From a8669ade5cb3add8b9ce08e2c3bd96e93190bca8 Mon Sep 17 00:00:00 2001
From: jsmall-nvidia <jsmall@nvidia.com>
Date: Fri, 17 Jan 2020 09:15:06 -0500
Subject: Slang -> CUDA kernel runs correctly in test infrastructure (#1167)

* First pass at BindLocation.

* Added BindSet::init - for initializing with two input constant buffers. Needs better name, and perhaps should be another class.

* Fix handling of constant buffer stripping.
Improved initialization.

* Trying to generalize BindLocation a little more.
Split out CPULikeBindRoot.

* More work to make BindLocation et al work with non uniform bindings.

* Added parsing to a location.

* WIP: Trying to get CPU working with BindLocation.

* Describe problem of knowing the type of the reference point in the binding table.

* More ideas on getBindings fix.

* Remove BindSet as member of BindLocation.

* Added BindLocation::Invalid

* Made BindLocation able to be key in hash

* Use BindLocation for bindings on BindingSet.

* Added cuda and nvrtc categories to test infrastructure.
Disabled CUDA synthetic tests by default.
Fixed such that all tests now produce something in BindLocation style.

* Use m_userIndex instead of m_userData on Resource.
Move the binding setup out of cpu-compute-util (as no longer CPU specific)

* Removed CPUBinding - used BindLocation/BindSet instead.
Fixed some bugs around indexOf around uniform indirection.

* Renamed BindSet::Resource -> BindSet::Value.

* Document BindLocation.

* Fixes for Clang/GCC
Improve invariant requirement handling when constructing from BindPoints.

* WIP: First attempt to run CUDA kernel.

* Fix some issues around doing CUDA kernel launch.

* Fix issues around use of cudaMemCpy .

* Better cuda runtime error checking mechanism.

* Fixed bug in passing parameters to cuda kernel launch.
Simplified initialisation of context.

* WIP: Fix CUDA runtime issues.

* Add explicit CUDA synchronize so failures don't appear on implicit ones.

* Fix problem emitting non shared variable on CUDA.

* Fix some typos in CUDA layout.
Use just a pointer for now for CUDA StucturedBuffer.

* Arg order for CUDA launch was wrong.

* First compute kernel runs on CUDA.
---
 tools/render-test/shader-input-layout.cpp | 58 +++++++++++++++++++++++++++++++
 1 file changed, 58 insertions(+)

(limited to 'tools/render-test/shader-input-layout.cpp')

diff --git a/tools/render-test/shader-input-layout.cpp b/tools/render-test/shader-input-layout.cpp
index 5ae35b90d..ee4f5fc2c 100644
--- a/tools/render-test/shader-input-layout.cpp
+++ b/tools/render-test/shader-input-layout.cpp
@@ -1,3 +1,6 @@
+// Stop warnings from Visual Studio
+#define _CRT_SECURE_NO_WARNINGS 1
+
 #include "shader-input-layout.h"
 #include "core/slang-token-reader.h"
 
@@ -676,6 +679,61 @@ namespace renderer_test
         return SLANG_OK;
     }
 
+    /* static */void ShaderInputLayout::getValueBuffers(const Slang::List<ShaderInputLayoutEntry>& entries, const BindSet& bindSet, List<BindSet::Value*>& outBuffers)
+    {
+        outBuffers.setCount(entries.getCount());
+
+        for (Index i = 0; i< outBuffers.getCount(); ++i)
+        {
+            outBuffers[i] = nullptr;
+        }
+
+        const auto& values = bindSet.getValues();
+        for (BindSet::Value* value : values)
+        {
+            if (value->m_userIndex >= 0)
+            {
+                outBuffers[value->m_userIndex] = value;
+            }
+        }
+    }
+
+
+    /* static */SlangResult ShaderInputLayout::writeBindings(const ShaderInputLayout& layout, const List<BindSet::Value*>& buffers, const String& fileName)
+    {
+        FILE * f = fopen(fileName.getBuffer(), "wb");
+        if (!f)
+        {
+            return SLANG_FAIL;
+        }
+
+        const auto& entries = layout.entries;
+
+        for (int i = 0; i < entries.getCount(); ++i)
+        {
+            const auto& entry = entries[i];
+            if (entry.isOutput)
+            {
+                BindSet::Value* buffer = buffers[i];
+
+                unsigned int* ptr = (unsigned int*)buffer->m_data;
+
+                const int size = int(entry.bufferData.getCount());
+                // Must be the same size or less than allocated buffer
+                SLANG_ASSERT(size * sizeof(unsigned int) <= buffer->m_sizeInBytes);
+
+                for (int i = 0; i < size; ++i)
+                {
+                    unsigned int v = ptr[i];
+
+                    fprintf(f, "%X\n", v);
+                }
+            }
+        }
+        fclose(f);
+        return SLANG_OK;
+    }
+
     void generateTextureData(TextureData& output, const InputTextureDesc& desc)
     {
         switch (desc.format)
-- 
cgit v1.2.3