Compare commits

..

3 Commits

Author SHA1 Message Date
lizzie b92ca83fa5 L;ICENSES 2026-07-01 06:32:09 +00:00
lizzie f768ba4170 ploo 2026-07-01 06:31:43 +00:00
lizzie 7dff851245 [vk] implement VK_EXT_border_color_swizzle
Signed-off-by: lizzie <lizzie@eden-emu.dev>
2026-07-01 05:22:23 +00:00
15 changed files with 269 additions and 176 deletions
+30 -30
View File
@@ -9,7 +9,7 @@
"boost": {
"artifact": "%TAG%-cmake.tar.xz",
"find_args": "CONFIG OPTIONAL_COMPONENTS headers context system fiber filesystem",
"hash": "935b3d516e996f6d25948ba8a54c1b7f70f7f0e3f517e36481fdf0196c2c5cfc2841f86e891f3df9517746b7fb605db47cdded1b8ff78d9482ddaa621db43a34",
"hash": "6ae6e94664fe7f2fb01976b59b276ac5df8085c7503fa829d810fbfe495960cfec44fa2c36e2cb23480bc19c956ed199d4952b02639a00a6c07625d4e7130c2d",
"min_version": "1.57",
"package": "Boost",
"patches": [
@@ -17,17 +17,17 @@
],
"repo": "boostorg/boost",
"tag": "boost-%VERSION%",
"version": "1.91.0"
"version": "1.90.0"
},
"boost_headers": {
"bundled": true,
"hash": "c5fa2cd72f6e6666b7963b97bc359c75284b8fb540c30f3629a028b85270c9bc66c8a051383964f2bd4c1e005a4691593d15696e7ef39ea87cf6cff9e5691fb2",
"hash": "4ef845775e2277a8104ded6ddf749aa262ce52cf8438042869a048f9a0156dd772fbbcfa74efa1378fecef339b7286f6fe4b4feb5c45d49966b35d08e3e83507",
"repo": "boostorg/headers",
"tag": "boost-%VERSION%",
"version": "1.91.0"
"version": "1.90.0"
},
"catch2": {
"hash": "2e32d3d41e4ef84d927eaaaff0a75eba8fd1e47a5ddeb6007b73934f9d1ac747bf9818d02d2cdfb6343830b5091ba4f3d66efc281fc376a4f0716753dd7c9a97",
"hash": "7eea385d79d88a5690cde131fe7ccda97d5c54ea09d6f515000d7bf07c828809d61c1ac99912c1ee507cf933f61c1c47ecdcc45df7850ffa82714034b0fccf35",
"min_version": "3.0.1",
"package": "Catch2",
"patches": [
@@ -35,7 +35,7 @@
],
"repo": "catchorg/Catch2",
"tag": "v%VERSION%",
"version": "3.15.1"
"version": "3.13.0"
},
"cpp-jwt": {
"find_args": "CONFIG",
@@ -96,11 +96,11 @@
"version": "8.0.1-c7b5f1537d"
},
"fmt": {
"hash": "5ac2ba0f54a484999ed5407d82b77aad170cea49a267decd2c0eedadf3b14413e2a83fcc8e9ca9c16640595e019b8636e160f72314d8be50653324e82ac745eb",
"hash": "f0da82c545b01692e9fd30fdfb613dbb8dd9716983dcd0ff19ac2a8d36f74beb5540ef38072fdecc1e34191b3682a8542ecbf3a61ef287dbba0a2679d4e023f2",
"min_version": "8",
"repo": "fmtlib/fmt",
"tag": "%VERSION%",
"version": "12.2.0"
"version": "12.1.0"
},
"frozen": {
"hash": "b8dfe741c82bc178dfc9749d4ab5a130cee718d9ee7b71d9b547cf5f7f23027ed0152ad250012a8546399fcc1e12187efc68d89d6731256c4d2df7d04eef8d5c",
@@ -119,7 +119,7 @@
},
"httplib": {
"find_args": "MODULE GLOBAL",
"hash": "5c8a151d0d40dd413975e12e4bbfd790a087c7e9c96bf86db31ce43921cbccece01a5cca78e47a8cfd12370f97e174a158aa8a826781e12fa256b157665e1310",
"hash": "159ed94965018f2a371d45a3bfc1961e5fb1549e501ded70a6b4532d7fe99d0579c18b5195aff6e35f96f399b426cea2650ec9fb75ef80d4c9edeccb51f2e6c9",
"options": [
"HTTPLIB_REQUIRE_OPENSSL ON",
"HTTPLIB_DISABLE_MACOSX_AUTOMATIC_ROOT_CERTIFICATES ON"
@@ -129,7 +129,7 @@
],
"repo": "yhirose/cpp-httplib",
"tag": "v%VERSION%",
"version": "0.49.0"
"version": "0.46.0"
},
"lagoon": {
"hash": "b9380f99c6effaeccc6d8f81d4942e852c11ad28613df637e155451556ae5826f93765bee57a5c87a9740d2bd1db463ad0f55a947772fe9d57eeabae3efa373e",
@@ -148,13 +148,13 @@
},
"libusb": {
"find_args": "MODULE",
"hash": "3251c9f41e900efa13caf981483f46886c8434bf4c30f3fd3073921b06be9977cebcf4a18f7bc46db86e33d7f19752296377be9050abfb8d887ffeb377864647",
"hash": "98c5f7940ff06b25c9aa65aa98e23de4c79a4c1067595f4c73cc145af23a1c286639e1ba11185cd91bab702081f307b973f08a4c9746576dc8d01b3620a3aeb5",
"patches": [
"0001-netbsd-gettime.patch"
],
"repo": "libusb/libusb",
"tag": "v%VERSION%",
"version": "1.0.30"
"version": "1.0.29"
},
"llvm-mingw": {
"artifact": "clang-rt-builtins.tar.zst",
@@ -203,7 +203,7 @@
"version": "1.10.0"
},
"openssl": {
"hash": "2405891660b67e4cc01c7868e100f2766e1c6290120213ae62f775ef3d9610c3527bdc43d4c831260417450c74353b0693ae3e24cca3e99b02e4daa548b93904",
"hash": "29002ce50cb95a4f4f1d0e9d3f684401fbd4eac34203dc2eef3b6334af5d44aa46bf788b63a6f5c139c383eafb7269ae87a58a9a3ad5912903b9773e545ccc0a",
"min_version": "3.0.0",
"package": "OpenSSL",
"patches": [
@@ -211,7 +211,7 @@
],
"repo": "openssl/openssl",
"tag": "openssl-%VERSION%",
"version": "4.0.1"
"version": "3.6.2"
},
"openssl-ci": {
"ci": true,
@@ -223,7 +223,7 @@
},
"openssl-cmake": {
"bundled": true,
