1
0
Fork 0

Compare commits

...

9 Commits

Author SHA1 Message Date
morgan 074e8e94d1 cxp downconn firmware: GTX setup
testing: add loopmode mode & tsusrclk mmcm drp bitbang
testing: add IDLE word printout
downconn: add QPLL and GTX setup
downconn: add DRP to support all CXP linerate up to 12.5Gbps
2024-09-05 17:24:43 +08:00
morgan 354949baab cxp upconn firmware: low speed serial setup
testing: add trigger & trigger ack CSR to test tranmission prioirty
upconn: add some upconn error for control flow
upconn: add control packet writer
control packet: add u32 & u64 register control
control packet: add CRC calculation & packet type inserter in fw
2024-09-05 17:24:43 +08:00
morgan b534129d08 zc706: add CXP_DEMO variant
zc706: add fmc pads
zc706: add constraint to fix comma alignment & setup/hold time issue
2024-09-05 17:21:09 +08:00
morgan 0f4a8754c3 cxp: add upconn interface, downconn PHY & crc
testing: add CSR control for tx trigger & trigger ack
upconn: connect trigger, trigger ack & command_packet to UpConnPHY
downconn: add GTX PHY
2024-09-05 17:21:09 +08:00
morgan d5096781d3 cxp pipeline: packet handling pipeline
tx pipeline: add CRC32 inserter
tx pipeline: add start & end of packet code inserter
tx pipeline: add packet wrapper for start & stop packet indication
tx pipeline: add code source for trigger & trigger ack packet
tx pipeline: add packet for trigger & trigger ack
tx pipeline: add tx_command_packet for firmware
tx command packet: add fifo to store control packet
2024-09-05 17:16:29 +08:00
morgan 030c4c13b9 cxp upconn gw: add low speed serial PHY
testing: add debug fifo output b4 encoder
cxp upconn: add low speed serial
cxp upconn: add reset, tx_busy, tx_enable
cxp upconn: add clockgen module for 20.83Mbps & 41.66Mbps using counters
cxp upconn: add oserdes using CEInserter
cxp upconn: add packet scheduler with CEInserter
scheduler: send priority packet during word/char boundary
scheduler: send IDLE every 10000 words
scheduler: encode packet before sending to oserdes
2024-09-05 17:07:59 +08:00
morgan 223c32a4dc cxp downconn gw: add gtx up to 12.5Gbps
testing: add txusrclk mmcm & loopback mode
testing: add debug output
downconn: add GTX and QPLL support
downconn: add DRP for GTX and QPLL to support all CXP linerates
GTX: add gtx with mmcm for TXUSRCLK freq requirement
GTX: add loopback mode parameter for testing
GTX: add gtx with 40bits internal width
GTX: use built-in comma aligner
GTX: add comma checker to ensure comma is aligner on highest linerate
GTX: set QPLL as CLK source for GTX
2024-09-05 17:07:38 +08:00
morgan 46f88a7793 fmc: add cxp_4r_fmc adepter io 2024-09-05 17:06:07 +08:00
morgan 8616d12b27 flake: add CXP_DEMO variant build options 2024-08-22 10:28:30 +08:00
10 changed files with 2417 additions and 4 deletions

View File

@ -342,6 +342,7 @@
{
inherit fastnumbers artiq-netboot ramda migen-axi binutils-arm;
} //
(board-package-set { target = "zc706"; variant = "cxp_demo"; }) //
(board-package-set { target = "zc706"; variant = "nist_clock"; }) //
(board-package-set { target = "zc706"; variant = "nist_clock_master"; }) //
(board-package-set { target = "zc706"; variant = "nist_clock_master_100mhz"; }) //

69
src/gateware/cxp.py Normal file
View File

@ -0,0 +1,69 @@
from migen import *
from misoc.interconnect.csr import *
from misoc.interconnect import stream
from cxp_downconn import CXP_DownConn
from cxp_upconn import CXP_UpConn_PHY
from cxp_pipeline import *
class CXP(Module, AutoCSR):
def __init__(self, refclk, downconn_pads, upconn_pads, sys_clk_freq, debug_sma, pmod_pads):
self.submodules.upconn = UpConn_Interface(upconn_pads, sys_clk_freq, debug_sma, pmod_pads)
self.submodules.downconn = CXP_DownConn(refclk, downconn_pads, sys_clk_freq, debug_sma, pmod_pads)
# TODO: support the option high speed upconn
# TODO: add link layer
class UpConn_Interface(Module, AutoCSR):
def __init__(self, upconn_pads, sys_clk_freq, debug_sma, pmod_pads):
self.clk_reset = CSRStorage(reset=1)
self.bitrate2x_enable = CSRStorage()
self.tx_enable = CSRStorage()
self.tx_busy = CSRStatus()
# # #
layout = [("data", 8), ("k", 1)]
self.submodules.upconn_phy = upconn_phy = CXP_UpConn_PHY(upconn_pads, sys_clk_freq, debug_sma, pmod_pads, layout)
self.sync += [
upconn_phy.bitrate2x_enable.eq(self.bitrate2x_enable.storage),
upconn_phy.tx_enable.eq(self.tx_enable.storage),
upconn_phy.clk_reset.eq(self.clk_reset.re),
self.tx_busy.status.eq(upconn_phy.tx_busy),
]
# Packet FIFOs with transmission priority
# 0: Trigger packet
self.submodules.trig = trig = TX_Trigger(layout)
self.comb += trig.source.connect(upconn_phy.sinks[0])
# DEBUG: INPUT
self.trig_stb = CSR()
self.trig_delay = CSRStorage(8)
self.linktrigger = CSRStorage(2)
self.sync += [
trig.trig_stb.eq(self.trig_stb.re),
trig.delay.eq(self.trig_delay.storage),
trig.linktrig_mode.eq(self.linktrigger.storage),
]
# 1: IO acknowledgment for trigger packet
self.submodules.trig_ack = trig_ack = Trigger_ACK(layout)
self.comb += trig_ack.source.connect(upconn_phy.sinks[1])
# DEBUG: INPUT
self.ack = CSR()
self.sync += trig_ack.ack.eq(self.ack.re),
# 2: All other packets
# Control is not timing dependent, all the link layer is done in firmware
self.submodules.command = command = TX_Command_Packet(layout)
self.comb += command.source.connect(upconn_phy.sinks[2])

View File

@ -0,0 +1,87 @@
from migen.build.generic_platform import *
fmc_adapter_io = [
# CoaXPress high speed link
("CXP_HS", 0,
Subsignal("txp", Pins("HPC:DP0_C2M_P")),
Subsignal("txn", Pins("HPC:DP0_C2M_N")),
Subsignal("rxp", Pins("HPC:DP0_M2C_P")),
Subsignal("rxn", Pins("HPC:DP0_M2C_N")),
),
("CXP_HS", 1,
Subsignal("txp", Pins("HPC:DP1_C2M_P")),
Subsignal("txn", Pins("HPC:DP1_C2M_N")),
Subsignal("rxp", Pins("HPC:DP1_M2C_P")),
Subsignal("rxn", Pins("HPC:DP1_M2C_N")),
),
("CXP_HS", 2,
Subsignal("txp", Pins("HPC:DP2_C2M_P")),
Subsignal("txn", Pins("HPC:DP2_C2M_N")),
Subsignal("rxp", Pins("HPC:DP2_M2C_P")),
Subsignal("rxn", Pins("HPC:DP2_M2C_N")),
),
("CXP_HS", 3,
Subsignal("txp", Pins("HPC:DP3_C2M_P")),
Subsignal("txn", Pins("HPC:DP3_C2M_N")),
Subsignal("rxp", Pins("HPC:DP3_M2C_P")),
Subsignal("rxn", Pins("HPC:DP3_M2C_N")),
),
# CoaXPress low speed link
("CXP_LS", 0, Pins("HPC:LA00_CC_P"), IOStandard("LVCMOS33")),
("CXP_LS", 1, Pins("HPC:LA01_CC_N"), IOStandard("LVCMOS33")),
("CXP_LS", 2, Pins("HPC:LA01_CC_P"), IOStandard("LVCMOS33")),
("CXP_LS", 3, Pins("HPC:LA02_N"), IOStandard("LVCMOS33")),
# CoaXPress green and red LED
("CXP_LED", 0,
Subsignal("green", Pins("HPC:LA11_P"), IOStandard("LVCMOS33")),
Subsignal("red", Pins("HPC:LA11_N"), IOStandard("LVCMOS33")),
),
("CXP_LED", 1,
Subsignal("green", Pins("HPC:LA12_P"), IOStandard("LVCMOS33")),
Subsignal("red", Pins("HPC:LA12_N"), IOStandard("LVCMOS33")),
),
("CXP_LED", 2,
Subsignal("green", Pins("HPC:LA13_P"), IOStandard("LVCMOS33")),
Subsignal("red", Pins("HPC:LA13_N"), IOStandard("LVCMOS33")),
),
("CXP_LED", 3,
Subsignal("green", Pins("HPC:LA14_P"), IOStandard("LVCMOS33")),
Subsignal("red", Pins("HPC:LA14_N"), IOStandard("LVCMOS33")),
),
# Power over CoaXPress
("PoCXP", 0,
Subsignal("enable", Pins("HPC:LA21_N"), IOStandard("LVCMOS33")),
Subsignal("alert", Pins("HPC:LA18_CC_P"), IOStandard("LVCMOS33")),
),
("PoCXP", 1,
Subsignal("enable", Pins("HPC:LA21_P"), IOStandard("LVCMOS33")),
Subsignal("alert", Pins("HPC:LA19_N"), IOStandard("LVCMOS33")),
),
("PoCXP", 2,
Subsignal("enable", Pins("HPC:LA22_N"), IOStandard("LVCMOS33")),
Subsignal("alert", Pins("HPC:LA19_P"), IOStandard("LVCMOS33")),
),
("PoCXP", 3,
Subsignal("enable", Pins("HPC:LA22_P"), IOStandard("LVCMOS33")),
Subsignal("alert", Pins("HPC:LA20_N"), IOStandard("LVCMOS33")),
),
("i2c_fmc", 0,
Subsignal("scl", Pins("HPC:IIC_SCL")),
Subsignal("sda", Pins("HPC:IIC_SDA")),
IOStandard("LVCMOS33")
),
("3V3", 0, Pins("HPC:PG_M2C")),
("GND", 0, Pins("HPC:PRSNT_M2C_L HPC:CLK0_M2C_P")),
("VADJ", 0, Pins("HPC:GBTCLK1_M2C_N CLK0_M2C_N")),
("clk125_fmc", 0,
Subsignal("p", Pins("HPC:GBTCLK0_M2C_P")),
Subsignal("n", Pins("HPC:GBTCLK0_M2C_n")),
),
]

