Merge from 'sycl' to 'sycl-web'

Author: iclsrc

.github/workflows/sycl_linux_build_and_test.yml

@@ -36,26 +36,14 @@ on:
       build_artifact_suffix:
         type: string
         required: true
-      intel_drivers_image:
-        type: string
-        required: false
-        default: "ghcr.io/intel/llvm/ubuntu2004_intel_drivers:latest"
-      lts_config:
+      lts_matrix:
         type: string
         required: false
         default: ""
       lts_cmake_extra_args:
         type: string
         required: false
         default: ""
-      amdgpu_image:
-        type: string
-        required: false
-        default: "ghcr.io/intel/llvm/ubuntu2004_build:latest"
-      cuda_image:
-        type: string
-        required: false
-        default: "ghcr.io/intel/llvm/ubuntu2004_build:latest"
       lts_ref:
         type: string
         required: false
@@ -167,38 +155,14 @@ jobs:
         name: sycl_lit_${{ inputs.build_artifact_suffix }}
         path: lit.tar.xz
 
-  # This job generates matrix of tests for LLVM Test Suite
-  resolve_matrix:
-    name: Resolve Test Matrix
-    runs-on: ubuntu-latest
-    outputs:
-      lts: ${{ steps.work.outputs.lts }}
-    steps:
-    - name: Download scripts and configs
-      run: |
-        wget raw.githubusercontent.com/intel/llvm/${{ github.sha }}/devops/scripts/generate_test_matrix.js
-        wget raw.githubusercontent.com/intel/llvm/${{ github.sha }}/devops/test_configs.json
-        wget raw.githubusercontent.com/intel/llvm/sycl/devops/dependencies.json
-        mv dependencies.json dependencies.sycl.json
-        wget raw.githubusercontent.com/intel/llvm/${{ github.sha }}/devops/dependencies.json
-    - id: work
-      uses: actions/github-script@v6
-      name: Generate matrix
-      env:
-        GHA_INPUTS: ${{ toJSON(inputs) }}
-      with:
-        script: |
-          const script = require('./generate_test_matrix.js');
-          script({core, process});
-
   llvm_test_suite:
-    needs: [build, resolve_matrix]
-    if: ${{ inputs.lts_config != '' }}
+    needs: build
+    if: ${{ inputs.lts_matrix != '' }}
     strategy:
       fail-fast: false
       max-parallel: ${{ inputs.max_parallel }}
       matrix:
-        include: ${{ fromJSON(needs.resolve_matrix.outputs.lts) }}
+        include: ${{ fromJSON(inputs.lts_matrix) }}
     name: ${{ matrix.name }}
     runs-on: ${{ matrix.runs-on }}
     env: ${{ matrix.env }}
@@ -239,4 +203,3 @@ jobs:
         check_sycl_all: ${{ matrix.check_sycl_all }}
         results_name_suffix: ${{ matrix.config }}_${{ inputs.build_artifact_suffix }}
         cmake_args: '${{ matrix.cmake_args }} ${{ inputs.lts_cmake_extra_args }}'
-

.github/workflows/sycl_post_commit.yml

@@ -11,12 +11,20 @@ on:
     - .github/workflows/sycl_post_commit.yml
 
 jobs:
+  # This job generates matrix of tests for LLVM Test Suite
+  resolve_matrix:
+    name: Resolve Test Matrix
+    uses: ./.github/workflows/sycl_resolve_test_matrix.yml
+    with:
+      lts_config: "l0_gen9"
   linux_default:
     name: Linux Default
+    needs: resolve_matrix
     uses: ./.github/workflows/sycl_linux_build_and_test.yml
     with:
       build_cache_root: "/__w/llvm"
-      build_artifact_suffix: default
+      build_artifact_suffix: "post_commit"
+      lts_matrix: ${{ needs.resolve_matrix.outputs.lts_matrix }}
   linux_no_assert:
     name: Linux (no assert)
     uses: ./.github/workflows/sycl_linux_build_and_test.yml

.github/workflows/sycl_precommit.yml

@@ -29,16 +29,23 @@ jobs:
     - name: Run clang-format
       uses: ./devops/actions/clang-format
 
