d2210af332
Current findBestLoopTop can find and move one kind of block to top, a latch block has one successor. Another common case is:
* a latch block
* it has two successors, one is loop header, another is exit
* it has more than one predecessors
If it is below one of its predecessors P, only P can fall through to it, all other predecessors need a jump to it, and another conditional jump to loop header. If it is moved before loop header, all its predecessors jump to it, then fall through to loop header. So all its predecessors except P can reduce one taken branch.
Differential Revision: https://reviews.llvm.org/D43256
llvm-svn: 363471
76 lines
1.7 KiB
LLVM
76 lines
1.7 KiB
LLVM
; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
|
|
|
|
; GCN-LABEL: {{^}}negated_cond:
|
|
; GCN: BB0_1:
|
|
; GCN: v_cmp_eq_u32_e64 [[CC:[^,]+]],
|
|
; GCN: BB0_3:
|
|
; GCN-NOT: v_cndmask_b32
|
|
; GCN-NOT: v_cmp
|
|
; GCN: s_andn2_b64 vcc, exec, [[CC]]
|
|
; GCN: s_cbranch_vccnz BB0_2
|
|
define amdgpu_kernel void @negated_cond(i32 addrspace(1)* %arg1) {
|
|
bb:
|
|
br label %bb1
|
|
|
|
bb1:
|
|
%tmp1 = load i32, i32 addrspace(1)* %arg1
|
|
%tmp2 = icmp eq i32 %tmp1, 0
|
|
br label %bb2
|
|
|
|
bb2:
|
|
%tmp3 = phi i32 [ 0, %bb1 ], [ %tmp6, %bb4 ]
|
|
%tmp4 = shl i32 %tmp3, 5
|
|
br i1 %tmp2, label %bb3, label %bb4
|
|
|
|
bb3:
|
|
%tmp5 = add i32 %tmp4, 1
|
|
br label %bb4
|
|
|
|
bb4:
|
|
%tmp6 = phi i32 [ %tmp5, %bb3 ], [ %tmp4, %bb2 ]
|
|
%gep = getelementptr inbounds i32, i32 addrspace(1)* %arg1, i32 %tmp6
|
|
store i32 0, i32 addrspace(1)* %gep
|
|
%tmp7 = icmp eq i32 %tmp6, 32
|
|
br i1 %tmp7, label %bb1, label %bb2
|
|
}
|
|
|
|
; GCN-LABEL: {{^}}negated_cond_dominated_blocks:
|
|
; GCN: v_cmp_eq_u32_e64 [[CC:[^,]+]],
|
|
; GCN: %bb4
|
|
; GCN-NOT: v_cndmask_b32
|
|
; GCN-NOT: v_cmp
|
|
; GCN: s_andn2_b64 vcc, exec, [[CC]]
|
|
; GCN: s_cbranch_vccnz BB1_1
|
|
define amdgpu_kernel void @negated_cond_dominated_blocks(i32 addrspace(1)* %arg1) {
|
|
bb:
|
|
br label %bb2
|
|
|
|
bb2:
|
|
%tmp1 = load i32, i32 addrspace(1)* %arg1
|
|
%tmp2 = icmp eq i32 %tmp1, 0
|
|
br label %bb4
|
|
|
|
bb3:
|
|
ret void
|
|
|
|
bb4:
|
|
%tmp3 = phi i32 [ 0, %bb2 ], [ %tmp7, %bb7 ]
|
|
%tmp4 = shl i32 %tmp3, 5
|
|
br i1 %tmp2, label %bb5, label %bb6
|
|
|
|
bb5:
|
|
%tmp5 = add i32 %tmp4, 1
|
|
br label %bb7
|
|
|
|
bb6:
|
|
%tmp6 = add i32 %tmp3, 1
|
|
br label %bb7
|
|
|
|
bb7:
|
|
%tmp7 = phi i32 [ %tmp5, %bb5 ], [ %tmp6, %bb6 ]
|
|
%gep = getelementptr inbounds i32, i32 addrspace(1)* %arg1, i32 %tmp7
|
|
store i32 0, i32 addrspace(1)* %gep
|
|
%tmp8 = icmp eq i32 %tmp7, 32
|
|
br i1 %tmp8, label %bb3, label %bb4
|
|
}
|