View File

@ -0,0 +1,794 @@
from migen import *
from migen.genlib.cdc import MultiReg
from migen.genlib.resetsync import AsyncResetSynchronizer
from misoc.cores.code_8b10b import Encoder, Decoder
from misoc.interconnect.csr import *
from artiq.gateware.drtio.transceiver.gtx_7series_init import *
from functools import reduce
from operator import add
class CXP_DownConn(Module, AutoCSR):
def __init__(self, refclk, pads, sys_clk_freq, debug_sma, pmod_pads):
nconn = len(pads)
self.rx_start_init = CSRStorage()
self.rx_restart = CSR()
self.tx_start_init = CSRStorage()
self.tx_restart = CSR()
self.txenable = CSRStorage()
self.rx_ready = CSRStatus(nconn)
self.qpll_reset = CSR()
self.qpll_locked = CSRStatus()
self.gtxs = []
# # #
self.submodules.qpll = qpll = QPLL(refclk, sys_clk_freq)
self.sync += [
qpll.reset.eq(self.qpll_reset.re),
self.qpll_locked.status.eq(qpll.lock),
]
for i in range(nconn):
if i != 0:
break
gtx = GTX(self.qpll, pads[i], sys_clk_freq, tx_mode="single", rx_mode="single")
self.gtxs.append(gtx)
setattr(self.submodules, "gtx"+str(i), gtx)
# TODO: add extension gtx connections
# TODO: add connection interface
# TODO: Connect slave cxp_gtx_rx clock tgt
# checkout channel interfaces & drtio_gtx
# GTPTXPhaseAlignement for inspiration
# Connect all GTX connections' DRP
self.gtx_daddr = CSRStorage(9)
self.gtx_dread = CSR()
self.gtx_din_stb = CSR()
self.gtx_din = CSRStorage(16)
self.gtx_dout = CSRStatus(16)
self.gtx_dready = CSR()
for gtx in self.gtxs:
self.sync += [
gtx.txenable.eq(self.txenable.storage[0]),
gtx.tx_restart.eq(self.tx_restart.re),
gtx.rx_restart.eq(self.rx_restart.re),
gtx.tx_init.clk_path_ready.eq(self.tx_start_init.storage),
gtx.rx_init.clk_path_ready.eq(self.rx_start_init.storage),
]
self.comb += gtx.dclk.eq(ClockSignal("sys"))
self.sync += [
gtx.den.eq(0),
gtx.dwen.eq(0),
If(self.gtx_dread.re,
gtx.den.eq(1),
gtx.daddr.eq(self.gtx_daddr.storage),
).Elif(self.gtx_din_stb.re,
gtx.den.eq(1),
gtx.dwen.eq(1),
gtx.daddr.eq(self.gtx_daddr.storage),
gtx.din.eq(self.gtx_din.storage),
),
]
# TODO: deal with 4 GTX instance of outpus
for n, gtx in enumerate(self.gtxs):
self.sync += [
self.rx_ready.status[n].eq(gtx.rx_ready),
If(gtx.dready,
self.gtx_dready.w.eq(1),
self.gtx_dout.status.eq(gtx.dout),
),
If(self.gtx_dready.re,
self.gtx_dready.w.eq(0),
),
]
# DEBUG: txusrclk PLL DRG
self.txpll_reset = CSRStorage()
self.pll_daddr = CSRStorage(7)
self.pll_dclk = CSRStorage()
self.pll_den = CSRStorage()
self.pll_din = CSRStorage(16)
self.pll_dwen = CSRStorage()
self.txpll_locked = CSRStatus()
self.pll_dout = CSRStatus(16)
self.pll_dready = CSRStatus()
self.txinit_phaligndone = CSRStatus()
self.rxinit_phaligndone = CSRStatus()
for n, gtx in enumerate(self.gtxs):
self.comb += [
gtx.txpll_reset.eq(self.txpll_reset.storage),
gtx.pll_daddr.eq(self.pll_daddr.storage),
gtx.pll_dclk.eq(self.pll_dclk.storage),
gtx.pll_den.eq(self.pll_den.storage),
gtx.pll_din.eq(self.pll_din.storage),
gtx.pll_dwen.eq(self.pll_dwen.storage),
self.txinit_phaligndone.status.eq(gtx.tx_init.Xxphaligndone),
self.rxinit_phaligndone.status.eq(gtx.rx_init.Xxphaligndone), self.txpll_locked.status.eq(gtx.txpll_locked),
self.pll_dout.status.eq(gtx.pll_dout),
self.pll_dready.status.eq(gtx.pll_dready),
]
# DEBUG:loopback
self.loopback_mode = CSRStorage(3)
self.comb += gtx.loopback_mode.eq(self.loopback_mode.storage)
# DEBUG: IO SMA & PMOD
if n == 0:
self.specials += [
# Instance("OBUF", i_I=gtx.cd_cxp_gtx_rx.clk, o_O=debug_sma.p_tx),
# Instance("OBUF", i_I=gtx.cd_cxp_gtx_tx.clk, o_O=debug_sma.n_rx),
# # pmod 0-7 pin
# Instance("OBUF", i_I=gtx.comma_checker.comma_aligned, o_O=pmod_pads[0]),
# Instance("OBUF", i_I=gtx.comma_checker.comma_det, o_O=pmod_pads[1]),
# Instance("OBUF", i_I=gtx.comma_checker.restart_sys, o_O=pmod_pads[2]),
# Instance("OBUF", i_I=gtx.comma_checker.aligner_en, o_O=pmod_pads[3]),
# Instance("OBUF", i_I=gtx.comma_checker.check_reset, o_O=pmod_pads[4]),
# Instance("OBUF", i_I=gtx.comma_checker.has_comma, o_O=pmod_pads[5]),
# Instance("OBUF", i_I=gtx.comma_checker.has_error, o_O=pmod_pads[6]),
# Instance("OBUF", i_I=gtx.comma_checker.ready_sys, o_O=pmod_pads[7]),
# Instance("OBUF", i_I=gtx.dclk, o_O=pmod_pads[0]),
# Instance("OBUF", i_I=gtx.den, o_O=pmod_pads[1]),
# Instance("OBUF", i_I=gtx.dwen, o_O=pmod_pads[2]),
# Instance("OBUF", i_I=gtx.dready, o_O=pmod_pads[3]),
]
# DEBUG: datain
self.sync.cxp_gtx_tx += [
gtx.encoder.d[0].eq(0xBC),
gtx.encoder.k[0].eq(1),
gtx.encoder.d[1].eq(0x3C),
gtx.encoder.k[1].eq(1),
gtx.encoder.d[2].eq(0x3C),
gtx.encoder.k[2].eq(1),
gtx.encoder.d[3].eq(0xB5),
gtx.encoder.k[3].eq(0),
]
for i in range(4):
gtx.decoders[i].input.attr.add("no_retiming")
gtx.decoders[i].d.attr.add("no_retiming")
gtx.decoders[i].k.attr.add("no_retiming")
rxdata_name = "rxdata_" + str(i)
rxdata_csr = CSRStatus(10, name=rxdata_name)
setattr(self, rxdata_name, rxdata_csr)
decoded_name = "decoded_data_" + str(i)
decoded_csr = CSRStatus(8, name=decoded_name)
setattr(self, decoded_name, decoded_csr)
k_name = "rxdata_" + str(i)
k_csr = CSRStatus(1, name=k_name)
setattr(self, k_name, k_csr)
self.sync.cxp_gtx_rx += [
rxdata_csr.status.eq(gtx.decoders[i].input),
decoded_csr.status.eq(gtx.decoders[i].d),
k_csr.status.eq(gtx.decoders[i].k),
]
class QPLL(Module, AutoCSR):
def __init__(self, refclk, sys_clk_freq):
self.clk = Signal()
self.refclk = Signal()
self.lock = Signal()
self.reset = Signal()
self.daddr = CSRStorage(8)
self.dread = CSR()
self.din_stb = CSR()
self.din = CSRStorage(16)
self.dout = CSRStatus(16)
self.dready = CSR()
# # #
# VCO @ 10GHz, linerate = 1.25Gbps
# feedback divider = 80
qpll_fbdiv = 0b0100100000
qpll_fbdiv_ratio = 1
refclk_div = 1
self.Xxout_div = 8
# DEBUG: txuserclk
fbdiv_real = 80
self.tx_usrclk_freq = (sys_clk_freq*fbdiv_real/self.Xxout_div)/40
dready = Signal()
self.specials += [
Instance("GTXE2_COMMON",
i_QPLLREFCLKSEL=0b001,
i_GTREFCLK0=refclk,
i_QPLLPD=0,
i_QPLLRESET=self.reset,
i_QPLLLOCKEN=1,
o_QPLLLOCK=self.lock,
o_QPLLOUTCLK=self.clk,
o_QPLLOUTREFCLK=self.refclk,
# See UG476 (v1.12.1) Table 2-16
p_QPLL_FBDIV=qpll_fbdiv,
p_QPLL_FBDIV_RATIO=qpll_fbdiv_ratio,
p_QPLL_REFCLK_DIV=refclk_div,
# From 7 Series FPGAs Transceivers Wizard
p_BIAS_CFG=0x0000040000001000,
p_COMMON_CFG=0x00000000,
p_QPLL_CFG=0x0680181,
p_QPLL_CLKOUT_CFG=0b0000,
p_QPLL_COARSE_FREQ_OVRD=0b010000,
p_QPLL_COARSE_FREQ_OVRD_EN=0b0,
p_QPLL_CP=0b0000011111,
p_QPLL_CP_MONITOR_EN=0b0,
p_QPLL_DMONITOR_SEL=0b0,
p_QPLL_FBDIV_MONITOR_EN= 0b0,
p_QPLL_INIT_CFG=0x000006,
p_QPLL_LOCK_CFG=0x21E8,
p_QPLL_LPF=0b1111,
# Reserved, values cannot be modified
i_BGBYPASSB=0b1,
i_BGMONITORENB=0b1,
i_BGPDB=0b1,
i_BGRCALOVRD=0b11111,
i_RCALENB=0b1,
i_QPLLRSVD1=0b0,
i_QPLLRSVD2=0b11111,
# Dynamic Reconfiguration Ports
i_DRPADDR=self.daddr.storage,
i_DRPCLK=ClockSignal("sys"),
i_DRPEN=(self.dread.re | self.din_stb.re),
i_DRPWE=self.din_stb.re,
i_DRPDI=self.din.storage,
o_DRPDO=self.dout.status,
o_DRPRDY=dready,
)
]
self.sync += [
If(dready,
self.dready.w.eq(1),
),
If(self.dready.re,
self.dready.w.eq(0),
),
]
# Warning: Xilinx transceivers are LSB first, and comma needs to be flipped
# compared to the usual 8b10b binary representation.
class Comma_Checker(Module):
def __init__(self, comma, reset_period=10_000_000):
self.data = Signal(20)
self.comma_aligned = Signal()
self.comma_realigned = Signal()
self.comma_det = Signal()
self.aligner_en = Signal()
self.ready_sys = Signal()
self.restart_sys = Signal()
# # #
# periodically reset rx until rx is connected and receiving valid data
# as after connecting RXP/RXN, the whole RX need to be reset
reset_counter = Signal(reset=reset_period-1, max=reset_period)
self.sync += [
self.restart_sys.eq(0),
If(~self.ready_sys,
If(reset_counter == 0,
reset_counter.eq(reset_counter.reset),
self.restart_sys.eq(1),
).Else(
reset_counter.eq(reset_counter - 1),
)
)
]
# Data and comma checker
# From UG476 (v1.12.1) p.228
# The built-in RXBYTEISALIGNED can be falsely asserted at linerate higher than 5Gbps
# The validity of data and comma needed to be checked externally
comma_n = ~comma & 0b1111111111
# DEBUG: remove after use
self.has_comma = Signal()
self.has_error = Signal()
comma_seen = Signal()
error_seen = Signal()
one_counts = Signal(max=11)
# From CXP-001-2021 section 9.2.5.1
# For high speed connection an IDLE word shall be transmitted at least once every 100 words
counter_period = 200
counter = Signal(reset=counter_period-1, max=counter_period)
check_reset = Signal()
check = Signal()
self.sync.cxp_gtx_rx += [
If(check_reset,
counter.eq(counter.reset),
check.eq(0),
).Elif(counter == 0,
check.eq(1),
).Else(
counter.eq(counter - 1),
),
If(check_reset,
comma_seen.eq(0),
).Elif((self.data[:10] == comma) | (self.data[:10] == comma_n),
comma_seen.eq(1)
),
one_counts.eq(reduce(add, [self.data[i] for i in range(10)])),
If(check_reset,
error_seen.eq(0),
).Elif((one_counts != 4) & (one_counts != 5) & (one_counts != 6),
error_seen.eq(1),
),
# DEBUG:
self.has_comma.eq(0),
If((self.data[:10] == comma) | (self.data[:10] == comma_n),
self.has_comma.eq(1),
),
self.has_error.eq(0),
If((one_counts != 4) & (one_counts != 5) & (one_counts != 6),
self.has_error.eq(1),
),
]
# DEBUG: expose signal
self.check_reset = Signal()
self.comb +=[
self.check_reset.eq(check_reset),
]
self.submodules.rxfsm = rxfsm = ClockDomainsRenamer("cxp_gtx_rx")(FSM(reset_state="WAIT_COMMA"))
rxfsm.act("WAIT_COMMA",
If(self.comma_det,
NextState("ALIGNING"),
)
)
rxfsm.act("ALIGNING",
If(self.comma_aligned & (~self.comma_realigned),
NextState("WAIT_ALIGNED_DATA"),
).Else(
self.aligner_en.eq(1),
)
)
# wait for the aligned data to arrive at the FPGA RX interface
# as there is a delay before the data is avaiable after RXBYTEISALIGNED is asserted
self.submodules.timer = timer = ClockDomainsRenamer("cxp_gtx_rx")(WaitTimer(10_000))
rxfsm.act("WAIT_ALIGNED_DATA",
timer.wait.eq(1),
If(timer.done,
check_reset.eq(1),
NextState("CHECKING"),
)
)
rxfsm.act("CHECKING",
If(check,
check_reset.eq(1),
If(comma_seen & (~error_seen),
NextState("READY"),
).Else(
NextState("WAIT_COMMA")
)
)
)
ready = Signal()
self.specials += MultiReg(ready, self.ready_sys)
rxfsm.act("READY",
ready.eq(1),
If(check,
check_reset.eq(1),
If(~(comma_seen & (~error_seen)),
NextState("WAIT_COMMA"),
)
)
)
class GTX(Module):
