10 files changed, 60 insertions, 68 deletions

diff --git a/CMakeLists.txt b/CMakeLists.txt
index 8ecc2e476..2f3c59c5d 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -161,7 +161,7 @@ macro(yuzu_find_packages)
     #    Cmake Pkg Prefix  Version     Conan Pkg
         "Boost             1.73        boost/1.73.0"
         "Catch2            2.13        catch2/2.13.0"
-        "fmt               7.0         fmt/7.0.1"
+        "fmt               7.0         fmt/7.0.3"
     # can't use until https://github.com/bincrafters/community/issues/1173
         #"libzip            1.5         libzip/1.5.2@bincrafters/stable"
         "lz4               1.8         lz4/1.9.2"
diff --git a/src/common/lz4_compression.cpp b/src/common/lz4_compression.cpp
index ade6759bb..8e2e4094b 100644
--- a/src/common/lz4_compression.cpp
+++ b/src/common/lz4_compression.cpp
@@ -10,14 +10,14 @@
 
 namespace Common::Compression {
 
-std::vector<u8> CompressDataLZ4(const u8* source, std::size_t source_size) {
-    ASSERT_MSG(source_size <= LZ4_MAX_INPUT_SIZE, "Source size exceeds LZ4 maximum input size");
+std::vector<u8> CompressDataLZ4(std::span<const u8> source) {
+    ASSERT_MSG(source.size() <= LZ4_MAX_INPUT_SIZE, "Source size exceeds LZ4 maximum input size");
 
-    const auto source_size_int = static_cast<int>(source_size);
+    const auto source_size_int = static_cast<int>(source.size());
     const int max_compressed_size = LZ4_compressBound(source_size_int);
     std::vector<u8> compressed(max_compressed_size);
 
-    const int compressed_size = LZ4_compress_default(reinterpret_cast<const char*>(source),
+    const int compressed_size = LZ4_compress_default(reinterpret_cast<const char*>(source.data()),
                                                      reinterpret_cast<char*>(compressed.data()),
                                                      source_size_int, max_compressed_size);
 
@@ -31,18 +31,17 @@ std::vector<u8> CompressDataLZ4(const u8* source, std::size_t source_size) {
     return compressed;
 }
 
-std::vector<u8> CompressDataLZ4HC(const u8* source, std::size_t source_size,
-                                  s32 compression_level) {
-    ASSERT_MSG(source_size <= LZ4_MAX_INPUT_SIZE, "Source size exceeds LZ4 maximum input size");
+std::vector<u8> CompressDataLZ4HC(std::span<const u8> source, s32 compression_level) {
+    ASSERT_MSG(source.size() <= LZ4_MAX_INPUT_SIZE, "Source size exceeds LZ4 maximum input size");
 
     compression_level = std::clamp(compression_level, LZ4HC_CLEVEL_MIN, LZ4HC_CLEVEL_MAX);
 
-    const auto source_size_int = static_cast<int>(source_size);
+    const auto source_size_int = static_cast<int>(source.size());
     const int max_compressed_size = LZ4_compressBound(source_size_int);
     std::vector<u8> compressed(max_compressed_size);
 
     const int compressed_size = LZ4_compress_HC(
-        reinterpret_cast<const char*>(source), reinterpret_cast<char*>(compressed.data()),
+        reinterpret_cast<const char*>(source.data()), reinterpret_cast<char*>(compressed.data()),
         source_size_int, max_compressed_size, compression_level);
 
     if (compressed_size <= 0) {
@@ -55,8 +54,8 @@ std::vector<u8> CompressDataLZ4HC(const u8* source, std::size_t source_size,
     return compressed;
 }
 
-std::vector<u8> CompressDataLZ4HCMax(const u8* source, std::size_t source_size) {
-    return CompressDataLZ4HC(source, source_size, LZ4HC_CLEVEL_MAX);
+std::vector<u8> CompressDataLZ4HCMax(std::span<const u8> source) {
+    return CompressDataLZ4HC(source, LZ4HC_CLEVEL_MAX);
 }
 
 std::vector<u8> DecompressDataLZ4(const std::vector<u8>& compressed,
diff --git a/src/common/lz4_compression.h b/src/common/lz4_compression.h
index 4c16f6e03..173f9b9ad 100644
--- a/src/common/lz4_compression.h
+++ b/src/common/lz4_compression.h
@@ -4,6 +4,7 @@
 
