[PATCH v6 0/10] MR458: vkd3d-shader/hlsl: Handle register(cX) reservations.
Currently we are not properly handling register(cX) reservations for SM1, this is one of the things required for the SNK shaders (CW Bug Bug 18092). register(cX) reservations also change the offset in the $Globals buffer in SM4, so support for this is also included. --- Patch 1/4 is required to specify: ``` [require] shader model < 4.0 ``` so that the tests that follow do not get run with the vulkan backend on SM4. I think nobody disagreed with that patch. -- v6: vkd3d-shader/hlsl: Turn register(cX) reservations into buffer offset for SM4. vkd3d-shader/hlsl: Make register(cX) reservations work for SM1. tests: Test register(cX) reservations. tests: Rename register-reservations.shader_test to register-reservations-resources.shader_test. tests/shader-runner: Run compilation tests with SM1 when SM1 models are selected. https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
From: Giovanni Mascellani <gmascellani(a)codeweavers.com> --- gitlab/test.yml | 24 +++++++----------------- 1 file changed, 7 insertions(+), 17 deletions(-) diff --git a/gitlab/test.yml b/gitlab/test.yml index 97bebba53..cfc4906f5 100644 --- a/gitlab/test.yml +++ b/gitlab/test.yml @@ -1,4 +1,4 @@ -test-win-64: +.test-win: stage: test rules: - if: $CI_PIPELINE_SOURCE == 'merge_request_event' @@ -9,27 +9,17 @@ test-win-64: - win10-21h2 script: - ./artifacts/driver.cross64.exe - variables: - TEST_ARCH: "64" artifacts: when: always paths: - artifacts +test-win-64: + extends: .test-win + variables: + TEST_ARCH: "64" + test-win-32: - stage: test - rules: - - if: $CI_PIPELINE_SOURCE == 'merge_request_event' - interruptible: true - needs: - - job: build-crosstest - tags: - - win10-21h2 - script: - - ./artifacts/driver.cross64.exe + extends: .test-win variables: TEST_ARCH: "32" - artifacts: - when: always - paths: - - artifacts -- GitLab https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
From: Giovanni Mascellani <gmascellani(a)codeweavers.com> --- gitlab/test.yml | 3 ++- tests/driver.c | 52 ++++++++++++++++++++++++++----------------------- 2 files changed, 30 insertions(+), 25 deletions(-) diff --git a/gitlab/test.yml b/gitlab/test.yml index cfc4906f5..65a75d0e9 100644 --- a/gitlab/test.yml +++ b/gitlab/test.yml @@ -8,7 +8,8 @@ tags: - win10-21h2 script: - - ./artifacts/driver.cross64.exe + - git rebase $CI_MERGE_REQUEST_DIFF_BASE_SHA --exec './artifacts/driver.cross64.exe $(git cherry $CI_MERGE_REQUEST_DIFF_BASE_SHA HEAD^ | wc -l) $(git rev-parse --short HEAD)' + - if (Test-Path "pipeline_failed") { exit 1 } artifacts: when: always paths: diff --git a/tests/driver.c b/tests/driver.c index 553417ada..00bf227a4 100644 --- a/tests/driver.c +++ b/tests/driver.c @@ -18,6 +18,7 @@ #include <stdio.h> #include <stdbool.h> +#include <stdlib.h> #include <windows.h> #include <shlobj.h> @@ -215,40 +216,43 @@ static bool run_tests_for_directory(const char *commit_dir) printf("# FAIL: %u\n", test_count - success_count); if (test_count != success_count) - ret = false; + { + HANDLE handle; + + handle = CreateFileA("pipeline_failed", GENERIC_WRITE, 0, NULL, CREATE_ALWAYS, FILE_ATTRIBUTE_NORMAL, NULL); + if (handle == INVALID_HANDLE_VALUE) + { + fprintf(stderr, "Cannot create failure file, last error %ld.\n", GetLastError()); + ret = false; + } + else + { + if (!CloseHandle(handle)) + fprintf(stderr, "Cannot close failure file, last error %ld.\n", GetLastError()); + } + } return ret; } -int wmain(void) +int wmain(int argc, WCHAR **wargv) { - WIN32_FIND_DATAA find_data; - HANDLE find_handle; - bool ret = true; + char commit_num[16], commit_hash[16], commit_dir[16]; SetErrorMode(SEM_FAILCRITICALERRORS | SEM_NOGPFAULTERRORBOX | SEM_NOOPENFILEERRORBOX); - find_handle = FindFirstFileA("artifacts/*-*", &find_data); - if (find_handle == INVALID_HANDLE_VALUE) + if (argc != 3) { - fprintf(stderr, "Cannot list commits, last error %ld.\n", GetLastError()); - ret = false; + fprintf(stderr, "Call with commit number and hash.\n"); + return 1; } - else - { - do - { - ret &= run_tests_for_directory(find_data.cFileName); - } while (FindNextFileA(find_handle, &find_data)); - if (GetLastError() != ERROR_NO_MORE_FILES) - { - fprintf(stderr, "Cannot list tests, last error %ld.\n", GetLastError()); - ret = false; - } - - FindClose(find_handle); - } + WideCharToMultiByte(CP_ACP, 0, wargv[1], -1, commit_num, sizeof(commit_num), NULL, NULL); + WideCharToMultiByte(CP_ACP, 0, wargv[2], -1, commit_hash, sizeof(commit_hash), NULL, NULL); + commit_num[sizeof(commit_num) - 1] = '\0'; + commit_hash[sizeof(commit_hash) - 1] = '\0'; + snprintf(commit_dir, sizeof(commit_dir), "%03d-%s", atoi(commit_num), commit_hash); + commit_dir[sizeof(commit_dir) - 1] = '\0'; - return !ret; + return !run_tests_for_directory(commit_dir); } -- GitLab https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
From: Francisco Casas <fcasas(a)codeweavers.com> We need these checks to properly handle tests that require target profiles 3.0 and lower, or tests that require target profile 6.0 and higher. Since all backeds require an ops->check_requirements now, expect it to always be defined. --- tests/shader_runner.c | 3 ++- tests/shader_runner.h | 4 ++-- tests/shader_runner_d3d11.c | 13 +++++++++++++ tests/shader_runner_d3d12.c | 14 ++++++++++++++ tests/shader_runner_vulkan.c | 13 +++++++++++++ 5 files changed, 44 insertions(+), 3 deletions(-) diff --git a/tests/shader_runner.c b/tests/shader_runner.c index dadfe3338..43949c01a 100644 --- a/tests/shader_runner.c +++ b/tests/shader_runner.c @@ -1089,8 +1089,9 @@ void run_shader_tests(struct shader_runner *runner, const struct shader_runner_o break; case STATE_REQUIRE: + assert(runner->ops->check_requirements); if (runner->maximum_shader_model < runner->minimum_shader_model - || (runner->ops->check_requirements && !runner->ops->check_requirements(runner))) + || !runner->ops->check_requirements(runner)) { skip_tests = true; } diff --git a/tests/shader_runner.h b/tests/shader_runner.h index bda44a429..ee0e9094d 100644 --- a/tests/shader_runner.h +++ b/tests/shader_runner.h @@ -142,9 +142,9 @@ struct shader_runner struct shader_runner_ops { - /* Returns false if unable to run the given tests. If NULL, all tests are - * run. */ + /* Returns false if unable to run the given tests. */ bool (*check_requirements)(struct shader_runner *runner); + struct resource *(*create_resource)(struct shader_runner *runner, const struct resource_params *params); void (*destroy_resource)(struct shader_runner *runner, struct resource *resource); bool (*draw)(struct shader_runner *runner, D3D_PRIMITIVE_TOPOLOGY primitive_topology, unsigned int vertex_count); diff --git a/tests/shader_runner_d3d11.c b/tests/shader_runner_d3d11.c index bd9d363ce..7a8f8dac0 100644 --- a/tests/shader_runner_d3d11.c +++ b/tests/shader_runner_d3d11.c @@ -100,6 +100,18 @@ static ID3D10Blob *compile_shader(const struct d3d11_shader_runner *runner, cons return blob; } +static bool d3d11_runner_check_requirements(struct shader_runner *r) +{ + struct d3d11_shader_runner *runner = d3d11_shader_runner(r); + + if (runner->r.maximum_shader_model < SHADER_MODEL_4_0) + return false; + if (runner->r.minimum_shader_model >= SHADER_MODEL_6_0) + return false; + + return true; +} + static IDXGIAdapter *create_adapter(void) { IDXGIFactory4 *factory4; @@ -724,6 +736,7 @@ static void d3d11_runner_release_readback(struct shader_runner *r, struct resour static const struct shader_runner_ops d3d11_runner_ops = { + .check_requirements = d3d11_runner_check_requirements, .create_resource = d3d11_runner_create_resource, .destroy_resource = d3d11_runner_destroy_resource, .dispatch = d3d11_runner_dispatch, diff --git a/tests/shader_runner_d3d12.c b/tests/shader_runner_d3d12.c index a05dfd059..bb50564fa 100644 --- a/tests/shader_runner_d3d12.c +++ b/tests/shader_runner_d3d12.c @@ -96,6 +96,19 @@ static ID3D10Blob *compile_shader(const struct d3d12_shader_runner *runner, cons return blob; } +static bool d3d12_runner_check_requirements(struct shader_runner *r) +{ + struct d3d12_shader_runner *runner = d3d12_shader_runner(r); + + if (runner->r.maximum_shader_model < SHADER_MODEL_4_0) + return false; + + if (runner->r.minimum_shader_model >= SHADER_MODEL_6_0 && !runner->dxc_compiler) + return false; + + return true; +} + #define MAX_RESOURCE_DESCRIPTORS (MAX_RESOURCES * 2) static struct resource *d3d12_runner_create_resource(struct shader_runner *r, const struct resource_params *params) @@ -561,6 +574,7 @@ static void d3d12_runner_release_readback(struct shader_runner *r, struct resour static const