# Settings:
# * GTX reference clock @ 125MHz
# * GTX data width = 20
# * GTX PLL frequency @ 3.125GHz
# * GTX line rate (TX & RX) @ 3.125Gb/s
# * GTX TX/RX USRCLK @ PLL/datawidth = 156MHz
def __init__(self, qpll, pads, sys_clk_freq, tx_mode="single", rx_mode="single"):
assert tx_mode in ["single", "master", "slave"]
assert rx_mode in ["single", "master", "slave"]
# linerate = USRCLK * datawidth
pll_fbout_mult = 8
txusr_pll_div = pll_fbout_mult*sys_clk_freq/qpll.tx_usrclk_freq
self.tx_restart = Signal()
self.rx_restart = Signal()
self.loopback_mode = Signal(3)
self.txenable = Signal()
self.rx_ready = Signal()
# Dynamic Reconfiguration Ports
self.daddr = Signal(9)
self.dclk = Signal()
self.den = Signal()
self.dwen = Signal()
self.din = Signal(16)
self.dout = Signal(16)
self.dready = Signal()
self.submodules.encoder = ClockDomainsRenamer("cxp_gtx_tx")(Encoder(4, True))
self.submodules.decoders = [ClockDomainsRenamer("cxp_gtx_rx")(
(Decoder(True))) for _ in range(4)]
# transceiver direct clock outputs
# useful to specify clock constraints in a way palatable to Vivado
self.txoutclk = Signal()
self.rxoutclk = Signal()
# # #
# TX generates cxp_tx clock, init must be in system domain
# FIXME: 500e6 is used to fix Xx reset by holding gtxXxreset for a couple cycle more
self.submodules.tx_init = tx_init = GTXInit(500e6, False, mode=tx_mode)
self.submodules.rx_init = rx_init = GTXInit(sys_clk_freq, True, mode=rx_mode)
# RX receives restart commands from txusrclk domain
# self.submodules.rx_init = rx_init = ClockDomainsRenamer("cxp_gtx_tx")(GTXInit(500e6, True, mode=rx_mode))
self.comb += [
tx_init.cplllock.eq(qpll.lock),
rx_init.cplllock.eq(qpll.lock)
]
txdata = Signal(40)
rxdata = Signal(40)
comma_aligned = Signal()
comma_realigned = Signal()
comma_det = Signal()
comma_aligner_en = Signal()
# Note: the following parameters were set after consulting AR45360
self.specials += \
Instance("GTXE2_CHANNEL",
# PMA Attributes
p_PMA_RSV=0x001E7080,
p_PMA_RSV2=0x2050, # PMA_RSV2[5] = 0: Eye scan feature disabled
p_PMA_RSV3=0,
p_PMA_RSV4=1, # PMA_RSV[4],RX_CM_TRIM[2:0] = 0b1010: Common mode 800mV
p_RX_BIAS_CFG=0b000000000100,
p_RX_OS_CFG=0b0000010000000,
p_RX_CLK25_DIV=5,
p_TX_CLK25_DIV=5,
# Power-Down Attributes
p_PD_TRANS_TIME_FROM_P2=0x3c,
p_PD_TRANS_TIME_NONE_P2=0x3c,
p_PD_TRANS_TIME_TO_P2=0x64,
i_CPLLPD=1,
# QPLL
i_QPLLCLK=qpll.clk,
i_QPLLREFCLK=qpll.refclk,
p_RXOUT_DIV=qpll.Xxout_div,
p_TXOUT_DIV=qpll.Xxout_div,
i_RXSYSCLKSEL=0b11, # use QPLL & QPLL's REFCLK
i_TXSYSCLKSEL=0b11, # use QPLL & CPLL's REFCLK
# TX clock
p_TXBUF_EN="FALSE",
p_TX_XCLK_SEL="TXUSR",
o_TXOUTCLK=self.txoutclk,
# i_TXSYSCLKSEL=0b00,
i_TXOUTCLKSEL=0b11,
# TX Startup/Reset
i_TXPHDLYRESET=0,
i_TXDLYBYPASS=0,
i_TXPHALIGNEN=1 if tx_mode != "single" else 0,
i_GTTXRESET=tx_init.gtXxreset,
o_TXRESETDONE=tx_init.Xxresetdone,
i_TXDLYSRESET=tx_init.Xxdlysreset,
o_TXDLYSRESETDONE=tx_init.Xxdlysresetdone,
i_TXPHINIT=tx_init.txphinit if tx_mode != "single" else 0,
o_TXPHINITDONE=tx_init.txphinitdone if tx_mode != "single" else Signal(),
i_TXPHALIGN=tx_init.Xxphalign if tx_mode != "single" else 0,
i_TXDLYEN=tx_init.Xxdlyen if tx_mode != "single" else 0,
o_TXPHALIGNDONE=tx_init.Xxphaligndone,
i_TXUSERRDY=tx_init.Xxuserrdy,
p_TXPMARESET_TIME=1,
p_TXPCSRESET_TIME=1,
i_TXINHIBIT=~self.txenable,
# TX data
p_TX_DATA_WIDTH=40,
p_TX_INT_DATAWIDTH=1, # 1 if a line rate is greater than 6.6 Gbps
i_TXCHARDISPMODE=Cat(txdata[9], txdata[19], txdata[29], txdata[39]),
i_TXCHARDISPVAL=Cat(txdata[8], txdata[18], txdata[28], txdata[38]),
i_TXDATA=Cat(txdata[:8], txdata[10:18], txdata[20:28], txdata[30:38]),
i_TXUSRCLK=ClockSignal("cxp_gtx_tx"),
i_TXUSRCLK2=ClockSignal("cxp_gtx_tx"),
# TX electrical
i_TXBUFDIFFCTRL=0b100,
i_TXDIFFCTRL=0b1000,
# RX Startup/Reset
i_RXPHDLYRESET=0,
i_RXDLYBYPASS=0,
i_RXPHALIGNEN=1 if rx_mode != "single" else 0,
i_GTRXRESET=rx_init.gtXxreset,
o_RXRESETDONE=rx_init.Xxresetdone,
i_RXDLYSRESET=rx_init.Xxdlysreset,
o_RXDLYSRESETDONE=rx_init.Xxdlysresetdone,
i_RXPHALIGN=rx_init.Xxphalign if rx_mode != "single" else 0,
i_RXDLYEN=rx_init.Xxdlyen if rx_mode != "single" else 0,
o_RXPHALIGNDONE=rx_init.Xxphaligndone,
i_RXUSERRDY=rx_init.Xxuserrdy,
p_RXPMARESET_TIME=1,
p_RXPCSRESET_TIME=1,
# RX AFE
p_RX_DFE_XYD_CFG=0,
p_RX_CM_SEL=0b11, # RX_CM_SEL = 0b11: Common mode is programmable
p_RX_CM_TRIM=0b010, # PMA_RSV[4],RX_CM_TRIM[2:0] = 0b1010: Common mode 800mV
i_RXDFEXYDEN=1,
i_RXDFEXYDHOLD=0,
i_RXDFEXYDOVRDEN=0,
i_RXLPMEN=1, # RXLPMEN = 1: LPM mode is enable for non scramble 8b10b data
p_RXLPM_HF_CFG=0b00000011110000,
p_RXLPM_LF_CFG=0b00000011110000,
p_RX_DFE_GAIN_CFG=0x0207EA,
p_RX_DFE_VP_CFG=0b00011111100000011,
p_RX_DFE_UT_CFG=0b10001000000000000,
p_RX_DFE_KL_CFG=0b0000011111110,
p_RX_DFE_KL_CFG2=0x3788140A,
p_RX_DFE_H2_CFG=0b000110000000,
p_RX_DFE_H3_CFG=0b000110000000,
p_RX_DFE_H4_CFG=0b00011100000,
p_RX_DFE_H5_CFG=0b00011100000,
p_RX_DFE_LPM_CFG=0x0904, # RX_DFE_LPM_CFG = 0x0904: linerate <= 6.6Gb/s
# = 0x0104: linerate > 6.6Gb/s
# RX clock
i_RXDDIEN=1,
# i_RXSYSCLKSEL=0b00,
i_RXOUTCLKSEL=0b010,
o_RXOUTCLK=self.rxoutclk,
i_RXUSRCLK=ClockSignal("cxp_gtx_rx"),
i_RXUSRCLK2=ClockSignal("cxp_gtx_rx"),
# RX Clock Correction Attributes
p_CLK_CORRECT_USE="FALSE",
p_CLK_COR_SEQ_1_1=0b0100000000,
p_CLK_COR_SEQ_2_1=0b0100000000,
p_CLK_COR_SEQ_1_ENABLE=0b1111,
p_CLK_COR_SEQ_2_ENABLE=0b1111,
# RX data
p_RX_DATA_WIDTH=40,
p_RX_INT_DATAWIDTH=1, # 1 if a line rate is greater than 6.6 Gbps
o_RXDISPERR=Cat(rxdata[9], rxdata[19], rxdata[29], rxdata[39]),
o_RXCHARISK=Cat(rxdata[8], rxdata[18], rxdata[28], rxdata[38]),
o_RXDATA=Cat(rxdata[:8], rxdata[10:18], rxdata[20:28], rxdata[30:38]),
# RX Byte and Word Alignment Attributes
p_ALIGN_COMMA_DOUBLE="FALSE",
p_ALIGN_COMMA_ENABLE=0b1111111111,
p_ALIGN_COMMA_WORD=4, # align comma to rxdata[:10] only
p_ALIGN_MCOMMA_DET="TRUE",
p_ALIGN_MCOMMA_VALUE=0b1010000011,
p_ALIGN_PCOMMA_DET="TRUE",
p_ALIGN_PCOMMA_VALUE=0b0101111100,
p_SHOW_REALIGN_COMMA="FALSE",
p_RXSLIDE_AUTO_WAIT=7,
p_RXSLIDE_MODE="OFF",
p_RX_SIG_VALID_DLY=10,
i_RXPCOMMAALIGNEN=comma_aligner_en,
i_RXMCOMMAALIGNEN=comma_aligner_en,
i_RXCOMMADETEN=1,
i_RXSLIDE=0,
o_RXBYTEISALIGNED=comma_aligned,
o_RXBYTEREALIGN=comma_realigned,
o_RXCOMMADET=comma_det,
# RX 8B/10B Decoder Attributes
p_RX_DISPERR_SEQ_MATCH="FALSE",
p_DEC_MCOMMA_DETECT="TRUE",
p_DEC_PCOMMA_DETECT="TRUE",
p_DEC_VALID_COMMA_ONLY="FALSE",
# RX Buffer Attributes
p_RXBUF_ADDR_MODE="FAST",
p_RXBUF_EIDLE_HI_CNT=0b1000,
p_RXBUF_EIDLE_LO_CNT=0b0000,
p_RXBUF_EN="FALSE",
p_RX_BUFFER_CFG=0b000000,
p_RXBUF_RESET_ON_CB_CHANGE="TRUE",
p_RXBUF_RESET_ON_COMMAALIGN="FALSE",
p_RXBUF_RESET_ON_EIDLE="FALSE", # RXBUF_RESET_ON_EIDLE = FALSE: OOB is disabled
p_RXBUF_RESET_ON_RATE_CHANGE="TRUE",
p_RXBUFRESET_TIME=0b00001,
p_RXBUF_THRESH_OVFLW=61,
p_RXBUF_THRESH_OVRD="FALSE",
p_RXBUF_THRESH_UNDFLW=4,
p_RXDLY_CFG=0x001F,
p_RXDLY_LCFG=0x030,
p_RXDLY_TAP_CFG=0x0000,
p_RXPH_CFG=0xC00002,
p_RXPHDLY_CFG=0x084020,
p_RXPH_MONITOR_SEL=0b00000,
p_RX_XCLK_SEL="RXUSR",
p_RX_DDI_SEL=0b000000,
p_RX_DEFER_RESET_BUF_EN="TRUE",
# CDR Attributes
p_RXCDR_CFG=0x03_0000_23FF_1008_0020, # LPM @ 0.5G-1.5625G , 8B/10B encoded data, CDR setting < +/- 200ppm
# (See UG476 (v1.12.1), p.206)
p_RXCDR_FR_RESET_ON_EIDLE=0b0,
p_RXCDR_HOLD_DURING_EIDLE=0b0,
p_RXCDR_PH_RESET_ON_EIDLE=0b0,
p_RXCDR_LOCK_CFG=0b010101,
# Pads
i_GTXRXP=pads.rxp,
i_GTXRXN=pads.rxn,
o_GTXTXP=pads.txp,
o_GTXTXN=pads.txn,
# Dynamic Reconfiguration Ports
p_IS_DRPCLK_INVERTED=0b0,
i_DRPADDR=self.daddr,
i_DRPCLK=self.dclk,
i_DRPEN=self.den,
i_DRPWE=self.dwen,
i_DRPDI=self.din,
o_DRPDO=self.dout,
o_DRPRDY=self.dready,
# ! loopback for debugging
i_LOOPBACK = self.loopback_mode,
p_TX_LOOPBACK_DRIVE_HIZ = "FALSE",
p_RXPRBS_ERR_LOOPBACK = 0b0,
# Other parameters
p_PCS_RSVD_ATTR=(
(tx_mode != "single") << 1 | # PCS_RSVD_ATTR[1] = 0: TX Single Lane Auto Mode
# = 1: TX Manual Mode
(rx_mode != "single") << 2 | # [2] = 0: RX Single Lane Auto Mode
# = 1: RX Manual Mode
0 << 8 # [8] = 0: OOB is disabled
),
i_RXELECIDLEMODE=0b11, # RXELECIDLEMODE = 0b11: OOB is disabled
p_RX_DFE_LPM_HOLD_DURING_EIDLE=0b0,
p_ES_EYE_SCAN_EN="TRUE", # Must be TRUE for GTX
)
# TX clocking
# A PLL is used to generate the correct frequency for TXUSRCLK (UG476 Equation 3-1)
self.clock_domains.cd_cxp_gtx_tx = ClockDomain()
txpll_fb_clk = Signal()
txoutclk_buf = Signal()
txpll_clkout = Signal()
self.txpll_reset = Signal()
self.pll_daddr = Signal(7)
self.pll_dclk = Signal()
self.pll_den = Signal()
self.pll_din = Signal(16)
self.pll_dwen = Signal()
self.txpll_locked = Signal()
self.pll_dout = Signal(16)
self.pll_dready = Signal()
self.specials += [
Instance("PLLE2_ADV",
p_BANDWIDTH="HIGH",
o_LOCKED=self.txpll_locked,
i_RST=self.txpll_reset,
p_CLKIN1_PERIOD=1e9/sys_clk_freq, # ns
i_CLKIN1=txoutclk_buf,
# VCO @ 1.25GHz
p_CLKFBOUT_MULT=pll_fbout_mult, p_DIVCLK_DIVIDE=1,
i_CLKFBIN=txpll_fb_clk, o_CLKFBOUT=txpll_fb_clk,
# frequency = linerate/40
p_CLKOUT0_DIVIDE=txusr_pll_div, p_CLKOUT0_PHASE=0.0, o_CLKOUT0=txpll_clkout,
# Dynamic Reconfiguration Ports
i_DADDR = self.pll_daddr,
i_DCLK = self.pll_dclk,
i_DEN = self.pll_den,
i_DI = self.pll_din,
i_DWE = self.pll_dwen,
o_DO = self.pll_dout,
o_DRDY = self.pll_dready,
),
Instance("BUFG", i_I=self.txoutclk, o_O=txoutclk_buf),
Instance("BUFG", i_I=txpll_clkout, o_O=self.cd_cxp_gtx_tx.clk),
AsyncResetSynchronizer(self.cd_cxp_gtx_tx, ~self.txpll_locked & ~tx_init.done)
]
# RX clocking
# the CDR matches the required frequency for RXUSRCLK, no need for PLL
self.clock_domains.cd_cxp_gtx_rx = ClockDomain()
self.specials += [
Instance("BUFG", i_I=self.rxoutclk, o_O=self.cd_cxp_gtx_rx.clk),
AsyncResetSynchronizer(self.cd_cxp_gtx_rx, ~rx_init.done)
]
self.comb += [
txdata.eq(Cat(self.encoder.output[0], self.encoder.output[1], self.encoder.output[2], self.encoder.output[3])),
self.decoders[0].input.eq(rxdata[:10]),
self.decoders[1].input.eq(rxdata[10:20]),
self.decoders[2].input.eq(rxdata[20:30]),
self.decoders[3].input.eq(rxdata[30:]),
]
self.submodules.comma_checker = comma_checker = Comma_Checker(0b0101111100)
self.comb += [
comma_checker.data.eq(rxdata),
comma_checker.comma_aligned.eq(comma_aligned),
comma_checker.comma_realigned.eq(comma_realigned),
comma_checker.comma_det.eq(comma_det),
comma_aligner_en.eq(comma_checker.aligner_en),
self.rx_ready.eq(comma_checker.ready_sys),
rx_init.restart.eq(self.rx_restart | comma_checker.restart_sys),
tx_init.restart.eq(self.tx_restart),
]

