[RISCV] Add matching of codegen patterns to RISCV Bit Manipulation Zbp asm instructions

[PaoloS]

Summary

This patch provides optimization of bit manipulation operations by enabling
the +experimental-b target feature.
It adds matching of single block patterns of instructions to specific bit-manip
instructions from the permutation subset (zbp subextension) of the experimental
B extension of RISC-V.
It adds also the correspondent codegen tests.

This patch is based on Clifford Wolf's proposal for the bit manipulation
extension of RISCV:
https://github.com/riscv/riscv-bitmanip/blob/master/bitmanip-0.92.pdf

[PaoloS]

Add missing pattern-matching for *w instructions.
Add tests.
Add both i32 and i64 code versions to both i32 and i64 test files.
Remove NOT labels from tests.

[PaoloS]

Updated the test:

• the tests have been updated from the top of all the sub-patches together so that they are exactly the same as they would be if updated with the whole final patch.
• labels specific to the sub-extension have been added alongside the generic RISCVIB label (that activates all the sub-extensions) so that we can see how differently the patterns are matched with the specific subextension or with all of them together.
• the tests will probably fail if run by checking out the commit of a subextension and if updated they'll change. These tests are designed to work with the final squashed patch.

Notice that rori takes over grevi on RV64 for the rotation as it is exactly the same, and because the activation of the Zbp subextension is eough to activate the "ZbbOrZbp" subextension.

[lewis-revill]

LGTM

[PaoloS]

Making sure the changes from patch zbb are maintained.

llvm/lib/Target/RISCV/RISCVISelLowering.cpp

RISCVTargetLowering::RISCVTargetLowering(const TargetMachine &TM,
   setOperationAction(ISD::UMUL_LOHI, XLenVT, Expand);
 
   setOperationAction(ISD::SHL_PARTS, XLenVT, Custom);
   setOperationAction(ISD::SRL_PARTS, XLenVT, Custom);
   setOperationAction(ISD::SRA_PARTS, XLenVT, Custom);
 
   setOperationAction(ISD::ROTL, XLenVT, Expand);
   setOperationAction(ISD::ROTR, XLenVT, Expand);
+
+  if (!Subtarget.hasStdExtZbp())
     setOperationAction(ISD::BSWAP, XLenVT, Expand);
 
   if (!Subtarget.hasStdExtZbb()) {
     setOperationAction(ISD::CTTZ, XLenVT, Expand);
     setOperationAction(ISD::CTLZ, XLenVT, Expand);
     setOperationAction(ISD::CTPOP, XLenVT, Expand);
   }
 
+  if (Subtarget.hasStdExtZbp())
+    setOperationAction(ISD::BITREVERSE, XLenVT, Legal);
+
   ISD::CondCode FPCCToExtend[] = {
       ISD::SETOGT, ISD::SETOGE, ISD::SETONE, ISD::SETUEQ, ISD::SETUGT,
       ISD::SETUGE, ISD::SETULT, ISD::SETULE, ISD::SETUNE, ISD::SETGT,
       ISD::SETGE,  ISD::SETNE};
 
   ISD::NodeType FPOpToExtend[] = {
       ISD::FSIN, ISD::FCOS, ISD::FSINCOS, ISD::FPOW, ISD::FREM, ISD::FP16_TO_FP,
       ISD::FP_TO_FP16};

llvm/lib/Target/RISCV/RISCVInstrInfoB.td

 def : Pat<(xor (shl (xor GPR:$rs1, -1), GPR:$rs2), -1),
           (SLO GPR:$rs1, GPR:$rs2)>;
 def : Pat<(xor (srl (xor GPR:$rs1, -1), GPR:$rs2), -1),
           (SRO GPR:$rs1, GPR:$rs2)>;
 def : Pat<(SLOIPat GPR:$rs1, uimmlog2xlen:$shamt),
           (SLOI GPR:$rs1, uimmlog2xlen:$shamt)>;
 def : Pat<(SROIPat GPR:$rs1, uimmlog2xlen:$shamt),
           (SROI GPR:$rs1, uimmlog2xlen:$shamt)>;
+} // Predicates = [HasStdExtZbb]
+
+let Predicates = [HasStdExtZbp, IsRV32] in {
+def : Pat<(or (or (and (srl GPR:$rs1, (i32 1)), (i32 0x55555555)), GPR:$rs1),
+              (and (shl GPR:$rs1, (i32 1)), (i32 0xAAAAAAAA))),
+          (GORCI GPR:$rs1, (i32 1))>;
+def : Pat<(or (or (and (srl GPR:$rs1, (i32 2)), (i32 0x33333333)), GPR:$rs1),
+              (and (shl GPR:$rs1, (i32 2)), (i32 0xCCCCCCCC))),
+          (GORCI GPR:$rs1, (i32 2))>;
+def : Pat<(or (or (and (srl GPR:$rs1, (i32 4)), (i32 0x0F0F0F0F)), GPR:$rs1),
+              (and (shl GPR:$rs1, (i32 4)), (i32 0xF0F0F0F0))),
+          (GORCI GPR:$rs1, (i32 4))>;
+def : Pat<(or (or (and (srl GPR:$rs1, (i32 8)), (i32 0x00FF00FF)), GPR:$rs1),
+              (and (shl GPR:$rs1, (i32 8)), (i32 0xFF00FF00))),
+          (GORCI GPR:$rs1, (i32 8))>;
+def : Pat<(or (or (srl GPR:$rs1, (i32 16)), GPR:$rs1),
+              (shl GPR:$rs1, (i32 16))),
+          (GORCI GPR:$rs1, (i32 16))>;
+} // Predicates = [HasStdExtZbp, IsRV32]
+
+let Predicates = [HasStdExtZbp, IsRV64] in {
+def : Pat<(or (or (and (srl GPR:$rs1, (i64 1)), (i64 0x5555555555555555)),
+                   GPR:$rs1),
+              (and (shl GPR:$rs1, (i64 1)), (i64 0xAAAAAAAAAAAAAAAA))),
+          (GORCI GPR:$rs1, (i64 1))>;
+def : Pat<(or (or (and (srl GPR:$rs1, (i64 2)), (i64 0x3333333333333333)),
+                   GPR:$rs1),
+              (and (shl GPR:$rs1, (i64 2)), (i64 0xCCCCCCCCCCCCCCCC))),
+          (GORCI GPR:$rs1, (i64 2))>;
+def : Pat<(or (or (and (srl GPR:$rs1, (i64 4)), (i64 0x0F0F0F0F0F0F0F0F)),
+                   GPR:$rs1),
+              (and (shl GPR:$rs1, (i64 4)), (i64 0xF0F0F0F0F0F0F0F0))),
+          (GORCI GPR:$rs1, (i64 4))>;
+def : Pat<(or (or (and (srl GPR:$rs1, (i64 8)), (i64 0x00FF00FF00FF00FF)),
+                   GPR:$rs1),
+              (and (shl GPR:$rs1, (i64 8)), (i64 0xFF00FF00FF00FF00))),
+          (GORCI GPR:$rs1, (i64 8))>;
+def : Pat<(or (or (and (srl GPR:$rs1, (i64 16)), (i64 0x0000FFFF0000FFFF)),
+                   GPR:$rs1),
+              (and (shl GPR:$rs1, (i64 16)), (i64 0xFFFF0000FFFF0000))),
+          (GORCI GPR:$rs1, (i64 16))>;
+def : Pat<(or (or (srl GPR:$rs1, (i64 32)), GPR:$rs1),
+              (shl GPR:$rs1, (i64 32))),
+          (GORCI GPR:$rs1, (i64 32))>;
+} // Predicates = [HasStdExtZbp, IsRV64]
+
+let Predicates = [HasStdExtZbp, IsRV32] in {
+def : Pat<(or (and (shl GPR:$rs1, (i32 1)), (i32 0xAAAAAAAA)),
+              (and (srl GPR:$rs1, (i32 1)), (i32 0x55555555))),
+          (GREVI GPR:$rs1, (i32 1))>;
+def : Pat<(or (and (shl GPR:$rs1, (i32 2)), (i32 0xCCCCCCCC)),
+              (and (srl GPR:$rs1, (i32 2)), (i32 0x33333333))),
+          (GREVI GPR:$rs1, (i32 2))>;
+def : Pat<(or (and (shl GPR:$rs1, (i32 4)), (i32 0xF0F0F0F0)),
+              (and (srl GPR:$rs1, (i32 4)), (i32 0x0F0F0F0F))),
+          (GREVI GPR:$rs1, (i32 4))>;
+def : Pat<(or (and (shl GPR:$rs1, (i32 8)), (i32 0xFF00FF00)),
+              (and (srl GPR:$rs1, (i32 8)), (i32 0x00FF00FF))),
+          (GREVI GPR:$rs1, (i32 8))>;
+def : Pat<(rotr (bswap GPR:$rs1), (i32 16)), (GREVI GPR:$rs1, (i32 8))>;
+def : Pat<(or (shl GPR:$rs1, (i32 16)), (srl GPR:$rs1, (i32 16))),
+          (GREVI GPR:$rs1, (i32 16))>;
+def : Pat<(rotl GPR:$rs1, (i32 16)), (GREVI GPR:$rs1, (i32 16))>;
+def : Pat<(bswap GPR:$rs1), (GREVI GPR:$rs1, (i32 24))>;
+def : Pat<(bitreverse GPR:$rs1), (GREVI GPR:$rs1, (i32 31))>;
+} // Predicates = [HasStdExtZbp, IsRV32]
+
+let Predicates = [HasStdExtZbp, IsRV64] in {
+def : Pat<(or (and (shl GPR:$rs1, (i64 1)), (i64 0xAAAAAAAAAAAAAAAA)),
+              (and (srl GPR:$rs1, (i64 1)), (i64 0x5555555555555555))),
+          (GREVI GPR:$rs1, (i64 1))>;
+def : Pat<(or (and (shl GPR:$rs1, (i64 2)), (i64 0xCCCCCCCCCCCCCCCC)),
+              (and (srl GPR:$rs1, (i64 2)), (i64 0x3333333333333333))),
+          (GREVI GPR:$rs1, (i64 2))>;
+def : Pat<(or (and (shl GPR:$rs1, (i64 4)), (i64 0xF0F0F0F0F0F0F0F0)),
+              (and (srl GPR:$rs1, (i64 4)), (i64 0x0F0F0F0F0F0F0F0F))),
+          (GREVI GPR:$rs1, (i64 4))>;
+def : Pat<(or (and (shl GPR:$rs1, (i64 8)), (i64 0xFF00FF00FF00FF00)),
+              (and (srl GPR:$rs1, (i64 8)), (i64 0x00FF00FF00FF00FF))),
+          (GREVI GPR:$rs1, (i64 8))>;
+def : Pat<(or (and (shl GPR:$rs1, (i64 16)), (i64 0xFFFF0000FFFF0000)),
+              (and (srl GPR:$rs1, (i64 16)), (i64 0x0000FFFF0000FFFF))),
+          (GREVI GPR:$rs1, (i64 16))>;
+def : Pat<(or (shl GPR:$rs1, (i64 32)), (srl GPR:$rs1, (i64 32))),
+          (GREVI GPR:$rs1, (i64 32))>;
+def : Pat<(rotl GPR:$rs1, (i64 32)), (GREVI GPR:$rs1, (i64 32))>;
+def : Pat<(bswap GPR:$rs1), (GREVI GPR:$rs1, (i64 56))>;
+def : Pat<(bitreverse GPR:$rs1), (GREVI GPR:$rs1, (i64 63))>;
+} // Predicates = [HasStdExtZbp, IsRV64]
+
+let Predicates = [HasStdExtZbb] in {
 def : Pat<(ctlz GPR:$rs1), (CLZ GPR:$rs1)>;
 def : Pat<(cttz GPR:$rs1), (CTZ GPR:$rs1)>;
 def : Pat<(ctpop GPR:$rs1), (PCNT GPR:$rs1)>;
 } // Predicates = [HasStdExtZbb]
 
 let Predicates = [HasStdExtZbb, IsRV32] in
 def : Pat<(sra (shl GPR:$rs1, (i32 24)), (i32 24)), (SEXTB GPR:$rs1)>;
 let Predicates = [HasStdExtZbb, IsRV64] in

 def : Pat<(umin GPR:$rs1, GPR:$rs2), (MINU GPR:$rs1, GPR:$rs2)>;
 def : Pat<(riscv_selectcc GPR:$rs1, GPR:$rs2, (XLenVT 12), GPR:$rs1, GPR:$rs2),
           (MINU  GPR:$rs1, GPR:$rs2)>;
 def : Pat<(umax GPR:$rs1, GPR:$rs2), (MAXU GPR:$rs1, GPR:$rs2)>;
 def : Pat<(riscv_selectcc GPR:$rs2, GPR:$rs1, (XLenVT 12), GPR:$rs1, GPR:$rs2),
           (MAXU  GPR:$rs1, GPR:$rs2)>;
 } // Predicates = [HasStdExtZbb]
 
