Merge branch 'sycl' into simplify-device-global

Author: zhaomaosu

.ci/generate-buildkite-pipeline-premerge

@@ -53,6 +53,8 @@ echo "Directories modified:" >&2
 echo "$modified_dirs" >&2
 
 function compute-projects-to-test() {
+  isForWindows=$1
+  shift
   projects=${@}
   for project in ${projects}; do
     echo "${project}"
@@ -63,20 +65,28 @@ function compute-projects-to-test() {
       done
     ;;
     llvm)
-      for p in bolt clang clang-tools-extra flang lld lldb mlir polly; do
+      for p in bolt clang clang-tools-extra lld lldb mlir polly; do
         echo $p
       done
+      # Flang is not stable in Windows CI at the moment
+      if [[ $isForWindows == 0 ]]; then
+        echo flang
+      fi
     ;;
     clang)
-      for p in clang-tools-extra compiler-rt lldb cross-project-tests; do
+      # lldb is temporarily removed to alleviate Linux pre-commit CI waiting times
+      for p in clang-tools-extra compiler-rt cross-project-tests; do
         echo $p
       done
     ;;
     clang-tools-extra)
       echo libc
     ;;
     mlir)
-      echo flang
+      # Flang is not stable in Windows CI at the moment
+      if [[ $isForWindows == 0 ]]; then
+        echo flang
+      fi
     ;;
     *)
       # Nothing to do
@@ -144,7 +154,6 @@ function exclude-linux() {
   for project in ${projects}; do
     case ${project} in
     cross-project-tests) ;; # tests failing
-    lldb)                ;; # tests failing
     openmp)              ;; # https://github.com/google/llvm-premerge-checks/issues/410
     *)
       echo "${project}"
@@ -161,7 +170,7 @@ function exclude-windows() {
     compiler-rt)         ;; # tests taking too long
     openmp)              ;; # TODO: having trouble with the Perl installation
     libc)                ;; # no Windows support
-    lldb)                ;; # tests failing
+    lldb)                ;; # custom environment requirements (https://github.com/llvm/llvm-project/pull/94208#issuecomment-2146256857)
     bolt)                ;; # tests are not supported yet
     *)
       echo "${project}"
@@ -204,7 +213,7 @@ function check-targets() {
       echo "check-unwind"
     ;;
     lldb)
-      echo "check-all" # TODO: check-lldb may not include all the LLDB tests?
+      echo "check-lldb"
     ;;
     pstl)
       echo "check-all"
@@ -241,15 +250,15 @@ fi
 all_projects="bolt clang clang-tools-extra compiler-rt cross-project-tests flang libc libclc lld lldb llvm mlir openmp polly pstl"
 modified_projects="$(keep-modified-projects ${all_projects})"
 
-linux_projects_to_test=$(exclude-linux $(compute-projects-to-test ${modified_projects}))
+linux_projects_to_test=$(exclude-linux $(compute-projects-to-test 0 ${modified_projects}))
 linux_check_targets=$(check-targets ${linux_projects_to_test} | sort | uniq)
 linux_projects=$(add-dependencies ${linux_projects_to_test} | sort | uniq)
 
 linux_runtimes_to_test=$(compute-runtimes-to-test ${linux_projects_to_test})
 linux_runtime_check_targets=$(check-targets ${linux_runtimes_to_test} | sort | uniq)
 linux_runtimes=$(echo ${linux_runtimes_to_test} | sort | uniq)
 
-windows_projects_to_test=$(exclude-windows $(compute-projects-to-test ${modified_projects}))
+windows_projects_to_test=$(exclude-windows $(compute-projects-to-test 1 ${modified_projects}))
 windows_check_targets=$(check-targets ${windows_projects_to_test} | sort | uniq)
 windows_projects=$(add-dependencies ${windows_projects_to_test} | sort | uniq)
 

