magnum

Commit Graph

Author	SHA1	Message	Date
Vladimír Vondruš	84bdb8076d	Vk: de-inline Queue members and test them. For consistency with other classes and to prepare for future extension. Interesting that I didn't even create a test file for this class yet.	5 years ago
Vladimír Vondruš	82d874040e	Vk: avoid redefining CORRADE_NO_ASSERT.	5 years ago
Vladimír Vondruš	9ad26feda0	Vk: store size in the Framebuffer and reuse it in RenderPassCreateInfo. Since this is the most common use case, it makes sense.	5 years ago
Vladimír Vondruš	4a49b3a770	Vk: implement render pass begin / end / next commands.	5 years ago
Vladimír Vondruš	495c1d9a7f	Vk: doc++	5 years ago
Vladimír Vondruš	349dc77322	Vk: don't use VK_STRUCTURE_TYPE_APPLICATION_INFO for sType in tests. The problem is that, of all structure type enums, this one is 0, and thus doesn't check for accidentally zeroed-out memory.	5 years ago
Vladimír Vondruš	9dbe377ac4	Vk: properly test Image::format().	5 years ago
Vladimír Vondruš	fd46515a21	Vk: account for cursed drivers doing cursed things with extensions.	5 years ago
Vladimír Vondruš	7ee30c55e3	Vk: command buffer begin and end.	5 years ago
Vladimír Vondruš	32cc6ed69d	Vk: fence status, reset and wait. I was hesitating a bit with the <chrono> include but I can't think of anything better right now. It's "only" 4k lines on C++11 (and I bet most of it is <type_traits>), so should be fine.	5 years ago
Vladimír Vondruš	2bcb5d6d9e	Vk: more useful MAGNUM_VK_INTERNAL_ASSERT_SUCCESS_OR(). Another breaking change here, sorry -- I don't expect anyone to be seriously using these ... yet :)	5 years ago
Vladimír Vondruš	b977a14214	Vk: unneeded include. Probably a leftover from before the moment when I realized I can't use a Reference due to the NoCreate constructor.	5 years ago
Vladimír Vondruš	5099b8c997	Vk: add a Fence wrapper.	5 years ago
Vladimír Vondruš	5c554a039e	Vk: doc++	5 years ago
Vladimír Vondruš	46a30e2ff1	Vk: doc++ Was changing this throughout the RenderPass implementation but this got forgotten.	5 years ago
Vladimír Vondruš	71af6a034c	Vk: implement a Framebuffer wrapper. EZ, for once.	5 years ago
Vladimír Vondruš	4a93c092b0	Vk: image view wrapper. This was done quite a long time ago but then I got stuck on device features for cubemaps. Sigh.	5 years ago
Vladimír Vondruš	0684894324	Vk: whoops, forgot these. What's the Developer Guidelines for if I don't use them.	5 years ago
Vladimír Vondruš	88d2acc66e	Vk: expand the Result enum with values for all extensions we have. So people can use the MAGNUM_VK_INTERNAL_ASSERT_SUCCESS() macro on their raw Vulkan calls and it reports a name back instead of a number.	5 years ago
Vladimír Vondruš	7bdf0feba1	Vk: expose the new RT extensions. Enabling them in the Vulkan header, function pointer loader, extension lists and feature queries / setup: - KHR_acceleration_structure - KHR_deferred_host_operations - KHR_pipeline_library - KHR_ray_tracing_pipeline - KHR_ray_query	5 years ago
Vladimír Vondruš	3f1a3eb1c2	Vk: a bit more robust feature enablement. Because the drivers don't seem to check that features are actually supported for anything beyond VkPhysicalDeviceFeatures2, we have to do that instead. Additionally, we also check if a feature is enabled but the extension that needs it is not (which isn't checked by drivers either).	5 years ago
Vladimír Vondruš	aad588054a	Vk: implement Device feature enablement. This was way more pain that initially expected, especially in regards to preserving externally-specified pNext chains without writing to them in any way.	5 years ago
Vladimír Vondruš	a4382c6305	Vk: more utilites for heavyweight structure chain processing. Gah, these device features are hell.	5 years ago
Vladimír Vondruš	6f867c2ab3	Vk: doc++	5 years ago
Vladimír Vondruš	51cf303585	Vk: list device features in magnum-vk-info. It's reusing an internal file to make them display nicely, I hope that doesn't make it too ugly / too sad that library users can't do the same thing in their code.	5 years ago
Vladimír Vondruš	1ee2b40c53	Vk: fetching of supported device features. This is what I needed BigEnumSet for -- good thing I didn't even try to have 128-bit enums because I'm now at 110 values and it's still far from complete. Next step is enabling those features when creating a device, which should hopefully be a lot less code, reusing most of what was here.	5 years ago