+let Predicates = [HasStdExtZbp, IsRV32] in {
+def : Pat<(or (or (and (shl GPR:$rs1, (i32 8)), (i32 0x00FF0000)),
+                  (and GPR:$rs1, (i32 0xFF0000FF))),
+              (and (srl GPR:$rs1, (i32 8)), (i32 0x0000FF00))),
+          (SHFLI GPR:$rs1, (i32 8))>;
+def : Pat<(or (or (and (shl GPR:$rs1, (i32 4)), (i32 0x0F000F00)),
+                  (and GPR:$rs1, (i32 0xF00FF00F))),
+              (and (srl GPR:$rs1, (i32 4)), (i32 0x00F000F0))),
+          (SHFLI GPR:$rs1, (i32 4))>;
+def : Pat<(or (or (and (shl GPR:$rs1, (i32 2)), (i32 0x30303030)),
+                  (and GPR:$rs1, (i32 0xC3C3C3C3))),
+              (and (srl GPR:$rs1, (i32 2)), (i32 0x0C0C0C0C))),
+          (SHFLI GPR:$rs1, (i32 2))>;
+def : Pat<(or (or (and (shl GPR:$rs1, (i32 1)), (i32 0x44444444)),
+                  (and GPR:$rs1, (i32 0x99999999))),
+              (and (srl GPR:$rs1, (i32 1)), (i32 0x22222222))),
+          (SHFLI GPR:$rs1, (i32 1))>;
+} // Predicates = [HasStdExtZbp, IsRV32]
+
+let Predicates = [HasStdExtZbp, IsRV64] in {
+def : Pat<(or (or (and (shl GPR:$rs1, (i64 16)), (i64 0x0000FFFF00000000)),
+                  (and GPR:$rs1, (i64 0xFFFF00000000FFFF))),
+              (and (srl GPR:$rs1, (i64 16)), (i64 0x00000000FFFF0000))),
+          (SHFLI GPR:$rs1, (i64 16))>;
+def : Pat<(or (or (and (shl GPR:$rs1, (i64 8)), (i64 0x00FF000000FF0000)),
+                  (and GPR:$rs1, (i64 0xFF0000FFFF0000FF))),
+              (and (srl GPR:$rs1, (i64 8)), (i64 0x0000FF000000FF00))),
+          (SHFLI GPR:$rs1, (i64 8))>;
+def : Pat<(or (or (and (shl GPR:$rs1, (i64 4)), (i64 0x0F000F000F000F00)),
+                  (and GPR:$rs1, (i64 0xF00FF00FF00FF00F))),
+              (and (srl GPR:$rs1, (i64 4)), (i64 0x00F000F000F000F0))),
+          (SHFLI GPR:$rs1, (i64 4))>;
+def : Pat<(or (or (and (shl GPR:$rs1, (i64 2)), (i64 0x3030303030303030)),
+                  (and GPR:$rs1, (i64 0xC3C3C3C3C3C3C3C3))),
+              (and (srl GPR:$rs1, (i64 2)), (i64 0x0C0C0C0C0C0C0C0C))),
+          (SHFLI GPR:$rs1, (i64 2))>;
+def : Pat<(or (or (and (shl GPR:$rs1, (i64 1)), (i64 0x4444444444444444)),
+                  (and GPR:$rs1, (i64 0x9999999999999999))),
+              (and (srl GPR:$rs1, (i64 1)), (i64 0x2222222222222222))),
+          (SHFLI GPR:$rs1, (i64 1))>;
+} // Predicates = [HasStdExtZbp, IsRV64]
+
 let Predicates = [HasStdExtZbb, IsRV64] in {
 def : Pat<(and (add GPR:$rs, simm12:$simm12), (i64 0xFFFFFFFF)),
           (ADDIWU GPR:$rs, simm12:$simm12)>;
 def : Pat<(SLLIUWPat GPR:$rs1, uimmlog2xlen:$shamt),
           (SLLIUW GPR:$rs1, uimmlog2xlen:$shamt)>;
 def : Pat<(and (add GPR:$rs1, GPR:$rs2), (i64 0xFFFFFFFF)),
           (ADDWU GPR:$rs1, GPR:$rs2)>;
 def : Pat<(and (sub GPR:$rs1, GPR:$rs2), (i64 0xFFFFFFFF)),
           (SUBWU GPR:$rs1, GPR:$rs2)>;
 def : Pat<(add GPR:$rs1, (and GPR:$rs2, (i64 0xFFFFFFFF))),
           (ADDUW GPR:$rs1, GPR:$rs2)>;
 def : Pat<(sub GPR:$rs1, (and GPR:$rs2, (i64 0xFFFFFFFF))),
           (SUBUW GPR:$rs1, GPR:$rs2)>;
 def : Pat<(xor (riscv_sllw (xor GPR:$rs1, -1), GPR:$rs2), -1),
           (SLOW GPR:$rs1, GPR:$rs2)>;
 def : Pat<(xor (riscv_srlw (xor GPR:$rs1, -1), GPR:$rs2), -1),
           (SROW GPR:$rs1, GPR:$rs2)>;
 def : Pat<(SLOIWPat GPR:$rs1, uimmlog2xlen:$shamt),
           (SLOIW GPR:$rs1, uimmlog2xlen:$shamt)>;
 def : Pat<(SROIWPat GPR:$rs1, uimmlog2xlen:$shamt),
           (SROIW GPR:$rs1, uimmlog2xlen:$shamt)>;
+} // Predicates = [HasStdExtZbb, IsRV64]
+
+let Predicates = [HasStdExtZbp, IsRV64] in {
+def : Pat<(sext_inreg (or (or (and (srl GPR:$rs1, (i64 1)), (i64 0x55555555)),
+                              GPR:$rs1),
+                          (and (shl GPR:$rs1, (i64 1)), (i64 0xAAAAAAAA))),
+                      i32),
+          (GORCIW GPR:$rs1, (i64 1))>;
+def : Pat<(sext_inreg (or (or (and (srl GPR:$rs1, (i64 2)), (i64 0x33333333)),
+                              GPR:$rs1),
+                          (and (shl GPR:$rs1, (i64 2)), (i64 0xCCCCCCCC))),
+                      i32),
+          (GORCIW GPR:$rs1, (i64 2))>;
+def : Pat<(sext_inreg (or (or (and (srl GPR:$rs1, (i64 4)), (i64 0x0F0F0F0F)),
+                              GPR:$rs1),
+                          (and (shl GPR:$rs1, (i64 4)), (i64 0xF0F0F0F0))),
+                      i32),
+          (GORCIW GPR:$rs1, (i64 4))>;
+def : Pat<(sext_inreg (or (or (and (srl GPR:$rs1, (i64 8)), (i64 0x00FF00FF)),
+                              GPR:$rs1),
+                          (and (shl GPR:$rs1, (i64 8)), (i64 0xFF00FF00))),
+                      i32),
+          (GORCIW GPR:$rs1, (i64 8))>;
+def : Pat<(sext_inreg (or (or (and (srl GPR:$rs1, (i64 16)), (i64 0x0000FFFF)),
+                              GPR:$rs1),
+                          (and (shl GPR:$rs1, (i64 16)), (i64 0xFFFF0000))),
+                      i32),
+          (GORCIW GPR:$rs1, (i64 16))>;
+def : Pat<(sext_inreg (or (or (srl (and GPR:$rs1, (i64 0xFFFF0000)), (i64 16)),
+                              GPR:$rs1),
+                          (shl GPR:$rs1, (i64 16))), i32),
+          (GORCIW GPR:$rs1, (i64 16))>;
+
+def : Pat<(sext_inreg (or (and (shl GPR:$rs1, (i64 1)), (i64 0xAAAAAAAA)),
+                          (and (srl GPR:$rs1, (i64 1)), (i64 0x55555555))),
+                      i32),
+          (GREVIW GPR:$rs1, (i64 1))>;
+def : Pat<(sext_inreg (or (and (shl GPR:$rs1, (i64 2)), (i64 0xCCCCCCCC)),
+                          (and (srl GPR:$rs1, (i64 2)), (i64 0x33333333))),
+                      i32),
+          (GREVIW GPR:$rs1, (i64 2))>;
+def : Pat<(sext_inreg (or (and (shl GPR:$rs1, (i64 4)), (i64 0xF0F0F0F0)),
+                          (and (srl GPR:$rs1, (i64 4)), (i64 0x0F0F0F0F))),
+                      i32),
+          (GREVIW GPR:$rs1, (i64 4))>;
+def : Pat<(sext_inreg (or (and (shl GPR:$rs1, (i64 8)), (i64 0xFF00FF00)),
+                          (and (srl GPR:$rs1, (i64 8)), (i64 0x00FF00FF))),
+                      i32),
+          (GREVIW GPR:$rs1, (i64 8))>;
+def : Pat<(sext_inreg (or (shl GPR:$rs1, (i64 16)),
+                          (srl (and GPR:$rs1, 0xFFFF0000), (i64 16))), i32),
+          (GREVIW GPR:$rs1, (i64 16))>;
+def : Pat<(sra (bswap GPR:$rs1), (i64 32)), (GREVIW GPR:$rs1, (i64 24))>;
+def : Pat<(sra (bitreverse GPR:$rs1), (i64 32)), (GREVIW GPR:$rs1, (i64 31))>;
+} // Predicates = [HasStdExtZbp, IsRV64]
+
+let Predicates = [HasStdExtZbb, IsRV64] in {
 def : Pat<(add (ctlz (and GPR:$rs1, (i64 0xFFFFFFFF))), (i64 -32)),
           (CLZW GPR:$rs1)>;
 // We don't pattern-match CTZW here as it has the same pattern and result as
 // RV64 CTZ
 def : Pat<(ctpop (and GPR:$rs1, (i64 0xFFFFFFFF))), (PCNTW GPR:$rs1)>;
 } // Predicates = [HasStdExtZbb, IsRV64]