 #pragma once
 
+#include <span>
 #include <vector>
 
 #include "common/common_types.h"
@@ -14,11 +15,10 @@ namespace Common::Compression {
  * Compresses a source memory region with LZ4 and returns the compressed data in a vector.
  *
  * @param source the uncompressed source memory region.
- * @param source_size the size in bytes of the uncompressed source memory region.
  *
  * @return the compressed data.
  */
-std::vector<u8> CompressDataLZ4(const u8* source, std::size_t source_size);
+std::vector<u8> CompressDataLZ4(std::span<const u8> source);
 
 /**
  * Utilizes the LZ4 subalgorithm LZ4HC with the specified compression level. Higher compression
@@ -27,22 +27,20 @@ std::vector<u8> CompressDataLZ4(const u8* source, std::size_t source_size);
  * also be decompressed with the default LZ4 decompression.
  *
  * @param source the uncompressed source memory region.
- * @param source_size the size in bytes of the uncompressed source memory region.
  * @param compression_level the used compression level. Should be between 3 and 12.
  *
  * @return the compressed data.
  */
-std::vector<u8> CompressDataLZ4HC(const u8* source, std::size_t source_size, s32 compression_level);
+std::vector<u8> CompressDataLZ4HC(std::span<const u8> source, s32 compression_level);
 
 /**
  * Utilizes the LZ4 subalgorithm LZ4HC with the highest possible compression level.
  *
  * @param source the uncompressed source memory region.
- * @param source_size the size in bytes of the uncompressed source memory region.
  *
  * @return the compressed data.
  */
-std::vector<u8> CompressDataLZ4HCMax(const u8* source, std::size_t source_size);
+std::vector<u8> CompressDataLZ4HCMax(std::span<const u8> source);
 
 /**
  * Decompresses a source memory region with LZ4 and returns the uncompressed data in a vector.
diff --git a/src/common/zstd_compression.cpp b/src/common/zstd_compression.cpp
index 978526492..770833ee7 100644
--- a/src/common/zstd_compression.cpp
+++ b/src/common/zstd_compression.cpp
@@ -5,19 +5,18 @@
 #include <algorithm>
 #include <zstd.h>
 
-#include "common/assert.h"
 #include "common/zstd_compression.h"
 
 namespace Common::Compression {
 
-std::vector<u8> CompressDataZSTD(const u8* source, std::size_t source_size, s32 compression_level) {
+std::vector<u8> CompressDataZSTD(std::span<const u8> source, s32 compression_level) {
     compression_level = std::clamp(compression_level, 1, ZSTD_maxCLevel());
 
-    const std::size_t max_compressed_size = ZSTD_compressBound(source_size);
+    const std::size_t max_compressed_size = ZSTD_compressBound(source.size());
     std::vector<u8> compressed(max_compressed_size);
 
-    const std::size_t compressed_size =
-        ZSTD_compress(compressed.data(), compressed.size(), source, source_size, compression_level);
+    const std::size_t compressed_size = ZSTD_compress(
+        compressed.data(), compressed.size(), source.data(), source.size(), compression_level);
 
     if (ZSTD_isError(compressed_size)) {
         // Compression failed
@@ -29,8 +28,8 @@ std::vector<u8> CompressDataZSTD(const u8* source, std::size_t source_size, s32
     return compressed;
 }
 
-std::vector<u8> CompressDataZSTDDefault(const u8* source, std::size_t source_size) {
-    return CompressDataZSTD(source, source_size, ZSTD_CLEVEL_DEFAULT);
+std::vector<u8> CompressDataZSTDDefault(std::span<const u8> source) {
+    return CompressDataZSTD(source, ZSTD_CLEVEL_DEFAULT);
 }
 
 std::vector<u8> DecompressDataZSTD(const std::vector<u8>& compressed) {
diff --git a/src/common/zstd_compression.h b/src/common/zstd_compression.h
index e9de941c8..b5edf19e7 100644
--- a/src/common/zstd_compression.h
+++ b/src/common/zstd_compression.h
@@ -4,6 +4,7 @@
 
