[AMDGPU] Do not reserve 16-bit registers

There should be no need to reserve all SGPR hi16/lo16 halves, or all
AGPR hi16 halves. This should be done by marking the corresponding
register classes as not allocatable instead.

Differential Revision: https://reviews.llvm.org/D147158

diff --git a/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp b/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
index c281e36..9d94344 100644 (file)
--- a/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
@@ -607,14 +607,6 @@ BitVector SIRegisterInfo::getReservedRegs(const MachineFunction &MF) const {
     reserveRegisterTuples(Reserved, Reg);
   }
 
-  for (auto Reg : AMDGPU::SReg_32RegClass) {
-    Reserved.set(getSubReg(Reg, AMDGPU::hi16));
-    Register Low = getSubReg(Reg, AMDGPU::lo16);
-    // This is to prevent BB vcc liveness errors.
-    if (!AMDGPU::SGPR_LO16RegClass.contains(Low))
-      Reserved.set(Low);
-  }
-
   Register ScratchRSrcReg = MFI->getScratchRSrcReg();
   if (ScratchRSrcReg != AMDGPU::NoRegister) {
     // Reserve 4 SGPRs for the scratch buffer resource descriptor in case we
@@ -650,10 +642,6 @@ BitVector SIRegisterInfo::getReservedRegs(const MachineFunction &MF) const {
   unsigned MaxNumAGPRs = MaxNumVGPRs;
   unsigned TotalNumVGPRs = AMDGPU::VGPR_32RegClass.getNumRegs();
 
-  for (auto Reg : AMDGPU::AGPR_32RegClass) {
-    Reserved.set(getSubReg(Reg, AMDGPU::hi16));
-  }
-
   // On GFX90A, the number of VGPRs and AGPRs need not be equal. Theoretically,
   // a wave may have up to 512 total vector registers combining together both
   // VGPRs and AGPRs. Hence, in an entry function without calls and without

diff --git a/llvm/lib/Target/AMDGPU/SIRegisterInfo.td b/llvm/lib/Target/AMDGPU/SIRegisterInfo.td
index 1baff1f..060f631 100644 (file)
--- a/llvm/lib/Target/AMDGPU/SIRegisterInfo.td
+++ b/llvm/lib/Target/AMDGPU/SIRegisterInfo.td
@@ -774,7 +774,7 @@ def SReg_LO16 : SIRegisterClass<"AMDGPU", [i16, f16], 16,
    SRC_PRIVATE_LIMIT_HI_LO16, SRC_POPS_EXITING_WAVE_ID_LO16, SRC_VCCZ_LO16,
    SRC_EXECZ_LO16, SRC_SCC_LO16, EXEC_LO_LO16, EXEC_HI_LO16, M0_CLASS_LO16)> {
   let Size = 16;
-  let AllocationPriority = 0;
+  let isAllocatable = 0;
   let BaseClassOrder = 16;
 }