summaryrefslogtreecommitdiff
path: root/src/video_core/textures/decoders.cpp
diff options
context:
space:
mode:
authorGravatar Fernando Sahmkow2019-07-24 19:18:17 -0400
committerGravatar FernandoS272019-07-25 20:41:42 -0400
commita452ff983dc736e16b0f1c0b01d162bfb6655c79 (patch)
treee2b6662a5f3dd884f4354bbabec82af33ad2cd77 /src/video_core/textures/decoders.cpp
parentMerge pull request #2704 from FernandoS27/conditional (diff)
downloadyuzu-a452ff983dc736e16b0f1c0b01d162bfb6655c79.tar.gz
yuzu-a452ff983dc736e16b0f1c0b01d162bfb6655c79.tar.xz
yuzu-a452ff983dc736e16b0f1c0b01d162bfb6655c79.zip
MaxwellDMA: Fixes, corrections and relaxations.
This commit fixes offsets on Linear -> Tiled copies, corrects z pos fortiled->linear copies, corrects bytes_per_pixel calculation in tiled -> linear copies and relaxes some limitations set by latest dma fixes refactors.
Diffstat (limited to 'src/video_core/textures/decoders.cpp')
-rw-r--r--src/video_core/textures/decoders.cpp14
1 files changed, 8 insertions, 6 deletions
diff --git a/src/video_core/textures/decoders.cpp b/src/video_core/textures/decoders.cpp
index 7e8295944..7df5f1452 100644
--- a/src/video_core/textures/decoders.cpp
+++ b/src/video_core/textures/decoders.cpp
@@ -257,19 +257,21 @@ std::vector<u8> UnswizzleTexture(u8* address, u32 tile_size_x, u32 tile_size_y,
257 257
258void SwizzleSubrect(u32 subrect_width, u32 subrect_height, u32 source_pitch, u32 swizzled_width, 258void SwizzleSubrect(u32 subrect_width, u32 subrect_height, u32 source_pitch, u32 swizzled_width,
259 u32 bytes_per_pixel, u8* swizzled_data, u8* unswizzled_data, 259 u32 bytes_per_pixel, u8* swizzled_data, u8* unswizzled_data,
260 u32 block_height_bit) { 260 u32 block_height_bit, u32 offset_x, u32 offset_y) {
261 const u32 block_height = 1U << block_height_bit; 261 const u32 block_height = 1U << block_height_bit;
262 const u32 image_width_in_gobs{(swizzled_width * bytes_per_pixel + (gob_size_x - 1)) / 262 const u32 image_width_in_gobs{(swizzled_width * bytes_per_pixel + (gob_size_x - 1)) /
263 gob_size_x}; 263 gob_size_x};
264 for (u32 line = 0; line < subrect_height; ++line) { 264 for (u32 line = 0; line < subrect_height; ++line) {
265 const u32 dst_y = line + offset_y;
265 const u32 gob_address_y = 266 const u32 gob_address_y =
266 (line / (gob_size_y * block_height)) * gob_size * block_height * image_width_in_gobs + 267 (dst_y / (gob_size_y * block_height)) * gob_size * block_height * image_width_in_gobs +
267 ((line % (gob_size_y * block_height)) / gob_size_y) * gob_size; 268 ((dst_y % (gob_size_y * block_height)) / gob_size_y) * gob_size;
268 const auto& table = legacy_swizzle_table[line % gob_size_y]; 269 const auto& table = legacy_swizzle_table[dst_y % gob_size_y];
269 for (u32 x = 0; x < subrect_width; ++x) { 270 for (u32 x = 0; x < subrect_width; ++x) {
271 const u32 dst_x = x + offset_x;
270 const u32 gob_address = 272 const u32 gob_address =
271 gob_address_y + (x * bytes_per_pixel / gob_size_x) * gob_size * block_height; 273 gob_address_y + (dst_x * bytes_per_pixel / gob_size_x) * gob_size * block_height;
272 const u32 swizzled_offset = gob_address + table[(x * bytes_per_pixel) % gob_size_x]; 274 const u32 swizzled_offset = gob_address + table[(dst_x * bytes_per_pixel) % gob_size_x];
273 u8* source_line = unswizzled_data + line * source_pitch + x * bytes_per_pixel; 275 u8* source_line = unswizzled_data + line * source_pitch + x * bytes_per_pixel;
274 u8* dest_addr = swizzled_data + swizzled_offset; 276 u8* dest_addr = swizzled_data + swizzled_offset;
275 277