"hash": "bc73a6c035f299711bb90e20b16fa3521e8ffdab32d8d17f9e7f5cde1ff5def37ec6f9e03de25c3466d18885815c0874331f56e592b8a2f7deeb484c69c5c621",
"hash": "2cc185c924fd70e7d886257ca0caa42b3b8f7f712f2052b4f94dde74759e27022de76178460e18c9bdfc57c366583999e198fbb6052d4e7d91c099d15a0ca63e",
"options": [
"OPENSSL_CONFIGURE_OPTIONS threads"
],
@@ -235,7 +235,7 @@
],
"repo": "jimmy-park/openssl-cmake",
"tag": "%VERSION%",
"version": "4.0.1"
"version": "3.6.2"
},
"opus": {
"find_args": "MODULE",
@@ -268,12 +268,12 @@
"version": "1.5"
},
"sdl3": {
"hash": "fc0a55ca01c32f613b9cd8c6cffad17ee0855ee542f03fa455632043c99a0a259599557a46c7c636032444260e64476867d9af43a1da689837f99c22942cd863",
"hash": "df5a323af7ac366661a3c0e887969c72584d232f3cc211419d59b0487b620b6b2859d4549c9e8df002ee489290062e466fcfddf7edc0872a37b1f2845e81c0f3",
"min_version": "3.2.10",
"package": "SDL3",
"repo": "libsdl-org/SDL",
"tag": "release-%VERSION%",
"version": "3.4.12"
"version": "3.4.8"
},
"sdl3-ci": {
"ci": true,
@@ -285,11 +285,11 @@
},
"simpleini": {
"find_args": "MODULE",
"hash": "a62c5748efe2473aae5bddab96ba9114d981a72f5b0d1a44d563daa085d5c231ed8c447794691d9bd67e1e0c6bfb44e4a8736be75fee59967d0c67ce3a59bb6e",
"hash": "b937c18a7b6277d77ca7ebfb216af4984810f77af4c32d101b7685369a4bd5eb61406223f82698e167e6311a728d07415ab59639fdf19eff71ad6dc2abfda989",
"package": "SimpleIni",
"repo": "brofield/simpleini",
"tag": "v%VERSION%",
"version": "4.26"
"version": "4.25"
},
"sirit": {
"artifact": "sirit-source-%VERSION%.tar.zst",
@@ -340,42 +340,42 @@
"version": "4.8.1"
},
"vulkan-headers": {
"hash": "2c336c163cc184f1aad147f8829dedabdffe0a801e1636c1ae8445feca76a64fc71815d63a69a85db6f5c485ecb6ff3197d94aef63b06f7f9daac0260f7853a0",
"hash": "d2846ea228415772645eea4b52a9efd33e6a563043dd3de059e798be6391a8f0ca089f455ae420ff22574939ed0f48ed7c6ff3d5a9987d5231dbf3b3f89b484b",
"min_version": "1.4.317",
"package": "VulkanHeaders",
"repo": "KhronosGroup/Vulkan-Headers",
"tag": "v%VERSION%",
"version": "1.4.356"
"version": "1.4.345"
},
"vulkan-memory-allocator": {
"find_args": "CONFIG",
"hash": "563acbcd8912d10d92c23715eba7acf0e7c1683af36021f415b36f359c2cce065f3906e395c32282a8410ec5c8179fbcb6412935c6629a49357475d4b4410e2a",
"hash": "deb5902ef8db0e329fbd5f3f4385eb0e26bdd9f14f3a2334823fb3fe18f36bc5d235d620d6e5f6fe3551ec3ea7038638899db8778c09f6d5c278f5ff95c3344b",
"package": "VulkanMemoryAllocator",
"repo": "GPUOpen-LibrariesAndSDKs/VulkanMemoryAllocator",
"tag": "v%VERSION%",
"version": "3.4.0"
"version": "3.3.0"
},
"vulkan-utility-libraries": {
"hash": "e35d715529aa96f82f2080e29c2f6da8da63ded70ff76edadf77e1f33b0f752ca4372d11478014d876d91c9313323c7a49c4e10bda35a324923ace0048dc91d5",
"hash": "114f6b237a6dcba923ccc576befb5dea3f1c9b3a30de7dc741f234a831d1c2d52d8a224afb37dd57dffca67ac0df461eaaab6a5ab5e503b393f91c166680c3e1",
"package": "VulkanUtilityLibraries",
"repo": "KhronosGroup/Vulkan-Utility-Libraries",
"tag": "v%VERSION%",
"version": "1.4.356"
"version": "1.4.345"
},
"vulkan-validation-layers": {
"artifact": "android-binaries-%VERSION%.zip",
"hash": "1888eaaa4f0ccd7f3d1f0c6485ae5508f39e7b402ea8486d8243756d63073def569ddee75ea40b403cbfcb2ead515c410d2ff16ae48d8a4d6b4fbacbd78e24e4",
"hash": "8812ae84cbe49e6a3418ade9c458d3be6d74a3dffd319d4502007b564d580998056e8190414368ec11b27bc83993c7a0dad713c31bcc3d9553b51243efee3753",
"package": "VVL",
"repo": "KhronosGroup/Vulkan-ValidationLayers",
"tag": "vulkan-sdk-%VERSION%",
"version": "1.4.350.1"
"version": "1.4.341.0"
},
"xbyak": {
"hash": "14a0a8e3a768e6256e220b5edfd783de5e89f36736ef62674c8532f2564df03719311db923a08d2c930491632f2f83e486c8264e9997b6394ef59d32cae3070b",
"hash": "b6475276b2faaeb315734ea8f4f8bd87ededcee768961b39679bee547e7f3e98884d8b7851e176d861dab30a80a76e6ea302f8c111483607dde969b4797ea95a",
"package": "xbyak",
"repo": "herumi/xbyak",
"tag": "v%VERSION%",
"version": "7.37.4"
"version": "7.35.2"
},
"zlib": {
"hash": "16fea4df307a68cf0035858abe2fd550250618a97590e202037acd18a666f57afc10f8836cbbd472d54a0e76539d0e558cb26f059d53de52ff90634bbf4f47d4",
@@ -11,8 +11,9 @@
#define BEGIN_PUSH_CONSTANTS layout(push_constant) uniform PushConstants {
#define END_PUSH_CONSTANTS };
#define UNIFORM(n)
#define BINDING_INPUT_BUFFER 0
#define BINDING_OUTPUT_IMAGE 1
#define BINDING_SWIZZLE_BUFFER 0
#define BINDING_INPUT_BUFFER 1
#define BINDING_OUTPUT_IMAGE 2
