dcache: Introduce an extra cycle latency to make timing

[microwatt.git] / dcache.vhdl

--
-- Set associative dcache write-through
--
-- TODO (in no specific order):
--
-- * See list in icache.vhdl
-- * Complete load misses on the cycle when WB data comes instead of
--   at the end of line (this requires dealing with requests coming in
--   while not idle...)
-- * Load with update could use one less non-pipelined cycle by moving
--   the register update to the pipeline bubble that exists when going
--   back to the IDLE state.
--
library ieee;
use ieee.std_logic_1164.all;
use ieee.numeric_std.all;

library work;
use work.common.all;
use work.helpers.all;
use work.wishbone_types.all;

entity dcache is
    generic (
        -- Line size in bytes
        LINE_SIZE : positive := 64;
        -- Number of lines in a set
        NUM_LINES : positive := 32;
        -- Number of ways
        NUM_WAYS  : positive := 4
        );
    port (
        clk          : in std_ulogic;
        rst          : in std_ulogic;

        d_in         : in Loadstore1ToDcacheType;
        d_out        : out DcacheToWritebackType;

        wishbone_out : out wishbone_master_out;
        wishbone_in  : in wishbone_slave_out
        );
end entity dcache;

architecture rtl of dcache is
    function log2(i : natural) return integer is
        variable tmp : integer := i;
        variable ret : integer := 0;
    begin
        while tmp > 1 loop
            ret  := ret + 1;
            tmp := tmp / 2;
        end loop;
        return ret;
    end function;

    function ispow2(i : integer) return boolean is
    begin
        if to_integer(to_unsigned(i, 32) and to_unsigned(i - 1, 32)) = 0 then
            return true;
        else
            return false;
        end if;
    end function;

    -- BRAM organisation: We never access more than wishbone_data_bits at
    -- a time so to save resources we make the array only that wide, and
    -- use consecutive indices for to make a cache "line"
    --
    -- ROW_SIZE is the width in bytes of the BRAM (based on WB, so 64-bits)
    constant ROW_SIZE      : natural := wishbone_data_bits / 8;
    -- ROW_PER_LINE is the number of row (wishbone transactions) in a line
    constant ROW_PER_LINE  : natural := LINE_SIZE / ROW_SIZE;
    -- BRAM_ROWS is the number of rows in BRAM needed to represent the full
    -- dcache
    constant BRAM_ROWS     : natural := NUM_LINES * ROW_PER_LINE;

    -- Bit fields counts in the address

    -- ROW_BITS is the number of bits to select a row 
    constant ROW_BITS      : natural := log2(BRAM_ROWS);
    -- ROW_LINEBITS is the number of bits to select a row within a line
    constant ROW_LINEBITS  : natural := log2(ROW_PER_LINE);
    -- LINE_OFF_BITS is the number of bits for the offset in a cache line
    constant LINE_OFF_BITS : natural := log2(LINE_SIZE);
    -- ROW_OFF_BITS is the number of bits for the offset in a row
    constant ROW_OFF_BITS  : natural := log2(ROW_SIZE);
    -- INDEX_BITS is the number if bits to select a cache line
    constant INDEX_BITS    : natural := log2(NUM_LINES);
    -- TAG_BITS is the number of bits of the tag part of the address
    constant TAG_BITS      : natural := 64 - LINE_OFF_BITS - INDEX_BITS;
    -- WAY_BITS is the number of bits to select a way
    constant WAY_BITS     : natural := log2(NUM_WAYS);

    -- Example of layout for 32 lines of 64 bytes:
    --
    -- ..  tag    |index|  line  |
    -- ..         |   row   |    |
    -- ..         |     |---|    | ROW_LINEBITS  (3)
    -- ..         |     |--- - --| LINE_OFF_BITS (6)
    -- ..         |         |- --| ROW_OFF_BITS  (3)
    -- ..         |----- ---|    | ROW_BITS      (8)
    -- ..         |-----|        | INDEX_BITS    (5)
    -- .. --------|              | TAG_BITS      (53)

    subtype row_t is integer range 0 to BRAM_ROWS-1;
    subtype index_t is integer range 0 to NUM_LINES-1;
    subtype way_t is integer range 0 to NUM_WAYS-1;

