WIP Geometry shaders #1815
WIP Geometry shaders #1815
Conversation
| @@ -304,7 +304,7 @@ if [ ! "$SPIRV_CROSS_ROOT" = "" ]; then | |||
| rm -rf ${REPO_NAME} | |||
| ln -sfn ${SPIRV_CROSS_ROOT} ${REPO_NAME} | |||
| else | |||
| REPO_URL="https://github.com/KhronosGroup/${REPO_NAME}.git" | |||
| REPO_URL="https://github.com/js6i/${REPO_NAME}.git" | |||
There was a problem hiding this comment.
For the time being He needs to pull from his fork of SPIRV-CROSS a lot of GS work is done there too.
|
Just asking if someone tryied this, does this solve Rocket League nameplates missing bug? Or possibly in future ... |
Yes among many other things, like unlocking the entire UE4 game library and much more. |
Thats great to hear. I can't imagine how much work this takes, I'm not really into graphical programming, but I hope this can be resolved this year. I will certainly try this on my Air when I have time. Thanks! |
Did you actually tried to lunch any UE4 game with this branch? |
I did some tests, nothing works yet, this is a work in progress. |
…explicit-lod-workaround Allow to disable Explicit LOD Workaround for Apple silicon.
- Fix memory leak when waiting on timeline semaphores. - For correctness, set VkPhysicalDeviceLimits::lineWidthGranularity to 1. - Update MoltenVK to version 1.2.4. - Update Whats_New.md document with recent changes.
…hore-mem-leak Several simple maintenance fixes.
- Add MVK_XCODE_14_3 macro to compile for iOS/tvOS 16.4 and above. - Add support for BC compression on iOS/tvOS 16.4 and above where supported. - Consolidate MVKPixelFormats::modifyMTLFormatCapabilities(mtlDev) and centralize querying MTLDevice format methods for all platforms.
of whitespace in MVKPixelFormats::modifyMTLFormatCapabilities().
…-compression Support BC compression on iOS/tvOS, where available.
…ated MTLBuffer. - MVKDevice track VkBuffers marked with VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT. - Add SPIRVToMSLConversionResultInfo::usesPhysicalStorageBufferAddressesCapability to detect and track shaders that use PhysicalStorageBufferAddresses capability, and track such shader stages within pipeline. - MVKResourcesCommandEncoderState encode usage of VkBuffers marked with VK_BUFFER_USAGE_SHADER_DEVICE_ADDRESS_BIT when pipeline uses PhysicalStorageBufferAddresses capability. - Rename MVKResourcesCommandEncoderState::encodeArgumentBufferResourceUsage() to encodeResourceUsage(). - MVKDevice move some functions to public scope and remove friend classes. - MVKDeviceMemory ensure _vkMemAllocFlags is always initialized (unrelated). - Rename MVKFoundation template method contains() to mvkContains() (unrelated).
Ensure shaders using PhysicalStorageBufferAddresses encode the associated MTLBuffer.
- MVKRenderSubpass add separate getDepthFormat() & getStencilFormat(), and isDepthAttachmentUsed() & isStencilAttachmentUsed() and use instead of testing pixel format for depth and stencil components. - Add MVKRenderingAttachmentIterator class to consistently iterate, and take actions, on the attachments in VkRenderingInfo to create synthetic MVKRenderPass and extract image views and clear colors. - Remove mvkCreateRenderPass() and mvkCreateFramebuffer() in favor of additional constructors, and remove mvkGetDepthStencilFormat() in favor of retrieving formats for separate depth and stencil attachments. - MVKRenderpass constructors reorganize order of adding attachments and subpasses, and connecting the two. - Renmame MVKRenderPassAttachment to MVKAttachmentDescription. - MVKPipeline reorganize member variables to minimize gaps in content and remove unnecessary _isRasterizingDepthStencil member var (unrelated).
|
Any updates on this PR? :) |
- MVKPipeline validate depth & stencil formats before setting frag shader depth and stencil builtins, and before setting formats in MTLRenderPipelineDescriptor. - MVKCmdClearAttachments always set depth/stencil format if subpass includes a depth or stencil attachment, even if they are not being cleared. - MVKRenderingAttachmentIterator add synthetic attachment if depth or stencil attachment is not provided, but image format supports both depth and stencil.
…ess-older-os-fix Disable pipeline cache compression prior to macOS 10.15 and iOS/tvOS 13.0
- Also rename kMVKCachedColorAttachmentCount to kMVKMaxColorAttachmentCount and kMVKCachedViewportScissorCount to kMVKMaxViewportScissorCount and (unrelated)
…ate-depth-stencil Support separate depth and stencil attachments during dynamic rendering
The VK_MVK_moltenvk extension has never been brought inside Vulkan, and
the functions have never been supported by the Vulkan Loader and Layers.
Most of the functionality has long been replaced by the official
VK_metal_objects extension.
- Remove VK_MVK_moltenvk as an advertised extension.