.ci/monolithic-linux.sh

@@ -39,6 +39,7 @@ targets="${2}"
 
 echo "--- cmake"
 pip install -q -r "${MONOREPO_ROOT}"/mlir/python/requirements.txt
+pip install -q -r "${MONOREPO_ROOT}"/lldb/test/requirements.txt
 cmake -S "${MONOREPO_ROOT}"/llvm -B "${BUILD_DIR}" \
       -D LLVM_ENABLE_PROJECTS="${projects}" \
       -G Ninja \

.ci/monolithic-windows.sh

@@ -44,6 +44,8 @@ pip install -q -r "${MONOREPO_ROOT}"/mlir/python/requirements.txt
 # see https://github.com/llvm/llvm-project/pull/82393 and
 # https://discourse.llvm.org/t/rfc-future-of-windows-pre-commit-ci/76840/40
 # for further information.
+# We limit the number of parallel compile jobs to 24 control memory
+# consumption and improve build reliability.
 cmake -S "${MONOREPO_ROOT}"/llvm -B "${BUILD_DIR}" \
       -D LLVM_ENABLE_PROJECTS="${projects}" \
       -G Ninja \
@@ -58,7 +60,9 @@ cmake -S "${MONOREPO_ROOT}"/llvm -B "${BUILD_DIR}" \
       -D MLIR_ENABLE_BINDINGS_PYTHON=ON \
       -D CMAKE_EXE_LINKER_FLAGS="/MANIFEST:NO" \
       -D CMAKE_MODULE_LINKER_FLAGS="/MANIFEST:NO" \
-      -D CMAKE_SHARED_LINKER_FLAGS="/MANIFEST:NO"
+      -D CMAKE_SHARED_LINKER_FLAGS="/MANIFEST:NO" \
+      -D LLVM_PARALLEL_COMPILE_JOBS=16 \
+      -D LLVM_PARALLEL_LINK_JOBS=4
 
 echo "--- ninja"
 # Targets are not escaped as they are passed as separate arguments.

.git-blame-ignore-revs

@@ -81,3 +81,12 @@ f6d557ee34b6bbdb1dc32f29e34b4a4a8ad35e81
 
 # [NFC] clang-format utils/TableGen (#80973)
 b9079baaddfed5e604fbfaa1d81a7a1c38e78c26
+
+# [libc++][NFC] Run clang-format on libcxx/include again (#95874)
+e2c2ffbe7a1b5d9e32a2ce64279475b50c4cba5b
+
+# [lldb][nfc] Deindent ProcessGDBRemote::SetThreadStopInfo by two levels
+b32931c5b32eb0d2cf37d688b34f8548c9674c19
+
+# [libc++][NFC] Fix inconsistent quoting and spacing in our CSV files
+64946fdaf9864d8279da1c30e4d7214fe13d1427

.github/CODEOWNERS

@@ -34,25 +34,18 @@ sycl/doc/design/ @intel/llvm-reviewers-runtime
 sycl/doc/design/spirv-extensions/ @intel/dpcpp-spirv-doc-reviewers
 sycl/doc/extensions/ @intel/dpcpp-specification-reviewers
 
-# Unified Runtime 
-sycl/plugins @intel/unified-runtime-reviewers
+# Unified Runtime
+sycl/cmake/modules/FetchUnifiedRuntime.cmake @intel/unified-runtime-reviewers
+sycl/include/sycl/detail/ur.hpp @intel/unified-runtime-reviewers
+sycl/source/detail/posix_ur.cpp @intel/unified-runtime-reviewers
+sycl/source/detail/ur.cpp @intel/unified-runtime-reviewers
+sycl/source/detail/windows_ur.cpp @intel/unified-runtime-reviewers
 sycl/test-e2e/Plugin/ @intel/unified-runtime-reviewers
 
 # Win Proxy Loader
 sycl/pi_win_proxy_loader @intel/llvm-reviewers-runtime
-sycl/plugins/common_win_pi_trace @intel/llvm-reviewers-runtime
 sycl/test-e2e/Plugin/dll-detach-order.cpp @intel/llvm-reviewers-runtime
 