struct shader_runner_ops d3d12_runner_ops = { + .check_requirements = d3d12_runner_check_requirements, .create_resource = d3d12_runner_create_resource, .destroy_resource = d3d12_runner_destroy_resource, .dispatch = d3d12_runner_dispatch, diff --git a/tests/shader_runner_vulkan.c b/tests/shader_runner_vulkan.c index f89b4d624..ec2d1fc9e 100644 --- a/tests/shader_runner_vulkan.c +++ b/tests/shader_runner_vulkan.c @@ -83,6 +83,18 @@ static struct vulkan_shader_runner *vulkan_shader_runner(struct shader_runner *r #define VK_CALL(f) (runner->f) +static bool vulkan_runner_check_requirements(struct shader_runner *r) +{ + struct vulkan_shader_runner *runner = vulkan_shader_runner(r); + + if (runner->r.maximum_shader_model < SHADER_MODEL_4_0) + return false; + if (runner->r.minimum_shader_model >= SHADER_MODEL_6_0) + return false; + + return true; +} + static void begin_command_buffer(struct vulkan_shader_runner *runner) { VkCommandBufferBeginInfo buffer_begin_desc = {.sType = VK_STRUCTURE_TYPE_COMMAND_BUFFER_BEGIN_INFO}; @@ -1157,6 +1169,7 @@ static void vulkan_runner_release_readback(struct shader_runner *r, struct resou static const struct shader_runner_ops vulkan_runner_ops = { + .check_requirements = vulkan_runner_check_requirements, .create_resource = vulkan_runner_create_resource, .destroy_resource = vulkan_runner_destroy_resource, .dispatch = vulkan_runner_dispatch, -- GitLab https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
From: Francisco Casas <fcasas(a)codeweavers.com> In case that the backend selects a shader model that is higher than the minimum in the provided range, it doesn't make sense to check todos(...), fails(...), or notimpl(...) against the minimum. So the backend must report the selected model for the given range (if any). This is done through the check_requirements() function. --- tests/shader_runner.c | 26 +++++++++++++++----------- tests/shader_runner.h | 6 ++++-- tests/shader_runner_d3d11.c | 8 ++++---- tests/shader_runner_d3d12.c | 12 ++++++------ tests/shader_runner_d3d9.c | 6 ++++-- tests/shader_runner_vulkan.c | 8 ++++---- 6 files changed, 37 insertions(+), 29 deletions(-) diff --git a/tests/shader_runner.c b/tests/shader_runner.c index 43949c01a..b9adec8b4 100644 --- a/tests/shader_runner.c +++ b/tests/shader_runner.c @@ -501,9 +501,9 @@ static void parse_test_directive(struct shader_runner *runner, const char *line) if (match_string(line, "todo", &line)) runner->is_todo = true; else if (match_string(line, "todo(sm<6)", &line)) - runner->is_todo = runner->minimum_shader_model < SHADER_MODEL_6_0; + runner->is_todo = runner->selected_shader_model < SHADER_MODEL_6_0; else if (match_string(line, "todo(sm>=6)", &line)) - runner->is_todo = runner->minimum_shader_model >= SHADER_MODEL_6_0; + runner->is_todo = runner->selected_shader_model >= SHADER_MODEL_6_0; if (match_string(line, "dispatch", &line)) { @@ -940,7 +940,7 @@ result_release: static void compile_shader(struct shader_runner *runner, IDxcCompiler3 *dxc_compiler, const char *source, size_t len, enum shader_type type, HRESULT expect) { - bool use_dxcompiler = runner->minimum_shader_model >= SHADER_MODEL_6_0; + bool use_dxcompiler = runner->selected_shader_model >= SHADER_MODEL_6_0; ID3D10Blob *blob = NULL, *errors = NULL; char profile[7]; HRESULT hr; @@ -972,9 +972,9 @@ static void compile_shader(struct shader_runner *runner, IDxcCompiler3 *dxc_comp else { if (type == SHADER_TYPE_FX) - sprintf(profile, "%s_%s", shader_type_string(type), effect_models[runner->minimum_shader_model]); + sprintf(profile, "%s_%s", shader_type_string(type), effect_models[runner->selected_shader_model]); else - sprintf(profile, "%s_%s", shader_type_string(type), shader_models[runner->minimum_shader_model]); + sprintf(profile, "%s_%s", shader_type_string(type), shader_models[runner->selected_shader_model]); hr = D3DCompile(source, len, NULL, NULL, NULL, "main", profile, runner->compile_options, 0, &blob, &errors); } hr = map_unidentified_hrs(hr); @@ -1004,7 +1004,7 @@ static enum parse_state read_shader_directive(struct shader_runner *runner, enum /* 'todo' is not meaningful when dxcompiler is in use, so it has no '(sm<6) qualifier. */ if (match_directive_substring(src, "todo", &src)) { - if (runner->minimum_shader_model >= SHADER_MODEL_6_0) + if (runner->selected_shader_model >= SHADER_MODEL_6_0) continue; if (state == STATE_SHADER_COMPUTE) state = STATE_SHADER_COMPUTE_TODO; @@ -1021,17 +1021,17 @@ static enum parse_state read_shader_directive(struct shader_runner *runner, enum } else if (match_directive_substring(src, "fail(sm<6)", &src)) { - if (runner->minimum_shader_model < SHADER_MODEL_6_0) + if (runner->selected_shader_model < SHADER_MODEL_6_0) *expect_hr = E_FAIL; } else if (match_directive_substring(src, "fail(sm>=6)", &src)) { - if (runner->minimum_shader_model >= SHADER_MODEL_6_0) + if (runner->selected_shader_model >= SHADER_MODEL_6_0) *expect_hr = E_FAIL; } else if (match_directive_substring(src, "notimpl(sm<6)", &src)) { - if (runner->minimum_shader_model < SHADER_MODEL_6_0) + if (runner->selected_shader_model < SHADER_MODEL_6_0) *expect_hr = E_NOTIMPL; } else @@ -1070,6 +1070,10 @@ void run_shader_tests(struct shader_runner *runner, const struct shader_runner_o runner->ops = ops; runner->minimum_shader_model = minimum_shader_model; runner->maximum_shader_model = maximum_shader_model; + runner->selected_shader_model = minimum_shader_model; + + assert(runner->ops->check_requirements); + skip_tests = !runner->ops->check_requirements(runner, &runner->selected_shader_model); for (;;) { @@ -1089,9 +1093,8 @@ void run_shader_tests(struct shader_runner *runner, const struct shader_runner_o break; case STATE_REQUIRE: - assert(runner->ops->check_requirements); if (runner->maximum_shader_model < runner->minimum_shader_model - || !runner->ops->check_requirements(runner)) + || !runner->ops->check_requirements(runner, &runner->selected_shader_model)) { skip_tests = true; } @@ -1249,6 +1252,7 @@ void run_shader_tests(struct shader_runner *runner, const struct shader_runner_o state = STATE_REQUIRE; runner->minimum_shader_model = minimum_shader_model; runner->maximum_shader_model = maximum_shader_model; + runner->selected_shader_model = minimum_shader_model; runner->compile_options = 0; skip_tests = false; } diff --git a/tests/shader_runner.h b/tests/shader_runner.h index ee0e9094d..89a01abff 100644 --- a/tests/shader_runner.h +++ b/tests/shader_runner.h @@ -122,6 +122,7 @@ struct shader_runner char *fx_source; enum shader_model minimum_shader_model; enum shader_model maximum_shader_model; + enum shader_model selected_shader_model; bool last_render_failed; @@ -142,8 +143,9 @@ struct shader_runner struct shader_runner_ops { - /* Returns false if unable to run the given tests. */ - bool (*check_requirements)(struct shader_runner *runner); + /* Returns false if unable to run the given tests. Otherwise, it also outputs the lowest model + * supported within the range provided in the runner. */ + bool (*check_requirements)(struct shader_runner *runner, enum shader_model *model); struct resource *(*create_resource)(struct shader_runner *runner, const struct resource_params *params); void (*destroy_resource)(struct shader_runner *runner, struct resource *resource); diff --git a/tests/shader_runner_d3d11.c b/tests/shader_runner_d3d11.c index 7a8f8dac0..df704da8c 100644 --- a/tests/shader_runner_d3d11.c +++ b/tests/shader_runner_d3d11.c @@ -80,15 +80,13 @@ static ID3D10Blob *compile_shader(const struct d3d11_shader_runner *runner, cons static const char *const shader_models[] = { - [SHADER_MODEL_2_0] = "4_0", - [SHADER_MODEL_3_0] = "4_0", [SHADER_MODEL_4_0] = "4_0", [SHADER_MODEL_4_1] = "4_1", [SHADER_MODEL_5_0] = "5_0", [SHADER_MODEL_5_1] = "5_1", }; - sprintf(profile, "%s_%s", type, shader_models[runner->r.minimum_shader_model]); + sprintf(profile, "%s_%s", type, shader_models[runner->r.selected_shader_model]); hr = D3DCompile(source, strlen(source), NULL, NULL, NULL, "main", profile, runner->r.compile_options, 0, &blob, &errors); ok(hr == S_OK, "Failed to compile shader, hr %#lx.