1 |
61 |
zero_gravi |
-- #################################################################################################
|
2 |
|
|
-- # << NEORV32 - CPU Co-Processor: Shifter (CPU Core ISA) >> #
|
3 |
|
|
-- # ********************************************************************************************* #
|
4 |
|
|
-- # Bit-shift unit for base ISA. #
|
5 |
|
|
-- # FAST_SHIFT_EN = false (default): Use bit-serial shifter architecture (small but slow) #
|
6 |
|
|
-- # FAST_SHIFT_EN = true: Use barrel shifter architecture (large but fast) #
|
7 |
|
|
-- # ********************************************************************************************* #
|
8 |
|
|
-- # BSD 3-Clause License #
|
9 |
|
|
-- # #
|
10 |
|
|
-- # Copyright (c) 2021, Stephan Nolting. All rights reserved. #
|
11 |
|
|
-- # #
|
12 |
|
|
-- # Redistribution and use in source and binary forms, with or without modification, are #
|
13 |
|
|
-- # permitted provided that the following conditions are met: #
|
14 |
|
|
-- # #
|
15 |
|
|
-- # 1. Redistributions of source code must retain the above copyright notice, this list of #
|
16 |
|
|
-- # conditions and the following disclaimer. #
|
17 |
|
|
-- # #
|
18 |
|
|
-- # 2. Redistributions in binary form must reproduce the above copyright notice, this list of #
|
19 |
|
|
-- # conditions and the following disclaimer in the documentation and/or other materials #
|
20 |
|
|
-- # provided with the distribution. #
|
21 |
|
|
-- # #
|
22 |
|
|
-- # 3. Neither the name of the copyright holder nor the names of its contributors may be used to #
|
23 |
|
|
-- # endorse or promote products derived from this software without specific prior written #
|
24 |
|
|
-- # permission. #
|
25 |
|
|
-- # #
|
26 |
|
|
-- # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS #
|
27 |
|
|
-- # OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF #
|
28 |
|
|
-- # MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE #
|
29 |
|
|
-- # COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, #
|
30 |
|
|
-- # EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE #
|
31 |
|
|
-- # GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED #
|
32 |
|
|
-- # AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING #
|
33 |
|
|
-- # NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED #
|
34 |
|
|
-- # OF THE POSSIBILITY OF SUCH DAMAGE. #
|
35 |
|
|
-- # ********************************************************************************************* #
|
36 |
|
|
-- # The NEORV32 Processor - https://github.com/stnolting/neorv32 (c) Stephan Nolting #
|
37 |
|
|
-- #################################################################################################
|
38 |
|
|
|
39 |
|
|
library ieee;
|
40 |
|
|
use ieee.std_logic_1164.all;
|
41 |
|
|
use ieee.numeric_std.all;
|
42 |
|
|
|
43 |
|
|
library neorv32;
|
44 |
|
|
use neorv32.neorv32_package.all;
|
45 |
|
|
|
46 |
|
|
entity neorv32_cpu_cp_shifter is
|
47 |
|
|
generic (
|
48 |
|
|
FAST_SHIFT_EN : boolean := false -- use barrel shifter for shift operations
|
49 |
|
|
);
|
50 |
|
|
port (
|
51 |
|
|
-- global control --
|
52 |
|
|
clk_i : in std_ulogic; -- global clock, rising edge
|
53 |
|
|
