llvmbot wrote:
<!--LLVM PR SUMMARY COMMENT--> @llvm/pr-subscribers-clangir Author: Sirui Mu (Lancern) <details> <summary>Changes</summary> This patch updates various out-of-sync OGCG checks in the test file `clang/test/CIR/CIRGen/builtin_bit.cpp`. These checks are all related to the original clang CodeGen for the bitwise rotate builtins. The OGCG patch #<!-- -->160259 inserts a new `urem` instruction that truncates the rotate amount against the input's bit width before calling the `llvm.fshr.*` intrinsic, which breaks these OGCG checks. I have not yet dug deep enough into the rationale behind the OGCG patch. The LLVM intrinsic `llvm.fshr.*` should already handle the truncation, and the new `urem` instruction seems redundant in terms of semantic correctness. Thus I choose not to hurry to also update relevant CIRGen code to match OGCG behavior in this patch. --- Full diff: https://github.com/llvm/llvm-project/pull/177189.diff 1 Files Affected: - (modified) clang/test/CIR/CodeGenBuiltins/builtin_bit.cpp (+16-8) ``````````diff diff --git a/clang/test/CIR/CodeGenBuiltins/builtin_bit.cpp b/clang/test/CIR/CodeGenBuiltins/builtin_bit.cpp index f252840c44900..fae19f5b1289e 100644 --- a/clang/test/CIR/CodeGenBuiltins/builtin_bit.cpp +++ b/clang/test/CIR/CodeGenBuiltins/builtin_bit.cpp @@ -504,7 +504,8 @@ unsigned char test_builtin_rotateleft8(unsigned char x, unsigned char y) { // OGCG-LABEL: @_Z24test_builtin_rotateleft8hh // OGCG: %[[INPUT:.+]] = load i8, ptr %{{.+}}, align 1 // OGCG-NEXT: %[[AMOUNT:.+]] = load i8, ptr %{{.+}}, align 1 -// OGCG-NEXT: %{{.+}} = call i8 @llvm.fshl.i8(i8 %[[INPUT]], i8 %[[INPUT]], i8 %[[AMOUNT]]) +// OGCG-NEXT: %[[CANON_AMOUNT:.+]] = urem i8 %[[AMOUNT]], 8 +// OGCG-NEXT: %{{.+}} = call i8 @llvm.fshl.i8(i8 %[[INPUT]], i8 %[[INPUT]], i8 %[[CANON_AMOUNT]]) unsigned short test_builtin_rotateleft16(unsigned short x, unsigned short y) { return __builtin_rotateleft16(x, y); @@ -521,7 +522,8 @@ unsigned short test_builtin_rotateleft16(unsigned short x, unsigned short y) { // OGCG-LABEL: @_Z25test_builtin_rotateleft16tt // OGCG: %[[INPUT:.+]] = load i16, ptr %{{.+}}, align 2 // OGCG-NEXT: %[[AMOUNT:.+]] = load i16, ptr %{{.+}}, align 2 -// OGCG-NEXT: %{{.+}} = call i16 @llvm.fshl.i16(i16 %[[INPUT]], i16 %[[INPUT]], i16 %[[AMOUNT]]) +// OGCG-NEXT: %[[CANON_AMOUNT:.+]] = urem i16 %[[AMOUNT]], 16 +// OGCG-NEXT: %{{.+}} = call i16 @llvm.fshl.i16(i16 %[[INPUT]], i16 %[[INPUT]], i16 %[[CANON_AMOUNT]]) unsigned test_builtin_rotateleft32(unsigned x, unsigned y) { return __builtin_rotateleft32(x, y); @@ -538,7 +540,8 @@ unsigned test_builtin_rotateleft32(unsigned x, unsigned y) { // OGCG-LABEL: @_Z25test_builtin_rotateleft32jj // OGCG: %[[INPUT:.+]] = load i32, ptr %{{.+}}, align 4 // OGCG-NEXT: %[[AMOUNT:.+]] = load i32, ptr %{{.+}}, align 4 -// OGCG-NEXT: %{{.+}} = call i32 @llvm.fshl.i32(i32 %[[INPUT]], i32 %[[INPUT]], i32 %[[AMOUNT]]) +// OGCG-NEXT: %[[CANON_AMOUNT:.+]] = urem i32 %[[AMOUNT]], 32 +// OGCG-NEXT: %{{.+}} = call i32 @llvm.fshl.i32(i32 %[[INPUT]], i32 %[[INPUT]], i32 %[[CANON_AMOUNT]]) unsigned long long test_builtin_rotateleft64(unsigned long long x, unsigned long long y) { @@ -556,7 +559,8 @@ unsigned long long test_builtin_rotateleft64(unsigned long long x, // OGCG-LABEL: @_Z25test_builtin_rotateleft64yy // OGCG: %[[INPUT:.