SPRAMArbiter latched the SPRAM word into rsp_buf and raised exi_rsp_valid a
cycle later. BBARegisterFile already took the response the cycle it arrived,
so the buffer bought nothing and cost a 24 MHz cycle (~42 ns) inside a deadline
that was short by about that much. Both are now combinational in the cycle
SPRAM's output is valid. The data sampling is unchanged -- the old code read
spram_dout in that same cycle, it just registered it before handing it over --
so only the handoff moved.
bba_top T9 goes from 1/6 to 6/6 byte-exact at EXI_HALF=1, and stays 6/6 at 2
and 8. The whole suite passes at the hardware rate. Nothing else changed:
BBARegisterFile is untouched.
Found by measuring instead of guessing, after four falsified attempts. A
counter in ExiCapture recording tx_load & ~tx_fifo.r_rdy, with a shift-register
history of which byte boundary underran, showed good and bad reads BOTH
underrun at pop 3 (harmless, mid-header) while a bad read adds pop 4 -- the pop
that loads data slot 0. So the late byte is the FIRST data byte, which rules
out every "prefetch further ahead" idea, because byte 0's request cannot be
issued before RHDR_LO where the low address byte arrives. That makes the
deadline arithmetic with no contention in it: rx CDC ~125ns + arbiter 42 + SPRAM
42 + rsp_buf 42 + tx CDC ~110 = ~361ns against a 296ns byte time. Over budget
idle, which is why fair arbitration changed nothing -- ETH contention was only
what pushed a marginal path over on some frames.
Two further changes were implemented, measured and then removed as unnecessary:
issuing byte 1's request during the pad byte, and two requests in flight with a
3-deep response hold. With rsp_buf gone, T9 is 6/6 with both removed, with
either alone, and with the register file reverted completely. They were real
pipelining but not the constraint, and they cost logic on sync, the binding
timing domain. CLAUDE.md records them as do-not-re-add, along with the trap
that sank the earlier 2-outstanding attempt: incrementing `outstanding` in the
FSM and decrementing it in the response handler are two m.d.sync assignments
that Amaranth resolves by PRIORITY, not arithmetic, so the increment is
silently discarded when both fire in one cycle.
Arbiter TB updated (the response is a cycle earlier) and mutation-tested:
re-registering the response fails T1/T2/T4, swapping the byte lanes fails
T1/T2. New T4 pins exi_rsp_valid to exactly one cycle.
Not yet confirmed on hardware; netboot's reader still probes both offsets.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Reading the frame body at 0x104 (first byte a known 0xDE) instead of the
descriptor at 0x100 (first byte 0x00, indistinguishable from a stale byte)
shows what actually happens:
good: DE AD BE EF 01 02
bad: DE DE AD BE EF 01 <-- the FIRST BYTE IS DUPLICATED
So it is a tx FIFO UNDERRUN, not an inserted byte. When the FIFO runs dry for
one byte-time the bit engine reloads the stale FIFO head and sends the previous
byte again, shifting the rest of the read.
The register file must sustain one byte per 296 ns at 27 MHz. The round trip
is request -> spram_req SyncFIFO -> arbiter (ETH has absolute priority and
preempts it) -> SPRAM -> response -> push -> tx AsyncFIFO CDC (2-3 EXI cycles).
That is marginal even idle and over budget during a frame write, and only ONE
request may be in flight so the latency is never hidden.
THREE FIXES TRIED AND FALSIFIED, recorded in CLAUDE.md so they are not
repeated:
- Fair round-robin arbitration instead of strict ETH priority: still 2/6.
Bounds the wait to ~2 cycles but does not close the gap, and breaks
spram_arbiter T3. Reverted.
- Deepen tx_fifo 2 -> 8 for cushion: 0/6 with leading zeros. The engine
drains leftovers with ONE POP PER HEADER BYTE, so a deeper FIFO leaks
stale bytes into the data phase. tx_depth=2 is deliberate. Reverted.
- Allow 2 outstanding requests to pipeline: 0/6 at EVERY rate, including
HALF=8 which previously passed. With no response FIFO, responses must be
consumed the cycle they are valid; two in flight loses bytes. The
existing comment was right. Reverted.
Tree is back to the T9-reproduction baseline: 6/6 at HALF=8, 1-2/6 at HALF=1.
PROMISING DIRECTION, not attempted: SPRAM is 16 BITS WIDE and the arbiter
reads a word then returns one byte selected by address bit 0. Returning both
bytes per request would push two bytes per round trip, doubling the budget to
592 ns and making it comfortable. That is the only approach identified that
addresses the constraint rather than working around it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
First time one of this session's RX faults has been reproducible on this side
rather than inferred from a console.
T9 performs EXI DMA reads from the SPRAM ring WHILE the ethernet side writes
frames into it. No earlier test did: T6 waits for the frame to complete before
reading, so reads and writes never overlapped -- and ETH has ABSOLUTE priority
in SPRAMArbiter, so only under that overlap can an EXI read be preempted.
EXI_HALF=8 6/6 byte-exact
EXI_HALF=2 6/6 byte-exact
EXI_HALF=1 2/6 byte-exact <-- the hardware rate
2-of-6 matches the console ratio closely (desc@+0=100 +1=343, 23% good), and
the corruption is exactly the hardware symptom -- ONE EXTRA BYTE inserted at
the front, shifting everything:
expected: 00 00 00 0C DE AD BE EF
got: 00 00 00 00 0C DE AD BE
That it is INSERTED rather than wrong is the useful part: an extra push into
the tx FIFO is a COUNT mismatch in the request/response path, not corrupted
SPRAM data. It only shows at 27 MHz because the read has to complete inside
the protocol's turnaround window while ETH preempts it.
This is the bug to chase first, ahead of the residual ACK retries: netboot
only works because its reader probes both offsets. Swiss does not probe -- it
would read a frame length of 1 -- so real networking on this adapter is broken
while our own tooling reports success.
bba_top now FAILS at EXI_HALF=1, deliberately. The failure is real and the
test is the reproduction; do not relax T9 to make the suite green. HALF=8 and
HALF=2 still pass for unrelated work.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
TIMING MEASUREMENT, which is the useful part. Raising the host timeout
0.1 -> 0.3 barely moved the retry count (110 -> 95) but tripled the cost of
each one: 16.9 s -> 40.1 s. If those were LATE ACKs, three times the deadline
would have all but eliminated them. They are LOST ACKs, so a SHORT timeout is
correct, not a long one -- the GC re-ACKs a duplicate block, so retrying early
costs nothing. Host default is now 0.05.