-# CUDA and HIP plugins
-sycl/plugins/**/cuda/ @intel/llvm-reviewers-cuda
-sycl/plugins/**/hip/ @intel/llvm-reviewers-cuda
-
-# Transform Dialect in MLIR.
-/mlir/include/mlir/Dialect/Transform/* @ftynse @nicolasvasilache
-/mlir/lib/Dialect/Transform/* @ftynse @nicolasvasilache
-/mlir/**/*TransformOps* @ftynse @nicolasvasilache
-
-
 # CUDA specific runtime implementations
 sycl/include/sycl/ext/oneapi/experimental/cuda/ @intel/llvm-reviewers-cuda
 
@@ -104,13 +97,16 @@ invoke_simd/ @intel/dpcpp-esimd-reviewers
 InvokeSimd/ @intel/dpcpp-esimd-reviewers
 sycl/test-e2e/InvokeSimd/ @intel/dpcpp-esimd-reviewers
 
+# dev-igc driver update
+devops/dependencies-igc-dev.json @intel/sycl-matrix-reviewers @intel/dpcpp-esimd-reviewers @intel/dpcpp-devops-reviewers
+
 # DevOps configs
 .github/ @intel/dpcpp-devops-reviewers
 buildbot/ @intel/dpcpp-devops-reviewers
 devops/ @intel/dpcpp-devops-reviewers
 
 # Kernel fusion JIT compiler
-sycl-fusion/ @intel/dpcpp-kernel-fusion-reviewers
+sycl-jit/ @intel/dpcpp-kernel-fusion-reviewers
 sycl/doc/design/KernelFusionJIT.md @intel/dpcpp-kernel-fusion-reviewers
 sycl/doc/extensions/experimental/sycl_ext_codeplay_kernel_fusion.asciidoc @intel/dpcpp-kernel-fusion-reviewers
 sycl/include/sycl/ext/codeplay/experimental/fusion_properties.hpp  @intel/dpcpp-kernel-fusion-reviewers
@@ -136,6 +132,7 @@ clang/test/Driver/sycl-native-cpu*.cpp @intel/dpcpp-nativecpu-pi-reviewers
 sycl/**/native_cpu/ @intel/dpcpp-nativecpu-pi-reviewers
 sycl/doc/design/SYCLNativeCPU.md @intel/dpcpp-nativecpu-pi-reviewers
 sycl/include/sycl/detail/native_cpu.hpp @intel/dpcpp-nativecpu-pi-reviewers
+libdevice/nativecpu* @intel/dpcpp-nativecpu-pi-reviewers
 
 # SYCL-Graphs extensions
 sycl/include/sycl/ext/oneapi/experimental/graph.hpp @intel/sycl-graphs-reviewers
@@ -155,7 +152,7 @@ sycl/include/syclcompat.hpp @intel/syclcompat-lib-reviewers
 sycl/doc/extensions/experimental/sycl_ext_oneapi_bindless_images.asciidoc @intel/bindless-images-reviewers
 sycl/include/sycl/ext/oneapi/bindless* @intel/bindless-images-reviewers
 sycl/source/detail/bindless* @intel/bindless-images-reviewers
-sycl/plugins/unified_runtime/ur/adapters/**/image.* @intel/bindless-images-reviewers
+sycl/test/check_device_code/extensions/bindless_images.cpp @intel/bindless-images-reviewers
 sycl/test-e2e/bindless_images/ @intel/bindless-images-reviewers
 
 # Miscellaneous sycl e2e tests
@@ -165,6 +162,7 @@ sycl/test-e2e/DeviceCodeSplit/ @intel/dpcpp-tools-reviewers
 sycl/test-e2e/SeparateCompile/ @intel/dpcpp-tools-reviewers
 sycl/test-e2e/Printf/ @intel/dpcpp-tools-reviewers @intel/llvm-reviewers-runtime
 sycl/test-e2e/SpecConstants/ @intel/dpcpp-tools-reviewers
+sycl/test-e2e/NewOffloadDriver/ @intel/dpcpp-tools-reviewers
 