\n", hr); if (errors) @@ -100,7 +98,7 @@ static ID3D10Blob *compile_shader(const struct d3d11_shader_runner *runner, cons return blob; } -static bool d3d11_runner_check_requirements(struct shader_runner *r) +static bool d3d11_runner_check_requirements(struct shader_runner *r, enum shader_model *model) { struct d3d11_shader_runner *runner = d3d11_shader_runner(r); @@ -109,6 +107,8 @@ static bool d3d11_runner_check_requirements(struct shader_runner *r) if (runner->r.minimum_shader_model >= SHADER_MODEL_6_0) return false; + *model = max(SHADER_MODEL_4_0, runner->r.minimum_shader_model); + return true; } diff --git a/tests/shader_runner_d3d12.c b/tests/shader_runner_d3d12.c index bb50564fa..2fd89d74e 100644 --- a/tests/shader_runner_d3d12.c +++ b/tests/shader_runner_d3d12.c @@ -67,8 +67,6 @@ static ID3D10Blob *compile_shader(const struct d3d12_shader_runner *runner, cons static const char *const shader_models[] = { - [SHADER_MODEL_2_0] = "4_0", - [SHADER_MODEL_3_0] = "4_0", [SHADER_MODEL_4_0] = "4_0", [SHADER_MODEL_4_1] = "4_1", [SHADER_MODEL_5_0] = "5_0", @@ -76,14 +74,14 @@ static ID3D10Blob *compile_shader(const struct d3d12_shader_runner *runner, cons [SHADER_MODEL_6_0] = "6_0", }; - if (runner->r.minimum_shader_model >= SHADER_MODEL_6_0) + if (runner->r.selected_shader_model >= SHADER_MODEL_6_0) { assert(runner->dxc_compiler); hr = dxc_compiler_compile_shader(runner->dxc_compiler, type, runner->r.compile_options, source, &blob, &errors); } else { - sprintf(profile, "%s_%s", shader_type_string(type), shader_models[runner->r.minimum_shader_model]); + sprintf(profile, "%s_%s", shader_type_string(type), shader_models[runner->r.selected_shader_model]); hr = D3DCompile(source, strlen(source), NULL, NULL, NULL, "main", profile, runner->r.compile_options, 0, &blob, &errors); } ok(FAILED(hr) == !blob, "Got unexpected hr %#x, blob %p.\n", hr, blob); @@ -96,7 +94,7 @@ static ID3D10Blob *compile_shader(const struct d3d12_shader_runner *runner, cons return blob; } -static bool d3d12_runner_check_requirements(struct shader_runner *r) +static bool d3d12_runner_check_requirements(struct shader_runner *r, enum shader_model *model) { struct d3d12_shader_runner *runner = d3d12_shader_runner(r); @@ -106,6 +104,8 @@ static bool d3d12_runner_check_requirements(struct shader_runner *r) if (runner->r.minimum_shader_model >= SHADER_MODEL_6_0 && !runner->dxc_compiler) return false; + *model = max(SHADER_MODEL_4_0, runner->r.minimum_shader_model); + return true; } @@ -334,7 +334,7 @@ static bool d3d12_runner_dispatch(struct shader_runner *r, unsigned int x, unsig size_t i; cs_code = compile_shader(runner, runner->r.cs_source, SHADER_TYPE_CS); - todo_if(runner->r.is_todo && runner->r.minimum_shader_model < SHADER_MODEL_6_0) ok(cs_code, "Failed to compile shader.\n"); + todo_if(runner->r.is_todo && runner->r.selected_shader_model < SHADER_MODEL_6_0) ok(cs_code, "Failed to compile shader.\n"); if (!cs_code) return false; diff --git a/tests/shader_runner_d3d9.c b/tests/shader_runner_d3d9.c index 0b33c1c53..fbd7d6b5b 100644 --- a/tests/shader_runner_d3d9.c +++ b/tests/shader_runner_d3d9.c @@ -68,7 +68,7 @@ static ID3D10Blob *compile_shader(const struct d3d9_shader_runner *runner, const [SHADER_MODEL_3_0] = "3_0", }; - sprintf(profile, "%s_%s", type, shader_models[runner->r.minimum_shader_model]); + sprintf(profile, "%s_%s", type, shader_models[runner->r.selected_shader_model]); hr = D3DCompile(source, strlen(source), NULL, NULL, NULL, "main", profile, runner->r.compile_options, 0, &blob, &errors); ok(hr == S_OK, "Failed to compile shader, hr %#lx.\n", hr); if (errors) @@ -183,13 +183,15 @@ static D3DTEXTUREADDRESS sampler_address_to_d3d9(D3D12_TEXTURE_ADDRESS_MODE addr vkd3d_unreachable(); } -static bool d3d9_runner_check_requirements(struct shader_runner *r) +static bool d3d9_runner_check_requirements(struct shader_runner *r, enum shader_model *model) { struct d3d9_shader_runner *runner = d3d9_shader_runner(r); if (runner->r.minimum_shader_model >= SHADER_MODEL_4_0) return false; + *model = runner->r.minimum_shader_model; + return true; } diff --git a/tests/shader_runner_vulkan.c b/tests/shader_runner_vulkan.c index ec2d1fc9e..131e394b0 100644 --- a/tests/shader_runner_vulkan.c +++ b/tests/shader_runner_vulkan.c @@ -83,7 +83,7 @@ static struct vulkan_shader_runner *vulkan_shader_runner(struct shader_runner *r #define VK_CALL(f) (runner->f) -static bool vulkan_runner_check_requirements(struct shader_runner *r) +static bool vulkan_runner_check_requirements(struct shader_runner *r, enum shader_model *model) { struct vulkan_shader_runner *runner = vulkan_shader_runner(r); @@ -92,6 +92,8 @@ static bool vulkan_runner_check_requirements(struct shader_runner *r) if (runner->r.minimum_shader_model >= SHADER_MODEL_6_0) return false; + *model = max(SHADER_MODEL_4_0, runner->r.minimum_shader_model); + return true; } @@ -407,8 +409,6 @@ static bool compile_shader(const struct vulkan_shader_runner *runner, const char static const char *const shader_models[] = { - [SHADER_MODEL_2_0] = "4_0", - [SHADER_MODEL_3_0] = "4_0", [SHADER_MODEL_4_0] = "4_0", [SHADER_MODEL_4_1] = "4_1", [SHADER_MODEL_5_0] = "5_0", @@ -451,7 +451,7 @@ static bool compile_shader(const struct vulkan_shader_runner *runner, const char } hlsl_info.entry_point = "main"; - sprintf(profile, "%s_%s", type, shader_models[runner->r.minimum_shader_model]); + sprintf(profile, "%s_%s", type, shader_models[runner->r.selected_shader_model]); hlsl_info.profile = profile; ret = vkd3d_shader_compile(&info, dxbc, &messages); -- GitLab https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
From: Francisco Casas <fcasas(a)codeweavers.com> --- tests/shader_runner.c | 48 +++++++++++++++++++++++++++++++++++-------- 1 file changed, 40 insertions(+), 8 deletions(-) diff --git a/tests/shader_runner.c b/tests/shader_runner.c index b9adec8b4..745f17b3a 100644 --- a/tests/shader_runner.c +++ b/tests/shader_runner.c @@ -500,6 +500,10 @@ static void parse_test_directive(struct shader_runner *runner, const char *line) if (match_string(line, "todo", &line)) runner->is_todo = true; + else if (match_string(line, "todo(sm<4)", &line)) + runner->is_todo = runner->selected_shader_model < SHADER_MODEL_4_0; + else if (match_string(line, "todo(sm>=4)", &line)) + runner->is_todo = runner->selected_shader_model >= SHADER_MODEL_4_0; else if (match_string(line, "todo(sm<6)", &line)) runner->is_todo = runner->selected_shader_model < SHADER_MODEL_6_0; else if (match_string(line, "todo(sm>=6)", &line)) @@ -996,6 +1000,18 @@ static void compile_shader(struct shader_runner *runner, IDxcCompiler3 *dxc_comp } } +static enum parse_state get_parse_state_todo(enum parse_state state) +{ + if (state == STATE_SHADER_COMPUTE) + return STATE_SHADER_COMPUTE_TODO; + else if (state == STATE_SHADER_PIXEL) + return STATE_SHADER_PIXEL_TODO; + else if (state == STATE_SHADER_VERTEX) + return STATE_SHADER_VERTEX_TODO; + else + return STATE_SHADER_EFFECT_TODO; +} + static enum parse_state read_shader_directive(struct shader_runner *runner, enum parse_state state, const char *line, const char *src, HRESULT *expect_hr) { @@ -1006,19 +1022,35 @@ static enum parse_state read_shader_directive(struct shader_runner *runner, enum { if (runner->selected_shader_model >= SHADER_MODEL_6_0) continue; - if (state == STATE_SHADER_COMPUTE) - state = STATE_SHADER_COMPUTE_TODO; - else if (state == STATE_SHADER_PIXEL) - state = STATE_SHADER_PIXEL_TODO; - else if (state == STATE_SHADER_VERTEX) - state = STATE_SHADER_VERTEX_TODO; - else - state = STATE_SHADER_EFFECT_TODO; + state = get_parse_state_todo(state); + } + else if (match_directive_substring(src, "todo(sm<4)", &src)) + { + if (runner->selected_shader_model >= SHADER_MODEL_4_0) + continue; + state = get_parse_state_todo(state); + } + else if (match_directive_substring(src, "todo(sm>=4)", &src)) + { + if (runner->selected_shader_model >= SHADER_MODEL_6_0 + || runner->selected_shader_model < SHADER_MODEL_4_0) + continue; + state = get_parse_state_todo(state); } else if (match_directive_substring(src, "fail", &src)) { *expect_hr = E_FAIL; } + else if (match_directive_substring(src, "fail(sm<4)", &src)) + { + if (runner->selected_shader_model < SHADER_MODEL_4_0) + *expect_hr = E_FAIL; + } + else if (match_directive_substring(src, "fail(sm>=4)", &src)) + { + if (runner->selected_shader_model >= SHADER_MODEL_4_0) + *expect_hr = E_FAIL; + } else if (match_directive_substring(src, "fail(sm<6)", &src)) { if (runner->selected_shader_model < SHADER_MODEL_6_0) -- GitLab https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