REGRESSION, mine, from 92f7d71. Inserting the RBFI handler split the TX-done
block, leaving these two lines attached to rx_overflow -- an RX event:
m.d.sync += r_ncra[1:3].eq(0) # clear ST bits
m.d.sync += tx_queued.eq(0) # ready for the next frame
So after the first transmit tx_queued stayed set forever. The NCRA ST
fallback trigger is guarded on ~tx_queued, so software driving TX via ST
rather than TWD would transmit exactly ONCE and then silently stop, and the ST
bits were left asserted in NCRA. Our own path arms via TWD, which is why it
kept working and why this was not noticed.
T9 covers it and is mutation-tested: with the clears misplaced, NCRA reads
0x0A with ST stuck set.
This does NOT explain the 1/3 ACK loss -- the TWD path was unaffected -- so
that remains open. txto from the next run is the discriminator: near zero
means the frames reach the W5100 and are lost after that, which points at the
chip's TX handling rather than the EXI side.
--seeds 8: clk 27.5-29.4 MHz, EXI 54.7-62.3 MHz, 8/8 PASS. Best seed 5
(clk 29.41, EXI 60.22), flashed. The .dol is unchanged; only send-dol.py's
default timeout moved, and that is read from the repo.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Two coupled fixes, both exposed by batching TXDATA writes in the test app.
1. eff_addr advanced on EVERY data byte of a register-space write. But
TXDATA (0x48) is a FIFO PORT, not a register: a whole frame is written to
that one address. So an 8-byte write put byte 0 in TXDATA and bytes 1..7
into 0x49+, which the switch silently ignores. The frame was an eighth of
its declared length, TXFrameDrain blocked waiting for the rest, and TI
never set (hardware: txto=16, "no ACK after 20 tries").
SWISS WOULD HAVE HIT THIS. It writes an entire frame with bba_outs to
BBA_WRTXFIFOD -- one transaction, one address -- so its TX would have
delivered ONE BYTE per frame. The path was never exercised because the
test app wrote TXDATA one byte per EXI transaction, giving every byte a
fresh header at 0x48. Every other register must still advance, so that
bba_outs can write PAR0-5 or MAR0-7 as a block; the exception is TXDATA
only.
2. tx_bytes was 16 deep. SPI cannot back-pressure -- there is no way to tell
the GC to wait -- so BBARegisterFile silently drops bytes when w_rdy is
low, which makes the depth the only thing between a drain stall and a
corrupted frame, not a buffering preference. 16 bytes is ~4.7 us at the
EXI byte rate, and TXFrameDrain shares the W5100 bus arbiter with RX and
the socket-3 UDP test, so any longer stall truncated a frame mid-flight.
With (1) fixed, Swiss streams up to 1514 bytes in a single transaction, so
16 could never have held.
Now 2048: a whole maximum-size frame plus margin, so a frame cannot be
truncated even if the drain makes no progress for the entire transaction.
Costs 3 of 27 free BRAMs (ICESTORM_RAM 3/30 -> 6/30, LC 68% -> 70%).
Dennis spotted that the 8-byte workaround could not survive whole-frame
writes, which is what prompted fixing the depth rather than shipping it.
bba_tx_bytes() consequently streams 64-byte chunks (bba_write()'s buffer
limit), so a 60-byte ACK is ONE transaction instead of sixty.
T8 covers both and is mutation-tested: restoring the increment gives
"1/8 bytes MISMATCH -- got [17], want [17, 34, 51, ...]". It now writes 64
bytes so it exercises the deepened FIFO too.
--seeds 8: clk 27.6-29.4 MHz, EXI 51.2-64.7 MHz, 8/8 PASS. Best seed 1
(clk 29.35, EXI 60.80), flashed.
STILL NO FLOW CONTROL on the TX path: a frame larger than 2048 would drop
silently rather than error. Fine for ethernet's 1518 ceiling; revisit if the
multiworld socket work ever wants larger payloads.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
SPRAMArbiter gives ETH writes ABSOLUTE priority -- eth_wr_ready is asserted
combinationally whenever eth_wr_valid is high, so the write port is never
stalled. Two consequences worth recording:
- The T5 stall model added to rx_frame_assembler does NOT reflect hardware.
Its value was eliminating a hypothesis, not modelling reality.
- EXI READS are what get preempted, so a read response can be delayed --
including past the end of the transaction that asked for it.
The register file did not guard against that. The response handler pushed
whatever arrived, and HEADER0 reset `outstanding` and `sp_req` but not
`sp_pending`. So a late response was pushed into tx_fifo anyway, a held byte
leaked across the transaction boundary, and `outstanding - 1` from zero
underflowed a 4-bit counter to 15 -- which then broke the prefetch bound for
the rest of that transaction.
Now guarded on (outstanding != 0), and sp_pending is cleared in HEADER0.
WHAT THIS DOES NOT CLAIM. I cannot demonstrate that this causes the +1
descriptor offset seen on hardware (17 of 22 frames, varying frame to frame).
A test written in bba_register_file passes with AND without the guard --
checked both ways -- because that testbench models TX as a capture list per
transaction, with no persistent tx FIFO and no header drain pops, which is
precisely the mechanism by which a leaked byte would shift the next read. A
test that cannot fail is worse than no test, so it is deleted rather than
banked, and the gap is documented where it was.
THE MISSING TEST, recorded in bba_register_file.py: a bba_top-level test that
performs an EXI DMA read from the SPRAM ring WHILE the ethernet side writes
frames into it. Only under that overlap can a response be delayed past its
transaction. No simulation does this today, which is why the offset is still
formally unexplained -- the fix here is a real latent bug, not a proven cause.