- Refactor vk_mvk_moltenvk.h header file into separate headers files:
- mvk_config.h - Valid public config functions
- mvk_private_api.h - Valid development debugging functions used with care
- mvk_deprecated_api.h - Formally deprecated functions.
- Retain skeleton vk_mvk_moltenvk.h header file for legacy compatibility only.
- Update documentation and header comments to explain changes.
- MVKInstance reorganize list of entry points for visual clarity (unrelated). - Fixes to deprecation documentation (unrelated). - mvkStringsAreEqual() add small performance optimization (unrelated).
…VK_moltenvk-extension Deprecate the obsolete and non-standard VK_MVK_moltenvk extension.
- Set Metal depth store action to MTLStoreActionDontCare when depth attachment exists, but depth component has been swizzled away.
…err-depth-swizzle Avoid Metal validation warning when depth component swizzled away
- Reinstate VK_MVK_moltenvk extension, but log warning message when it is enabled. - Add vkGetMoltenVKConfiguration2MVK() and vkSetMoltenVKConfiguration2MVK() to set config without passing a dummy VkInstance, and deprecate vkGetMoltenVKConfigurationMVK() and vkSetMoltenVKConfigurationMVK().
…extn Reinstate VK_MVK_moltenvk extension and add simplified config functions.
When compiling tessellation vertex shaders, MVKGraphicsPipeline pass array of MVKMTLFunction instead of MTLFunctions to retain MTLFunctions for duration of processing.
…-regression Fix bad access regression to prematurely-deallocated MTLFunction.
This allows rebuilding without an internet connection and avoids random build failures on flaky connections.
Don't fetch a submodule if the commit is already known
…devices. The [MTLDevice sampleTimestamps:gpuTimestamp:] function turns out to be synchronized with other queue activities, and can block GPU execution if it is called between MTLCommandBuffer submissions. On non-Apple-Silicon devices, it was called before and after every vkQueueSubmit() submission, to track the correlation between GPU and CPU timestamps, and was delaying the start of GPU work on the next submission (on Apple Silicon, both CPU & GPU timestamps are specified in nanoseconds, and the call was bypassed). Move timestamp correlation from vkQueueSubmit() to vkGetPhysicalDeviceProperties(), where it is used to update VkPhysicalDeviceLimits::timestampPeriod on non-Apple-Silicon devices. Delegate MVKPhysicalDevice::getProperties(VkPhysicalDeviceProperties2*) to MVKPhysicalDevice::getProperties(VkPhysicalDeviceProperties*), plus minimize wasted effort if pNext is empty (unrelated). Move the declaration of several MVKPhysicalDevice member structs to potentially reduce member spacing (unrelated).
Xcode simulator always requires 256B buffer alignment, even when running on Apple Silicon. Previously, it was assumed that Apple Silicon would use it's native 16B buffer alignment.
…c-delay Fix sync delay between calls to vkQueueSubmit() on non-Apple-Silicon devices.
…lign Ensure Xcode simulator always uses 256B buffer alignment.
- In MoltenVK Xcode projects, set iOS & tvOS deployment targets to 12.0, to avoid warnings while building MoltenVK. - Add DYLD_LIBRARY_PATH to runcts script, to ensure Vulkan and MoltenVK libraries are found during CTS runs. - Update Whats_New.md and MoltenVK_Runtime_UserGuide.md documents.
Update dependency libraries to match Vulkan SDK 1.3.261.
- Guard against Intel returning zero values for CPU & GPU timestamps. - Apply lowpass filter on timestampPeriod updates, to avoid wild temporary changes, particularly at startup before GPU has been really exercised.
…stampPeriod Fix VkPhysicalDeviceLimits::timestampPeriod calculations on Intel GPU.
In the rare case where vertex attribute buffers are bound to MVKCommandEncoder, are not used by first pipeline, but are used by a subsequent pipeline, and no other bindings are changed, the MVKResourcesCommandEncoderState will not appear to be dirty to the second pipeline, and the buffer will not be bound to Metal. When reverting a binding to dirty if it is not used by a pipeline, also revert the enclosing MVKResourcesCommandEncoderState to dirty state. Update MoltenVK to version 1.2.6 (unrelated).
…x-attr-buff Fix rare case where vertex attribute buffers are not bound to Metal.