    -- The cache data BRAM organized as described above for each way
    subtype cache_row_t is std_ulogic_vector(wishbone_data_bits-1 downto 0);

    -- The cache tags LUTRAM has a row per set. Vivado is a pain and will
    -- not handle a clean (commented) definition of the cache tags as a 3d
    -- memory. For now, work around it by putting all the tags
    subtype cache_tag_t is std_logic_vector(TAG_BITS-1 downto 0);
--    type cache_tags_set_t is array(way_t) of cache_tag_t;
--    type cache_tags_array_t is array(index_t) of cache_tags_set_t;
    constant TAG_RAM_WIDTH : natural := TAG_BITS * NUM_WAYS;
    subtype cache_tags_set_t is std_logic_vector(TAG_RAM_WIDTH-1 downto 0);
    type cache_tags_array_t is array(index_t) of cache_tags_set_t;

    -- The cache valid bits
    subtype cache_way_valids_t is std_ulogic_vector(NUM_WAYS-1 downto 0);
    type cache_valids_t is array(index_t) of cache_way_valids_t;

    -- Storage. Hopefully "cache_rows" is a BRAM, the rest is LUTs
    signal cache_tags   : cache_tags_array_t;
    signal cache_valids : cache_valids_t;

    attribute ram_style : string;
    attribute ram_style of cache_tags : signal is "distributed";

    -- Type of operation on a "valid" input
    type op_t is (OP_NONE,
                  OP_LOAD_HIT,      -- Cache hit on load
                  OP_LOAD_MISS,     -- Load missing cache
                  OP_LOAD_NC,       -- Non-cachable load
                  OP_BAD,           -- BAD: Cache hit on NC load/store
                  OP_STORE_HIT,     -- Store hitting cache
                  OP_STORE_MISS);   -- Store missing cache
                      
    -- Cache state machine
    type state_t is (IDLE,             -- Normal load hit processing
                     LOAD_UPDATE,      -- Load with update extra cycle
                     LOAD_UPDATE2,     -- Load with update extra cycle
                     RELOAD_WAIT_ACK,  -- Cache reload wait ack
                     STORE_WAIT_ACK,   -- Store wait ack
                     NC_LOAD_WAIT_ACK);-- Non-cachable load wait ack

    type reg_internal_t is record                        
        req_latch : Loadstore1ToDcacheType;
        
        -- Cache hit state (Latches for 1 cycle BRAM access)
        hit_way          : way_t;
        hit_load_valid   : std_ulogic;

        -- 1-cycle delayed signals to account for the BRAM extra
        -- buffer that seems necessary to make timing on load hits
        --
        hit_way_delayed        : way_t;
        hit_load_delayed       : std_ulogic;
        hit_load_upd_delayed   : std_ulogic;
        hit_load_reg_delayed   : std_ulogic_vector(4 downto 0);
        hit_data_shift_delayed : std_ulogic_vector(2 downto 0);
        hit_dlength_delayed    : std_ulogic_vector(3 downto 0);
        hit_sign_ext_delayed   : std_ulogic;
        hit_byte_rev_delayed   : std_ulogic;

        -- Register update (load/store with update)
        update_valid : std_ulogic;

        -- Data buffer for "slow" read ops (load miss and NC loads).
        slow_data    : std_ulogic_vector(63 downto 0);
        slow_valid   : std_ulogic;

        -- Cache miss state (reload state machine)
        state            : state_t;
        wb               : wishbone_master_out;
        store_way        : way_t;
        store_index      : index_t;
    end record;

    signal r : reg_internal_t;