 # Sanitizer
 clang/lib/Driver/SanitizerArgs.cpp @intel/dpcpp-sanitizers-review

.github/new-prs-labeler.yml

@@ -54,6 +54,9 @@ llvm-lit:
   - llvm/utils/lit/**/*
 
 PGO:
+  - llvm/**/ProfileData/**/*
+  - llvm/**/SampleProfile*
+  - llvm/**/CodeGen/MIRSampleProfile*
   - llvm/lib/Transforms/Instrumentation/CGProfile.cpp
   - llvm/lib/Transforms/Instrumentation/ControlHeightReduction.cpp
   - llvm/lib/Transforms/Instrumentation/IndirectCallPromotion.cpp
@@ -62,9 +65,9 @@ PGO:
   - llvm/lib/Transforms/Instrumentation/ValueProfile*
   - llvm/test/Instrumentation/InstrProfiling/**/*
   - llvm/test/Transforms/PGOProfile/**/*
+  - llvm/test/Transforms/SampleProfile/**/*
   - llvm/**/llvm-profdata/**/*
   - llvm/**/llvm-profgen/**/*
-  - llvm/unittests/ProfileData/**/*
 
 vectorization:
   - llvm/lib/Transforms/Vectorize/**/*
@@ -714,6 +717,16 @@ backend:AMDGPU:
   - '**/*amdgpu*/**'
   - '**/*AMDGPU*/**'
 
+backend:NVPTX:
+  - 'llvm/**/*nvvm*'
+  - 'llvm/**/*NVVM*'
+  - 'llvm/**/*nvptx*'
+  - 'llvm/**/*NVPTX*'
+  - 'llvm/**/*nvvm*/**'
+  - 'llvm/**/*NVVM*/**'
+  - 'llvm/**/*nvptx*/**'
+  - 'llvm/**/*NVPTX*/**'
+
 backend:RISC-V:
   - clang/**/*riscv*
   - clang/**/*RISCV*
@@ -746,6 +759,8 @@ backend:ARM:
   - clang/lib/CodeGen/Targets/ARM.cpp
   - clang/include/clang/Basic/BuiltinsARM*
   - llvm/test/MC/DisasemblerARM/**
+  - clang/include/clang/Sema/SemaARM.h
+  - clang/lib/Sema/SemaARM.cpp
 
 backend:AArch64:
   - llvm/include/llvm/IR/IntrinsicsAArch64.td
@@ -757,6 +772,31 @@ backend:AArch64:
   - clang/lib/CodeGen/Targets/AArch64.cpp
   - clang/include/clang/Basic/BuiltinsAArch64*
   - llvm/test/MC/Disassembler/AArch64/**
+  - clang/include/clang/Sema/SemaARM.h
+  - clang/lib/Sema/SemaARM.cpp
+
+backend:Hexagon:
+  - clang/include/clang/Basic/BuiltinsHexagon*.def
+  - clang/include/clang/Sema/SemaHexagon.h
+  - clang/lib/Basic/Targets/Hexagon.*
+  - clang/lib/CodeGen/Targets/Hexagon.cpp
+  - clang/lib/Driver/ToolChains/Hexagon.*
+  - clang/lib/Sema/SemaHexagon.cpp
+  - lld/ELF/Arch/Hexagon.cpp
+  - lldb/source/Plugins/ABI/Hexagon/**
+  - lldb/source/Plugins/DynamicLoader/Hexagon-DYLD/**
+  - llvm/include/llvm/BinaryFormat/ELFRelocs/Hexagon.def
+  - llvm/include/llvm/IR/IntrinsicsHexagon*
+  - llvm/include/llvm/Support/Hexagon*
+  - llvm/lib/Support/Hexagon*
+  - llvm/lib/Target/Hexagon/**
+  - llvm/test/CodeGen/Hexagon/**
+  - llvm/test/CodeGen/*/Hexagon/**
+  - llvm/test/DebugInfo/*/Hexagon/**
+  - llvm/test/Transforms/*/Hexagon
+  - llvm/test/MC/Disassembler/Hexagon/**
+  - llvm/test/MC/Hexagon/**
+  - llvm/test/tools/llvm-objdump/ELF/Hexagon/**
 