--seeds 8: clk 28.1-30.3 MHz, EXI 54.9-62.3 MHz, 8/8 PASS. Best seed 7
(clk 30.26, EXI 61.09), flashed. Gateware only -- .dol unchanged.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Hardware: "timeout: no match (0 seen, 0 bad desc, 0 lapped)". Zero frames
reached the ring at all -- a regression from 83a62d6, which reset the REGISTER
FILE's copy of RWP on an NCRA reset but not RXFrameAssembler's own `rwp`.
They are two copies of the same pointer. Resetting one leaves them
disagreeing about where the ring is: a driver that syncs RRP to the register
file's value (page 1) points somewhere the writer is not (wherever it left
off), the free-space sum shrinks with every frame, and once it falls below
_MAX_FRAME_PAGES the new overflow protection drops EVERY frame. Silently,
because a dropped frame never reaches the ring to be counted -- hence 0 seen
rather than an error.
So fixing one half of a two-copy pointer created exactly the inconsistency the
other half had just been fixed for. That is the third time in this session
that two copies of one piece of state has caused a hardware fault: the W5100's
init, then RRP, now RWP. When adding state that the GC can also observe,
reset BOTH copies from the same event or do not keep two.
The reset is placed AFTER the FSM so it takes priority over ADVANCE_RWP in the
same cycle. No in-flight frame needs abandoning: the same NCRA write that
pulses ncra_rst also clears the SR bit, so rx_enabled is low and the FSM is
already in IDLE.
T6 covers it and is mutation-tested. Without the reset the frame lands on the
wrong page entirely and carries a previous frame's bytes:
rwp back to page 1: got 4, want 2
descriptor: got [0,0,1,48], want [0,0,0,44]
body index 0: got 9, want 192
--seeds 8: clk 28.9-30.8 MHz, EXI 61.0-65.7 MHz, 8/8 PASS. Best seed 2
(clk 30.83, EXI 65.72), flashed. Gateware only -- the .dol is unchanged.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
eth_wr_ready was held constantly HIGH by every test this file ever had, which
is a serious blind spot: on hardware the SPRAM arbiter stalls that port
constantly, because the GC reads the ring over EXI while frames are written
into it. Any address bookkeeping that depends on ready timing is invisible
until ready actually goes low.
T5 repeats T2's 300-byte frame with ready asserted 1 cycle in 3 and requires
the descriptor and body offset to be IDENTICAL to the unstalled case.
Result: contention does NOT reproduce the one-byte body offset seen on
hardware. Under stalls the descriptor is still [0,0,1,48] and all 300 body
bytes land at base+4. So that hypothesis is eliminated, which was the point
of writing it.
Two harness faults found and fixed while building it, both of which produced
false results rather than no result:
- rx_irq and rx_overflow are 1-cycle pulses, and rx_overflow fires on the
very tick that consumes the EOF byte -- inside send_frame. Polling for it
afterwards missed it entirely, so a correctly DROPPED frame looked like a
stall. Both are now latched in the tick helper.
- send_frame read rx_ready BEFORE tick() applied the next eth_wr_ready.
rx_ready is combinational on it, so the loop was a cycle out of phase with
the accept the DUT actually made: it waited on a byte already taken, and
reported "rx_ready never asserted" for a frame that had stored correctly.
Ready is now applied, then read, then clocked.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
RXFrameAssembler ignored RRP entirely, so it lapped the reader: a traffic
burst overwrote pages the GC had not read, and the "descriptor" the driver
then parsed was frame data from whatever had replaced it. That is where the
hardware's 00 00 00 01 came from -- not a dropped write.
This is a product bug, not just a netboot one. Until now ANY burst could
corrupt a frame Swiss was midway through reading, with nothing to indicate it
had happened.
The assembler now takes the GC's read pointer and computes free space as
(rrp - rwp - 1) mod 15. On a frame start with fewer than 6 free pages -- the
worst case for 1518 bytes plus a 4-byte descriptor -- it enters DROP_FRAME,
discards to EOF, and pulses rx_overflow, which sets IR[7] (RBFI). That is the
bit the GC's IMR = 0x86 already enables, so a driver that loses a frame is
told so instead of reading corruption.
Deliberate details:
- rx_ready stays HIGH while dropping. Stalling the ethernet master instead
would back-pressure into the W5100's own buffer and turn a dropped frame
into a wedged receiver.
- Six pages are reserved up front because a frame's length is not known
until EOF. Starting with less risks running into unread data partway
through, which is worse than refusing cleanly.
- have_room is REGISTERED, not combinational. Feeding a mod-15 subtract and
comparator into the FSM branch cost real depth in `sync`, the domain with
the least margin: clk fell 27.8-32.2 -> 27.0-28.0 MHz. Registering it
restored 28.4-31.4. One cycle of staleness is harmless -- rwp moves only
at end-of-frame and rrp only when the GC writes it.
T4 proves the property that matters, and the negative control REPRODUCES THE
HARDWARE FAULT IN SIMULATION: with the check removed the unread page reads
back as [245, 248, 251, 254, ...] -- frame data where a descriptor belongs.
First time one of today's RX faults has been reproducible on this side rather
than inferred from the console. T4 also checks the receiver un-latches once
RRP advances, so a full ring cannot kill RX permanently.
Fixed a test artifact in the same pass: rx_overflow fires on the very tick
that consumes the EOF byte, inside send_frame, so polling for a 1-cycle pulse
afterwards missed it and a correctly-dropped frame looked like a stall. Both
pulses are now latched in the tick helper.
netboot also clears IR (including RBFI) at init so a stale bit is not
inherited from an earlier run.
--seeds 8: clk 28.4-31.4 MHz, EXI 55.2-62.7 MHz, 8/8 PASS. Best seed 4
(clk 31.39, EXI 61.85), flashed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Both from one mistake: eth_wr_valid is scheduled with m.d.sync and so is LOW
for the whole first cycle in a state, while eth_wr_ready was tested
immediately -- and `ready` is usually already high on entry. Both
valid.eq(1) and valid.eq(0) then fired in the same cycle, the later won, the
write was never issued, and it was dropped SILENTLY.
