February 2024 - Wine-commits

Jinoh Kang : kernel32/tests: Fix console test with odd-sized consoles.
by Alexandre Julliard 22 Feb '24

22 Feb '24

Module: wine Branch: master Commit: 232b18d820e08fe6534249a64d3afdbf35d0b688 URL: https://gitlab.winehq.org/wine/wine/-/commit/232b18d820e08fe6534249a64d3afd… Author: Jinoh Kang <jinoh.kang.kr(a)gmail.com> Date: Tue Feb 20 20:45:44 2024 +0900 kernel32/tests: Fix console test with odd-sized consoles. --- dlls/kernel32/tests/process.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/dlls/kernel32/tests/process.c b/dlls/kernel32/tests/process.c index 2904a31d014..8fadc624e05 100644 --- a/dlls/kernel32/tests/process.c +++ b/dlls/kernel32/tests/process.c @@ -545,8 +545,8 @@ static void doChild(const char* file, const char* option) ok( ret, "Setting mode (%ld)\n", GetLastError()); ret = SetConsoleMode(hConOut, modeOut ^ 1); ok( ret, "Setting mode (%ld)\n", GetLastError()); - sbi.dwCursorPosition.X ^= 1; - sbi.dwCursorPosition.Y ^= 1; + sbi.dwCursorPosition.X = !sbi.dwCursorPosition.X; + sbi.dwCursorPosition.Y = !sbi.dwCursorPosition.Y; ret = SetConsoleCursorPosition(hConOut, sbi.dwCursorPosition); ok( ret, "Setting cursor position (%ld)\n", GetLastError()); }

1 0

Conor McCarthy : tests/shader-runner: Support structured buffer UAVs.
by Alexandre Julliard 22 Feb '24

22 Feb '24

Module: vkd3d Branch: master Commit: 081c9dbc96bdc8f567489ecfdb76b0fccfe4a697 URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/081c9dbc96bdc8f567489ecfdb76b… Author: Conor McCarthy <cmccarthy(a)codeweavers.com> Date: Thu Feb 1 16:01:45 2024 +1000 tests/shader-runner: Support structured buffer UAVs. --- tests/hlsl/uav-rwstructuredbuffer.shader_test | 12 ++++++++++++ tests/shader_runner.c | 7 +++++++ tests/shader_runner.h | 1 + tests/shader_runner_d3d11.c | 17 +++++++++-------- tests/shader_runner_d3d12.c | 1 + tests/shader_runner_gl.c | 1 + tests/shader_runner_vulkan.c | 4 ++++ 7 files changed, 35 insertions(+), 8 deletions(-) diff --git a/tests/hlsl/uav-rwstructuredbuffer.shader_test b/tests/hlsl/uav-rwstructuredbuffer.shader_test index f962e18e..bbe0ac95 100644 --- a/tests/hlsl/uav-rwstructuredbuffer.shader_test +++ b/tests/hlsl/uav-rwstructuredbuffer.shader_test @@ -1,6 +1,12 @@ [require] shader model >= 5.0 +[uav 2] +stride 16 +size (buffer, 2) + +0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 + [pixel shader todo] struct s { @@ -22,9 +28,15 @@ RWStructuredBuffer<struct s2> u5; float4 main() : sv_target1 { u[0] = float4(11.1, 12.2, 13.3, 14.4); + u[1] = float4(15.5, 16.6, 17.7, 18.8); return 0; } +[test] +todo draw quad +probe uav 2 (0) rgba (11.1, 12.2, 13.3, 14.4) +probe uav 2 (1) rgba (15.5, 16.6, 17.7, 18.8) + [pixel shader todo] struct s { diff --git a/tests/shader_runner.c b/tests/shader_runner.c index 85c88d7b..c20932cc 100644 --- a/tests/shader_runner.c +++ b/tests/shader_runner.c @@ -427,6 +427,13 @@ static void parse_resource_directive(struct resource_params *resource, const cha { resource->format = parse_format(line, &resource->data_type, &resource->texel_size, &line); } + else if (match_string(line, "stride", &line)) + { + if (sscanf(line, "%u", &resource->stride) < 1) + fatal_error("Malformed texture stride '%s'.\n", line); + resource->texel_size = resource->stride; + resource->format = DXGI_FORMAT_UNKNOWN; + } else if (match_string(line, "size", &line)) { if (sscanf(line, "( buffer , %u ) ", &resource->width) == 1) diff --git a/tests/shader_runner.h b/tests/shader_runner.h index 5ad6a278..62358d78 100644 --- a/tests/shader_runner.h +++ b/tests/shader_runner.h @@ -87,6 +87,7 @@ struct resource_params DXGI_FORMAT format; enum texture_data_type data_type; unsigned int texel_size; + unsigned int stride; unsigned int width, height; unsigned int level_count; uint8_t *data; diff --git a/tests/shader_runner_d3d11.c b/tests/shader_runner_d3d11.c index 0d13fc00..528c04e9 100644 --- a/tests/shader_runner_d3d11.c +++ b/tests/shader_runner_d3d11.c @@ -330,7 +330,8 @@ static void destroy_test_context(struct d3d11_shader_runner *runner) ok(!ref, "Device has %lu references left.\n", ref); } -static ID3D11Buffer *create_buffer(ID3D11Device *device, unsigned int bind_flags, unsigned int size, const void *data) +static ID3D11Buffer *create_buffer(ID3D11Device *device, unsigned int bind_flags, unsigned int size, + unsigned int stride, const void *data) { D3D11_SUBRESOURCE_DATA resource_data; D3D11_BUFFER_DESC buffer_desc; @@ -341,8 +342,8 @@ static ID3D11Buffer *create_buffer(ID3D11Device *device, unsigned int bind_flags buffer_desc.Usage = D3D11_USAGE_DEFAULT; buffer_desc.BindFlags = bind_flags; buffer_desc.CPUAccessFlags = 0; - buffer_desc.MiscFlags = 0; - buffer_desc.StructureByteStride = 0; + buffer_desc.MiscFlags = stride ? D3D11_RESOURCE_MISC_BUFFER_STRUCTURED : 0; + buffer_desc.StructureByteStride = stride; resource_data.pSysMem = data; resource_data.SysMemPitch = 0; @@ -417,7 +418,7 @@ static void init_resource_srv_buffer(struct d3d11_shader_runner *runner, struct ID3D11Device *device = runner->device; HRESULT hr; - resource->buffer = create_buffer(device, D3D11_BIND_SHADER_RESOURCE, params->data_size, params->data); + resource->buffer = create_buffer(device, D3D11_BIND_SHADER_RESOURCE, params->data_size, params->stride, params->data); resource->resource = (ID3D11Resource *)resource->buffer; srv_desc.Format = params->format; @@ -435,7 +436,7 @@ static void init_resource_uav_buffer(struct d3d11_shader_runner *runner, struct ID3D11Device *device = runner->device; HRESULT hr; - resource->buffer = create_buffer(device, D3D11_BIND_UNORDERED_ACCESS, params->data_size, params->data); + resource->buffer = create_buffer(device, D3D11_BIND_UNORDERED_ACCESS, params->data_size, params->stride, params->data); resource->resource = (ID3D11Resource *)resource->buffer; uav_desc.Format = params->format; @@ -474,7 +475,7 @@ static struct resource *d3d11_runner_create_resource(struct shader_runner *r, co break; case RESOURCE_TYPE_VERTEX_BUFFER: - resource->buffer = create_buffer(device, D3D11_BIND_VERTEX_BUFFER, params->data_size, params->data); + resource->buffer = create_buffer(device, D3D11_BIND_VERTEX_BUFFER, params->data_size, params->stride, params->data); resource->resource = (ID3D11Resource *)resource->buffer; break; } @@ -538,7 +539,7 @@ static bool d3d11_runner_dispatch(struct shader_runner *r, unsigned int x, unsig ID3D11Buffer *cb; cb = create_buffer(device, D3D11_BIND_CONSTANT_BUFFER, - runner->r.uniform_count * sizeof(*runner->r.uniforms), runner->r.uniforms); + runner->r.uniform_count * sizeof(*runner->r.uniforms), 0, runner->r.uniforms); ID3D11DeviceContext_CSSetConstantBuffers(context, 0, 1, &cb); ID3D11Buffer_Release(cb); } @@ -618,7 +619,7 @@ static bool d3d11_runner_draw(struct shader_runner *r, if (runner->r.uniform_count) { cb = create_buffer(device, D3D11_BIND_CONSTANT_BUFFER, - runner->r.uniform_count * sizeof(*runner->r.uniforms), runner->r.uniforms); + runner->r.uniform_count * sizeof(*runner->r.uniforms), 0, runner->r.uniforms); ID3D11DeviceContext_PSSetConstantBuffers(context, 0, 1, &cb); } diff --git a/tests/shader_runner_d3d12.c b/tests/shader_runner_d3d12.c index b7a29431..ca137512 100644 --- a/tests/shader_runner_d3d12.c +++ b/tests/shader_runner_d3d12.c @@ -200,6 +200,7 @@ static struct resource *d3d12_runner_create_resource(struct shader_runner *r, co uav_desc.Format = params->format; uav_desc.ViewDimension = D3D12_UAV_DIMENSION_BUFFER; uav_desc.Buffer.NumElements = params->width * params->height; + uav_desc.Buffer.StructureByteStride = params->stride; ID3D12Device_CreateUnorderedAccessView(device, resource->resource, NULL, &uav_desc, get_cpu_descriptor_handle(test_context, runner->heap, resource->r.slot + MAX_RESOURCES)); diff --git a/tests/shader_runner_gl.c b/tests/shader_runner_gl.c index f4e4fa75..263d1ba7 100644 --- a/tests/shader_runner_gl.c +++ b/tests/shader_runner_gl.c @@ -290,6 +290,7 @@ static const struct format_info *get_format_info(enum DXGI_FORMAT format) static const struct format_info format_info[] = { + {DXGI_FORMAT_UNKNOWN, 1, true, GL_R32UI, GL_RED_INTEGER, GL_UNSIGNED_INT}, {DXGI_FORMAT_R32G32B32A32_FLOAT, 4, false, GL_RGBA32F, GL_RGBA, GL_FLOAT}, {DXGI_FORMAT_R32G32B32A32_UINT, 4, true, GL_RGBA32UI, GL_RGBA_INTEGER, GL_UNSIGNED_INT}, {DXGI_FORMAT_R32G32_FLOAT, 2, false, GL_RG32F, GL_RG, GL_FLOAT}, diff --git a/tests/shader_runner_vulkan.c b/tests/shader_runner_vulkan.c index f9ab204f..35db729b 100644 --- a/tests/shader_runner_vulkan.c +++ b/tests/shader_runner_vulkan.c @@ -330,6 +330,10 @@ static void resource_init_buffer(struct vulkan_shader_runner *runner, struct vul else usage = VK_BUFFER_USAGE_UNIFORM_TEXEL_BUFFER_BIT; + /* d3d12 requires DXGI_FORMAT_UNKNOWN for structured buffers, but Vulkan requires a defined format. */ + if (format == VK_FORMAT_UNDEFINED && params->stride) + format = VK_FORMAT_R32_UINT; + resource->buffer = create_buffer(runner, params->data_size, usage, VK_MEMORY_PROPERTY_HOST_VISIBLE_BIT, &resource->memory); resource->buffer_view = create_buffer_view(runner, resource->buffer, format);

