844c188c79
Update getDependenceDistanceStrideAndSize to reason about different combinations of strides directly and explicitly. Update getPtrStride to return 0 for invariant pointers. Then proceed by checking the strides. If either source or sink are not strided by a constant (i.e. not a non-wrapping AddRec) or invariant, the accesses may overlap with earlier or later iterations and we cannot generate runtime checks to disambiguate them. Otherwise they are either loop invariant or strided. In that case, we can generate a runtime check to disambiguate them. If both are strided by constants, we proceed as previously. This is an alternative to https://github.com/llvm/llvm-project/pull/99239 and also replaces additional checks if the underlying object is loop-invariant. Fixes https://github.com/llvm/llvm-project/issues/87189. PR: https://github.com/llvm/llvm-project/pull/99577
93 lines
3.6 KiB
LLVM
93 lines
3.6 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 5
|
|
; RUN: opt -passes='print<access-info>' -disable-output %s 2>&1 | FileCheck %s
|
|
|
|
; Test case for https://github.com/llvm/llvm-project/issues/87189.
|
|
; It is not safe to vectorize because %indices are loaded in the loop and the
|
|
; same indices could be loaded in later iterations.
|
|
; FIXME: currently this is incorrectly considered safe for vectorization with
|
|
; runtime checks
|
|
define void @B_indices_loaded_in_loop_A_stored(ptr %A, ptr noalias %B, i64 %N, i64 %off) {
|
|
; CHECK-LABEL: 'B_indices_loaded_in_loop_A_stored'
|
|
; CHECK-NEXT: loop:
|
|
; CHECK-NEXT: Report: unsafe dependent memory operations in loop. Use #pragma clang loop distribute(enable) to allow loop distribution to attempt to isolate the offending operations into a separate loop
|
|
; CHECK-NEXT: Unsafe indirect dependence.
|
|
; CHECK-NEXT: Dependences:
|
|
; CHECK-NEXT: IndirectUnsafe:
|
|
; CHECK-NEXT: %l = load i32, ptr %gep.B, align 4 ->
|
|
; CHECK-NEXT: store i32 %inc, ptr %gep.B, align 4
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: Unknown:
|
|
; CHECK-NEXT: %indices = load i8, ptr %gep.A, align 1 ->
|
|
; CHECK-NEXT: store i32 %l, ptr %gep.C, align 4
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: Run-time memory checks:
|
|
; CHECK-NEXT: Grouped accesses:
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: Non vectorizable stores to invariant address were not found in loop.
|
|
; CHECK-NEXT: SCEV assumptions:
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: Expressions re-written:
|
|
;
|
|
entry:
|
|
br label %loop
|
|
|
|
loop:
|
|
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
|
|
%iv.off = add nuw nsw i64 %iv, %off
|
|
%gep.A = getelementptr inbounds i8, ptr %A, i64 %iv.off
|
|
%indices = load i8, ptr %gep.A, align 1
|
|
%indices.ext = zext i8 %indices to i64
|
|
%gep.B = getelementptr inbounds i32, ptr %B, i64 %indices.ext
|
|
%l = load i32, ptr %gep.B, align 4
|
|
%inc = add i32 %l, 1
|
|
store i32 %inc, ptr %gep.B, align 4
|
|
%gep.C = getelementptr inbounds i32, ptr %A, i64 %iv
|
|
store i32 %l, ptr %gep.C, align 4
|
|
%iv.next = add nuw nsw i64 %iv, 1
|
|
%ec = icmp eq i64 %iv.next, %N
|
|
br i1 %ec, label %exit, label %loop
|
|
|
|
exit:
|
|
ret void
|
|
}
|
|
|
|
; It is not safe to vectorize because %indices are loaded in the loop and the
|
|
; same indices could be loaded in later iterations.
|
|
define void @B_indices_loaded_in_loop_A_not_stored(ptr %A, ptr noalias %B, i64 %N) {
|
|
; CHECK-LABEL: 'B_indices_loaded_in_loop_A_not_stored'
|
|
; CHECK-NEXT: loop:
|
|
; CHECK-NEXT: Report: unsafe dependent memory operations in loop. Use #pragma clang loop distribute(enable) to allow loop distribution to attempt to isolate the offending operations into a separate loop
|
|
; CHECK-NEXT: Unsafe indirect dependence.
|
|
; CHECK-NEXT: Dependences:
|
|
; CHECK-NEXT: IndirectUnsafe:
|
|
; CHECK-NEXT: %l = load i32, ptr %gep.B, align 4 ->
|
|
; CHECK-NEXT: store i32 %inc, ptr %gep.B, align 4
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: Run-time memory checks:
|
|
; CHECK-NEXT: Grouped accesses:
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: Non vectorizable stores to invariant address were not found in loop.
|
|
; CHECK-NEXT: SCEV assumptions:
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: Expressions re-written:
|
|
;
|
|
entry:
|
|
br label %loop
|
|
|
|
loop:
|
|
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
|
|
%gep.A = getelementptr inbounds i8, ptr %A, i64 %iv
|
|
%indices = load i8, ptr %gep.A, align 1
|
|
%indices.ext = zext i8 %indices to i64
|
|
%gep.B = getelementptr inbounds i32, ptr %B, i64 %indices.ext
|
|
%l = load i32, ptr %gep.B, align 4
|
|
%inc = add i32 %l, 1
|
|
store i32 %inc, ptr %gep.B, align 4
|
|
%iv.next = add nuw nsw i64 %iv, 1
|
|
%ec = icmp eq i64 %iv.next, %N
|
|
br i1 %ec, label %exit, label %loop
|
|
|
|
exit:
|
|
ret void
|
|
}
|