1. WRITE_LEN_HI dropped the descriptor's HIGH length byte, leaving it at its
placeholder 0x00. Hardware read a descriptor of 00 00 00 01 -- a 1-byte
frame length, impossible for ethernet -- and rejected the ring.
2. WRITE_PLACEHOLDER counted a phantom first write, advancing wr_addr without
issuing one. Only THREE placeholder bytes were written and wr_addr ended
at desc_base+3, so the frame body started at descriptor byte 3 and EVERY
RECEIVED FRAME WAS SHIFTED ONE BYTE EARLY. The GC's ethertype test then
read the wrong offset and silently dropped every frame: the host saw "no
ACK" while the ring filled and RWP advanced normally.
Both now advance on `valid & ready`, and the body offset is set explicitly
(desc_base + 4) rather than relying on an incremented pointer, so it cannot
drift with the handshake again.
WHY NO TEST CAUGHT THIS, which is the more important half:
- The testbench drove the RX stream OPEN-LOOP -- one byte per tick, never
checking rx_ready, which is low through IDLE and the whole placeholder
write. It silently lost bytes, including the SOF byte. A real master
must hold each byte until accepted.
- It only ever asserted on rwp and rx_irq. It declared a spram_writes list
and never filled it, so NOTHING had ever checked what was written.
- Every frame it sent was short, and a short frame's high length byte is
legitimately zero -- so a dropped high byte was indistinguishable from a
correct one.
The testbench now respects back-pressure, records every write, and checks the
descriptor bytes and the body offset. T2 sends a 300-byte frame specifically
so the high length byte is non-zero. Mutation-tested: restoring either bug
fails T2 with got [0,0,0,48] want [0,0,1,48] -- the same shape as the
hardware's 00 00 00 01 -- while T1 (short frame) still passes, which is
exactly the blind spot that let this ship.
Bring-up's 342-byte DHCP frame read a correct 00 00 01 5A only because the
arbiter happened to hold `ready` low on that entry cycle; the domain merge
and the SPRAM response-FIFO removal changed that timing.
--seeds 8: clk 27.8-32.2 MHz, EXI 54.9-64.5 MHz, 8/8 PASS. Best seed 8
(clk 32.22, EXI 62.25), flashed. Gateware only -- the .dol is unchanged.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
An NCRA reset returns RXFrameAssembler's write pointer to page 1, but the
register file's COPY of RWP was never reset -- it kept whatever the last frame
left. So after any reset the GC reads a stale RWP pointing at a page the
assembler is not using, and its first RX after a reset reads the wrong page.
CLAUDE.md has stated the intent all along ("NCRA = 0x01 ... resets W5500 +
SPRAM ptrs"); the register file simply never did it.
This is NOT only a netboot problem. Any driver that resets and then trusts
RWP hits it, Swiss included. It stayed invisible because the suite's RX test
only checks that RWP CHANGES, never that it points at the right page, and
t_dma used to read a hardcoded 0x100 so it never consulted RWP either. It
surfaced through netboot, whose RRP = RWP sync landed on a stale page and
found garbage where a descriptor should be.
T1c covers it, and BOTH halves have teeth. My first version did not: RWP read
[1] with or without the fix because nothing in the testbench had moved it. It
now drives the assembler's RWP update path to page 9 and writes RRP to 7
first, so the negative control fails on both:
T1c RWP after NCRA reset: expected [1], got [9]
T1c RRP after NCRA reset: expected [1], got [7]
A test that passes whether or not the fix is present is worse than no test.
Gateware only -- the test app needs no change and the existing .dol is
compatible: after its reset it reads RWP, now gets 1, and writes RRP = 1.
--seeds 8: clk 28.5-32.2 MHz, EXI 51.3-63.4 MHz, 8/8 PASS. Best seed 2
(clk 32.19, EXI 63.23), flashed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The DMA descriptor test reported "RRP page 0 outside ring pages 1-15" on a
board whose RX path was working (t_rx passed: RWP advanced and RI fired), so
the pointer itself was wrong.
RXFrameAssembler's write pointer resets to _RX_PAGE_FIRST = page 1 -- the ring
is pages 1-15 and page 0 holds the register region -- but the register file's
r_rwp_lo and r_rrp_lo were plain Signal(8) and reset to 0. Two consequences:
the register file and the assembler disagreed about where the ring starts, and
a driver reading its first frame from RRP=0 gets the REGISTER page instead of
the frame, which is at page 1.
It hid because RRP == RWP == 0 looks self-consistently empty, so nothing
complained until something actually read the pointer. RRP is driver-owned
(the GC writes it after consuming frames) but its RESET value is ours, and
RRP == RWP == 1 is the correct initial state: empty, pointing at the ring
start.
T1b locks down both reset values AND imports the assembler's constant to
assert it matches, because two copies of a page number in two modules is
exactly how the off-by-one got in.
--seeds 8: clk 28.x-32.2 MHz, EXI 56-60 MHz, 8/8 PASS. Best seed 8, flashed.
NOT fixed here, and worth separate attention: nothing in the gateware consumes
RRP, so RXFrameAssembler can overwrite pages the GC has not read yet -- there
is no ring-overflow protection (the real BBA has RBFI for this).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
bba_devid fails on a console with 00 04 02 02 -- the ID correct but one byte
late -- and it fails in its REAL shape, with the 16 us gap and exactly four
bytes clocked, so this is not the t_raw_id probe artefact. The register suite
passes, so register-space reads are fine; only command space is affected.
Cause is latency, not phase. A command-space header is two bytes, so the
reply must be in the tx FIFO by the LAST RISING EDGE of the header transfer:
one byte time, 296 ns at 27 MHz. The round trip is rx AsyncFIFO (~125 ns) +
a sync cycle + tx AsyncFIFO (~110 ns) = 280-320 ns, marginal by construction.
HEADER0 already pushed in the same cycle it consumed byte 0, so the FSM side
was already as early as it can be.
The inter-transfer gap cannot rescue it: the load happens on that last rising
edge and the clock is PAUSED afterwards, so no edge exists during the gap for
anything to update. That is why bba_devid's 16 us gap makes no difference.
Pre-pushing the ID at CS-fall does not work either -- the engine's
unconditional per-byte pops would eat it, and the header length is not known
until byte 0 has been decoded, which is the latency in question.