+  # This job generates matrix of tests for LLVM Test Suite
+  resolve_matrix:
+    name: Resolve Test Matrix
+    uses: ./.github/workflows/sycl_resolve_test_matrix.yml
+    with:
+      lts_config: "hip_amdgpu;ocl_x64;ocl_gen9;l0_gen9;esimd_emu;cuda"
+
   linux_default:
     name: Linux
     # Only build and test patches, that have passed all linter checks, because
     # the next commit is likely to be a follow-up on that job.
-    needs: lint
+    needs: [lint, resolve_matrix]
     if: always() && (success() || contains(github.event.pull_request.labels.*.name, 'ignore-lint'))
     uses: ./.github/workflows/sycl_linux_build_and_test.yml
     with:
       build_cache_root: "/__w/"
       build_cache_size: "8G"
       build_artifact_suffix: "default"
       build_cache_suffix: "default"
-      lts_config: "hip_amdgpu;ocl_x64;ocl_gen9;l0_gen9;esimd_emu;cuda"
+      lts_matrix: ${{ needs.resolve_matrix.outputs.lts_matrix }}

.github/workflows/sycl_resolve_test_matrix.yml

@@ -0,0 +1,49 @@
+name: Reusable test matrix generation
+
+on:
+  workflow_call:
+    inputs:
+      intel_drivers_image:
+        type: string
+        required: false
+        default: "ghcr.io/intel/llvm/ubuntu2004_intel_drivers:latest"
+      amdgpu_image:
+        type: string
+        required: false
+        default: "ghcr.io/intel/llvm/ubuntu2004_build:latest"
+      cuda_image:
+        type: string
+        required: false
+        default: "ghcr.io/intel/llvm/ubuntu2004_build:latest"
+      lts_config:
+        type: string
+        required: true
+        default: ""
+    outputs:
+      lts_matrix:
+        description: "Generated Matrix"
+        value: ${{ jobs.resolve_matrix.outputs.lts_matrix }}
+jobs:
+  resolve_matrix:
+    name: Resolve Test Matrix
+    runs-on: ubuntu-latest
+    outputs:
+      lts_matrix: ${{ steps.work.outputs.lts_matrix }}
+    steps:
+    - name: Download scripts and configs
+      shell: bash
+      run: |
+        wget raw.githubusercontent.com/intel/llvm/${{ github.sha }}/devops/scripts/generate_test_matrix.js
+        wget raw.githubusercontent.com/intel/llvm/${{ github.sha }}/devops/test_configs.json
+        wget raw.githubusercontent.com/intel/llvm/sycl/devops/dependencies.json
+        mv dependencies.json dependencies.sycl.json
+        wget raw.githubusercontent.com/intel/llvm/${{ github.sha }}/devops/dependencies.json
+    - id: work
+      uses: actions/github-script@v6
+      name: Generate matrix
+      env:
+        GHA_INPUTS: ${{ toJSON(inputs) }}
+      with:
+        script: |
+          const script = require('./generate_test_matrix.js');
+          script({core, process});

.github/workflows/sycl_windows_build_and_test.yml

@@ -10,7 +10,7 @@ on:
 
 jobs:
   build:
-    name: Build
+    name: Build + LIT
     runs-on: [Windows, build]
     # TODO use cached checkout
     steps:

devops/scripts/generate_test_matrix.js

@@ -59,7 +59,7 @@ module.exports = ({core, process}) => {
             "ghcr.io/intel/llvm/ubuntu2004_base:latest");
       }
 
-      core.setOutput('lts', ltsString);
+      core.setOutput('lts_matrix', ltsString);
     }
   });
 }

libclc/amdgcn-amdhsa/libspirv/SOURCES

@@ -65,3 +65,4 @@ workitem/get_sub_group_local_id.cl
 workitem/get_sub_group_size.cl
 misc/sub_group_shuffle.cl
 async/wait_group_events.cl
+assert/__assert_fail.ll