+]] = load i64, ptr %{{.+}}, align 8 // OGCG-NEXT: %[[AMOUNT:.+]] = load i64, ptr %{{.+}}, align 8 -// OGCG-NEXT: %{{.+}} = call i64 @llvm.fshl.i64(i64 %[[INPUT]], i64 %[[INPUT]], i64 %[[AMOUNT]]) +// OGCG-NEXT: %[[CANON_AMOUNT:.+]] = urem i64 %[[AMOUNT]], 64 +// OGCG-NEXT: %{{.+}} = call i64 @llvm.fshl.i64(i64 %[[INPUT]], i64 %[[INPUT]], i64 %[[CANON_AMOUNT]]) unsigned char test_builtin_rotateright8(unsigned char x, unsigned char y) { return __builtin_rotateright8(x, y); @@ -573,7 +577,8 @@ unsigned char test_builtin_rotateright8(unsigned char x, unsigned char y) { // OGCG-LABEL: @_Z25test_builtin_rotateright8hh // OGCG: %[[INPUT:.+]] = load i8, ptr %{{.+}}, align 1 // OGCG-NEXT: %[[AMOUNT:.+]] = load i8, ptr %{{.+}}, align 1 -// OGCG-NEXT: %{{.+}} = call i8 @llvm.fshr.i8(i8 %[[INPUT]], i8 %[[INPUT]], i8 %[[AMOUNT]]) +// OGCG-NEXT: %[[CANON_AMOUNT:.+]] = urem i8 %[[AMOUNT]], 8 +// OGCG-NEXT: %{{.+}} = call i8 @llvm.fshr.i8(i8 %[[INPUT]], i8 %[[INPUT]], i8 %[[CANON_AMOUNT]]) unsigned short test_builtin_rotateright16(unsigned short x, unsigned short y) { return __builtin_rotateright16(x, y); @@ -590,7 +595,8 @@ unsigned short test_builtin_rotateright16(unsigned short x, unsigned short y) { // OGCG-LABEL: @_Z26test_builtin_rotateright16tt // OGCG: %[[INPUT:.+]] = load i16, ptr %{{.+}}, align 2 // OGCG-NEXT: %[[AMOUNT:.+]] = load i16, ptr %{{.+}}, align 2 -// OGCG-NEXT: %{{.+}} = call i16 @llvm.fshr.i16(i16 %[[INPUT]], i16 %[[INPUT]], i16 %[[AMOUNT]]) +// OGCG-NEXT: %[[CANON_AMOUNT:.+]] = urem i16 %[[AMOUNT]], 16 +// OGCG-NEXT: %{{.+}} = call i16 @llvm.fshr.i16(i16 %[[INPUT]], i16 %[[INPUT]], i16 %[[CANON_AMOUNT]]) unsigned test_builtin_rotateright32(unsigned x, unsigned y) { return __builtin_rotateright32(x, y); @@ -607,7 +613,8 @@ unsigned test_builtin_rotateright32(unsigned x, unsigned y) { // OGCG-LABEL: @_Z26test_builtin_rotateright32jj // OGCG: %[[INPUT:.+]] = load i32, ptr %{{.+}}, align 4 // OGCG-NEXT: %[[AMOUNT:.+]] = load i32, ptr %{{.+}}, align 4 -// OGCG-NEXT: %{{.+}} = call i32 @llvm.fshr.i32(i32 %[[INPUT]], i32 %[[INPUT]], i32 %[[AMOUNT]]) +// OGCG-NEXT: %[[CANON_AMOUNT:.+]] = urem i32 %[[AMOUNT]], 32 +// OGCG-NEXT: %{{.+}} = call i32 @llvm.fshr.i32(i32 %[[INPUT]], i32 %[[INPUT]], i32 %[[CANON_AMOUNT]]) unsigned long long test_builtin_rotateright64(unsigned long long x, unsigned long long y) { @@ -625,4 +632,5 @@ unsigned long long test_builtin_rotateright64(unsigned long long x, // OGCG-LABEL: @_Z26test_builtin_rotateright64yy // OGCG: %[[INPUT:.+]] = load i64, ptr %{{.+}}, align 8 // OGCG-NEXT: %[[AMOUNT:.+]] = load i64, ptr %{{.+}}, align 8 -// OGCG-NEXT: %{{.+}} = call i64 @llvm.fshr.i64(i64 %[[INPUT]], i64 %[[INPUT]], i64 %[[AMOUNT]]) +// OGCG-NEXT: %[[CANON_AMOUNT:.+]] = urem i64 %[[AMOUNT]], 64 +// OGCG-NEXT: %{{.+}} = call i64 @llvm.fshr.i64(i64 %[[INPUT]], i64 %[[INPUT]], i64 %[[CANON_AMOUNT]]) `````````` </details> https://github.com/llvm/llvm-project/pull/177189 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