#else // ^^^ Vulkan ^^^ // vvv OpenGL vvv
@@ -25,7 +26,8 @@
#define BEGIN_PUSH_CONSTANTS
#define END_PUSH_CONSTANTS
#define UNIFORM(n) layout (location = n) uniform
#define BINDING_INPUT_BUFFER 0
#define BINDING_SWIZZLE_BUFFER 0
#define BINDING_INPUT_BUFFER 1
#define BINDING_OUTPUT_IMAGE 0
#endif
@@ -41,6 +43,10 @@ UNIFORM(6) uint block_height;
UNIFORM(7) uint block_height_mask;
END_PUSH_CONSTANTS
layout(binding = BINDING_SWIZZLE_BUFFER, std430) readonly buffer SwizzleTable {
uint swizzle_table[];
};
#if HAS_EXTENDED_TYPES
layout(binding = BINDING_INPUT_BUFFER, std430) buffer InputBufferU8 { uint8_t u8data[]; };
layout(binding = BINDING_INPUT_BUFFER, std430) buffer InputBufferU16 { uint16_t u16data[]; };
@@ -65,19 +71,9 @@ const uint GOB_SIZE_SHIFT = GOB_SIZE_X_SHIFT + GOB_SIZE_Y_SHIFT + GOB_SIZE_Z_SHI
const uvec2 SWIZZLE_MASK = uvec2(GOB_SIZE_X - 1, GOB_SIZE_Y - 1);
uint SwizzleTable(uint pos) {
const uint t[8] = uint[](
0x12100200, 0x13110301, 0x16140604, 0x17150705,
0x1a180a08, 0x1b190b09, 0x1e1c0e0c, 0x1f1d0f0d
);
const uint i = pos >> 4;
const uint h = (t[i / 4] >> ((i % 4) * 8)) & 0xff;
return (h << 4) | (pos & 0xf);
}
uint SwizzleOffset(uvec2 pos) {
pos = pos & SWIZZLE_MASK;
return SwizzleTable(pos.y * 64 + pos.x);
return swizzle_table[pos.y * 64 + pos.x];
}
uvec4 ReadTexel(uint offset) {
@@ -11,8 +11,9 @@
#define BEGIN_PUSH_CONSTANTS layout(push_constant) uniform PushConstants {
#define END_PUSH_CONSTANTS };
#define UNIFORM(n)
#define BINDING_INPUT_BUFFER 0
#define BINDING_OUTPUT_IMAGE 1
#define BINDING_SWIZZLE_BUFFER 0
#define BINDING_INPUT_BUFFER 1
#define BINDING_OUTPUT_IMAGE 2
#else // ^^^ Vulkan ^^^ // vvv OpenGL vvv
@@ -25,7 +26,8 @@
#define BEGIN_PUSH_CONSTANTS
#define END_PUSH_CONSTANTS
#define UNIFORM(n) layout (location = n) uniform
#define BINDING_INPUT_BUFFER 0
#define BINDING_SWIZZLE_BUFFER 0
#define BINDING_INPUT_BUFFER 1
#define BINDING_OUTPUT_IMAGE 0
#endif
@@ -43,6 +45,10 @@ UNIFORM(8) uint block_depth;
UNIFORM(9) uint block_depth_mask;
END_PUSH_CONSTANTS
layout(binding = BINDING_SWIZZLE_BUFFER, std430) readonly buffer SwizzleTable {
uint swizzle_table[];
};
#if HAS_EXTENDED_TYPES
layout(binding = BINDING_INPUT_BUFFER, std430) buffer InputBufferU8 { uint8_t u8data[]; };
layout(binding = BINDING_INPUT_BUFFER, std430) buffer InputBufferU16 { uint16_t u16data[]; };
@@ -67,19 +73,9 @@ const uint GOB_SIZE_SHIFT = GOB_SIZE_X_SHIFT + GOB_SIZE_Y_SHIFT + GOB_SIZE_Z_SHI
const uvec2 SWIZZLE_MASK = uvec2(GOB_SIZE_X - 1, GOB_SIZE_Y - 1);
uint SwizzleTable(uint pos) {
const uint t[8] = uint[](
0x12100200, 0x13110301, 0x16140604, 0x17150705,
0x1a180a08, 0x1b190b09, 0x1e1c0e0c, 0x1f1d0f0d
);
const uint i = pos >> 4;
const uint h = (t[i / 4] >> ((i % 4) * 8)) & 0xff;
return (h << 4) | (pos & 0xf);
}
uint SwizzleOffset(uvec2 pos) {
pos = pos & SWIZZLE_MASK;
return SwizzleTable(pos.y * 64 + pos.x);
return swizzle_table[pos.y * 64 + pos.x];
}
uvec4 ReadTexel(uint offset) {
@@ -10,8 +10,9 @@
#define BEGIN_PUSH_CONSTANTS layout(push_constant) uniform PushConstants {
#define END_PUSH_CONSTANTS };
#define UNIFORM(n)
#define BINDING_INPUT_BUFFER 0
#define BINDING_OUTPUT_BUFFER 1
#define BINDING_SWIZZLE_BUFFER 0
#define BINDING_INPUT_BUFFER 1
#define BINDING_OUTPUT_BUFFER 2
#else
#extension GL_NV_gpu_shader5 : enable
#ifdef GL_NV_gpu_shader5
@@ -22,6 +23,7 @@
#define BEGIN_PUSH_CONSTANTS
#define END_PUSH_CONSTANTS
#define UNIFORM(n) layout(location = n) uniform
#define BINDING_SWIZZLE_BUFFER 0
#define BINDING_INPUT_BUFFER 1
#define BINDING_OUTPUT_BUFFER 0
#endif
@@ -64,9 +66,13 @@ END_PUSH_CONSTANTS
#endif
// --- Buffers ---
layout(binding = BINDING_SWIZZLE_BUFFER, std430) readonly buffer SwizzleTable {
uint swizzle_table[];
};
#if HAS_EXTENDED_TYPES
layout(binding = BINDING_INPUT_BUFFER, std430) buffer InputBufferU8 { uint8_t u8data[]; };
layout(binding = BINDING_INPUT_BUFFER, std430) buffer InputBufferU16 { uint16_t u16data[]; };
layout(binding = BINDING_INPUT_BUFFER, std430) buffer InputBufferU8 { uint8_t u8data[]; };
layout(binding = BINDING_INPUT_BUFFER, std430) buffer InputBufferU16 { uint16_t u16data[]; };
#endif
layout(binding = BINDING_INPUT_BUFFER, std430) buffer InputBufferU32 { uint u32data[]; };
layout(binding = BINDING_INPUT_BUFFER, std430) buffer InputBufferU64 { uvec2 u64data[]; };
@@ -90,20 +96,10 @@ const uint GOB_SIZE_Z_SHIFT = 0;
