X-Git-Url: https://notaz.gp2x.de/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=plugins%2Fgpu_neon%2Fpsx_gpu%2Fpsx_gpu_4x.c;h=942b3d30ff266fbab7d84f1e7b7191bb47a7fe8d;hb=9088aca124430b38f473c931492b333cd998dfda;hp=19c4a9ef42fc241d3a758275a69d00f0b87a00dc;hpb=59d15d23d97d4347d8046057013f8979db0914f0;p=pcsx_rearmed.git

diff --git a/plugins/gpu_neon/psx_gpu/psx_gpu_4x.c b/plugins/gpu_neon/psx_gpu/psx_gpu_4x.c
index 19c4a9ef..942b3d30 100644
--- a/plugins/gpu_neon/psx_gpu/psx_gpu_4x.c
+++ b/plugins/gpu_neon/psx_gpu/psx_gpu_4x.c
@@ -1,4 +1,22 @@
-#ifndef NEON_BUILD
+#define select_enhancement_buf_ptr(psx_gpu, x) \
+  ((psx_gpu)->enhancement_buf_ptr + \
+   ((psx_gpu)->enhancement_buf_by_x16[(x) / 16] << 20))
+
+#if !defined(NEON_BUILD) || defined(SIMD_BUILD)
+
+#ifndef zip_4x32b
+
+#define vector_cast(vec_to, source) source
+
+#define zip_4x32b(dest, source_a, source_b) {                                  \
+  u32 _i; for(_i = 0; _i < 4; _i++) {                                          \
+    (dest).e[_i * 2 + 0] = (source_a).e[_i];                                   \
+    (dest).e[_i * 2 + 1] = (source_b).e[_i];                                   \
+  }                                                                            \
+}
+
+#endif
+
 void setup_sprite_16bpp_4x(psx_gpu_struct *psx_gpu, s32 x, s32 y, s32 u,
  s32 v, s32 width, s32 height, u32 color)
 {
@@ -52,7 +70,8 @@ void setup_sprite_16bpp_4x(psx_gpu_struct *psx_gpu, s32 x, s32 y, s32 u,
       texture_block_ptr =
        texture_page_ptr + (texture_offset_base & texture_mask);
 
-      load_128b(texels, texture_block_ptr);
+      //load_128b(texels, texture_block_ptr);
+      texels = *(vec_8x16u *)texture_block_ptr;
       
       zip_4x32b(vector_cast(vec_4x32u, texels_wide), texels.low, texels.low);
       block->texels = texels_wide;
@@ -113,7 +132,8 @@ void setup_sprite_16bpp_4x(psx_gpu_struct *psx_gpu, s32 x, s32 y, s32 u,
 
       texture_block_ptr = texture_page_ptr + (texture_offset & texture_mask);
       
-      load_128b(texels, texture_block_ptr);
+      //load_128b(texels, texture_block_ptr);
+      texels = *(vec_8x16u *)texture_block_ptr;
 
       zip_4x32b(vector_cast(vec_4x32u, texels_wide), texels.low, texels.low);
       block->texels = texels_wide;
@@ -143,7 +163,8 @@ void setup_sprite_16bpp_4x(psx_gpu_struct *psx_gpu, s32 x, s32 y, s32 u,
       while(blocks_remaining)
       {
         texture_block_ptr = texture_page_ptr + (texture_offset & texture_mask);
-        load_128b(texels, texture_block_ptr);
+        //load_128b(texels, texture_block_ptr);
+        texels = *(vec_8x16u *)texture_block_ptr;
 
         zip_4x32b(vector_cast(vec_4x32u, texels_wide), texels.low, texels.low);
         block->texels = texels_wide;
@@ -174,7 +195,8 @@ void setup_sprite_16bpp_4x(psx_gpu_struct *psx_gpu, s32 x, s32 y, s32 u,
       }
 
       texture_block_ptr = texture_page_ptr + (texture_offset & texture_mask);
-      load_128b(texels, texture_block_ptr);
+      //load_128b(texels, texture_block_ptr);
+      texels = *(vec_8x16u *)texture_block_ptr;
       
       zip_4x32b(vector_cast(vec_4x32u, texels_wide), texels.low, texels.low);
       block->texels = texels_wide;
@@ -287,7 +309,7 @@ static void setup_sprite_untextured_4x(psx_gpu_struct *psx_gpu, s32 x, s32 y,
   render_sprite_blocks_switch_block_texture_mode_4x(4bpp),                     \
   render_sprite_blocks_switch_block_texture_mode_4x(8bpp),                     \
   render_sprite_blocks_switch_block_texture_mode_4x(16bpp),                    \
-  render_sprite_blocks_switch_block_texture_mode_4x(4bpp)                      \
+  render_sprite_blocks_switch_block_texture_mode_4x(16bpp)                     \
 
 
 render_block_handler_struct render_sprite_block_handlers_4x[] =
@@ -331,6 +353,8 @@ void render_sprite_4x(psx_gpu_struct *psx_gpu, s32 x, s32 y, u32 u, u32 v,
   if((width <= 0) || (height <= 0))
     return;
 
+  psx_gpu->vram_out_ptr = select_enhancement_buf_ptr(psx_gpu, x);
+
   x *= 2;
   y *= 2;