So take the round trip out of the path. The device ID is a hardwired constant
and belongs to the physical interface rather than to the register file, and
byte 0 == 0x00 identifies the query exactly (bit 7 clear = command space, bit
6 clear = read, reg 0). SPIMode3Slave now recognises it and answers
04 02 02 00 from the EXI clock domain, with no CDC at all. The only thing it
needs from sync is id_phase, a single bit that changes once, crossed with an
FFSynchronizer.
The register file's ID path is REMOVED rather than kept as a fallback: left
in, it would push bytes that a 2-byte header has too few pops to drain, which
is exactly how an earlier version made the NEXT enumeration come back a slot
late.
Tests follow the responsibility:
- spi_mode3_slave T7: the query returns 04 02 02 00 with the tx FIFO
deliberately feeding 0x00, so only the fast path can be answering.
- spi_mode3_slave T8: and it is GATED -- with id_arm low, 0x5A comes
through from the FIFO. Without this the fast path would hijack address 0
forever and the GC could never read NCRA back after its own reset.
- bba_register_file T1: retargeted to assert the FSM pushes NOTHING for an
ID query, which is the property that stops stale bytes corrupting the
following transaction.
Note what these tests do NOT establish: bba_top already passed the device ID
at EXI_HALF=1 before this change, because simulation cannot reproduce a
296 ns window miss -- the sync domain's latency relative to the EXI clock in
sim does not match silicon closely enough. T7/T8 prove the fast path works;
the evidence that it is NECESSARY is the hardware result plus the paused-clock
argument above.
--seeds 8: clk 28.4-31.0 MHz, EXI 56.7-64.4 MHz, 8/8 PASS. Best seed 2
(clk 30.98, EXI 59.47), flashed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
PHASE_METER on a console settled what two rounds of delay-cutting could not.
Per-speed dump, first two slots per EXI speed index, with the register file
and both byte FIFOs out of the MISO path:
0:8080 1:8080 2:8080 3:8080 4:8080 5:C040
Perfect to 16 MHz, one bit late at 27 MHz. The 8-slot dump at 27 MHz read
C0 40 40 40 C0 40 40 40 -- a UNIFORM one-bit lag (ones at bits 1, 9, 17, 25,
33, ...), with extra highs at bits 0 and 32 where libogc splits the read into
two 4-byte transfers and MISO tri-states to its pull-up in the gap. The ID
dump agreed exactly.
Decisive detail: ~9 ns had already been removed from the fabric output path
(the EXI_J4 pad merge, then the CS mux, then the entire fabric route by moving
the output into an SB_IO register) and the hardware result stayed BYTE
IDENTICAL each time. Nine nanoseconds out of an 18.5 ns budget cannot change
nothing if propagation were the whole story. What is left is the part no RTL
change can reach: input pad + GLOBAL CLOCK NETWORK INSERTION + IOB clock-to-Q
+ output pad, plus the host's own setup. There was no fabric delay left to
cut, so cutting delay was the wrong move.
So double the budget instead. MISO is now launched on the RISING edge
carrying the bit the host will sample at the NEXT rising edge, so the whole
path gets a FULL period (37 ns at 27 MHz) rather than half. This is safe in
both directions: the insertion delay that caused the problem now covers the
host's hold time on the previous bit, and a full period covers its setup.
- tx_out is clocked on the rising edge and loads at rx_bit == 7, one
bit-time earlier than a falling-edge shifter loading at rx_bit == 0.
- tx_hold and the entire exiclk_n domain are gone: the engine now has ONE
clock edge and ONE phase reference.
- synth.py's SB_IO output register moves to NEG_TRIGGER=0.
- Byte alignment is unchanged -- T2/T5 still land A5 and 3C in the same
slots. Only the sub-bit launch timing moved.
Testbench consequence, and it matters: bba_top's master sampled MISO AFTER
raising the clock. With a zero-delay simulation and a rising-edge launch that
reads the NEXT bit, so the suite would have reported everything one bit early
while hardware was correct. Both masters now sample in the CLK-LOW phase --
the value the host latches ON the rising edge -- which is also exactly what
the hardware notes say a probe must do.
T2 slot 0 now reads 0xFF (the CS-reset parking value that keeps MISO high
while deselected) instead of 0x00; it is a header byte the host ignores.
T6 re-anchored to the rising edge and still proves miso_next predicts
spi_miso bit for bit, so the SB_IO register stays equivalent to the fabric
model rather than an untested divergence.
--seeds 8: clk 28.9-32.1 MHz, EXI 62.1-65.4 MHz, 8/8 PASS. The EXI numbers
rise sharply (41.6-49.7 -> 62.1-65.4) because the setup path into the IOB
register is now posedge->posedge against a FULL period instead of
posedge->negedge against a half -- the same doubling, visible internally.
Best seed 3 (clk 32.09, EXI 62.78), flashed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Removing ~2 ns from the fabric MISO path (EXI_J4 merge + the CS mux) did NOT
move the hardware result: the GC still returned a byte-identical 0B 80 at
27 MHz while reading correctly at 16 MHz and below. Identical rather than
intermittent means the path is well outside the window, not marginally inside
it, so trimming fabric delay further was not going to get there.
MISO is source-synchronous -- driven on the falling edge, sampled by the GC on
the next rising edge -- so the whole path must fit a HALF period, 18.5 ns at
27 MHz, minus global-buffer insertion, pad, connector and the GC's setup.
Driving from fabric spent 8.4-11.5 ns of that on routing alone.
An SB_IO output register sits in the IO tile: clock-to-pad is the IO flop's
own delay with no fabric routing. nextpnr confirms the path left the fabric
entirely -- miso_io.D_OUT_0 packs at die (9,0), the pad itself, SB_IO count is
unchanged at 24, and no negedge -> <async> path is reported any more. What
replaces it is a setup path INTO the IOB register (10.07 ns against a
half-period), which is internal and has nothing stacked on top of it.
