[AMDGPU] Mark relevant rematerializable VOP3 instructions

Differential Revision: https://reviews.llvm.org/D106110
llvm · Jul 21, 2021 · fe197ef · fe197ef
1 parent 9ef7de7
commit fe197ef
Show file tree

Hide file tree

Showing 3 changed files with 1,196 additions and 12 deletions.
diff --git a/llvm/lib/Target/AMDGPU/VOP3Instructions.td b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
@@ -275,6 +275,7 @@ class VOP3_INTERP16 <list<ValueType> ArgVT> : VOPProfile<ArgVT> {
 
 let isCommutable = 1 in {
 
+let isReMaterializable = 1 in {
 let mayRaiseFPException = 0 in {
 let SubtargetPredicate = HasMadMacF32Insts in {
 defm V_MAD_LEGACY_F32 : VOP3Inst <"v_mad_legacy_f32", VOP3_Profile<VOP_F32_F32_F32_F32>>;
@@ -308,6 +309,7 @@ defm V_MUL_HI_U32 : VOP3Inst <"v_mul_hi_u32", VOP3_Profile<VOP_I32_I32_I32>, mul
 defm V_MUL_LO_I32 : VOP3Inst <"v_mul_lo_i32", VOP3_Profile<VOP_I32_I32_I32>>;
 defm V_MUL_HI_I32 : VOP3Inst <"v_mul_hi_i32", VOP3_Profile<VOP_I32_I32_I32>, mulhs>;
 } // End SchedRW = [WriteIntMul]
+} // End isReMaterializable = 1
 
 let Uses = [MODE, VCC, EXEC] in {
 // v_div_fmas_f32:
@@ -328,6 +330,7 @@ defm V_DIV_FMAS_F64 : VOP3Inst_Pseudo_Wrapper  <"v_div_fmas_f64", VOP_F64_F64_F6
 
 } // End isCommutable = 1
 
+let isReMaterializable = 1 in {
 let mayRaiseFPException = 0 in {
 defm V_CUBEID_F32 : VOP3Inst <"v_cubeid_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, int_amdgcn_cubeid>;
 defm V_CUBESC_F32 : VOP3Inst <"v_cubesc_f32", VOP3_Profile<VOP_F32_F32_F32_F32>, int_amdgcn_cubesc>;
@@ -370,6 +373,7 @@ let SchedRW = [WriteDoubleAdd], FPDPRounding = 1 in {
   defm V_DIV_FIXUP_F64 : VOP3Inst <"v_div_fixup_f64", VOP3_Profile<VOP_F64_F64_F64_F64>, AMDGPUdiv_fixup>;
   defm V_LDEXP_F64 : VOP3Inst <"v_ldexp_f64", VOP3_Profile<VOP_F64_F64_I32>, AMDGPUldexp, 1>;
 } // End SchedRW = [WriteDoubleAdd], FPDPRounding = 1
+} // End isReMaterializable = 1
 
 
 let mayRaiseFPException = 0 in { // Seems suspicious but manual doesn't say it does.
@@ -381,13 +385,15 @@ let mayRaiseFPException = 0 in { // Seems suspicious but manual doesn't say it d
   defm V_DIV_SCALE_F64 : VOP3Inst_Pseudo_Wrapper <"v_div_scale_f64", VOP3b_F64_I1_F64_F64_F64, [], 1>;
 } // End mayRaiseFPException = 0
 
+let isReMaterializable = 1 in
 defm V_MSAD_U8 : VOP3Inst <"v_msad_u8", VOP3_Profile<VOP_I32_I32_I32_I32, VOP3_CLAMP>>;
 
 let Constraints = "@earlyclobber $vdst" in {
 defm V_MQSAD_PK_U16_U8 : VOP3Inst <"v_mqsad_pk_u16_u8", VOP3_Profile<VOP_I64_I64_I32_I64, VOP3_CLAMP>>;
 } // End Constraints = "@earlyclobber $vdst"
 
 
+let isReMaterializable = 1 in {
 let SchedRW = [WriteDouble] in {
 defm V_TRIG_PREOP_F64 : VOP3Inst <"v_trig_preop_f64", VOP3_Profile<VOP_F64_F64_I32>, int_amdgcn_trig_preop>;
 } // End SchedRW = [WriteDouble]
@@ -405,12 +411,14 @@ let SchedRW = [Write64Bit] in {
   defm V_ASHRREV_I64 : VOP3Inst <"v_ashrrev_i64", VOP3_Profile<VOP_I64_I32_I64>, ashr_rev>;
   } // End SubtargetPredicate = isGFX8Plus
 } // End SchedRW = [Write64Bit]
+} // End isReMaterializable = 1
 
 def : GCNPat<
   (i32 (getDivergentFrag<sext>.ret i16:$src)),
   (i32 (V_BFE_I32_e64 $src, (S_MOV_B32 (i32 0)), (S_MOV_B32 (i32 0x10))))
 >;
 
