From a41c886c551156a458bbd932f3d816add9d98a25 Mon Sep 17 00:00:00 2001 From: Sanjay Patel Date: Sun, 15 Jul 2018 16:27:07 +0000 Subject: [DAGCombiner] extend(ifpositive(X)) -> shift-right (not X) This is almost the same as an existing IR canonicalization in instcombine, so I'm assuming this is a good early generic DAG combine too. The motivation comes from reduced bit-hacking for select-of-constants in IR after rL331486. We want to restore that functionality in the DAG as noted in the commit comments for that change and the llvm-dev discussion here: http://lists.llvm.org/pipermail/llvm-dev/2018-July/124433.html The PPC and AArch tests show that those targets are already doing something similar. x86 will be neutral in the minimal case and generally better when this pattern is extended with other ops as shown in the signbit-shift.ll tests. Note the asymmetry: we don't include the (extend (ifneg X)) transform because it already exists in SimplifySelectCC(), and that is verified in the later unchanged tests in the signbit-shift.ll files. Without the 'not' op, the general transform to use a shift is always a win because that's a single instruction. Alive proofs: https://rise4fun.com/Alive/ysli Name: if pos, get -1 %c = icmp sgt i16 %x, -1 %r = sext i1 %c to i16 => %n = xor i16 %x, -1 %r = ashr i16 %n, 15 Name: if pos, get 1 %c = icmp sgt i16 %x, -1 %r = zext i1 %c to i16 => %n = xor i16 %x, -1 %r = lshr i16 %n, 15 Differential Revision: https://reviews.llvm.org/D48970 llvm-svn: 337130 --- llvm/test/CodeGen/AMDGPU/commute-compares.ll | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'llvm/test/CodeGen/AMDGPU') diff --git a/llvm/test/CodeGen/AMDGPU/commute-compares.ll b/llvm/test/CodeGen/AMDGPU/commute-compares.ll index e12a968a987..20d1ea5ac54 100644 --- a/llvm/test/CodeGen/AMDGPU/commute-compares.ll +++ b/llvm/test/CodeGen/AMDGPU/commute-compares.ll @@ -114,7 +114,7 @@ define amdgpu_kernel void @commute_ule_64_i32(i32 addrspace(1)* %out, i32 addrsp } ; GCN-LABEL: {{^}}commute_sgt_neg1_i32: -; GCN: v_cmp_lt_i32_e32 vcc, -1, v{{[0-9]+}} +; GCN: v_ashrrev_i32_e32 v2, 31, v2 define amdgpu_kernel void @commute_sgt_neg1_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %in) #1 { %tid = call i32 @llvm.amdgcn.workitem.id.x() #0 %gep.in = getelementptr i32, i32 addrspace(1)* %in, i32 %tid -- cgit v1.2.3