From: Francisco Casas <fcasas(a)codeweavers.com> --- tests/hlsl/any.shader_test | 4 ++-- .../hlsl/arithmetic-float-uniform.shader_test | 8 +++---- tests/hlsl/arithmetic-int-uniform.shader_test | 16 +++++++------- tests/hlsl/arithmetic-int.shader_test | 2 ++ tests/hlsl/combined-samplers.shader_test | 4 ++-- tests/hlsl/conditional.shader_test | 8 +++---- tests/hlsl/ddxddy.shader_test | 2 +- tests/hlsl/discard.shader_test | 2 +- tests/hlsl/distance.shader_test | 2 +- tests/hlsl/expr-indexing.shader_test | 6 +++--- tests/hlsl/floor.shader_test | 4 ++-- tests/hlsl/fmod.shader_test | 4 ++-- tests/hlsl/for.shader_test | 4 ++-- tests/hlsl/function-return.shader_test | 12 ++++------- tests/hlsl/fwidth.shader_test | 2 +- tests/hlsl/half.shader_test | 2 +- tests/hlsl/lit.shader_test | 4 ++-- tests/hlsl/loop.shader_test | 14 ++++++------- tests/hlsl/matrix-indexing.shader_test | 2 +- tests/hlsl/non-const-indexing.shader_test | 4 ++-- tests/hlsl/return.shader_test | 21 +++++++------------ tests/hlsl/round.shader_test | 4 ++-- tests/hlsl/sample-bias.shader_test | 2 +- tests/hlsl/sample-level.shader_test | 2 +- tests/hlsl/sampler.shader_test | 4 ++-- tests/hlsl/saturate.shader_test | 2 +- tests/hlsl/sign.shader_test | 12 +++++------ tests/hlsl/smoothstep.shader_test | 2 +- tests/hlsl/static-initializer.shader_test | 5 +++-- tests/hlsl/step.shader_test | 2 +- tests/hlsl/ternary.shader_test | 8 +++---- tests/hlsl/trigonometry.shader_test | 8 +++---- tests/hlsl/trunc.shader_test | 4 ++-- .../hlsl/vector-indexing-uniform.shader_test | 2 +- tests/shader_runner.c | 6 +++--- 35 files changed, 91 insertions(+), 99 deletions(-) diff --git a/tests/hlsl/any.shader_test b/tests/hlsl/any.shader_test index 9b8b922c9..2b5bfbd0b 100644 --- a/tests/hlsl/any.shader_test +++ b/tests/hlsl/any.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 f; float4 main() : sv_target @@ -29,7 +29,7 @@ uniform 0 float4 -1.0 -1.0 -1.0 -1.0 todo(sm>=6) draw quad probe all rgba (1.0, 1.0, 1.0, 1.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float f; float4 main() : sv_target diff --git a/tests/hlsl/arithmetic-float-uniform.shader_test b/tests/hlsl/arithmetic-float-uniform.shader_test index 4812d053a..61d369497 100644 --- a/tests/hlsl/arithmetic-float-uniform.shader_test +++ b/tests/hlsl/arithmetic-float-uniform.shader_test @@ -13,7 +13,7 @@ uniform 0 float4 5.0 15.0 0.0 0.0 draw quad probe all rgba (20.0, -10.0, 75.0, 0.33333333) 1 -[pixel shader] +[pixel shader todo(sm<4)] uniform float2 a; float4 main() : SV_TARGET @@ -28,7 +28,7 @@ uniform 0 float4 5.0 15.0 0.0 0.0 draw quad probe all rgba (5.0, 5.0, -5.0, 3.0) 1 -[pixel shader] +[pixel shader todo(sm<4)] uniform float2 a; float4 main() : SV_TARGET @@ -43,7 +43,7 @@ uniform 0 float4 42.0 5.0 0.0 0.0 draw quad probe all rgba (2.0, -2.0, 2.0, -2.0) 16 -[pixel shader] +[pixel shader todo(sm<4)] uniform float2 a; float4 main() : SV_TARGET @@ -58,7 +58,7 @@ uniform 0 float4 45.0 5.0 0.0 0.0 draw quad probe all rgba (0.0, 0.0, 0.0, 0.0) -[pixel shader] +[pixel shader todo(sm<4)] float4 x, y; float4 main() : sv_target diff --git a/tests/hlsl/arithmetic-int-uniform.shader_test b/tests/hlsl/arithmetic-int-uniform.shader_test index 726a191a7..b0d538ed9 100644 --- a/tests/hlsl/arithmetic-int-uniform.shader_test +++ b/tests/hlsl/arithmetic-int-uniform.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float2 a; float4 main() : SV_TARGET @@ -13,7 +13,7 @@ uniform 0 float4 5.0 16.0 0.0 0.0 todo(sm>=6) draw quad probe all rgba (21.0, -11.0, 80.0, 0.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float2 a; float4 main() : SV_TARGET @@ -28,7 +28,7 @@ uniform 0 float4 5.0 16.0 0.0 0.0 todo(sm>=6) draw quad probe all rgba (5.0, 5.0, -5.0, 3.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float2 a; float4 main() : SV_TARGET @@ -43,7 +43,7 @@ uniform 0 float4 42.0 5.0 0.0 0.0 todo(sm>=6) draw quad probe all rgba (8.0, -8.0, -8.0, 8.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float2 a; float4 main() : SV_TARGET @@ -58,7 +58,7 @@ uniform 0 float4 42.0 5.0 0.0 0.0 todo(sm>=6) draw quad probe all rgba (2.0, -2.0, 2.0, -2.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float2 a; float4 main() : SV_TARGET @@ -73,7 +73,7 @@ uniform 0 float4 45.0 5.0 0.0 0.0 todo(sm>=6) draw quad probe all rgba (9.0, -9.0, -9.0, 9.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float2 a; float4 main() : SV_TARGET @@ -88,7 +88,7 @@ uniform 0 float4 45.0 5.0 0.0 0.0 todo(sm>=6) draw quad probe all rgba (0.0, 0.0, 0.0, 0.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 a; float4 main() : SV_TARGET @@ -101,7 +101,7 @@ uniform 0 float4 5.0 -7.0 0.0 -10.0 todo(sm>=6) draw quad probe all rgba (5.0, 7.0, 0.0, 10.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 a; uniform float4 b; diff --git a/tests/hlsl/arithmetic-int.shader_test b/tests/hlsl/arithmetic-int.shader_test index f2044c42c..46b641811 100644 --- a/tests/hlsl/arithmetic-int.shader_test +++ b/tests/hlsl/arithmetic-int.shader_test @@ -77,6 +77,7 @@ draw quad probe all rgba (0.0, 0.0, 0.0, 0.0) [pixel shader fail(sm<6)] +// On SM1 this gives hr 0x88760b59. float4 main() : SV_TARGET { int x = 1; @@ -90,6 +91,7 @@ draw quad probe all rgba (0.0, 0.0, 0.0, 0.0) [pixel shader fail(sm<6)] +// On SM1 this gives hr 0x88760b59. float4 main() : SV_TARGET { int x = 1; diff --git a/tests/hlsl/combined-samplers.shader_test b/tests/hlsl/combined-samplers.shader_test index 235537594..4d743aecf 100644 --- a/tests/hlsl/combined-samplers.shader_test +++ b/tests/hlsl/combined-samplers.shader_test @@ -64,7 +64,7 @@ draw quad probe all rgba (10, 10, 10, 11) -[pixel shader] +[pixel shader todo(sm<4)] Texture2D tex; sampler sam[2]; @@ -111,7 +111,7 @@ probe all rgba (104, 104, 104, 111) % Sampler arrays with components that have different usage dimensions are only forbidden in SM4 upwards. % However, tex2D and tex1D are considered the same dimension for these purposes. -[pixel shader fail] +[pixel shader fail(sm>=4)] sampler sam[2]; float4 main() : sv_target diff --git a/tests/hlsl/conditional.shader_test b/tests/hlsl/conditional.shader_test index 4175813bd..f06fcc4d1 100644 --- a/tests/hlsl/conditional.shader_test +++ b/tests/hlsl/conditional.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target @@ -17,7 +17,7 @@ uniform 0 float4 0.1 0.0 0.0 0.0 todo(sm>=6) draw quad probe all rgba (0.1, 0.2, 0.3, 0.4) -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target @@ -29,7 +29,7 @@ float4 main() : sv_target return float4(0.9, 0.8, 0.7, 0.6); } -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target @@ -74,7 +74,7 @@ float main() : sv_target [require] shader model >= 3.0 -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target diff --git a/tests/hlsl/ddxddy.shader_test b/tests/hlsl/ddxddy.shader_test index ba7215ca0..e31456cb4 100644 --- a/tests/hlsl/ddxddy.shader_test +++ b/tests/hlsl/ddxddy.shader_test @@ -12,7 +12,7 @@ todo(sm>=6) draw quad probe all rgba (1.0, 1.0, 0.0, 0.0) -[pixel shader] +[pixel shader todo(sm<4)] float4 main(float4 pos : sv_position) : sv_target { // Shader models < 4 don't add 0.5 to sv_position, so this adjustment is required to get the diff --git a/tests/hlsl/discard.shader_test b/tests/hlsl/discard.shader_test index cecb8c1fd..956cfe574 100644 --- a/tests/hlsl/discard.shader_test +++ b/tests/hlsl/discard.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 x; float4 main() : sv_target diff --git a/tests/hlsl/distance.shader_test b/tests/hlsl/distance.shader_test index 6527c218a..b4022aa5d 100644 --- a/tests/hlsl/distance.shader_test +++ b/tests/hlsl/distance.shader_test @@ -13,7 +13,7 @@ uniform 4 float4 2.0 -1.0 4.0 5.0 todo(sm>=6) draw quad probe all rgba (7.483983, 7.483983, 7.483983, 7.483983) 1 -[pixel shader] +[pixel shader todo(sm<4)] uniform int4 x; uniform int4 y; diff --git a/tests/hlsl/expr-indexing.shader_test b/tests/hlsl/expr-indexing.shader_test index c11fa6540..6e314f070 100644 --- a/tests/hlsl/expr-indexing.shader_test +++ b/tests/hlsl/expr-indexing.shader_test @@ -13,7 +13,7 @@ draw quad probe all rgba (8.0, 8.0, 8.0, 8.0) -[pixel shader] +[pixel shader todo(sm<4)] float4 a, b; float i; @@ -44,7 +44,7 @@ draw quad probe all rgba (3.0, 3.0, 3.0, 3.0) -[pixel shader] +[pixel shader todo(sm<4)] float4 a; float i; @@ -82,7 +82,7 @@ draw quad probe all rgba (4.0, 4.0, 4.0, 4.0) -[pixel shader] +[pixel shader todo(sm<4)] float4 a; float i; diff --git a/tests/hlsl/floor.shader_test b/tests/hlsl/floor.shader_test index deb9acf91..c40bee733 100644 --- a/tests/hlsl/floor.shader_test +++ b/tests/hlsl/floor.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target @@ -11,7 +11,7 @@ uniform 0 float4 -0.5 6.5 7.5 3.4 todo(sm>=6) draw quad probe all rgba (-1.0, 6.0, 7.0, 3.0) 4 -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target diff --git a/tests/hlsl/fmod.shader_test b/tests/hlsl/fmod.shader_test index 05518c7cd..70e993899 100644 --- a/tests/hlsl/fmod.shader_test +++ b/tests/hlsl/fmod.