1 0

Conor McCarthy : tests/shader-runner: Add a test for a signed int typed buffer UAV.
by Alexandre Julliard 22 Feb '24

22 Feb '24

Module: vkd3d Branch: master Commit: 9180ea6591b0566e5040628822311fa334699317 URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/9180ea6591b0566e5040628822311… Author: Conor McCarthy <cmccarthy(a)codeweavers.com> Date: Thu Feb 1 16:02:58 2024 +1000 tests/shader-runner: Add a test for a signed int typed buffer UAV. Check for backend type mismatches resulting from the absence of signedness in SM 6. --- tests/hlsl/uav-rwbuffer.shader_test | 20 ++++++++++++++++++++ tests/shader_runner.c | 24 ++++++++++++++++++------ tests/shader_runner_gl.c | 1 + tests/utils.h | 1 + 4 files changed, 40 insertions(+), 6 deletions(-) diff --git a/tests/hlsl/uav-rwbuffer.shader_test b/tests/hlsl/uav-rwbuffer.shader_test index ec8937b3..c813d0d1 100644 --- a/tests/hlsl/uav-rwbuffer.shader_test +++ b/tests/hlsl/uav-rwbuffer.shader_test @@ -164,6 +164,26 @@ float4 main() : sv_target1 return 0; } +[uav 1] +format r32g32b32a32 sint +size (buffer, 2) +1 2 3 4 5 6 7 8 + +[pixel shader] +RWBuffer<int4> u : register(u1); + +float4 main() : sv_target +{ + u[0] = int4(11, -12, 13, -14); + u[1] = int4(-15, 16, -17, 18); + return 0; +} + +[test] +todo(sm>=6) draw quad +probe uav 1 (0) rgbai (11, -12, 13, -14) +probe uav 1 (1) rgbai (-15, 16, -17, 18) + [uav 2] size (buffer, 1) 0.1 0.2 0.3 0.4 diff --git a/tests/shader_runner.c b/tests/shader_runner.c index c20932cc..d0d80650 100644 --- a/tests/shader_runner.c +++ b/tests/shader_runner.c @@ -336,6 +336,7 @@ static DXGI_FORMAT parse_format(const char *line, enum texture_data_type *data_t formats[] = { {"r32g32b32a32 float", TEXTURE_DATA_FLOAT, 16, DXGI_FORMAT_R32G32B32A32_FLOAT}, + {"r32g32b32a32 sint", TEXTURE_DATA_SINT, 16, DXGI_FORMAT_R32G32B32A32_SINT}, {"r32g32b32a32 uint", TEXTURE_DATA_UINT, 16, DXGI_FORMAT_R32G32B32A32_UINT}, {"r32g32 float", TEXTURE_DATA_FLOAT, 8, DXGI_FORMAT_R32G32_FLOAT}, {"r32g32 int", TEXTURE_DATA_SINT, 8, DXGI_FORMAT_R32G32_SINT}, @@ -622,12 +623,12 @@ static void read_uint(const char **line, unsigned int *u, bool is_uniform) *line = rest + (!is_uniform && *rest == ','); } -static void read_int4(const char **line, struct ivec4 *v) +static void read_int4(const char **line, struct ivec4 *v, bool is_uniform) { - read_int(line, &v->x, true); - read_int(line, &v->y, true); - read_int(line, &v->z, true); - read_int(line, &v->w, true); + read_int(line, &v->x, is_uniform); + read_int(line, &v->y, is_uniform); + read_int(line, &v->z, is_uniform); + read_int(line, &v->w, is_uniform); } static void read_uint4(const char **line, struct uvec4 *v, bool is_uniform) @@ -907,6 +908,17 @@ static void parse_test_directive(struct shader_runner *runner, const char *line) line = close_parentheses(line); todo_if(runner->is_todo) check_readback_data_uvec4(rb, &rect, &v); } + else if (match_string(line, "rgbai", &line)) + { + struct ivec4 v; + + if (*line != '(') + fatal_error("Malformed probe arguments '%s'.\n", line); + ++line; + read_int4(&line, &v, false); + line = close_parentheses(line); + todo_if(runner->is_todo) check_readback_data_ivec4(rb, &rect, &v); + } else if (match_string(line, "rgba", &line)) { struct vec4 v; @@ -993,7 +1005,7 @@ static void parse_test_directive(struct shader_runner *runner, const char *line) { struct ivec4 v; - read_int4(&line, &v); + read_int4(&line, &v, true); set_uniforms(runner, offset, 4, &v); } else if (match_string(line, "uint4", &line)) diff --git a/tests/shader_runner_gl.c b/tests/shader_runner_gl.c index 263d1ba7..6cec33a4 100644 --- a/tests/shader_runner_gl.c +++ b/tests/shader_runner_gl.c @@ -293,6 +293,7 @@ static const struct format_info *get_format_info(enum DXGI_FORMAT format) {DXGI_FORMAT_UNKNOWN, 1, true, GL_R32UI, GL_RED_INTEGER, GL_UNSIGNED_INT}, {DXGI_FORMAT_R32G32B32A32_FLOAT, 4, false, GL_RGBA32F, GL_RGBA, GL_FLOAT}, {DXGI_FORMAT_R32G32B32A32_UINT, 4, true, GL_RGBA32UI, GL_RGBA_INTEGER, GL_UNSIGNED_INT}, + {DXGI_FORMAT_R32G32B32A32_SINT, 4, true, GL_RGBA32I, GL_RGBA_INTEGER, GL_INT}, {DXGI_FORMAT_R32G32_FLOAT, 2, false, GL_RG32F, GL_RG, GL_FLOAT}, {DXGI_FORMAT_R32G32_UINT, 2, true, GL_RG32UI, GL_RG_INTEGER, GL_UNSIGNED_INT}, {DXGI_FORMAT_R32G32_SINT, 2, true, GL_RG32I, GL_RG_INTEGER, GL_INT}, diff --git a/tests/utils.h b/tests/utils.h index 98dceba8..cd5215fe 100644 --- a/tests/utils.h +++ b/tests/utils.h @@ -275,6 +275,7 @@ static inline void check_readback_data_vec4_(unsigned int line, const struct res got.x, got.y, got.z, got.w, expected->x, expected->y, expected->z, expected->w, x, y); } +#define check_readback_data_ivec4(a, b, c) check_readback_data_uvec4_(__LINE__, a, b, (const struct uvec4 *)(c)) #define check_readback_data_uvec4(a, b, c) check_readback_data_uvec4_(__LINE__, a, b, c) static inline void check_readback_data_uvec4_(unsigned int line, const struct resource_readback *rb, const RECT *rect, const struct uvec4 *expected)