const uint GOB_SIZE_SHIFT = GOB_SIZE_X_SHIFT + GOB_SIZE_Y_SHIFT + GOB_SIZE_Z_SHIFT;
const uvec2 SWIZZLE_MASK = uvec2(GOB_SIZE_X - 1u, GOB_SIZE_Y - 1u);
uint SwizzleTable(uint pos) {
const uint t[8] = uint[](
0x12100200, 0x13110301, 0x16140604, 0x17150705,
0x1a180a08, 0x1b190b09, 0x1e1c0e0c, 0x1f1d0f0d
);
const uint i = pos >> 4;
const uint h = (t[i / 4] >> ((i % 4) * 8)) & 0xff;
return (h << 4) | (pos & 0xf);
}
// --- Helpers ---
uint SwizzleOffset(uvec2 pos) {
pos &= SWIZZLE_MASK;
return SwizzleTable(pos.y * 64u + pos.x);
return swizzle_table[pos.y * 64u + pos.x];
}
uvec4 ReadTexel(uint offset) {
@@ -1,6 +1,3 @@
// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
// SPDX-License-Identifier: GPL-3.0-or-later
// SPDX-FileCopyrightText: Copyright 2020 yuzu Emulator Project
// SPDX-License-Identifier: GPL-2.0-or-later
@@ -59,8 +56,10 @@ UtilShaders::UtilShaders(ProgramManager& program_manager_)
copy_bc4_program(MakeProgram(OPENGL_COPY_BC4_COMP)),
convert_s8d24_program(MakeProgram(OPENGL_CONVERT_S8D24_COMP)),
convert_ms_to_nonms_program(MakeProgram(CONVERT_MSAA_TO_NON_MSAA_COMP)),
convert_nonms_to_ms_program(MakeProgram(CONVERT_NON_MSAA_TO_MSAA_COMP))
{
convert_nonms_to_ms_program(MakeProgram(CONVERT_NON_MSAA_TO_MSAA_COMP)) {
const auto swizzle_table = Tegra::Texture::MakeSwizzleTable();
swizzle_table_buffer.Create();
glNamedBufferStorage(swizzle_table_buffer.handle, sizeof(swizzle_table), &swizzle_table, 0);
}
UtilShaders::~UtilShaders() = default;
@@ -117,11 +116,13 @@ void UtilShaders::ASTCDecode(Image& image, const StagingBufferMap& map,
void UtilShaders::BlockLinearUpload2D(Image& image, const StagingBufferMap& map,
std::span<const SwizzleParameters> swizzles) {
static constexpr Extent3D WORKGROUP_SIZE{32, 32, 1};
static constexpr GLuint BINDING_INPUT_BUFFER = 0;
static constexpr GLuint BINDING_SWIZZLE_BUFFER = 0;
static constexpr GLuint BINDING_INPUT_BUFFER = 1;
static constexpr GLuint BINDING_OUTPUT_IMAGE = 0;
program_manager.BindComputeProgram(block_linear_unswizzle_2d_program.handle);
glFlushMappedNamedBufferRange(map.buffer, map.offset, image.guest_size_bytes);
glBindBufferBase(GL_SHADER_STORAGE_BUFFER, BINDING_SWIZZLE_BUFFER, swizzle_table_buffer.handle);
const GLenum store_format = StoreFormat(BytesPerBlock(image.info.format));
for (const SwizzleParameters& swizzle : swizzles) {
@@ -152,11 +153,14 @@ void UtilShaders::BlockLinearUpload2D(Image& image, const StagingBufferMap& map,
void UtilShaders::BlockLinearUpload3D(Image& image, const StagingBufferMap& map,
std::span<const SwizzleParameters> swizzles) {
static constexpr Extent3D WORKGROUP_SIZE{16, 8, 8};
static constexpr GLuint BINDING_INPUT_BUFFER = 0;
static constexpr GLuint BINDING_SWIZZLE_BUFFER = 0;
static constexpr GLuint BINDING_INPUT_BUFFER = 1;
static constexpr GLuint BINDING_OUTPUT_IMAGE = 0;
glFlushMappedNamedBufferRange(map.buffer, map.offset, image.guest_size_bytes);
program_manager.BindComputeProgram(block_linear_unswizzle_3d_program.handle);
glBindBufferBase(GL_SHADER_STORAGE_BUFFER, BINDING_SWIZZLE_BUFFER, swizzle_table_buffer.handle);
const GLenum store_format = StoreFormat(BytesPerBlock(image.info.format));
for (const SwizzleParameters& swizzle : swizzles) {
@@ -1,6 +1,3 @@
// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
// SPDX-License-Identifier: GPL-3.0-or-later
// SPDX-FileCopyrightText: Copyright 2020 yuzu Emulator Project
// SPDX-License-Identifier: GPL-2.0-or-later
@@ -48,6 +45,9 @@ public:
private:
ProgramManager& program_manager;
OGLBuffer swizzle_table_buffer;
OGLProgram astc_decoder_program;
OGLProgram block_linear_unswizzle_2d_program;
OGLProgram block_linear_unswizzle_3d_program;
@@ -203,11 +203,7 @@ inline void PushImageDescriptors(TextureCache& texture_cache,
const VideoCommon::ImageViewId image_view_id{(views++)->id};
const VideoCommon::SamplerId sampler_id{*(samplers++)};
ImageView& image_view{texture_cache.GetImageView(image_view_id)};
VkImageView vk_image_view{image_view.Handle(desc.type)};
if (vk_image_view == VK_NULL_HANDLE) {
const VkImageView null_image_view{texture_cache.GetImageView(VideoCommon::NULL_IMAGE_VIEW_ID).Handle(desc.type)};
if (null_image_view != VK_NULL_HANDLE) vk_image_view = null_image_view;
}
const VkImageView vk_image_view{image_view.Handle(desc.type)};
const Sampler& sampler{texture_cache.GetSampler(sampler_id)};