PIN_TYPE was decoded from yosys's own SB_IO model rather than from memory,
because wrong bits there produce a silently dead or mistimed output:
[5:4]=10 tri-state, combinational OUTPUT_ENABLE
[3]=0,[2]=1 dout = dout_q_0, single-rate REGISTERED output
[1:0]=01 simple input (unused)
NEG_TRIGGER=1 moves the IO registers to the FALLING edge.
To keep simulation and silicon agreeing about MISO, the slave exposes
miso_next -- the value the shift register WILL hold after the falling edge --
and spi_mode3_slave T6 proves it predicts spi_miso bit for bit. Without that
the IOB register would be an untested sim/hardware divergence, which is the
exact class of bug that has cost this design the most time.
BEHAVIOUR CHANGE: MISO is now TRI-STATED when CS is high, held high by the
pad's PULLUP, instead of actively driven high. EXI MISO is a SHARED bus --
SP1 is on channel 0 alongside memory-card slot A -- so driving it while
deselected was wrong and could fight a card.
--seeds 8: clk 28.5-30.7 MHz, EXI 41.6-49.7 MHz, 8/8 PASS. The EXI numbers
fall (62 -> 49) because the posedge->negedge setup path into the IOB register
is now charged to that domain at a half-period constraint; 41.6 MHz worst case
is still 1.5x the 27 MHz target. Best seed 3 (clk 30.68, EXI 49.14), flashed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Remaining after the bit-phase fix was a SPEED-DEPENDENT failure -- the GC read
correctly at 1/2/4/8/16 MHz and failed only at index 5 (27 MHz), returning
0B 80 = 0x17 late by exactly ONE bit. Opposite signature to the phase slip it
was hiding under, and this one really is propagation delay.
MISO is source-synchronous: driven on the falling edge, sampled by the GC on
the next rising edge, so the whole path must fit a HALF period -- 18.5 ns at
27 MHz -- minus global-buffer insertion, pad, connector and the GC's setup.
The failing build measured ~10.5 ns of that internally. Critical path:
clk-to-q 1.39 tx_shift (3,12)
routing 3.60 tx_shift[7] (3,12) -> (5,5) to the CS mux
logic 1.21 Mux(cs_n, 1, ...) LUT
routing 4.79 miso (5,5) -> (17,0) to the J4 MISO pad
10.99 ns (2.60 logic, 8.39 routing)
Two removals, both of things that were pure cost here:
1. EXI_J4 is GONE entirely (exi_j4.py, the resource, the AND merge, the docs
section, scripts/pi-exi-wirecheck.py) -- recoverable from git history. The
AND put a LUT in the CLOCK path, denying the EXI clock pad its dedicated
global-buffer route, and MISO fanned out to the J4 pad at die (17,0) which
was the far end of the critical output path: 4.79 ns of routing to a header
with nothing plugged into it. 10.2-11.5 ns -> 9.3-9.6 ns.
2. The Mux(cs_n, 1, ...) LUT is gone from the MISO data path. It only parked
MISO high when deselected, which the flop's own reset value does for free:
tx_shift now has init=0xFF and CS is that domain's async reset, so MISO is
tx_shift[7] with nothing in between. Removes 1.21 ns of logic plus the
3.60 ns route feeding it and lets the placer sit the flop by the pad.
synth.py now PARSES AND RANKS BY the MISO clock-to-pad delay. Ranking by fmax
was how a board that failed at 27 MHz passed every seed at >=1.29x margin --
fmax was never the binding constraint for this design.
--seeds 8: clk 28.7-32.5 MHz, EXI 49.6-62.1 MHz, MISO 8.38-10.22 ns, 8/8 PASS.
Best seed 8 (clk 30.35, EXI 49.56, MISO 8.38 ns), flashed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The bit engine had TWO independent phase references: rx_bit counting rising
edges and tx_bit counting falling edges. Those only stay in step while the
host emits exactly as many falls as rises, and it does not -- a transfer that
begins on a RISING edge (CS asserted while CLK is already low, or the clock
gating between two back-to-back host transfers) delivers a rise with no
matching fall. tx_bit then slips one bit and never recovers.
RX never notices, because the rising count always equals the host's bit count.
So the header decodes perfectly, the right register is selected, and ONLY MISO
is wrong -- which is what made this look like an output-edge or output-delay
problem for two rounds of debugging.
Measured on a GameCube: every MISO byte shifted right by 2 bits, identically
at every EXI speed index. NWAYS read 05 C0 (= 0x17 with two leading zero
bits), device ID 01 00 80 80, speed sweep ok=[0,0,0,0,0,0].
Two diagnoses were wrong before this one, and both were ruled out by evidence
rather than argument:
- NOT propagation delay. Index 0 is 1 MHz, a 1000 ns bit time; no output
path fails there. Speed-independence is the tell.
- NOT the output edge. The old rising-edge build and the new falling-edge
build returned BYTE-IDENTICAL 05 C0. Flipping the edge would have moved
it; it did not, so the slip is upstream of the output stage.
What settled it was a measurement. PHASE_METER=1 builds a diagnostic,
non-functional bitstream whose engine transmits a constant 0x80 per byte
instead of the FIFO head, taking the register file and both byte FIFOs out of
the MISO path so the host reads the engine's byte PHASE directly: 0x80
aligned, 0x40 one bit late, 0x20 two bits late. Hardware read 0x20 in all
eight slots, localising the fault to the bit engine in a single run.
Fix: the shift register loads on rx_bit == 0. One phase reference, on the
edge the host counts, so the two cannot drift.
Regression test (spi_mode3_slave T5) asserts CS while CLK is LOW so the
transaction begins on a rising edge. The old counter fails it by exactly one
bit: A5 3C comes back as 52 9E. EVERY earlier testbench asserted CS with CLK
high, so falls and rises always matched and the slip was structurally
invisible -- the same blind spot as the Mode 2/3 mix-up, which was invisible
to a master that parked MOSI early. Vary the starting edge in SPI tests.