1 0

Conor McCarthy : tests/shader-runner: Add a test for a signed int structured buffer UAV.
by Alexandre Julliard 22 Feb '24

22 Feb '24

Module: vkd3d Branch: master Commit: b447a725ecaa4cc17b1f3a8e6730ce4c5ee56cd4 URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/b447a725ecaa4cc17b1f3a8e6730c… Author: Conor McCarthy <cmccarthy(a)codeweavers.com> Date: Mon Feb 12 11:44:36 2024 +1000 tests/shader-runner: Add a test for a signed int structured buffer UAV. Check for backend type mismatches resulting from the absence of signedness in SM 6. --- tests/hlsl/uav-rwstructuredbuffer.shader_test | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/tests/hlsl/uav-rwstructuredbuffer.shader_test b/tests/hlsl/uav-rwstructuredbuffer.shader_test index bbe0ac95..93864f0f 100644 --- a/tests/hlsl/uav-rwstructuredbuffer.shader_test +++ b/tests/hlsl/uav-rwstructuredbuffer.shader_test @@ -135,3 +135,24 @@ float4 main() : sv_target u[0].x = float3(30.0, 40.0, 50.0); return 0; } + +[uav 2] +stride 16 +size (buffer, 2) + +1 2 3 4 5 6 7 8 + +[pixel shader todo] +RWStructuredBuffer<int4> u : register(u2); + +float4 main() : sv_target1 +{ + u[0] = int4(11, -12, 13, -14); + u[1] = int4(-15, 16, -17, 18); + return 0; +} + +[test] +todo draw quad +probe uav 2 (0) rgbai (11, -12, 13, -14) +probe uav 2 (1) rgbai (-15, 16, -17, 18)

1 0

Conor McCarthy : vkd3d-shader/dxil: Implement DX intrinsic BufferStore.
by Alexandre Julliard 22 Feb '24