    -- Async signals on incoming request
    signal req_index   : index_t;
    signal req_row     : row_t;
    signal req_hit_way : way_t;
    signal req_tag     : cache_tag_t;
    signal req_op      : op_t;

    -- Cache RAM interface
    type cache_ram_out_t is array(way_t) of cache_row_t;
    signal cache_out   : cache_ram_out_t;

    -- PLRU output interface
    type plru_out_t is array(index_t) of std_ulogic_vector(WAY_BITS-1 downto 0);
    signal plru_victim : plru_out_t;

    -- Wishbone read/write/cache write formatting signals
    signal bus_sel                  : wishbone_sel_type;
    signal store_data               : wishbone_data_type;
    
    -- Return the cache line index (tag index) for an address
    function get_index(addr: std_ulogic_vector(63 downto 0)) return index_t is
    begin
        return to_integer(unsigned(addr(63-TAG_BITS downto LINE_OFF_BITS)));
    end;

    -- Return the cache row index (data memory) for an address
    function get_row(addr: std_ulogic_vector(63 downto 0)) return row_t is
    begin
        return to_integer(unsigned(addr(63-TAG_BITS downto ROW_OFF_BITS)));
    end;

    -- Returns whether this is the last row of a line
    function is_last_row(addr: std_ulogic_vector(63 downto 0)) return boolean is
        constant ones : std_ulogic_vector(ROW_LINEBITS-1 downto 0) := (others => '1');
    begin
        return addr(LINE_OFF_BITS-1 downto ROW_OFF_BITS) = ones;
    end;

    -- Return the address of the next row in the current cache line
    function next_row_addr(addr: std_ulogic_vector(63 downto 0)) return std_ulogic_vector is
        variable row_idx : std_ulogic_vector(ROW_LINEBITS-1 downto 0);
        variable result  : std_ulogic_vector(63 downto 0);
    begin
        -- Is there no simpler way in VHDL to generate that 3 bits adder ?
        row_idx := addr(LINE_OFF_BITS-1 downto ROW_OFF_BITS);
        row_idx := std_ulogic_vector(unsigned(row_idx) + 1);
        result := addr;
        result(LINE_OFF_BITS-1 downto ROW_OFF_BITS) := row_idx;
        return result;
    end;

    -- Get the tag value from the address
    function get_tag(addr: std_ulogic_vector(63 downto 0)) return cache_tag_t is
    begin
        return addr(63 downto 64-TAG_BITS);
    end;

    -- Read a tag from a tag memory row
    function read_tag(way: way_t; tagset: cache_tags_set_t) return cache_tag_t is
    begin
        return tagset((way+1) * TAG_BITS - 1 downto way * TAG_BITS);
    end;

    -- Write a tag to tag memory row
    procedure write_tag(way: in way_t; tagset: inout cache_tags_set_t;
                        tag: cache_tag_t) is
    begin
        tagset((way+1) * TAG_BITS - 1 downto way * TAG_BITS) := tag;
    end;

    -- Generate byte enables from sizes
    function length_to_sel(length : in std_logic_vector(3 downto 0)) return std_ulogic_vector is
    begin
        case length is
            when "0001" =>
                return "00000001";
            when "0010" =>
                return "00000011";
            when "0100" =>
                return "00001111";
            when "1000" =>
                return "11111111";
            when others =>
                return "00000000";
        end case;
    end function length_to_sel;