View File

@ -0,0 +1,280 @@
from migen import *
from misoc.interconnect.csr import *
from misoc.interconnect import stream
from misoc.cores.liteeth_mini.mac.crc import LiteEthMACCRCEngine, LiteEthMACCRCChecker
def K(x, y):
return ((y << 5) | x)
class Code_Source(Module):
def __init__(self, layout, counts=4):
self.source = stream.Endpoint(layout)
self.stb = Signal()
self.data = Signal.like(self.source.data)
self.k = Signal.like(self.source.k)
# # #
cnt = Signal(max=counts)
clr_cnt = Signal()
inc_cnt = Signal()
self.sync += [
If(clr_cnt,
cnt.eq(cnt.reset),
).Elif(inc_cnt,
cnt.eq(cnt + 1),
)
]
self.submodules.fsm = fsm = FSM(reset_state="IDLE")
fsm.act("IDLE",
clr_cnt.eq(1),
If(self.stb,
NextState("WRITE")
)
)
fsm.act("WRITE",
self.source.stb.eq(1),
self.source.data.eq(self.data),
self.source.k.eq(self.k),
If(cnt == counts - 1,
self.source.eop.eq(1),
If(self.source.ack, NextState("IDLE"))
).Else(
inc_cnt.eq(self.source.ack)
)
)
class Code_Inserter(Module):
def __init__(self, layout, insert_infront=True, counts=4):
self.sink = sink = stream.Endpoint(layout)
self.source = source = stream.Endpoint(layout)
self.data = Signal.like(sink.data)
self.k = Signal.like(sink.k)
# # #
assert counts > 0
cnt = Signal() if counts == 1 else Signal(max=counts)
clr_cnt = Signal()
inc_cnt = Signal()
self.sync += [
If(clr_cnt,
cnt.eq(cnt.reset),
).Elif(inc_cnt,
cnt.eq(cnt + 1),
)
]
self.submodules.fsm = fsm = FSM(reset_state="IDLE")
if insert_infront:
fsm.act("IDLE",
sink.ack.eq(1),
clr_cnt.eq(1),
If(sink.stb,
sink.ack.eq(0),
NextState("INSERT"),
)
)
fsm.act("INSERT",
sink.ack.eq(0),
source.stb.eq(1),
source.data.eq(self.data),
source.k.eq(self.k),
If(cnt == counts - 1,
If(source.ack, NextState("COPY"))
).Else(
inc_cnt.eq(source.ack)
)
)
fsm.act("COPY",
sink.connect(source),
If(sink.stb & sink.eop & source.ack,
NextState("IDLE"),
)
)
else:
fsm.act("IDLE",
sink.ack.eq(1),
clr_cnt.eq(1),
If(sink.stb,
sink.ack.eq(0),
NextState("COPY"),
)
)
fsm.act("COPY",
sink.connect(source),
source.eop.eq(0),
If(sink.stb & sink.eop & source.ack,
NextState("INSERT"),
)
)
fsm.act("INSERT",
sink.ack.eq(0),
source.stb.eq(1),
source.data.eq(self.data),
source.k.eq(self.k),
If(cnt == counts - 1,
source.eop.eq(1),
If(source.ack, NextState("IDLE"))
).Else(
inc_cnt.eq(source.ack)
),
)
class Packet_Wrapper(Module):
def __init__(self, layout):
self.submodules.pak_start = pak_start = Code_Inserter(layout)
self.submodules.pak_end = pak_end = Code_Inserter(layout, insert_infront=False)
self.sink = pak_start.sink
self.source = pak_end.source
self.comb += [
pak_start.data.eq(K(27, 7)),
pak_start.k.eq(1),
pak_end.data.eq(K(29, 7)),
pak_end.k.eq(1),
pak_start.source.connect(pak_end.sink),
]
@ResetInserter()
@CEInserter()
class CXPCRC32(Module):
# Section 9.2.2.2 (CXP-001-2021)
width = 32
polynom = 0x04C11DB7
seed = 2**width-1
check = 0x00000000
def __init__(self, data_width):
self.data = Signal(data_width)
self.value = Signal(self.width)
self.error = Signal()
# # #
self.submodules.engine = LiteEthMACCRCEngine(data_width, self.width, self.polynom)
reg = Signal(self.width, reset=self.seed)
self.sync += reg.eq(self.engine.next)
self.comb += [
self.engine.data.eq(self.data),
self.engine.last.eq(reg),
self.value.eq(reg[::-1]),
self.error.eq(self.engine.next != self.check)
]
class CXPCRC32Checker(LiteEthMACCRCChecker):
def __init__(self, layout):
LiteEthMACCRCChecker.__init__(self, CXPCRC32, layout)
class TX_Trigger(Module, AutoCSR):
def __init__(self, layout):
self.trig_stb = Signal()
self.delay = Signal(8)
self.linktrig_mode = Signal(max=4)
# # #
self.submodules.code_src = code_src = Code_Source(layout, counts=3)
self.comb += [
code_src.stb.eq(self.trig_stb),
code_src.data.eq(self.delay),
code_src.k.eq(0)
]
self.submodules.inserter_once = inserter_once = Code_Inserter(layout, counts=1)
self.submodules.inserter_twice = inserter_twice = Code_Inserter(layout, counts=2)
self.comb += [
inserter_once.k.eq(1),
inserter_twice.k.eq(1),
If((self.linktrig_mode == 0) | (self.linktrig_mode == 2),
inserter_once.data.eq(K(28, 2)),
inserter_twice.data.eq(K(28, 4)),
).Else(
inserter_once.data.eq(K(28, 4)),
inserter_twice.data.eq(K(28, 2)),
)
]
tx_pipeline = [ code_src, inserter_twice, inserter_once]
for s, d in zip(tx_pipeline, tx_pipeline[1:]):
self.comb += s.source.connect(d.sink)
self.source = tx_pipeline[-1].source
class Trigger_ACK(Module):
def __init__(self, layout):
self.ack = Signal()
# # #
# Section 9.3.2 (CXP-001-2021)
# Send 4x K28.6 and 4x 0x01 as trigger packet ack
self.submodules.code_src = code_src = Code_Source(layout)
self.submodules.k_code_inserter = k_code_inserter = Code_Inserter(layout)
self.comb += [
code_src.stb.eq(self.ack),
code_src.data.eq(0x01),
code_src.k.eq(0),
k_code_inserter.data.eq(K(28, 6)),
k_code_inserter.k.eq(1),
code_src.source.connect(k_code_inserter.sink)
]
self.source = k_code_inserter.source
class TX_Command_Packet(Module, AutoCSR):
def __init__(self, layout):
self.len = CSRStorage(6)
self.data = CSR(8)
self.writeable = CSRStatus()
# # #
# TODO: use RAM instead of FIFO ?
# Section 12.1.2 (CXP-001-2021)
# Max control packet size is 128 bytes
# NOTE: The firmware will lock up if there is not enough space for the packet
self.submodules.fifo = fifo = stream.SyncFIFO(layout, 128)
self.submodules.pak_wrp = pak_wrp = Packet_Wrapper(layout)
self.source = pak_wrp.source
self.comb += fifo.source.connect(pak_wrp.sink)
len = Signal(6, reset=1)
self.sync += [
self.writeable.status.eq(fifo.sink.ack),
If(fifo.sink.ack, fifo.sink.stb.eq(0)),
If(self.data.re,
fifo.sink.stb.eq(1),
fifo.sink.data.eq(self.data.r),
fifo.sink.k.eq(0),
If(len == self.len.storage,
fifo.sink.eop.eq(1),
len.eq(len.reset),
).Else(
fifo.sink.eop.eq(0),
len.eq(len + 1),
),
)
]