libclc/amdgcn-amdhsa/libspirv/assert/__assert_fail.ll

@@ -0,0 +1,70 @@
+;;===----------------------------------------------------------------------===//
+;;
+;; Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+;; See https://llvm.org/LICENSE.txt for license information.
+;; SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+;;
+;;===----------------------------------------------------------------------===//
+
+#if __clang_major__ >= 7
+target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5-G1-ni:7"
+#else
+target datalayout = "e-p:64:64-p1:64:64-p2:32:32-p3:32:32-p4:64:64-p5:32:32-p6:32:32-i64:64-v16:16-v24:32-v32:32-v48:64-v96:128-v192:256-v256:256-v512:512-v1024:1024-v2048:2048-n32:64-S32-A5-G1-ni:7"
+#endif
+
+@.assert_fmt = private unnamed_addr constant [79 x i8] c"%s:%u: %s: global id: [%u,%u,%u], local id: [%u,%u,%u] Assertion `%s` failed.\0A\00", align 1
+
+declare void @llvm.trap() cold noreturn nounwind
+
+declare i64 @__ockl_fprintf_stderr_begin() local_unnamed_addr
+declare i64 @__ockl_fprintf_append_string_n(i64, i8* readonly, i64, i32) local_unnamed_addr
+declare i64 @__ockl_fprintf_append_args(i64, i32, i64, i64, i64, i64, i64, i64, i64, i32) local_unnamed_addr
+
+declare dso_local i64 @_Z28__spirv_GlobalInvocationId_xv() local_unnamed_addr
+declare dso_local i64 @_Z28__spirv_GlobalInvocationId_yv() local_unnamed_addr
+declare dso_local i64 @_Z28__spirv_GlobalInvocationId_zv() local_unnamed_addr
+
+declare dso_local i64 @_Z27__spirv_LocalInvocationId_xv() local_unnamed_addr
+declare dso_local i64 @_Z27__spirv_LocalInvocationId_yv() local_unnamed_addr
+declare dso_local i64 @_Z27__spirv_LocalInvocationId_zv() local_unnamed_addr
+
+define dso_local hidden noundef i64 @__strlen_assert(i8* noundef %str) local_unnamed_addr {
+entry:
+  br label %while.cond
+
+while.cond:
+  %tmp.0 = phi i8* [ %str, %entry ], [ %incdec.ptr, %while.cond ]
+  %incdec.ptr = getelementptr inbounds i8, i8* %tmp.0, i64 1
+  %0 = load i8, i8* %tmp.0, align 1
+  %tobool.not = icmp eq i8 %0, 0
+  br i1 %tobool.not, label %while.end, label %while.cond
+
+while.end:
+  %sub.ptr.lhs.cast = ptrtoint i8* %incdec.ptr to i64
+  %sub.ptr.rhs.cast = ptrtoint i8* %str to i64
+  %sub.ptr.sub = sub i64 %sub.ptr.lhs.cast, %sub.ptr.rhs.cast
+  ret i64 %sub.ptr.sub
+}
+
+define hidden void @__assert_fail(i8* %assertion, i8* %file, i32 %line, i8* %function) nounwind alwaysinline {
+entry:
+  %msg = call i64 @__ockl_fprintf_stderr_begin()
+  %msg.1 = call i64 @__ockl_fprintf_append_string_n(i64 %msg, i8* readonly getelementptr inbounds ([79 x i8], [79 x i8]* @.assert_fmt, i64 0, i64 0), i64 79, i32 0)
+  %len.file = call i64 @__strlen_assert(i8* %file)
+  %msg.2 = call i64 @__ockl_fprintf_append_string_n(i64 %msg.1, i8* readonly %file, i64 %len.file, i32 0)
+  %line.i64 = sext i32 %line to i64
+  %msg.3 = call i64 @__ockl_fprintf_append_args(i64 %msg.2, i32 1, i64 %line.i64, i64 0, i64 0, i64 0, i64 0, i64 0, i64 0, i32 0)
+  %len.func = call i64 @__strlen_assert(i8* %function)
+  %msg.4 = call i64 @__ockl_fprintf_append_string_n(i64 %msg.3, i8* readonly %function, i64 %len.func, i32 0)
+  %gidx = tail call i64 @_Z28__spirv_GlobalInvocationId_xv()
+  %gidy = tail call i64 @_Z28__spirv_GlobalInvocationId_yv()
+  %gidz = tail call i64 @_Z28__spirv_GlobalInvocationId_zv()
+  %lidx = tail call i64 @_Z27__spirv_LocalInvocationId_xv()
+  %lidy = tail call i64 @_Z27__spirv_LocalInvocationId_yv()
+  %lidz = tail call i64 @_Z27__spirv_LocalInvocationId_zv()
+  %msg.5 = call i64 @__ockl_fprintf_append_args(i64 %msg.4, i32 6, i64 %gidx, i64 %gidy, i64 %gidz, i64 %lidx, i64 %lidy, i64 %lidz, i64 0, i32 0)
+  %len.assertion = call i64 @__strlen_assert(i8* %assertion)
+  %msg.6 = call i64 @__ockl_fprintf_append_string_n(i64 %msg.4, i8* readonly %assertion, i64 %len.assertion, i32 1)
+  tail call void @llvm.trap()
+  unreachable
+}

llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp

@@ -555,11 +555,14 @@ StringRef AMDGPUTargetMachine::getFeatureString(const Function &F) const {
 }
 
