GPU/HW: Add fallback path for no-texture-buffers

Yay, old Mali drivers without SSBOs...

src/core/gpu_hw.cpp

@@ -645,10 +645,15 @@ bool GPU_HW::CreateBuffers()
   GL_OBJECT_NAME(m_vram_read_texture, "VRAM Read Texture");
   GL_OBJECT_NAME(m_vram_readback_texture, "VRAM Readback Texture");
 
-  if (!(m_vram_upload_buffer =
-          g_gpu_device->CreateTextureBuffer(GPUTextureBuffer::Format::R16UI, GPUDevice::MIN_TEXEL_BUFFER_ELEMENTS)))
+  if (g_gpu_device->GetFeatures().supports_texture_buffers)
   {
-    return false;
+    if (!(m_vram_upload_buffer =
+            g_gpu_device->CreateTextureBuffer(GPUTextureBuffer::Format::R16UI, GPUDevice::MIN_TEXEL_BUFFER_ELEMENTS)))
+    {
+      return false;
+    }
+
+    GL_OBJECT_NAME(m_vram_upload_buffer, "VRAM Upload Buffer");
   }
 
   Log_InfoFmt("Created HW framebuffer of {}x{}", texture_width, texture_height);
@@ -999,9 +1004,10 @@ bool GPU_HW::CompilePipelines()
 
   // VRAM write
   {
+    const bool use_buffer = features.supports_texture_buffers;
     const bool use_ssbo = features.texture_buffers_emulated_with_ssbo;
-    std::unique_ptr<GPUShader> fs =
-      g_gpu_device->CreateShader(GPUShaderStage::Fragment, shadergen.GenerateVRAMWriteFragmentShader(use_ssbo));
+    std::unique_ptr<GPUShader> fs = g_gpu_device->CreateShader(
+      GPUShaderStage::Fragment, shadergen.GenerateVRAMWriteFragmentShader(use_buffer, use_ssbo));
     if (!fs)
       return false;
 
@@ -2438,11 +2444,28 @@ void GPU_HW::UpdateVRAM(u32 x, u32 y, u32 width, u32 height, const void* data, b
     }
   }
 
-  const u32 num_pixels = width * height;
-  void* map = m_vram_upload_buffer->Map(num_pixels);
-  const u32 map_index = m_vram_upload_buffer->GetCurrentPosition();
-  std::memcpy(map, data, num_pixels * sizeof(u16));
-  m_vram_upload_buffer->Unmap(num_pixels);
+  std::unique_ptr<GPUTexture> upload_texture;
+  u32 map_index;
+
+  if (!g_gpu_device->GetFeatures().supports_texture_buffers)
+  {
+    map_index = 0;
+    upload_texture = g_gpu_device->FetchTexture(width, height, 1, 1, 1, GPUTexture::Type::Texture,
+                                                GPUTexture::Format::R16U, data, width * sizeof(u16));
+    if (!upload_texture)
+    {
+      Log_ErrorFmt("Failed to get {}x{} upload texture. Things are gonna break.", width, height);
+      return;
+    }
+  }
+  else
+  {
+    const u32 num_pixels = width * height;
+    void* map = m_vram_upload_buffer->Map(num_pixels);
+    map_index = m_vram_upload_buffer->GetCurrentPosition();
+    std::memcpy(map, data, num_pixels * sizeof(u16));
+    m_vram_upload_buffer->Unmap(num_pixels);
+  }
 
   struct VRAMWriteUBOData
   {
@@ -2465,8 +2488,17 @@ void GPU_HW::UpdateVRAM(u32 x, u32 y, u32 width, u32 height, const void* data, b
   g_gpu_device->SetScissor(scaled_bounds.left, scaled_bounds.top, scaled_bounds.GetWidth(), scaled_bounds.GetHeight());
   g_gpu_device->SetPipeline(m_vram_write_pipelines[BoolToUInt8(check_mask && !m_pgxp_depth_buffer)].get());
   g_gpu_device->PushUniformBuffer(&uniforms, sizeof(uniforms));
-  g_gpu_device->SetTextureBuffer(0, m_vram_upload_buffer.get());
-  g_gpu_device->Draw(3, 0);
+  if (upload_texture)
+  {
+    g_gpu_device->SetTextureSampler(0, upload_texture.get(), g_gpu_device->GetNearestSampler());
+    g_gpu_device->Draw(3, 0);
+    g_gpu_device->RecycleTexture(std::move(upload_texture));
+  }
+  else
+  {
+    g_gpu_device->SetTextureBuffer(0, m_vram_upload_buffer.get());
+    g_gpu_device->Draw(3, 0);
+  }
 