381
src/gateware/cxp_upconn.py Normal file
View File

@ -0,0 +1,381 @@
from math import ceil
from migen import *
from migen.genlib.coding import PriorityEncoder
from misoc.cores.code_8b10b import SingleEncoder
from misoc.interconnect import stream
from misoc.interconnect.csr import *
IDLE_CHARS = Array([
#[char, k]
[0xBC, 1], #K28.5
[0x3C, 1], #K28.1
[0x3C, 1], #K28.1
[0xB5, 0], #D21.5
])
@ResetInserter()
class UpConn_ClockGen(Module):
def __init__(self, sys_clk_freq):
self.clk = Signal()
self.clk_10x = Signal() # 20.83MHz 48ns or 41.66MHz 24ns
self.freq2x_enable = Signal()
# # #
period = 1e9/sys_clk_freq
max_count = ceil(48/period)
counter = Signal(max=max_count, reset=max_count-1)
clk_div = Signal(max=10, reset=9)
self.sync += [
self.clk.eq(0),
self.clk_10x.eq(0),
If(counter == 0,
self.clk_10x.eq(1),
If(self.freq2x_enable,
counter.eq(int(max_count/2)-1),
).Else(
counter.eq(counter.reset),
),
).Else(
counter.eq(counter-1),
),
If(counter == 0,
If(clk_div == 0,
self.clk.eq(1),
clk_div.eq(clk_div.reset),
).Else(
clk_div.eq(clk_div-1),
)
)
]
@ResetInserter()
@CEInserter()
class SERDES_10bits(Module):
def __init__(self, pad):
self.oe = Signal()
self.d = Signal(10)
# # #
o = Signal()
tx_bitcount = Signal(max=10)
tx_reg = Signal(10)
# DEBUG:
self.o = Signal()
self.comb += self.o.eq(o)
self.specials += Instance("OBUF", i_I=o, o_O=pad),
self.sync += [
If(self.oe,
# send LSB first
o.eq(tx_reg[0]),
tx_reg.eq(Cat(tx_reg[1:], 0)),
tx_bitcount.eq(tx_bitcount + 1),
If(tx_bitcount == 9,
tx_bitcount.eq(0),
tx_reg.eq(self.d),
),
).Else(
o.eq(0),
tx_bitcount.eq(0),
)
]
@ResetInserter()
@CEInserter()
class Packets_Scheduler(Module):
def __init__(self, tx_fifos, debug_buf):
self.tx_enable = Signal()
self.oe = Signal()
# # #
self.submodules.startup_fsm = startup_fsm = FSM(reset_state="WAIT_TX_ENABLE")
self.submodules.encoder = encoder = SingleEncoder(True)
tx_charcount = Signal(max=4)
tx_wordcount = Signal(max=10000)
idling = Signal()
priorities = Signal.like(tx_fifos.pe.o)
# DEBUG:
self.idling = Signal()
self.tx_charcount = Signal(max=4)
self.comb += [
self.idling.eq(idling),
self.tx_charcount.eq(tx_charcount),
]
startup_fsm.act("WAIT_TX_ENABLE",
If(self.tx_enable,
NextValue(idling, 1),
NextValue(tx_charcount, 0),
NextValue(encoder.d, IDLE_CHARS[0][0]),
NextValue(encoder.k, IDLE_CHARS[0][1]),
NextState("START_TX"),
# DEBUG:
If(debug_buf.sink_ack,
NextValue(debug_buf.sink_stb, 1),
NextValue(debug_buf.sink_data, IDLE_CHARS[0][0]),
NextValue(debug_buf.sink_k, IDLE_CHARS[0][1]),
)
)
)
startup_fsm.act("START_TX",
self.oe.eq(1),
If((~self.tx_enable) & (tx_charcount == 3),
NextState("WAIT_TX_ENABLE")
)
)
self.sync += [
If(self.oe,
encoder.disp_in.eq(encoder.disp_out),
If((~tx_fifos.pe.n) & (tx_fifos.pe.o == 0),
# trigger packets are inserted at char boundary and don't contribute to word count
tx_fifos.source_ack[0].eq(1),
encoder.d.eq(tx_fifos.source_data[0]),
encoder.k.eq(tx_fifos.source_k[0]),
# DEBUG:
If(debug_buf.sink_ack,
debug_buf.sink_stb.eq(1),
debug_buf.sink_data.eq(tx_fifos.source_data[0]),
debug_buf.sink_k.eq(tx_fifos.source_k[0]),
)
).Else(
If(tx_charcount == 3,
tx_charcount.eq(0),
# Section 9.2.4 (CXP-001-2021)
# other priorities packets are inserted at word boundary
If((~tx_fifos.pe.n) & (tx_wordcount != 9999),
idling.eq(0),
priorities.eq(tx_fifos.pe.o),
tx_wordcount.eq(tx_wordcount + 1),
tx_fifos.source_ack[tx_fifos.pe.o].eq(1),
encoder.d.eq(tx_fifos.source_data[tx_fifos.pe.o]),
encoder.k.eq(tx_fifos.source_k[tx_fifos.pe.o]),
# DEBUG:
If(debug_buf.sink_ack,
debug_buf.sink_stb.eq(1),
debug_buf.sink_data.eq(tx_fifos.source_data[tx_fifos.pe.o]),
debug_buf.sink_k.eq(tx_fifos.source_k[tx_fifos.pe.o]),
)
).Else(
# Section 9.2.5.1 (CXP-001-2021)
# IDLE word shall be transmitted at least once every 10,000 words, but should not be inserted into trigger packet
idling.eq(1),
tx_wordcount.eq(0),
encoder.d.eq(IDLE_CHARS[0][0]),
encoder.k.eq(IDLE_CHARS[0][1]),
# DEBUG:
If(debug_buf.sink_ack,
debug_buf.sink_stb.eq(1),
debug_buf.sink_data.eq(IDLE_CHARS[0][0]),
debug_buf.sink_k.eq(IDLE_CHARS[0][1]),
)
)
).Else(
tx_charcount.eq(tx_charcount + 1),
If(~idling,
tx_fifos.source_ack[priorities].eq(1),
encoder.d.eq(tx_fifos.source_data[priorities]),
encoder.k.eq(tx_fifos.source_k[priorities]),
# DEBUG:
If(debug_buf.sink_ack,
debug_buf.sink_stb.eq(1),
debug_buf.sink_data.eq(tx_fifos.source_data[priorities]),
debug_buf.sink_k.eq(tx_fifos.source_k[priorities]),
)
).Else(
encoder.d.eq(IDLE_CHARS[tx_charcount + 1][0]),
encoder.k.eq(IDLE_CHARS[tx_charcount + 1][1]),
# DEBUG:
If(debug_buf.sink_ack,
debug_buf.sink_stb.eq(1),
debug_buf.sink_data.eq(IDLE_CHARS[tx_charcount + 1][0]),
debug_buf.sink_k.eq(IDLE_CHARS[tx_charcount + 1][1]),
)
)
),
),
)
]
class PHY_Interface(Module):
def __init__(self, layout, nsink):
self.source_stb = Signal(nsink)
self.source_ack = Array(Signal() for _ in range(nsink))
self.source_data = Array(Signal(8) for _ in range(nsink))
self.source_k = Array(Signal() for _ in range(nsink))
# # #
self.sinks = []
for i in range(nsink):
sink = stream.Endpoint(layout)
self.sinks += [sink]
self.sync += [
If(self.source_ack[i],
# reset ack after asserted
# since upconn clk run much slower, the ack will be high for longer than expected which will result in data loss
self.source_ack[i].eq(0),
sink.ack.eq(1),
).Else(
sink.ack.eq(0),
),
self.source_stb[i].eq(sink.stb),
self.source_data[i].eq(sink.data),
self.source_k[i].eq(sink.k),
]
# FIFOs transmission priority
self.submodules.pe = PriorityEncoder(nsink)
self.comb += self.pe.i.eq(self.source_stb)
class Debug_buffer(Module,AutoCSR):
def __init__(self, layout):
self.sink_stb = Signal()
self.sink_ack = Signal()
self.sink_data = Signal(8)
self.sink_k = Signal()
# # #
self.submodules.buf_out = buf_out = stream.SyncFIFO(layout, 128)
self.sync += [
If(self.sink_stb,
# reset ack after asserted
# since upconn clk run much slower, the stb will be high for longer than expected which will result in multiple data entry
self.sink_stb.eq(0),
buf_out.sink.stb.eq(1),
).Else(
buf_out.sink.stb.eq(0),
),
self.sink_ack.eq(buf_out.sink.ack),
buf_out.sink.data.eq(self.sink_data),
buf_out.sink.k.eq(self.sink_k),
]
self.inc = CSR()
self.dout_pak = CSRStatus(8)
self.kout_pak = CSRStatus()
self.dout_valid = CSRStatus()
self.sync += [
# output
buf_out.source.ack.eq(self.inc.re),
self.dout_pak.status.eq(buf_out.source.data),
self.kout_pak.status.eq(buf_out.source.k),
self.dout_valid.status.eq(buf_out.source.stb),
]
class CXP_UpConn_PHY(Module, AutoCSR):
def __init__(self, pad, sys_clk_freq, debug_sma, pmod_pads, layout, nsink=3):
self.bitrate2x_enable = Signal()
self.clk_reset = Signal()
self.tx_enable = Signal()
self.tx_busy = Signal()
# # #
self.submodules.cg = cg = UpConn_ClockGen(sys_clk_freq)
self.submodules.interface = interface = PHY_Interface(layout, nsink)
self.sinks = interface.sinks
# DEBUG:
self.submodules.debug_buf = debug_buf = Debug_buffer(layout)
self.submodules.scheduler = scheduler = Packets_Scheduler(interface, debug_buf)
self.submodules.serdes = serdes = SERDES_10bits(pad)
self.comb += [
self.tx_busy.eq(interface.source_stb != 0),
cg.reset.eq(self.clk_reset),
cg.freq2x_enable.eq(self.bitrate2x_enable),
scheduler.reset.eq(self.clk_reset),
scheduler.ce.eq(cg.clk),
scheduler.tx_enable.eq(self.tx_enable),
serdes.reset.eq(self.clk_reset),
serdes.ce.eq(cg.clk_10x),
serdes.d.eq(scheduler.encoder.output),
serdes.oe.eq(scheduler.oe),
]
# DEBUG: remove pads
prioity_0 = Signal()
word_bound = Signal()
p0 = Signal()
p3 = Signal()
self.comb += [
prioity_0.eq((~interface.pe.n) & (interface.pe.o == 0)),
word_bound.eq(scheduler.tx_charcount == 3),
# because of clk delay
p0.eq(scheduler.tx_charcount == 2),
p3.eq(scheduler.tx_charcount == 1),
]
self.specials += [
# # debug sma
# Instance("OBUF", i_I=cg.clk, o_O=debug_sma.p_tx),
# Instance("OBUF", i_I=cg.clk_10x, o_O=debug_sma.n_rx),
# # pmod 0-7 pin
# Instance("OBUF", i_I=serdes.o, o_O=pmod_pads[0]),
# Instance("OBUF", i_I=cg.clk_10x, o_O=pmod_pads[1]),
# Instance("OBUF", i_I=~tx_fifos.pe.n, o_O=pmod_pads[2]),
# Instance("OBUF", i_I=prioity_0, o_O=pmod_pads[3]),
# Instance("OBUF", i_I=word_bound, o_O=pmod_pads[4]),
# Instance("OBUF", i_I=debug_buf.buf_out.sink.stb, o_O=pmod_pads[4]),
# Instance("OBUF", i_I=debug_buf.buf_out.sink.ack, o_O=pmod_pads[5]),
# Instance("OBUF", i_I=debug_buf.buf_out.source.stb, o_O=pmod_pads[6]),
# Instance("OBUF", i_I=debug_buf.buf_out.source.ack, o_O=pmod_pads[7]),
# Instance("OBUF", i_I=scheduler.idling, o_O=pmod_pads[5]),
# # Instance("OBUF", i_I=tx_fifos.source_ack[0], o_O=pmod[6]),
# # Instance("OBUF", i_I=tx_fifos.source_ack[2], o_O=pmod[6]),
# # Instance("OBUF", i_I=tx_fifos.source_ack[1], o_O=pmod[7]),
# Instance("OBUF", i_I=p0, o_O=pmod_pads[6]),
# Instance("OBUF", i_I=p3, o_O=pmod_pads[7]),
]

