URL
https://opencores.org/ocsvn/openrisc/openrisc/trunk
Subversion Repositories openrisc
[/] [openrisc/] [trunk/] [or1200/] [rtl/] [verilog/] [or1200_fpu_post_norm_mul.v] - Rev 463
Go to most recent revision | Compare with Previous | Blame | View Log
////////////////////////////////////////////////////////////////////// //// //// //// or1200_fpu_post_norm_mul //// //// //// //// This file is part of the OpenRISC 1200 project //// //// http://opencores.org/project,or1k //// //// //// //// Description //// //// post-normalization entity for the multiplication unit //// //// //// //// To Do: //// //// //// //// //// //// Author(s): //// //// - Original design (FPU100) - //// //// Jidan Al-eryani, jidan@gmx.net //// //// - Conv. to Verilog and inclusion in OR1200 - //// //// Julius Baxter, julius@opencores.org //// //// //// ////////////////////////////////////////////////////////////////////// // // Copyright (C) 2006, 2010 // // This source file may be used and distributed without // restriction provided that this copyright statement is not // removed from the file and that any derivative work contains // the original copyright notice and the associated disclaimer. // // THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY // EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED // TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS // FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL THE AUTHOR // OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, // INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES // (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE // GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR // BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF // LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT // OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE // POSSIBILITY OF SUCH DAMAGE. // module or1200_fpu_post_norm_mul( clk_i, opa_i, opb_i, exp_10_i, fract_48_i, sign_i, rmode_i, output_o, ine_o ); parameter FP_WIDTH = 32; parameter MUL_SERIAL = 0; // 0 for parallel multiplier, 1 for serial parameter MUL_COUNT = 11; //11 for parallel multiplier, 34 for serial parameter FRAC_WIDTH = 23; parameter EXP_WIDTH = 8; parameter ZERO_VECTOR = 31'd0; parameter INF = 31'b1111111100000000000000000000000; parameter QNAN = 31'b1111111110000000000000000000000; parameter SNAN = 31'b1111111100000000000000000000001; input clk_i; input [FP_WIDTH-1:0] opa_i; input [FP_WIDTH-1:0] opb_i; input [EXP_WIDTH+1:0] exp_10_i; input [2*FRAC_WIDTH+1:0] fract_48_i; input sign_i; input [1:0] rmode_i; output reg [FP_WIDTH-1:0] output_o; output reg ine_o; reg [EXP_WIDTH-1:0] s_expa; reg [EXP_WIDTH-1:0] s_expb; reg [EXP_WIDTH+1:0] s_exp_10_i; reg [2*FRAC_WIDTH+1:0] s_fract_48_i; reg s_sign_i; wire [FP_WIDTH-1:0] s_output_o; wire s_ine_o; wire s_overflow; reg [FP_WIDTH-1:0] s_opa_i; reg [FP_WIDTH-1:0] s_opb_i; reg [1:0] s_rmode_i; reg [5:0] s_zeros; wire s_carry; reg [5:0] s_shr2; reg [5:0] s_shl2; reg [8:0] s_expo1; wire [8:0] s_expo2b; wire [9:0] s_exp_10a; wire [9:0] s_exp_10b; reg [47:0] s_frac2a; wire s_sticky, s_guard, s_round; wire s_roundup; reg [24:0] s_frac_rnd; wire [24:0] s_frac3; wire s_shr3; reg [5:0] s_r_zeros; wire s_lost; wire s_op_0; wire [8:0] s_expo3; wire s_infa, s_infb; wire s_nan_in, s_nan_op, s_nan_a, s_nan_b; // Input Register always @(posedge clk_i) begin