 /// Predicate for Internalize pass.
+/// Functions with the 'sycl-module-id' attribute are SYCL_EXTERNAL functions
+/// and must be preserved.
 static bool mustPreserveGV(const GlobalValue &GV) {
   if (const Function *F = dyn_cast<Function>(&GV))
     return F->isDeclaration() || F->getName().startswith("__asan_") ||
            F->getName().startswith("__sanitizer_") ||
-           AMDGPU::isEntryFunctionCC(F->getCallingConv());
+           AMDGPU::isEntryFunctionCC(F->getCallingConv()) ||
+           F->hasFnAttribute("sycl-module-id");
 
   GV.removeDeadConstantUsers();
   return !GV.use_empty();

sycl/doc/EnvironmentVariables.md

@@ -155,7 +155,7 @@ variables in production code.</span>
 | `SYCL_PI_LEVEL_ZERO_COPY_BATCH_SIZE` | Integer | Sets a preferred number of copy commands to batch into a command list before executing the command list. A value of 0 causes the batch size to be adjusted dynamically. A value greater than 0 specifies fixed size batching, with the batch size set to the specified value. The default is 0. |
 | `SYCL_PI_LEVEL_ZERO_FILTER_EVENT_WAIT_LIST` | Integer | When set to 0, disables filtering of signaled events from wait lists when using the Level Zero backend. The default is 1. |
 | `SYCL_PI_LEVEL_ZERO_USE_COPY_ENGINE` | Any(\*) | This environment variable enables users to control use of copy engines for copy operations. If the value is an integer, it will allow the use of copy engines, if available in the device, in Level Zero plugin to transfer SYCL buffer or image data between the host and/or device(s) and to fill SYCL buffer or image data in device or shared memory. The value of this environment variable can also be a pair of the form "lower_index:upper_index" where the indices point to copy engines in a list of all available copy engines. The default is 1. |
-| `SYCL_PI_LEVEL_ZERO_USE_COMPUTE_ENGINE` | Integer | It can be set to an integer (>=0) in which case all compute commands will be submitted to the command-queue with the given index in the compute command group. If it is instead set to a negative value then all available compute engines may be used. The default value is "-1" |
+| `SYCL_PI_LEVEL_ZERO_USE_COMPUTE_ENGINE` | Integer | It can be set to an integer (>=0) in which case all compute commands will be submitted to the command-queue with the given index in the compute command group. If it is instead set to a negative value then all available compute engines may be used. The default value is "0" |
 | `SYCL_PI_LEVEL_ZERO_USE_COPY_ENGINE_FOR_D2D_COPY` (experimental) | Integer | Allows the use of copy engine, if available in the device, in Level Zero plugin for device to device copy operations. The default is 0. This option is experimental and will be removed once heuristics are added to make a decision about use of copy engine for device to device copy operations. |
 | `SYCL_PI_LEVEL_ZERO_DEVICE_SCOPE_EVENTS` | Any(\*) | Enable support of device-scope events whose state is not visible to the host. If enabled mode is SYCL_PI_LEVEL_ZERO_DEVICE_SCOPE_EVENTS=1 the Level Zero plugin would create all events having device-scope only and create proxy host-visible events for them when their status is needed (wait/query) on the host. If enabled mode is SYCL_PI_LEVEL_ZERO_DEVICE_SCOPE_EVENTS=2 the Level Zero plugin would create all events having device-scope and add proxy host-visible event at the end of each command-list submission. The default is 2, meaning only the last event in a batch is host-visible. |
 | `SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS` | Integer | When set to a positive value enables use of Level Zero immediate commandlists, which means there is no batching and all commands are immediately submitted for execution. Default is 0. Note: When immediate commandlist usage is enabled it is necessary to also set SYCL_PI_LEVEL_ZERO_DEVICE_SCOPE_EVENTS to either 0 or 1.  |