View File

@ -25,6 +25,7 @@ import analyzer
import acpki
import drtio_aux_controller
import zynq_clocking
import cxp_4r_fmc, cxp
from config import write_csr_file, write_mem_file, write_rustc_cfg_file
class SMAClkinForward(Module):
@ -138,7 +139,7 @@ class ZC706(SoCCore):
platform.add_extension(si5324_fmc33)
self.comb += platform.request("si5324_33").rst_n.eq(1)
cdr_clk = Signal()
self.cdr_clk = Signal()
cdr_clk_buf = Signal()
si5324_out = platform.request("si5324_clkout")
platform.add_period_constraint(si5324_out.p, 8.0)
@ -146,11 +147,11 @@ class ZC706(SoCCore):
Instance("IBUFDS_GTE2",
i_CEB=0,
i_I=si5324_out.p, i_IB=si5324_out.n,
o_O=cdr_clk,
o_O=self.cdr_clk,
p_CLKCM_CFG="TRUE",
p_CLKRCV_TRST="TRUE",
p_CLKSWING_CFG=3),
Instance("BUFG", i_I=cdr_clk, o_O=cdr_clk_buf)
Instance("BUFG", i_I=self.cdr_clk, o_O=cdr_clk_buf)
]
self.config["HAS_SI5324"] = None
self.config["SI5324_AS_SYNTHESIZER"] = None
@ -652,6 +653,74 @@ class _NIST_QC2_RTIO:
self.add_rtio(rtio_channels)
class CXP_FMC():
"""
CoaXpress FMC with 4 CXP channel and 1 SMA trigger
"""
def __init__(self):
platform = self.platform
platform.add_extension(cxp_4r_fmc.fmc_adapter_io)
platform.add_extension(leds_fmc33)
debug_sma = [
("user_sma_clock_33", 0,
Subsignal("p_tx", Pins("AD18"), IOStandard("LVCMOS33")),
Subsignal("n_rx", Pins("AD19"), IOStandard("LVCMOS33")),
),
]
pmod1_33 = [
("pmod1_33", 0, Pins("AJ21"), IOStandard("LVCMOS33")),
("pmod1_33", 1, Pins("AK21"), IOStandard("LVCMOS33")),
("pmod1_33", 2, Pins("AB21"), IOStandard("LVCMOS33")),
("pmod1_33", 3, Pins("AB16"), IOStandard("LVCMOS33")),
("pmod1_33", 4, Pins("Y20"), IOStandard("LVCMOS33")),
("pmod1_33", 5, Pins("AA20"), IOStandard("LVCMOS33")),
("pmod1_33", 6, Pins("AC18"), IOStandard("LVCMOS33")),
("pmod1_33", 7, Pins("AC19"), IOStandard("LVCMOS33")),
]
platform.add_extension(debug_sma)
platform.add_extension(pmod1_33)
pmod_pads = [platform.request("pmod1_33", i) for i in range(8)]
clk_freq = 125e6
gtx_pads = [platform.request("CXP_HS", i) for i in range(4)]
self.submodules.cxp = cxp.CXP(
refclk=self.cdr_clk,
downconn_pads=gtx_pads,
upconn_pads=platform.request("CXP_LS", 0),
sys_clk_freq=clk_freq,
debug_sma=platform.request("user_sma_clock_33"),
pmod_pads = pmod_pads
)
self.csr_devices.append("cxp")
# max freq of cxp_gtx_rx = linerate/internal_datawidth = 12.5Gbps/40 = 312.5MHz
# zc706 use speed grade 2 which only support up to 10.3125Gbps (4ns)
# pushing to 12.5Gbps (3.2ns) will result in Pulse width violation but setup/hold times are met
for gtx in self.cxp.downconn.gtxs:
platform.add_period_constraint(gtx.cd_cxp_gtx_tx.clk, 3.2)
platform.add_period_constraint(gtx.cd_cxp_gtx_rx.clk, 3.2)
# constraint the CLK path
platform.add_false_path_constraints(self.sys_crg.cd_sys.clk, gtx.cd_cxp_gtx_tx.clk, gtx.cd_cxp_gtx_rx.clk)
rtio_channels = []
# FIXME remove this placeholder RTIO channel
# There are too few RTIO channels and cannot be compiled (adr width issue of the lane distributor)
# see https://github.com/m-labs/artiq/pull/2158 for similar issue
print("USER LED at RTIO channel 0x{:06x}".format(len(rtio_channels)))
phy = ttl_simple.Output(self.platform.request("user_led_33", 0))
self.submodules += phy
rtio_channels.append(rtio.Channel.from_phy(phy))
self.config["HAS_RTIO_LOG"] = None
rtio_channels.append(rtio.LogChannel())
self.config["RTIO_LOG_CHANNEL"] = len(rtio_channels)
self.add_rtio(rtio_channels)
class NIST_CLOCK(ZC706, _NIST_CLOCK_RTIO):
def __init__(self, acpki, drtio100mhz):
ZC706.__init__(self, acpki)
@ -684,8 +753,13 @@ class NIST_QC2_Satellite(_SatelliteBase, _NIST_QC2_RTIO):
_SatelliteBase.__init__(self, acpki, drtio100mhz)
_NIST_QC2_RTIO.__init__(self)
class CXP_Demo(ZC706, CXP_FMC):
def __init__(self, acpki, drtio100mhz):
ZC706.__init__(self, acpki)
CXP_FMC.__init__(self)
VARIANTS = {cls.__name__.lower(): cls for cls in [NIST_CLOCK, NIST_CLOCK_Master, NIST_CLOCK_Satellite,
NIST_QC2, NIST_QC2_Master, NIST_QC2_Satellite]}
NIST_QC2, NIST_QC2_Master, NIST_QC2_Satellite, CXP_Demo]}
def main():
parser = argparse.ArgumentParser(

