prefetch_buffer.sv

// Copyright 2015 ETH Zurich and University of Bologna.
// Copyright and related rights are licensed under the Solderpad Hardware
// License, Version 0.51 (the “License”); you may not use this file except in
// compliance with the License.  You may obtain a copy of the License at
// http://solderpad.org/licenses/SHL-0.51. Unless required by applicable law
// or agreed to in writing, software, hardware and materials distributed under
// this License is distributed on an “AS IS” BASIS, WITHOUT WARRANTIES OR
// CONDITIONS OF ANY KIND, either express or implied. See the License for the
// specific language governing permissions and limitations under the License.

////////////////////////////////////////////////////////////////////////////////
// Engineer:       Andreas Traber - atraber@iis.ee.ethz.ch                    //
//                                                                            //
// Design Name:    Prefetcher Buffer for 32 bit memory interface              //
// Project Name:   RI5CY                                                      //
// Language:       SystemVerilog                                              //
//                                                                            //
// Description:    Prefetch Buffer that caches instructions. This cuts overly //
//                 long critical paths to the instruction cache               //
//                                                                            //
////////////////////////////////////////////////////////////////////////////////

// input port: send address one cycle before the data
// clear_i clears the FIFO for the following cycle. in_addr_i can be sent in
// this cycle already
module riscv_fetch_fifo
(
    input  logic        clk,
    input  logic        rst_n,

    // control signals
    input  logic        clear_i,          // clears the contents of the fifo

    // input port
    input  logic [31:0] in_addr_i,
    input  logic [31:0] in_rdata_i,
    input  logic        in_valid_i,
    output logic        in_ready_o,

    input  logic        in_replace2_i, // replaces second entry if there is one: "to be served after this instr"
    input  logic        in_is_hwlp_i,

    // output port
    output logic        out_valid_o,
    input  logic        out_ready_i,
    output logic [31:0] out_rdata_o,
    output logic [31:0] out_addr_o,

    output logic        out_valid_stored_o, // same as out_valid_o, except that if something is incoming now it is not included. This signal is available immediately as it comes directly out of FFs
    output logic        out_is_hwlp_o
  );

  localparam DEPTH = 4; // must be 3 or greater

  // index 0 is used for output
  logic [0:DEPTH-1] [31:0]  addr_n,    addr_int,    addr_Q;
  logic [0:DEPTH-1] [31:0]  rdata_n,   rdata_int,   rdata_Q;
  logic [0:DEPTH-1]         valid_n,   valid_int,   valid_Q;
  logic [0:1      ]         is_hwlp_n, is_hwlp_int, is_hwlp_Q;

  logic             [31:0]  addr_next;
  logic             [31:0]  rdata, rdata_unaligned;
  logic                     valid, valid_unaligned;

  logic                     aligned_is_compressed, unaligned_is_compressed;
  logic                     aligned_is_compressed_st, unaligned_is_compressed_st;

  //////////////////////////////////////////////////////////////////////////////
  // output port
  //////////////////////////////////////////////////////////////////////////////


  assign rdata = (valid_Q[0]) ? rdata_Q[0] : in_rdata_i;
  assign valid = valid_Q[0] || in_valid_i || is_hwlp_Q[1];

  assign rdata_unaligned = (valid_Q[1]) ? {rdata_Q[1][15:0], rdata[31:16]} : {in_rdata_i[15:0], rdata[31:16]};
  // it is implied that rdata_valid_Q[0] is set
  assign valid_unaligned = (valid_Q[1] || (valid_Q[0] && in_valid_i));

  assign unaligned_is_compressed    = rdata[17:16] != 2'b11;
  assign aligned_is_compressed      = rdata[1:0] != 2'b11;
  assign unaligned_is_compressed_st = rdata_Q[0][17:16] != 2'b11;
  assign aligned_is_compressed_st   = rdata_Q[0][1:0] != 2'b11;

  //////////////////////////////////////////////////////////////////////////////
  // instruction aligner (if unaligned)
  //////////////////////////////////////////////////////////////////////////////

  always_comb
  begin
    // serve the aligned case even though the output address is unaligned when
    // the next instruction will be from a hardware loop target
    // in this case the current instruction is already prealigned in element 0
    if (out_addr_o[1] && (~is_hwlp_Q[1])) begin
      // unaligned case
      out_rdata_o = rdata_unaligned;

      if (unaligned_is_compressed)
        out_valid_o = valid;
      else
        out_valid_o = valid_unaligned;
    end else begin
      // aligned case
      out_rdata_o = rdata;
      out_valid_o = valid;
    end
  end

  assign out_addr_o    = (valid_Q[0]) ? addr_Q[0] : in_addr_i;
  assign out_is_hwlp_o = (valid_Q[0]) ? is_hwlp_Q[0] : in_is_hwlp_i;

