Merged main:190b9179a5de into amd-gfx:c84b9d216f14

Local branch amd-gfx c84b9d2 Merged main:1eaf926f2c61 into amd-gfx:c3eb9e85dcf1
Remote branch main 190b917 [clang][Interp] Handle SizeOfPackExprs (llvm#71929)

Author: SC llvm team

.github/workflows/pr-code-format.yml

@@ -11,14 +11,33 @@ jobs:
       - name: Fetch LLVM sources
         uses: actions/checkout@v4
         with:
-          fetch-depth: 2 # Fetches only the last 2 commits
+          ref: ${{ github.event.pull_request.head.sha }}
+
+      - name: Checkout through merge base
+        uses: rmacklin/fetch-through-merge-base@v0
+        with:
+          base_ref: ${{ github.event.pull_request.base.ref }}
+          head_ref: ${{ github.event.pull_request.head.sha }}
+          deepen_length: 500
 
       - name: Get changed files
         id: changed-files
         uses: tj-actions/changed-files@v39
         with:
           separator: ","
-          fetch_depth: 2000 # Fetches only the last 2000 commits
+          skip_initial_fetch: true
+
+      # We need to make sure that we aren't executing/using any code from the
+      # PR for security reasons as we're using pull_request_target. Checkout
+      # the target branch with the necessary files.
+      - name: Fetch code formatting utils
+        uses: actions/checkout@v4
+        with:
+          sparse-checkout: |
+            llvm/utils/git/requirements_formatting.txt
+            llvm/utils/git/code-format-helper.py
+          sparse-checkout-cone-mode: false
+          path: code-format-tools
 
       - name: "Listed files"
         run: |
@@ -35,10 +54,10 @@ jobs:
         with:
           python-version: '3.11'
           cache: 'pip'
-          cache-dependency-path: 'llvm/utils/git/requirements_formatting.txt'
+          cache-dependency-path: 'code-format-tools/llvm/utils/git/requirements_formatting.txt'
 
       - name: Install python dependencies
-        run: pip install -r llvm/utils/git/requirements_formatting.txt
+        run: pip install -r code-format-tools/llvm/utils/git/requirements_formatting.txt
 
       - name: Run code formatter
         env:
@@ -47,7 +66,7 @@ jobs:
           END_REV: ${{ github.event.pull_request.head.sha }}
           CHANGED_FILES: ${{ steps.changed-files.outputs.all_changed_files }}
         run: |
-          python llvm/utils/git/code-format-helper.py \
+          python ./code-format-tools/llvm/utils/git/code-format-helper.py \
             --token ${{ secrets.GITHUB_TOKEN }} \
             --issue-number $GITHUB_PR_NUMBER \
             --start-rev $START_REV \

clang/lib/AST/Interp/ByteCodeExprGen.cpp

@@ -1624,6 +1624,11 @@ bool ByteCodeExprGen<Emitter>::VisitCXXScalarValueInitExpr(
   return this->visitZeroInitializer(classifyPrim(Ty), Ty, E);
 }
 
+template <class Emitter>
+bool ByteCodeExprGen<Emitter>::VisitSizeOfPackExpr(const SizeOfPackExpr *E) {
+  return this->emitConst(E->getPackLength(), E);
+}
+
 template <class Emitter> bool ByteCodeExprGen<Emitter>::discard(const Expr *E) {
   if (E->containsErrors())
     return false;

clang/lib/AST/Interp/ByteCodeExprGen.h

@@ -107,6 +107,7 @@ class ByteCodeExprGen : public ConstStmtVisitor<ByteCodeExprGen<Emitter>, bool>,
   bool VisitSourceLocExpr(const SourceLocExpr *E);
   bool VisitOffsetOfExpr(const OffsetOfExpr *E);
   bool VisitCXXScalarValueInitExpr(const CXXScalarValueInitExpr *E);
+  bool VisitSizeOfPackExpr(const SizeOfPackExpr *E);
 
 protected:
   bool visitExpr(const Expr *E) override;