const bool use_fallback_sampler{sampler.HasAddedAnisotropy() &&
!image_view.SupportsAnisotropy()};
@@ -653,8 +653,71 @@ void ASTCDecoderPass::Assemble(Image& image, const StagingBufferRef& map,
scheduler.Finish();
}
constexpr u32 BL3D_BINDING_INPUT_BUFFER = 0;
constexpr u32 BL3D_BINDING_OUTPUT_BUFFER = 1;
constexpr u32 BL3D_BINDING_SWIZZLE_TABLE = 0;
constexpr u32 BL3D_BINDING_INPUT_BUFFER = 1;
constexpr u32 BL3D_BINDING_OUTPUT_BUFFER = 2;
constexpr std::array<VkDescriptorSetLayoutBinding, 3> BL3D_DESCRIPTOR_SET_BINDINGS{{
{
.binding = BL3D_BINDING_SWIZZLE_TABLE,
.descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER, // swizzle_table[]
.descriptorCount = 1,
.stageFlags = VK_SHADER_STAGE_COMPUTE_BIT,
.pImmutableSamplers = nullptr,
},
{
.binding = BL3D_BINDING_INPUT_BUFFER,
.descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER, // block-linear input
.descriptorCount = 1,
.stageFlags = VK_SHADER_STAGE_COMPUTE_BIT,
.pImmutableSamplers = nullptr,
},
{
.binding = BL3D_BINDING_OUTPUT_BUFFER,
.descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER,
.descriptorCount = 1,
.stageFlags = VK_SHADER_STAGE_COMPUTE_BIT,
.pImmutableSamplers = nullptr,
},
}};
constexpr DescriptorBankInfo BL3D_BANK_INFO{
.uniform_buffers = 0,
.storage_buffers = 3,
.texture_buffers = 0,
.image_buffers = 0,
.textures = 0,
.images = 0,
.score = 3,
};
constexpr std::array<VkDescriptorUpdateTemplateEntry, 3>
BL3D_DESCRIPTOR_UPDATE_TEMPLATE_ENTRY{{
{
.dstBinding = BL3D_BINDING_SWIZZLE_TABLE,
.dstArrayElement = 0,
.descriptorCount = 1,
.descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER,
.offset = BL3D_BINDING_SWIZZLE_TABLE * sizeof(DescriptorUpdateEntry),
.stride = sizeof(DescriptorUpdateEntry),
},
{
.dstBinding = BL3D_BINDING_INPUT_BUFFER,
.dstArrayElement = 0,
.descriptorCount = 1,
.descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER,
.offset = BL3D_BINDING_INPUT_BUFFER * sizeof(DescriptorUpdateEntry),
.stride = sizeof(DescriptorUpdateEntry),
},
{
.dstBinding = BL3D_BINDING_OUTPUT_BUFFER,
.dstArrayElement = 0,
.descriptorCount = 1,
.descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER,
.offset = BL3D_BINDING_OUTPUT_BUFFER * sizeof(DescriptorUpdateEntry),
.stride = sizeof(DescriptorUpdateEntry),
}
}};
struct alignas(16) BlockLinearUnswizzle3DPushConstants {
u32 blocks_dim[3]; // Offset 0
@@ -682,50 +745,11 @@ BlockLinearUnswizzle3DPass::BlockLinearUnswizzle3DPass(
DescriptorPool& descriptor_pool_,
StagingBufferPool& staging_buffer_pool_,
ComputePassDescriptorQueue& compute_pass_descriptor_queue_)
: ComputePass(device_, scheduler_, descriptor_pool_,
std::array<VkDescriptorSetLayoutBinding, 2>{{
{
.binding = BL3D_BINDING_INPUT_BUFFER,
.descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER, // block-linear input
.descriptorCount = 1,
.stageFlags = VK_SHADER_STAGE_COMPUTE_BIT,
.pImmutableSamplers = nullptr,
},
{
.binding = BL3D_BINDING_OUTPUT_BUFFER,
.descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER,
.descriptorCount = 1,
.stageFlags = VK_SHADER_STAGE_COMPUTE_BIT,
.pImmutableSamplers = nullptr,
},
}},
std::array<VkDescriptorUpdateTemplateEntry, 2>{{
{
.dstBinding = BL3D_BINDING_INPUT_BUFFER,
.dstArrayElement = 0,
.descriptorCount = 1,
.descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER,
.offset = BL3D_BINDING_INPUT_BUFFER * sizeof(DescriptorUpdateEntry),
.stride = sizeof(DescriptorUpdateEntry),
},
{
.dstBinding = BL3D_BINDING_OUTPUT_BUFFER,
.dstArrayElement = 0,
.descriptorCount = 1,
.descriptorType = VK_DESCRIPTOR_TYPE_STORAGE_BUFFER,
.offset = BL3D_BINDING_OUTPUT_BUFFER * sizeof(DescriptorUpdateEntry),
.stride = sizeof(DescriptorUpdateEntry),
}
}},
DescriptorBankInfo{
.uniform_buffers = 0,
.storage_buffers = 2,
.texture_buffers = 0,
.image_buffers = 0,
.textures = 0,
.images = 0,
.score = 2,
},
: ComputePass(
device_, scheduler_, descriptor_pool_,
BL3D_DESCRIPTOR_SET_BINDINGS,
BL3D_DESCRIPTOR_UPDATE_TEMPLATE_ENTRY,
BL3D_BANK_INFO,
COMPUTE_PUSH_CONSTANT_RANGE<sizeof(BlockLinearUnswizzle3DPushConstants)>,
BLOCK_LINEAR_UNSWIZZLE_3D_BCN_COMP_SPV),
scheduler{scheduler_},
@@ -798,6 +822,8 @@ void BlockLinearUnswizzle3DPass::UnswizzleChunk(
pc.blocks_dim[2] = z_count; // Only process the count
compute_pass_descriptor_queue.Acquire(scheduler, 3);
compute_pass_descriptor_queue.AddBuffer(*image.runtime->swizzle_table_buffer, 0,
