shader/texture: Join separate image and sampler pairs offline

Games using D3D idioms can join images and samplers when a shader
executes, instead of baking them into a combined sampler image. This is
also possible on Vulkan.

One approach to this solution would be to use separate samplers on
Vulkan and leave this unimplemented on OpenGL, but we can't do this
because there's no consistent way of determining which constant buffer
holds a sampler and which one an image. We could in theory find the
first bit and if it's in the TIC area, it's an image; but this falls
apart when an image or sampler handle use an index of zero.

The used approach is to track for a LOP.OR operation (this is done at an
IR level, not at an ISA level), track again the constant buffers used as
source and store this pair. Then, outside of shader execution, join
the sample and image pair with a bitwise or operation.

This approach won't work on games that truly use separate samplers in a
meaningful way. For example, pooling textures in a 2D array and
determining at runtime what sampler to use.

This invalidates OpenGL's disk shader cache :)

- Used mostly by D3D ports to Switch

1 files changed, 20 insertions, 0 deletions

diff --git a/src/video_core/shader/registry.cpp b/src/video_core/shader/registry.cpp
index af70b3f35..cdf274e54 100644
--- a/src/video_core/shader/registry.cpp
+++ b/src/video_core/shader/registry.cpp
@@ -93,6 +93,26 @@ std::optional<SamplerDescriptor> Registry::ObtainBoundSampler(u32 offset) {
     return value;
 }
 
+std::optional<Tegra::Engines::SamplerDescriptor> Registry::ObtainSeparateSampler(
+    std::pair<u32, u32> buffers, std::pair<u32, u32> offsets) {
+    SeparateSamplerKey key;
+    key.buffers = buffers;
+    key.offsets = offsets;
+    const auto iter = separate_samplers.find(key);
+    if (iter != separate_samplers.end()) {
+        return iter->second;
+    }
+    if (!engine) {
+        return std::nullopt;
+    }
+
+    const u32 handle_1 = engine->AccessConstBuffer32(stage, key.buffers.first, key.offsets.first);
+    const u32 handle_2 = engine->AccessConstBuffer32(stage, key.buffers.second, key.offsets.second);
+    const SamplerDescriptor value = engine->AccessSampler(handle_1 | handle_2);
+    separate_samplers.emplace(key, value);
+    return value;
+}
+
 std::optional<Tegra::Engines::SamplerDescriptor> Registry::ObtainBindlessSampler(u32 buffer,
                                                                                  u32 offset) {
     const std::pair key = {buffer, offset};