micheleandClaude Sonnet 5
7e2711fa27
Bumps ADDR_WIDTH's default from 22 to 23 bits across every RTL module (neuron_memory, layer_sequencer, spi_engine, spi_neuron_top, mem_arbiter, int8_memory_access, memory_interface, psram_controller, memory_model) and every testbench that mirrors it, so the system's byte-address space reaches the full 8 MiB the recommended PSRAM part (ISSI IS66WVE4M16EBLL-70BLI, docs/FPGA-Neural-Hardware-Design.md §3) actually provides -- previously only 4 MiB (half the chip) was reachable, since int8_memory_access.v's byte->word address shift (addr >> 1) turned the old 22-bit byte address into only 21 real word bits, one short of the chip's real 22-bit word address (A0-A21). At 23 bits, that same shift lands exactly on all 22 chip address lines, so the whole part is usable now instead of deferred to a future widening. Also fixes a stray 22'd11-sized literal in layer_sequencer.v's descriptor-table address increment (numerically already safe via Verilog's zero-extension, but now correctly unsized so it always matches ADDR_WIDTH instead of silently assuming 22). Updated docs/FPGA-NeuralNetwork-Engine.md's SPI protocol address-field note (23 bits, top 1 reserved bit instead of 2) and docs/FPGA-Neural-Hardware-Design.md's PSRAM section (the "chip has one spare address line" framing is gone now that all 22 are wired and used). Full regression (all 11 ADDR_WIDTH-touching testbenches, plus a Yosys elaboration check of spi_neuron_top with the new default and no override) passes clean. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01WQV3vS9TXaGDJ5cRfnfidt
105 lines
2.8 KiB
Verilog
105 lines
2.8 KiB
Verilog
module memory_model #(
|
|
parameter ADDR_WIDTH = 23,
|
|
parameter DATA_WIDTH = 16,
|
|
parameter DEPTH = 4096,
|
|
parameter READ_LATENCY = 2
|
|
)(
|
|
input wire clk,
|
|
input wire rst,
|
|
|
|
input wire req,
|
|
input wire wr,
|
|
input wire [ADDR_WIDTH-1:0] addr,
|
|
input wire [DATA_WIDTH-1:0] wdata,
|
|
|
|
output reg [DATA_WIDTH-1:0] rdata,
|
|
output reg ready
|
|
);
|
|
|
|
reg [DATA_WIDTH-1:0] mem [0:DEPTH-1];
|
|
|
|
reg busy;
|
|
reg pending_wr;
|
|
reg [ADDR_WIDTH-1:0] pending_addr;
|
|
reg [DATA_WIDTH-1:0] pending_wdata;
|
|
|
|
integer delay_count;
|
|
integer i;
|
|
|
|
always @(posedge clk) begin
|
|
if (rst) begin
|
|
|
|
rdata <= {DATA_WIDTH{1'b0}};
|
|
ready <= 1'b0;
|
|
|
|
busy <= 1'b0;
|
|
pending_wr <= 1'b0;
|
|
pending_addr <= {ADDR_WIDTH{1'b0}};
|
|
pending_wdata <= {DATA_WIDTH{1'b0}};
|
|
|
|
delay_count <= 0;
|
|
|
|
for (i = 0; i < DEPTH; i = i + 1)
|
|
mem[i] <= {DATA_WIDTH{1'b0}};
|
|
|
|
end else begin
|
|
|
|
// ready is a one-cycle pulse
|
|
ready <= 1'b0;
|
|
|
|
// ----------------------------------------------------
|
|
// Accept request
|
|
// ----------------------------------------------------
|
|
|
|
if (!busy) begin
|
|
|
|
if (req) begin
|
|
|
|
busy <= 1'b1;
|
|
pending_wr <= wr;
|
|
pending_addr <= addr;
|
|
pending_wdata <= wdata;
|
|
|
|
delay_count <= READ_LATENCY;
|
|
end
|
|
|
|
end else begin
|
|
|
|
// ------------------------------------------------
|
|
// Wait
|
|
// ------------------------------------------------
|
|
|
|
if (delay_count > 0) begin
|
|
|
|
delay_count <= delay_count - 1;
|
|
|
|
end else begin
|
|
|
|
// --------------------------------------------
|
|
// Complete transaction
|
|
// --------------------------------------------
|
|
|
|
if (pending_wr) begin
|
|
|
|
// WRITE
|
|
if (pending_addr < DEPTH)
|
|
mem[pending_addr] <= pending_wdata;
|
|
|
|
end else begin
|
|
|
|
// READ
|
|
if (pending_addr < DEPTH)
|
|
rdata <= mem[pending_addr];
|
|
else
|
|
rdata <= {DATA_WIDTH{1'b0}};
|
|
|
|
end
|
|
|
|
ready <= 1'b1;
|
|
busy <= 1'b0;
|
|
end
|
|
end
|
|
end
|
|
end
|
|
|
|
endmodule |