NuBusFPGA/nubus-to-ztex-gateware/nubus_full_unified.py
Romain Dolbeau e82802de4a V1.2 bringup
2023-04-17 22:55:22 +02:00

827 lines
38 KiB
Python

from migen import *
from migen.genlib.fifo import *
from migen.genlib.cdc import *
from migen.fhdl.specials import Tristate
import litex
from litex.soc.interconnect import wishbone
class NuBus(Module):
def __init__(self, soc, version,
burst_size, tosbus_fifo, fromsbus_fifo, fromsbus_req_fifo,
wb_read, wb_write, wb_dma,
usesampling=False,
cd_nubus="nubus", cd_nubus90="nubus90"):
platform = soc.platform
self.add_sources(platform, version)
#led0 = platform.request("user_led", 0)
#led1 = platform.request("user_led", 1)
if (usesampling):
# when using 'sampling', the NuBus clock is sampled at sys_clk frequency instead of used directly
# the other signals are still sampled synchronously, and the buffer used from sysclk afterwards
# I'm not completely sure about timings, but in practice it seems to work...
# The major benefit is that when a read is detected, it is sent to the Wishbone synchronously as the signals are already in sys_clk
# And when Wishbone answers, we just wait for the next detected NuBus edge to answer
# It significantly improves read latency
# Writes don't see the same improvement, as they always are fire-and-forget in a FIFO anyway
nub_clk = ClockSignal(cd_nubus)
nub_resetn = ~ResetSignal(cd_nubus)
nub_clk_prev_bits = 4 # how many cycles after posedge do we still dare set some signals (i.e. still before setup time before negedge)
nub_clk_prev = Signal(nub_clk_prev_bits)
nub_clk_negedge = Signal()
nub_clk_posedge = Signal()
nub_clk_insetup = Signal()
self.sync += [
nub_clk_prev[0].eq(nub_clk),
]
self.sync += [
nub_clk_prev[i].eq(nub_clk_prev[i-1]) for i in range(1, nub_clk_prev_bits)
]
#self.sync += [
# nub_clk_negedge.eq(~nub_clk & nub_clk_prev[0]),
# nub_clk_posedge.eq( nub_clk & ~nub_clk_prev[0]),
# nub_clk_insetup.eq( nub_clk & (nub_clk_prev != ((2**nub_clk_prev_bits)-1))), # if one of the previous X cycles is zero, we're early enough to set up signals
#]
# this should use double sampling; however using [1] and [2] break, perhaps the detection is too late?
self.comb += [
nub_clk_negedge.eq(~nub_clk_prev[0] & nub_clk_prev[1]),
nub_clk_posedge.eq( nub_clk_prev[0] & ~nub_clk_prev[1]),
nub_clk_insetup.eq( nub_clk_prev[0] & (nub_clk_prev[1:nub_clk_prev_bits] != ((2**(nub_clk_prev_bits-1))-1))), # if one of the previous X cycles is zero, we're early enough to set up signals
]
# Signals for tri-stated nubus access
# slave
tmo_oe = Signal() # output enable
tm0_i_n = Signal()
tm0_o_n = Signal(reset = 1)
tm1_i_n = Signal()
tm1_o_n = Signal(reset = 1)
ack_i_n = Signal()
ack_o_n = Signal(reset = 1)
ad_oe = Signal()
ad_i_n = Signal(32)
ad_o_n = Signal(32)
id_i_n = Signal(4)
start_i_n = Signal()
start_o_n = Signal(reset = 1) # master via master_oe
# master
rqst_oe = Signal()
rqst_i_n = Signal()
rqst_o_n = Signal(reset = 1)
# sampled signals, exposing the value of the register acquired on the falling edge
# they can change every cycle *on falling edge*
# slave
sampled_tm0 = Signal() # high is byte (which byte is in ad0/ad1); low is halfword/word/block depending on ad0/ad1
sampled_tm1 = Signal() # high is write
sampled_start = Signal()
sampled_ack = Signal()
sampled_ad = Signal(32)
sampled_ad_byterev = Signal(32)
# master
sampled_rqst = Signal()
# address rewriting
# can change every cycle *on falling edge*
processed_ad = Signal(32)
processed_super_ad = Signal(32)
self.comb += [
processed_ad[0:23].eq(sampled_ad[0:23]),
If(~sampled_ad[23], # first 8 MiB of slot space: remap to last 8 Mib of SDRAM
processed_ad[23:32].eq(Cat(Signal(1, reset=1), Signal(8, reset = 0x8f))), # 0x8f8...