 #pragma once
 
+#include <span>
 #include <vector>
 
 #include "common/common_types.h"
@@ -14,23 +15,21 @@ namespace Common::Compression {
  * Compresses a source memory region with Zstandard and returns the compressed data in a vector.
  *
  * @param source the uncompressed source memory region.
- * @param source_size the size in bytes of the uncompressed source memory region.
  * @param compression_level the used compression level. Should be between 1 and 22.
  *
  * @return the compressed data.
  */
-std::vector<u8> CompressDataZSTD(const u8* source, std::size_t source_size, s32 compression_level);
+std::vector<u8> CompressDataZSTD(std::span<const u8> source, s32 compression_level);
 
 /**
  * Compresses a source memory region with Zstandard with the default compression level and returns
  * the compressed data in a vector.
  *
  * @param source the uncompressed source memory region.
- * @param source_size the size in bytes of the uncompressed source memory region.
  *
  * @return the compressed data.
  */
-std::vector<u8> CompressDataZSTDDefault(const u8* source, std::size_t source_size);
+std::vector<u8> CompressDataZSTDDefault(std::span<const u8> source);
 
 /**
  * Decompresses a source memory region with Zstandard and returns the uncompressed data in a vector.
diff --git a/src/core/hle/service/time/time_zone_content_manager.cpp b/src/core/hle/service/time/time_zone_content_manager.cpp
index c070d6e97..320672add 100644
--- a/src/core/hle/service/time/time_zone_content_manager.cpp
+++ b/src/core/hle/service/time/time_zone_content_manager.cpp
@@ -73,10 +73,8 @@ TimeZoneContentManager::TimeZoneContentManager(TimeManager& time_manager, Core::
 
     std::string location_name;
     const auto timezone_setting = Settings::GetTimeZoneString();
-    if (timezone_setting == "auto") {
+    if (timezone_setting == "auto" || timezone_setting == "default") {
         location_name = Common::TimeZone::GetDefaultTimeZone();
-    } else if (timezone_setting == "default") {
-        location_name = location_name;
     } else {
         location_name = timezone_setting;
     }
diff --git a/src/video_core/engines/maxwell_dma.cpp b/src/video_core/engines/maxwell_dma.cpp
index a2d3d7823..e88290754 100644
--- a/src/video_core/engines/maxwell_dma.cpp
+++ b/src/video_core/engines/maxwell_dma.cpp
@@ -94,7 +94,8 @@ void MaxwellDMA::CopyPitchToPitch() {
 }
 
 void MaxwellDMA::CopyBlockLinearToPitch() {
-    ASSERT(regs.src_params.block_size.depth == 0);
+    UNIMPLEMENTED_IF(regs.src_params.block_size.depth != 0);
+    UNIMPLEMENTED_IF(regs.src_params.layer != 0);
 
     // Optimized path for micro copies.
     const size_t dst_size = static_cast<size_t>(regs.pitch_out) * regs.line_count;
@@ -123,17 +124,12 @@ void MaxwellDMA::CopyBlockLinearToPitch() {
         write_buffer.resize(dst_size);
     }
 
-    if (Settings::IsGPULevelExtreme()) {
-        memory_manager.ReadBlock(regs.offset_in, read_buffer.data(), src_size);
-        memory_manager.ReadBlock(regs.offset_out, write_buffer.data(), dst_size);
-    } else {
-        memory_manager.ReadBlockUnsafe(regs.offset_in, read_buffer.data(), src_size);
-        memory_manager.ReadBlockUnsafe(regs.offset_out, write_buffer.data(), dst_size);
-    }
+    memory_manager.ReadBlock(regs.offset_in, read_buffer.data(), src_size);
+    memory_manager.ReadBlock(regs.offset_out, write_buffer.data(), dst_size);
 