  // this valid signal must not depend on signals from outside!
  always_comb
  begin
    out_valid_stored_o = 1'b1;

    if (out_addr_o[1] && (~is_hwlp_Q[1])) begin
      if (unaligned_is_compressed_st)
        out_valid_stored_o = 1'b1;
      else
        out_valid_stored_o = valid_Q[1];
    end else begin
      out_valid_stored_o = valid_Q[0];
    end
  end


  //////////////////////////////////////////////////////////////////////////////
  // input port
  //////////////////////////////////////////////////////////////////////////////

  // we accept data as long as our fifo is not full
  // we don't care about clear here as the data will be received one cycle
  // later anyway
  assign in_ready_o = ~valid_Q[DEPTH-2];


  //////////////////////////////////////////////////////////////////////////////
  // FIFO management
  //////////////////////////////////////////////////////////////////////////////

  int j;
  always_comb
  begin
    addr_int    = addr_Q;
    rdata_int   = rdata_Q;
    valid_int   = valid_Q;
    is_hwlp_int = is_hwlp_Q;

    if (in_valid_i) begin
      for(j = 0; j < DEPTH; j++) begin
        if (~valid_Q[j]) begin
          addr_int[j]  = in_addr_i;
          rdata_int[j] = in_rdata_i;
          valid_int[j] = 1'b1;

          break;
        end
      end

      // replace 2nd entry
      if (in_replace2_i) begin
        if (valid_Q[0]) begin
          addr_int[1]          = in_addr_i;

          // if we replace the 2nd entry, let's cache the output word in case we
          // still need it and it would span two words in the FIFO
          rdata_int[0]         = out_rdata_o;
          rdata_int[1]         = in_rdata_i;
          valid_int[1]         = 1'b1;
          valid_int[2:DEPTH-1] = '0;

          // hardware loop incoming?
          is_hwlp_int[1] = in_is_hwlp_i;
        end else begin
          is_hwlp_int[0] = in_is_hwlp_i;
        end
      end
    end
  end

  assign addr_next = {addr_int[0][31:2], 2'b00} + 32'h4;

  // move everything by one step
  always_comb
  begin
    addr_n     = addr_int;
    rdata_n    = rdata_int;
    valid_n    = valid_int;
    is_hwlp_n  = is_hwlp_int;

    if (out_ready_i && out_valid_o) begin
      is_hwlp_n = {is_hwlp_int[1], 1'b0};

      if (is_hwlp_int[1]) begin
        addr_n[0] = addr_int[1][31:0];
        rdata_n   = {rdata_int[1:DEPTH-1], 32'b0};
        valid_n   = {valid_int[1:DEPTH-1], 1'b0};
      end else begin
        if (addr_int[0][1]) begin
          // unaligned case
          if (unaligned_is_compressed) begin
            addr_n[0] = {addr_next[31:2], 2'b00};
          end else begin
            addr_n[0] = {addr_next[31:2], 2'b10};
          end

          rdata_n  = {rdata_int[1:DEPTH-1], 32'b0};
          valid_n  = {valid_int[1:DEPTH-1], 1'b0};
        end else begin
          // aligned case
          if (aligned_is_compressed) begin
            // just increase address, do not move to next entry in FIFO
            addr_n[0] = {addr_int[0][31:2], 2'b10};
          end else begin
            // move to next entry in FIFO
            addr_n[0] = {addr_next[31:2], 2'b00};
            rdata_n   = {rdata_int[1:DEPTH-1], 32'b0};
            valid_n   = {valid_int[1:DEPTH-1], 1'b0};
          end
        end
      end
    end
  end

  //////////////////////////////////////////////////////////////////////////////
  // registers
  //////////////////////////////////////////////////////////////////////////////

  always_ff @(posedge clk, negedge rst_n)
  begin
    if(rst_n == 1'b0)
    begin
      addr_Q    <= '{default: '0};
      rdata_Q   <= '{default: '0};
      valid_Q   <= '0;
      is_hwlp_Q <= '0;
    end
    else
    begin
      // on a clear signal from outside we invalidate the content of the FIFO
      // completely and start from an empty state
      if (clear_i) begin
        valid_Q    <= '0;
        is_hwlp_Q  <= '0;
      end else begin
        addr_Q    <= addr_n;
        rdata_Q   <= rdata_n;
        valid_Q   <= valid_n;
        is_hwlp_Q <= is_hwlp_n;
      end
    end
  end

  //----------------------------------------------------------------------------
  // Assertions
  //----------------------------------------------------------------------------