).Else( # second 8 MiB: direct access
processed_ad[23:32].eq(Cat(sampled_ad[23], Signal(8, reset = 0xf0)))), # 24 bits, a.k.a 22 bits of words
processed_super_ad[0:28].eq(sampled_ad[0:28]),
processed_super_ad[28:32].eq(Signal(4, reset = 0x8)),
sampled_ad_byterev[ 0: 8].eq(sampled_ad[24:32]),
sampled_ad_byterev[ 8:16].eq(sampled_ad[16:24]),
sampled_ad_byterev[16:24].eq(sampled_ad[ 8:16]),
sampled_ad_byterev[24:32].eq(sampled_ad[ 0: 8]),
]
# decoded signals, exposing decoded results from the sampled signals
# they can change every cycle *on falling edge*
# from sampling (fixme?)
decoded_sel = Signal(4)
decoded_block = Signal()
decoded_busy = Signal()
# locally evaluated
decoded_myslot = Signal()
decoded_mysuperslot = Signal()
self.comb += [
decoded_myslot.eq(
(sampled_ad[28:32] == 0xF) &
(sampled_ad[27] == ~id_i_n[3]) &
(sampled_ad[26] == ~id_i_n[2]) &
(sampled_ad[25] == ~id_i_n[1]) &
(sampled_ad[24] == ~id_i_n[0])),
decoded_mysuperslot.eq(
(sampled_ad[31] == ~id_i_n[3]) &
(sampled_ad[30] == ~id_i_n[2]) &
(sampled_ad[29] == ~id_i_n[1]) &
(sampled_ad[28] == ~id_i_n[0])),
#led0.eq(decoded_block),
]
# current value, registered from the sampled/processed/decoded signals
# change is controlled by the FSM
current_adr = Signal(32)
#current_tm0 = Signal()
#current_tm1 = Signal()
current_sel = Signal(4)
#current_block = Signal()
current_data = Signal(32)
# write FIFO to speed up bus turnaround on NuBus side
write_fifo_layout = [
("adr", 32),
("data", 32),
("sel", 4),
]
if (usesampling):
self.submodules.write_fifo = write_fifo = SyncFIFOBuffered(width=layout_len(write_fifo_layout), depth=16)
else:
self.submodules.write_fifo = write_fifo = ClockDomainsRenamer({"read": "sys", "write": "nubus"})(AsyncFIFOBuffered(width=layout_len(write_fifo_layout), depth=16))
write_fifo_dout = Record(write_fifo_layout)
self.comb += write_fifo_dout.raw_bits().eq(write_fifo.dout)
write_fifo_din = Record(write_fifo_layout)
self.comb += write_fifo.din.eq(write_fifo_din.raw_bits())
# nubus-synchronous sampling (in Verilog for negedge)
self.specials += Instance("nubus_sampling",
i_nub_clkn = ClockSignal(cd_nubus),
i_nub_resetn = ~ResetSignal(cd_nubus),
i_nub_tm0n = tm0_i_n,
i_nub_tm1n = tm1_i_n,
i_nub_startn = start_i_n,
i_nub_rqstn = rqst_i_n,
i_nub_ackn = ack_i_n,
i_nub_adn = ad_i_n,
o_tm0 = sampled_tm0,
o_tm1 = sampled_tm1,
o_start = sampled_start,
o_rqst = sampled_rqst,
o_ack = sampled_ack,
o_ad = sampled_ad,
o_sel = decoded_sel,
o_block = decoded_block,
o_busy = decoded_busy,
)
self.read_ctr = read_ctr = Signal(32)
self.writ_ctr = writ_ctr = Signal(32)
if (usesampling):
# ############# usesampling FSM
self.submodules.slave_fsm = slave_fsm = FSM(reset_state="Reset")
slave_fsm.act("Reset",
NextState("Idle")
)
slave_fsm.act("Idle",
# only react to transaction start at posedge
If(nub_clk_posedge & (decoded_myslot | decoded_mysuperslot) & sampled_start & ~sampled_ack & ~sampled_tm1,# & ~decoded_block, # regular read (we always send back 32 bits, so don't worry about byte/word)
If(decoded_myslot,
