From ece0c1095d29d940796568d7ce20fd5f414c4066 Mon Sep 17 00:00:00 2001
From: Liam <byteslice@airmail.cc>
Date: Wed, 16 Nov 2022 22:28:58 -0500
Subject: [PATCH 1/2] maxwell3d: HLE multi-layer clear macro
---
src/video_core/engines/maxwell_3d.h | 1 +
src/video_core/macro/macro_hle.cpp | 22 +++++++++++++++++++++-
2 files changed, 22 insertions(+), 1 deletion(-)
diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h
index 34b0853886..f9cdfbc0b9 100644
--- a/src/video_core/engines/maxwell_3d.h
+++ b/src/video_core/engines/maxwell_3d.h
@@ -1129,6 +1129,7 @@ public:
Tegra::RenderTargetFormat format;
TileMode tile_mode;
union {
+ u32 depth_volume;
BitField<0, 16, u32> depth;
BitField<16, 1, u32> volume;
};
diff --git a/src/video_core/macro/macro_hle.cpp b/src/video_core/macro/macro_hle.cpp
index f896591bfe..6d35387d92 100644
--- a/src/video_core/macro/macro_hle.cpp
+++ b/src/video_core/macro/macro_hle.cpp
@@ -126,11 +126,31 @@ void HLE_3F5E74B9C9A50164(Engines::Maxwell3D& maxwell3d, const std::vector<u32>&
}
}
-constexpr std::array<std::pair<u64, HLEFunction>, 4> hle_funcs{{
+// Multi-layer Clear
+void HLE_EAD26C3E2109B06B(Engines::Maxwell3D& maxwell3d, const std::vector<u32>& parameters) {
+ ASSERT(parameters.size() == 1);
+
+ Engines::Maxwell3D::Regs::ClearSurface clear_params{parameters[0]};
+
+ const u32 rt_index = clear_params.RT;
+ const u32 num_layers = maxwell3d.regs.rt[rt_index].depth_volume;
+
+ for (u32 i = 0; i < num_layers; i++) {
+ // 0x674 = regs.clear_surface
+ maxwell3d.CallMethod(0x674, clear_params.raw, true);
+ clear_params.layer.Assign(clear_params.layer + 1);
+
+ // FIXME: remove this when amdvlk can clear multiple layers without crashing
+ break;
+ }
+}
+
+constexpr std::array<std::pair<u64, HLEFunction>, 5> hle_funcs{{
{0x771BB18C62444DA0, &HLE_771BB18C62444DA0},
{0x0D61FC9FAAC9FCAD, &HLE_0D61FC9FAAC9FCAD},
{0x0217920100488FF7, &HLE_0217920100488FF7},
{0x3F5E74B9C9A50164, &HLE_3F5E74B9C9A50164},
+ {0xEAD26C3E2109B06B, &HLE_EAD26C3E2109B06B},
}};
class HLEMacroImpl final : public CachedMacro {
From 4c42655a2defddcd8faacaaa4e0bf109693b5a9c Mon Sep 17 00:00:00 2001
From: Liam <byteslice@airmail.cc>
Date: Wed, 16 Nov 2022 22:41:40 -0500
Subject: [PATCH 2/2] maxwell3d: full HLE for multi-layer clears
---
src/video_core/engines/maxwell_3d.cpp | 6 +++---
src/video_core/engines/maxwell_3d.h | 7 +++----
src/video_core/macro/macro_hle.cpp | 16 +++++-----------
src/video_core/rasterizer_interface.h | 2 +-
src/video_core/renderer_opengl/gl_rasterizer.cpp | 2 +-
src/video_core/renderer_opengl/gl_rasterizer.h | 2 +-
src/video_core/renderer_vulkan/vk_rasterizer.cpp | 4 ++--
src/video_core/renderer_vulkan/vk_rasterizer.h | 2 +-
8 files changed, 17 insertions(+), 24 deletions(-)
diff --git a/src/video_core/engines/maxwell_3d.cpp b/src/video_core/engines/maxwell_3d.cpp
index d502d181cb..5bb1427c17 100644
--- a/src/video_core/engines/maxwell_3d.cpp
+++ b/src/video_core/engines/maxwell_3d.cpp