llvm/test/CodeGen/RISCV/rv32Zbp.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefix=RV32I
+; RUN: llc -mtriple=riscv32 -mattr=+experimental-b -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefix=RV32IB
+; RUN: llc -mtriple=riscv32 -mattr=+experimental-zbp -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefix=RV32IBP
+
+define i32 @gorc1_i32(i32 %a) nounwind {
+; RV32I-LABEL: gorc1_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a1, a0, 1
+; RV32I-NEXT:    lui a2, 699051
+; RV32I-NEXT:    addi a2, a2, -1366
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a2, a0, 1
+; RV32I-NEXT:    lui a3, 349525
+; RV32I-NEXT:    addi a3, a3, 1365
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: gorc1_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    orc.p a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: gorc1_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    orc.p a0, a0
+; RV32IBP-NEXT:    ret
+  %and = shl i32 %a, 1
+  %shl = and i32 %and, -1431655766
+  %and1 = lshr i32 %a, 1
+  %shr = and i32 %and1, 1431655765
+  %or = or i32 %shr, %a
+  %or2 = or i32 %or, %shl
+  ret i32 %or2
+}
+
+define i64 @gorc1_i64(i64 %a) nounwind {
+; RV32I-LABEL: gorc1_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a2, a0, 1
+; RV32I-NEXT:    slli a3, a1, 1
+; RV32I-NEXT:    lui a4, 699051
+; RV32I-NEXT:    addi a4, a4, -1366
+; RV32I-NEXT:    and a6, a3, a4
+; RV32I-NEXT:    and a2, a2, a4
+; RV32I-NEXT:    srli a4, a1, 1
+; RV32I-NEXT:    srli a5, a0, 1
+; RV32I-NEXT:    lui a3, 349525
+; RV32I-NEXT:    addi a3, a3, 1365
+; RV32I-NEXT:    and a5, a5, a3
+; RV32I-NEXT:    and a3, a4, a3
+; RV32I-NEXT:    or a1, a3, a1
+; RV32I-NEXT:    or a0, a5, a0
+; RV32I-NEXT:    or a0, a0, a2
+; RV32I-NEXT:    or a1, a1, a6
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: gorc1_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    orc.p a0, a0
+; RV32IB-NEXT:    orc.p a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: gorc1_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    orc.p a0, a0
+; RV32IBP-NEXT:    orc.p a1, a1
+; RV32IBP-NEXT:    ret
+  %and = shl i64 %a, 1
+  %shl = and i64 %and, -6148914691236517206
+  %and1 = lshr i64 %a, 1
+  %shr = and i64 %and1, 6148914691236517205
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define i32 @gorc2_i32(i32 %a) nounwind {
+; RV32I-LABEL: gorc2_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a1, a0, 2
+; RV32I-NEXT:    lui a2, 838861
+; RV32I-NEXT:    addi a2, a2, -820
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a2, a0, 2
+; RV32I-NEXT:    lui a3, 209715
+; RV32I-NEXT:    addi a3, a3, 819
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: gorc2_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    orc2.n a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: gorc2_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    orc2.n a0, a0
+; RV32IBP-NEXT:    ret
+  %and = shl i32 %a, 2
+  %shl = and i32 %and, -858993460
+  %and1 = lshr i32 %a, 2
+  %shr = and i32 %and1, 858993459
+  %or = or i32 %shr, %a
+  %or2 = or i32 %or, %shl
+  ret i32 %or2
+}
+
+define i64 @gorc2_i64(i64 %a) nounwind {
+; RV32I-LABEL: gorc2_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a2, a0, 2
+; RV32I-NEXT:    slli a3, a1, 2
+; RV32I-NEXT:    lui a4, 838861
+; RV32I-NEXT:    addi a4, a4, -820
+; RV32I-NEXT:    and a6, a3, a4
+; RV32I-NEXT:    and a2, a2, a4
+; RV32I-NEXT:    srli a4, a1, 2
+; RV32I-NEXT:    srli a5, a0, 2
+; RV32I-NEXT:    lui a3, 209715
+; RV32I-NEXT:    addi a3, a3, 819
+; RV32I-NEXT:    and a5, a5, a3
+; RV32I-NEXT:    and a3, a4, a3
+; RV32I-NEXT:    or a1, a3, a1
+; RV32I-NEXT:    or a0, a5, a0
+; RV32I-NEXT:    or a0, a0, a2
+; RV32I-NEXT:    or a1, a1, a6
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: gorc2_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    orc2.n a0, a0
+; RV32IB-NEXT:    orc2.n a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: gorc2_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    orc2.n a0, a0
+; RV32IBP-NEXT:    orc2.n a1, a1
+; RV32IBP-NEXT:    ret
+  %and = shl i64 %a, 2
+  %shl = and i64 %and, -3689348814741910324
+  %and1 = lshr i64 %a, 2
+  %shr = and i64 %and1, 3689348814741910323
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define i32 @gorc4_i32(i32 %a) nounwind {
+; RV32I-LABEL: gorc4_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a1, a0, 4
+; RV32I-NEXT:    lui a2, 986895
+; RV32I-NEXT:    addi a2, a2, 240
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a2, a0, 4
+; RV32I-NEXT:    lui a3, 61681
+; RV32I-NEXT:    addi a3, a3, -241
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: gorc4_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    orc4.b a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: gorc4_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    orc4.b a0, a0
+; RV32IBP-NEXT:    ret
+  %and = shl i32 %a, 4
+  %shl = and i32 %and, -252645136
+  %and1 = lshr i32 %a, 4
+  %shr = and i32 %and1, 252645135
+  %or = or i32 %shr, %a
+  %or2 = or i32 %or, %shl
+  ret i32 %or2
+}
+
+define i64 @gorc4_i64(i64 %a) nounwind {
+; RV32I-LABEL: gorc4_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a2, a0, 4
+; RV32I-NEXT:    slli a3, a1, 4
+; RV32I-NEXT:    lui a4, 986895
+; RV32I-NEXT:    addi a4, a4, 240
+; RV32I-NEXT:    and a6, a3, a4
+; RV32I-NEXT:    and a2, a2, a4
+; RV32I-NEXT:    srli a4, a1, 4
+; RV32I-NEXT:    srli a5, a0, 4
+; RV32I-NEXT:    lui a3, 61681
+; RV32I-NEXT:    addi a3, a3, -241
+; RV32I-NEXT:    and a5, a5, a3
+; RV32I-NEXT:    and a3, a4, a3
+; RV32I-NEXT:    or a1, a3, a1
+; RV32I-NEXT:    or a0, a5, a0
+; RV32I-NEXT:    or a0, a0, a2
+; RV32I-NEXT:    or a1, a1, a6
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: gorc4_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    orc4.b a0, a0
+; RV32IB-NEXT:    orc4.b a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: gorc4_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    orc4.b a0, a0
+; RV32IBP-NEXT:    orc4.b a1, a1
+; RV32IBP-NEXT:    ret
+  %and = shl i64 %a, 4
+  %shl = and i64 %and, -1085102592571150096
+  %and1 = lshr i64 %a, 4
+  %shr = and i64 %and1, 1085102592571150095
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define i32 @gorc8_i32(i32 %a) nounwind {
+; RV32I-LABEL: gorc8_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a1, a0, 8
+; RV32I-NEXT:    lui a2, 1044496
+; RV32I-NEXT:    addi a2, a2, -256
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a2, a0, 8
+; RV32I-NEXT:    lui a3, 4080
+; RV32I-NEXT:    addi a3, a3, 255
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: gorc8_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    orc8.h a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: gorc8_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    orc8.h a0, a0
+; RV32IBP-NEXT:    ret
+  %and = shl i32 %a, 8
+  %shl = and i32 %and, -16711936
+  %and1 = lshr i32 %a, 8
+  %shr = and i32 %and1, 16711935
+  %or = or i32 %shr, %a
+  %or2 = or i32 %or, %shl
+  ret i32 %or2
+}
+
+define i64 @gorc8_i64(i64 %a) nounwind {
+; RV32I-LABEL: gorc8_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a2, a0, 8
+; RV32I-NEXT:    slli a3, a1, 8
+; RV32I-NEXT:    lui a4, 1044496
+; RV32I-NEXT:    addi a4, a4, -256
+; RV32I-NEXT:    and a6, a3, a4
+; RV32I-NEXT:    and a2, a2, a4
+; RV32I-NEXT:    srli a4, a1, 8
+; RV32I-NEXT:    srli a5, a0, 8
+; RV32I-NEXT:    lui a3, 4080
+; RV32I-NEXT:    addi a3, a3, 255
+; RV32I-NEXT:    and a5, a5, a3
+; RV32I-NEXT:    and a3, a4, a3
+; RV32I-NEXT:    or a1, a3, a1
+; RV32I-NEXT:    or a0, a5, a0
+; RV32I-NEXT:    or a0, a0, a2
+; RV32I-NEXT:    or a1, a1, a6
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: gorc8_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    orc8.h a0, a0
+; RV32IB-NEXT:    orc8.h a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: gorc8_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    orc8.h a0, a0
+; RV32IBP-NEXT:    orc8.h a1, a1
+; RV32IBP-NEXT:    ret
+  %and = shl i64 %a, 8
+  %shl = and i64 %and, -71777214294589696
+  %and1 = lshr i64 %a, 8
+  %shr = and i64 %and1, 71777214294589695
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define i32 @gorc16_i32(i32 %a) nounwind {
+; RV32I-LABEL: gorc16_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a1, a0, 16
+; RV32I-NEXT:    srli a2, a0, 16
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: gorc16_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    orc16 a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: gorc16_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    orc16 a0, a0
+; RV32IBP-NEXT:    ret
+  %shl = shl i32 %a, 16
+  %shr = lshr i32 %a, 16
+  %or = or i32 %shr, %a
+  %or2 = or i32 %or, %shl
+  ret i32 %or2
+}
+
+define i64 @gorc16_i64(i64 %a) nounwind {
+; RV32I-LABEL: gorc16_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a2, a1, 16
+; RV32I-NEXT:    slli a3, a0, 16
+; RV32I-NEXT:    srli a4, a0, 16
+; RV32I-NEXT:    srli a5, a1, 16
+; RV32I-NEXT:    or a1, a5, a1
+; RV32I-NEXT:    or a0, a4, a0
+; RV32I-NEXT:    or a0, a0, a3
+; RV32I-NEXT:    or a1, a1, a2
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: gorc16_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    orc16 a0, a0
+; RV32IB-NEXT:    orc16 a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: gorc16_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    orc16 a0, a0
+; RV32IBP-NEXT:    orc16 a1, a1
+; RV32IBP-NEXT:    ret
+  %and = shl i64 %a, 16
+  %shl = and i64 %and, -281470681808896
+  %and1 = lshr i64 %a, 16
+  %shr = and i64 %and1, 281470681808895
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define i32 @grev1_i32(i32 %a) nounwind {
+; RV32I-LABEL: grev1_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a1, a0, 1
+; RV32I-NEXT:    lui a2, 699051
+; RV32I-NEXT:    addi a2, a2, -1366
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a0, a0, 1
+; RV32I-NEXT:    lui a2, 349525
+; RV32I-NEXT:    addi a2, a2, 1365
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    or a0, a1, a0
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: grev1_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev.p a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: grev1_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev.p a0, a0
+; RV32IBP-NEXT:    ret
+  %and = shl i32 %a, 1
+  %shl = and i32 %and, -1431655766
+  %and1 = lshr i32 %a, 1
+  %shr = and i32 %and1, 1431655765
+  %or = or i32 %shl, %shr
+  ret i32 %or
+}
+
+define i64 @grev1_i64(i64 %a) nounwind {
+; RV32I-LABEL: grev1_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a2, a0, 1
+; RV32I-NEXT:    slli a3, a1, 1
+; RV32I-NEXT:    lui a4, 699051
+; RV32I-NEXT:    addi a4, a4, -1366
+; RV32I-NEXT:    and a3, a3, a4
+; RV32I-NEXT:    and a2, a2, a4
+; RV32I-NEXT:    srli a0, a0, 1
+; RV32I-NEXT:    srli a1, a1, 1
+; RV32I-NEXT:    lui a4, 349525
+; RV32I-NEXT:    addi a4, a4, 1365
+; RV32I-NEXT:    and a1, a1, a4
+; RV32I-NEXT:    and a0, a0, a4
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    or a1, a3, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: grev1_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev.p a0, a0
+; RV32IB-NEXT:    rev.p a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: grev1_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev.p a0, a0
+; RV32IBP-NEXT:    rev.p a1, a1
+; RV32IBP-NEXT:    ret
+  %and = shl i64 %a, 1
+  %shl = and i64 %and, -6148914691236517206
+  %and1 = lshr i64 %a, 1
+  %shr = and i64 %and1, 6148914691236517205
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+define i32 @grev2_i32(i32 %a) nounwind {
+; RV32I-LABEL: grev2_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a1, a0, 2