View File

@ -0,0 +1,542 @@
use embedded_hal::prelude::_embedded_hal_blocking_delay_DelayUs;
use libboard_zynq::{println, timer::GlobalTimer};
use log::info;
// use log::info;
use crate::pl::csr;
#[derive(Clone, Copy, Debug)]
#[allow(non_camel_case_types)]
pub enum CXP_SPEED {
CXP_1,
CXP_2,
CXP_3,
CXP_5,
CXP_6,
CXP_10,
CXP_12,
}
pub fn loopback_testing(timer: &mut GlobalTimer, speed: CXP_SPEED) {
println!("==============================================================================");
cxp_gtx::change_linerate(timer, speed);
unsafe {
info!("waiting for tx&rx setup...");
timer.delay_us(50_000);
info!(
"tx_phaligndone = {} | rx_phaligndone = {}",
csr::cxp::downconn_txinit_phaligndone_read(),
csr::cxp::downconn_rxinit_phaligndone_read(),
);
// enable txdata tranmission thought MGTXTXP, required by PMA loopback
csr::cxp::downconn_txenable_write(1);
info!("waiting for rx to align...");
while csr::cxp::downconn_rx_ready_read() != 1 {}
info!("rx ready!");
loop {
// for _ in 0..20 {
// NOTE: raw bits
// let data0 = csr::cxp::downconn_rxdata_0_read();
// let data1 = csr::cxp::downconn_rxdata_1_read();
// let data2 = csr::cxp::downconn_rxdata_2_read();
// let data3 = csr::cxp::downconn_rxdata_3_read();
// let rxready = csr::cxp::downconn_rx_ready_read();
// timer.delay_us(100);
// if data0 == 0b0101111100 || data0 == 0b1010000011 {
// println!(
// "data[0] = {:#012b} comma = {} | rx ready = {}",
// data0,
// data0 == 0b0101111100 || data0 == 0b1010000011,
// rxready,
// );
// timer.delay_us(1_000_000);
// } else if data0 == 0b1001111100 || data0 == 0b0110000011 {
// println!(
// "data[0] = {:#012b} K28.1 | rx ready = {}",
// data0,
// rxready,
// );
// timer.delay_us(1_000_000);
// } else {
// println!(
// "data[0] = {:#012b} | rx ready = {}",
// data0,
// rxready,
// );
// timer.delay_us(1_000_000);
// }
timer.delay_us(1_000_000);
// NOTE: raw bits
// let data0 = csr::cxp::downconn_rxdata_0_read();
// let data1 = csr::cxp::downconn_rxdata_1_read();
// let data2 = csr::cxp::downconn_rxdata_2_read();
// let data3 = csr::cxp::downconn_rxdata_3_read();
// println!(
// "0b{:010b} {:010b} {:010b} {:010b}",
// data0, data1, data2, data3
// );
// NOTE:decode data
// let data0_k = csr::cxp::downconn_decoded_k_0_read();
// let data1_k = csr::cxp::downconn_decoded_k_1_read();
// let data2_k = csr::cxp::downconn_decoded_k_2_read();
// let data3_k = csr::cxp::downconn_decoded_k_3_read();
let data0_decoded = csr::cxp::downconn_decoded_data_0_read();
let data1_decoded = csr::cxp::downconn_decoded_data_1_read();
let data2_decoded = csr::cxp::downconn_decoded_data_2_read();
let data3_decoded = csr::cxp::downconn_decoded_data_3_read();
println!(
"{:#04x} {:#04x} {:#04x} {:#04x}",
data0_decoded, data1_decoded, data2_decoded, data3_decoded,
);
// println!(
// "decoded_data[0] = {:#04x} decoded_k[0] = {:#b} decoded_data[1] = {:#04x} decoded_k[1] = {:#b}",
// data0_decoded,
// data0_k,
// data1_decoded,
// data1_k,
// );
}
}
}
pub fn setup(timer: &mut GlobalTimer) {
unsafe {
info!("turning on pmc loopback mode...");
csr::cxp::downconn_loopback_mode_write(0b010); // Near-End PMA Loopback
// QPLL setup
csr::cxp::downconn_qpll_reset_write(1);
info!("waiting for QPLL/CPLL to lock...");
while csr::cxp::downconn_qpll_locked_read() != 1 {}
info!("QPLL locked");
// tx/rx setup
csr::cxp::downconn_tx_start_init_write(1);
csr::cxp::downconn_rx_start_init_write(1);
info!("waiting for tx & rx setup...");
timer.delay_us(50_000);
info!(
"tx_phaligndone = {} | rx_phaligndone = {}",
csr::cxp::downconn_txinit_phaligndone_read(),
csr::cxp::downconn_rxinit_phaligndone_read(),
);
}
cxp_gtx::change_linerate(timer, CXP_SPEED::CXP_1);
}
pub mod cxp_gtx {
use super::*;
struct CdrConfig {
pub cfg_reg0: u16, //0x0A8
pub cfg_reg1: u16, //0x0A9
pub cfg_reg2: u16, //0x0AA
pub cfg_reg3: u16, //0x0AB
pub cfg_reg4: u16, //0x0AC
}
pub fn change_linerate(timer: &mut GlobalTimer, speed: CXP_SPEED) {
info!("Changing datarate to {:?}", speed);
// DEBUG: DRP pll for TXUSRCLK = freq(linerate)/20
let settings = txusrclk::get_txusrclk_config(speed);
txusrclk::setup(timer, settings);
change_qpll_settings(speed);
change_cdr_cfg(speed);
unsafe {
csr::cxp::downconn_qpll_reset_write(1);
info!("waiting for QPLL/CPLL to lock...");
while csr::cxp::downconn_qpll_locked_read() != 1 {}
info!("QPLL locked");
}
unsafe {
csr::cxp::downconn_tx_restart_write(1);
csr::cxp::downconn_rx_restart_write(1);
}
}
fn change_qpll_settings(speed: CXP_SPEED) {
// Change QPLL_FBDIV
let qpll_div_reg = match speed {
CXP_SPEED::CXP_1 | CXP_SPEED::CXP_2 | CXP_SPEED::CXP_5 | CXP_SPEED::CXP_10 => 0x0120, // FB_Divider = 80
CXP_SPEED::CXP_3 | CXP_SPEED::CXP_6 | CXP_SPEED::CXP_12 => 0x0170, // FB_Divider = 100
};
println!("0x36 = {:#06x}", qpll_read(0x36));
qpll_write(0x36, qpll_div_reg);
println!("0x36 = {:#06x}", qpll_read(0x36));
// DEBUG: remove txoutdiv
let txrxout_div = match speed {
CXP_SPEED::CXP_1 => 0x33, // 8
CXP_SPEED::CXP_2 | CXP_SPEED::CXP_3 => 0x22, // 4
CXP_SPEED::CXP_5 | CXP_SPEED::CXP_6 => 0x11, // 2
CXP_SPEED::CXP_10 | CXP_SPEED::CXP_12 => 0x00, // 1
};
// OUT_DIV
println!("0x88 = {:#06x}", gtx_read(0x88));
gtx_write(0x88, txrxout_div);
println!("0x88 = {:#06x}", gtx_read(0x88));
}
fn change_cdr_cfg(speed: CXP_SPEED) {
let cdr_cfg = match speed {
// rxout_div = 8
CXP_SPEED::CXP_1 => {
CdrConfig {
cfg_reg0: 0x0020, //0x0A8
cfg_reg1: 0x1008, //0x0A9
cfg_reg2: 0x23FF, //0x0AA
cfg_reg3: 0x0000, //0x0AB
cfg_reg4: 0x0003, //0x0AC
}
}
// rxout_div = 4
CXP_SPEED::CXP_2 | CXP_SPEED::CXP_5 => {
CdrConfig {
cfg_reg0: 0x0020, //0x0A8
cfg_reg1: 0x1010, //0x0A9
cfg_reg2: 0x23FF, //0x0AA
cfg_reg3: 0x0000, //0x0AB
cfg_reg4: 0x0003, //0x0AC
}
}
// rxout_div = 2
CXP_SPEED::CXP_3 | CXP_SPEED::CXP_6 => {
CdrConfig {
cfg_reg0: 0x0020, //0x0A8
cfg_reg1: 0x1020, //0x0A9
cfg_reg2: 0x23FF, //0x0AA
cfg_reg3: 0x0000, //0x0AB
cfg_reg4: 0x0003, //0x0AC
}
}
// rxout_div = 1
CXP_SPEED::CXP_10 | CXP_SPEED::CXP_12 => {
CdrConfig {
cfg_reg0: 0x0020, //0x0A8
cfg_reg1: 0x1040, //0x0A9
cfg_reg2: 0x23FF, //0x0AA
cfg_reg3: 0x0000, //0x0AB
cfg_reg4: 0x000B, //0x0AC
}
}
};
gtx_write(0x0A8, cdr_cfg.cfg_reg0);
gtx_write(0x0A9, cdr_cfg.cfg_reg1);
gtx_write(0x0AA, cdr_cfg.cfg_reg2);
gtx_write(0x0AB, cdr_cfg.cfg_reg3);
gtx_write(0x0AC, cdr_cfg.cfg_reg4);
}
#[allow(dead_code)]
fn gtx_read(address: u16) -> u16 {
// DEBUG:
unsafe {
csr::cxp::downconn_gtx_daddr_write(address);
csr::cxp::downconn_gtx_dread_write(1);
while csr::cxp::downconn_gtx_dready_read() != 1 {}
csr::cxp::downconn_gtx_dout_read()
}
}
fn gtx_write(address: u16, value: u16) {
unsafe {
csr::cxp::downconn_gtx_daddr_write(address);
csr::cxp::downconn_gtx_din_write(value);
csr::cxp::downconn_gtx_din_stb_write(1);
while csr::cxp::downconn_gtx_dready_read() != 1 {}
}
}
#[allow(dead_code)]
fn qpll_read(address: u8) -> u16 {
// DEBUG:
unsafe {
csr::cxp::downconn_qpll_daddr_write(address);
csr::cxp::downconn_qpll_dread_write(1);
while csr::cxp::downconn_qpll_dready_read() != 1 {}
csr::cxp::downconn_qpll_dout_read()
}
}
fn qpll_write(address: u8, value: u16) {
unsafe {
csr::cxp::downconn_qpll_daddr_write(address);
csr::cxp::downconn_qpll_din_write(value);
csr::cxp::downconn_qpll_din_stb_write(1);
while csr::cxp::downconn_qpll_dready_read() != 1 {}
}
}
}
pub mod txusrclk {
use super::*;
pub struct PLLSetting {
pub clkout0_reg1: u16, //0x08
pub clkout0_reg2: u16, //0x09
pub clkfbout_reg1: u16, //0x14
pub clkfbout_reg2: u16, //0x15
pub div_reg: u16, //0x16
pub lock_reg1: u16, //0x18
pub lock_reg2: u16, //0x19
pub lock_reg3: u16, //0x1A
pub power_reg: u16, //0x28
pub filt_reg1: u16, //0x4E
pub filt_reg2: u16, //0x4F
}
fn one_clock_cycle() {
unsafe {
csr::cxp::downconn_pll_dclk_write(1);
csr::cxp::downconn_pll_dclk_write(0);
}
}
fn set_addr(address: u8) {
unsafe {
csr::cxp::downconn_pll_daddr_write(address);
}
}
fn set_data(value: u16) {
unsafe {
csr::cxp::downconn_pll_din_write(value);
}
}
fn set_enable(en: bool) {
unsafe {
let val = if en { 1 } else { 0 };
csr::cxp::downconn_pll_den_write(val);
}
}
fn set_write_enable(en: bool) {
unsafe {
let val = if en { 1 } else { 0 };
csr::cxp::downconn_pll_dwen_write(val);
}
}
fn get_data() -> u16 {
unsafe { csr::cxp::downconn_pll_dout_read() }
}
fn drp_ready() -> bool {
unsafe { csr::cxp::downconn_pll_dready_read() == 1 }
}
#[allow(dead_code)]
fn read(address: u8) -> u16 {
set_addr(address);
set_enable(true);
// Set DADDR on the mmcm and assert DEN for one clock cycle
one_clock_cycle();
set_enable(false);
while !drp_ready() {
// keep the clock signal until data is ready
one_clock_cycle();
}
get_data()
}
fn write(address: u8, value: u16) {
set_addr(address);
set_data(value);
set_write_enable(true);
set_enable(true);
// Set DADDR, DI on the mmcm and assert DWE, DEN for one clock cycle
one_clock_cycle();
set_write_enable(false);
set_enable(false);
while !drp_ready() {
// keep the clock signal until write is finished
one_clock_cycle();
}
}
fn reset(rst: bool) {
unsafe {
let val = if rst { 1 } else { 0 };
csr::cxp::downconn_txpll_reset_write(val)
}
}
pub fn setup(timer: &mut GlobalTimer, settings: PLLSetting) {
if false {
info!("0x08 = {:#06x}", read(0x08));
info!("0x09 = {:#06x}", read(0x09));
info!("0x14 = {:#06x}", read(0x14));
info!("0x15 = {:#06x}", read(0x15));
info!("0x16 = {:#06x}", read(0x16));
info!("0x18 = {:#06x}", read(0x18));
info!("0x19 = {:#06x}", read(0x19));
info!("0x1A = {:#06x}", read(0x1A));
info!("0x28 = {:#06x}", read(0x28));
info!("0x4E = {:#06x}", read(0x4E));
info!("0x4F = {:#06x}", read(0x4F));
} else {
// Based on "DRP State Machine" from XAPP888
// hold reset HIGH during pll config
reset(true);
write(0x08, settings.clkout0_reg1);
write(0x09, settings.clkout0_reg2);
write(0x14, settings.clkfbout_reg1);
write(0x15, settings.clkfbout_reg2);
write(0x16, settings.div_reg);
write(0x18, settings.lock_reg1);
write(0x19, settings.lock_reg2);
write(0x1A, settings.lock_reg3);
write(0x28, settings.power_reg);
write(0x4E, settings.filt_reg1);
write(0x4F, settings.filt_reg2);
reset(false);
// wait for the pll to lock
timer.delay_us(100);
let locked = unsafe { csr::cxp::downconn_txpll_locked_read() == 1 };
info!("txusrclk locked = {}", locked);
}
}
pub fn get_txusrclk_config(speed: CXP_SPEED) -> PLLSetting {
match speed {
CXP_SPEED::CXP_1 => {
// CLKFBOUT_MULT = 8, DIVCLK_DIVIDE = 1 , CLKOUT0_DIVIDE = 32
// TXUSRCLK=62.5MHz
PLLSetting {
clkout0_reg1: 0x1410, //0x08
clkout0_reg2: 0x0000, //0x09
clkfbout_reg1: 0x1104, //0x14
clkfbout_reg2: 0x0000, //0x15
div_reg: 0x1041, //0x16
lock_reg1: 0x03e8, //0x18
lock_reg2: 0x5801, //0x19
lock_reg3: 0xdbe9, //0x1A
power_reg: 0x0000, //0x28
filt_reg1: 0x9808, //0x4E
filt_reg2: 0x9100, //0x4F
}
}
CXP_SPEED::CXP_2 => {
// CLKFBOUT_MULT = 8, DIVCLK_DIVIDE = 1 , CLKOUT0_DIVIDE = 16
// TXUSRCLK=62.5MHz
PLLSetting {
clkout0_reg1: 0x1208, //0x08
clkout0_reg2: 0x0000, //0x09
clkfbout_reg1: 0x1104, //0x14
clkfbout_reg2: 0x0000, //0x15
div_reg: 0x1041, //0x16
lock_reg1: 0x03e8, //0x18
lock_reg2: 0x5801, //0x19
lock_reg3: 0xdbe9, //0x1A
power_reg: 0x0000, //0x28
filt_reg1: 0x9808, //0x4E
filt_reg2: 0x9100, //0x4F
}
}
CXP_SPEED::CXP_3 => {
// CLKFBOUT_MULT = 10, DIVCLK_DIVIDE = 1 , CLKOUT0_DIVIDE = 16
// TXUSRCLK=78.125MHz
PLLSetting {
clkout0_reg1: 0x1208, //0x08
clkout0_reg2: 0x0000, //0x09
clkfbout_reg1: 0x1145, //0x14
clkfbout_reg2: 0x0000, //0x15
div_reg: 0x1041, //0x16
lock_reg1: 0x03e8, //0x18
lock_reg2: 0x7001, //0x19
lock_reg3: 0xf3e9, //0x1A
power_reg: 0x0000, //0x28
filt_reg1: 0x9908, //0x4E
filt_reg2: 0x1900, //0x4F
}
}
CXP_SPEED::CXP_5 => {
// CLKFBOUT_MULT = 8, DIVCLK_DIVIDE = 1 , CLKOUT0_DIVIDE = 8
// TXUSRCLK=125MHz
PLLSetting {
clkout0_reg1: 0x1104, //0x08
clkout0_reg2: 0x0000, //0x09
clkfbout_reg1: 0x1104, //0x14
clkfbout_reg2: 0x0000, //0x15
div_reg: 0x1041, //0x16
lock_reg1: 0x03e8, //0x18
lock_reg2: 0x5801, //0x19
lock_reg3: 0xdbe9, //0x1A
power_reg: 0x0000, //0x28
filt_reg1: 0x9808, //0x4E
filt_reg2: 0x9100, //0x4F
}
}
CXP_SPEED::CXP_6 => {
// CLKFBOUT_MULT = 10, DIVCLK_DIVIDE = 1 , CLKOUT0_DIVIDE = 8
// TXUSRCLK=156.25MHz
PLLSetting {
clkout0_reg1: 0x1104, //0x08
clkout0_reg2: 0x0000, //0x09
clkfbout_reg1: 0x1145, //0x14
clkfbout_reg2: 0x0000, //0x15
div_reg: 0x1041, //0x16
lock_reg1: 0x03e8, //0x18
lock_reg2: 0x7001, //0x19
lock_reg3: 0xf3e9, //0x1A
power_reg: 0x0000, //0x28
filt_reg1: 0x9908, //0x4E
filt_reg2: 0x1900, //0x4F
}
}
CXP_SPEED::CXP_10 => {
// CLKFBOUT_MULT = 8, DIVCLK_DIVIDE = 1 , CLKOUT0_DIVIDE = 4
// TXUSRCLK=250MHz
PLLSetting {
clkout0_reg1: 0x1082, //0x08
clkout0_reg2: 0x0000, //0x09
clkfbout_reg1: 0x1104, //0x14
clkfbout_reg2: 0x0000, //0x15
div_reg: 0x1041, //0x16
lock_reg1: 0x03e8, //0x18
lock_reg2: 0x5801, //0x19
lock_reg3: 0xdbe9, //0x1A
power_reg: 0x0000, //0x28
filt_reg1: 0x9808, //0x4E
filt_reg2: 0x9100, //0x4F
}
}
CXP_SPEED::CXP_12 => {
// CLKFBOUT_MULT = 10, DIVCLK_DIVIDE = 1 , CLKOUT0_DIVIDE = 4
// TXUSRCLK=312.5MHz
PLLSetting {
clkout0_reg1: 0x1082, //0x08
clkout0_reg2: 0x0000, //0x09
clkfbout_reg1: 0x1145, //0x14
clkfbout_reg2: 0x0000, //0x15
div_reg: 0x1041, //0x16
lock_reg1: 0x03e8, //0x18
lock_reg2: 0x7001, //0x19
lock_reg3: 0xf3e9, //0x1A
power_reg: 0x0000, //0x28
filt_reg1: 0x9908, //0x4E
filt_reg2: 0x1900, //0x4F
}
}
}
}
}