rstn_i : in std_ulogic; -- global reset, low-active, async
|
54 |
|
|
ctrl_i : in std_ulogic_vector(ctrl_width_c-1 downto 0); -- main control bus
|
55 |
|
|
start_i : in std_ulogic; -- trigger operation
|
56 |
|
|
-- data input --
|
57 |
|
|
rs1_i : in std_ulogic_vector(data_width_c-1 downto 0); -- rf source 1
|
58 |
|
|
rs2_i : in std_ulogic_vector(data_width_c-1 downto 0); -- rf source 2
|
59 |
|
|
imm_i : in std_ulogic_vector(data_width_c-1 downto 0); -- immediate
|
60 |
|
|
-- result and status --
|
61 |
|
|
res_o : out std_ulogic_vector(data_width_c-1 downto 0); -- operation result
|
62 |
|
|
valid_o : out std_ulogic -- data output valid
|
63 |
|
|
);
|
64 |
|
|
end neorv32_cpu_cp_shifter;
|
65 |
|
|
|
66 |
|
|
architecture neorv32_cpu_cp_shifter_rtl of neorv32_cpu_cp_shifter is
|
67 |
|
|
|
68 |
|
|
-- operands --
|
69 |
|
|
signal shift_amount : std_ulogic_vector(index_size_f(data_width_c)-1 downto 0);
|
70 |
|
|
|
71 |
|
|
-- serial shifter --
|
72 |
|
|
type shifter_t is record
|
73 |
|
|
busy : std_ulogic;
|
74 |
|
|
busy_ff : std_ulogic;
|
75 |
|
|
done : std_ulogic;
|
76 |
|
|
cnt : std_ulogic_vector(index_size_f(data_width_c)-1 downto 0);
|
77 |
|
|
sreg : std_ulogic_vector(data_width_c-1 downto 0);
|
78 |
|
|
end record;
|
79 |
|
|
signal shifter : shifter_t;
|
80 |
|
|
|
81 |
|
|
-- barrel shifter --
|
82 |
|
|
type bs_level_t is array (index_size_f(data_width_c) downto 0) of std_ulogic_vector(data_width_c-1 downto 0);
|
83 |
|
|
signal bs_level : bs_level_t;
|
84 |
|
|
signal bs_result : std_ulogic_vector(data_width_c-1 downto 0);
|
85 |
|
|
|
86 |
|
|
begin
|
87 |
|
|
|
88 |
|
|
-- Shift Amount ---------------------------------------------------------------------------
|
89 |
|
|
-- -------------------------------------------------------------------------------------------
|
90 |
|
|
shift_amount <= imm_i(index_size_f(data_width_c)-1 downto 0) when (ctrl_i(ctrl_alu_opb_mux_c) = '1') else -- immediate source
|
91 |
|
|
rs2_i(index_size_f(data_width_c)-1 downto 0); -- register source
|
92 |
|
|
|
93 |
|
|
|
94 |
|
|
-- Iterative Shifter Core (small but slow) ------------------------------------------------
|
95 |
|
|
-- -------------------------------------------------------------------------------------------
|
96 |
|
|
serial_shifter_sync:
|
97 |
|
|
if (FAST_SHIFT_EN = false) generate
|
98 |
|
|
shifter_unit_sync: process(rstn_i, clk_i)
|
99 |
|
|
begin
|
100 |
|
|
if (rstn_i = '0') then
|
101 |
|
|
shifter.busy <= '0';
|
102 |
|
|
shifter.busy_ff <= def_rst_val_c;
|
103 |
|
|
shifter.sreg <= (others => def_rst_val_c);
|
104 |
|
|
shifter.cnt <= (others => def_rst_val_c);
|
105 |
|
|
elsif rising_edge(clk_i) then
|
106 |
|
|
shifter.busy_ff <= shifter.busy;
|
107 |
|
|
if (start_i = '1') then
|
108 |
|
|
shifter.busy <= '1';
|
109 |
|
|
elsif (shifter.done = '1') then
|
110 |
|
|
shifter.busy <= '0';
|
111 |
|
|
end if;
|
112 |
|
|
--
|
113 |
|
|
if (start_i = '1') then -- trigger new shift
|
114 |
|
|
shifter.sreg <= rs1_i; -- shift operand
|
115 |
|
|
shifter.cnt <= shift_amount; -- shift amount
|
116 |
|
|
elsif (or_reduce_f(shifter.cnt) = '1') then -- running shift (cnt != 0)
|
117 |
|
|
shifter.cnt <= std_ulogic_vector(unsigned(shifter.cnt) - 1);