Vladimír Vondruš	ff916e1934	Vk: internal helper for checking both extension and its core version. Not exactly sure about the semantics yet, so keeping it private. In particular, if this should be the default behavior for isExtensionEnabled() (would make sense probably?), or for isExtensionSupported(), and what should be the API that separates the version from the extension so we can decide which function pointer to fetch.	5 years ago
Vladimír Vondruš	613a0d8cf6	Vk: write an internal helper for connecting Vulkan structures.	5 years ago
Vladimír Vondruš	75d2e20c97	Vk: move constructors should be implicit, so test that property.	5 years ago
Vladimír Vondruš	d06e64d770	Vk: rename DeviceVkTest test case for clarity. Was copied from InstanceVkTest where it's for both layers and extensions. Not here.	5 years ago
Vladimír Vondruš	ccab98920f	Vk: DeviceCreateInfo state is allocated always. Probably some leftover from the early days.	5 years ago
Vladimír Vondruš	71b2fca3d8	Vk: rename private Device member for clarity.	5 years ago
Vladimír Vondruš	6acd7b72b1	Vk: doc++	5 years ago
Vladimír Vondruš	4f789ce92b	Vk: okay, I need these weird extension definitions now.	5 years ago
Vladimír Vondruš	4f22a6a3fe	Test: forgot to set IDE folders for some of these.	5 years ago
Vladimír Vondruš	cf60ea3d0d	Vk: doc-- Yeah, these were rather redundant, so...	5 years ago
Vladimír Vondruš	909dd781df	Vk: put CreateInfo structures to dedicated headers. After I implemented the render pass wrapper, seeing how the RenderPassCreateInfo structure and its dependencies were HUGE compared to the actual tiny and lean RenderPass, I felt uneasy dragging their definition along to every place where a RenderPass gets used. It's not as bad with the others, but as new extensions are implemented I expect that to get the same. This change makes it easier for me to accept that Image.h / Buffer.h depends on Memory.h. There isn't a real measurable difference when building Magnum itself (50 ms out of 7 seconds for the Vk library alone), but that's because most of the code (and tests) needs the CreateInfo structures anyway.	5 years ago
Vladimír Vondruš	98a69fc568	Vk: rethink "creation" docs. Although the APIs don't look like that right now, in many cases creation of a particular Vulkan object isn't all that's there for it. So change the section names from a generic "Usage" to "Creation".	5 years ago
Vladimír Vondruš	299a17f95b	Vk: clarify what's the extra Memory::wrap() parameter for, add assert. Similar thing will be done for all upcoming wrap() calls as well.	5 years ago
Vladimír Vondruš	8ea64e814a	Vk: DeviceCreateInfo wasn't properly movable. Interestingly I didn't run into this until now. Doing what I just learned while implementing the RenderPass.	5 years ago
Vladimír Vondruš	4933583f7f	Vk: Device(..., const DeviceCreateInfo&) wasn't covered. Interestingly enough, apparently I was always constructing it in-place in the Device constructor. Which leads me to the next issue...	5 years ago
Vladimír Vondruš	f6532e1a5c	Vk: implicitly enable VK_KHR_create_renderpass2.	5 years ago
Vladimír Vondruš	eb42342d16	Vk: minor cleanup.	5 years ago
Vladimír Vondruš	c32f75ddf3	Vk: implement a RenderPass wrapper. Quite a big chunk of work, further expanded due to how VK_KHR_create_renderpass2 is designed -- basically, due to the tightly-packed nested structures that got replaced with their "version 2", we can no longer just extract the previous structure for backwards compatibility, but instead have to deep-copy everything to a newly allocated memory. Thanks to the the new ArrayTuple structure and a few design iterations I managed to kick the backwards-compatiblity code into just a single allocation, while still keeping it possible for the "version 2" code path to be fully allocation-free (if one passes a completely filled VkRenderPassCreateInfo2 structure there).	5 years ago
Vladimír Vondruš	de1d520caf	Vk: no need to prefix with Vk:: in tests.	5 years ago
abgita	f9b07ac290	Typo: simultaenous -> simultaneous.	5 years ago
Vladimír Vondruš	1579a89e47	Vk: expose VK_IMAGE_LAYOUT, finally. I was pretending to ignore it until now, but it's just easier to expose it. All newly gained knowledge is now put in the enum docs.	5 years ago
Vladimír Vondruš	271f8b6fec	Vk: tighten rules for extension-dependent code paths. To ensure a non-success VkResult is never missed.	5 years ago
Vladimír Vondruš	5a838c9220	Vk: this can be written in a shorter way.	5 years ago
Vladimír Vondruš	4b7b74715d	doc: forgot to list this Vulkan API. ... and Doxygen is shitty once again, totally ignoring the doc block on a forward declaration. FFS.	5 years ago

1 2 3 4

166 Commits (84bdb8076dfa095d224a1b04da4ea3348bc62a87)