    -- Calculate shift and byte enables for wishbone
    function wishbone_data_shift(address : in std_ulogic_vector(63 downto 0)) return natural is
    begin
        return to_integer(unsigned(address(2 downto 0))) * 8;
    end function wishbone_data_shift;

    function wishbone_data_sel(size : in std_logic_vector(3 downto 0);
                               address : in std_logic_vector(63 downto 0))
        return std_ulogic_vector is
    begin
        return std_ulogic_vector(shift_left(unsigned(length_to_sel(size)),
                                            to_integer(unsigned(address(2 downto 0)))));
    end function wishbone_data_sel;

begin

    assert LINE_SIZE mod ROW_SIZE = 0 report "LINE_SIZE not multiple of ROW_SIZE" severity FAILURE;
    assert ispow2(LINE_SIZE)    report "LINE_SIZE not power of 2" severity FAILURE;
    assert ispow2(NUM_LINES)    report "NUM_LINES not power of 2" severity FAILURE;
    assert ispow2(ROW_PER_LINE) report "ROW_PER_LINE not power of 2" severity FAILURE;
    assert (ROW_BITS = INDEX_BITS + ROW_LINEBITS)
        report "geometry bits don't add up" severity FAILURE;
    assert (LINE_OFF_BITS = ROW_OFF_BITS + ROW_LINEBITS)
        report "geometry bits don't add up" severity FAILURE;
    assert (64 = TAG_BITS + INDEX_BITS + LINE_OFF_BITS)
        report "geometry bits don't add up" severity FAILURE;
    assert (64 = TAG_BITS + ROW_BITS + ROW_OFF_BITS)
        report "geometry bits don't add up" severity FAILURE;
    assert (64 = wishbone_data_bits)
        report "Can't yet handle a wishbone width that isn't 64-bits" severity FAILURE;
    
    -- Generate PLRUs
    maybe_plrus: if NUM_WAYS > 1 generate
    begin
        plrus: for i in 0 to NUM_LINES-1 generate
            -- PLRU interface
            signal plru_acc    : std_ulogic_vector(WAY_BITS-1 downto 0);
            signal plru_acc_en : std_ulogic;
            signal plru_out    : std_ulogic_vector(WAY_BITS-1 downto 0);
            
        begin
            plru : entity work.plru
                generic map (
                    BITS => WAY_BITS
                    )
                port map (
                    clk => clk,
                    rst => rst,
                    acc => plru_acc,
                    acc_en => plru_acc_en,
                    lru => plru_out
                    );

            process(req_index, req_op, req_hit_way, plru_out)
            begin
                -- PLRU interface
                if (req_op = OP_LOAD_HIT or
                    req_op = OP_STORE_HIT) and req_index = i then
                    plru_acc_en <= '1';
                else
                    plru_acc_en <= '0';
                end if;
                plru_acc <= std_ulogic_vector(to_unsigned(req_hit_way, WAY_BITS));
                plru_victim(i) <= plru_out;
            end process;
        end generate;
    end generate;

    -- Cache request parsing and hit detection
    dcache_request : process(all)
        variable is_hit  : std_ulogic;
        variable hit_way : way_t;
        variable op      : op_t;
        variable tmp     : std_ulogic_vector(63 downto 0);
        variable data    : std_ulogic_vector(63 downto 0);
        variable opsel   : std_ulogic_vector(3 downto 0);
    begin
        -- Extract line, row and tag from request
        req_index <= get_index(d_in.addr);
        req_row <= get_row(d_in.addr);
        req_tag <= get_tag(d_in.addr);

        -- Test if pending request is a hit on any way
        hit_way := 0;
        is_hit := '0';
        for i in way_t loop
            if d_in.valid = '1' and cache_valids(req_index)(i) = '1' then
                if read_tag(i, cache_tags(req_index)) = req_tag then
                    hit_way := i;
                    is_hit := '1';
                end if;
            end if;
        end loop;

        -- The way that matched on a hit               
        req_hit_way <= hit_way;

        -- Combine the request and cache his status to decide what
        -- operation needs to be done
        --
        opsel := d_in.valid & d_in.load & d_in.nc & is_hit;
        case opsel is
        when "1101" => op := OP_LOAD_HIT;
        when "1100" => op := OP_LOAD_MISS;
        when "1110" => op := OP_LOAD_NC;
        when "1001" => op := OP_STORE_HIT;
        when "1000" => op := OP_STORE_MISS;
        when "1010" => op := OP_STORE_MISS;
        when "1011" => op := OP_BAD;
        when "1111" => op := OP_BAD;
        when others => op := OP_NONE;
        end case;

        req_op <= op;