 backend:loongarch:
   - llvm/include/llvm/IR/IntrinsicsLoongArch.td
@@ -767,6 +807,8 @@ backend:loongarch:
   - clang/lib/Driver/ToolChains/Arch/LoongArch.*
   - clang/lib/CodeGen/Targets/LoongArch.cpp
   - clang/include/clang/Basic/BuiltinsLoongArch*
+  - clang/include/clang/Sema/SemaLoongArch.h
+  - clang/lib/Sema/SemaLoongArch.cpp
 
 backend:MSP430:
   - llvm/include/llvm/IR/IntrinsicsMSP430.td
@@ -814,6 +856,8 @@ backend:WebAssembly:
   - llvm/unittests/Target/WebAssembly/**
   - llvm/test/DebugInfo/WebAssembly/**
   - llvm/test/MC/WebAssembly/**
+  - clang/include/clang/Sema/SemaWasm.h
+  - clang/lib/Sema/SemaLoongWasm.cpp
 
 backend:X86:
   - llvm/include/llvm/IR/IntrinsicsX86.td
@@ -833,6 +877,8 @@ backend:X86:
   - llvm/include/llvm/TargetParser/X86*
   - llvm/lib/TargetParser/X86*
   - llvm/utils/TableGen/X86*
+  - clang/include/clang/Sema/SemaX86.h
+  - clang/lib/Sema/SemaX86.cpp
 
 backend:PowerPC:
   - llvm/include/llvm/BinaryFormat/ELFRelocs/PowerPC*
@@ -857,6 +903,8 @@ backend:PowerPC:
   - clang/lib/Driver/ToolChains/AIX*
   - clang/lib/Driver/ToolChains/Arch/PPC.*
   - clang/test/CodeGen/PowerPC/**
+  - clang/include/clang/Sema/SemaPPC.h
+  - clang/lib/Sema/SemaPPC.cpp
 
 backend:SystemZ:
   - llvm/include/llvm/BinaryFormat/ELFRelocs/SystemZ*
@@ -877,6 +925,8 @@ backend:SystemZ:
   - clang/lib/Driver/ToolChains/ZOS*
   - clang/lib/Driver/ToolChains/Arch/SystemZ.*
   - clang/test/CodeGen/SystemZ/**
+  - clang/include/clang/Sema/SemaSystemZ.h
+  - clang/lib/Sema/SemaSystemZ.cpp
 
 third-party:unittests:
   - third-party/unittests/**

.github/workflows/ci-post-commit-analyzer-run.py

@@ -0,0 +1,34 @@
+import json
+import multiprocessing
+import os
+import re
+import subprocess
+import sys
+
+
+def run_analyzer(data):
+    os.chdir(data["directory"])
+    command = (
+        data["command"]
+        + f" --analyze --analyzer-output html -o analyzer-results -Xclang -analyzer-config -Xclang max-nodes=75000"
+    )
+    print(command)
+    subprocess.run(command, shell=True, check=True)
+
+
+def pool_error(e):
+    print("Error analyzing file:", e)
+
+
+def main():
+    db_path = sys.argv[1]
+    database = json.load(open(db_path))
+
+    with multiprocessing.Pool() as pool:
+        pool.map_async(run_analyzer, [k for k in database], error_callback=pool_error)
+        pool.close()
+        pool.join()
+
+
+if __name__ == "__main__":
+    main()