+; RV32I-NEXT:    lui a2, 838861
+; RV32I-NEXT:    addi a2, a2, -820
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a0, a0, 2
+; RV32I-NEXT:    lui a2, 209715
+; RV32I-NEXT:    addi a2, a2, 819
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    or a0, a1, a0
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: grev2_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev2.n a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: grev2_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev2.n a0, a0
+; RV32IBP-NEXT:    ret
+  %and = shl i32 %a, 2
+  %shl = and i32 %and, -858993460
+  %and1 = lshr i32 %a, 2
+  %shr = and i32 %and1, 858993459
+  %or = or i32 %shl, %shr
+  ret i32 %or
+}
+
+define i64 @grev2_i64(i64 %a) nounwind {
+; RV32I-LABEL: grev2_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a2, a0, 2
+; RV32I-NEXT:    slli a3, a1, 2
+; RV32I-NEXT:    lui a4, 838861
+; RV32I-NEXT:    addi a4, a4, -820
+; RV32I-NEXT:    and a3, a3, a4
+; RV32I-NEXT:    and a2, a2, a4
+; RV32I-NEXT:    srli a0, a0, 2
+; RV32I-NEXT:    srli a1, a1, 2
+; RV32I-NEXT:    lui a4, 209715
+; RV32I-NEXT:    addi a4, a4, 819
+; RV32I-NEXT:    and a1, a1, a4
+; RV32I-NEXT:    and a0, a0, a4
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    or a1, a3, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: grev2_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev2.n a0, a0
+; RV32IB-NEXT:    rev2.n a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: grev2_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev2.n a0, a0
+; RV32IBP-NEXT:    rev2.n a1, a1
+; RV32IBP-NEXT:    ret
+  %and = shl i64 %a, 2
+  %shl = and i64 %and, -3689348814741910324
+  %and1 = lshr i64 %a, 2
+  %shr = and i64 %and1, 3689348814741910323
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+define i32 @grev4_i32(i32 %a) nounwind {
+; RV32I-LABEL: grev4_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a1, a0, 4
+; RV32I-NEXT:    lui a2, 986895
+; RV32I-NEXT:    addi a2, a2, 240
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a0, a0, 4
+; RV32I-NEXT:    lui a2, 61681
+; RV32I-NEXT:    addi a2, a2, -241
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    or a0, a1, a0
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: grev4_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev4.b a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: grev4_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev4.b a0, a0
+; RV32IBP-NEXT:    ret
+  %and = shl i32 %a, 4
+  %shl = and i32 %and, -252645136
+  %and1 = lshr i32 %a, 4
+  %shr = and i32 %and1, 252645135
+  %or = or i32 %shl, %shr
+  ret i32 %or
+}
+
+define i64 @grev4_i64(i64 %a) nounwind {
+; RV32I-LABEL: grev4_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a2, a0, 4
+; RV32I-NEXT:    slli a3, a1, 4
+; RV32I-NEXT:    lui a4, 986895
+; RV32I-NEXT:    addi a4, a4, 240
+; RV32I-NEXT:    and a3, a3, a4
+; RV32I-NEXT:    and a2, a2, a4
+; RV32I-NEXT:    srli a0, a0, 4
+; RV32I-NEXT:    srli a1, a1, 4
+; RV32I-NEXT:    lui a4, 61681
+; RV32I-NEXT:    addi a4, a4, -241
+; RV32I-NEXT:    and a1, a1, a4
+; RV32I-NEXT:    and a0, a0, a4
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    or a1, a3, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: grev4_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev4.b a0, a0
+; RV32IB-NEXT:    rev4.b a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: grev4_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev4.b a0, a0
+; RV32IBP-NEXT:    rev4.b a1, a1
+; RV32IBP-NEXT:    ret
+  %and = shl i64 %a, 4
+  %shl = and i64 %and, -1085102592571150096
+  %and1 = lshr i64 %a, 4
+  %shr = and i64 %and1, 1085102592571150095
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+define i32 @grev8_i32(i32 %a) nounwind {
+; RV32I-LABEL: grev8_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a1, a0, 8
+; RV32I-NEXT:    lui a2, 1044496
+; RV32I-NEXT:    addi a2, a2, -256
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a0, a0, 8
+; RV32I-NEXT:    lui a2, 4080
+; RV32I-NEXT:    addi a2, a2, 255
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    or a0, a1, a0
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: grev8_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev8.h a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: grev8_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev8.h a0, a0
+; RV32IBP-NEXT:    ret
+  %and = shl i32 %a, 8
+  %shl = and i32 %and, -16711936
+  %and1 = lshr i32 %a, 8
+  %shr = and i32 %and1, 16711935
+  %or = or i32 %shl, %shr
+  ret i32 %or
+}
+
+define i64 @grev8_i64(i64 %a) nounwind {
+; RV32I-LABEL: grev8_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a2, a0, 8
+; RV32I-NEXT:    slli a3, a1, 8
+; RV32I-NEXT:    lui a4, 1044496
+; RV32I-NEXT:    addi a4, a4, -256
+; RV32I-NEXT:    and a3, a3, a4
+; RV32I-NEXT:    and a2, a2, a4
+; RV32I-NEXT:    srli a0, a0, 8
+; RV32I-NEXT:    srli a1, a1, 8
+; RV32I-NEXT:    lui a4, 4080
+; RV32I-NEXT:    addi a4, a4, 255
+; RV32I-NEXT:    and a1, a1, a4
+; RV32I-NEXT:    and a0, a0, a4
+; RV32I-NEXT:    or a0, a2, a0
+; RV32I-NEXT:    or a1, a3, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: grev8_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev8.h a0, a0
+; RV32IB-NEXT:    rev8.h a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: grev8_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev8.h a0, a0
+; RV32IBP-NEXT:    rev8.h a1, a1
+; RV32IBP-NEXT:    ret
+  %and = shl i64 %a, 8
+  %shl = and i64 %and, -71777214294589696
+  %and1 = lshr i64 %a, 8
+  %shr = and i64 %and1, 71777214294589695
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+define i32 @grev16_i32(i32 %a) nounwind {
+; RV32I-LABEL: grev16_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a1, a0, 16
+; RV32I-NEXT:    srli a0, a0, 16
+; RV32I-NEXT:    or a0, a1, a0
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: grev16_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rori a0, a0, 16
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: grev16_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rori a0, a0, 16
+; RV32IBP-NEXT:    ret
+  %shl = shl i32 %a, 16
+  %shr = lshr i32 %a, 16
+  %or = or i32 %shl, %shr
+  ret i32 %or
+}
+
+define i64 @grev16_i64(i64 %a) nounwind {
+; RV32I-LABEL: grev16_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    slli a2, a1, 16
+; RV32I-NEXT:    srli a3, a0, 16
+; RV32I-NEXT:    slli a0, a0, 16
+; RV32I-NEXT:    or a0, a0, a3
+; RV32I-NEXT:    srli a1, a1, 16
+; RV32I-NEXT:    or a1, a2, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: grev16_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rori a0, a0, 16
+; RV32IB-NEXT:    rori a1, a1, 16
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: grev16_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rori a0, a0, 16
+; RV32IBP-NEXT:    rori a1, a1, 16
+; RV32IBP-NEXT:    ret
+  %and = shl i64 %a, 16
+  %shl = and i64 %and, -281470681808896
+  %and1 = lshr i64 %a, 16
+  %shr = and i64 %and1, 281470681808895
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+declare i32 @llvm.bswap.i32(i32)
+
+define i32 @bswap_i32(i32 %a) nounwind {
+; RV32I-LABEL: bswap_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    lui a2, 16
+; RV32I-NEXT:    addi a2, a2, -256
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a2, a0, 24
+; RV32I-NEXT:    or a1, a1, a2
+; RV32I-NEXT:    slli a2, a0, 8
+; RV32I-NEXT:    lui a3, 4080
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    slli a0, a0, 24
+; RV32I-NEXT:    or a0, a0, a2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: bswap_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev8 a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: bswap_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev8 a0, a0
+; RV32IBP-NEXT:    ret
+  %1 = tail call i32 @llvm.bswap.i32(i32 %a)
+  ret i32 %1
+}
+
+declare i64 @llvm.bswap.i64(i64)
+
+define i64 @bswap_i64(i64 %a) {
+; RV32I-LABEL: bswap_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    srli a2, a1, 8
+; RV32I-NEXT:    lui a3, 16
+; RV32I-NEXT:    addi a3, a3, -256
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    srli a4, a1, 24
+; RV32I-NEXT:    or a2, a2, a4
+; RV32I-NEXT:    slli a4, a1, 8
+; RV32I-NEXT:    lui a5, 4080
+; RV32I-NEXT:    and a4, a4, a5
+; RV32I-NEXT:    slli a1, a1, 24
+; RV32I-NEXT:    or a1, a1, a4
+; RV32I-NEXT:    or a2, a1, a2
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    and a1, a1, a3
+; RV32I-NEXT:    srli a3, a0, 24
+; RV32I-NEXT:    or a1, a1, a3
+; RV32I-NEXT:    slli a3, a0, 8
+; RV32I-NEXT:    and a3, a3, a5
+; RV32I-NEXT:    slli a0, a0, 24
+; RV32I-NEXT:    or a0, a0, a3
+; RV32I-NEXT:    or a1, a0, a1
+; RV32I-NEXT:    mv a0, a2
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: bswap_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev8 a2, a1
+; RV32IB-NEXT:    rev8 a1, a0
+; RV32IB-NEXT:    mv a0, a2
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: bswap_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev8 a2, a1
+; RV32IBP-NEXT:    rev8 a1, a0
+; RV32IBP-NEXT:    mv a0, a2
+; RV32IBP-NEXT:    ret
+  %1 = call i64 @llvm.bswap.i64(i64 %a)
+  ret i64 %1
+}
+
+declare i32 @llvm.bitreverse.i32(i32)
+
+define i32 @bitreverse_i32(i32 %a) nounwind {
+; RV32I-LABEL: bitreverse_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    lui a2, 16
+; RV32I-NEXT:    addi a2, a2, -256
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a2, a0, 24
+; RV32I-NEXT:    or a1, a1, a2
+; RV32I-NEXT:    slli a2, a0, 8
+; RV32I-NEXT:    lui a3, 4080
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    slli a0, a0, 24
+; RV32I-NEXT:    or a0, a0, a2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    lui a1, 61681
+; RV32I-NEXT:    addi a1, a1, -241
+; RV32I-NEXT:    and a1, a0, a1
+; RV32I-NEXT:    slli a1, a1, 4
+; RV32I-NEXT:    lui a2, 986895
+; RV32I-NEXT:    addi a2, a2, 240
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    srli a0, a0, 4
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    lui a1, 209715
+; RV32I-NEXT:    addi a1, a1, 819
+; RV32I-NEXT:    and a1, a0, a1
+; RV32I-NEXT:    slli a1, a1, 2
+; RV32I-NEXT:    lui a2, 838861
+; RV32I-NEXT:    addi a2, a2, -820
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    srli a0, a0, 2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    lui a1, 349525
+; RV32I-NEXT:    addi a1, a1, 1365
+; RV32I-NEXT:    and a1, a0, a1
+; RV32I-NEXT:    slli a1, a1, 1
+; RV32I-NEXT:    lui a2, 699051
+; RV32I-NEXT:    addi a2, a2, -1366
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    srli a0, a0, 1
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: bitreverse_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: bitreverse_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev a0, a0
+; RV32IBP-NEXT:    ret
+  %1 = tail call i32 @llvm.bitreverse.i32(i32 %a)
+  ret i32 %1
+}
+
+declare i64 @llvm.bitreverse.i64(i64)
+
+define i64 @bitreverse_i64(i64 %a) nounwind {
+; RV32I-LABEL: bitreverse_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    srli a2, a1, 8
+; RV32I-NEXT:    lui a3, 16
+; RV32I-NEXT:    addi t0, a3, -256
+; RV32I-NEXT:    and a2, a2, t0
+; RV32I-NEXT:    srli a4, a1, 24
+; RV32I-NEXT:    or a2, a2, a4
+; RV32I-NEXT:    slli a4, a1, 8
+; RV32I-NEXT:    lui a6, 4080
+; RV32I-NEXT:    and a4, a4, a6
+; RV32I-NEXT:    slli a1, a1, 24
+; RV32I-NEXT:    or a1, a1, a4
+; RV32I-NEXT:    or a1, a1, a2
+; RV32I-NEXT:    lui a2, 61681
+; RV32I-NEXT:    addi t1, a2, -241
+; RV32I-NEXT:    and a2, a1, t1
+; RV32I-NEXT:    slli a2, a2, 4
+; RV32I-NEXT:    lui a5, 986895