Timing after the fix, EXI_J4=1, --seeds 8: clk 26.9-31.1 MHz, EXI 45.8-61.7
MHz, 8/8 PASS. Best seed 4 (clk 31.07, EXI 56.70), flashed.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Replaces the PLL-fed 54 MHz `capture` domain, which oversampled the GC's
27 MHz EXI clock 2x and so put BIT-level timing across a CDC boundary: the
engine's view of the clock lagged the pin by 2 FFSynchronizer stages plus a
register (~55 ns) against a 37 ns bit time, so MISO came out ~2 bits late and
the GC read 0x17 as 05 C0. No amount of oversampling fixes that -- 54 MHz
gives only 2 ticks per bit and the part cannot close the 81 MHz that 3x needs.
The engine is now clocked by the pin itself (pin 44 = global buffer input net
6, no board change), making MISO clock-to-Q. ExiCapture creates four domains:
exiclk (posedge, samples MOSI), exiclk_n (negedge, drives MISO), both
async-reset by CS, plus reset-less exififo_w/exififo_r for the byte FIFO sides
-- reset-less because resetting one side of an AsyncFIFO desynchronises its
gray pointers, and posedge because AsyncFIFO has a posedge-only reset CDC
internally and cannot host a negedge side.
Also removes the SPRAM response FIFO. The DMA read was one byte late because
the round trip did not fit the protocol's turnaround byte; the arbiter already
exposed valid/ready, so the FIFO bought a cycle of latency and nothing else.
The response is handled outside the FSM (the reply lands in RHDR_PAD, not
SPRAM_STREAM) with a one-entry hold for when tx_fifo is momentarily full.
The PLL was left instantiated but unused after the rewrite; removing it frees
the hard macro and pin 35 (which stays unconnected only because the board is
already built that way). This is what lifted the pass rate:
domain target before after
clk (sync) 24 MHz 28.1-30.2 26.4-32.2 8/8
capture_clk 54 MHz 2 of 8 PASS (gone)
EXI pin domain 27 MHz -- 45.6-64.4 8/8
Best seed 3: clk 32.18 MHz, EXI 55.86 MHz. The binding constraint has FLIPPED
-- `clk` (1.10-1.34x) is now tighter than the EXI domain (1.69-2.39x) -- so
synth.py ranks seeds by worst RELATIVE margin, not by the EXI number.
synth.py could not have scored this design at all: it required the vanished
capture_clk, and its regex '(\w+)' cannot match nextpnr's exi_clk_$glb_clk
(no $ in \w). add_clock_constraint was also landing on the wrong net -- under
EXI_J4 the clock is the AND of two pads, not either pad -- so nextpnr silently
fell back to a 12 MHz default and under-optimised the whole EXI domain.
Testbenches:
- spi_mode3_slave had lost its testbench entirely in the rewrite. Restored
with a strict Mode 3 master (changes MOSI on the falling edge, samples MISO
in the clock-low phase) and MUTATION-TESTED: flipping either clock edge in
the RTL makes it fail. Both of those bugs previously reached silicon
against a lax master that parked MOSI well before the edge.
- exi_j4 T2/T3 returned [0,4,2,2]. Real bug: the engine latches the FIFO head
at each byte boundary, so a 2-byte command header leaves only ONE drain
slot. DEVID streamed zeros until CS and left two stale bytes -- one too
many -- so the NEXT enumeration came back a slot late. DEVID now pushes
exactly four bytes and stops. Register-space reads may still over-push:
their 4-byte header gives three drain slots.
- Dead ports removed from SPIMode3Slave (hdr_bytes, frame_start, cs_active,
byte_idx) and the mode2 parameter is gone -- there is one correct edge
convention and it is now enforced by tests.
All 14 module suites pass, plus bba_top at EXI_HALF=1/2/8.
CLAUDE.md: rewrote the clocking, CDC inventory and module tables; marked the
2026-09-30 capture_clk measurements historical; and corrected the line in
"Critical Protocol Notes" that still stated the INVERTED edge convention
("MOSI sampled on falling, MISO driven on rising" is Mode 2) -- the original
implementation followed that doc error.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
bba_top wired ClockSignal("sync") to ClockSignal("exi"), so the two "domains"
were always the same physical clock net. Everything between them -- six
AsyncFIFOs and three PulseSynchronizers -- was pure latency and logic for a
crossing that did not exist.
* the six AsyncFIFOs become plain SyncFIFOs
* the three PulseSynchronizers become wires
* BBARegisterFile and ExiCapture move from m.d.exi to m.d.sync
* the exi ClockDomain is gone; HFOSC drives sync directly
The SPRAM request write port is also combinational now. A registered w_en
costs a 24 MHz cycle, and the protocol's turnaround byte (~296 ns at 27 MHz)
has no cycles to spare.
Passing: bba_register_file, spram_arbiter, rx_frame_assembler, tx_frame_drain,
w5100_parallel_master, uart_shell, status_panel. bba_top at EXI_HALF=1
(27 MHz, the hardware rate) passes device ID, PAR read/write, NWAYS, RWP and
the 34-byte TX frame.
STILL OPEN: the SPRAM/DMA read is one byte late. Removing the CDC was not
enough -- the remaining round trip is the rx crossing (exiclk->sync, a REAL
CDC, ~125 ns) plus FSM decode plus the spram_req FIFO, arbiter, SPRAM access
and spram_rsp FIFO, which still exceeds the turnaround byte. Next step: now
that it is all one domain, drop the two SPRAM FIFOs entirely and give the
register file a direct request/grant port on the arbiter -- that should take
the round trip to ~2-3 cycles and fit.
exi_capture and exi_j4 testbenches still assert removed behaviour (the CS-fall
flush and the old MISO slot timing) and need rewriting for the clocked engine.
Still WIP -- do not flash.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The TX failure was a TESTBENCH artifact, not a design bug. Between
transactions the bench raised CS for one HP -- 18.5 ns at EXI_HALF=1 -- but
cs_active reaches the exi domain through a 2-stage FFSynchronizer at 24 MHz,
so the register file never saw one transaction end and swallowed the next
one's header as write data. A real host leaves microseconds between
exi_deselect() and the next exi_select(). With a realistic gap the 34-byte TX
frame now matches byte-for-byte at the hardware rate.
At EXI_HALF=1 (27 MHz) bba_top now passes device ID, PAR read/write, NWAYS,
RWP and TX. The one remaining failure is the SPRAM/DMA stream, offset by
exactly one byte.