NextValue(current_adr, processed_ad),
).Else( # decoded_mysuperslot,
NextValue(current_adr, processed_super_ad),
),
NextValue(read_ctr, read_ctr + 1),
NextState("WaitWBRead"),
).Elif(nub_clk_posedge & (decoded_myslot | decoded_mysuperslot) & sampled_start & ~sampled_ack & sampled_tm1,# & ~decoded_block, # regular write
If(decoded_myslot,
NextValue(current_adr, processed_ad),
).Else( # decoded_mysuperslot,
NextValue(current_adr, processed_super_ad),
),
NextValue(current_sel, decoded_sel),
NextValue(writ_ctr, writ_ctr + 1),
If(write_fifo.writable,
NextState("NubusWriteDataToFIFO"),
).Else(
NextState("NubusWaitForFIFO"),
)
)
)
slave_fsm.act("WaitWBRead",
wb_read.cyc.eq(1),
wb_read.stb.eq(1),
wb_read.we.eq(0),
wb_read.sel.eq(0xf),
wb_read.adr.eq(current_adr[2:32]),
tmo_oe.eq(1),
tm0_o_n.eq(1),
tm1_o_n.eq(1),
ack_o_n.eq(1),
If(wb_read.ack,
NextValue(current_data, wb_read.dat_r),
If(nub_clk_insetup,
ad_oe.eq(1),
ad_o_n.eq(~wb_read.dat_r),
tm0_o_n.eq(0),
tm1_o_n.eq(0),
ack_o_n.eq(0),
NextState("FinishRead"),
).Else(
NextState("WaitBeforeFinishRead"),
)
)
)
slave_fsm.act("WaitBeforeFinishRead",
tmo_oe.eq(1),
tm0_o_n.eq(1),
tm1_o_n.eq(1),
ack_o_n.eq(1),
If(nub_clk_insetup,
ad_oe.eq(1),
ad_o_n.eq(~current_data),
tm0_o_n.eq(0),
tm1_o_n.eq(0),
ack_o_n.eq(0),
NextState("FinishRead"),
),
)
slave_fsm.act("FinishRead",
tmo_oe.eq(1),
ad_oe.eq(1),
ad_o_n.eq(~current_data),
tm0_o_n.eq(0),
tm1_o_n.eq(0),
ack_o_n.eq(0),
#If((~nub_clk & nub_clk_prev[0]), # simultaneous with setting negedge
If(nub_clk_negedge,
NextState("ReadCleanup"),
)
)
slave_fsm.act("ReadCleanup",
tmo_oe.eq(1),
ad_oe.eq(1),
ad_o_n.eq(~current_data),
tm0_o_n.eq(0),
tm1_o_n.eq(0),
ack_o_n.eq(0),
NextState("Idle"),
),
slave_fsm.act("NubusWriteDataToFIFO",
tmo_oe.eq(1),
tm0_o_n.eq(0),
tm1_o_n.eq(0),
ack_o_n.eq(0),
#If((~nub_clk & nub_clk_prev[0]), # simultaneous with setting negedge
If(nub_clk_negedge,
write_fifo.we.eq(1),
NextState("WriteCleanup"),
)
)
slave_fsm.act("NubusWaitForFIFO",
tmo_oe.eq(1),
tm0_o_n.eq(1),
tm1_o_n.eq(1),
ack_o_n.eq(1),
If(nub_clk_posedge & write_fifo.writable,
NextState("NubusWriteDataToFIFO"),
)
)
slave_fsm.act("WriteCleanup", # extra sysclk cycle after negedge
tmo_oe.eq(1),
tm0_o_n.eq(0),
tm1_o_n.eq(0),
ack_o_n.eq(0),
NextState("Idle"),
)
# ############# end of usesampling FSM
else:
# ############# non-usesampling FSM
self.submodules.slave_fsm = slave_fsm = ClockDomainsRenamer(cd_nubus)(FSM(reset_state="Reset"))
slave_fsm.act("Reset",
NextState("Idle")
)
slave_fsm.act("Idle",
If((decoded_myslot | decoded_mysuperslot) & sampled_start & ~sampled_ack & ~sampled_tm1,# & ~decoded_block, # regular read (we always send back 32 bits, so don't worry about byte/word)
If(decoded_myslot,
NextValue(current_adr, processed_ad),
).Else( # decoded_mysuperslot,
NextValue(current_adr, processed_super_ad),
),
NextValue(read_ctr, read_ctr + 1),
NextState("WaitWBRead"),
).Elif((decoded_myslot | decoded_mysuperslot) & sampled_start & ~sampled_ack & sampled_tm1,# & ~decoded_block, # regular write
If(decoded_myslot,
NextValue(current_adr, processed_ad),