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target @@ -14,7 +14,7 @@ uniform 0 float4 1.1 0.3 0.0 0.0 todo(sm>=6) draw quad probe all rgba (0.2, 0.0, 0.0, 0.0) 4 -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target diff --git a/tests/hlsl/for.shader_test b/tests/hlsl/for.shader_test index e5faa28fd..e7a379c65 100644 --- a/tests/hlsl/for.shader_test +++ b/tests/hlsl/for.shader_test @@ -4,7 +4,7 @@ void main(out float tex : texcoord, inout float4 pos : sv_position) tex = pos.x; } -[pixel shader] +[pixel shader todo(sm<4)] float4 main(float tex : texcoord) : sv_target { int i; @@ -27,7 +27,7 @@ probe ( 0, 0, 159, 480) rgba (10.0, 35.0, 0.0, 0.0) probe (161, 0, 479, 480) rgba (10.0, 38.0, 0.0, 0.0) probe (481, 0, 640, 480) rgba ( 5.0, 10.0, 0.0, 0.0) -[pixel shader] +[pixel shader todo(sm<4)] float4 main(float tex : texcoord) : sv_target { int i; diff --git a/tests/hlsl/function-return.shader_test b/tests/hlsl/function-return.shader_test index be997d0c3..038fb8d4f 100644 --- a/tests/hlsl/function-return.shader_test +++ b/tests/hlsl/function-return.shader_test @@ -32,8 +32,7 @@ float4 main() : sv_target draw quad probe all rgba (0.2, 0.1, 0.8, 0.5); -[pixel shader] - +[pixel shader todo(sm<4)] uniform float f; float func(out float o) @@ -92,8 +91,7 @@ uniform 0 float 0.8 todo(sm>=6) draw quad probe all rgba (0.8, 0.7, 0.4, 0.5) 1 -[pixel shader] - +[pixel shader todo(sm<4)] uniform float f; float func(out float o) @@ -145,8 +143,7 @@ uniform 0 float 0.9 todo(sm>=6) draw quad probe all rgba (1.0, 0.9, 1.0, 0.6) 1 -[pixel shader] - +[pixel shader todo(sm<4)] float func(out float o) { o = 0.1; @@ -187,8 +184,7 @@ float4 main() : sv_target draw quad probe all rgba (0.4, 0.3, 0.3, 0.9) 1 -[pixel shader] - +[pixel shader todo(sm<4)] uniform float f; float func(out float o) diff --git a/tests/hlsl/fwidth.shader_test b/tests/hlsl/fwidth.shader_test index d4c20f4e0..04b47222e 100644 --- a/tests/hlsl/fwidth.shader_test +++ b/tests/hlsl/fwidth.shader_test @@ -1,7 +1,7 @@ [require] shader model >= 3.0 -[pixel shader] +[pixel shader todo(sm<4)] float4 main(float4 pos : sv_position) : sv_target { // Shader models < 4 don't add 0.5 to sv_position, so this adjustment is required to get the diff --git a/tests/hlsl/half.shader_test b/tests/hlsl/half.shader_test index fe7074e45..2497ac36a 100644 --- a/tests/hlsl/half.shader_test +++ b/tests/hlsl/half.shader_test @@ -9,7 +9,7 @@ float4 main() : sv_target [require] options: backcompat -[pixel shader] +[pixel shader todo(sm<4)] uniform half h; float4 main() : sv_target diff --git a/tests/hlsl/lit.shader_test b/tests/hlsl/lit.shader_test index 5014c1ed0..0802c2718 100644 --- a/tests/hlsl/lit.shader_test +++ b/tests/hlsl/lit.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target @@ -21,7 +21,7 @@ uniform 0 float4 1.2 2.0 3.0 0.0 todo(sm>=6) draw quad probe all rgba (1.0, 1.2, 8.0, 1.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target diff --git a/tests/hlsl/loop.shader_test b/tests/hlsl/loop.shader_test index 35a303595..9c142c1c3 100644 --- a/tests/hlsl/loop.shader_test +++ b/tests/hlsl/loop.shader_test @@ -1,6 +1,6 @@ % TODO: dxcompiler emits no loops for any of these test shaders. -[pixel shader] +[pixel shader todo(sm<4)] float a; float4 main() : sv_target @@ -22,7 +22,7 @@ draw quad probe all rgba (50.0, 50.0, 50.0, 50.0) -[pixel shader] +[pixel shader todo(sm<4)] float a; float4 main() : sv_target @@ -44,7 +44,7 @@ uniform 0 float 4.0 draw quad probe all rgba (20.0, 20.0, 20.0, 20.0) -[pixel shader] +[pixel shader todo(sm<4)] float a; float4 main() : sv_target @@ -73,7 +73,7 @@ uniform 0 float 4.0 draw quad probe all rgba (409.1, 409.1, 409.1, 409.1) -[pixel shader] +[pixel shader todo(sm<4)] float a; float4 main() : sv_target @@ -104,7 +104,7 @@ draw quad probe all rgba (410.1, 410.1, 410.1, 410.1) % loop attribute by itself -[pixel shader] +[pixel shader todo(sm<4)] float4 main() : sv_target { float ret = 0.0; @@ -121,7 +121,7 @@ float4 main() : sv_target todo(sm>=6) draw quad probe all rgba (10.0, 10.0, 10.0, 10.0) -[pixel shader] +[pixel shader todo(sm<4)] float4 main() : sv_target { float ret = 0.0; @@ -140,7 +140,7 @@ float4 main() : sv_target todo(sm>=6) draw quad probe all rgba (10.0, 10.0, 10.0, 10.0) -[pixel shader] +[pixel shader todo(sm<4)] float4 main() : sv_target { float ret = 0.0; diff --git a/tests/hlsl/matrix-indexing.shader_test b/tests/hlsl/matrix-indexing.shader_test index b8e6dec68..24f81357e 100644 --- a/tests/hlsl/matrix-indexing.shader_test +++ b/tests/hlsl/matrix-indexing.shader_test @@ -108,7 +108,7 @@ draw quad probe all rgba (3.0, 4.0, 50.0, 60.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float i; float4 main() : sv_target diff --git a/tests/hlsl/non-const-indexing.shader_test b/tests/hlsl/non-const-indexing.shader_test index bc550738c..488d75534 100644 --- a/tests/hlsl/non-const-indexing.shader_test +++ b/tests/hlsl/non-const-indexing.shader_test @@ -25,7 +25,7 @@ todo(sm>=6) draw quad todo(sm>=6) probe all rgba (9.0, 10.0, 11.0, 12.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float i; float4 main() : SV_TARGET @@ -65,7 +65,7 @@ todo(sm>=6) draw quad todo(sm>=6) probe all rgba (3, 3, 3, 3) -[pixel shader] +[pixel shader todo(sm<4)] uniform float i; float4 main() : SV_TARGET diff --git a/tests/hlsl/return.shader_test b/tests/hlsl/return.shader_test index 29621b006..3d73e728a 100644 --- a/tests/hlsl/return.shader_test +++ b/tests/hlsl/return.shader_test @@ -25,8 +25,7 @@ void main(out float4 ret : sv_target) draw quad probe all rgba (0.1, 0.2, 0.3, 0.4) -[pixel shader] - +[pixel shader todo(sm<4)] uniform float f; float4 main() : sv_target @@ -44,8 +43,7 @@ uniform 0 float 0.8 todo(sm>=6) draw quad probe all rgba (0.5, 0.6, 0.7, 0.8) -[pixel shader] - +[pixel shader todo(sm<4)] uniform float f; void main(out float4 ret : sv_target) @@ -71,8 +69,7 @@ uniform 0 float 0.8 todo(sm>=6) draw quad probe all rgba (0.1, 0.2, 0.3, 0.4) -[pixel shader] - +[pixel shader todo(sm<4)] uniform float f; void main(out float4 ret : sv_target) @@ -101,8 +98,7 @@ uniform 0 float 0.9 todo(sm>=6) draw quad probe all rgba (0.5, 0.6, 0.7, 0.8) 1 -[pixel shader] - +[pixel shader todo(sm<4)] uniform float f; void main(out float4 ret : sv_target) @@ -128,8 +124,7 @@ uniform 0 float 0.9 todo(sm>=6) draw quad probe all rgba (0.4, 0.5, 0.6, 0.7) 1 -[pixel shader] - +[pixel shader todo(sm<4)] void main(out float4 ret : sv_target) { ret = float4(0.1, 0.2, 0.3, 0.4); @@ -146,8 +141,7 @@ void main(out float4 ret : sv_target) draw quad probe all rgba (0.2, 0.4, 0.6, 0.8) -[pixel shader] - +[pixel shader todo(sm<4)] uniform float f; void main(out float4 ret : sv_target) @@ -185,8 +179,7 @@ uniform 0 float 0.9 todo(sm>=6) draw quad probe all rgba (0.9, 0.9, 0.9, 0.9) 1 -[pixel shader] - +[pixel shader todo(sm<4)] uniform float f; void main(out float4 ret : sv_target) diff --git a/tests/hlsl/round.shader_test b/tests/hlsl/round.shader_test index 2a32b3ec2..51e273634 100644 --- a/tests/hlsl/round.shader_test +++ b/tests/hlsl/round.shader_test @@ -13,7 +13,7 @@ probe all rgba (0.0, -7.0, 8.0, 3.0) 4 -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target @@ -31,7 +31,7 @@ probe all rgba (-7.0, 8.0, 0.0, 3.0) 4 -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target diff --git a/tests/hlsl/sample-bias.shader_test b/tests/hlsl/sample-bias.shader_test index e56945d06..3f5b28095 100644 --- a/tests/hlsl/sample-bias.shader_test +++ b/tests/hlsl/sample-bias.shader_test @@ -17,7 +17,7 @@ void main(out float2 tex : texcoord, inout float4 pos : sv_position) tex = pos.xy; } -[pixel shader] +[pixel shader todo(sm<4)] sampler s; Texture2D t; uniform float bias; diff --git a/tests/hlsl/sample-level.shader_test b/tests/hlsl/sample-level.shader_test index c91c65493..8ce855274 100644 --- a/tests/hlsl/sample-level.shader_test +++ b/tests/hlsl/sample-level.shader_test @@ -14,7 +14,7 @@ levels 2 0.0 0.0 1.0 0.0 -[pixel shader] +[pixel shader todo(sm<4)] sampler s; Texture2D t; uniform float level; diff --git a/tests/hlsl/sampler.shader_test b/tests/hlsl/sampler.shader_test index caff7b2fa..d857c4b18 100644 --- a/tests/hlsl/sampler.shader_test +++ b/tests/hlsl/sampler.shader_test @@ -7,7 +7,7 @@ size (2, 2) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 0.0 1.0 0.0 -[pixel shader] +[pixel shader todo(sm<4)] sampler s; Texture2D t; @@ -20,7 +20,7 @@ float4 main() : sv_target todo(sm>=6) draw quad probe all rgba (0.25, 0, 0.25, 0) -[pixel shader] +[pixel shader todo(sm<4)] SamplerState s; Texture2D t; diff --git a/tests/hlsl/saturate.shader_test b/tests/hlsl/saturate.shader_test index 2ed83cf66..56f84c773 100644 --- a/tests/hlsl/saturate.shader_test +++ b/tests/hlsl/saturate.shader_test @@ -11,7 +11,7 @@ uniform 0 float4 0.7 -0.1 0.0 0.0 todo(sm>=6) draw quad probe all rgba (0.7, 0.0, 1.0, 0.