        -- XXX GENERATE ERRORS
        -- err_nc_collision <= '1' when op = OP_BAD else '0';

        -- XXX Generate stalls
        -- stall_out <= r.state /= IDLE ?

    end process;

    -- Wire up wishbone request latch
    wishbone_out <= r.wb;

    -- Writeback (loads and reg updates) & completion control logic
    --
    writeback_control: process(all)
    begin

        -- The mux on d_out.write reg defaults to the normal load hit case.
        d_out.write_enable <= '0';
        d_out.valid <= '0';
        d_out.write_reg <= r.hit_load_reg_delayed;
        d_out.write_data <= cache_out(r.hit_way_delayed);
        d_out.write_len <= r.hit_dlength_delayed;
        d_out.write_shift <= r.hit_data_shift_delayed;
        d_out.sign_extend <= r.hit_sign_ext_delayed;
        d_out.byte_reverse <= r.hit_byte_rev_delayed;
        d_out.second_word <= '0';

        -- We have a valid load or store hit or we just completed a slow
        -- op such as a load miss, a NC load or a store
        --
        -- Note: the load hit is delayed by one cycle. However it can still
        -- not collide with r.slow_valid (well unless I miscalculated) because
        -- slow_valid can only be set on a subsequent request and not on its
        -- first cycle (the state machine must have advanced), which makes
        -- slow_valid at least 2 cycles from the previous hit_load_valid.
        --

        -- Sanity: Only one of these must be set in any given cycle
        assert (r.update_valid and r.hit_load_delayed) /= '1' report
            "unexpected hit_load_delayed collision with update_valid"
            severity FAILURE;
        assert (r.slow_valid and r.hit_load_delayed) /= '1' report
            "unexpected hit_load_delayed collision with slow_valid"
            severity FAILURE;
        assert (r.slow_valid and r.update_valid) /= '1' report
            "unexpected update_valid collision with slow_valid"
            severity FAILURE;

        -- Delayed load hit case is the standard path
        if r.hit_load_delayed = '1' then
            d_out.write_enable <= '1';

            -- If it's not a load with update, complete it now
            if r.hit_load_upd_delayed = '0' then
                d_out.valid <= '1';
                end if;
        end if;

        -- Slow ops (load miss, NC, stores)
        if r.slow_valid = '1' then
            -- If it's a load, enable register writeback and switch
            -- mux accordingly
            --
            if r.req_latch.load then        
                d_out.write_reg <= r.req_latch.write_reg;
                d_out.write_enable <= '1';

                -- Read data comes from the slow data latch, formatter
                -- from the latched request.
                --
                d_out.write_data <= r.slow_data;
                d_out.write_shift <= r.req_latch.addr(2 downto 0);
                d_out.sign_extend <= r.req_latch.sign_extend;
                d_out.byte_reverse <= r.req_latch.byte_reverse;
                d_out.write_len <= r.req_latch.length;
            end if;

            -- If it's a store or a non-update load form, complete now
            if r.req_latch.load = '0' or r.req_latch.update = '0' then
                d_out.valid <= '1';
            end if;
        end if;

        -- We have a register update to do.
        if r.update_valid = '1' then
            d_out.write_enable <= '1';
            d_out.write_reg <= r.req_latch.update_reg;

            -- Change the read data mux to the address that's going into
            -- the register and the formatter does nothing.
            --
            d_out.write_data <= r.req_latch.addr;
            d_out.write_shift <= "000";
            d_out.write_len <= "1000";
            d_out.sign_extend <= '0';
            d_out.byte_reverse <= '0';