).Else( # decoded_mysuperslot,
NextValue(current_adr, processed_super_ad),
),
NextValue(current_sel, decoded_sel),
NextValue(writ_ctr, writ_ctr + 1),
NextState("NubusWriteDataToFIFO"),
)
)
slave_fsm.act("WaitWBRead",
wb_read.cyc.eq(1),
wb_read.stb.eq(1),
wb_read.we.eq(0),
wb_read.sel.eq(0xf),
wb_read.adr.eq(current_adr[2:32]),
tmo_oe.eq(1),
tm0_o_n.eq(1),
tm1_o_n.eq(1),
ack_o_n.eq(1),
If(wb_read.ack,
ad_oe.eq(1),
ad_o_n.eq(~wb_read.dat_r),
tm0_o_n.eq(0),
tm1_o_n.eq(0),
ack_o_n.eq(0),
NextState("Idle"),
)
)
slave_fsm.act("NubusWriteDataToFIFO",
tmo_oe.eq(1),
tm0_o_n.eq(1),
tm1_o_n.eq(1),
ack_o_n.eq(1),
If(write_fifo.writable,
write_fifo.we.eq(1),
tm0_o_n.eq(0),
tm1_o_n.eq(0),
ack_o_n.eq(0),
NextState("Idle"),
)
)
# ############# end of non-usesampling FSM
# connect the write FIFO inputs
self.comb += [ write_fifo_din.adr.eq(current_adr), # recorded
write_fifo_din.data.eq(~ad_i_n if usesampling else sampled_ad),
write_fifo_din.sel.eq(current_sel), # recorded
]
# deal with emptying the Write FIFO to the write WB
self.comb += [ wb_write.cyc.eq(write_fifo.readable),
wb_write.stb.eq(write_fifo.readable),
wb_write.we.eq(1),
wb_write.adr.eq(write_fifo_dout.adr[2:32]),
wb_write.dat_w.eq(write_fifo_dout.data),
wb_write.sel.eq(write_fifo_dout.sel),
write_fifo.re.eq(wb_write.ack),
]
owning_bus = Signal(reset = 0) # fixme ; theoretically one can bypass arbitration when owning the bus
start_arbitration = Signal()
grant = Signal()
master_oe = Signal()
nubus_sync = getattr(self.sync, cd_nubus)
nubus_sync += [
If(sampled_rqst & ~start_arbitration,
owning_bus.eq(0),
)
]
self.submodules.dma_fsm = dma_fsm = ClockDomainsRenamer(cd_nubus)(FSM(reset_state="Reset"))
ctr = Signal(log2_int(burst_size)) # burst counter
burst = Signal()
burst_we = Signal()
data_width = burst_size * 4
data_width_bits = burst_size * 32
blk_addr_width = 32 - log2_int(data_width) # 27 for burst_size == 8, 28 for burst_size == 4
fifo_addr = Signal(blk_addr_width)
fifo_blk_addr = Signal(blk_addr_width)
fifo_buffer = Signal(data_width_bits)
tosbus_fifo_dout = Record(soc.tosbus_layout)
self.comb += tosbus_fifo_dout.raw_bits().eq(tosbus_fifo.dout)
tosbus_fifo_dout_data_byterev = Signal(data_width_bits)
tosbus_fifo_dout_bytereversal_stmts = [ tosbus_fifo_dout_data_byterev[k*32+j*8:k*32+j*8+8].eq(tosbus_fifo_dout.data[k*32+32-j*8-8:k*32+32-j*8]) for k in range(burst_size) for j in range(4) ]
self.comb += tosbus_fifo_dout_bytereversal_stmts
fromsbus_req_fifo_dout = Record(soc.fromsbus_req_layout)
self.comb += fromsbus_req_fifo_dout.raw_bits().eq(fromsbus_req_fifo.dout)
fromsbus_fifo_din = Record(soc.fromsbus_layout)
self.comb += fromsbus_fifo.din.eq(fromsbus_fifo_din.raw_bits())
#self.comb += led0.eq(~dma_fsm.ongoing("Idle"))
#self.comb += led1.eq(burst)
dma_fsm.act("Reset",
NextState("Idle")
)
dma_fsm.act("Idle",
If(wb_dma.cyc & wb_dma.stb & ~sampled_rqst, # we need the bus and it's not being requested
NextValue(burst, 0),
If(owning_bus, # we own the bus, skip arbitration
NextState("AdrCycle"),
).Else( # go for arbitration