image.runtime->swizzle_table_size);
compute_pass_descriptor_queue.AddBuffer(swizzled.buffer,
sw.buffer_offset + swizzled.offset,
image.guest_size_bytes - sw.buffer_offset);
@@ -909,6 +909,40 @@ TextureCacheRuntime::TextureCacheRuntime(const Device& device_, Scheduler& sched
bl3d_unswizzle_pass.emplace(device, scheduler, descriptor_pool,
staging_buffer_pool, compute_pass_descriptor_queue);
}
// --- Create swizzle table buffer ---
{
auto table = Tegra::Texture::MakeSwizzleTable();
swizzle_table_size = static_cast<VkDeviceSize>(table.size() * sizeof(table[0]));
auto staging = staging_buffer_pool.Request(swizzle_table_size, MemoryUsage::Upload);
std::memcpy(staging.mapped_span.data(), table.data(), static_cast<size_t>(swizzle_table_size));
VkBufferCreateInfo ci{
.sType = VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO,
.size = swizzle_table_size,
.usage = VK_BUFFER_USAGE_STORAGE_BUFFER_BIT |
VK_BUFFER_USAGE_TRANSFER_DST_BIT |
VK_BUFFER_USAGE_TRANSFER_SRC_BIT,
.sharingMode = VK_SHARING_MODE_EXCLUSIVE,
};
swizzle_table_buffer = memory_allocator.CreateBuffer(ci, MemoryUsage::DeviceLocal);
scheduler.RequestOutsideRenderPassOperationContext();
scheduler.Record([staging_buf = staging.buffer,
dst_buf = *swizzle_table_buffer,
size = swizzle_table_size,
src_off = staging.offset](vk::CommandBuffer cmdbuf) {
const VkBufferCopy region{
.srcOffset = src_off,
.dstOffset = 0,
.size = size,
};
cmdbuf.CopyBuffer(staging_buf, dst_buf, region);
});
}
}
void TextureCacheRuntime::Finish() {
@@ -2280,29 +2314,46 @@ vk::ImageView ImageView::MakeView(VkFormat vk_format, VkImageAspectFlags aspect_
}
Sampler::Sampler(TextureCacheRuntime& runtime, const Tegra::Texture::TSCEntry& tsc) {
const auto& device = runtime.device;
const bool has_custom_border_extension = runtime.device.IsExtCustomBorderColorSupported();
const bool has_format_undefined =
has_custom_border_extension && runtime.device.IsCustomBorderColorWithoutFormatSupported();
const bool has_custom_border_colors =
has_format_undefined && runtime.device.IsCustomBorderColorsSupported();
const bool has_custom_border_ext = runtime.device.IsExtCustomBorderColorSupported();
const bool has_format_undefined = has_custom_border_ext && runtime.device.IsCustomBorderColorWithoutFormatSupported();
const bool has_custom_border_colors = has_format_undefined && runtime.device.IsCustomBorderColorsSupported();
const bool has_border_color_swizzle = has_custom_border_ext && runtime.device.IsExtBorderColorSwizzleSupported();
const auto color = tsc.BorderColor();
const void* pnext = nullptr;
const VkSamplerCustomBorderColorCreateInfoEXT border_ci{
.sType = VK_STRUCTURE_TYPE_SAMPLER_CUSTOM_BORDER_COLOR_CREATE_INFO_EXT,
.pNext = nullptr,
.pNext = pnext,
.customBorderColor = std::bit_cast<VkClearColorValue>(color),
.format = VK_FORMAT_UNDEFINED,
};
const void* pnext = nullptr;
if (has_custom_border_colors) {
pnext = &border_ci;
// Log extension usage for custom border color
if (GPU::Logging::IsActive()) {
GPU::Logging::GPULogger::GetInstance().LogExtensionUsage(
"VK_EXT_custom_border_color", "Sampler::Sampler");
GPU::Logging::GPULogger::GetInstance().LogExtensionUsage("VK_EXT_custom_border_color", "Sampler::Sampler");
}
}
const VkSamplerBorderColorComponentMappingCreateInfoEXT border_swizzle_ci{
.sType = VK_STRUCTURE_TYPE_SAMPLER_BORDER_COLOR_COMPONENT_MAPPING_CREATE_INFO_EXT,
.pNext = pnext,
.components = VkComponentMapping{
.r = VK_COMPONENT_SWIZZLE_IDENTITY,
.g = VK_COMPONENT_SWIZZLE_IDENTITY,
.b = VK_COMPONENT_SWIZZLE_IDENTITY,
.a = VK_COMPONENT_SWIZZLE_IDENTITY,
},
.srgb = tsc.srgb_conversion
};
if (has_border_color_swizzle) {
pnext = &border_swizzle_ci;
// Log extension usage for custom border color
if (GPU::Logging::IsActive()) {
GPU::Logging::GPULogger::GetInstance().LogExtensionUsage("VK_EXT_border_color_swizzle", "Sampler::Sampler");
}
}
const VkSamplerReductionModeCreateInfoEXT reduction_ci{
.sType = VK_STRUCTURE_TYPE_SAMPLER_REDUCTION_MODE_CREATE_INFO_EXT,
.pNext = pnext,
@@ -2313,36 +2364,33 @@ Sampler::Sampler(TextureCacheRuntime& runtime, const Tegra::Texture::TSCEntry& t
} else if (reduction_ci.reductionMode != VK_SAMPLER_REDUCTION_MODE_WEIGHTED_AVERAGE_EXT) {
LOG_WARNING(Render_Vulkan, "VK_EXT_sampler_filter_minmax is required");
}
// Some games have samplers with garbage. Sanitize them here.