22 Feb '24

Module: vkd3d Branch: master Commit: ce1875c4351a86a275d9fe281e7fc5687eaba6e4 URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/ce1875c4351a86a275d9fe281e7fc… Author: Conor McCarthy <cmccarthy(a)codeweavers.com> Date: Thu Feb 1 16:04:28 2024 +1000 vkd3d-shader/dxil: Implement DX intrinsic BufferStore. --- libs/vkd3d-shader/dxil.c | 63 +++++++++++++++++++++++++++ tests/hlsl/uav-rwbuffer.shader_test | 4 +- tests/hlsl/uav-rwstructuredbuffer.shader_test | 12 ++--- 3 files changed, 71 insertions(+), 8 deletions(-) diff --git a/libs/vkd3d-shader/dxil.c b/libs/vkd3d-shader/dxil.c index 2ca3aa95..d461fad2 100644 --- a/libs/vkd3d-shader/dxil.c +++ b/libs/vkd3d-shader/dxil.c @@ -383,6 +383,7 @@ enum dx_intrinsic_opcode DX_TEXTURE_LOAD = 66, DX_TEXTURE_STORE = 67, DX_BUFFER_LOAD = 68, + DX_BUFFER_STORE = 69, DX_ATOMIC_BINOP = 78, DX_ATOMIC_CMP_XCHG = 79, DX_DERIV_COARSEX = 83, @@ -4179,6 +4180,67 @@ static void sm6_parser_emit_dx_buffer_load(struct sm6_parser *sm6, enum dx_intri instruction_dst_param_init_ssa_vector(ins, VKD3D_VEC4_SIZE, sm6); } +static void sm6_parser_emit_dx_buffer_store(struct sm6_parser *sm6, enum dx_intrinsic_opcode op, + const struct sm6_value **operands, struct function_emission_state *state) +{ + struct vkd3d_shader_src_param *src_params; + struct vkd3d_shader_dst_param *dst_param; + unsigned int write_mask, component_count; + struct vkd3d_shader_instruction *ins; + struct vkd3d_shader_register texel; + const struct sm6_value *resource; + + resource = operands[0]; + if (!sm6_value_validate_is_handle(resource, sm6)) + return; + + if (resource->u.handle.d->kind != RESOURCE_KIND_TYPEDBUFFER) + { + WARN("Resource is not a typed buffer.\n"); + vkd3d_shader_parser_warning(&sm6->p, VKD3D_SHADER_WARNING_DXIL_INVALID_OPERATION, + "Resource for a typed buffer store is not a typed buffer."); + } + + write_mask = sm6_value_get_constant_uint(operands[7]); + if (!write_mask || write_mask > VKD3DSP_WRITEMASK_ALL) + { + WARN("Invalid write mask %#x.\n", write_mask); + vkd3d_shader_parser_error(&sm6->p, VKD3D_SHADER_ERROR_DXIL_INVALID_OPERAND, + "Write mask %#x for a typed buffer store operation is invalid.", write_mask); + return; + } + else if (write_mask & (write_mask + 1)) + { + /* In this case, it is unclear which source operands will be defined unless we encounter it in a shader. */ + FIXME("Unhandled write mask %#x.\n", write_mask); + vkd3d_shader_parser_error(&sm6->p, VKD3D_SHADER_ERROR_DXIL_INVALID_OPERAND, + "Write mask %#x for a typed buffer store operation is unhandled.", write_mask); + } + component_count = vsir_write_mask_component_count(write_mask); + + if (!sm6_parser_emit_composite_construct(sm6, &operands[3], component_count, state, &texel)) + return; + + ins = state->ins; + vsir_instruction_init(ins, &sm6->p.location, VKD3DSIH_STORE_UAV_TYPED); + + if (!(src_params = instruction_src_params_alloc(ins, 2, sm6))) + return; + src_param_init_from_value(&src_params[0], operands[1]); + if (!sm6_value_is_undef(operands[2])) + { + /* Constant zero would have no effect, but is not worth checking for unless it shows up. */ + WARN("Ignoring structure offset.\n"); + vkd3d_shader_parser_warning(&sm6->p, VKD3D_SHADER_WARNING_DXIL_IGNORING_OPERANDS, + "Ignoring structure offset for a typed buffer store."); + } + src_param_init_vector_from_reg(&src_params[1], &texel); + + dst_param = instruction_dst_params_alloc(ins, 1, sm6); + dst_param_init_with_mask(dst_param, write_mask); + dst_param->reg = resource->u.handle.reg; +} + static unsigned int sm6_value_get_texel_offset(const struct sm6_value *value) { return sm6_value_is_undef(value) ? 0 : sm6_value_get_constant_uint(value); @@ -4478,6 +4540,7 @@ static const struct sm6_dx_opcode_info sm6_dx_op_table[] = [DX_ATOMIC_CMP_XCHG ] = {"o", "HiiiRR", sm6_parser_emit_dx_atomic_binop}, [DX_BFREV ] = {"m", "R", sm6_parser_emit_dx_unary}, [DX_BUFFER_LOAD ] = {"o", "Hii", sm6_parser_emit_dx_buffer_load}, + [DX_BUFFER_STORE ] = {"v", "Hiiooooc", sm6_parser_emit_dx_buffer_store}, [DX_CBUFFER_LOAD_LEGACY ] = {"o", "Hi", sm6_parser_emit_dx_cbuffer_load}, [DX_COS ] = {"g", "R", sm6_parser_emit_dx_sincos}, [DX_COUNT_BITS ] = {"i", "m", sm6_parser_emit_dx_unary}, diff --git a/tests/hlsl/uav-rwbuffer.shader_test b/tests/hlsl/uav-rwbuffer.shader_test index c813d0d1..bcd5c695 100644 --- a/tests/hlsl/uav-rwbuffer.shader_test +++ b/tests/hlsl/uav-rwbuffer.shader_test @@ -180,7 +180,7 @@ float4 main() : sv_target } [test] -todo(sm>=6) draw quad +draw quad probe uav 1 (0) rgbai (11, -12, 13, -14) probe uav 1 (1) rgbai (-15, 16, -17, 18) @@ -198,5 +198,5 @@ float4 main() : sv_target1 } [test] -todo(sm>=6) draw quad +draw quad probe uav 2 (0) rgba (11.1, 12.2, 13.3, 14.4) diff --git a/tests/hlsl/uav-rwstructuredbuffer.shader_test b/tests/hlsl/uav-rwstructuredbuffer.shader_test index 93864f0f..d2fe7a3f 100644 --- a/tests/hlsl/uav-rwstructuredbuffer.shader_test +++ b/tests/hlsl/uav-rwstructuredbuffer.shader_test @@ -33,9 +33,9 @@ float4 main() : sv_target1 } [test] -todo draw quad -probe uav 2 (0) rgba (11.1, 12.2, 13.3, 14.4) -probe uav 2 (1) rgba (15.5, 16.6, 17.7, 18.8) +todo(sm<6) draw quad +todo probe uav 2 (0) rgba (11.1, 12.2, 13.3, 14.4) +todo probe uav 2 (1) rgba (15.5, 16.6, 17.7, 18.8) [pixel shader todo] struct s @@ -153,6 +153,6 @@ float4 main() : sv_target1 } [test] -todo draw quad -probe uav 2 (0) rgbai (11, -12, 13, -14) -probe uav 2 (1) rgbai (-15, 16, -17, 18) +todo(sm<6) draw quad +todo probe uav 2 (0) rgbai (11, -12, 13, -14) +todo probe uav 2 (1) rgbai (-15, 16, -17, 18)

1 0

Conor McCarthy : vkd3d-shader/dxil: Handle raw and structured buffers in sm6_parser_emit_dx_buffer_store().
by Alexandre Julliard 22 Feb '24