+; RV32I-NEXT:    addi t2, a5, 240
+; RV32I-NEXT:    and a1, a1, t2
+; RV32I-NEXT:    srli a1, a1, 4
+; RV32I-NEXT:    or a1, a1, a2
+; RV32I-NEXT:    lui a2, 209715
+; RV32I-NEXT:    addi t3, a2, 819
+; RV32I-NEXT:    and a3, a1, t3
+; RV32I-NEXT:    slli a3, a3, 2
+; RV32I-NEXT:    lui a4, 838861
+; RV32I-NEXT:    addi a4, a4, -820
+; RV32I-NEXT:    and a1, a1, a4
+; RV32I-NEXT:    srli a1, a1, 2
+; RV32I-NEXT:    or a1, a1, a3
+; RV32I-NEXT:    lui a3, 349525
+; RV32I-NEXT:    addi a3, a3, 1365
+; RV32I-NEXT:    and a5, a1, a3
+; RV32I-NEXT:    slli a5, a5, 1
+; RV32I-NEXT:    lui a2, 699051
+; RV32I-NEXT:    addi a2, a2, -1366
+; RV32I-NEXT:    and a1, a1, a2
+; RV32I-NEXT:    srli a1, a1, 1
+; RV32I-NEXT:    or a7, a1, a5
+; RV32I-NEXT:    srli a1, a0, 8
+; RV32I-NEXT:    and a1, a1, t0
+; RV32I-NEXT:    srli a5, a0, 24
+; RV32I-NEXT:    or a1, a1, a5
+; RV32I-NEXT:    slli a5, a0, 8
+; RV32I-NEXT:    and a5, a5, a6
+; RV32I-NEXT:    slli a0, a0, 24
+; RV32I-NEXT:    or a0, a0, a5
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    and a1, a0, t1
+; RV32I-NEXT:    slli a1, a1, 4
+; RV32I-NEXT:    and a0, a0, t2
+; RV32I-NEXT:    srli a0, a0, 4
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    and a1, a0, t3
+; RV32I-NEXT:    slli a1, a1, 2
+; RV32I-NEXT:    and a0, a0, a4
+; RV32I-NEXT:    srli a0, a0, 2
+; RV32I-NEXT:    or a0, a0, a1
+; RV32I-NEXT:    and a1, a0, a3
+; RV32I-NEXT:    slli a1, a1, 1
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    srli a0, a0, 1
+; RV32I-NEXT:    or a1, a0, a1
+; RV32I-NEXT:    mv a0, a7
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: bitreverse_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    rev a2, a1
+; RV32IB-NEXT:    rev a1, a0
+; RV32IB-NEXT:    mv a0, a2
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: bitreverse_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    rev a2, a1
+; RV32IBP-NEXT:    rev a1, a0
+; RV32IBP-NEXT:    mv a0, a2
+; RV32IBP-NEXT:    ret
+  %1 = call i64 @llvm.bitreverse.i64(i64 %a)
+  ret i64 %1
+}
+
+define i32 @shfl1_i32(i32 %a, i32 %b) nounwind {
+; RV32I-LABEL: shfl1_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    lui a1, 629146
+; RV32I-NEXT:    addi a1, a1, -1639
+; RV32I-NEXT:    and a1, a0, a1
+; RV32I-NEXT:    slli a2, a0, 1
+; RV32I-NEXT:    lui a3, 279620
+; RV32I-NEXT:    addi a3, a3, 1092
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    or a1, a2, a1
+; RV32I-NEXT:    srli a0, a0, 1
+; RV32I-NEXT:    lui a2, 139810
+; RV32I-NEXT:    addi a2, a2, 546
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    or a0, a1, a0
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: shfl1_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    zip.n a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: shfl1_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    zip.n a0, a0
+; RV32IBP-NEXT:    ret
+  %and = and i32 %a, -1717986919
+  %shl = shl i32 %a, 1
+  %and1 = and i32 %shl, 1145324612
+  %or = or i32 %and1, %and
+  %shr = lshr i32 %a, 1
+  %and2 = and i32 %shr, 572662306
+  %or3 = or i32 %or, %and2
+  ret i32 %or3
+}
+
+define i64 @shfl1_i64(i64 %a, i64 %b) nounwind {
+; RV32I-LABEL: shfl1_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    lui a2, 629146
+; RV32I-NEXT:    addi a2, a2, -1639
+; RV32I-NEXT:    and a6, a0, a2
+; RV32I-NEXT:    and a2, a1, a2
+; RV32I-NEXT:    slli a4, a1, 1
+; RV32I-NEXT:    slli a5, a0, 1
+; RV32I-NEXT:    lui a3, 279620
+; RV32I-NEXT:    addi a3, a3, 1092
+; RV32I-NEXT:    and a5, a5, a3
+; RV32I-NEXT:    and a3, a4, a3
+; RV32I-NEXT:    or a2, a3, a2
+; RV32I-NEXT:    or a3, a5, a6
+; RV32I-NEXT:    srli a0, a0, 1
+; RV32I-NEXT:    srli a1, a1, 1
+; RV32I-NEXT:    lui a4, 139810
+; RV32I-NEXT:    addi a4, a4, 546
+; RV32I-NEXT:    and a1, a1, a4
+; RV32I-NEXT:    and a0, a0, a4
+; RV32I-NEXT:    or a0, a3, a0
+; RV32I-NEXT:    or a1, a2, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: shfl1_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    zip.n a0, a0
+; RV32IB-NEXT:    zip.n a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: shfl1_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    zip.n a0, a0
+; RV32IBP-NEXT:    zip.n a1, a1
+; RV32IBP-NEXT:    ret
+  %and = and i64 %a, -7378697629483820647
+  %shl = shl i64 %a, 1
+  %and1 = and i64 %shl, 4919131752989213764
+  %or = or i64 %and1, %and
+  %shr = lshr i64 %a, 1
+  %and2 = and i64 %shr, 2459565876494606882
+  %or3 = or i64 %or, %and2
+  ret i64 %or3
+}
+
+define i32 @shfl2_i32(i32 %a, i32 %b) nounwind {
+; RV32I-LABEL: shfl2_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    lui a1, 801852
+; RV32I-NEXT:    addi a1, a1, 963
+; RV32I-NEXT:    and a1, a0, a1
+; RV32I-NEXT:    slli a2, a0, 2
+; RV32I-NEXT:    lui a3, 197379
+; RV32I-NEXT:    addi a3, a3, 48
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    or a1, a2, a1
+; RV32I-NEXT:    srli a0, a0, 2
+; RV32I-NEXT:    lui a2, 49345
+; RV32I-NEXT:    addi a2, a2, -1012
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    or a0, a1, a0
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: shfl2_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    zip2.b a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: shfl2_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    zip2.b a0, a0
+; RV32IBP-NEXT:    ret
+  %and = and i32 %a, -1010580541
+  %shl = shl i32 %a, 2
+  %and1 = and i32 %shl, 808464432
+  %or = or i32 %and1, %and
+  %shr = lshr i32 %a, 2
+  %and2 = and i32 %shr, 202116108
+  %or3 = or i32 %or, %and2
+  ret i32 %or3
+}
+
+define i64 @shfl2_i64(i64 %a, i64 %b) nounwind {
+; RV32I-LABEL: shfl2_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    lui a2, 801852
+; RV32I-NEXT:    addi a2, a2, 963
+; RV32I-NEXT:    and a6, a0, a2
+; RV32I-NEXT:    and a2, a1, a2
+; RV32I-NEXT:    slli a4, a1, 2
+; RV32I-NEXT:    slli a5, a0, 2
+; RV32I-NEXT:    lui a3, 197379
+; RV32I-NEXT:    addi a3, a3, 48
+; RV32I-NEXT:    and a5, a5, a3
+; RV32I-NEXT:    and a3, a4, a3
+; RV32I-NEXT:    or a2, a3, a2
+; RV32I-NEXT:    or a3, a5, a6
+; RV32I-NEXT:    srli a0, a0, 2
+; RV32I-NEXT:    srli a1, a1, 2
+; RV32I-NEXT:    lui a4, 49345
+; RV32I-NEXT:    addi a4, a4, -1012
+; RV32I-NEXT:    and a1, a1, a4
+; RV32I-NEXT:    and a0, a0, a4
+; RV32I-NEXT:    or a0, a3, a0
+; RV32I-NEXT:    or a1, a2, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: shfl2_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    zip2.b a0, a0
+; RV32IB-NEXT:    zip2.b a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: shfl2_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    zip2.b a0, a0
+; RV32IBP-NEXT:    zip2.b a1, a1
+; RV32IBP-NEXT:    ret
+  %and = and i64 %a, -4340410370284600381
+  %shl = shl i64 %a, 2
+  %and1 = and i64 %shl, 3472328296227680304
+  %or = or i64 %and1, %and
+  %shr = lshr i64 %a, 2
+  %and2 = and i64 %shr, 868082074056920076
+  %or3 = or i64 %or, %and2
+  ret i64 %or3
+}
+
+define i32 @shfl4_i32(i32 %a, i32 %b) nounwind {
+; RV32I-LABEL: shfl4_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    lui a1, 983295
+; RV32I-NEXT:    addi a1, a1, 15
+; RV32I-NEXT:    and a1, a0, a1
+; RV32I-NEXT:    slli a2, a0, 4
+; RV32I-NEXT:    lui a3, 61441
+; RV32I-NEXT:    addi a3, a3, -256
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    or a1, a2, a1
+; RV32I-NEXT:    srli a0, a0, 4
+; RV32I-NEXT:    lui a2, 3840
+; RV32I-NEXT:    addi a2, a2, 240
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    or a0, a1, a0
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: shfl4_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    zip4.h a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: shfl4_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    zip4.h a0, a0
+; RV32IBP-NEXT:    ret
+  %and = and i32 %a, -267390961
+  %shl = shl i32 %a, 4
+  %and1 = and i32 %shl, 251662080
+  %or = or i32 %and1, %and
+  %shr = lshr i32 %a, 4
+  %and2 = and i32 %shr, 15728880
+  %or3 = or i32 %or, %and2
+  ret i32 %or3
+}
+
+define i64 @shfl4_i64(i64 %a, i64 %b) nounwind {
+; RV32I-LABEL: shfl4_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    lui a2, 983295
+; RV32I-NEXT:    addi a2, a2, 15
+; RV32I-NEXT:    and a6, a0, a2
+; RV32I-NEXT:    and a2, a1, a2
+; RV32I-NEXT:    slli a4, a1, 4
+; RV32I-NEXT:    slli a5, a0, 4
+; RV32I-NEXT:    lui a3, 61441
+; RV32I-NEXT:    addi a3, a3, -256
+; RV32I-NEXT:    and a5, a5, a3
+; RV32I-NEXT:    and a3, a4, a3
+; RV32I-NEXT:    or a2, a3, a2
+; RV32I-NEXT:    or a3, a5, a6
+; RV32I-NEXT:    srli a0, a0, 4
+; RV32I-NEXT:    srli a1, a1, 4
+; RV32I-NEXT:    lui a4, 3840
+; RV32I-NEXT:    addi a4, a4, 240
+; RV32I-NEXT:    and a1, a1, a4
+; RV32I-NEXT:    and a0, a0, a4
+; RV32I-NEXT:    or a0, a3, a0
+; RV32I-NEXT:    or a1, a2, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: shfl4_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    zip4.h a0, a0
+; RV32IB-NEXT:    zip4.h a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: shfl4_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    zip4.h a0, a0
+; RV32IBP-NEXT:    zip4.h a1, a1
+; RV32IBP-NEXT:    ret
+  %and = and i64 %a, -1148435428713435121
+  %shl = shl i64 %a, 4
+  %and1 = and i64 %shl, 1080880403494997760
+  %or = or i64 %and1, %and
+  %shr = lshr i64 %a, 4
+  %and2 = and i64 %shr, 67555025218437360
+  %or3 = or i64 %or, %and2
+  ret i64 %or3
+}
+
+define i32 @shfl8_i32(i32 %a, i32 %b) nounwind {
+; RV32I-LABEL: shfl8_i32:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    lui a1, 1044480
+; RV32I-NEXT:    addi a1, a1, 255
+; RV32I-NEXT:    and a1, a0, a1
+; RV32I-NEXT:    slli a2, a0, 8
+; RV32I-NEXT:    lui a3, 4080
+; RV32I-NEXT:    and a2, a2, a3
+; RV32I-NEXT:    or a1, a2, a1
+; RV32I-NEXT:    srli a0, a0, 8
+; RV32I-NEXT:    lui a2, 16
+; RV32I-NEXT:    addi a2, a2, -256
+; RV32I-NEXT:    and a0, a0, a2
+; RV32I-NEXT:    or a0, a1, a0
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: shfl8_i32:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    zip8 a0, a0
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: shfl8_i32:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    zip8 a0, a0
+; RV32IBP-NEXT:    ret
+  %and = and i32 %a, -16776961
+  %shl = shl i32 %a, 8
+  %and1 = and i32 %shl, 16711680
+  %or = or i32 %and1, %and
+  %shr = lshr i32 %a, 8
+  %and2 = and i32 %shr, 65280
+  %or3 = or i32 %or, %and2
+  ret i32 %or3
+}
+
+define i64 @shfl8_i64(i64 %a, i64 %b) nounwind {
+; RV32I-LABEL: shfl8_i64:
+; RV32I:       # %bb.0:
+; RV32I-NEXT:    lui a2, 1044480
+; RV32I-NEXT:    addi a2, a2, 255
+; RV32I-NEXT:    and a3, a0, a2
+; RV32I-NEXT:    and a2, a1, a2
+; RV32I-NEXT:    slli a4, a1, 8
+; RV32I-NEXT:    slli a5, a0, 8
+; RV32I-NEXT:    lui a6, 4080
+; RV32I-NEXT:    and a5, a5, a6
+; RV32I-NEXT:    and a4, a4, a6
+; RV32I-NEXT:    or a2, a4, a2
+; RV32I-NEXT:    or a3, a5, a3
+; RV32I-NEXT:    srli a0, a0, 8
+; RV32I-NEXT:    srli a1, a1, 8
+; RV32I-NEXT:    lui a4, 16
+; RV32I-NEXT:    addi a4, a4, -256
+; RV32I-NEXT:    and a1, a1, a4
+; RV32I-NEXT:    and a0, a0, a4
+; RV32I-NEXT:    or a0, a3, a0
+; RV32I-NEXT:    or a1, a2, a1
+; RV32I-NEXT:    ret
+;
+; RV32IB-LABEL: shfl8_i64:
+; RV32IB:       # %bb.0:
+; RV32IB-NEXT:    zip8 a0, a0
+; RV32IB-NEXT:    zip8 a1, a1
+; RV32IB-NEXT:    ret
+;
+; RV32IBP-LABEL: shfl8_i64:
+; RV32IBP:       # %bb.0:
+; RV32IBP-NEXT:    zip8 a0, a0
+; RV32IBP-NEXT:    zip8 a1, a1
+; RV32IBP-NEXT:    ret
+  %and = and i64 %a, -72056494543077121
+  %shl = shl i64 %a, 8
+  %and1 = and i64 %shl, 71776119077928960
+  %or = or i64 %and1, %and
+  %shr = lshr i64 %a, 8
+  %and2 = and i64 %shr, 280375465148160
+  %or3 = or i64 %or, %and2
+  ret i64 %or3
+}

