564d85e090
In LLVM IR, `AlignmentBitfieldElementT` is 5-bit wide But that means that the maximal alignment exponent is `(1<<5)-2`, which is `30`, not `29`. And indeed, alignment of `1073741824` roundtrips IR serialization-deserialization. While this doesn't seem all that important, this doubles the maximal supported alignment from 512MiB to 1GiB, and there's actually one noticeable use-case for that; On X86, the huge pages can have sizes of 2MiB and 1GiB (!). So while this doesn't add support for truly huge alignments, which i think we can easily-ish do if wanted, i think this adds zero-cost support for a not-trivially-dismissable case. I don't believe we need any upgrade infrastructure, and since we don't explicitly record the IR version, we don't need to bump one either. As @craig.topper speculates in D108661#2963519, this might be an artificial limit imposed by the original implementation of the `getAlignment()` functions. Differential Revision: https://reviews.llvm.org/D108661
33 lines
2.9 KiB
C++
33 lines
2.9 KiB
C++
// RUN: %clang_cc1 -fopenmp-simd -emit-llvm %s -o - -triple x86_64-linux-gnu | FileCheck %s
|
|
// RUN: %clang_cc1 -fopenmp-simd -fsanitize=alignment -fno-sanitize-recover=alignment -emit-llvm %s -o - -triple x86_64-linux-gnu | FileCheck %s -implicit-check-not="call void @__ubsan_handle_alignment_assumption" --check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-ANYRECOVER,CHECK-SANITIZE-NORECOVER,CHECK-SANITIZE-UNREACHABLE
|
|
// RUN: %clang_cc1 -fopenmp-simd -fsanitize=alignment -fsanitize-recover=alignment -emit-llvm %s -o - -triple x86_64-linux-gnu | FileCheck %s -implicit-check-not="call void @__ubsan_handle_alignment_assumption" --check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-ANYRECOVER,CHECK-SANITIZE-RECOVER
|
|
// RUN: %clang_cc1 -fopenmp-simd -fsanitize=alignment -fsanitize-trap=alignment -emit-llvm %s -o - -triple x86_64-linux-gnu | FileCheck %s -implicit-check-not="call void @__ubsan_handle_alignment_assumption" --check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-TRAP,CHECK-SANITIZE-UNREACHABLE
|
|
|
|
// CHECK-SANITIZE-ANYRECOVER: @[[CHAR:.*]] = {{.*}} c"'char *'\00" }
|
|
// CHECK-SANITIZE-ANYRECOVER: @[[LINE_100_ALIGNMENT_ASSUMPTION:.*]] = {{.*}}, i32 100, i32 30 }, {{.*}}* @[[CHAR]] }
|
|
|
|
void func(char *data) {
|
|
// CHECK: define{{.*}} void @{{.*}}(i8* %[[DATA:.*]])
|
|
// CHECK-NEXT: [[ENTRY:.*]]:
|
|
// CHECK-NEXT: %[[DATA_ADDR:.*]] = alloca i8*, align 8
|
|
// CHECK: store i8* %[[DATA]], i8** %[[DATA_ADDR]], align 8
|
|
// CHECK: %[[DATA_RELOADED:.*]] = load i8*, i8** %[[DATA_ADDR]], align 8
|
|
// CHECK-SANITIZE-NEXT: %[[PTRINT:.*]] = ptrtoint i8* %[[DATA_RELOADED]] to i64
|
|
// CHECK-SANITIZE-NEXT: %[[MASKEDPTR:.*]] = and i64 %[[PTRINT]], 2147483647
|
|
// CHECK-SANITIZE-NEXT: %[[MASKCOND:.*]] = icmp eq i64 %[[MASKEDPTR]], 0
|
|
// CHECK-SANITIZE-NEXT: %[[PTRINT_DUP:.*]] = ptrtoint i8* %[[DATA_RELOADED]] to i64, !nosanitize
|
|
// CHECK-SANITIZE-NEXT: br i1 %[[MASKCOND]], label %[[CONT:.*]], label %[[HANDLER_ALIGNMENT_ASSUMPTION:[^,]+]],{{.*}} !nosanitize
|
|
// CHECK-SANITIZE: [[HANDLER_ALIGNMENT_ASSUMPTION]]:
|
|
// CHECK-SANITIZE-NORECOVER-NEXT: call void @__ubsan_handle_alignment_assumption_abort(i8* bitcast ({ {{{.*}}}, {{{.*}}}, {{{.*}}}* }* @[[LINE_100_ALIGNMENT_ASSUMPTION]] to i8*), i64 %[[PTRINT_DUP]], i64 2147483648, i64 0){{.*}}, !nosanitize
|
|
// CHECK-SANITIZE-RECOVER-NEXT: call void @__ubsan_handle_alignment_assumption(i8* bitcast ({ {{{.*}}}, {{{.*}}}, {{{.*}}}* }* @[[LINE_100_ALIGNMENT_ASSUMPTION]] to i8*), i64 %[[PTRINT_DUP]], i64 2147483648, i64 0){{.*}}, !nosanitize
|
|
// CHECK-SANITIZE-TRAP-NEXT: call void @llvm.ubsantrap(i8 23){{.*}}, !nosanitize
|
|
// CHECK-SANITIZE-UNREACHABLE-NEXT: unreachable, !nosanitize
|
|
// CHECK-SANITIZE: [[CONT]]:
|
|
// CHECK-NEXT: call void @llvm.assume(i1 true) [ "align"(i8* %[[DATA_RELOADED]], i64 2147483648) ]
|
|
|
|
#line 100
|
|
#pragma omp for simd aligned(data : 0x80000000)
|
|
for (int x = 0; x < 1; x++)
|
|
data[x] = data[x];
|
|
}
|