s_opa_i <= opa_i; s_opb_i <= opb_i; s_expa <= opa_i[30:23]; s_expb <= opb_i[30:23]; s_exp_10_i <= exp_10_i; s_fract_48_i <= fract_48_i; s_sign_i <= sign_i; s_rmode_i <= rmode_i; end // Output register always @(posedge clk_i) begin output_o <= s_output_o; ine_o <= s_ine_o; end //*** Stage 1 **** // figure out the exponent and howmuch the fraction has to be shiftd // right/left assign s_carry = s_fract_48_i[47]; always @(posedge clk_i) if (!s_fract_48_i[47]) casez(s_fract_48_i[46:1]) // synopsys full_case parallel_case 46'b1?????????????????????????????????????????????: s_zeros <= 0; 46'b01????????????????????????????????????????????: s_zeros <= 1; 46'b001???????????????????????????????????????????: s_zeros <= 2; 46'b0001??????????????????????????????????????????: s_zeros <= 3; 46'b00001?????????????????????????????????????????: s_zeros <= 4; 46'b000001????????????????????????????????????????: s_zeros <= 5; 46'b0000001???????????????????????????????????????: s_zeros <= 6; 46'b00000001??????????????????????????????????????: s_zeros <= 7; 46'b000000001?????????????????????????????????????: s_zeros <= 8; 46'b0000000001????????????????????????????????????: s_zeros <= 9; 46'b00000000001???????????????????????????????????: s_zeros <= 10; 46'b000000000001??????????????????????????????????: s_zeros <= 11; 46'b0000000000001?????????????????????????????????: s_zeros <= 12; 46'b00000000000001????????????????????????????????: s_zeros <= 13; 46'b000000000000001???????????????????????????????: s_zeros <= 14; 46'b0000000000000001??????????????????????????????: s_zeros <= 15; 46'b00000000000000001?????????????????????????????: s_zeros <= 16; 46'b000000000000000001????????????????????????????: s_zeros <= 17; 46'b0000000000000000001???????????????????????????: s_zeros <= 18; 46'b00000000000000000001??????????????????????????: s_zeros <= 19; 46'b000000000000000000001?????????????????????????: s_zeros <= 20; 46'b0000000000000000000001????????????????????????: s_zeros <= 21; 46'b00000000000000000000001???????????????????????: s_zeros <= 22; 46'b000000000000000000000001??????????????????????: s_zeros <= 23; 46'b0000000000000000000000001?????????????????????: s_zeros <= 24; 46'b00000000000000000000000001????????????????????: s_zeros <= 25; 46'b000000000000000000000000001???????????????????: s_zeros <= 26; 46'b0000000000000000000000000001??????????????????: s_zeros <= 27; 46'b00000000000000000000000000001?????????????????: s_zeros <= 28; 46'b000000000000000000000000000001????????????????: s_zeros <= 29; 46'b0000000000000000000000000000001???????????????: s_zeros <= 30; 46'b00000000000000000000000000000001??????????????: s_zeros <= 31; 46'b000000000000000000000000000000001?????????????: s_zeros <= 32; 46'b0000000000000000000000000000000001????????????: s_zeros <= 33; 46'b00000000000000000000000000000000001???????????: s_zeros <= 34; 46'b000000000000000000000000000000000001??????????: s_zeros <= 35; 46'b0000000000000000000000000000000000001?????????: s_zeros <= 36; 46'b00000000000000000000000000000000000001????????: s_zeros <= 37; 46'b000000000000000000000000000000000000001???????: s_zeros <= 38; 46'b0000000000000000000000000000000000000001??????: s_zeros <= 39; 46'b00000000000000000000000000000000000000001?????: s_zeros <= 40; 46'b000000000000000000000000000000000000000001????: s_zeros <= 41; 46'b0000000000000000000000000000000000000000001???: s_zeros <= 42; 46'b00000000000000000000000000000000000000000001??: s_zeros <= 43; 46'b000000000000000000000000000000000000000000001?: s_zeros <= 44; 46'b0000000000000000000000000000000000000000000001: s_zeros <= 45; 46'b0000000000000000000000000000000000000000000000: s_zeros <= 46; endcase // casex (s_fract_48_i[46:1]) else s_zeros <= 0; always @(posedge clk_i) casez(s_fract_48_i) // synopsys full_case parallel_case 48'b???????????????????????????????????????????????1: s_r_zeros <= 0; 48'b??????????????????????????????????????????????10: s_r_zeros <= 1; 48'b?????????????????????????????????????????????100: s_r_zeros <= 2; 48'b????????????????????????????????????????????1000: s_r_zeros <= 3; 48'b???????????????????????????????????????????10000: s_r_zeros <= 4; 48'b??????????????????????????????????????????100000: s_r_zeros <= 5; 48'b?????????????????????????????????????????1000000: s_r_zeros <= 6; 48'b????????????????????????????????????????10000000: s_r_zeros <= 7; 48'b???????????????????????????????????????100000000: s_r_zeros <= 8; 48'b??????????????????????????????????????1000000000: s_r_zeros <= 9; 48'b?????????????????????????????????????10000000000: s_r_zeros <= 10; 48'b????????????????????????????????????100000000000: s_r_zeros <= 11; 48'b???????????????????????????????????1000000000000: s_r_zeros <= 12; 48'b??????????????????????????????????10000000000000: s_r_zeros <= 13; 48'b?????????????????????????????????100000000000000: s_r_zeros <= 14; 48'b????????????????????????????????1000000000000000: s_r_zeros <= 15; 48'b???????????????????????????????10000000000000000: s_r_zeros <= 16; 48'b??????????????????????????????100000000000000000: s_r_zeros <= 17; 48'b?????????????????????????????1000000000000000000: s_r_zeros <= 18; 48'b????????????????????????????10000000000000000000: s_r_zeros <= 19; 48'b???????????????????????????100000000000000000000: s_r_zeros <= 20; 48'b??????????????????????????1000000000000000000000: s_r_zeros <= 21; 48'b?????????????????????????10000000000000000000000: s_r_zeros <= 22; 48'b????????????????????????100000000000000000000000: s_r_zeros <= 23; 48'b???????????????????????1000000000000000000000000: s_r_zeros <= 24; 48'b??????????????????????10000000000000000000000000: s_r_zeros <= 25; 48'b?????????????????????100000000000000000000000000: s_r_zeros <= 26; 48'b????????????????????1000000000000000000000000000: s_r_zeros <= 27; 48'b???????????????????10000000000000000000000000000: s_r_zeros <= 28; 48'b??????????????????100000000000000000000000000000: s_r_zeros <= 29; 48'b?????????????????1000000000000000000000000000000: s_r_zeros <= 30; 48'b????????????????10000000000000000000000000000000: s_r_zeros <= 31; 48'b???????????????100000000000000000000000000000000: s_r_zeros <= 32; 48'b??????????????1000000000000000000000000000000000: s_r_zeros <= 33; 48'b?????????????10000000000000000000000000000000000: s_r_zeros <= 34; 48'b????????????100000000000000000000000000000000000: s_r_zeros <= 35; 48'b???????????1000000000000000000000000000000000000: s_r_zeros <= 36; 48'b??????????10000000000000000000000000000000000000: s_r_zeros <= 37; 48'b?????????100000000000000000000000000000000000000: s_r_zeros <= 38; 48'b????????1000000000000000000000000000000000000000: s_r_zeros <= 39; 48'b???????10000000000000000000000000000000000000000: s_r_zeros <= 40; 48'b??????100000000000000000000000000000000000000000: s_r_zeros <= 41; 48'b?????1000000000000000000000000000000000000000000: s_r_zeros <= 42; 48'b????10000000000000000000000000000000000000000000: s_r_zeros <= 43; 48'b???100000000000000000000000000000000000000000000: s_r_zeros <= 44; 48'b??1000000000000000000000000000000000000000000000: s_r_zeros <= 45; 48'b?10000000000000000000000000000000000000000000000: s_r_zeros <= 46; 48'b100000000000000000000000000000000000000000000000: s_r_zeros <= 47; 48'b000000000000000000000000000000000000000000000000: s_r_zeros <= 48; endcase // casex (s_fract_48_i) assign s_exp_10a = s_exp_10_i + {9'd0,s_carry}; assign s_exp_10b = s_exp_10a - {4'd0,s_zeros}; wire [9:0] v_shr1; wire [9:0] v_shl1; assign v_shr1 = (s_exp_10a[9] | !(|s_exp_10a)) ? 10'd1 - s_exp_10a + {9'd0,s_carry} : (s_exp_10b[9] | !(|s_exp_10b)) ? 0 : s_exp_10b[8] ? 0 : {9'd0,s_carry}; assign v_shl1 = (s_exp_10a[9] | !(|s_exp_10a)) ? 0 : (s_exp_10b[9] | !(|s_exp_10b)) ? {4'd0,s_zeros} - s_exp_10a : s_exp_10b[8] ? 0 : {4'd0,s_zeros}; always @(posedge clk_i) begin if ((s_exp_10a[9] | !(|s_exp_10a))) s_expo1 <= 9'd1; else if (s_exp_10b[9] | !(|s_exp_10b)) s_expo1 <= 9'd1; else if (s_exp_10b[8]) s_expo1 <= 9'b011111111; else s_expo1 <= s_exp_10b[8:0]; if (v_shr1[6]) s_shr2 <= {6{1'b1}}; else s_shr2 <= v_shr1[5:0]; s_shl2 <= v_shl1[5:0]; end // always @ (posedge clk_i) // *** Stage 2 *** // Shifting the fraction and rounding // shift the fraction always @(posedge clk_i) if (|s_shr2) s_frac2a <= s_fract_48_i >> s_shr2; else s_frac2a <= s_fract_48_i << s_shl2; assign s_expo2b = s_frac2a[46] ? s_expo1 : s_expo1 - 9'd1; // signals if precision was last during the right-shift above assign s_lost = (s_shr2 + {5'd0,s_shr3}) > s_r_zeros; // ***Stage 3*** // Rounding // 23 // | // xx00000000000000000000000grsxxxxxxxxxxxxxxxxxxxx // guard bit: s_frac2a[23] (LSB of output) // round bit: s_frac2a[22] assign s_guard = s_frac2a[22]; assign s_round = s_frac2a[21]; assign s_sticky = (|s_frac2a[20:0]) | s_lost; assign s_roundup = s_rmode_i==2'b00 ? // round to nearest even s_guard & ((s_round | s_sticky) | s_frac2a[23]) : s_rmode_i==2'b10 ? // round up (s_guard | s_round | s_sticky) & !s_sign_i : s_rmode_i==2'b11 ? // round down (s_guard | s_round | s_sticky) & s_sign_i : 0; // round to zero(truncate = no rounding) always @(posedge clk_i) if (s_roundup) s_frac_rnd <= s_frac2a[47:23] + 1; else s_frac_rnd <= s_frac2a[47:23]; assign s_shr3 = s_frac_rnd[24]; assign s_expo3 = (s_shr3 & (s_expo2b!=9'b011111111)) ? s_expo2b + 1 : s_expo2b; assign s_frac3 = (s_shr3 & (s_expo2b!=9'b011111111)) ? {1'b0,s_frac_rnd[24:1]} : s_frac_rnd; //-***Stage 4**** // Output assign s_op_0 = !((|s_opa_i[30:0]) & (|s_opb_i[30:0])); assign s_infa = &s_expa; assign s_infb = &s_expb; assign s_nan_a = s_infa & (|s_opa_i[22:0]); assign s_nan_b = s_infb & (|s_opb_i[22:0]); assign s_nan_in = s_nan_a | s_nan_b; assign s_nan_op = (s_infa | s_infb) & s_op_0; // 0 * inf = nan assign s_overflow = (s_expo3==9'b011111111) & !(s_infa | s_infb); assign s_ine_o = !s_op_0 & (s_lost | (|s_frac2a[22:0]) | s_overflow); assign s_output_o = (s_nan_in | s_nan_op) ? {s_sign_i,QNAN} : (s_infa | s_infb) | s_overflow ? {s_sign_i,INF} : s_r_zeros==48 ? {s_sign_i,ZERO_VECTOR} : {s_sign_i,s_expo3[7:0],s_frac3[22:0]}; endmodule // or1200_fpu_post_norm_mul
Go to most recent revision | Compare with Previous | Blame | View Log