video_core: replace Fortune's Weave's DLSS with AMD FSR 3.1

Fire Emblem: Fortune's Weave's DLSS network can't run here, and the
temporal stand-in is only an approximation of it. FSR 3.1 takes the
same inputs DLSS does: the jittered colour, motion vectors, the jitter
and the depth, which the game's first DLSS pass reads through
cbuf0+0x298 (Z32, reversed, the colour's size). Kepler compute keeps it
next to the motion vectors.

Fsr3Upscaler ports the host side of the SDK's ffx_fsr3upscaler.cpp: the
eight passes (prepare inputs, luma and shading-change pyramids, shading
change, reactivity, luma instability, accumulate, RCAS), the internal
images and the constants, recorded through the scheduler with the
shaders from externals/FidelityFX-SDK. It takes HDR colour, render-size
motion vectors and reversed depth, computes its own exposure, and uses
the no-wave SPD variants on devices without compute quad operations.
The SDK's GLSL uses separate textures and samplers, so DescriptorBankInfo
gains sampled image and sampler counts (default 0). FSR's RCAS follows
the RCAS Sharpness slider. The result goes to FSR's own RGBA16F image
and is drawn into the game's output, which keeps its format.

TextureCacheRuntime::TemporalUpscale runs FSR 3.1 first. Without the
depth, at mismatched scales, on a device lacking FSR's formats or
features, or if its pipelines fail to build, the stand-in runs instead,
and the bilinear blit stays last. EDEN_NX2_DLSS_REPLACEMENT=standin
forces the stand-in. A gap of more than 250 ms between evaluations
restarts the history.

The BlitColor overload that takes a raw image view transitions its
source to SHADER_READ_ONLY_OPTIMAL but described it as GENERAL in the
descriptor (a validation error); it now describes the layout it uses.

Assisted-by: Claude Opus 5.5 (debugging, implementation and drafting)
Signed-off-by: Duck912 <350528965+Duck912@users.noreply.github.com>

1 files changed, 14 insertions, 4 deletions

diff --git a/src/video_core/engines/kepler_compute.cpp b/src/video_core/engines/kepler_compute.cpp
index 4732167369..2d35b603ae 100644
--- a/src/video_core/engines/kepler_compute.cpp
+++ b/src/video_core/engines/kepler_compute.cpp
@@ -591,8 +591,9 @@ void KeplerCompute::DescribeNativeResources(const NativeQmdScheduler::Descriptor
 std::atomic<u32> g_nx2_trace_draws_after_upscale{};
 
 // Fire Emblem: Fortune's Weave's first DLSS pass (rejected, before the network) has the input colour
-// at cbuf0+0x278, its motion vectors (R16G16 float, the input's size) at +0x290 and this frame's
-// camera jitter (x, y; input pixels, Halton bases 2 and 3 less 0.5) at +0x1b0.
+// at cbuf0+0x278, its motion vectors (R16G16 float, the input's size) at +0x290, its depth (Z32, the
+// input's size) at +0x298 and this frame's camera jitter (x, y; input pixels, Halton bases 2 and 3
+// less 0.5) at +0x1b0.
 void KeplerCompute::RememberDlssMotionVectors() {
     const auto& cbuf = launch_description.const_buffer_config[0];
     if (cbuf.size.Value() < 0x294) return;
@@ -610,6 +611,14 @@ void KeplerCompute::RememberDlssMotionVectors() {
         return;
     dlss_motion_vectors = *motion;
     dlss_motion_color = color->Address();
+    const auto depth = cbuf.size.Value() >= 0x29c ? image(0x298) : std::nullopt;
+    if (depth && depth->format == Texture::TextureFormat::Z32 && depth->Address() != 0 &&
+        depth->Width() == color->Width() && depth->Height() == color->Height() &&
+        depth->IsBlockLinear()) {
+        dlss_depth = *depth;
+    } else {
+        dlss_depth.reset();
+    }
     dlss_jitter = {std::bit_cast<f32>(memory_manager.Read<u32>(cbuf_address + 0x1b0)),
                    std::bit_cast<f32>(memory_manager.Read<u32>(cbuf_address + 0x1b4))};
 }
@@ -770,8 +779,9 @@ bool KeplerCompute::UpscaleInsteadOfNativeProgram() {
             return false;
         Fermi2D::Surface motion = surface(*dlss_motion_vectors);
         motion.format = RTFormat::R16G16_FLOAT;
-        return rasterizer->AccelerateTemporalUpscale(source, surface(*output), motion, dlss_jitter,
-                                                     {1.0f, 1.0f});
+        return rasterizer->AccelerateTemporalUpscale(source, surface(*output), motion,
+                                                     dlss_depth ? &*dlss_depth : nullptr,
+                                                     dlss_jitter, {1.0f, 1.0f});
     };
     const bool done = upscale_temporally() ||
                       rasterizer->AccelerateSurfaceCopy(source, surface(*output), config);