Merge remote-tracking branch 'origin/main' into 2024q4-preiselintrinsiclowering-learn-to-emit-memsetpattern16-libcall

Author: asb

.github/workflows/libcxx-build-and-test.yaml

@@ -48,8 +48,8 @@ jobs:
           'generic-cxx26',
           'generic-modules'
         ]
-        cc: [  'clang-19' ]
-        cxx: [ 'clang++-19' ]
+        cc: [  'clang-20' ]
+        cxx: [ 'clang++-20' ]
         include:
           - config: 'generic-gcc'
             cc: 'gcc-14'
@@ -88,18 +88,18 @@ jobs:
           'generic-cxx20',
           'generic-cxx23'
         ]
-        cc: [ 'clang-19' ]
-        cxx: [ 'clang++-19' ]
+        cc: [ 'clang-20' ]
+        cxx: [ 'clang++-20' ]
         include:
           - config: 'generic-gcc-cxx11'
             cc: 'gcc-14'
             cxx: 'g++-14'
           - config: 'generic-cxx23'
-            cc: 'clang-17'
-            cxx: 'clang++-17'
-          - config: 'generic-cxx26'
             cc: 'clang-18'
             cxx: 'clang++-18'
+          - config: 'generic-cxx26'
+            cc: 'clang-19'
+            cxx: 'clang++-19'
     steps:
       - uses: actions/checkout@v4
       - name: ${{ matrix.config }}
@@ -169,8 +169,8 @@ jobs:
       - name: ${{ matrix.config }}
         run: libcxx/utils/ci/run-buildbot ${{ matrix.config }}
         env:
-          CC: clang-19
-          CXX: clang++-19
+          CC: clang-20
+          CXX: clang++-20
       - uses: actions/upload-artifact@26f96dfa697d77e81fd5907df203aa23a56210a8 # v4.3.0
         if: always()
         with:

.github/workflows/release-binaries.yml

@@ -226,14 +226,6 @@ jobs:
       id: setup-stage
       uses: ./workflows-main/.github/workflows/release-binaries-setup-stage
 
-    - name: Setup sccache
-      uses: hendrikmuhs/ccache-action@ca3acd2731eef11f1572ccb126356c2f9298d35e # v1.2.9
-      with:
-        # Default to 2G to workaround: https://github.com/hendrikmuhs/ccache-action/issues/174
-        max-size: 2G
-        key: ${{ needs.prepare.outputs.ccache }}-${{ runner.os }}-${{ runner.arch }}-release
-        variant: ${{ needs.prepare.outputs.ccache }}
-
     - name: Configure
       id: build
       shell: bash
@@ -246,9 +238,8 @@ jobs:
             ${{ needs.prepare.outputs.target-cmake-flags }} \
             -C clang/cmake/caches/Release.cmake \
             -DBOOTSTRAP_LLVM_PARALLEL_LINK_JOBS=1 \
-            -DBOOTSTRAP_CPACK_PACKAGE_FILE_NAME="${{ needs.prepare.outputs.release-binary-basename }}" \
-            -DCMAKE_C_COMPILER_LAUNCHER=$CCACHE_BIN \
-            -DCMAKE_CXX_COMPILER_LAUNCHER=$CCACHE_BIN
+            -DBOOTSTRAP_CPACK_PACKAGE_FILE_NAME="${{ needs.prepare.outputs.release-binary-basename }}"
+
     - name: Build
       shell: bash
       run: |

bolt/lib/Core/BinaryFunction.cpp

@@ -795,7 +795,6 @@ BinaryFunction::processIndirectBranch(MCInst &Instruction, unsigned Size,
 
   auto Begin = Instructions.begin();
   if (BC.isAArch64()) {
-    PreserveNops = BC.HasRelocations;
     // Start at the last label as an approximation of the current basic block.
     // This is a heuristic, since the full set of labels have yet to be
     // determined
@@ -2300,6 +2299,10 @@ Error BinaryFunction::buildCFG(MCPlusBuilder::AllocatorIdTy AllocatorId) {
       BC.errs() << "BOLT-WARNING: failed to post-process indirect branches for "
                 << *this << '\n';
     }
+
+    if (BC.isAArch64())
+      PreserveNops = BC.HasRelocations;
+
     // In relocation mode we want to keep processing the function but avoid
     // optimizing it.
     setSimple(false);

bolt/test/AArch64/pad-before-funcs.s

@@ -15,7 +15,7 @@
 # RUN: llvm-bolt %t.exe -o %t.bolt.4.4 --pad-funcs-before=_start:4 --pad-funcs=_start:4
 # RUN: llvm-bolt %t.exe -o %t.bolt.4.8 --pad-funcs-before=_start:4 --pad-funcs=_start:8
 
-# RUN: not llvm-bolt %t.exe -o %t.bolt.8 --pad-funcs-before=_start:1 2>&1 | FileCheck --check-prefix=CHECK-BAD-ALIGN %s
+# RUN: not llvm-bolt %t.exe -o %t.bolt.1 --pad-funcs-before=_start:1 2>&1 | FileCheck --check-prefix=CHECK-BAD-ALIGN %s
 
 # CHECK-BAD-ALIGN: user-requested 1 padding bytes before function _start(*2) is not a multiple of the minimum function alignment (4).
 

bolt/test/AArch64/remove-nops.s

@@ -0,0 +1,28 @@
+## Verify that llvm-bolt removes nop instructions from functions with indirect
+## branches that have defined control flow.
+
+# REQUIRES: system-linux
+
+# RUN: llvm-mc -filetype=obj -triple aarch64-unknown-unknown %s -o %t.o
+# RUN: %clang %cflags --target=aarch64-unknown-linux %t.o -o %t.exe -Wl,-q
+# RUN: llvm-bolt %t.exe -o %t.bolt --print-normalized 2>&1 | FileCheck %s
+# RUN: llvm-objdump -d --disassemble-symbols=_start %t.bolt \
+# RUN:   | FileCheck %s --check-prefix=CHECK-OBJDUMP
+
+# CHECK-OBJDUMP-LABEL: _start
+# CHECK-OBJDUMP-NOT: nop
+
+  .section .text
+  .align 4
+  .globl _start
+  .type  _start, %function
+_start:
+# CHECK-LABEL: Binary Function "_start"
+  nop
+# CHECK-NOT: nop
+  br      x0
+# CHECK: br x0 # TAILCALL
+.size _start, .-_start
+
+## Force relocation mode.
+  .reloc 0, R_AARCH64_NONE