Michael Mestnik
Michael Mestnik
Along with #573 some patterns, useful for Layers, are applicable to device drivers. * An Iterator over vk::BaseInStructure * Picking members out of such an iterator. https://gitlab.com/cheako/vk-layer-cache/-/blob/ff04a089e3335ef1823ec5280781b9be14fcc0f3/src/handlers/types.rs
There are a few new strucks not included: https://github.com/KhronosGroup/Vulkan-LoaderAndValidationLayers/blob/25d5884746a2de7b51a8ef3ec88e1cd8066460e8/include/vulkan/vk_layer.h#L112 It becomes a pattern of descending chains of pnext looking for a particular strict type. Many tools used in implementing vulkan...
I'm trying to get_physical_device_features2() for PhysicalDeviceDescriptorIndexingFeatures. Edit: maybe VkPhysicalDeviceVulkan12Features? Same issue.
Looks like instead of None or something, functions that panic are being propped up. This came up when I didn't set my api version correctly when creating the instance.
What's the purpose of the marker(s) on builders, is it because of pointers? There are a number of builders that don't contain pointers, not even p_next. Is there any point...
wait_semaphore_count is shared by p_wait_semaphores and p_wait_dst_stage_mask, but there are two functions used to set these and they can be passed different counts. This silently leads to crashes that validation...
If Align::copy_from_slice() is passed a big enough slice and it takes the no padding route, it'll happily map too much memory and then write to it.
Use case: [patch.crates-io] ash = { git = 'https://github.com/MaikKlein/ash.git' } If the version in git is not ahead of creats-io then things get confusing. If you NEED to increment the...
> Sorry if this is already implemented, I just heard of luxtorpeda and am happy to hear I'm not the only one. > > **Is your feature request related to...
I'm looking to test my downlink capability on a multi-homed setup. There are several layers of NAT and I'd rather not, and in one instance can't, hardcode holes for this....