llvm/test/CodeGen/RISCV/rv64Zbp.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv64 -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefix=RV64I
+; RUN: llc -mtriple=riscv64 -mattr=+experimental-b -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefix=RV64IB
+; RUN: llc -mtriple=riscv64 -mattr=+experimental-zbp -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefix=RV64IBP
+
+define signext i32 @gorc1_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: gorc1_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 1
+; RV64I-NEXT:    lui a2, 171
+; RV64I-NEXT:    addiw a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1366
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 1
+; RV64I-NEXT:    lui a3, 349525
+; RV64I-NEXT:    addiw a3, a3, 1365
+; RV64I-NEXT:    and a2, a2, a3
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    sext.w a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc1_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    gorciw a0, a0, 1
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc1_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    gorciw a0, a0, 1
+; RV64IBP-NEXT:    ret
+  %and = shl i32 %a, 1
+  %shl = and i32 %and, -1431655766
+  %and1 = lshr i32 %a, 1
+  %shr = and i32 %and1, 1431655765
+  %or = or i32 %shr, %a
+  %or2 = or i32 %or, %shl
+  ret i32 %or2
+}
+
+define i64 @gorc1_i64(i64 %a) nounwind {
+; RV64I-LABEL: gorc1_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 1
+; RV64I-NEXT:    lui a2, 1026731
+; RV64I-NEXT:    addiw a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1366
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 1
+; RV64I-NEXT:    lui a3, 21845
+; RV64I-NEXT:    addiw a3, a3, 1365
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, 1365
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, 1365
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, 1365
+; RV64I-NEXT:    and a2, a2, a3
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc1_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    orc.p a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc1_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    orc.p a0, a0
+; RV64IBP-NEXT:    ret
+  %and = shl i64 %a, 1
+  %shl = and i64 %and, -6148914691236517206
+  %and1 = lshr i64 %a, 1
+  %shr = and i64 %and1, 6148914691236517205
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define signext i32 @gorc2_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: gorc2_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 2
+; RV64I-NEXT:    lui a2, 205
+; RV64I-NEXT:    addiw a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -820
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 2
+; RV64I-NEXT:    lui a3, 209715
+; RV64I-NEXT:    addiw a3, a3, 819
+; RV64I-NEXT:    and a2, a2, a3
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    sext.w a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc2_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    gorciw a0, a0, 2
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc2_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    gorciw a0, a0, 2
+; RV64IBP-NEXT:    ret
+  %and = shl i32 %a, 2
+  %shl = and i32 %and, -858993460
+  %and1 = lshr i32 %a, 2
+  %shr = and i32 %and1, 858993459
+  %or = or i32 %shr, %a
+  %or2 = or i32 %or, %shl
+  ret i32 %or2
+}
+
+define i64 @gorc2_i64(i64 %a) nounwind {
+; RV64I-LABEL: gorc2_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 2
+; RV64I-NEXT:    lui a2, 1035469
+; RV64I-NEXT:    addiw a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -820
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 2
+; RV64I-NEXT:    lui a3, 13107
+; RV64I-NEXT:    addiw a3, a3, 819
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, 819
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, 819
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, 819
+; RV64I-NEXT:    and a2, a2, a3
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc2_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    orc2.n a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc2_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    orc2.n a0, a0
+; RV64IBP-NEXT:    ret
+  %and = shl i64 %a, 2
+  %shl = and i64 %and, -3689348814741910324
+  %and1 = lshr i64 %a, 2
+  %shr = and i64 %and1, 3689348814741910323
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define signext i32 @gorc4_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: gorc4_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 4
+; RV64I-NEXT:    lui a2, 241
+; RV64I-NEXT:    addiw a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 240
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 4
+; RV64I-NEXT:    lui a3, 61681
+; RV64I-NEXT:    addiw a3, a3, -241
+; RV64I-NEXT:    and a2, a2, a3
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    sext.w a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc4_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    gorciw a0, a0, 4
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc4_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    gorciw a0, a0, 4
+; RV64IBP-NEXT:    ret
+  %and = shl i32 %a, 4
+  %shl = and i32 %and, -252645136
+  %and1 = lshr i32 %a, 4
+  %shr = and i32 %and1, 252645135
+  %or = or i32 %shr, %a
+  %or2 = or i32 %or, %shl
+  ret i32 %or2
+}
+
+define i64 @gorc4_i64(i64 %a) nounwind {
+; RV64I-LABEL: gorc4_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 4
+; RV64I-NEXT:    lui a2, 1044721
+; RV64I-NEXT:    addiw a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 240
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 4
+; RV64I-NEXT:    lui a3, 3855
+; RV64I-NEXT:    addiw a3, a3, 241
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, -241
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, 241
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, -241
+; RV64I-NEXT:    and a2, a2, a3
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc4_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    orc4.b a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc4_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    orc4.b a0, a0
+; RV64IBP-NEXT:    ret
+  %and = shl i64 %a, 4
+  %shl = and i64 %and, -1085102592571150096
+  %and1 = lshr i64 %a, 4
+  %shr = and i64 %and1, 1085102592571150095
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define signext i32 @gorc8_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: gorc8_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 8
+; RV64I-NEXT:    lui a2, 16
+; RV64I-NEXT:    addiw a2, a2, -255
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, -256
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 8
+; RV64I-NEXT:    lui a3, 4080
+; RV64I-NEXT:    addiw a3, a3, 255
+; RV64I-NEXT:    and a2, a2, a3
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    sext.w a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc8_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    gorciw a0, a0, 8
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc8_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    gorciw a0, a0, 8
+; RV64IBP-NEXT:    ret
+  %and = shl i32 %a, 8
+  %shl = and i32 %and, -16711936
+  %and1 = lshr i32 %a, 8
+  %shr = and i32 %and1, 16711935
+  %or = or i32 %shr, %a
+  %or2 = or i32 %or, %shl
+  ret i32 %or2
+}
+
+define i64 @gorc8_i64(i64 %a) nounwind {
+; RV64I-LABEL: gorc8_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 8
+; RV64I-NEXT:    lui a2, 1044496
+; RV64I-NEXT:    addiw a2, a2, -255
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, -255
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, -256
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 8
+; RV64I-NEXT:    lui a3, 4080
+; RV64I-NEXT:    addiw a3, a3, 255
+; RV64I-NEXT:    slli a3, a3, 16
+; RV64I-NEXT:    addi a3, a3, 255
+; RV64I-NEXT:    slli a3, a3, 16
+; RV64I-NEXT:    addi a3, a3, 255
+; RV64I-NEXT:    and a2, a2, a3
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc8_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    orc8.h a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc8_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    orc8.h a0, a0
+; RV64IBP-NEXT:    ret
+  %and = shl i64 %a, 8
+  %shl = and i64 %and, -71777214294589696
+  %and1 = lshr i64 %a, 8
+  %shr = and i64 %and1, 71777214294589695
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define signext i32 @gorc16_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: gorc16_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 16
+; RV64I-NEXT:    srliw a2, a0, 16
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    sext.w a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc16_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    gorciw a0, a0, 16
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc16_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    gorciw a0, a0, 16
+; RV64IBP-NEXT:    ret
+  %shl = shl i32 %a, 16
+  %shr = lshr i32 %a, 16
+  %or = or i32 %shr, %a
+  %or2 = or i32 %or, %shl
+  ret i32 %or2
+}
+
+define i64 @gorc16_i64(i64 %a) nounwind {
+; RV64I-LABEL: gorc16_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 16
+; RV64I-NEXT:    lui a2, 1048560
+; RV64I-NEXT:    addiw a2, a2, 1
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, -1
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 16
+; RV64I-NEXT:    lui a3, 16
+; RV64I-NEXT:    addiw a3, a3, -1
+; RV64I-NEXT:    slli a3, a3, 16
+; RV64I-NEXT:    addi a3, a3, 1
+; RV64I-NEXT:    slli a3, a3, 16
+; RV64I-NEXT:    addi a3, a3, -1
+; RV64I-NEXT:    and a2, a2, a3
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc16_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    orc16.w a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc16_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    orc16.w a0, a0
+; RV64IBP-NEXT:    ret
+  %and = shl i64 %a, 16
+  %shl = and i64 %and, -281470681808896
+  %and1 = lshr i64 %a, 16
+  %shr = and i64 %and1, 281470681808895
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define i64 @gorc32(i64 %a) nounwind {
+; RV64I-LABEL: gorc32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 32
+; RV64I-NEXT:    srli a2, a0, 32
+; RV64I-NEXT:    or a0, a2, a0
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: gorc32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    orc32 a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: gorc32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    orc32 a0, a0
+; RV64IBP-NEXT:    ret
+  %shl = shl i64 %a, 32
+  %shr = lshr i64 %a, 32
+  %or = or i64 %shr, %a
+  %or2 = or i64 %or, %shl
+  ret i64 %or2
+}
+
+define signext i32 @grev1_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: grev1_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 1
+; RV64I-NEXT:    lui a2, 171
+; RV64I-NEXT:    addiw a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1366
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a0, a0, 1
+; RV64I-NEXT:    lui a2, 349525
+; RV64I-NEXT:    addiw a2, a2, 1365
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    sext.w a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev1_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    greviw a0, a0, 1
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev1_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    greviw a0, a0, 1
+; RV64IBP-NEXT:    ret
+  %and = shl i32 %a, 1
+  %shl = and i32 %and, -1431655766
+  %and1 = lshr i32 %a, 1
+  %shr = and i32 %and1, 1431655765
+  %or = or i32 %shl, %shr
+  ret i32 %or
+}
+
+define i64 @grev1_i64(i64 %a) nounwind {
+; RV64I-LABEL: grev1_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 1
+; RV64I-NEXT:    lui a2, 1026731
+; RV64I-NEXT:    addiw a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1366
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a0, a0, 1
+; RV64I-NEXT:    lui a2, 21845