NextState("Arbitration"),
),
).Elif(tosbus_fifo.readable & ~sampled_rqst,
NextValue(burst, 1),
NextValue(burst_we, 1),
NextValue(fifo_addr, tosbus_fifo_dout.address[(32-blk_addr_width):32]),
If(owning_bus, # we own the bus, skip arbitration
NextState("Burst4AdrCycle"),
).Else( # go for arbitration
NextState("Arbitration"),
)
).Elif(fromsbus_req_fifo.readable & fromsbus_fifo.writable & ~sampled_rqst,
NextValue(burst, 1),
NextValue(burst_we, 0),
NextValue(fifo_addr, fromsbus_req_fifo_dout.dmaaddress[(32-blk_addr_width):32]),
NextValue(fifo_blk_addr, fromsbus_req_fifo_dout.blkaddress),
If(owning_bus, # we own the bus, skip arbitration
NextState("Burst4AdrCycle"),
).Else( # go for arbitration
NextState("Arbitration"),
)
)
)
dma_fsm.act("Arbitration",
start_arbitration.eq(1),
rqst_oe.eq(1),
rqst_o_n.eq(0),
NextState("WaitForGrant"),
)
dma_fsm.act("WaitForGrant",
start_arbitration.eq(1),
rqst_oe.eq(1),
rqst_o_n.eq(0),
If(grant & ~decoded_busy, # I'm now 'owner'
NextValue(owning_bus, 1),
If(burst,
NextState("Burst4AdrCycle"),
).Else(
NextState("AdrCycle"),
),
)
)
dma_fsm.act("AdrCycle",
start_arbitration.eq(0),
master_oe.eq(1), # for start
tmo_oe.eq(1), # for tm0, tm1, ack
ad_oe.eq(1), # for write address
start_o_n.eq(0),
tm0_o_n.eq(~((wb_dma.sel == 0x1) | (wb_dma.sel == 0x2) | (wb_dma.sel == 0x4) | (wb_dma.sel == 0x8))), # byte only
tm1_o_n.eq(~wb_dma.we),
ad_o_n[0].eq(~((wb_dma.sel == 0x2) | (wb_dma.sel == 0x3) | (wb_dma.sel == 0x8) | (wb_dma.sel == 0xc))), # odd bytes, both half-words
ad_o_n[1].eq(~((wb_dma.sel == 0x4) | (wb_dma.sel == 0x8) | (wb_dma.sel == 0xc))), # upper bytes and half-word
ad_o_n[2:32].eq(~wb_dma.adr),
ack_o_n.eq(1),
If(wb_dma.we,
NextState("DatCycle"),
).Else(
NextState("ReadWaitForAck"),
)
)
dma_fsm.act("DatCycle",
master_oe.eq(1), # for start
ad_oe.eq(1), # for write data
start_o_n.eq(1), # start finished, but still need to be driven
ad_o_n.eq(~wb_dma.dat_w),
If(sampled_ack,
wb_dma.ack.eq(1),
# fixme: check status ??? (tm0 and tm1 should be active for no-error)
NextState("FinishCycle"),
)
)
dma_fsm.act("FinishCycle",
NextValue(burst, 0),
master_oe.eq(1), # for start
start_o_n.eq(1), # start finished, but still need to be driven
tmo_oe.eq(1), # for tm0, tm1, ack, need to be driven to inactive
tm0_o_n.eq(1),
tm1_o_n.eq(1),
ack_o_n.eq(1),
NextState("Idle"),
)
dma_fsm.act("ReadWaitForAck",
master_oe.eq(1), # for start
start_o_n.eq(1), # start finished, but still need to be driven
wb_dma.dat_r.eq(sampled_ad),
If(sampled_ack,
wb_dma.ack.eq(1),
# fixme: check status ??? (tm0 and tm1 should be active for no-error)
NextState("FinishCycle"),
)
)
if (burst_size == 4):
handle_ad_for_burst = [
ad_o_n[0].eq(1), # burst
ad_o_n[1].eq(0), # burst
ad_o_n[2].eq(0), # burst == 4
ad_o_n[3].eq(1), # burst == 4
ad_o_n[4:32].eq(~fifo_addr), # adr
]
elif (burst_size == 8):
handle_ad_for_burst = [
ad_o_n[0].eq(1), # burst
ad_o_n[1].eq(0), # burst
ad_o_n[2].eq(0), # burst == 8
ad_o_n[3].eq(0), # burst == 8
ad_o_n[4].eq(1), # burst == 8
ad_o_n[5:32].eq(~fifo_addr), # adr
]
else:
raise ValueError(f"Unsupported burst_size {burst_size}")
dma_fsm.act("Burst4AdrCycle",
start_arbitration.eq(0),
master_oe.eq(1), # for start
tmo_oe.eq(1), # for tm0, tm1, ack
ad_oe.eq(1), # for write address
start_o_n.eq(0),
tm0_o_n.eq(1), # burst
tm1_o_n.eq(~burst_we),
*handle_ad_for_burst,
ack_o_n.eq(1),
NextValue(ctr, 0),
If(burst_we,
NextState("Burst4DatCycleTM0"),
).Else(
NextState("Burst4ReadWaitForTM0"),
)
)
if (burst_size == 4):
handle_buffer_read_for_burst = [
Case(ctr, {
0x0: NextValue(fifo_buffer[ 0: 32], sampled_ad),
0x1: NextValue(fifo_buffer[32: 64], sampled_ad),
0x2: NextValue(fifo_buffer[64: 96], sampled_ad),
#0x3: NextValue(fifo_buffer[96:128], sampled_ad),
#0x0: NextValue(fifo_buffer[ 0: 32], sampled_ad_byterev),
#0x1: NextValue(fifo_buffer[32: 64], sampled_ad_byterev),
#0x2: NextValue(fifo_buffer[64: 96], sampled_ad_byterev),
##0x3: NextValue(fifo_buffer[96:128], sampled_ad_byterev),
}),
]
handle_final_buffer_read_for_burst = [
fromsbus_fifo_din.data.eq(Cat(fifo_buffer[0:96], sampled_ad)), # we use sampled_ad directly for 96:128
]
elif (burst_size == 8):
handle_buffer_read_for_burst = [
Case(ctr, {
0x0: NextValue(fifo_buffer[ 0: 32], sampled_ad),
0x1: NextValue(fifo_buffer[ 32: 64], sampled_ad),
0x2: NextValue(fifo_buffer[ 64: 96], sampled_ad),
0x3: NextValue(fifo_buffer[ 96:128], sampled_ad),
0x4: NextValue(fifo_buffer[128:160], sampled_ad),
0x5: NextValue(fifo_buffer[160:192], sampled_ad),
0x6: NextValue(fifo_buffer[192:224], sampled_ad),
#0x7: NextValue(fifo_buffer[224:256], sampled_ad),
}),
]
handle_final_buffer_read_for_burst = [
fromsbus_fifo_din.data.eq(Cat(fifo_buffer[0:224], sampled_ad)), # we use sampled_ad directly for 224:256
]
else:
raise ValueError(f"Unsupported burst_size {burst_size}")
dma_fsm.act("Burst4ReadWaitForTM0",
master_oe.eq(1), # for start
start_o_n.eq(1), # start finished, but still need to be driven
If(sampled_ack, # oups
fromsbus_req_fifo.re.eq(1), # remove request to avoid infinite repeat
#NextValue(led0, 1),
#NextValue(led1, 1),
NextState("FinishCycle"),
).Elif(sampled_tm0,
*handle_buffer_read_for_burst,
NextValue(ctr, ctr + 1),
If(ctr == (burst_size - 2), # burst next-to-last
NextState("Burst4ReadWaitForAck"),
).Else(
NextState("Burst4ReadWaitForTM0"),
)
)
)
dma_fsm.act("Burst4ReadWaitForAck",
master_oe.eq(1), # for start
start_o_n.eq(1), # start finished, but still need to be driven
If(sampled_ack,
fromsbus_req_fifo.re.eq(1), # remove request
fromsbus_fifo.we.eq(1),
fromsbus_fifo_din.blkaddress.eq(fifo_blk_addr),
*handle_final_buffer_read_for_burst,
# fixme: check status ??? (tm0 and tm1 should be active for no-error)
#NextValue(led0, (~sampled_tm0 | ~sampled_tm1)),
NextState("FinishCycle"),
)
)
if (burst_size == 4):
handle_buffer_write_for_burst = [
Case(ctr, {
0x0: ad_o_n.eq(~tosbus_fifo_dout.data[ 0: 32]),
0x1: ad_o_n.eq(~tosbus_fifo_dout.data[32: 64]),
0x2: ad_o_n.eq(~tosbus_fifo_dout.data[64: 96]),
##0x3: ad_o_n.eq(~tosbus_fifo_dout.data[96:128]),
#0x0: ad_o_n.eq(~tosbus_fifo_dout_data_byterev[ 0: 32]),