+let isReMaterializable = 1 in {
 let SubtargetPredicate = isGFX6GFX7GFX10 in {
 defm V_MULLIT_F32 : VOP3Inst <"v_mullit_f32", VOP3_Profile<VOP_F32_F32_F32_F32>>;
 } // End SubtargetPredicate = isGFX6GFX7GFX10
@@ -420,6 +428,7 @@ let SubtargetPredicate = isGFX8Plus in {
 defm V_PERM_B32 : VOP3Inst <"v_perm_b32", VOP3_Profile<VOP_I32_I32_I32_I32>, AMDGPUperm>;
 } // End SubtargetPredicate = isGFX8Plus
 } // End SchedRW = [Write32Bit]
+} // End isReMaterializable = 1
 
 let SubtargetPredicate = isGFX7Plus in {
 
@@ -601,14 +610,14 @@ class ThreeOpFrag<SDPatternOperator op1, SDPatternOperator op2> : PatFrag<
 }
 
 let SubtargetPredicate = isGFX9Plus in {
-let isCommutable = 1 in {
+let isCommutable = 1, isReMaterializable = 1 in {
   defm V_ADD3_U32 : VOP3Inst <"v_add3_u32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
   defm V_AND_OR_B32 : VOP3Inst <"v_and_or_b32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
   defm V_OR3_B32 : VOP3Inst <"v_or3_b32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
   defm V_XAD_U32 : VOP3Inst <"v_xad_u32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
   defm V_ADD_I32 : VOP3Inst <"v_add_i32", VOP3_Profile<VOP_I32_I32_I32_ARITH>>;
   defm V_ADD_LSHL_U32 : VOP3Inst <"v_add_lshl_u32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
-} // End isCommutable = 1
+} // End isCommutable = 1, isReMaterializable = 1
 // TODO src0 contains the opsel bit for dst, so if we commute, need to mask and swap this
 // to the new src0.
 defm V_MED3_F16 : VOP3Inst <"v_med3_f16", VOP3_Profile<VOP_F16_F16_F16_F16, VOP3_OPSEL>, AMDGPUfmed3>;
@@ -632,11 +641,13 @@ defm V_MAD_I32_I16 : VOP3Inst <"v_mad_i32_i16", VOP3_Profile<VOP_I32_I16_I16_I32
 defm V_CVT_PKNORM_I16_F16 : VOP3Inst <"v_cvt_pknorm_i16_f16", VOP3_Profile<VOP_B32_F16_F16, VOP3_OPSEL>>;
 defm V_CVT_PKNORM_U16_F16 : VOP3Inst <"v_cvt_pknorm_u16_f16", VOP3_Profile<VOP_B32_F16_F16, VOP3_OPSEL>>;
 
-defm V_SUB_I32 : VOP3Inst <"v_sub_i32", VOP3_Profile<VOP_I32_I32_I32_ARITH>>;
 defm V_PACK_B32_F16 : VOP3Inst <"v_pack_b32_f16", VOP3_Profile<VOP_B32_F16_F16, VOP3_OPSEL>>;
+
+let isReMaterializable = 1 in {
+defm V_SUB_I32 : VOP3Inst <"v_sub_i32", VOP3_Profile<VOP_I32_I32_I32_ARITH>>;
 defm V_LSHL_ADD_U32 : VOP3Inst <"v_lshl_add_u32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
 defm V_LSHL_OR_B32 : VOP3Inst <"v_lshl_or_b32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
-
+} // End isReMaterializable = 1
 
 
 class ThreeOp_i32_Pats <SDPatternOperator op1, SDPatternOperator op2, Instruction inst> : GCNPat <
@@ -715,9 +726,9 @@ class PermlaneDiscardVDstIn<SDPatternOperator permlane,
 
 
 let SubtargetPredicate = isGFX10Plus in {
-  let isCommutable = 1 in {
+  let isCommutable = 1, isReMaterializable = 1 in {
     defm V_XOR3_B32 : VOP3Inst <"v_xor3_b32", VOP3_Profile<VOP_I32_I32_I32_I32>>;
-  } // End isCommutable = 1
+  } // End isCommutable = 1, isReMaterializable = 1
   def : ThreeOp_i32_Pats<xor, xor, V_XOR3_B32_e64>;
 
   let Constraints = "$vdst = $vdst_in", DisableEncoding="$vdst_in" in {