            -- If it was a load, this completes the operation (load with
            -- update case).
            --
            if r.req_latch.load = '1' then
                d_out.valid <= '1';
            end if;
        end if;

    end process;

    -- Misc data & sel signals
    misc: process(d_in)
    begin
        -- Wishbone & BRAM write data formatting for stores (most of it already
        -- happens in loadstore1, this is the remaining sel generation and shifting)
        --
        store_data  <= std_logic_vector(shift_left(unsigned(d_in.data),
                                                   wishbone_data_shift(d_in.addr)));

        -- Wishbone read and write and BRAM write sel bits generation
        bus_sel     <= wishbone_data_sel(d_in.length, d_in.addr);
    end process;

    -- Generate a cache RAM for each way. This handles the normal
    -- reads, writes from reloads and the special store-hit update
    -- path as well.
    --
    -- Note: the BRAMs have an extra read buffer, meaning the output
    --       is pipelined an extra cycle. This differs from the
    --       icache. The writeback logic needs to take that into
    --       account by using 1-cycle delayed signals for load hits.
    --
    rams: for i in 0 to NUM_WAYS-1 generate
        signal do_read  : std_ulogic;
        signal rd_addr  : std_ulogic_vector(ROW_BITS-1 downto 0);
        signal do_write : std_ulogic;
        signal wr_addr  : std_ulogic_vector(ROW_BITS-1 downto 0);
        signal wr_data  : std_ulogic_vector(wishbone_data_bits-1 downto 0);
        signal wr_sel   : std_ulogic_vector(ROW_SIZE-1 downto 0);
        signal dout     : cache_row_t;
    begin
        way: entity work.cache_ram
            generic map (
                ROW_BITS => ROW_BITS,
                WIDTH => wishbone_data_bits,
                ADD_BUF => true
                )
            port map (
                clk     => clk,
                rd_en   => do_read,
                rd_addr => rd_addr,
                rd_data => dout,
                wr_en   => do_write,
                wr_sel  => wr_sel,
                wr_addr => wr_addr,
                wr_data => wr_data
                );
        process(all)
        begin
            -- Cache hit reads
            do_read <= '1';
            rd_addr <= std_ulogic_vector(to_unsigned(req_row, ROW_BITS));
            cache_out(i) <= dout;

            -- Write mux:
            --
            -- Defaults to wishbone read responses (cache refill),
            --
            -- For timing, the mux on wr_data/sel/addr is not dependent on anything
            -- other than the current state. Only the do_write signal is.
            --
            if r.state = IDLE then
                -- When IDLE, the only write path is the store-hit update case
                wr_addr  <= std_ulogic_vector(to_unsigned(req_row, ROW_BITS));
                wr_data  <= store_data;
                wr_sel   <= bus_sel;
            else
                -- Otherwise, we might be doing a reload
                wr_data <= wishbone_in.dat;
                wr_sel  <= (others => '1');
                wr_addr <= std_ulogic_vector(to_unsigned(get_row(r.wb.adr), ROW_BITS));
            end if;

            -- The two actual write cases here
            do_write <= '0';
            if r.state = RELOAD_WAIT_ACK and wishbone_in.ack = '1' and r.store_way = i then
                do_write <= '1';
            end if;
            if req_op = OP_STORE_HIT and req_hit_way = i then
                assert r.state /= RELOAD_WAIT_ACK report "Store hit while in state:" &
                    state_t'image(r.state)
                    severity FAILURE;
                do_write <= '1';
            end if;
        end process;
    end generate;