#0x1: ad_o_n.eq(~tosbus_fifo_dout_data_byterev[32: 64]),
#0x2: ad_o_n.eq(~tosbus_fifo_dout_data_byterev[64: 96]),
##0x3: ad_o_n.eq(~tosbus_fifo_dout_data_byterev[96:128]),
}),
]
handle_last_buffer_write_for_burst = [
ad_o_n.eq(~tosbus_fifo_dout.data[96:128]), # last word
]
elif (burst_size == 8):
handle_buffer_write_for_burst = [
Case(ctr, {
0x0: ad_o_n.eq(~tosbus_fifo_dout.data[ 0: 32]),
0x1: ad_o_n.eq(~tosbus_fifo_dout.data[ 32: 64]),
0x2: ad_o_n.eq(~tosbus_fifo_dout.data[ 64: 96]),
0x3: ad_o_n.eq(~tosbus_fifo_dout.data[ 96:128]),
0x4: ad_o_n.eq(~tosbus_fifo_dout.data[128:160]),
0x5: ad_o_n.eq(~tosbus_fifo_dout.data[160:192]),
0x6: ad_o_n.eq(~tosbus_fifo_dout.data[192:224]),
#0x7: ad_o_n.eq(~tosbus_fifo_dout.data[224:256]),
}),
]
handle_last_buffer_write_for_burst = [
ad_o_n.eq(~tosbus_fifo_dout.data[224:256]), # last word
]
else:
raise ValueError(f"Unsupported burst_size {burst_size}")
dma_fsm.act("Burst4DatCycleTM0",
master_oe.eq(1), # for start
ad_oe.eq(1), # for write data
start_o_n.eq(1), # start finished, but still need to be driven
*handle_buffer_write_for_burst,
If(sampled_ack, # oups
#NextValue(led0, 1),
#NextValue(led1, 1),
tosbus_fifo.re.eq(1), # remove FIFO entry to avoid infinite repeat
NextState("FinishCycle"),
).Elif(sampled_tm0,
NextValue(ctr, ctr + 1),
If(ctr == (burst_size - 2), # burst next-to-last
NextState("Burst4DatCycleAck"),
).Else(
NextState("Burst4DatCycleTM0"),
)
)
)
dma_fsm.act("Burst4DatCycleAck",
master_oe.eq(1), # for start
ad_oe.eq(1), # for write data
start_o_n.eq(1), # start finished, but still need to be driven
*handle_last_buffer_write_for_burst,
If(sampled_ack,
tosbus_fifo.re.eq(1), # remove FIFO entry at last
# fixme: check status ??? (tm0 and tm1 should be active for no-error)
#NextValue(led0, (~sampled_tm0 | ~sampled_tm1)),
NextState("FinishCycle"),
)
)
# stuff at this end so we don't use the signals inadvertantly
# real NuBus signals
nub_tm0n = platform.request("tm0_3v3_n") # V1.0: from CPLD ; V1.2: from shifters
nub_tm1n = platform.request("tm1_3v3_n") # V1.0: from CPLD ; V1.2: from shifters
nub_startn = platform.request("start_3v3_n") # V1.0: from CPLD ; V1.2: from shifters
nub_ackn = platform.request("ack_3v3_n") # V1.0: from CPLD ; V1.2: from shifters
nub_adn = platform.request("ad_3v3_n") # V1.0: from CPLD ; V1.2: from shifters
nub_idn = platform.request("id_3v3_n") # V1.0: from CPLD (4 bits) ; V1.2: from shifters (3 bits, /ID3 is always 0)
# idem between V1.0 and V1.2
self.specials += Tristate(nub_adn, ad_o_n, ad_oe, ad_i_n)
# Tri-state
if (version == "V1.0"):
# tri-state communication with CPLD
self.specials += Tristate(nub_tm0n, tm0_o_n, tmo_oe, tm0_i_n)
self.specials += Tristate(nub_tm1n, tm1_o_n, tmo_oe, tm1_i_n)
self.specials += Tristate(nub_ackn, ack_o_n, tmo_oe, ack_i_n)
self.specials += Tristate(nub_startn, start_o_n, master_oe, start_i_n)
elif (version == "V1.2"):
# input only
self.comb += [
tm0_i_n.eq(nub_tm0n),
tm1_i_n.eq(nub_tm1n),
ack_i_n.eq(nub_ackn),
start_i_n.eq(nub_startn),
]
else:
raise ValueError(f"Unsupported version {version}")