     UnswizzleSubrect(regs.line_length_in, regs.line_count, regs.pitch_out, width, bytes_per_pixel,
-                     read_buffer.data() + src_layer_size * src_params.layer, write_buffer.data(),
-                     block_height, src_params.origin.x, src_params.origin.y);
+                     block_height, src_params.origin.x, src_params.origin.y, write_buffer.data(),
+                     read_buffer.data());
 
     memory_manager.WriteBlock(regs.offset_out, write_buffer.data(), dst_size);
 }
@@ -198,7 +194,6 @@ void MaxwellDMA::FastCopyBlockLinearToPitch() {
     if (read_buffer.size() < src_size) {
         read_buffer.resize(src_size);
     }
-
     if (write_buffer.size() < dst_size) {
         write_buffer.resize(dst_size);
     }
@@ -212,8 +207,8 @@ void MaxwellDMA::FastCopyBlockLinearToPitch() {
     }
 
     UnswizzleSubrect(regs.line_length_in, regs.line_count, regs.pitch_out, regs.src_params.width,
-                     bytes_per_pixel, read_buffer.data(), write_buffer.data(),
-                     regs.src_params.block_size.height, pos_x, pos_y);
+                     bytes_per_pixel, regs.src_params.block_size.height, pos_x, pos_y,
+                     write_buffer.data(), read_buffer.data());
 
     memory_manager.WriteBlock(regs.offset_out, write_buffer.data(), dst_size);
 }
diff --git a/src/video_core/renderer_opengl/gl_shader_disk_cache.cpp b/src/video_core/renderer_opengl/gl_shader_disk_cache.cpp
index 2dcc2b0eb..c0e73789b 100644
--- a/src/video_core/renderer_opengl/gl_shader_disk_cache.cpp
+++ b/src/video_core/renderer_opengl/gl_shader_disk_cache.cpp
@@ -422,8 +422,7 @@ void ShaderDiskCacheOpenGL::SavePrecompiledHeaderToVirtualPrecompiledCache() {
 void ShaderDiskCacheOpenGL::SaveVirtualPrecompiledFile() {
     precompiled_cache_virtual_file_offset = 0;
     const std::vector<u8> uncompressed = precompiled_cache_virtual_file.ReadAllBytes();
-    const std::vector<u8> compressed =
-        Common::Compression::CompressDataZSTDDefault(uncompressed.data(), uncompressed.size());
+    const std::vector<u8> compressed = Common::Compression::CompressDataZSTDDefault(uncompressed);
 
     const auto precompiled_path{GetPrecompiledPath()};
     FileUtil::IOFile file(precompiled_path, "wb");
diff --git a/src/video_core/textures/decoders.cpp b/src/video_core/textures/decoders.cpp
index 474ae620a..16d46a018 100644
--- a/src/video_core/textures/decoders.cpp
+++ b/src/video_core/textures/decoders.cpp
@@ -228,24 +228,30 @@ void SwizzleSubrect(u32 subrect_width, u32 subrect_height, u32 source_pitch, u32
     }
 }
 