@@ -232,7 +232,7 @@ void Maxwell3D::ProcessMethodCall(u32 method, u32 argument, u32 nonshadow_argume
use_topology_override = true;
return;
case MAXWELL3D_REG_INDEX(clear_surface):
- return ProcessClearBuffers();
+ return ProcessClearBuffers(1);
case MAXWELL3D_REG_INDEX(report_semaphore.query):
return ProcessQueryGet();
case MAXWELL3D_REG_INDEX(render_enable.mode):
@@ -596,8 +596,8 @@ u32 Maxwell3D::GetRegisterValue(u32 method) const {
return regs.reg_array[method];
}
-void Maxwell3D::ProcessClearBuffers() {
- rasterizer->Clear();
+void Maxwell3D::ProcessClearBuffers(u32 layer_count) {
+ rasterizer->Clear(layer_count);
}
void Maxwell3D::ProcessDraw(u32 instance_count) {
diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h
index f9cdfbc0b9..c3099f9a6b 100644
--- a/src/video_core/engines/maxwell_3d.h
+++ b/src/video_core/engines/maxwell_3d.h
@@ -1129,7 +1129,6 @@ public:
Tegra::RenderTargetFormat format;
TileMode tile_mode;
union {
- u32 depth_volume;
BitField<0, 16, u32> depth;
BitField<16, 1, u32> volume;
};
@@ -3087,6 +3086,9 @@ public:
std::vector<u8> inline_index_draw_indexes;
+ /// Handles a write to the CLEAR_BUFFERS register.
+ void ProcessClearBuffers(u32 layer_count);
+
private:
void InitializeRegisterDefaults();
@@ -3121,9 +3123,6 @@ private:
/// Handles firmware blob 4
void ProcessFirmwareCall4();
- /// Handles a write to the CLEAR_BUFFERS register.
- void ProcessClearBuffers();
-
/// Handles a write to the QUERY_GET register.
void ProcessQueryGet();
diff --git a/src/video_core/macro/macro_hle.cpp b/src/video_core/macro/macro_hle.cpp
index 6d35387d92..0f3262edb7 100644
--- a/src/video_core/macro/macro_hle.cpp
+++ b/src/video_core/macro/macro_hle.cpp
@@ -130,19 +130,13 @@ void HLE_3F5E74B9C9A50164(Engines::Maxwell3D& maxwell3d, const std::vector<u32>&
void HLE_EAD26C3E2109B06B(Engines::Maxwell3D& maxwell3d, const std::vector<u32>& parameters) {
ASSERT(parameters.size() == 1);
- Engines::Maxwell3D::Regs::ClearSurface clear_params{parameters[0]};
-
+ const Engines::Maxwell3D::Regs::ClearSurface clear_params{parameters[0]};
const u32 rt_index = clear_params.RT;
- const u32 num_layers = maxwell3d.regs.rt[rt_index].depth_volume;
+ const u32 num_layers = maxwell3d.regs.rt[rt_index].depth;
+ ASSERT(clear_params.layer == 0);
- for (u32 i = 0; i < num_layers; i++) {
- // 0x674 = regs.clear_surface
- maxwell3d.CallMethod(0x674, clear_params.raw, true);
- clear_params.layer.Assign(clear_params.layer + 1);
-
- // FIXME: remove this when amdvlk can clear multiple layers without crashing
- break;
- }
+ maxwell3d.regs.clear_surface.raw = clear_params.raw;
+ maxwell3d.ProcessClearBuffers(num_layers);
}
constexpr std::array<std::pair<u64, HLEFunction>, 5> hle_funcs{{
diff --git a/src/video_core/rasterizer_interface.h b/src/video_core/rasterizer_interface.h
index 1cbfef0902..cfd872a40e 100644
--- a/src/video_core/rasterizer_interface.h
+++ b/src/video_core/rasterizer_interface.h