22 Feb '24

Module: vkd3d Branch: master Commit: cab8b781ab39f1b917b2ce668cecf44b5cc90a53 URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/cab8b781ab39f1b917b2ce668cecf… Author: Conor McCarthy <cmccarthy(a)codeweavers.com> Date: Thu Feb 1 16:05:35 2024 +1000 vkd3d-shader/dxil: Handle raw and structured buffers in sm6_parser_emit_dx_buffer_store(). --- libs/vkd3d-shader/dxil.c | 57 +++++++++++++++++++++++++++ tests/hlsl/uav-rwstructuredbuffer.shader_test | 8 ++-- 2 files changed, 61 insertions(+), 4 deletions(-) diff --git a/libs/vkd3d-shader/dxil.c b/libs/vkd3d-shader/dxil.c index d461fad2..5d98a38c 100644 --- a/libs/vkd3d-shader/dxil.c +++ b/libs/vkd3d-shader/dxil.c @@ -4138,6 +4138,57 @@ static void sm6_parser_emit_dx_raw_buffer_load(struct sm6_parser *sm6, enum dx_i instruction_dst_param_init_ssa_vector(ins, component_count, sm6); } +static void sm6_parser_emit_dx_raw_buffer_store(struct sm6_parser *sm6, enum dx_intrinsic_opcode op, + const struct sm6_value **operands, struct function_emission_state *state) +{ + unsigned int write_mask, component_count, operand_count; + struct vkd3d_shader_src_param *src_params; + struct vkd3d_shader_dst_param *dst_param; + struct vkd3d_shader_instruction *ins; + struct vkd3d_shader_register data; + const struct sm6_value *resource; + bool raw; + + resource = operands[0]; + if (!sm6_value_validate_is_handle(resource, sm6)) + return; + raw = resource->u.handle.d->kind == RESOURCE_KIND_RAWBUFFER; + + write_mask = sm6_value_get_constant_uint(operands[7]); + if (!write_mask || write_mask > VKD3DSP_WRITEMASK_ALL) + { + WARN("Invalid write mask %#x.\n", write_mask); + vkd3d_shader_parser_error(&sm6->p, VKD3D_SHADER_ERROR_DXIL_INVALID_OPERAND, + "Write mask %#x for a raw/structured buffer store operation is invalid.", write_mask); + return; + } + else if (write_mask & (write_mask + 1)) + { + /* In this case, it is unclear which source operands will be defined unless we encounter it in a shader. */ + FIXME("Unhandled write mask %#x.\n", write_mask); + vkd3d_shader_parser_error(&sm6->p, VKD3D_SHADER_ERROR_DXIL_INVALID_OPERAND, + "Write mask %#x for a raw/structured buffer store operation is unhandled.", write_mask); + } + component_count = vsir_write_mask_component_count(write_mask); + + if (!sm6_parser_emit_composite_construct(sm6, &operands[3], component_count, state, &data)) + return; + + ins = state->ins; + vsir_instruction_init(ins, &sm6->p.location, raw ? VKD3DSIH_STORE_RAW : VKD3DSIH_STORE_STRUCTURED); + operand_count = 2 + !raw; + + if (!(src_params = instruction_src_params_alloc(ins, operand_count, sm6))) + return; + src_params_init_from_operands(src_params, &operands[1], operand_count - 1); + data.data_type = VKD3D_DATA_UINT; + src_param_init_vector_from_reg(&src_params[operand_count - 1], &data); + + dst_param = instruction_dst_params_alloc(ins, 1, sm6); + dst_param_init_with_mask(dst_param, write_mask); + dst_param->reg = resource->u.handle.reg; +} + static void sm6_parser_emit_dx_buffer_load(struct sm6_parser *sm6, enum dx_intrinsic_opcode op, const struct sm6_value **operands, struct function_emission_state *state) { @@ -4194,6 +4245,12 @@ static void sm6_parser_emit_dx_buffer_store(struct sm6_parser *sm6, enum dx_intr if (!sm6_value_validate_is_handle(resource, sm6)) return; + if (resource->u.handle.d->kind == RESOURCE_KIND_RAWBUFFER + || resource->u.handle.d->kind == RESOURCE_KIND_STRUCTUREDBUFFER) + { + return sm6_parser_emit_dx_raw_buffer_store(sm6, op, operands, state); + } + if (resource->u.handle.d->kind != RESOURCE_KIND_TYPEDBUFFER) { WARN("Resource is not a typed buffer.\n"); diff --git a/tests/hlsl/uav-rwstructuredbuffer.shader_test b/tests/hlsl/uav-rwstructuredbuffer.shader_test index d2fe7a3f..15026d06 100644 --- a/tests/hlsl/uav-rwstructuredbuffer.shader_test +++ b/tests/hlsl/uav-rwstructuredbuffer.shader_test @@ -34,8 +34,8 @@ float4 main() : sv_target1 [test] todo(sm<6) draw quad -todo probe uav 2 (0) rgba (11.1, 12.2, 13.3, 14.4) -todo probe uav 2 (1) rgba (15.5, 16.6, 17.7, 18.8) +probe uav 2 (0) rgba (11.1, 12.2, 13.3, 14.4) +probe uav 2 (1) rgba (15.5, 16.6, 17.7, 18.8) [pixel shader todo] struct s @@ -154,5 +154,5 @@ float4 main() : sv_target1 [test] todo(sm<6) draw quad -todo probe uav 2 (0) rgbai (11, -12, 13, -14) -todo probe uav 2 (1) rgbai (-15, 16, -17, 18) +probe uav 2 (0) rgbai (11, -12, 13, -14) +probe uav 2 (1) rgbai (-15, 16, -17, 18)

1 0

Conor McCarthy : vkd3d-shader/dxil: Implement DX intrinsic RawBufferStore.
by Alexandre Julliard 22 Feb '24

22 Feb '24

Module: vkd3d Branch: master Commit: 1dffc281c95103d774faad5a1f67f1c6ff948f10 URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/1dffc281c95103d774faad5a1f67f… Author: Conor McCarthy <cmccarthy(a)codeweavers.com> Date: Fri Feb 9 14:14:32 2024 +1000 vkd3d-shader/dxil: Implement DX intrinsic RawBufferStore. --- libs/vkd3d-shader/dxil.c | 23 ++++++++++++++++++++++- 1 file changed, 22 insertions(+), 1 deletion(-) diff --git a/libs/vkd3d-shader/dxil.c b/libs/vkd3d-shader/dxil.c index 5d98a38c..93ed3e06 100644 --- a/libs/vkd3d-shader/dxil.c +++ b/libs/vkd3d-shader/dxil.c @@ -394,6 +394,7 @@ enum dx_intrinsic_opcode DX_LEGACY_F32TOF16 = 130, DX_LEGACY_F16TOF32 = 131, DX_RAW_BUFFER_LOAD = 139, + DX_RAW_BUFFER_STORE = 140, }; enum dxil_cast_code @@ -4141,7 +4142,7 @@ static void sm6_parser_emit_dx_raw_buffer_load(struct sm6_parser *sm6, enum dx_i static void sm6_parser_emit_dx_raw_buffer_store(struct sm6_parser *sm6, enum dx_intrinsic_opcode op, const struct sm6_value **operands, struct function_emission_state *state) { - unsigned int write_mask, component_count, operand_count; + unsigned int write_mask, component_count, alignment = 0, operand_count; struct vkd3d_shader_src_param *src_params; struct vkd3d_shader_dst_param *dst_param; struct vkd3d_shader_instruction *ins; @@ -4171,6 +4172,24 @@ static void sm6_parser_emit_dx_raw_buffer_store(struct sm6_parser *sm6, enum dx_ } component_count = vsir_write_mask_component_count(write_mask); + if (op == DX_RAW_BUFFER_STORE) + { + if (!raw && resource->u.handle.d->kind != RESOURCE_KIND_STRUCTUREDBUFFER) + { + WARN("Resource is not a raw or structured buffer.\n"); + vkd3d_shader_parser_warning(&sm6->p, VKD3D_SHADER_WARNING_DXIL_INVALID_OPERATION, + "Resource for a raw buffer store is not a raw or structured buffer."); + } + + alignment = sm6_value_get_constant_uint(operands[8]); + if (alignment & (alignment - 1)) + { + FIXME("Invalid alignment %#x.\n", alignment); + vkd3d_shader_parser_error(&sm6->p, VKD3D_SHADER_ERROR_DXIL_INVALID_OPERAND, + "Alignment %#x for a raw/structured buffer store operation is invalid.", alignment); + } + } + if (!sm6_parser_emit_composite_construct(sm6, &operands[3], component_count, state, &data)) return; @@ -4187,6 +4206,7 @@ static void sm6_parser_emit_dx_raw_buffer_store(struct sm6_parser *sm6, enum dx_ dst_param = instruction_dst_params_alloc(ins, 1, sm6); dst_param_init_with_mask(dst_param, write_mask); dst_param->reg = resource->u.handle.reg; + dst_param->reg.alignment = alignment; } static void sm6_parser_emit_dx_buffer_load(struct sm6_parser *sm6, enum dx_intrinsic_opcode op, @@ -4627,6 +4647,7 @@ static const struct sm6_dx_opcode_info sm6_dx_op_table[] = [DX_LOAD_INPUT ] = {"o", "ii8i", sm6_parser_emit_dx_load_input}, [DX_LOG ] = {"g", "R", sm6_parser_emit_dx_unary}, [DX_RAW_BUFFER_LOAD ] = {"o", "Hii8i", sm6_parser_emit_dx_raw_buffer_load}, + [DX_RAW_BUFFER_STORE ] = {"v", "Hiioooocc", sm6_parser_emit_dx_raw_buffer_store}, [DX_ROUND_NE ] = {"g", "R", sm6_parser_emit_dx_unary}, [DX_ROUND_NI ] = {"g", "R", sm6_parser_emit_dx_unary}, [DX_ROUND_PI ] = {"g", "R", sm6_parser_emit_dx_unary},

1 0

Alex Henrie : include: Add debugstr_time to wine/strmbase.h.
by Alexandre Julliard 22 Feb '24