clang/test/AST/Interp/functions.cpp

@@ -371,3 +371,10 @@ namespace Variadic {
   constexpr int (*VFP)(...) = variadic_function2;
   static_assert(VFP() == 12, "");
 }
+
+namespace Packs {
+  template<typename...T>
+  constexpr int foo() { return sizeof...(T); }
+  static_assert(foo<int, char>() == 2, "");
+  static_assert(foo<>() == 0, "");
+}

llvm/include/llvm/Config/llvm-config.h.cmake

@@ -16,7 +16,7 @@
 
 /* Indicate that this is LLVM compiled from the amd-gfx branch. */
 #define LLVM_HAVE_BRANCH_AMD_GFX
-#define LLVM_MAIN_REVISION 480508
+#define LLVM_MAIN_REVISION 480514
 
 /* Define if LLVM_ENABLE_DUMP is enabled */
 #cmakedefine LLVM_ENABLE_DUMP

llvm/lib/Target/RISCV/GISel/RISCVLegalizerInfo.cpp

@@ -81,6 +81,10 @@ RISCVLegalizerInfo::RISCVLegalizerInfo(const RISCVSubtarget &ST) {
         .clampScalar(BigTyIdx, sXLen, sXLen);
   }
 
+  getActionDefinitionsBuilder({G_FSHL, G_FSHR}).lower();
+
+  getActionDefinitionsBuilder({G_ROTL, G_ROTR}).lower();
+
   getActionDefinitionsBuilder({G_BSWAP, G_BITREVERSE})
       .maxScalar(0, sXLen)
       .lower();

llvm/lib/Target/RISCV/RISCVInstrInfo.td

@@ -1864,6 +1864,12 @@ def : Pat<(i64 (shl (and GPR:$rs1, 0xffffffff), uimm5:$shamt)),
           (SRLI (SLLI GPR:$rs1, 32), (ImmSubFrom32 uimm5:$shamt))>;
 }
 
+class binop_allhusers<SDPatternOperator operator>
+    : PatFrag<(ops node:$lhs, node:$rhs),
+              (XLenVT (operator node:$lhs, node:$rhs)), [{
+  return hasAllHUsers(Node);
+}]>;
+
 // PatFrag to allow ADDW/SUBW/MULW/SLLW to be selected from i64 add/sub/mul/shl
 // if only the lower 32 bits of their result is used.
 class binop_allwusers<SDPatternOperator operator>

llvm/lib/Target/RISCV/RISCVInstrInfoZb.td

@@ -643,6 +643,10 @@ def : Pat<(or (shl (zexti8 (XLenVT GPR:$rs2)), (XLenVT 8)),
 def : Pat<(and (or (shl GPR:$rs2, (XLenVT 8)),
                    (zexti8 (XLenVT GPR:$rs1))), 0xFFFF),
           (PACKH GPR:$rs1, GPR:$rs2)>;
+
+def : Pat<(binop_allhusers<or> (shl GPR:$rs2, (XLenVT 8)),
+                               (zexti8 (XLenVT GPR:$rs1))),
+          (PACKH GPR:$rs1, GPR:$rs2)>;
 } // Predicates = [HasStdExtZbkb]
 
 let Predicates = [HasStdExtZbkb, IsRV32] in
@@ -853,6 +857,9 @@ def : Pat<(or (and (shl GPR:$rs2, (i64 8)), 0xFFFF),
 def : Pat<(or (shl (zexti8i32 (i32 GPR:$rs2)), (i64 8)),
               (zexti8i32 (i32 GPR:$rs1))),
           (PACKH GPR:$rs1, GPR:$rs2)>;
+def : Pat<(and (anyext (or (shl GPR:$rs2, (XLenVT 8)),
+                           (zexti8i32 (i32 GPR:$rs1)))), 0xFFFF),
+          (PACKH GPR:$rs1, GPR:$rs2)>;
 
 def : Pat<(i32 (or (shl GPR:$rs2, (i64 16)), (zexti16i32 (i32 GPR:$rs1)))),
           (PACKW GPR:$rs1, GPR:$rs2)>;