-void UnswizzleSubrect(u32 subrect_width, u32 subrect_height, u32 dest_pitch, u32 swizzled_width,
-                      u32 bytes_per_pixel, u8* swizzled_data, u8* unswizzled_data,
-                      u32 block_height_bit, u32 offset_x, u32 offset_y) {
-    const u32 block_height = 1U << block_height_bit;
-    for (u32 line = 0; line < subrect_height; ++line) {
-        const u32 y2 = line + offset_y;
-        const u32 gob_address_y = (y2 / (GOB_SIZE_Y * block_height)) * GOB_SIZE * block_height +
-                                  ((y2 % (GOB_SIZE_Y * block_height)) / GOB_SIZE_Y) * GOB_SIZE;
-        const auto& table = LEGACY_SWIZZLE_TABLE[y2 % GOB_SIZE_Y];
-        for (u32 x = 0; x < subrect_width; ++x) {
-            const u32 x2 = (x + offset_x) * bytes_per_pixel;
-            const u32 gob_address = gob_address_y + (x2 / GOB_SIZE_X) * GOB_SIZE * block_height;
-            const u32 swizzled_offset = gob_address + table[x2 % GOB_SIZE_X];
-            const u32 unswizzled_offset = line * dest_pitch + x * bytes_per_pixel;
-            u8* dest_line = unswizzled_data + unswizzled_offset;
-            u8* source_addr = swizzled_data + swizzled_offset;
+void UnswizzleSubrect(u32 line_length_in, u32 line_count, u32 pitch, u32 width, u32 bytes_per_pixel,
+                      u32 block_height, u32 origin_x, u32 origin_y, u8* output, const u8* input) {
+    const u32 stride = width * bytes_per_pixel;
+    const u32 gobs_in_x = (stride + GOB_SIZE_X - 1) / GOB_SIZE_X;
+    const u32 block_size = gobs_in_x << (GOB_SIZE_SHIFT + block_height);
+
+    const u32 block_height_mask = (1U << block_height) - 1;
+    const u32 x_shift = static_cast<u32>(GOB_SIZE_SHIFT) + block_height;
+
+    for (u32 line = 0; line < line_count; ++line) {
+        const u32 src_y = line + origin_y;
+        const auto& table = LEGACY_SWIZZLE_TABLE[src_y % GOB_SIZE_Y];
+
+        const u32 block_y = src_y >> GOB_SIZE_Y_SHIFT;
+        const u32 src_offset_y = (block_y >> block_height) * block_size +
+                                 ((block_y & block_height_mask) << GOB_SIZE_SHIFT);
+        for (u32 column = 0; column < line_length_in; ++column) {
+            const u32 src_x = (column + origin_x) * bytes_per_pixel;
+            const u32 src_offset_x = (src_x >> GOB_SIZE_X_SHIFT) << x_shift;
+
+            const u32 swizzled_offset = src_offset_y + src_offset_x + table[src_x % GOB_SIZE_X];
+            const u32 unswizzled_offset = line * pitch + column * bytes_per_pixel;
 
-            std::memcpy(dest_line, source_addr, bytes_per_pixel);
+            std::memcpy(output + unswizzled_offset, input + swizzled_offset, bytes_per_pixel);
         }
     }
 }
@@ -261,7 +267,7 @@ void SwizzleSliceToVoxel(u32 line_length_in, u32 line_count, u32 pitch, u32 widt
     const u32 block_size = gobs_in_x << (GOB_SIZE_SHIFT + block_height + block_depth);
 
     const u32 block_height_mask = (1U << block_height) - 1;
-    const u32 x_shift = Common::CountTrailingZeroes32(GOB_SIZE << (block_height + block_depth));
+    const u32 x_shift = static_cast<u32>(GOB_SIZE_SHIFT) + block_height + block_depth;
 
     for (u32 line = 0; line < line_count; ++line) {
         const auto& table = LEGACY_SWIZZLE_TABLE[line % GOB_SIZE_Y];
diff --git a/src/video_core/textures/decoders.h b/src/video_core/textures/decoders.h
index d6fe35d37..01e156bc8 100644
--- a/src/video_core/textures/decoders.h
+++ b/src/video_core/textures/decoders.h
@@ -48,9 +48,8 @@ void SwizzleSubrect(u32 subrect_width, u32 subrect_height, u32 source_pitch, u32
                     u32 block_height_bit, u32 offset_x, u32 offset_y);
 
 /// Copies a tiled subrectangle into a linear surface.
-void UnswizzleSubrect(u32 subrect_width, u32 subrect_height, u32 dest_pitch, u32 swizzled_width,
-                      u32 bytes_per_pixel, u8* swizzled_data, u8* unswizzled_data, u32 block_height,
-                      u32 offset_x, u32 offset_y);
+void UnswizzleSubrect(u32 line_length_in, u32 line_count, u32 pitch, u32 width, u32 bytes_per_pixel,
+                      u32 block_height, u32 origin_x, u32 origin_y, u8* output, const u8* input);
 
 /// @brief Swizzles a 2D array of pixels into a 3D texture
 /// @param line_length_in  Number of pixels per line