22 Feb '24

Module: wine Branch: master Commit: 2e9a57a0b823140f8cf514ac7bf8812b0f15f15d URL: https://gitlab.winehq.org/wine/wine/-/commit/2e9a57a0b823140f8cf514ac7bf881… Author: Alex Henrie <alexhenrie24(a)gmail.com> Date: Sat Feb 17 23:11:27 2024 -0700 include: Add debugstr_time to wine/strmbase.h. --- dlls/evr/evr_private.h | 21 --------------------- dlls/qasf/asfreader.c | 21 --------------------- dlls/quartz/quartz_private.h | 21 --------------------- dlls/winegstreamer/gst_private.h | 21 --------------------- include/wine/strmbase.h | 22 ++++++++++++++++++++++ libs/strmbase/strmbase_private.h | 21 --------------------- 6 files changed, 22 insertions(+), 105 deletions(-) diff --git a/dlls/evr/evr_private.h b/dlls/evr/evr_private.h index 225e60348b4..c24dd01a745 100644 --- a/dlls/evr/evr_private.h +++ b/dlls/evr/evr_private.h @@ -24,27 +24,6 @@ #include "wine/strmbase.h" #include "wine/debug.h" -static inline const char *debugstr_time(LONGLONG time) -{ - ULONGLONG abstime = time >= 0 ? time : -time; - unsigned int i = 0, j = 0; - char buffer[23], rev[23]; - - while (abstime || i <= 8) - { - buffer[i++] = '0' + (abstime % 10); - abstime /= 10; - if (i == 7) buffer[i++] = '.'; - } - if (time < 0) buffer[i++] = '-'; - - while (i--) rev[j++] = buffer[i]; - while (rev[j-1] == '0' && rev[j-2] != '.') --j; - rev[j] = 0; - - return wine_dbg_sprintf("%s", rev); -} - static inline const char *debugstr_normalized_rect(const MFVideoNormalizedRect *rect) { if (!rect) return "(null)"; diff --git a/dlls/qasf/asfreader.c b/dlls/qasf/asfreader.c index 4cf077e9dd3..7a832e9a718 100644 --- a/dlls/qasf/asfreader.c +++ b/dlls/qasf/asfreader.c @@ -27,27 +27,6 @@ WINE_DEFAULT_DEBUG_CHANNEL(quartz); -static inline const char *debugstr_time(REFERENCE_TIME time) -{ - ULONGLONG abstime = time >= 0 ? time : -time; - unsigned int i = 0, j = 0; - char buffer[23], rev[23]; - - while (abstime || i <= 8) - { - buffer[i++] = '0' + (abstime % 10); - abstime /= 10; - if (i == 7) buffer[i++] = '.'; - } - if (time < 0) buffer[i++] = '-'; - - while (i--) rev[j++] = buffer[i]; - while (rev[j-1] == '0' && rev[j-2] != '.') --j; - rev[j] = 0; - - return wine_dbg_sprintf("%s", rev); -} - struct buffer { INSSBuffer INSSBuffer_iface; diff --git a/dlls/quartz/quartz_private.h b/dlls/quartz/quartz_private.h index 8a761e540b9..946709f936a 100644 --- a/dlls/quartz/quartz_private.h +++ b/dlls/quartz/quartz_private.h @@ -44,27 +44,6 @@ 0, 0, { (DWORD_PTR)(__FILE__ ": " # cs) }}; \ static CRITICAL_SECTION cs = { &cs##_debug, -1, 0, 0, 0, 0 }; -static inline const char *debugstr_time(REFERENCE_TIME time) -{ - ULONGLONG abstime = time >= 0 ? time : -time; - unsigned int i = 0, j = 0; - char buffer[23], rev[23]; - - while (abstime || i <= 8) - { - buffer[i++] = '0' + (abstime % 10); - abstime /= 10; - if (i == 7) buffer[i++] = '.'; - } - if (time < 0) buffer[i++] = '-'; - - while (i--) rev[j++] = buffer[i]; - while (rev[j-1] == '0' && rev[j-2] != '.') --j; - rev[j] = 0; - - return wine_dbg_sprintf("%s", rev); -} - /* see IAsyncReader::Request on MSDN for the explanation of this */ #define MEDIATIME_FROM_BYTES(x) ((LONGLONG)(x) * 10000000) #define BYTES_FROM_MEDIATIME(time) ((time) / 10000000) diff --git a/dlls/winegstreamer/gst_private.h b/dlls/winegstreamer/gst_private.h index 5e6cbbab5d5..2374fb6fddd 100644 --- a/dlls/winegstreamer/gst_private.h +++ b/dlls/winegstreamer/gst_private.h @@ -41,27 +41,6 @@ bool array_reserve(void **elements, size_t *capacity, size_t count, size_t size); -static inline const char *debugstr_time(REFERENCE_TIME time) -{ - ULONGLONG abstime = time >= 0 ? time : -time; - unsigned int i = 0, j = 0; - char buffer[23], rev[23]; - - while (abstime || i <= 8) - { - buffer[i++] = '0' + (abstime % 10); - abstime /= 10; - if (i == 7) buffer[i++] = '.'; - } - if (time < 0) buffer[i++] = '-'; - - while (i--) rev[j++] = buffer[i]; - while (rev[j-1] == '0' && rev[j-2] != '.') --j; - rev[j] = 0; - - return wine_dbg_sprintf("%s", rev); -} - #define MEDIATIME_FROM_BYTES(x) ((LONGLONG)(x) * 10000000) struct wg_sample_queue; diff --git a/include/wine/strmbase.h b/include/wine/strmbase.h index 1b4cc3f7657..978d1c67160 100644 --- a/include/wine/strmbase.h +++ b/include/wine/strmbase.h @@ -20,6 +20,7 @@ */ #include "dshow.h" +#include "wine/debug.h" HRESULT WINAPI CopyMediaType(AM_MEDIA_TYPE * pDest, const AM_MEDIA_TYPE *pSrc); void WINAPI FreeMediaType(AM_MEDIA_TYPE * pMediaType); @@ -28,6 +29,27 @@ void WINAPI DeleteMediaType(AM_MEDIA_TYPE * pMediaType); void strmbase_dump_media_type(const AM_MEDIA_TYPE *mt); +static inline const char *debugstr_time(REFERENCE_TIME time) +{ + ULONGLONG abstime = time >= 0 ? time : -time; + unsigned int i = 0, j = 0; + char buffer[23], rev[23]; + + while (abstime || i <= 8) + { + buffer[i++] = '0' + (abstime % 10); + abstime /= 10; + if (i == 7) buffer[i++] = '.'; + } + if (time < 0) buffer[i++] = '-'; + + while (i--) rev[j++] = buffer[i]; + while (rev[j-1] == '0' && rev[j-2] != '.') --j; + rev[j] = 0; + + return wine_dbg_sprintf("%s", rev); +} + /* Pin functions */ struct strmbase_pin diff --git a/libs/strmbase/strmbase_private.h b/libs/strmbase/strmbase_private.h index cf702a279c2..374b10e7d51 100644 --- a/libs/strmbase/strmbase_private.h +++ b/libs/strmbase/strmbase_private.h @@ -30,25 +30,4 @@ #include "wine/list.h" #include "wine/strmbase.h" -static inline const char *debugstr_time(REFERENCE_TIME time) -{ - ULONGLONG abstime = time >= 0 ? time : -time; - unsigned int i = 0, j = 0; - char buffer[23], rev[23]; - - while (abstime || i <= 8) - { - buffer[i++] = '0' + (abstime % 10); - abstime /= 10; - if (i == 7) buffer[i++] = '.'; - } - if (time < 0) buffer[i++] = '-'; - - while (i--) rev[j++] = buffer[i]; - while (rev[j-1] == '0' && rev[j-2] != '.') --j; - rev[j] = 0; - - return wine_dbg_sprintf("%s", rev); -} - #endif /* __WINE_STRMBASE_PRIVATE_H */