   RestoreDeviceContext();
 }

src/core/gpu_hw_shadergen.cpp

@@ -1271,18 +1271,23 @@ uint SampleVRAM(uint2 coords)
   return ss.str();
 }
 
-std::string GPU_HW_ShaderGen::GenerateVRAMWriteFragmentShader(bool use_ssbo)
+std::string GPU_HW_ShaderGen::GenerateVRAMWriteFragmentShader(bool use_buffer, bool use_ssbo)
 {
   std::stringstream ss;
   WriteHeader(ss);
   WriteCommonFunctions(ss);
   DefineMacro(ss, "PGXP_DEPTH", m_pgxp_depth);
+  DefineMacro(ss, "USE_BUFFER", use_buffer);
   DeclareUniformBuffer(ss,
                        {"uint2 u_base_coords", "uint2 u_end_coords", "uint2 u_size", "uint u_buffer_base_offset",
                         "uint u_mask_or_bits", "float u_depth_value"},
                        true);
 
-  if (use_ssbo && m_glsl)
+  if (!use_buffer)
+  {
+    DeclareTexture(ss, "samp0", 0, false, true, true);
+  }
+  else if (use_ssbo && m_glsl)
   {
     ss << "layout(std430";
     if (IsVulkan())
@@ -1316,14 +1321,17 @@ std::string GPU_HW_ShaderGen::GenerateVRAMWriteFragmentShader(bool use_ssbo)
     discard;
   }
 
-
   // find offset from the start of the row/column
   uint2 offset;
   offset.x = (coords.x < u_base_coords.x) ? ((VRAM_SIZE.x / RESOLUTION_SCALE) - u_base_coords.x + coords.x) : (coords.x - u_base_coords.x);
   offset.y = (coords.y < u_base_coords.y) ? ((VRAM_SIZE.y / RESOLUTION_SCALE) - u_base_coords.y + coords.y) : (coords.y - u_base_coords.y);
 
+#if !USE_BUFFER
+  uint value = LOAD_TEXTURE(samp0, int2(offset), 0).x;
+#else
   uint buffer_offset = u_buffer_base_offset + (offset.y * u_size.x) + offset.x;
   uint value = GET_VALUE(buffer_offset) | u_mask_or_bits;
+#endif
   
   o_col0 = RGBA5551ToRGBA8(value);
 #if !PGXP_DEPTH

src/core/gpu_hw_shadergen.h

@@ -22,7 +22,7 @@ public:
   std::string GenerateWireframeGeometryShader();
   std::string GenerateWireframeFragmentShader();
   std::string GenerateVRAMReadFragmentShader();
-  std::string GenerateVRAMWriteFragmentShader(bool use_ssbo);
+  std::string GenerateVRAMWriteFragmentShader(bool use_buffer, bool use_ssbo);
   std::string GenerateVRAMCopyFragmentShader();
   std::string GenerateVRAMFillFragmentShader(bool wrapped, bool interlaced);
   std::string GenerateVRAMUpdateDepthFragmentShader();

src/util/opengl_device.cpp

@@ -492,10 +492,7 @@ bool OpenGLDevice::CheckFeatures(bool* buggy_pbo, FeatureMask disabled_features)
     }
     else
     {
-      Host::ReportErrorAsync(
-        TRANSLATE_SV("GPUDevice", "Error"),
-        TRANSLATE_SV("Error", "Both texture buffers and SSBOs are not supported, or are of inadequate size."));
-      return false;
+      Log_WarningPrint("Both texture buffers and SSBOs are not supported. Performance will suffer.");
     }
   }