|
118 |
|
|
if (ctrl_i(ctrl_alu_shift_dir_c) = '0') then -- SLL: shift left logical
|
119 |
|
|
shifter.sreg <= shifter.sreg(shifter.sreg'left-1 downto 0) & '0';
|
120 |
|
|
else -- SRL: shift right logical / SRA: shift right arithmetical
|
121 |
|
|
shifter.sreg <= (shifter.sreg(shifter.sreg'left) and ctrl_i(ctrl_alu_shift_ar_c)) & shifter.sreg(shifter.sreg'left downto 1);
|
122 |
|
|
end if;
|
123 |
|
|
end if;
|
124 |
|
|
end if;
|
125 |
|
|
end process shifter_unit_sync;
|
126 |
|
|
end generate;
|
127 |
|
|
|
128 |
|
|
-- shift control --
|
129 |
|
|
serial_shifter_ctrl:
|
130 |
|
|
if (FAST_SHIFT_EN = false) generate
|
131 |
|
|
shifter.done <= not or_reduce_f(shifter.cnt(shifter.cnt'left downto 1));
|
132 |
|
|
valid_o <= shifter.busy and shifter.done;
|
133 |
|
|
res_o <= shifter.sreg when (shifter.busy = '0') and (shifter.busy_ff = '1') else (others => '0');
|
134 |
|
|
end generate;
|
135 |
|
|
|
136 |
|
|
|
137 |
|
|
-- Barrel Shifter Core (fast but large) ---------------------------------------------------
|
138 |
|
|
-- -------------------------------------------------------------------------------------------
|
139 |
|
|
barrel_shifter_async:
|
140 |
|
|
if (FAST_SHIFT_EN = true) generate
|
141 |
|
|
shifter_unit_async: process(rs1_i, shift_amount, ctrl_i, bs_level)
|
142 |
|
|
begin
|
143 |
|
|
-- input level: convert left shifts to right shifts --
|
144 |
|
|
if (ctrl_i(ctrl_alu_shift_dir_c) = '0') then -- is left shift?
|
145 |
|
|
bs_level(index_size_f(data_width_c)) <= bit_rev_f(rs1_i); -- reverse bit order of input operand
|
146 |
|
|
else
|
147 |
|
|
bs_level(index_size_f(data_width_c)) <= rs1_i;
|
148 |
|
|
end if;
|
149 |
|
|
|
150 |
|
|
-- shifter array --
|
151 |
|
|
for i in index_size_f(data_width_c)-1 downto 0 loop
|
152 |
|
|
if (shift_amount(i) = '1') then
|
153 |
|
|
bs_level(i)(data_width_c-1 downto data_width_c-(2**i)) <= (others => (bs_level(i+1)(data_width_c-1) and ctrl_i(ctrl_alu_shift_ar_c)));
|
154 |
|
|
bs_level(i)((data_width_c-(2**i))-1 downto 0) <= bs_level(i+1)(data_width_c-1 downto 2**i);
|
155 |
|
|
else
|
156 |
|
|
bs_level(i) <= bs_level(i+1);
|
157 |
|
|
end if;
|
158 |
|
|
end loop;
|
159 |
|
|
|
160 |
|
|
-- re-convert original left shifts --
|
161 |
|
|
if (ctrl_i(ctrl_alu_shift_dir_c) = '0') then
|
162 |
|
|
bs_result <= bit_rev_f(bs_level(0));
|
163 |
|
|
else
|
164 |
|
|
bs_result <= bs_level(0);
|
165 |
|
|
end if;
|
166 |
|
|
end process shifter_unit_async;
|
167 |
|
|
end generate;
|
168 |
|
|
|
169 |
|
|
-- output register --
|
170 |
|
|
barrel_shifter_sync:
|
171 |
|
|
if (FAST_SHIFT_EN = true) generate
|
172 |
|
|
shifter_unit_sync: process(clk_i)
|
173 |
|
|
begin
|
174 |
|
|
if rising_edge(clk_i) then
|
175 |
|
|
res_o <= (others => '0');
|
176 |
|
|
if (start_i = '1') then
|
177 |
|
|
res_o <= bs_result;
|
178 |
|
|
end if;
|
179 |
|
|
end if;
|
180 |
|
|
end process shifter_unit_sync;
|
181 |
|
|
end generate;
|
182 |
|
|
|
183 |
|
|
-- shift control --
|
184 |
|
|
barrel_shifter_ctrl:
|
185 |
|
|
if (FAST_SHIFT_EN = true) generate
|
186 |
|
|
valid_o <= start_i;
|
187 |
|
|
end generate;
|
188 |
|
|
|
189 |
|
|
|
190 |
|
|
end neorv32_cpu_cp_shifter_rtl;
|