1 0

Petrichor Park : vkd3d-shader/hlsl: Implement acos and asin trig intrinsics.
by Alexandre Julliard 21 Feb '24

21 Feb '24

Module: vkd3d Branch: master Commit: 69294c290bad83470e55046f02e2d1c6167150cb URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/69294c290bad83470e55046f02e2d… Author: Petrichor Park <ppark(a)codeweavers.com> Date: Sun Feb 11 11:06:01 2024 -0600 vkd3d-shader/hlsl: Implement acos and asin trig intrinsics. Tests have already been implemented in 92044d5e; this commit also reduces the scope of some of the todos (because now they're implemented!). Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=55154 --- libs/vkd3d-shader/hlsl.y | 57 +++++++++++++++++++++++++++++++++++++ tests/hlsl/inverse-trig.shader_test | 24 ++++++++-------- 2 files changed, 69 insertions(+), 12 deletions(-) diff --git a/libs/vkd3d-shader/hlsl.y b/libs/vkd3d-shader/hlsl.y index dc7b4c1b..111fec17 100644 --- a/libs/vkd3d-shader/hlsl.y +++ b/libs/vkd3d-shader/hlsl.y @@ -2666,6 +2666,55 @@ static bool intrinsic_abs(struct hlsl_ctx *ctx, return !!add_unary_arithmetic_expr(ctx, params->instrs, HLSL_OP1_ABS, params->args[0], loc); } +static bool write_acos_or_asin(struct hlsl_ctx *ctx, + const struct parse_initializer *params, const struct vkd3d_shader_location *loc, bool asin_mode) +{ + struct hlsl_ir_function_decl *func; + struct hlsl_type *type; + char *body; + + static const char template[] = + "%s %s(%s x)\n" + "{\n" + " %s abs_arg = abs(x);\n" + " %s poly_approx = (((-0.018729\n" + " * abs_arg + 0.074261)\n" + " * abs_arg - 0.212114)\n" + " * abs_arg + 1.570729);\n" + " %s correction = sqrt(1.0 - abs_arg);\n" + " %s zero_flip = (x < 0.0) * (-2.0 * correction * poly_approx + 3.141593);\n" + " %s result = poly_approx * correction + zero_flip;\n" + " return %s;\n" + "}"; + static const char fn_name_acos[] = "acos"; + static const char fn_name_asin[] = "asin"; + static const char return_stmt_acos[] = "result"; + static const char return_stmt_asin[] = "-result + 1.570796"; + + const char *fn_name = asin_mode ? fn_name_asin : fn_name_acos; + + type = params->args[0]->data_type; + type = hlsl_get_numeric_type(ctx, type->class, HLSL_TYPE_FLOAT, type->dimx, type->dimy); + + if (!(body = hlsl_sprintf_alloc(ctx, template, + type->name, fn_name, type->name, + type->name, type->name, type->name, type->name, type->name, + (asin_mode ? return_stmt_asin : return_stmt_acos)))) + return false; + func = hlsl_compile_internal_function(ctx, fn_name, body); + vkd3d_free(body); + if (!func) + return false; + + return add_user_call(ctx, func, params, loc); +} + +static bool intrinsic_acos(struct hlsl_ctx *ctx, + const struct parse_initializer *params, const struct vkd3d_shader_location *loc) +{ + return write_acos_or_asin(ctx, params, loc, false); +} + static bool intrinsic_all(struct hlsl_ctx *ctx, const struct parse_initializer *params, const struct vkd3d_shader_location *loc) { @@ -2743,6 +2792,12 @@ static bool intrinsic_any(struct hlsl_ctx *ctx, return false; } +static bool intrinsic_asin(struct hlsl_ctx *ctx, + const struct parse_initializer *params, const struct vkd3d_shader_location *loc) +{ + return write_acos_or_asin(ctx, params, loc, true); +} + /* Find the type corresponding to the given source type, with the same * dimensions but a different base type. */ static struct hlsl_type *convert_numeric_type(const struct hlsl_ctx *ctx, @@ -3970,9 +4025,11 @@ intrinsic_functions[] = /* Note: these entries should be kept in alphabetical order. */ {"D3DCOLORtoUBYTE4", 1, true, intrinsic_d3dcolor_to_ubyte4}, {"abs", 1, true, intrinsic_abs}, + {"acos", 1, true, intrinsic_acos}, {"all", 1, true, intrinsic_all}, {"any", 1, true, intrinsic_any}, {"asfloat", 1, true, intrinsic_asfloat}, + {"asin", 1, true, intrinsic_asin}, {"asuint", -1, true, intrinsic_asuint}, {"ceil", 1, true, intrinsic_ceil}, {"clamp", 3, true, intrinsic_clamp}, diff --git a/tests/hlsl/inverse-trig.shader_test b/tests/hlsl/inverse-trig.shader_test index 0c1fdc01..01344204 100644 --- a/tests/hlsl/inverse-trig.shader_test +++ b/tests/hlsl/inverse-trig.shader_test @@ -3,7 +3,7 @@ % implementations. DXIL defines intrinsics for inverse trig, to be implemented % by the backend. -[pixel shader todo] +[pixel shader todo(sm<4)] uniform float4 a; float4 main() : sv_target @@ -13,26 +13,26 @@ float4 main() : sv_target [test] uniform 0 float4 -1.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (3.14159274, 0.0, 0.0, 0.0) 128 uniform 0 float4 -0.5 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (2.094441441, 0.0, 0.0, 0.0) 256 uniform 0 float4 0.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (1.57072878, 0.0, 0.0, 0.0) 1024 uniform 0 float4 0.5 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (1.04715133, 0.0, 0.0, 0.0) 512 uniform 0 float4 1.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (0.0, 0.0, 0.0, 0.0) 128 -[pixel shader todo] +[pixel shader todo(sm<4)] uniform float4 a; float4 main() : sv_target @@ -44,7 +44,7 @@ float4 main() : sv_target [test] uniform 0 float4 -1.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (-31416.0, 0.0, 0.0, 0.0) [require] @@ -52,15 +52,15 @@ shader model < 6.0 [test] uniform 0 float4 -0.5 0.0 0.0 0.0 -todo draw quad +todo(sm<4) draw quad probe all rgba (-10473.0, 0.0, 0.0, 0.0) uniform 0 float4 0.0 0.0 0.0 0.0 -todo draw quad +todo(sm<4) draw quad probe all rgba (1.0, 0.0, 0.0, 0.0) uniform 0 float4 0.5 0.0 0.0 0.0 -todo draw quad +todo(sm<4) draw quad probe all rgba (10473.0, 0.0, 0.0, 0.0) [require] @@ -88,7 +88,7 @@ probe all rgba (10472.0, 0.0, 0.0, 0.0) 4096 [test] uniform 0 float4 1.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (31416.0, 0.0, 0.0, 0.0)

1 0

Petrichor Park : vkd3d-shader/hlsl: Implement atan and atan2.
by Alexandre Julliard 21 Feb '24