  // check for FIFO overflows
  assert property (
    @(posedge clk) (in_valid_i) |-> ((valid_Q[DEPTH-1] == 1'b0) || (clear_i == 1'b1) || (in_replace2_i == 1'b1)) );

endmodule


module riscv_prefetch_buffer
(
  input  logic        clk,
  input  logic        rst_n,

  input  logic        req_i,

  input  logic        branch_i,
  input  logic [31:0] addr_i,

  input  logic        hwloop_i,
  input  logic [31:0] hwloop_target_i,

  input  logic        ready_i,
  output logic        valid_o,
  output logic [31:0] rdata_o,
  output logic [31:0] addr_o,
  output logic        is_hwlp_o, // is set when the currently served data is from a hwloop

  // goes to instruction memory / instruction cache
  output logic        instr_req_o,
  input  logic        instr_gnt_i,
  output logic [31:0] instr_addr_o,
  input  logic [31:0] instr_rdata_i,
  input  logic        instr_rvalid_i,

  // Prefetch Buffer Status
  output logic        busy_o
);

  enum logic [1:0] {IDLE, WAIT_GNT, WAIT_RVALID, WAIT_ABORTED } CS, NS;
  enum logic [1:0] {HWLP_NONE, HWLP_IN, HWLP_FETCHING, HWLP_DONE } hwlp_CS, hwlp_NS;

  logic [31:0] instr_addr_q, fetch_addr;
  logic        fetch_is_hwlp;
  logic        addr_valid;

  logic        fifo_valid;
  logic        fifo_ready;
  logic        fifo_clear;
  logic        fifo_hwlp;

  logic        valid_stored;
  logic        hwlp_masked;


  //////////////////////////////////////////////////////////////////////////////
  // prefetch buffer status
  //////////////////////////////////////////////////////////////////////////////

  assign busy_o = (CS != IDLE) || instr_req_o;

  //////////////////////////////////////////////////////////////////////////////
  // fetch fifo
  // consumes addresses and rdata
  //////////////////////////////////////////////////////////////////////////////

  riscv_fetch_fifo fifo_i
  (
    .clk                   ( clk               ),
    .rst_n                 ( rst_n             ),

    .clear_i               ( fifo_clear        ),

    .in_addr_i             ( instr_addr_q      ),
    .in_rdata_i            ( instr_rdata_i     ),
    .in_valid_i            ( fifo_valid        ),
    .in_ready_o            ( fifo_ready        ),

    .in_replace2_i         ( fifo_hwlp         ),
    .in_is_hwlp_i          ( fifo_hwlp         ),

    .out_valid_o           ( valid_o           ),
    .out_ready_i           ( ready_i           ),
    .out_rdata_o           ( rdata_o           ),
    .out_addr_o            ( addr_o            ),

    .out_valid_stored_o    ( valid_stored      ),
    .out_is_hwlp_o         ( is_hwlp_o         )
  );


  //////////////////////////////////////////////////////////////////////////////
  // fetch addr
  //////////////////////////////////////////////////////////////////////////////

  assign fetch_addr = {instr_addr_q[31:2], 2'b00} + 32'd4;

  always_comb
  begin
    hwlp_NS     = hwlp_CS;
    fifo_hwlp   = 1'b0;
    hwlp_masked = 1'b0;
    fifo_clear  = 1'b0;

    unique case (hwlp_CS)
      HWLP_NONE: begin
        if (hwloop_i) begin
          hwlp_masked = 1'b1;

          if (fetch_is_hwlp)
            hwlp_NS = HWLP_FETCHING;
          else
            hwlp_NS = HWLP_IN;

          if (ready_i)
            fifo_clear = 1'b1;
        end
      end

      HWLP_IN: begin
        hwlp_masked = 1'b1;

        if (fetch_is_hwlp)
          hwlp_NS = HWLP_FETCHING;

        if (ready_i)
          fifo_clear = 1'b1;
      end

      // just waiting for rvalid really
      HWLP_FETCHING: begin
        fifo_hwlp = 1'b1;

        if (instr_rvalid_i & (CS != WAIT_ABORTED)) begin
          if (valid_o & is_hwlp_o)
            hwlp_NS = HWLP_NONE;
          else
            hwlp_NS = HWLP_DONE;
        end else begin
          if (ready_i)
            fifo_clear = 1'b1;
        end
      end

      HWLP_DONE: begin
        if (valid_o & is_hwlp_o)
          hwlp_NS = HWLP_NONE;
      end

      default: begin
        hwlp_NS = HWLP_NONE;
      end
    endcase

    if (branch_i) begin
      hwlp_NS    = HWLP_NONE;
      fifo_clear = 1'b1;
    end
  end