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target diff --git a/tests/hlsl/sign.shader_test b/tests/hlsl/sign.shader_test index 54b953681..aaba3e69a 100644 --- a/tests/hlsl/sign.shader_test +++ b/tests/hlsl/sign.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float f; float4 main() : sv_target @@ -17,7 +17,7 @@ uniform 0 float4 0.0 0.0 0.0 0.0 todo(sm>=6) draw quad probe all rgba (0.0, 0.0, 0.0, 0.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 f; float4 main() : sv_target @@ -30,7 +30,7 @@ uniform 0 float4 1.0 2.0 3.0 4.0 todo(sm>=6) draw quad probe all rgba (1.0, 1.0, 1.0, 1.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float2x2 f; float4 main() : sv_target @@ -48,7 +48,7 @@ probe all rgba (1.0, 1.0, 1.0, 1.0) % SM1-3 doesn't support integral types shader model >= 4.0 -[pixel shader] +[pixel shader todo(sm<4)] uniform int f; float4 main() : sv_target @@ -67,7 +67,7 @@ uniform 0 int4 0 0 0 0 todo(sm>=6) draw quad probe all rgba (0, 0, 0, 0) -[pixel shader] +[pixel shader todo(sm<4)] uniform int4 f; float4 main() : sv_target @@ -80,7 +80,7 @@ uniform 0 int4 1 2 3 4 todo(sm>=6) draw quad probe all rgba (1, 1, 1, 1) -[pixel shader] +[pixel shader todo(sm<4)] uniform int2x2 f; float4 main() : sv_target diff --git a/tests/hlsl/smoothstep.shader_test b/tests/hlsl/smoothstep.shader_test index 971f6d5d8..38e1432ca 100644 --- a/tests/hlsl/smoothstep.shader_test +++ b/tests/hlsl/smoothstep.shader_test @@ -109,7 +109,7 @@ draw quad probe all rgba (0.028, 0.104, 0.216, 0.352) 6 -[pixel shader] +[pixel shader todo(sm<4)] // 4 division by zero warnings. // Only test compilation because result is implementation-dependent. float4 main() : sv_target diff --git a/tests/hlsl/static-initializer.shader_test b/tests/hlsl/static-initializer.shader_test index 217444308..aec1dac1e 100644 --- a/tests/hlsl/static-initializer.shader_test +++ b/tests/hlsl/static-initializer.shader_test @@ -17,6 +17,7 @@ probe all rgba (0.8, 0.0, 0.0, 0.0) [pixel shader fail(sm<6)] +// On SM1 this gives hr 0x88760b59. static uint i; float4 main() : sv_target @@ -135,7 +136,7 @@ float4 main(Texture2D tex2) : sv_target } -[pixel shader] +[pixel shader todo(sm<4)] Texture2D real_tex; static Texture2D tex = real_tex; sampler sam; @@ -150,7 +151,7 @@ todo(sm>=6) draw quad probe all rgba (1, 2, 3, 4) -[pixel shader] +[pixel shader todo(sm<4)] Texture2D real_tex; static Texture2D tex; sampler sam; diff --git a/tests/hlsl/step.shader_test b/tests/hlsl/step.shader_test index b857f00b2..6dc147bd7 100644 --- a/tests/hlsl/step.shader_test +++ b/tests/hlsl/step.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 f, p; float4 main() : sv_target diff --git a/tests/hlsl/ternary.shader_test b/tests/hlsl/ternary.shader_test index 587b3d556..a695984bf 100644 --- a/tests/hlsl/ternary.shader_test +++ b/tests/hlsl/ternary.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 x; float4 main() : sv_target @@ -14,7 +14,7 @@ uniform 0 float4 0.0 10.0 11.0 12.0 todo(sm>=6) draw quad probe all rgba (-1.0, 9.0, 10.0, 11.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 x; float4 main() : sv_target @@ -32,7 +32,7 @@ uniform 0 float4 1.1 3.0 4.0 5.0 todo(sm>=6) draw quad probe all rgba (1.1, 2.0, 0.0, 0.0) -[pixel shader] +[pixel shader todo(sm<4)] float4 f; float4 main() : sv_target @@ -47,7 +47,7 @@ uniform 0 float4 1.0 0.0 0.0 0.0 todo(sm>=6) draw quad probe all rgba (0.5, 0.6, 0.7, 0.0) -[pixel shader fail(sm>=6)] +[pixel shader fail(sm>=6) todo(sm<4)] float4 x, y, z; float4 main() : sv_target diff --git a/tests/hlsl/trigonometry.shader_test b/tests/hlsl/trigonometry.shader_test index f52d01dea..5736696f6 100644 --- a/tests/hlsl/trigonometry.shader_test +++ b/tests/hlsl/trigonometry.shader_test @@ -4,7 +4,7 @@ void main(out float tex : texcoord, inout float4 pos : sv_position) tex = (pos.x + 1) * 320; } -[pixel shader] +[pixel shader todo(sm<4)] float4 main(float tex : texcoord) : sv_target { tex = floor(tex + 0.25); @@ -31,7 +31,7 @@ probe (14, 0) rgba ( 0.99060736, 0.13673722, 7.24460662, 0.0) 1024 probe (15, 0) rgba ( 0.65028784, -0.75968791, -0.85599340, 0.0) 1024 -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 a; float4 main() : sv_target @@ -45,7 +45,7 @@ todo(sm>=6) draw quad probe all rgba (0.0, 500.0, 500.0, 0.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 a; float4 main() : sv_target @@ -59,7 +59,7 @@ todo(sm>=6) draw quad probe all rgba (1000.0, 707.0, -0.0, -707.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 a; float4 main() : sv_target diff --git a/tests/hlsl/trunc.shader_test b/tests/hlsl/trunc.shader_test index a359cd03a..85631f25e 100644 --- a/tests/hlsl/trunc.shader_test +++ b/tests/hlsl/trunc.shader_test @@ -1,4 +1,4 @@ -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target @@ -14,7 +14,7 @@ uniform 0 float4 -1.5 6.5 7.5 3.4 todo(sm>=6) draw quad probe all rgba (-1.0, 6.0, 7.0, 3.0) -[pixel shader] +[pixel shader todo(sm<4)] uniform float4 u; float4 main() : sv_target diff --git a/tests/hlsl/vector-indexing-uniform.shader_test b/tests/hlsl/vector-indexing-uniform.shader_test index 3501f3af7..63e7ec054 100644 --- a/tests/hlsl/vector-indexing-uniform.shader_test +++ b/tests/hlsl/vector-indexing-uniform.shader_test @@ -1,6 +1,6 @@ % Use a uniform to prevent the compiler from optimizing. -[pixel shader] +[pixel shader todo(sm<4)] uniform float i; float4 main() : SV_TARGET { diff --git a/tests/shader_runner.c b/tests/shader_runner.c index 745f17b3a..a48f58003 100644 --- a/tests/shader_runner.c +++ b/tests/shader_runner.c @@ -805,7 +805,7 @@ unsigned int get_vb_stride(const struct shader_runner *runner, unsigned int slot static HRESULT map_unidentified_hrs(HRESULT hr) { - if (hr == 0x80010064) + if (hr == 0x80010064 || hr == 0x88760b59) { trace("Mapping unindentified hr %#x as %#x.\n", hr, E_FAIL); return E_FAIL; @@ -951,8 +951,8 @@ static void compile_shader(struct shader_runner *runner, IDxcCompiler3 *dxc_comp static const char *const shader_models[] = { - [SHADER_MODEL_2_0] = "4_0", - [SHADER_MODEL_3_0] = "4_0", + [SHADER_MODEL_2_0] = "2_0", + [SHADER_MODEL_3_0] = "3_0", [SHADER_MODEL_4_0] = "4_0", [SHADER_MODEL_4_1] = "4_1", [SHADER_MODEL_5_0] = "5_0", -- GitLab https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
From: Francisco Casas <fcasas(a)codeweavers.com> --- Makefile.am | 2 +- ....shader_test => register-reservations-resources.shader_test} | 0 2 files changed, 1 insertion(+), 1 deletion(-) rename tests/hlsl/{register-reservations.shader_test => register-reservations-resources.shader_test} (100%) diff --git a/Makefile.am b/Makefile.am index 14b5cf096..dbaac9373 100644 --- a/Makefile.am +++ b/Makefile.am @@ -143,7 +143,7 @@ vkd3d_shader_tests = \ tests/hlsl/object-references.shader_test \ tests/hlsl/pow.shader_test \ tests/hlsl/reflect.shader_test \ - tests/hlsl/register-reservations.shader_test \ + tests/hlsl/register-reservations-resources.shader_test \ tests/hlsl/return-implicit-conversion.shader_test \ tests/hlsl/return.shader_test \ tests/hlsl/round.shader_test \ diff --git a/tests/hlsl/register-reservations.shader_test b/tests/hlsl/register-reservations-resources.shader_test similarity index 100% rename from tests/hlsl/register-reservations.shader_test rename to tests/hlsl/register-reservations-resources.shader_test -- GitLab https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