const f32 max_anisotropy = std::clamp(tsc.MaxAnisotropy(), 1.0f, 16.0f);
const auto create_sampler = [&](const f32 anisotropy) {
return device.GetLogical().CreateSampler(VkSamplerCreateInfo{
return runtime.device.GetLogical().CreateSampler(VkSamplerCreateInfo{
.sType = VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO,
.pNext = pnext,
.flags = 0,
.magFilter = MaxwellToVK::Sampler::Filter(tsc.mag_filter),
.minFilter = MaxwellToVK::Sampler::Filter(tsc.min_filter),
.mipmapMode = MaxwellToVK::Sampler::MipmapMode(tsc.mipmap_filter),
.addressModeU = MaxwellToVK::Sampler::WrapMode(device, tsc.wrap_u, tsc.mag_filter),
.addressModeV = MaxwellToVK::Sampler::WrapMode(device, tsc.wrap_v, tsc.mag_filter),
.addressModeW = MaxwellToVK::Sampler::WrapMode(device, tsc.wrap_p, tsc.mag_filter),
.addressModeU = MaxwellToVK::Sampler::WrapMode(runtime.device, tsc.wrap_u, tsc.mag_filter),
.addressModeV = MaxwellToVK::Sampler::WrapMode(runtime.device, tsc.wrap_v, tsc.mag_filter),
.addressModeW = MaxwellToVK::Sampler::WrapMode(runtime.device, tsc.wrap_p, tsc.mag_filter),
.mipLodBias = tsc.LodBias(),
.anisotropyEnable = static_cast<VkBool32>(anisotropy > 1.0f ? VK_TRUE : VK_FALSE),
.anisotropyEnable = VkBool32(anisotropy > 1.0f ? VK_TRUE : VK_FALSE),
.maxAnisotropy = anisotropy,
.compareEnable = tsc.depth_compare_enabled,
.compareOp = MaxwellToVK::Sampler::DepthCompareFunction(tsc.depth_compare_func),
.minLod = tsc.mipmap_filter == TextureMipmapFilter::None ? 0.0f : tsc.MinLod(),
.maxLod = tsc.mipmap_filter == TextureMipmapFilter::None ? 0.25f : tsc.MaxLod(),
.borderColor = has_custom_border_colors ? VK_BORDER_COLOR_FLOAT_CUSTOM_EXT
: ConvertBorderColor(color),
.borderColor = has_custom_border_colors ? VK_BORDER_COLOR_FLOAT_CUSTOM_EXT : ConvertBorderColor(color),
.unnormalizedCoordinates = VK_FALSE,
});
};
sampler = create_sampler(max_anisotropy);
const f32 max_anisotropy_default = static_cast<f32>(1U << tsc.max_anisotropy);
const f32 max_anisotropy_default = f32(1U << tsc.max_anisotropy);
if (max_anisotropy > max_anisotropy_default) {
sampler_default_anisotropy = create_sampler(max_anisotropy_default);
}
@@ -2450,13 +2498,18 @@ void TextureCacheRuntime::AccelerateImageUpload(
if (!Settings::values.gpu_unswizzle_enabled.GetValue() || !bl3d_unswizzle_pass) {
if (IsPixelFormatBCn(image.info.format) && image.info.type == ImageType::e3D) {
ASSERT(false && "GPU unswizzle is disabled for BCn 3D texture");
ASSERT_MSG(false, "GPU unswizzle is disabled for BCn 3D texture");
}
ASSERT(false);
return;
}
if (bl3d_unswizzle_pass && IsPixelFormatBCn(image.info.format) && image.info.type == ImageType::e3D && image.info.resources.levels == 1 && image.info.resources.layers == 1) {
if (bl3d_unswizzle_pass &&
IsPixelFormatBCn(image.info.format) &&
image.info.type == ImageType::e3D &&
image.info.resources.levels == 1 &&
image.info.resources.layers == 1) {
return bl3d_unswizzle_pass->Unswizzle(image, map, swizzles, z_start, z_count);
}
@@ -1,4 +1,4 @@
// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
// SPDX-License-Identifier: GPL-3.0-or-later
// SPDX-FileCopyrightText: Copyright 2019 yuzu Emulator Project
@@ -130,6 +130,9 @@ public:
std::optional<ASTCDecoderPass> astc_decoder_pass;
std::optional<BlockLinearUnswizzle3DPass> bl3d_unswizzle_pass;
vk::Buffer swizzle_table_buffer;
VkDeviceSize swizzle_table_size = 0;
std::optional<MSAACopyPass> msaa_copy_pass;
const Settings::ResolutionScalingInfo& resolution;
std::array<std::vector<VkFormat>, VideoCore::Surface::MaxPixelFormat> view_formats;
+3 -2
View File
@@ -255,7 +255,8 @@ std::optional<u64> GenericEnvironment::TryFindSize() {
static constexpr u64 SELF_BRANCH_A = 0xE2400FFFFF87000FULL;
static constexpr u64 SELF_BRANCH_B = 0xE2400FFFFF07000FULL;
static constexpr u64 EXIT_VALUE = 0xE30000000007000FULL;
static constexpr u64 MESA_EXIT_MASK = 0xFFF00000000F001FULL;
static constexpr u64 MESA_EXIT_VALUE = (0xE30ULL << 52) | (0x7ULL << 16) | 0xFULL;
code.resize(MAXIMUM_SIZE / INST_SIZE);
@@ -270,7 +271,7 @@ std::optional<u64> GenericEnvironment::TryFindSize() {
if (inst == SELF_BRANCH_A || inst == SELF_BRANCH_B) {
return offset + index;
}
if (!is_proprietary_driver && inst == EXIT_VALUE) {
if ((inst & MESA_EXIT_MASK) == MESA_EXIT_VALUE) {
return offset + index + INST_SIZE;
}
}
+18 -3
View File
@@ -1,6 +1,3 @@
// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
// SPDX-License-Identifier: GPL-3.0-or-later
// SPDX-FileCopyrightText: Copyright 2018 yuzu Emulator Project
// SPDX-License-Identifier: GPL-2.0-or-later
@@ -26,6 +23,24 @@ constexpr u32 GOB_SIZE_SHIFT = GOB_SIZE_X_SHIFT + GOB_SIZE_Y_SHIFT + GOB_SIZE_Z_
constexpr u32 SWIZZLE_X_BITS = 0b100101111;
constexpr u32 SWIZZLE_Y_BITS = 0b011010000;
using SwizzleTable = std::array<std::array<u32, GOB_SIZE_X>, GOB_SIZE_Y>;
/**
* This table represents the internal swizzle of a gob, in format 16 bytes x 2 sector packing.