+; RV64I-NEXT:    addiw a2, a2, 1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 1365
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev1_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    rev.p a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev1_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    rev.p a0, a0
+; RV64IBP-NEXT:    ret
+  %and = shl i64 %a, 1
+  %shl = and i64 %and, -6148914691236517206
+  %and1 = lshr i64 %a, 1
+  %shr = and i64 %and1, 6148914691236517205
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+define signext i32 @grev2_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: grev2_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 2
+; RV64I-NEXT:    lui a2, 205
+; RV64I-NEXT:    addiw a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -820
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a0, a0, 2
+; RV64I-NEXT:    lui a2, 209715
+; RV64I-NEXT:    addiw a2, a2, 819
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    sext.w a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev2_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    greviw a0, a0, 2
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev2_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    greviw a0, a0, 2
+; RV64IBP-NEXT:    ret
+  %and = shl i32 %a, 2
+  %shl = and i32 %and, -858993460
+  %and1 = lshr i32 %a, 2
+  %shr = and i32 %and1, 858993459
+  %or = or i32 %shl, %shr
+  ret i32 %or
+}
+
+define i64 @grev2_i64(i64 %a) nounwind {
+; RV64I-LABEL: grev2_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 2
+; RV64I-NEXT:    lui a2, 1035469
+; RV64I-NEXT:    addiw a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -820
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a0, a0, 2
+; RV64I-NEXT:    lui a2, 13107
+; RV64I-NEXT:    addiw a2, a2, 819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 819
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev2_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    rev2.n a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev2_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    rev2.n a0, a0
+; RV64IBP-NEXT:    ret
+  %and = shl i64 %a, 2
+  %shl = and i64 %and, -3689348814741910324
+  %and1 = lshr i64 %a, 2
+  %shr = and i64 %and1, 3689348814741910323
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+define signext i32 @grev4_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: grev4_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 4
+; RV64I-NEXT:    lui a2, 241
+; RV64I-NEXT:    addiw a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 240
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a0, a0, 4
+; RV64I-NEXT:    lui a2, 61681
+; RV64I-NEXT:    addiw a2, a2, -241
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    sext.w a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev4_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    greviw a0, a0, 4
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev4_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    greviw a0, a0, 4
+; RV64IBP-NEXT:    ret
+  %and = shl i32 %a, 4
+  %shl = and i32 %and, -252645136
+  %and1 = lshr i32 %a, 4
+  %shr = and i32 %and1, 252645135
+  %or = or i32 %shl, %shr
+  ret i32 %or
+}
+
+define i64 @grev4_i64(i64 %a) nounwind {
+; RV64I-LABEL: grev4_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 4
+; RV64I-NEXT:    lui a2, 1044721
+; RV64I-NEXT:    addiw a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 240
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a0, a0, 4
+; RV64I-NEXT:    lui a2, 3855
+; RV64I-NEXT:    addiw a2, a2, 241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -241
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev4_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    rev4.b a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev4_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    rev4.b a0, a0
+; RV64IBP-NEXT:    ret
+  %and = shl i64 %a, 4
+  %shl = and i64 %and, -1085102592571150096
+  %and1 = lshr i64 %a, 4
+  %shr = and i64 %and1, 1085102592571150095
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+define signext i32 @grev8_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: grev8_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 8
+; RV64I-NEXT:    lui a2, 16
+; RV64I-NEXT:    addiw a2, a2, -255
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, -256
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a0, a0, 8
+; RV64I-NEXT:    lui a2, 4080
+; RV64I-NEXT:    addiw a2, a2, 255
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    sext.w a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev8_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    greviw a0, a0, 8
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev8_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    greviw a0, a0, 8
+; RV64IBP-NEXT:    ret
+  %and = shl i32 %a, 8
+  %shl = and i32 %and, -16711936
+  %and1 = lshr i32 %a, 8
+  %shr = and i32 %and1, 16711935
+  %or = or i32 %shl, %shr
+  ret i32 %or
+}
+
+define i64 @grev8_i64(i64 %a) nounwind {
+; RV64I-LABEL: grev8_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 8
+; RV64I-NEXT:    lui a2, 1044496
+; RV64I-NEXT:    addiw a2, a2, -255
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, -255
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, -256
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a0, a0, 8
+; RV64I-NEXT:    lui a2, 4080
+; RV64I-NEXT:    addiw a2, a2, 255
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, 255
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, 255
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev8_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    rev8.h a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev8_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    rev8.h a0, a0
+; RV64IBP-NEXT:    ret
+  %and = shl i64 %a, 8
+  %shl = and i64 %and, -71777214294589696
+  %and1 = lshr i64 %a, 8
+  %shr = and i64 %and1, 71777214294589695
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+define signext i32 @grev16_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: grev16_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 16
+; RV64I-NEXT:    srliw a0, a0, 16
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    sext.w a0, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev16_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    greviw a0, a0, 16
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev16_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    greviw a0, a0, 16
+; RV64IBP-NEXT:    ret
+  %shl = shl i32 %a, 16
+  %shr = lshr i32 %a, 16
+  %or = or i32 %shl, %shr
+  ret i32 %or
+}
+
+define i64 @grev16_i64(i64 %a) nounwind {
+; RV64I-LABEL: grev16_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 16
+; RV64I-NEXT:    lui a2, 1048560
+; RV64I-NEXT:    addiw a2, a2, 1
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, -1
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a0, a0, 16
+; RV64I-NEXT:    lui a2, 16
+; RV64I-NEXT:    addiw a2, a2, -1
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, 1
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, -1
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev16_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    rev16.w a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev16_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    rev16.w a0, a0
+; RV64IBP-NEXT:    ret
+  %and = shl i64 %a, 16
+  %shl = and i64 %and, -281470681808896
+  %and1 = lshr i64 %a, 16
+  %shr = and i64 %and1, 281470681808895
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+define i64 @grev32(i64 %a) nounwind {
+; RV64I-LABEL: grev32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 32
+; RV64I-NEXT:    srli a0, a0, 32
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: grev32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    rori a0, a0, 32
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: grev32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    rori a0, a0, 32
+; RV64IBP-NEXT:    ret
+  %shl = shl i64 %a, 32
+  %shr = lshr i64 %a, 32
+  %or = or i64 %shl, %shr
+  ret i64 %or
+}
+
+declare i32 @llvm.bswap.i32(i32)
+
+define signext i32 @bswap_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: bswap_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    slli a1, a0, 8
+; RV64I-NEXT:    addi a2, zero, 255
+; RV64I-NEXT:    slli a3, a2, 32
+; RV64I-NEXT:    and a1, a1, a3
+; RV64I-NEXT:    slli a3, a0, 24
+; RV64I-NEXT:    slli a4, a2, 40
+; RV64I-NEXT:    and a3, a3, a4
+; RV64I-NEXT:    or a1, a3, a1
+; RV64I-NEXT:    slli a3, a0, 40
+; RV64I-NEXT:    slli a2, a2, 48
+; RV64I-NEXT:    and a2, a3, a2
+; RV64I-NEXT:    slli a0, a0, 56
+; RV64I-NEXT:    or a0, a0, a2
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srai a0, a0, 32
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: bswap_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    greviw a0, a0, 24
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: bswap_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    greviw a0, a0, 24
+; RV64IBP-NEXT:    ret
+  %1 = tail call i32 @llvm.bswap.i32(i32 %a)
+  ret i32 %1
+}
+
+declare i64 @llvm.bswap.i64(i64)
+
+define i64 @bswap_i64(i64 %a) {
+; RV64I-LABEL: bswap_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    srli a1, a0, 24
+; RV64I-NEXT:    lui a2, 4080
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 8
+; RV64I-NEXT:    addi a3, zero, 255
+; RV64I-NEXT:    slli a4, a3, 24
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    or a1, a2, a1
+; RV64I-NEXT:    srli a2, a0, 40
+; RV64I-NEXT:    lui a4, 16
+; RV64I-NEXT:    addiw a4, a4, -256
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    srli a4, a0, 56
+; RV64I-NEXT:    or a2, a2, a4
+; RV64I-NEXT:    or a1, a1, a2
+; RV64I-NEXT:    slli a2, a0, 8
+; RV64I-NEXT:    slli a4, a3, 32
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    slli a4, a0, 24
+; RV64I-NEXT:    slli a5, a3, 40
+; RV64I-NEXT:    and a4, a4, a5
+; RV64I-NEXT:    or a2, a4, a2
+; RV64I-NEXT:    slli a4, a0, 40
+; RV64I-NEXT:    slli a3, a3, 48
+; RV64I-NEXT:    and a3, a4, a3
+; RV64I-NEXT:    slli a0, a0, 56
+; RV64I-NEXT:    or a0, a0, a3
+; RV64I-NEXT:    or a0, a0, a2
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: bswap_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    rev8 a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: bswap_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    rev8 a0, a0
+; RV64IBP-NEXT:    ret
+  %1 = call i64 @llvm.bswap.i64(i64 %a)
+  ret i64 %1
+}
+
+declare i32 @llvm.bitreverse.i32(i32)
+
+define signext i32 @bitreverse_i32(i32 signext %a) nounwind {
+; RV64I-LABEL: bitreverse_i32:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    srli a1, a0, 24
+; RV64I-NEXT:    lui a2, 4080
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 8
+; RV64I-NEXT:    addi a3, zero, 255
+; RV64I-NEXT:    slli a4, a3, 24
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    or a1, a2, a1
+; RV64I-NEXT:    srli a2, a0, 40
+; RV64I-NEXT:    lui a4, 16
+; RV64I-NEXT:    addiw a4, a4, -256
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    srli a4, a0, 56
+; RV64I-NEXT:    or a2, a2, a4
+; RV64I-NEXT:    or a1, a1, a2
+; RV64I-NEXT:    slli a2, a0, 8
+; RV64I-NEXT:    slli a4, a3, 32
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    slli a4, a0, 24
+; RV64I-NEXT:    slli a5, a3, 40
+; RV64I-NEXT:    and a4, a4, a5
+; RV64I-NEXT:    or a2, a4, a2
+; RV64I-NEXT:    slli a4, a0, 40
+; RV64I-NEXT:    slli a3, a3, 48
+; RV64I-NEXT:    and a3, a4, a3
+; RV64I-NEXT:    slli a0, a0, 56
+; RV64I-NEXT:    or a0, a0, a3
+; RV64I-NEXT:    or a0, a0, a2
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    lui a1, 3855
+; RV64I-NEXT:    addiw a1, a1, 241