View File

@ -0,0 +1,180 @@
use core_io::{Error as IoError, Write};
use crc::crc32;
use embedded_hal::prelude::_embedded_hal_blocking_delay_DelayUs;
use io::Cursor;
use libboard_zynq::{println, timer::GlobalTimer};
use crate::pl::csr;
#[derive(Debug)]
pub enum Error {
BufferError,
LinkDown,
}
impl From<IoError> for Error {
fn from(_: IoError) -> Error {
Error::BufferError
}
}
pub fn tx_test(timer: &mut GlobalTimer) {
const LEN: usize = 4 * 20;
let mut pak_arr: [u8; LEN] = [0; LEN];
unsafe {
csr::cxp::upconn_clk_reset_write(1);
// csr::cxp::upconn_bitrate2x_enable_write(1);
csr::cxp::upconn_clk_reset_write(0);
send(&Packet::ControlU32Reg(Command::Read { addr: 0x00 })).expect("Cannot send CoaXpress packet");
csr::cxp::upconn_tx_enable_write(1);
timer.delay_us(2);
// DEBUG: Trigger packet
let linktrig_mode: u8 = 0x01;
csr::cxp::upconn_trig_delay_write(0x05);
csr::cxp::upconn_linktrigger_write(linktrig_mode);
csr::cxp::upconn_trig_stb_write(1); // send trig
// DEBUG: Trigger ACK packet
// csr::cxp::upconn_ack_write(1);
timer.delay_us(20);
csr::cxp::upconn_tx_enable_write(0);
// Collect data
let mut i: usize = 0;
while csr::cxp::upconn_upconn_phy_debug_buf_dout_valid_read() == 1 {
pak_arr[i] = csr::cxp::upconn_upconn_phy_debug_buf_dout_pak_read();
// println!("received {:#04X}", pak_arr[i]);
csr::cxp::upconn_upconn_phy_debug_buf_inc_write(1);
i += 1;
if i == LEN {
break;
}
}
print_packet(&pak_arr);
}
}
pub enum Command<T> {
Read { addr: u32 },
Write { addr: u32, data: T },
ReadWithTag { addr: u32, tag: u8 },
WriteWithTag { addr: u32, data: T, tag: u8 },
}
pub enum Packet {
ControlU32Reg(Command<u32>),
ControlU64Reg(Command<u64>),
}
impl Packet {
pub fn write_to<W>(&self, writer: &mut W) -> Result<(), Error>
where W: Write {
match self {
Packet::ControlU32Reg(cmd) => match cmd {
Command::Read { addr } => {
writer.write(&[0x02; 4])?;
writer.write(&[0x00, 0x00, 0x00, 0x04])?;
writer.write(&addr.to_be_bytes())?;
}
Command::Write { addr, data } => {
writer.write(&[0x02; 4])?;
writer.write(&[0x01, 0x00, 0x00, 0x04])?;
writer.write(&addr.to_be_bytes())?;
writer.write(&data.to_be_bytes())?;
}
Command::ReadWithTag { addr, tag } => {
writer.write(&[0x05; 4])?;
writer.write(&[*tag; 4])?;
writer.write(&[0x00, 0x00, 0x00, 0x04])?;
writer.write(&addr.to_be_bytes())?;
}
Command::WriteWithTag { addr, data, tag } => {
writer.write(&[0x05; 4])?;
writer.write(&[*tag; 4])?;
writer.write(&[0x01, 0x00, 0x00, 0x04])?;
writer.write(&addr.to_be_bytes())?;
writer.write(&data.to_be_bytes())?;
}
},
Packet::ControlU64Reg(cmd) => match cmd {
Command::Read { addr } => {
writer.write(&[0x02; 4])?;
writer.write(&[0x00, 0x00, 0x00, 0x08])?;
writer.write(&addr.to_be_bytes())?;
}
Command::Write { addr, data } => {
writer.write(&[0x02; 4])?;
writer.write(&[0x01, 0x00, 0x00, 0x08])?;
writer.write(&addr.to_be_bytes())?;
writer.write(&data.to_be_bytes())?;
}
Command::ReadWithTag { addr, tag } => {
writer.write(&[0x05; 4])?;
writer.write(&[*tag; 4])?;
writer.write(&[0x00, 0x00, 0x00, 0x08])?;
writer.write(&addr.to_be_bytes())?;
}
Command::WriteWithTag { addr, data, tag } => {
writer.write(&[0x05; 4])?;
writer.write(&[*tag; 4])?;
writer.write(&[0x01, 0x00, 0x00, 0x08])?;
writer.write(&addr.to_be_bytes())?;
writer.write(&data.to_be_bytes())?;
}
},
}
Ok(())
}
}
pub fn send(packet: &Packet) -> Result<(), Error> {
// DEBUG: remove the comment out section
// if unsafe { csr::cxp::upconn_tx_enable_read() } == 0 {
// Err(Error::LinkDown)?
// }
const LEN: usize = 4 * 20;
let mut buffer: [u8; LEN] = [0; LEN];
let mut writer = Cursor::new(&mut buffer[..]);
packet.write_to(&mut writer)?;
// Section 9.2.2.2 (CXP-001-2021)
// CoaXpress use the polynomial of IEEE-802.3 (Ethernet) CRC but the checksum calculation is different
// Also, the calculation does not include the first 4 bytes of packet_type
let checksum = crc32::checksum_ieee(&writer.get_ref()[4..writer.position()]);
writer.write(&(!checksum).to_le_bytes())?;
unsafe {
let len = writer.position();
csr::cxp::upconn_command_len_write(len as u8);
for data in writer.get_ref()[..len].iter() {
while csr::cxp::upconn_command_writeable_read() == 0 {}
csr::cxp::upconn_command_data_write(*data);
}
}
Ok(())
}
fn print_packet(pak: &[u8]) {
println!("pak = [");
for i in 0..(pak.len() / 4) {
println!(
"{:#03} {:#04X} {:#04X} {:#04X} {:#04X},",
i + 1,
pak[i * 4],
pak[i * 4 + 1],
pak[i * 4 + 2],
pak[i * 4 + 3]
)
}
println!("]");
println!("============================================");
}

View File

@ -42,6 +42,11 @@ pub mod si5324;
pub mod si549;
use core::{cmp, str};
#[cfg(has_cxp)]
pub mod cxp_downconn;
#[cfg(has_cxp)]
pub mod cxp_upconn;
pub fn identifier_read(buf: &mut [u8]) -> &str {
unsafe {
pl::csr::identifier::address_write(0);