  //////////////////////////////////////////////////////////////////////////////
  // instruction fetch FSM
  // deals with instruction memory / instruction cache
  //////////////////////////////////////////////////////////////////////////////

  always_comb
  begin
    instr_req_o   = 1'b0;
    instr_addr_o  = fetch_addr;
    fifo_valid    = 1'b0;
    addr_valid    = 1'b0;
    fetch_is_hwlp = 1'b0;
    NS            = CS;

    unique case(CS)
      // default state, not waiting for requested data
      IDLE:
      begin
        instr_addr_o = fetch_addr;
        instr_req_o  = 1'b0;

        if (branch_i)
          instr_addr_o = addr_i;
        else if(hwlp_masked & valid_stored)
          instr_addr_o = hwloop_target_i;

        if (req_i & (fifo_ready | branch_i | (hwlp_masked & valid_stored))) begin
          instr_req_o = 1'b1;
          addr_valid  = 1'b1;

          if (hwlp_masked & valid_stored) begin
            fetch_is_hwlp = 1'b1;
          end

          if(instr_gnt_i) //~>  granted request
            NS = WAIT_RVALID;
          else begin //~> got a request but no grant
            NS = WAIT_GNT;
          end
        end
      end // case: IDLE

      // we sent a request but did not yet get a grant
      WAIT_GNT:
      begin
        instr_addr_o = instr_addr_q;
        instr_req_o  = 1'b1;

        if (branch_i) begin
          instr_addr_o = addr_i;
          addr_valid   = 1'b1;
        end else if (hwlp_masked & valid_stored) begin
          instr_addr_o  = hwloop_target_i;
          addr_valid    = 1'b1;
          fetch_is_hwlp = 1'b1;
        end

        if(instr_gnt_i)
          NS = WAIT_RVALID;
        else
          NS = WAIT_GNT;
      end // case: WAIT_GNT

      // we wait for rvalid, after that we are ready to serve a new request
      WAIT_RVALID: begin
        instr_addr_o = fetch_addr;

        if (branch_i)
          instr_addr_o  = addr_i;
        else if (hwlp_masked)
          instr_addr_o  = hwloop_target_i;

        if (req_i & (fifo_ready | branch_i | hwlp_masked)) begin
          // prepare for next request

          if (instr_rvalid_i) begin
            instr_req_o = 1'b1;
            fifo_valid  = 1'b1;
            addr_valid  = 1'b1;

            if (hwlp_masked) begin
              fetch_is_hwlp = 1'b1;
            end

            if (instr_gnt_i) begin
              NS = WAIT_RVALID;
            end else begin
              NS = WAIT_GNT;
            end
          end else begin
            // we are requested to abort our current request
            // we didn't get an rvalid yet, so wait for it
            if (branch_i) begin
              addr_valid = 1'b1;
              NS         = WAIT_ABORTED;
            end else if (hwlp_masked & valid_o) begin
              addr_valid    = 1'b1;
              fetch_is_hwlp = 1'b1;
              NS            = WAIT_ABORTED;
            end
          end
        end else begin
          // just wait for rvalid and go back to IDLE, no new request

          if (instr_rvalid_i) begin
            fifo_valid = 1'b1;
            NS         = IDLE;
          end
        end
      end // case: WAIT_RVALID

      // our last request was aborted, but we didn't yet get a rvalid and
      // there was no new request sent yet
      // we assume that req_i is set to high
      WAIT_ABORTED: begin
        instr_addr_o = instr_addr_q;

        if (branch_i) begin
          instr_addr_o = addr_i;
          addr_valid   = 1'b1;
        end

        if (instr_rvalid_i) begin
          instr_req_o  = 1'b1;
          // no need to send address, already done in WAIT_RVALID

          if (instr_gnt_i) begin
            NS = WAIT_RVALID;
          end else begin
            NS = WAIT_GNT;
          end
        end
      end

      default:
      begin
        NS          = IDLE;
        instr_req_o = 1'b0;
      end
    endcase
  end

  //////////////////////////////////////////////////////////////////////////////
  // registers
  //////////////////////////////////////////////////////////////////////////////

  always_ff @(posedge clk, negedge rst_n)
  begin
    if(rst_n == 1'b0)
    begin
      CS              <= IDLE;
      hwlp_CS         <= HWLP_NONE;
      instr_addr_q    <= '0;
    end
    else
    begin
      CS              <= NS;
      hwlp_CS         <= hwlp_NS;

      if (addr_valid) begin
        instr_addr_q    <= instr_addr_o;
      end
    end
  end

endmodule