    -- Cache hit synchronous machine for the easy case. This handles
    -- non-update form load hits.
    --
    dcache_fast_hit : process(clk)
    begin
        if rising_edge(clk) then
            -- 1-cycle delayed signals for load hit response
            r.hit_load_delayed       <= r.hit_load_valid;
            r.hit_way_delayed        <= r.hit_way;
            r.hit_load_upd_delayed   <= r.req_latch.update;
            r.hit_load_reg_delayed   <= r.req_latch.write_reg;
            r.hit_data_shift_delayed <= r.req_latch.addr(2 downto 0);
            r.hit_sign_ext_delayed   <= r.req_latch.sign_extend;
            r.hit_byte_rev_delayed   <= r.req_latch.byte_reverse;
            r.hit_dlength_delayed    <= r.req_latch.length;

            -- On-cycle pulse values get reset on every cycle
            r.hit_load_valid <= '0';

            -- If we have a request incoming, we have to latch it as d_in.valid
            -- is only set for a single cycle. It's up to the control logic to
            -- ensure we don't override an uncompleted request (for now we are
            -- single issue on load/stores so we are fine, later, we can generate
            -- a stall output if necessary).

            if d_in.valid = '1' then
                r.req_latch <= d_in;

                report "op:" & op_t'image(req_op) &
                    " addr:" & to_hstring(d_in.addr) &
                    " upd:" & std_ulogic'image(d_in.update) &
                    " nc:" & std_ulogic'image(d_in.nc) &
                    " reg:" & to_hstring(d_in.write_reg) &
                    " idx:" & integer'image(req_index) &
                    " tag:" & to_hstring(req_tag) &
                    " way: " & integer'image(req_hit_way);
            end if;

            -- Fast path for load/store hits. Set signals for the writeback controls.
            if req_op = OP_LOAD_HIT then
                r.hit_way <= req_hit_way;
                r.hit_load_valid <= '1';
            end if;
        end if;
    end process;

    -- Every other case is handled by this stage machine:
    --
    --   * Cache load miss/reload (in conjunction with "rams")
    --   * Load hits for update forms
    --   * Load hits for non-cachable forms
    --   * Stores (the collision case is handled in "rams")
    --
    -- All wishbone requests generation is done here
    --
    dcache_slow : process(clk)
        variable way : integer range 0 to NUM_WAYS-1;
        variable tagset : cache_tags_set_t;
    begin
        if rising_edge(clk) then
            -- On reset, clear all valid bits to force misses
            if rst = '1' then
                for i in index_t loop
                    cache_valids(i) <= (others => '0');
                end loop;
                r.state <= IDLE;
                r.slow_valid <= '0';
                r.update_valid <= '0';
                r.wb.cyc <= '0';
                r.wb.stb <= '0';

                -- Not useful normally but helps avoiding tons of sim warnings
                r.wb.adr <= (others => '0');
            else
                -- One cycle pulses reset
                r.slow_valid <= '0';
                r.update_valid <= '0';

                -- We cannot currently process a new request when not idle
                assert req_op = OP_NONE or r.state = IDLE report "request " &
                    op_t'image(req_op) & " while in state " & state_t'image(r.state)
                    severity FAILURE;

                -- Main state machine
                case r.state is
                when IDLE =>
                    case req_op is
                    when OP_LOAD_HIT =>
                        -- We have a load with update hit, we need the delayed update cycle
                        if d_in.update = '1' then
                            r.state <= LOAD_UPDATE;
                        end if;

                    when OP_LOAD_MISS =>
                        -- Normal load cache miss, start the reload machine
                        --
                        -- First find a victim way from the PLRU
                        --
                        way := to_integer(unsigned(plru_victim(req_index)));

                        report "cache miss addr:" & to_hstring(d_in.addr) &
                            " idx:" & integer'image(req_index) &
                            " way:" & integer'image(way) &
                            " tag:" & to_hstring(req_tag);

                        -- Force misses on that way while reloading that line
                        cache_valids(req_index)(way) <= '0';

                        -- Store new tag in selected way
                        for i in 0 to NUM_WAYS-1 loop
                            if i = way then
                                tagset := cache_tags(req_index);
                                write_tag(i, tagset, req_tag);
                                cache_tags(req_index) <= tagset;
                            end if;
                        end loop;