@@ -43,7 +43,7 @@ public:
virtual void Draw(bool is_indexed, u32 instance_count) = 0;
/// Clear the current framebuffer
- virtual void Clear() = 0;
+ virtual void Clear(u32 layer_count) = 0;
/// Dispatches a compute shader invocation
virtual void DispatchCompute() = 0;
diff --git a/src/video_core/renderer_opengl/gl_rasterizer.cpp b/src/video_core/renderer_opengl/gl_rasterizer.cpp
index d05a5f60be..115a5e0104 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer.cpp
+++ b/src/video_core/renderer_opengl/gl_rasterizer.cpp
@@ -136,7 +136,7 @@ void RasterizerOpenGL::LoadDiskResources(u64 title_id, std::stop_token stop_load
shader_cache.LoadDiskResources(title_id, stop_loading, callback);
}
-void RasterizerOpenGL::Clear() {
+void RasterizerOpenGL::Clear(u32 layer_count) {
MICROPROFILE_SCOPE(OpenGL_Clears);
if (!maxwell3d->ShouldExecute()) {
return;
diff --git a/src/video_core/renderer_opengl/gl_rasterizer.h b/src/video_core/renderer_opengl/gl_rasterizer.h
index 793e0d608d..449a14f12b 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer.h
+++ b/src/video_core/renderer_opengl/gl_rasterizer.h
@@ -69,7 +69,7 @@ public:
~RasterizerOpenGL() override;
void Draw(bool is_indexed, u32 instance_count) override;
- void Clear() override;
+ void Clear(u32 layer_count) override;
void DispatchCompute() override;
void ResetCounter(VideoCore::QueryType type) override;
void Query(GPUVAddr gpu_addr, VideoCore::QueryType type, std::optional<u64> timestamp) override;
diff --git a/src/video_core/renderer_vulkan/vk_rasterizer.cpp b/src/video_core/renderer_vulkan/vk_rasterizer.cpp
index f69c0c50fa..67b88621ab 100644
--- a/src/video_core/renderer_vulkan/vk_rasterizer.cpp
+++ b/src/video_core/renderer_vulkan/vk_rasterizer.cpp
@@ -213,7 +213,7 @@ void RasterizerVulkan::Draw(bool is_indexed, u32 instance_count) {
EndTransformFeedback();
}
-void RasterizerVulkan::Clear() {
+void RasterizerVulkan::Clear(u32 layer_count) {
MICROPROFILE_SCOPE(Vulkan_Clearing);
if (!maxwell3d->ShouldExecute()) {
@@ -256,7 +256,7 @@ void RasterizerVulkan::Clear() {
.rect = regs.clear_control.use_scissor ? GetScissorState(regs, 0, up_scale, down_shift)
: default_scissor,
.baseArrayLayer = regs.clear_surface.layer,
- .layerCount = 1,
+ .layerCount = layer_count,
};
if (clear_rect.rect.extent.width == 0 || clear_rect.rect.extent.height == 0) {
return;
diff --git a/src/video_core/renderer_vulkan/vk_rasterizer.h b/src/video_core/renderer_vulkan/vk_rasterizer.h
index b0bc306f5a..70f36d58a7 100644
--- a/src/video_core/renderer_vulkan/vk_rasterizer.h
+++ b/src/video_core/renderer_vulkan/vk_rasterizer.h
@@ -65,7 +65,7 @@ public:
~RasterizerVulkan() override;
void Draw(bool is_indexed, u32 instance_count) override;
- void Clear() override;
+ void Clear(u32 layer_count) override;
void DispatchCompute() override;
void ResetCounter(VideoCore::QueryType type) override;
void Query(GPUVAddr gpu_addr, VideoCore::QueryType type, std::optional<u64> timestamp) override;