js6i
force-pushed
the
geometry-shaders
branch
from
23f70f8 to
8cbaaa6
Compare
|
I built libMoltenVK.dylib with @js6i's latest changes and I get the error > xcodebuild -version
Xcode 15.0
Build version 15A240d
> sw_vers
ProductName: macOS
ProductVersion: 14.0
BuildVersion: 23A339I also replaced MACOSX_DEPLOYMENT_TARGET diff --git a/ExternalDependencies.xcodeproj/project.pbxproj b/ExternalDependencies.xcodeproj/project.pbxproj
index f92a13b8..6496f6ed 100644
--- a/ExternalDependencies.xcodeproj/project.pbxproj
+++ b/ExternalDependencies.xcodeproj/project.pbxproj
@@ -5407,7 +5407,7 @@
GCC_WARN_UNUSED_VARIABLE = NO;
IPHONEOS_DEPLOYMENT_TARGET = 9.0;
MACH_O_TYPE = staticlib;
- MACOSX_DEPLOYMENT_TARGET = 10.11;
+ MACOSX_DEPLOYMENT_TARGET = 10.14;
SKIP_INSTALL = YES;
TVOS_DEPLOYMENT_TARGET = 9.0;
};
@@ -5456,7 +5456,7 @@
GCC_WARN_UNUSED_VARIABLE = NO;
IPHONEOS_DEPLOYMENT_TARGET = 9.0;
MACH_O_TYPE = staticlib;
- MACOSX_DEPLOYMENT_TARGET = 10.11;
+ MACOSX_DEPLOYMENT_TARGET = 10.14;
SKIP_INSTALL = YES;
TVOS_DEPLOYMENT_TARGET = 9.0;
VALIDATE_PRODUCT = YES;
diff --git a/MoltenVK/MoltenVK.xcodeproj/project.pbxproj b/MoltenVK/MoltenVK.xcodeproj/project.pbxproj
index 650a8219..b50158d6 100644
--- a/MoltenVK/MoltenVK.xcodeproj/project.pbxproj
+++ b/MoltenVK/MoltenVK.xcodeproj/project.pbxproj
@@ -1594,7 +1594,7 @@
);
IPHONEOS_DEPLOYMENT_TARGET = 9.0;
MACH_O_TYPE = staticlib;
- MACOSX_DEPLOYMENT_TARGET = 10.11;
+ MACOSX_DEPLOYMENT_TARGET = 10.14;
MTL_ENABLE_DEBUG_INFO = YES;
MVK_SKIP_DYLIB = "";
"MVK_SKIP_DYLIB[sdk=appletvsimulator*]" = YES;
@@ -1664,7 +1664,7 @@
);
IPHONEOS_DEPLOYMENT_TARGET = 9.0;
MACH_O_TYPE = staticlib;
- MACOSX_DEPLOYMENT_TARGET = 10.11;
+ MACOSX_DEPLOYMENT_TARGET = 10.14;
MTL_ENABLE_DEBUG_INFO = NO;
MVK_SKIP_DYLIB = "";
"MVK_SKIP_DYLIB[sdk=appletvsimulator*]" = YES;
diff --git a/MoltenVKShaderConverter/MoltenVKShaderConverter.xcodeproj/project.pbxproj b/MoltenVKShaderConverter/MoltenVKShaderConverter.xcodeproj/project.pbxproj
index 924a8987..9ddbf11b 100644
--- a/MoltenVKShaderConverter/MoltenVKShaderConverter.xcodeproj/project.pbxproj
+++ b/MoltenVKShaderConverter/MoltenVKShaderConverter.xcodeproj/project.pbxproj
@@ -632,7 +632,7 @@
);
IPHONEOS_DEPLOYMENT_TARGET = 9.0;
MACH_O_TYPE = staticlib;
- MACOSX_DEPLOYMENT_TARGET = 10.11;
+ MACOSX_DEPLOYMENT_TARGET = 10.14;
PRODUCT_NAME = MoltenVKShaderConverter;
SKIP_INSTALL = YES;
TVOS_DEPLOYMENT_TARGET = 9.0;
@@ -691,7 +691,7 @@
);
IPHONEOS_DEPLOYMENT_TARGET = 9.0;
MACH_O_TYPE = staticlib;
- MACOSX_DEPLOYMENT_TARGET = 10.11;
+ MACOSX_DEPLOYMENT_TARGET = 10.14;
PRODUCT_NAME = MoltenVKShaderConverter;
SKIP_INSTALL = YES;
TVOS_DEPLOYMENT_TARGET = 9.0;Build of the latest release version works |
It's probably missing shaderCullDistance and, if you're running wined3d, pipelineStatisticsQuery features, didn't you apply a hack adding those on top of release to make it work? |
|
I'm using dxvk. I don't think I can to apply any hacks, I'm a regular user who just follows this MR :D. I looked at the game launch logs with version 1.2.3, which I built myself with hacks to properly display shaders instead of black textures and my game runs fine And used your changes + merged the main branch I tried to compare the logs, but it's not obvious to me what the difference is and what went wrong |
|
I can help with the debug if needed |
|
Any News on this pr as if I try to build it I get this error when trying to fetch the dependency "SPIRV-Cross" fatal: reference is not a tree: 7a1e3e7586b9b10a6286cad912940b7a501e93ea |
That's because you don't have access to this repo |
no seen any branch with this id, and there is no latest code related to geometry-shaders branch(https://github.com/js6i/SPIRV-Cross/tree/geometry-shaders). 
The latest code seems to be missing here.How can I get this branch? I want to try its code. |
|
So Alyssa Rosenzweig seems to have added this feature to her Linux driver in the mean time. Is that enough guidance for someone to make this work? See https://rosenzweig.io/blog/aaa-gaming-on-m1.html |
Works in some cases.