                        -- Keep track of our index and way for subsequent stores.
                        r.store_index <= req_index;
                        r.store_way <= way;

                        -- Prep for first wishbone read. We calculate the address of
                        -- the start of the cache line
                        --
                        r.wb.adr <= d_in.addr(63 downto LINE_OFF_BITS) &
                                    (LINE_OFF_BITS-1 downto 0 => '0');
                        r.wb.sel <= (others => '1');
                        r.wb.we  <= '0';
                        r.wb.cyc <= '1';
                        r.wb.stb <= '1';
                        r.state <= RELOAD_WAIT_ACK;

                    when OP_LOAD_NC =>
                        r.wb.sel <= bus_sel;
                        r.wb.adr <= d_in.addr(63 downto 3) & "000";
                        r.wb.cyc <= '1';
                        r.wb.stb <= '1';
                        r.wb.we <= '0';
                        r.state <= NC_LOAD_WAIT_ACK;

                    when OP_STORE_HIT | OP_STORE_MISS =>
                        -- For store-with-update do the register update
                        if d_in.update = '1' then
                            r.update_valid <= '1';
                        end if;
                        r.wb.sel <= bus_sel;
                        r.wb.adr <= d_in.addr(63 downto 3) & "000";
                        r.wb.dat <= store_data;
                        r.wb.cyc <= '1';
                        r.wb.stb <= '1';
                        r.wb.we <= '1';
                        r.state <= STORE_WAIT_ACK;

                    -- OP_NONE and OP_BAD do nothing
                    when OP_NONE =>
                    when OP_BAD =>
                    end case;

                when RELOAD_WAIT_ACK =>
                    if wishbone_in.ack = '1' then
                        -- Is this the data we were looking for ? Latch it so
                        -- we can respond later. We don't currently complete the
                        -- pending miss request immediately, we wait for the
                        -- whole line to be loaded. The reason is that if we
                        -- did, we would potentially get new requests in while
                        -- not idle, which we don't currently know how to deal
                        -- with.
                        --
                        if r.wb.adr(LINE_OFF_BITS-1 downto ROW_OFF_BITS) =
                            r.req_latch.addr(LINE_OFF_BITS-1 downto ROW_OFF_BITS) then
                            r.slow_data <= wishbone_in.dat;
                        end if;

                        -- That was the last word ? We are done
                        if is_last_row(r.wb.adr) then
                            cache_valids(r.store_index)(way) <= '1';
                            r.wb.cyc <= '0';
                            r.wb.stb <= '0';

                            -- Complete the load that missed. For load with update
                            -- we also need to do the deferred update cycle.
                            --
                            r.slow_valid <= '1';
                            if r.req_latch.load = '1' and r.req_latch.update = '1' then
                                r.state <= LOAD_UPDATE;
                                report "completing miss with load-update !";
                            else
                                r.state <= IDLE;
                                report "completing miss !";
                            end if;
                        else
                            -- Otherwise, calculate the next row address
                            r.wb.adr <= next_row_addr(r.wb.adr);
                        end if;
                    end if;

                when LOAD_UPDATE =>
                    -- We need the extra cycle to complete a load with update
                    r.state <= LOAD_UPDATE2;
                when LOAD_UPDATE2 =>
                    -- We need the extra cycle to complete a load with update
                    r.update_valid <= '1';
                    r.state <= IDLE;

                when STORE_WAIT_ACK | NC_LOAD_WAIT_ACK =>
                    if wishbone_in.ack = '1' then
                        if r.state = NC_LOAD_WAIT_ACK then
                            r.slow_data <= wishbone_in.dat;
                        end if;
                        r.slow_valid <= '1';
                        r.wb.cyc <= '0';
                        r.wb.stb <= '0';
                        r.state <= IDLE;
                    end if;
                end case;
            end if;
        end if;
    end process;
end;