From eebe418df9b24a13c81c3f7888ae45aec73da729 Mon Sep 17 00:00:00 2001 From: Tom Stellard Date: Mon, 8 Jul 2013 21:20:22 +0000 Subject: R600: Fix R600ControlFlowFinalizer not considering VTX_READ 128 bit dst reg Patch by: Vincent Lejeune https://bugs.freedesktop.org/show_bug.cgi?id=64877 NOTE: This is a candidate for the 3.3 branch. Merged from r182600 Author: Tom Stellard Date: Thu May 23 18:26:42 2013 +0000 llvm-svn: 185868 --- llvm/lib/Target/R600/R600ControlFlowFinalizer.cpp | 11 +++++++++-- llvm/test/CodeGen/R600/vtx-schedule.ll | 22 ++++++++++++++++++++++ 2 files changed, 31 insertions(+), 2 deletions(-) create mode 100644 llvm/test/CodeGen/R600/vtx-schedule.ll diff --git a/llvm/lib/Target/R600/R600ControlFlowFinalizer.cpp b/llvm/lib/Target/R600/R600ControlFlowFinalizer.cpp index ffe34144132a..7d3a0f52de5a 100644 --- a/llvm/lib/Target/R600/R600ControlFlowFinalizer.cpp +++ b/llvm/lib/Target/R600/R600ControlFlowFinalizer.cpp @@ -116,8 +116,15 @@ private: const MachineOperand &MO = *I; if (!MO.isReg()) continue; - if (MO.isDef()) - DstMI = MO.getReg(); + if (MO.isDef()) { + unsigned Reg = MO.getReg(); + if (AMDGPU::R600_Reg128RegClass.contains(Reg)) + DstMI = Reg; + else + DstMI = TRI.getMatchingSuperReg(Reg, + TRI.getSubRegFromChannel(TRI.getHWRegChan(Reg)), + &AMDGPU::R600_Reg128RegClass); + } if (MO.isUse()) { unsigned Reg = MO.getReg(); if (AMDGPU::R600_Reg128RegClass.contains(Reg)) diff --git a/llvm/test/CodeGen/R600/vtx-schedule.ll b/llvm/test/CodeGen/R600/vtx-schedule.ll new file mode 100644 index 000000000000..a0c79e36d3c5 --- /dev/null +++ b/llvm/test/CodeGen/R600/vtx-schedule.ll @@ -0,0 +1,22 @@ +; RUN: llc < %s -march=r600 -mcpu=redwood | FileCheck %s + +; This test is for a scheduler bug where VTX_READ instructions that used +; the result of another VTX_READ instruction were being grouped in the +; same fetch clasue. + +; CHECK: @test +; CHECK: Fetch clause +; CHECK_VTX_READ_32 [[IN0:T[0-9]+\.X]], [[IN0]], 40 +; CHECK_VTX_READ_32 [[IN1:T[0-9]+\.X]], [[IN1]], 44 +; CHECK: Fetch clause +; CHECK_VTX_READ_32 [[IN0:T[0-9]+\.X]], [[IN0]], 0 +; CHECK_VTX_READ_32 [[IN1:T[0-9]+\.X]], [[IN1]], 0 +define void @test(i32 addrspace(1)* nocapture %out, i32 addrspace(1)* nocapture %in0, i32 addrspace(1)* nocapture %in1) { +entry: + %0 = load i32 addrspace(1)* %in0, align 4 + %1 = load i32 addrspace(1)* %in1, align 4 + %cmp.i = icmp slt i32 %0, %1 + %cond.i = select i1 %cmp.i, i32 %0, i32 %1 + store i32 %cond.i, i32 addrspace(1)* %out, align 4 + ret void +} -- cgit v1.2.3