21 Feb '24

Module: vkd3d Branch: master Commit: 02a366782263f6eb06b8cd839113dff308f2cf05 URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/02a366782263f6eb06b8cd839113d… Author: Petrichor Park <ppark(a)codeweavers.com> Date: Sun Feb 11 11:06:24 2024 -0600 vkd3d-shader/hlsl: Implement atan and atan2. Also narrows some more todos on the tests. --- libs/vkd3d-shader/hlsl.y | 95 +++++++++++++++++++++++++++++++++++++ tests/hlsl/inverse-trig.shader_test | 44 ++++++++--------- 2 files changed, 117 insertions(+), 22 deletions(-) diff --git a/libs/vkd3d-shader/hlsl.y b/libs/vkd3d-shader/hlsl.y index 111fec17..b59dbf32 100644 --- a/libs/vkd3d-shader/hlsl.y +++ b/libs/vkd3d-shader/hlsl.y @@ -2798,6 +2798,99 @@ static bool intrinsic_asin(struct hlsl_ctx *ctx, return write_acos_or_asin(ctx, params, loc, true); } +static bool write_atan_or_atan2(struct hlsl_ctx *ctx, + const struct parse_initializer *params, + const struct vkd3d_shader_location *loc, bool atan2_mode) +{ + struct hlsl_ir_function_decl *func; + struct hlsl_type *type; + struct vkd3d_string_buffer *buf; + int ret; + + static const char atan2_name[] = "atan2"; + static const char atan_name[] = "atan"; + + static const char atan2_header_template[] = + "%s atan2(%s y, %s x)\n" + "{\n" + " %s in_y, in_x;\n" + " in_y = y;\n" + " in_x = x;\n"; + static const char atan_header_template[] = + "%s atan(%s y)\n" + "{\n" + " %s in_y, in_x;\n" + " in_y = y;\n" + " in_x = 1.0;\n"; + + static const char body_template[] = + " %s recip, input, x2, poly_approx, flipped;" + " recip = 1.0 / max(abs(in_y), abs(in_x));\n" + " input = recip * min(abs(in_y), abs(in_x));\n" + " x2 = input * input;\n" + " poly_approx = ((((0.020835\n" + " * x2 - 0.085133)\n" + " * x2 + 0.180141)\n" + " * x2 - 0.330299)\n" + " * x2 + 0.999866)\n" + " * input;\n" + " flipped = poly_approx * -2.0 + 1.570796;\n" + " poly_approx += abs(in_x) < abs(in_y) ? flipped : 0.0;\n" + " poly_approx += in_x < 0.0 ? -3.1415927 : 0.0;\n" + " return (min(in_x, in_y) < 0.0 && max(in_x, in_y) >= 0.0)\n" + " ? -poly_approx\n" + " : poly_approx;\n" + "}"; + + if (!(type = elementwise_intrinsic_get_common_type(ctx, params, loc))) + return false; + type = hlsl_get_numeric_type(ctx, type->class, HLSL_TYPE_FLOAT, type->dimx, type->dimy); + + if (!(buf = hlsl_get_string_buffer(ctx))) + return false; + + if (atan2_mode) + ret = vkd3d_string_buffer_printf(buf, atan2_header_template, + type->name, type->name, type->name, type->name); + else + ret = vkd3d_string_buffer_printf(buf, atan_header_template, + type->name, type->name, type->name); + if (ret < 0) + { + vkd3d_string_buffer_cleanup(buf); + return false; + } + + ret = vkd3d_string_buffer_printf(buf, body_template, type->name); + if (ret < 0) + { + vkd3d_string_buffer_cleanup(buf); + return false; + } + + func = hlsl_compile_internal_function(ctx, + atan2_mode ? atan2_name : atan_name, buf->buffer); + vkd3d_string_buffer_cleanup(buf); + if (!func) + return false; + + return add_user_call(ctx, func, params, loc); +} + +static bool intrinsic_atan(struct hlsl_ctx *ctx, + const struct parse_initializer *params, const struct vkd3d_shader_location *loc) +{ + return write_atan_or_atan2(ctx, params, loc, false); +} + + +static bool intrinsic_atan2(struct hlsl_ctx *ctx, + const struct parse_initializer *params, const struct vkd3d_shader_location *loc) +{ + return write_atan_or_atan2(ctx, params, loc, true); +} + + /* Find the type corresponding to the given source type, with the same * dimensions but a different base type. */ static struct hlsl_type *convert_numeric_type(const struct hlsl_ctx *ctx, @@ -4031,6 +4124,8 @@ intrinsic_functions[] = {"asfloat", 1, true, intrinsic_asfloat}, {"asin", 1, true, intrinsic_asin}, {"asuint", -1, true, intrinsic_asuint}, + {"atan", 1, true, intrinsic_atan}, + {"atan2", 2, true, intrinsic_atan2}, {"ceil", 1, true, intrinsic_ceil}, {"clamp", 3, true, intrinsic_clamp}, {"clip", 1, true, intrinsic_clip}, diff --git a/tests/hlsl/inverse-trig.shader_test b/tests/hlsl/inverse-trig.shader_test index 01344204..e428d309 100644 --- a/tests/hlsl/inverse-trig.shader_test +++ b/tests/hlsl/inverse-trig.shader_test @@ -92,7 +92,7 @@ todo(sm<4) draw quad probe all rgba (31416.0, 0.0, 0.0, 0.0) -[pixel shader todo] +[pixel shader todo(sm<4)] uniform float4 a; float4 main() : sv_target @@ -102,26 +102,26 @@ float4 main() : sv_target [test] uniform 0 float4 -1.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (-0.785409629, 0.0, 0.0, 0.0) 512 uniform 0 float4 -0.5 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (-0.4636476, 0.0, 0.0, 0.0) 256 uniform 0 float4 0.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (0.0, 0.0, 0.0, 0.0) 256 uniform 0 float4 0.5 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (0.4636476, 0.0, 0.0, 0.0) 256 uniform 0 float4 1.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (0.785409629, 0.0, 0.0, 0.0) 512 -[pixel shader todo] +[pixel shader todo(sm<4)] uniform float4 a; float4 main() : sv_target @@ -133,64 +133,64 @@ float4 main() : sv_target [test] % Non-degenerate cases uniform 0 float4 1.0 1.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (0.785385, 0.0, 0.0, 0.0) 512 uniform 0 float4 5.0 -5.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (2.356194, 0.0, 0.0, 0.0) 256 uniform 0 float4 -3.0 -3.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (-2.356194, 0.0, 0.0, 0.0) 256 uniform 0 float4 1.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (1.570796, 0.0, 0.0, 0.0) 256 uniform 0 float4 -1.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (-1.570796, 0.0, 0.0, 0.0) 256 uniform 0 float4 0.0 1.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (0.0, 0.0, 0.0, 0.0) 256 uniform 0 float4 0.0 -1.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (3.1415927, 0.0, 0.0, 0.0) 256 % Degenerate cases uniform 0 float4 0.00001 0.00002 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (0.463647, 0.0, 0.0, 0.0) 256 uniform 0 float4 0.00001 -0.00002 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (2.677945, 0.0, 0.0, 0.0) 256 uniform 0 float4 -0.00001 100000.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (-0.000000000099986595, 0.0, 0.0, 0.0) 2048 uniform 0 float4 10000000.0 0.00000001 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (1.570796, 0.0, 0.0, 0.0) 256 % Negative zero behavior should be to treat it the % same as normal zero. uniform 0 float4 1000000000.0 0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (1.570796, 0.0, 0.0, 0.0) 256 uniform 0 float4 1000000000.0 -0.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (1.570796, 0.0, 0.0, 0.0) 256 uniform 0 float4 0.0 -1.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (3.1415927, 0.0, 0.0, 0.0) 256 uniform 0 float4 -0.0 -1.0 0.0 0.0 -todo(sm<6) draw quad +todo(sm<4) draw quad probe all rgba (3.1415927, 0.0, 0.0, 0.0) 256

1 0