From e510a287deb25f2542a68bf21382f2557740d70c Mon Sep 17 00:00:00 2001
From: jsmall-nvidia <jsmall@nvidia.com>
Date: Thu, 6 May 2021 12:45:00 -0400
Subject: Half texture support (#1836)

* #include an absolute path didn't work - because paths were taken to always be relative.

* Split out StringEscapeUtil.

* Added StringEscapeUtil.

* Fix typo in unix quoting type.

* Small comment improvements.

* Try to fix linux linking issue.

* Fix typo.

* Attempt to fix linux link issue.

* Update VS proj even though nothing really changed.

* Fix another typo issue.

* Fix for windows issue.
Fixed bug.

* Make separate Utils for escaping.

* Fix typo.

* Split out into StringEscapeHandler.

* Windows shell does handle removing quotes (so remove code to remove them).

* Handle unescaping if not initiating using the shell.

* Slight improvement around shell like decoding.

* Simplify command extraction.

* Add shared-library category type.

* Fix bug in command extraction.

* Typo in transcendental category.

* Enable unit-test on in smoke test category.

* Make parsing failing output as a failing test.

* Fixes for transcendental tests. Disable tests that do not work.

* Changed category parsing.

* Removed the TestResult parameter from _gatherTestsForFile.
Made testsList only output.

* Remove testing if all tests were disabled.

* Make args of CommandLine always unescaped.

* Add category.

* Don't need escaping on unix/linux.

* Remove some no longer used functions.

* Add requireSMVersion to CUDAExtensionTracker.

* half-calc.slang now works for CUDA.

* bit-cast-16-bit works on CUDA.

* WIP handling of CUDA vector<half> types.

* Half swizzle CUDA.

* Half vector test.

* Fix swizzle half bug.

* Fix compilation issue with narrowing to Index.

* Add unary ops.

* Add some vector scalar maths ops.

* Add half vector conversions for CUDA.

* Fix erroneous comment.

* Support for half comparisons.

* First pass test for half compare.

* Fix bug in CUDA specialized emit control.
Updated tests to have pre and post inc/dec.

* Removed unneeded parts of the cuda prelude.

* Half structured buffer works on CUDA.

* Added name lookup for Gfx::Format

* Support half texture type in test system.

* Test for half reading on CUDA.

* Add half formats to Vk and D3D utils.

* Fix getAt for CUDA - where there might not be a .x member in a vector.
---
 tools/gfx/render.cpp | 75 +++++++++++++++++++++++++++++++++++++++++++---------
 1 file changed, 63 insertions(+), 12 deletions(-)

(limited to 'tools/gfx/render.cpp')

diff --git a/tools/gfx/render.cpp b/tools/gfx/render.cpp
index d14ce904a..cbf1c6d26 100644
--- a/tools/gfx/render.cpp
+++ b/tools/gfx/render.cpp
@@ -19,24 +19,70 @@ static bool debugLayerEnabled = false;
 
 /* !!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Global Renderer Functions !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! */
 
-static const uint8_t s_formatSize[] = {
-    0, // Unknown,
+#define GFX_FORMAT_SIZE(name, size) uint8_t(size),
 
-    uint8_t(sizeof(float) * 4), // RGBA_Float32,
-    uint8_t(sizeof(float) * 3), // RGB_Float32,
-    uint8_t(sizeof(float) * 2), // RG_Float32,
-    uint8_t(sizeof(float) * 1), // R_Float32,
+static const uint8_t s_formatSize[] =
+{
+    GFX_FORMAT(GFX_FORMAT_SIZE)
+};
+
+static bool _checkFormat()
+{
+    Index value = 0;
+    Index count = 0;
+
+    // Check the values are in the same order
+#define GFX_FORMAT_CHECK(name, size) count += Index(Index(Format::name) == value++);
+    GFX_FORMAT(GFX_FORMAT_CHECK)
+
+    const bool r = (count == Index(Format::CountOf));
+    SLANG_ASSERT(r);
+    return r;
+}
+
+// We don't make static because we will get a warning that it's unused
+static const bool _checkFormatResult = _checkFormat();
+
+struct FormatInfoMap
+{
+    FormatInfoMap()
+    {
+        // Set all to nothing initially
+        for (auto& info : m_infos)
+        {
+            info.channelCount = 0;
+            info.channelType = SLANG_SCALAR_TYPE_NONE;
+        }
+
+        set(Format::RGBA_Float16, SLANG_SCALAR_TYPE_FLOAT16, 4);
+        set(Format::RG_Float16, SLANG_SCALAR_TYPE_FLOAT16, 2);
+        set(Format::R_Float16, SLANG_SCALAR_TYPE_FLOAT16, 1);
+
+        set(Format::RGBA_Float32, SLANG_SCALAR_TYPE_FLOAT32, 4);
+        set(Format::RGB_Float32, SLANG_SCALAR_TYPE_FLOAT32, 3);
+        set(Format::RG_Float32, SLANG_SCALAR_TYPE_FLOAT32, 2);
+        set(Format::R_Float32, SLANG_SCALAR_TYPE_FLOAT32, 1);
+
+        set(Format::R_UInt16, SLANG_SCALAR_TYPE_UINT16, 1);
+        set(Format::R_UInt32, SLANG_SCALAR_TYPE_UINT32, 1);
+
+        set(Format::D_Float32, SLANG_SCALAR_TYPE_FLOAT32, 1);
+    }
 
-    uint8_t(sizeof(uint32_t)), // RGBA_Unorm_UInt8,
-    uint8_t(sizeof(uint32_t)), // BGRA_Unorm_UInt8,
+    void set(Format format, SlangScalarType type, Index channelCount)
+    {
+        FormatInfo& info = m_infos[Index(format)];
+        info.channelCount = uint8_t(channelCount);
+        info.channelType = uint8_t(type);
+    }
 
-    uint8_t(sizeof(uint16_t)), // R_UInt16,
-    uint8_t(sizeof(uint32_t)), // R_UInt32,
+    const FormatInfo& get(Format format) const { return m_infos[Index(format)]; }
 
-    uint8_t(sizeof(float)), // D_Float32,
-    uint8_t(sizeof(uint32_t)), // D_Unorm24_S8,
+    FormatInfo m_infos[Index(Format::CountOf)];
 };
 
+static const FormatInfoMap s_formatInfoMap;
+
 static void _compileTimeAsserts()
 {
     SLANG_COMPILE_TIME_ASSERT(SLANG_COUNT_OF(s_formatSize) == int(Format::CountOf));
@@ -49,6 +95,11 @@ extern "C"
         return s_formatSize[int(format)];
     }
 
+    SLANG_GFX_API FormatInfo gfxGetFormatInfo(Format format)
+    {
+        return s_formatInfoMap.get(format);
+    }
+
     SlangResult _createDevice(const IDevice::Desc* desc, IDevice** outDevice)
     {
         switch (desc->deviceType)
-- 
cgit v1.2.3