebb45fa8832207d3fc1b86c;m=c603ce39;t=56abbccecfbff;x=d5cd88c8aa741919 kernel: 4.16.3-301.fc28.x86_64 rootdir: / runlevel: N 5 type: CCpp uid: 1000 Truncated backtrace: Thread no. 1 (3 frames) #4 vk::Instance::enumeratePhysicalDevices<std::allocator<vk::PhysicalDevice> > at /usr/include/c++/8/bits/char_traits.h:350 #5 VulkanState::choose_physical_device(VulkanWSI&) at ../src/vulkan_state.cpp:69 #6 VulkanState::VulkanState(VulkanWSI&) at ../src/vulkan_state.cpp:34
Created attachment 1427026 [details] File: backtrace
Created attachment 1427027 [details] File: cgroup
Created attachment 1427028 [details] File: core_backtrace
Created attachment 1427029 [details] File: cpuinfo
Created attachment 1427030 [details] File: dso_list
Created attachment 1427031 [details] File: environ
Created attachment 1427032 [details] File: limits
Created attachment 1427033 [details] File: maps
Created attachment 1427034 [details] File: mountinfo
Created attachment 1427035 [details] File: open_fds
Created attachment 1427036 [details] File: proc_pid_status
The assertion is on the vulkan side and probably specific to your hardware config + fedora drivers. I'll move to vulkan for now where people might have a better idea.
What graphics hardware do you have? as vulkan will run the first driver icd even if it's headless.
(In reply to leigh scott from comment #13) > What graphics hardware do you have? as vulkan will run the first driver icd > even if it's headless. 01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Oland XT [Radeon HD 8670 / R7 250/350] (prog-if 00 [VGA controller]) Subsystem: Dell Radeon R7 250 Flags: bus master, fast devsel, latency 0, IRQ 30 Memory at e0000000 (64-bit, prefetchable) [size=256M] Memory at f7e00000 (64-bit, non-prefetchable) [size=256K] I/O ports at e000 [size=256] Expansion ROM at 000c0000 [disabled] [size=128K] Capabilities: <access denied> Kernel driver in use: radeon Kernel modules: radeon, amdgpu
I can't reproduce the issue here, maybe the open source driver is at fault. $ vkmark -p immediate ======================================================= vkmark 2017.08 ======================================================= Vendor ID: 0x10DE Device ID: 0x1C82 Device Name: GeForce GTX 1050 Ti Driver Version: 1661239296 ======================================================= [vertex] device-local=true: FPS: 5396 FrameTime: 0.185 ms [vertex] device-local=false: FPS: 2883 FrameTime: 0.347 ms [texture] anisotropy=0: FPS: 5164 FrameTime: 0.194 ms [texture] anisotropy=16: FPS: 5117 FrameTime: 0.195 ms [shading] shading=gouraud: FPS: 5137 FrameTime: 0.195 ms [shading] shading=blinn-phong-inf: FPS: 5114 FrameTime: 0.196 ms [shading] shading=phong: FPS: 5013 FrameTime: 0.199 ms [shading] shading=cel: FPS: 5046 FrameTime: 0.198 ms [effect2d] kernel=edge: FPS: 5340 FrameTime: 0.187 ms [effect2d] kernel=blur: FPS: 4050 FrameTime: 0.247 ms [desktop] <default>: FPS: 4322 FrameTime: 0.231 ms [cube] <default>: FPS: 6806 FrameTime: 0.147 ms [clear] <default>: FPS: 6776 FrameTime: 0.148 ms ======================================================= vkmark Score: 5089 ======================================================= $ rpm -q vulkan vkmark vulkan-1.1.73.0-1.fc28.x86_64 vkmark-2017.08-0.3.20180123git68b6f23.fc28.x86_64
I can't debug further as I only have nvidia hardware with nvidia.com driver and no easy way to test under wayland.
Can you post dmesg, I'm expecting you don't have drivers in place for this, we need to use the amdgpu driver on oland hw to support vulkan.
Created attachment 1427577 [details] Dmesg output
I become aware that the issue originated by an hardware problem. Infacts the PCI express slot where the card was inserted has a damaged connection. I have changed slot and the card works flawlessly now. So, I close this bug.