From: Francisco Casas <fcasas(a)codeweavers.com> --- Makefile.am | 1 + .../register-reservations-numeric.shader_test | 237 ++++++++++++++++++ 2 files changed, 238 insertions(+) create mode 100644 tests/hlsl/register-reservations-numeric.shader_test diff --git a/Makefile.am b/Makefile.am index dbaac9373..b27211540 100644 --- a/Makefile.am +++ b/Makefile.am @@ -143,6 +143,7 @@ vkd3d_shader_tests = \ tests/hlsl/object-references.shader_test \ tests/hlsl/pow.shader_test \ tests/hlsl/reflect.shader_test \ + tests/hlsl/register-reservations-numeric.shader_test \ tests/hlsl/register-reservations-resources.shader_test \ tests/hlsl/return-implicit-conversion.shader_test \ tests/hlsl/return.shader_test \ diff --git a/tests/hlsl/register-reservations-numeric.shader_test b/tests/hlsl/register-reservations-numeric.shader_test new file mode 100644 index 000000000..ad11669a2 --- /dev/null +++ b/tests/hlsl/register-reservations-numeric.shader_test @@ -0,0 +1,237 @@ +[pixel shader fail(sm<6) todo] +// Overlapping register(cX) reservations are not allowed except on SM6, where they are aliased. +// On SM1 this gives hr 0x88760b59. +float a : register(c0); +float b : register(c0); + +float4 main() : sv_target +{ + return a + b; +} + + +[pixel shader] +// It is not required to provide a register(cX) for all elements in the $Globals buffer. +float4 a; // will get register(c1) +float4 b : register(c0); + +float4 main() : sv_target +{ + return float4(a.xw, b.yz); +} + +[test] +uniform 0 float4 0.1 0.2 0.3 0.4 +uniform 4 float4 1.1 1.2 1.3 1.4 +draw quad +todo(sm<6) probe all rgba (1.1, 1.4, 0.2, 0.3) + + +[pixel shader] +float4 a[3]; // will get register(c3) +float4 b[2] : register(c1); + +float4 main() : sv_target +{ + return float4(a[1].xy, b[0].zw); +} + +[test] +uniform 0 float4 0.1 0.2 0.3 0.4 +uniform 4 float4 1.1 1.2 1.3 1.4 +uniform 8 float4 2.1 2.2 2.3 2.4 +uniform 12 float4 3.1 3.2 3.3 3.4 +uniform 16 float4 4.1 4.2 4.3 4.4 +draw quad +todo(sm<6) probe all rgba (4.1, 4.2, 1.3, 1.4) + + +[require] +shader model >= 6.0 + +[pixel shader] +// Variables with overlapping register(cX) reservations are aliased in SM6. +float2 a : register(c2); +float3 b : register(c2); + +float4 main() : sv_target +{ + return float4(a, b.yz); +} + +[test] +uniform 0 float4 0.1 0.2 0.3 0.4 +uniform 4 float4 1.1 1.2 1.3 1.4 +uniform 8 float4 2.1 2.2 2.3 2.4 +draw quad +probe all rgba (2.1, 2.2, 2.2, 2.3) + + +% Results differ between SM1 and SM4 because in the latter variables can share the same register, +% using different writemasks. +[require] +shader model < 4.0 + +[pixel shader] +struct +{ + float2 a; + float b; +} apple : register(c2); + +float4 main() : sv_target +{ + return float4(apple.a, apple.b, 0); +} + +[test] +uniform 0 float4 0.1 0.2 0.3 0.4 +uniform 4 float4 1.1 1.2 1.3 1.4 +uniform 8 float4 2.1 2.2 2.3 2.4 +uniform 12 float4 3.1 3.2 3.3 3.4 +draw quad +todo probe all rgba (2.1, 2.2, 3.1, 0.0) + + +[require] +shader model >= 4.0 + +[pixel shader] +struct +{ + float2 a; + float b; +} apple : register(c2); + +float4 main() : sv_target +{ + return float4(apple.a, apple.b, 0); +} + +[test] +uniform 0 float4 0.1 0.2 0.3 0.4 +uniform 4 float4 1.1 1.2 1.3 1.4 +uniform 8 float4 2.1 2.2 2.3 2.4 +uniform 12 float4 3.1 3.2 3.3 3.4 +draw quad +todo(sm<6) probe all rgba (2.1, 2.2, 2.3, 0.0) + + +[pixel shader] +// On SM4, register(cX) has no effect unless in the $Globals buffer. +cbuffer extra +{ + float a : register(c1); +}; + +float4 main() : sv_target +{ + return a; +} + +[test] +uniform 0 float 100 +uniform 4 float 101 +draw quad +probe all rgba (100, 100, 100, 100) + + +[pixel shader fail(sm>=6)] +// On SM4 register(cX) has no effect unless in the $Globals buffer. +float4 main(uniform float a : register(c1)) : sv_target +{ + return a; +} + +[test] +uniform 0 float 100 +uniform 4 float 101 +draw quad +probe all rgba (100, 100, 100, 100) + +[pixel shader todo] +cbuffer c +{ + float a : packoffset(c1); + float b : packoffset(c2) : register(c1); + // ^ register(c1) ignored for cbuffer that is not $Globals. +} + +float4 main() : sv_target +{ + return float4(a, b, 0, 0); +} + +[test] +uniform 0 float 200 +uniform 4 float 201 +uniform 8 float 202 +todo(sm<6) draw quad +todo(sm<6) probe all rgba (201.0, 202.0, 0.0, 0.0) + + +[pixel shader fail(sm<4)] +int k : register(i0); // register(cX) is also required. + +float4 main() : sv_target +{ + return k; +} + + +[require] +% All shader models. + +% In SM1, most variables are needed in the "c" register group, for float operations. +% If a variable is needed in the "c" register group, register() reservations in other groups can be +% provided only if a register(cX) reservation is also provided. + +[pixel shader fail(sm<4)] +int k : register(i0); +// ^^ register(cX) is also required in SM1. + +float4 main() : sv_target +{ + return k; +} + +[pixel shader todo] +int k : register(i0) : register(c1); +// Shader compiles because a "c" register reservation is provided for "k". + +float4 main() : sv_target +{ + return k; +} + + +[require] +shader model >= 3.0 +% model 2.0 doesn't support unrollable loops. + +[pixel shader todo(sm<4)] +int k : register(i0); +// ^^ register(cX) is not required since "k" is just needed in the "i" register group. + +float4 main() : sv_target +{ + float f = 0; + + for (int i = 0; i < k; ++i) + f += i; + return f; +} + + +[pixel shader todo] +int k : register(c0) : register(b0); +// ^^ unlike the "c" register group, a reservation is not required for the "i" group, even though "k" is needed on it. + +float4 main() : sv_target +{ + float f = 0; + + for (int i = 0; i < k; ++i) + f += i; + return f; +} -- GitLab https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
From: Francisco Casas <fcasas(a)codeweavers.com> --- libs/vkd3d-shader/hlsl_codegen.c | 41 ++++++++++++++++++++++++++++---- 1 file changed, 37 insertions(+), 4 deletions(-) diff --git a/libs/vkd3d-shader/hlsl_codegen.c b/libs/vkd3d-shader/hlsl_codegen.c index 76a010fb3..5ea6df57c 100644 --- a/libs/vkd3d-shader/hlsl_codegen.c +++ b/libs/vkd3d-shader/hlsl_codegen.c @@ -3947,13 +3947,46 @@ static void allocate_const_registers(struct hlsl_ctx *ctx, struct hlsl_ir_functi LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) { - if (var->is_uniform && var->last_read) + unsigned int reg_size = var->data_type->reg_size[HLSL_REGSET_NUMERIC]; + + if (!var->is_uniform || !var->last_read || reg_size == 0) + continue; + + if (var->reg_reservation.reg_type == 'c') { - unsigned int reg_size = var->data_type->reg_size[HLSL_REGSET_NUMERIC]; + unsigned int reg_idx = var->reg_reservation.reg_index; + unsigned int i; - if (reg_size == 0) - continue; + assert(reg_size % 4 == 0); + for (i = 0; i < reg_size / 4; ++i) + { + if (get_available_writemask(&allocator, 1, UINT_MAX, reg_idx + i) != VKD3DSP_WRITEMASK_ALL) + { + hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_INVALID_RESERVATION, + "Overlapping register() reservations on 'c%u'.", reg_idx + i); + } + + record_allocation(ctx, &allocator, reg_idx + i, VKD3DSP_WRITEMASK_ALL, 1, UINT_MAX); + } + var->regs[HLSL_REGSET_NUMERIC].id = reg_idx; + var->regs[HLSL_REGSET_NUMERIC].allocation_size = reg_size / 4; + var->regs[HLSL_REGSET_NUMERIC].writemask = VKD3DSP_WRITEMASK_ALL; + var->regs[HLSL_REGSET_NUMERIC].allocated = true; + TRACE("Allocated reserved %s to %s.\n", var->name, + debug_register('c', var->regs[HLSL_REGSET_NUMERIC], var->data_type)); + } + } + + LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) + { + unsigned int reg_size = var->data_type->reg_size[HLSL_REGSET_NUMERIC]; + + if (!var->is_uniform || !var->last_read || reg_size == 0) + continue; + + if (!var->regs[HLSL_REGSET_NUMERIC].allocated) + { var->regs[HLSL_REGSET_NUMERIC] = allocate_numeric_registers_for_type(ctx, &allocator, 1, UINT_MAX, var->data_type); TRACE("Allocated %s to %s.\n", var->name, -- GitLab https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
From: Francisco Casas <fcasas(a)codeweavers.com> --- libs/vkd3d-shader/hlsl_codegen.c | 96 ++++++++++++------- .../register-reservations-numeric.shader_test | 8 +- 2 files changed, 66 insertions(+), 38 deletions(-) diff --git a/libs/vkd3d-shader/hlsl_codegen.c b/libs/vkd3d-shader/hlsl_codegen.c index 5ea6df57c..80b75a88d 100644 --- a/libs/vkd3d-shader/hlsl_codegen.c +++ b/libs/vkd3d-shader/hlsl_codegen.c @@ -4124,45 +4124,52 @@ static const struct hlsl_buffer *get_reserved_buffer(struct hlsl_ctx *ctx, uint3 return NULL; } -static void calculate_buffer_offset(struct hlsl_ctx *ctx, struct hlsl_ir_var *var) +static void calculate_buffer_offset(struct hlsl_ctx *ctx, struct hlsl_ir_var *var, bool register_reservation) { unsigned int var_reg_size = var->data_type->reg_size[HLSL_REGSET_NUMERIC]; enum hlsl_type_class var_class = var->data_type->class; struct hlsl_buffer *buffer = var->buffer; - if (var->reg_reservation.offset_type == 'c') + if (register_reservation) { - if (var->reg_reservation.offset_index % 4) + var->buffer_offset = 4 * var->reg_reservation.reg_index; + } + else + { + if (var->reg_reservation.offset_type == 'c') { - if (var_class == HLSL_CLASS_MATRIX) - { - hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_INVALID_RESERVATION, - "packoffset() reservations with matrix types must be aligned with the beginning of a register."); - } - else if (var_class == HLSL_CLASS_ARRAY) - { - hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_INVALID_RESERVATION, - "packoffset() reservations with array types must be aligned with the beginning of a register."); - } - else if (var_class == HLSL_CLASS_STRUCT) - { - hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_INVALID_RESERVATION, - "packoffset() reservations with struct types must be aligned with the beginning of a register."); - } - else if (var_class == HLSL_CLASS_VECTOR) + if (var->reg_reservation.offset_index % 4) { - unsigned