* Calculates the offset of an (x, y) position within a swizzled texture.
* Taken from the Tegra X1 Technical Reference Manual. pages 1187-1188
*/
constexpr SwizzleTable MakeSwizzleTable() {
SwizzleTable table{};
for (u32 y = 0; y < table.size(); ++y) {
for (u32 x = 0; x < table[0].size(); ++x) {
table[y][x] = ((x % 64) / 32) * 256 + ((y % 8) / 2) * 64 + ((x % 32) / 16) * 32 +
(y % 2) * 16 + (x % 16);
}
}
return table;
}
/// Unswizzles a block linear texture into linear memory.
void UnswizzleTexture(std::span<u8> output, std::span<const u8> input, u32 bytes_per_pixel,
u32 width, u32 height, u32 depth, u32 block_height, u32 block_depth,
+11 -8
View File
@@ -1,4 +1,4 @@
// SPDX-FileCopyrightText: Copyright 2025 Eden Emulator Project
// SPDX-FileCopyrightText: Copyright 2026 Eden Emulator Project
// SPDX-License-Identifier: GPL-3.0-or-later
// SPDX-FileCopyrightText: Copyright 2020 yuzu Emulator Project
@@ -55,13 +55,16 @@ namespace {
} // Anonymous namespace
std::array<float, 4> TSCEntry::BorderColor() const noexcept {
// TODO: Handle SRGB correctly. Using this breaks shadows in some games (Xenoblade).
// if (!srgb_conversion) {
// return border_color;
//}
// return {SRGB_CONVERSION_LUT[srgb_border_color_r], SRGB_CONVERSION_LUT[srgb_border_color_g],
// SRGB_CONVERSION_LUT[srgb_border_color_b], border_color[3]};
return border_color;
if (!srgb_conversion)
return border_color;
// SRGB will be handled by custom border color swizzle
// if not, we are in big, big trouble and may break shadows on Xenoblade
return {
f32(srgb_border_color_r) / 255.f,
f32(srgb_border_color_g) / 255.f,
f32(srgb_border_color_b) / 255.f,
border_color[3]
};
}
float TSCEntry::MaxAnisotropy() const noexcept {
@@ -1172,12 +1172,10 @@ bool Device::GetSuitability(bool requires_swapchain) {
void Device::RemoveUnsuitableExtensions() {
// VK_EXT_custom_border_color
if (extensions.custom_border_color) {
extensions.custom_border_color =
features.custom_border_color.customBorderColors &&
features.custom_border_color.customBorderColorWithoutFormat;
extensions.custom_border_color = features.custom_border_color.customBorderColors
&& features.custom_border_color.customBorderColorWithoutFormat;
}
RemoveExtensionFeatureIfUnsuitable(extensions.custom_border_color, features.custom_border_color,
VK_EXT_CUSTOM_BORDER_COLOR_EXTENSION_NAME);
RemoveExtensionFeatureIfUnsuitable(extensions.custom_border_color, features.custom_border_color, VK_EXT_CUSTOM_BORDER_COLOR_EXTENSION_NAME);
// VK_EXT_depth_bias_control
extensions.depth_bias_control =
@@ -50,6 +50,7 @@ VK_DEFINE_HANDLE(VmaAllocator)
// Define all features which may be used by the implementation and require an extension here.
#define FOR_EACH_VK_FEATURE_EXT(FEATURE) \
FEATURE(EXT, CustomBorderColor, CUSTOM_BORDER_COLOR, custom_border_color) \
FEATURE(EXT, BorderColorSwizzle, BORDER_COLOR_SWIZZLE, border_color_swizzle) \
FEATURE(EXT, DepthBiasControl, DEPTH_BIAS_CONTROL, depth_bias_control) \
FEATURE(EXT, DepthClipControl, DEPTH_CLIP_CONTROL, depth_clip_control) \
FEATURE(EXT, ExtendedDynamicState, EXTENDED_DYNAMIC_STATE, extended_dynamic_state) \
@@ -582,6 +583,11 @@ FN_MAX_LIMIT_LIST
return features.custom_border_color.customBorderColorWithoutFormat;
}
/// Returns true if the device supports VK_EXT_border_color_swizzle .
bool IsExtBorderColorSwizzleSupported() const {
return extensions.border_color_swizzle;
}
/// Returns true if the device supports VK_EXT_extended_dynamic_state.
bool IsExtExtendedDynamicStateSupported() const {
return extensions.extended_dynamic_state;