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, -241
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 241
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, -241
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a1, a1, 4
+; RV64I-NEXT:    lui a2, 1044721
+; RV64I-NEXT:    addiw a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 240
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    srli a0, a0, 4
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    lui a1, 13107
+; RV64I-NEXT:    addiw a1, a1, 819
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 819
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 819
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 819
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a1, a1, 2
+; RV64I-NEXT:    lui a2, 1035469
+; RV64I-NEXT:    addiw a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -820
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    srli a0, a0, 2
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    lui a1, 349525
+; RV64I-NEXT:    addiw a1, a1, 1365
+; RV64I-NEXT:    slli a1, a1, 32
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a1, a1, 1
+; RV64I-NEXT:    lui a2, 873813
+; RV64I-NEXT:    addiw a2, a2, 1365
+; RV64I-NEXT:    slli a2, a2, 33
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    srli a0, a0, 1
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    srai a0, a0, 32
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: bitreverse_i32:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    greviw a0, a0, 31
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: bitreverse_i32:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    greviw a0, a0, 31
+; RV64IBP-NEXT:    ret
+  %1 = tail call i32 @llvm.bitreverse.i32(i32 %a)
+  ret i32 %1
+}
+
+declare i64 @llvm.bitreverse.i64(i64)
+
+define i64 @bitreverse_i64(i64 %a) nounwind {
+; RV64I-LABEL: bitreverse_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    srli a1, a0, 24
+; RV64I-NEXT:    lui a2, 4080
+; RV64I-NEXT:    and a1, a1, a2
+; RV64I-NEXT:    srli a2, a0, 8
+; RV64I-NEXT:    addi a3, zero, 255
+; RV64I-NEXT:    slli a4, a3, 24
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    or a1, a2, a1
+; RV64I-NEXT:    srli a2, a0, 40
+; RV64I-NEXT:    lui a4, 16
+; RV64I-NEXT:    addiw a4, a4, -256
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    srli a4, a0, 56
+; RV64I-NEXT:    or a2, a2, a4
+; RV64I-NEXT:    or a1, a1, a2
+; RV64I-NEXT:    slli a2, a0, 8
+; RV64I-NEXT:    slli a4, a3, 32
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    slli a4, a0, 24
+; RV64I-NEXT:    slli a5, a3, 40
+; RV64I-NEXT:    and a4, a4, a5
+; RV64I-NEXT:    or a2, a4, a2
+; RV64I-NEXT:    slli a4, a0, 40
+; RV64I-NEXT:    slli a3, a3, 48
+; RV64I-NEXT:    and a3, a4, a3
+; RV64I-NEXT:    slli a0, a0, 56
+; RV64I-NEXT:    or a0, a0, a3
+; RV64I-NEXT:    or a0, a0, a2
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    lui a1, 3855
+; RV64I-NEXT:    addiw a1, a1, 241
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, -241
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 241
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, -241
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a1, a1, 4
+; RV64I-NEXT:    lui a2, 1044721
+; RV64I-NEXT:    addiw a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -241
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, 240
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    srli a0, a0, 4
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    lui a1, 13107
+; RV64I-NEXT:    addiw a1, a1, 819
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 819
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 819
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 819
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a1, a1, 2
+; RV64I-NEXT:    lui a2, 1035469
+; RV64I-NEXT:    addiw a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -819
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -820
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    srli a0, a0, 2
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    lui a1, 21845
+; RV64I-NEXT:    addiw a1, a1, 1365
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 1365
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 1365
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 1365
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a1, a1, 1
+; RV64I-NEXT:    lui a2, 1026731
+; RV64I-NEXT:    addiw a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1365
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1366
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    srli a0, a0, 1
+; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: bitreverse_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    rev a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: bitreverse_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    rev a0, a0
+; RV64IBP-NEXT:    ret
+  %1 = call i64 @llvm.bitreverse.i64(i64 %a)
+  ret i64 %1
+}
+
+; There's no [un]shfliw instruction as slliu.w occupies the encoding slot that
+; would be occupied by shfliw.
+
+define i64 @shfl1_i64(i64 %a, i64 %b) nounwind {
+; RV64I-LABEL: shfl1_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    lui a1, 1035469
+; RV64I-NEXT:    addiw a1, a1, -819
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, -819
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, -819
+; RV64I-NEXT:    slli a1, a1, 13
+; RV64I-NEXT:    addi a1, a1, -1639
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a2, a0, 1
+; RV64I-NEXT:    lui a3, 4369
+; RV64I-NEXT:    addiw a3, a3, 273
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, 273
+; RV64I-NEXT:    slli a3, a3, 12
+; RV64I-NEXT:    addi a3, a3, 273
+; RV64I-NEXT:    slli a4, a3, 14
+; RV64I-NEXT:    addi a4, a4, 1092
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    or a1, a2, a1
+; RV64I-NEXT:    srli a0, a0, 1
+; RV64I-NEXT:    slli a2, a3, 13
+; RV64I-NEXT:    addi a2, a2, 546
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: shfl1_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    zip.n a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: shfl1_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    zip.n a0, a0
+; RV64IBP-NEXT:    ret
+  %and = and i64 %a, -7378697629483820647
+  %shl = shl i64 %a, 1
+  %and1 = and i64 %shl, 4919131752989213764
+  %or = or i64 %and1, %and
+  %shr = lshr i64 %a, 1
+  %and2 = and i64 %shr, 2459565876494606882
+  %or3 = or i64 %or, %and2
+  ret i64 %or3
+}
+
+define i64 @shfl2_i64(i64 %a, i64 %b) nounwind {
+; RV64I-LABEL: shfl2_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    lui a1, 1044721
+; RV64I-NEXT:    addiw a1, a1, -241
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 241
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, -241
+; RV64I-NEXT:    slli a1, a1, 14
+; RV64I-NEXT:    addi a1, a1, 963
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a2, a0, 2
+; RV64I-NEXT:    lui a3, 48
+; RV64I-NEXT:    addiw a3, a3, 771
+; RV64I-NEXT:    slli a3, a3, 16
+; RV64I-NEXT:    addi a3, a3, 771
+; RV64I-NEXT:    slli a4, a3, 16
+; RV64I-NEXT:    addi a4, a4, 771
+; RV64I-NEXT:    slli a4, a4, 12
+; RV64I-NEXT:    addi a4, a4, 48
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    or a1, a2, a1
+; RV64I-NEXT:    srli a0, a0, 2
+; RV64I-NEXT:    slli a2, a3, 14
+; RV64I-NEXT:    addi a2, a2, 193
+; RV64I-NEXT:    slli a2, a2, 12
+; RV64I-NEXT:    addi a2, a2, -1012
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: shfl2_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    zip2.b a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: shfl2_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    zip2.b a0, a0
+; RV64IBP-NEXT:    ret
+  %and = and i64 %a, -4340410370284600381
+  %shl = shl i64 %a, 2
+  %and1 = and i64 %shl, 3472328296227680304
+  %or = or i64 %and1, %and
+  %shr = lshr i64 %a, 2
+  %and2 = and i64 %shr, 868082074056920076
+  %or3 = or i64 %or, %and2
+  ret i64 %or3
+}
+
+define i64 @shfl4_i64(i64 %a, i64 %b) nounwind {
+; RV64I-LABEL: shfl4_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    lui a1, 1048560
+; RV64I-NEXT:    addiw a1, a1, 255
+; RV64I-NEXT:    slli a1, a1, 16
+; RV64I-NEXT:    addi a1, a1, 255
+; RV64I-NEXT:    slli a1, a1, 16
+; RV64I-NEXT:    addi a1, a1, 255
+; RV64I-NEXT:    slli a1, a1, 12
+; RV64I-NEXT:    addi a1, a1, 15
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a2, a0, 4
+; RV64I-NEXT:    lui a3, 240
+; RV64I-NEXT:    addiw a3, a3, 15
+; RV64I-NEXT:    slli a3, a3, 16
+; RV64I-NEXT:    addi a3, a3, 15
+; RV64I-NEXT:    slli a4, a3, 12
+; RV64I-NEXT:    addi a4, a4, 1
+; RV64I-NEXT:    slli a4, a4, 12
+; RV64I-NEXT:    addi a4, a4, -256
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    or a1, a2, a1
+; RV64I-NEXT:    srli a0, a0, 4
+; RV64I-NEXT:    slli a2, a3, 20
+; RV64I-NEXT:    addi a2, a2, 240
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: shfl4_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    zip4.h a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: shfl4_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    zip4.h a0, a0
+; RV64IBP-NEXT:    ret
+  %and = and i64 %a, -1148435428713435121
+  %shl = shl i64 %a, 4
+  %and1 = and i64 %shl, 1080880403494997760
+  %or = or i64 %and1, %and
+  %shr = lshr i64 %a, 4
+  %and2 = and i64 %shr, 67555025218437360
+  %or3 = or i64 %or, %and2
+  ret i64 %or3
+}
+
+define i64 @shfl8_i64(i64 %a, i64 %b) nounwind {
+; RV64I-LABEL: shfl8_i64:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    lui a1, 1048560
+; RV64I-NEXT:    addiw a1, a1, 1
+; RV64I-NEXT:    slli a1, a1, 16
+; RV64I-NEXT:    addi a1, a1, -1
+; RV64I-NEXT:    slli a1, a1, 24
+; RV64I-NEXT:    addi a1, a1, 255
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a2, a0, 8
+; RV64I-NEXT:    addi a3, zero, 255
+; RV64I-NEXT:    slli a4, a3, 32
+; RV64I-NEXT:    addi a4, a4, 255
+; RV64I-NEXT:    slli a4, a4, 16
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    or a1, a2, a1
+; RV64I-NEXT:    srli a0, a0, 8
+; RV64I-NEXT:    slli a2, a3, 24
+; RV64I-NEXT:    addi a2, a2, 1
+; RV64I-NEXT:    slli a2, a2, 16
+; RV64I-NEXT:    addi a2, a2, -256
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: shfl8_i64:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    zip8.w a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: shfl8_i64:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    zip8.w a0, a0
+; RV64IBP-NEXT:    ret
+  %and = and i64 %a, -72056494543077121
+  %shl = shl i64 %a, 8
+  %and1 = and i64 %shl, 71776119077928960
+  %or = or i64 %and1, %and
+  %shr = lshr i64 %a, 8
+  %and2 = and i64 %shr, 280375465148160
+  %or3 = or i64 %or, %and2
+  ret i64 %or3
+}
+
+define i64 @shfl16(i64 %a, i64 %b) nounwind {
+; RV64I-LABEL: shfl16:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    addi a1, zero, -1
+; RV64I-NEXT:    slli a1, a1, 32
+; RV64I-NEXT:    addi a1, a1, 1
+; RV64I-NEXT:    slli a1, a1, 16
+; RV64I-NEXT:    addi a1, a1, -1
+; RV64I-NEXT:    and a1, a0, a1
+; RV64I-NEXT:    slli a2, a0, 16
+; RV64I-NEXT:    lui a3, 16
+; RV64I-NEXT:    addiw a3, a3, -1
+; RV64I-NEXT:    slli a4, a3, 32
+; RV64I-NEXT:    and a2, a2, a4
+; RV64I-NEXT:    or a1, a2, a1
+; RV64I-NEXT:    srli a0, a0, 16
+; RV64I-NEXT:    slli a2, a3, 16
+; RV64I-NEXT:    and a0, a0, a2
+; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    ret
+;
+; RV64IB-LABEL: shfl16:
+; RV64IB:       # %bb.0:
+; RV64IB-NEXT:    zip16 a0, a0
+; RV64IB-NEXT:    ret
+;
+; RV64IBP-LABEL: shfl16:
+; RV64IBP:       # %bb.0:
+; RV64IBP-NEXT:    zip16 a0, a0
+; RV64IBP-NEXT:    ret
+  %and = and i64 %a, -281474976645121
+  %shl = shl i64 %a, 16
+  %and1 = and i64 %shl, 281470681743360
+  %or = or i64 %and1, %and
+  %shr = lshr i64 %a, 16
+  %and2 = and i64 %shr, 4294901760
+  %or3 = or i64 %or, %and2
+  ret i64 %or3
+}