int aligned_offset = hlsl_type_get_sm4_offset(var->data_type, var->reg_reservation.offset_index); - - if (var->reg_reservation.offset_index != aligned_offset) + if (var_class == HLSL_CLASS_MATRIX) + { hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_INVALID_RESERVATION, - "packoffset() reservations with vector types cannot span multiple registers."); + "packoffset() reservations with matrix types must be aligned with the beginning of a register."); + } + else if (var_class == HLSL_CLASS_ARRAY) + { + hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_INVALID_RESERVATION, + "packoffset() reservations with array types must be aligned with the beginning of a register."); + } + else if (var_class == HLSL_CLASS_STRUCT) + { + hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_INVALID_RESERVATION, + "packoffset() reservations with struct types must be aligned with the beginning of a register."); + } + else if (var_class == HLSL_CLASS_VECTOR) + { + unsigned int aligned_offset = hlsl_type_get_sm4_offset(var->data_type, var->reg_reservation.offset_index); + + if (var->reg_reservation.offset_index != aligned_offset) + hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_INVALID_RESERVATION, + "packoffset() reservations with vector types cannot span multiple registers."); + } } + var->buffer_offset = var->reg_reservation.offset_index; + } + else + { + var->buffer_offset = hlsl_type_get_sm4_offset(var->data_type, buffer->size); } - var->buffer_offset = var->reg_reservation.offset_index; - } - else - { - var->buffer_offset = hlsl_type_get_sm4_offset(var->data_type, buffer->size); } TRACE("Allocated buffer offset %u to %s.\n", var->buffer_offset, var->name); @@ -4231,6 +4238,11 @@ static void validate_buffer_offsets(struct hlsl_ctx *ctx) } } +static bool var_has_buffer_offset_register_reservation(struct hlsl_ctx *ctx, const struct hlsl_ir_var *var) +{ + return var->reg_reservation.reg_type == 'c' && var->buffer == ctx->globals_buffer; +} + static void allocate_buffers(struct hlsl_ctx *ctx) { struct hlsl_buffer *buffer; @@ -4239,13 +4251,29 @@ static void allocate_buffers(struct hlsl_ctx *ctx) LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) { - if (var->is_uniform && !hlsl_type_is_resource(var->data_type)) - { - if (var->is_param) - var->buffer = ctx->params_buffer; + if (!var->is_uniform || hlsl_type_is_resource(var->data_type)) + continue; - calculate_buffer_offset(ctx, var); - } + if (var->is_param) + var->buffer = ctx->params_buffer; + } + + LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) + { + if (!var->is_uniform || hlsl_type_is_resource(var->data_type)) + continue; + + if (var_has_buffer_offset_register_reservation(ctx, var)) + calculate_buffer_offset(ctx, var, true); + } + + LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) + { + if (!var->is_uniform || hlsl_type_is_resource(var->data_type)) + continue; + + if (!var_has_buffer_offset_register_reservation(ctx, var)) + calculate_buffer_offset(ctx, var, false); } validate_buffer_offsets(ctx); diff --git a/tests/hlsl/register-reservations-numeric.shader_test b/tests/hlsl/register-reservations-numeric.shader_test index ad11669a2..2f30bcd52 100644 --- a/tests/hlsl/register-reservations-numeric.shader_test +++ b/tests/hlsl/register-reservations-numeric.shader_test @@ -1,4 +1,4 @@ -[pixel shader fail(sm<6) todo] +[pixel shader fail(sm<6)] // Overlapping register(cX) reservations are not allowed except on SM6, where they are aliased. // On SM1 this gives hr 0x88760b59. float a : register(c0); @@ -24,7 +24,7 @@ float4 main() : sv_target uniform 0 float4 0.1 0.2 0.3 0.4 uniform 4 float4 1.1 1.2 1.3 1.4 draw quad -todo(sm<6) probe all rgba (1.1, 1.4, 0.2, 0.3) +probe all rgba (1.1, 1.4, 0.2, 0.3) [pixel shader] @@ -43,7 +43,7 @@ uniform 8 float4 2.1 2.2 2.3 2.4 uniform 12 float4 3.1 3.2 3.3 3.4 uniform 16 float4 4.1 4.2 4.3 4.4 draw quad -todo(sm<6) probe all rgba (4.1, 4.2, 1.3, 1.4) +probe all rgba (4.1, 4.2, 1.3, 1.4) [require] @@ -114,7 +114,7 @@ uniform 4 float4 1.1 1.2 1.3 1.4 uniform 8 float4 2.1 2.2 2.3 2.4 uniform 12 float4 3.1 3.2 3.3 3.4 draw quad -todo(sm<6) probe all rgba (2.1, 2.2, 2.3, 0.0) +probe all rgba (2.1, 2.2, 2.3, 0.0) [pixel shader] -- GitLab https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458
The pipeline test failed again because there were tests in `static-initializers.shader_test` and `arithmetic-int.shader_test` that were retrieving the unexpected hr 0x88760b59. I had previously added this mapping to E_FAIL in 7/10 because this also happened for the first test in `register-reservations-numeric.shader_test` but now I had to move that check to 6/10. By the way, this shows a little inconvenience with 6/10: For now, it is hard to catch the behavior of sm1 compilation on all the tests before the pipeline does, because that only happens on Windows in the shader_runner (and I don't want to overload this MR even more including the patches necessary to run SM1 compilation tests on Linux!). I would probably have to add some more `todo(sm<4)` qualifiers once I rebase this series. -- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458#note_52439
On Fri Nov 17 09:57:15 2023 +0000, Francisco Casas wrote:
Added 6/10 so that the aforementioned tests with the `fail(sm<4)` qualifier is actually compiled with ps_2_0 and fails. I'm not sure I understand this: I removed some of the `fail(sm<4)` markers and I saw no failure. Also, the only runner able to compile SM1-3 shaders is D3D9, which only runs on Windows, where `todo` is irrelevant. Am I missing anything?
-- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458#note_52669
On Fri Nov 17 09:57:15 2023 +0000, Giovanni Mascellani wrote:
I'm not sure I understand this: I removed some of the `fail(sm<4)` markers and I saw no failure. Also, the only runner able to compile SM1-3 shaders is D3D9, which only runs on Windows, where `todo` is irrelevant. Am I missing anything? Without 6/10 vkd3d_shader, on Windows is using the native compiler with SM4 profiles, so the tests don't fail as expected and the pipeline complains.
6/10 is required so that vkd3d_shader, on Windows, actually uses SM1 profiles and fails as expected (unlike todos, fails make sense in Windows). -- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458#note_52685
On Mon Nov 13 12:02:04 2023 +0000, Francisco Casas wrote:
I think I added those already on [this line](https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458/diffs?commit_id=48...). Oh I see, the test is copy-pasted. I didn't think we had to do that currently?
-- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458#note_52770
It may look ugly, but I could think of of a less ugly way :(.
Actually I think it looks pretty reasonable :-) I'm mildly surprised that the unallocated ones don't try to go back and fill in the gaps, but that does make our job easier. The one thing I don't love is the name of var_has_buffer_offset_register_reservation(). I'd honestly be tempted to inline the condition. Or maybe replace the word "buffer" with "globals". One other test that I'm tempted to add just to be safe: ``` float a : register(c2); float b; float4 main() : sv_target { return a + b; } ``` -- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458#note_52771
Actually I think it looks pretty reasonable :-) I'm mildly surprised that the unallocated ones don't try to go back and fill in the gaps, but that does make our job easier.
They go back and try to fill the gaps! I just failed to see that calculate_buffer_offset() doesn't fill the gaps like allocate_numeric_registers_for_type() does. In my test ``` float4 a[3]; // will get register(c3) float4 b[2] : register(c1); ``` `a` cannot fill the gap because it is too big. So it is good to add your test. And this patch requires more work. I will see if it makes sense to use `struct register_allocator` here. Thanks! -- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458#note_52901
Actually I think it looks pretty reasonable :-) I'm mildly surprised that the unallocated ones don't try to go back and fill in the gaps, but that does make our job easier.
They go back and try to fill the gaps! I just failed to see that calculate_buffer_offset() doesn't fill the gaps like allocate_numeric_registers_for_type() does.
Wait, what? When I tested they didn't. I.e. in that example I posted, "b" gets allocated to c2.y. -- https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/458#note_52918
participants (5)
-
Francisco Casas -
Francisco Casas (＠fcasas)
-
Giovanni Mascellani
-
Giovanni Mascellani (＠giomasce)
-
Zebediah Figura (＠zfigura)