# input only
if (version == "V1.0"):
self.comb += [
id_i_n.eq(nub_idn),
]
elif (version == "V1.2"):
self.comb += [
id_i_n.eq(Cat(nub_idn, Signal(1, reset = 0))),
]
else:
raise ValueError(f"Unsupported version {version}")
# NubusFPGA-only signals
nf_nubus_ad_dir = platform.request("nubus_ad_dir") # to drivers
self.comb += [
nf_nubus_ad_dir.eq(~ad_oe),
]
if (version == "V1.0"):
nf_tmoen = platform.request("tmoen") # to cpld
self.comb += [
nf_tmoen.eq(~tmo_oe),
]
# real Nubus signal, for master
nub_rqstn = platform.request("rqst_3v3_n") # V1.0: from CPLD ; V1.2: from shifters
# Tri-state
if (version == "V1.0"):
# tri-state communication with CPLD
self.specials += Tristate(nub_rqstn, rqst_o_n, rqst_oe, rqst_i_n)
elif (version == "V1.2"):
# input only
self.comb += [
rqst_i_n.eq(nub_rqstn),
]
else:
raise ValueError(f"Unsupported version {version}")
# NubusFPGA-only signals, for master
if (version == "V1.0"):
nub_arbcy_n = platform.request("arbcy_n") # V1.0: from cpld
nf_grant = platform.request("grant") # V1.0: from cpld
nf_nubus_master_dir = platform.request("nubus_master_dir") # V1.0: to cpld
nf_fpga_to_cpld_signal = platform.request("fpga_to_cpld_signal") # V1.0: to cpld, 'rqstoen'
# NuBus90 signals, , for completeness
nub_clk2xn = ClockSignal(cd_nubus90)
nub_tm2n = platform.request("tm2_3v3_n") # V1.0: from CPLD ; V1.2: from shifters
if (version == "V1.0"):
self.comb += [
nf_nubus_master_dir.eq(master_oe),
nub_arbcy_n.eq(~start_arbitration),
grant.eq(nf_grant),
nf_fpga_to_cpld_signal.eq(~rqst_oe),
]
if (version == "V1.2"):
self.specials += Instance("nubus_cpldinfpga",
i_nubus_oe = soc.hold_reset, # improveme, handled in SoC
i_tmoen = ~tmo_oe,
i_nubus_master_dir = master_oe,
i_rqst_oe_n = ~rqst_oe,
i_id_n_3v3 = id_i_n, # input only
i_arbcy_n = ~start_arbitration,
i_arb_n_3v3 = platform.request("arb_3v3_n"), # arb only seen by cpld
o_arb_o_n = platform.request("arb_o_n"),
o_grant = grant,
i_tm0_n_3v3 = tm0_o_n, # tm0 driving controlled by tmoen
o_tm0_o_n = platform.request("tm0_o_n"),
i_tm1_n_3v3 = tm1_o_n, # tm1 driving controlled by tmoen
o_tm1_o_n = platform.request("tm1_o_n"),
o_tmx_oe_n = platform.request("tmx_oe_n"),
i_tm2_n_3v3 = nub_tm2n, # tm2 currently never driven
o_tm2_o_n = platform.request("tm2_o_n"),
o_tm2_oe_n = platform.request("tm2_oe_n"),
i_start_n_3v3 = start_o_n, # start driving enabled by nubus_master_dir
o_start_o_n = platform.request("start_o_n"),
o_start_oe_n = platform.request("start_oe_n"),
i_ack_n_3v3 = ack_o_n, # ack driving controlled by tmoen
o_ack_o_n = platform.request("ack_o_n"),
o_ack_oe_n = platform.request("ack_oe_n"),
i_rqst_n_3v3 = rqst_o_n, # rqst driving controller by rqst_oe_n
o_rqst_o_n = platform.request("rqst_o_n")
)
def add_sources(self, platform, version):
# sampling of data on falling edge of clock, done in verilog
platform.add_source("nubus_sampling.v", "verilog")
if (version == "V1.2"):
platform.add_source("nubus_arbiter.v", "verilog") # for CPLDinfpga
platform.add_source("nubus_cpldinfpga.v", "verilog") # internal now