Root cause of that offset, and it is not architectural: bba_top.py:171 does
m.d.comb += ClockSignal("sync").eq(ClockSignal("exi"))
so exi and sync are LITERALLY THE SAME CLOCK NET. Every AsyncFIFO between
them is pure overhead. The SPRAM round trip pays two needless CDC crossings
(~6 cycles, ~250 ns) on top of the arbiter and SPRAM access, giving ~375 ns
against the ~170 ns the protocol's turnaround byte actually leaves once the rx
crossing is subtracted. Register reads are unaffected because their response
is immediate; only SPRAM is late.
Fix direction: merge the two domains and replace the exi<->sync AsyncFIFOs and
PulseSynchronizers with plain SyncFIFOs/signals. That removes the latency,
drops a lot of logic, and cannot change behaviour because the clocks are
already identical.
Known failing, both testing removed behaviour rather than new bugs:
* exi_capture's own testbench asserts the CS-fall flush and the old MISO
slot timing; the flush no longer exists (header pops drain the FIFO).
* exi_j4's T3 re-enumerate is off by a byte.
Still WIP -- do not flash.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Replaces 54 MHz oversampling with a slave clocked by EXI CLK itself:
exiclk posedge sample MOSI (Mode 3 samples on edge 2)
exiclk_n negedge drive MISO (must settle between the master's samples)
CS_N async reset for both
EXI CLK (pin 44) is at die (6,0) = global buffer net 6 on the 5k-sg48, so it
can drive a real clock network with no board change.
This removes the capture domain entirely. Bit-level timing no longer
crosses a clock boundary -- MISO is clock-to-Q instead of 3 ticks of a foreign
clock -- and the only crossings left are the byte FIFOs, which have a whole
byte-time (~296 ns) of slack rather than a fraction of a bit.
AsyncFIFO needs a posedge clock for its internal reset CDC, so the FIFO sides
use reset-less POSEDGE domains on the same pin; the engine pops on the posedge
and hands the byte to the negedge output stage through a register. The FIFOs
must not see CS: resetting one side of an AsyncFIFO desynchronises its gray
pointers.
Two bugs found and fixed along the way:
* one pop per byte boundary is all that is needed -- pops during the header
drain whatever the previous transaction left (replacing the CS-fall flush)
and the turnaround-byte push lands in data slot 0. No slot-threshold or
header-length logic at all.
* rx_byte/rx_valid had to become COMBINATIONAL. Registered, they need one
more clock edge, and after a transaction's final bit the clock stops and
CS rises -- so the last byte of every transaction was lost (a 4-byte write
landed as DE AD BE 00, a 34-byte TX as 15 bytes).
At the HARDWARE rate (EXI_HALF=1, 27 MHz) the device ID, NWAYS, PAR
read/write and the W5100->SPRAM RX path now all pass -- none of which has ever
worked at 27 MHz before. Testbenches now drive the SPI clock with real
delays, since there is no capture domain to count ticks against.
STILL BROKEN: the SPRAM/DMA stream is offset by one byte (its round trip
exceeds the single turnaround byte), and TX lands 15 of 34 bytes. Do not
flash this yet.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The docs call this 'Mode 3 (CPOL=1, CPHA=1)' and then describe sampling MOSI
on the FALLING edge. Idle-high plus sample-on-falling is Mode 2. Real Mode 3
samples on the second edge of each cycle, which with an idle-high clock is the
RISING one, and that is what the GameCube does.
Sampling on the same edge the GC drives MOSI latched the PREVIOUS bit, so
every byte arrived one bit late in both directions, at every clock speed:
the 0x80 register-space header byte was read as 0x00, which decodes as a
command-space device-ID query -- so NWAYS reads returned the device ID and the
speed sweep failed at every index. Hardware raw dump, first transaction after
an FPGA reset:
00 00 02 01 01 00 00 00 02 01 01 00 = 04 02 02 00 >> 1
The Pi probe never caught it because that script was written to match this
device, and every simulation testbench drives MOSI many capture ticks before
toggling the clock, so MOSI is stable at BOTH edges and the convention cannot
be distinguished. The bba_top master now models a true Mode 3 master (drive
on falling, sample on rising) so the suite can tell the difference.
SPIMode3Slave(mode2=True) restores the old behaviour for the Pi/J4 probe.
Known remaining issue: at EXI_HALF=1 (27 MHz, 2x oversampling) reads are still
shifted, because the 2-stage input FFSynchronizer costs 2 capture ticks which
at 27 MHz is a whole bit. That is a separate fault and only bites at 2x.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
bba_top derived its EXI clock from HALF, defaulting to 3.4 MHz against the
real GC's ~27 MHz. At that rate the whole CDC response latency fits inside
one bit time, so the suite passed green on a design that failed on console --
this is how both the wrong-header-format and the TX-pop-threshold bugs
reached hardware with everything green.
HALF is now EXI_HALF-configurable (default 8 for fast regression). At the
real rate, EXI_HALF=1, every read comes back shifted right by ONE BIT:
04 02 02 00 -> 02 01 01 00, NWAYS 0x17 -> 0x0B. That is a bit-level sampling
misalignment, not the byte-level latency chased earlier: at 54 MHz capture
there are only 2 ticks per EXI bit and the 2-stage FFSynchronizer on EXI CLK
costs 2 ticks, a whole bit.
Recorded as an open bug in CLAUDE.md with the measurements.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
tx_fifo.r_en popped from txld_cnt >= 2, hard-coded for the old invented
2-byte header where the data phase began at slot 2. The real protocol's
register space uses a 4-byte header, so data starts at slot 4 -- two bytes
were being consumed before the GC clocked any data, and every register read
came back wrong. On hardware: HIPR and NWAYS both read 0x00, and the device
ID returned 01 80 c0 c0 (stale bytes, including our own header command bytes
0x80/0xC0 echoing back out of the FIFO).
The threshold is now derived from byte 0 bit 7, which is exactly what selects
the header form: register space -> 4 slots, command space -> 2 slots.
Also removes the pad-slot filler byte the register file was pushing. That was
compensating for this bug rather than fixing it -- it made the simulation pass
while hardware still failed, which is how it slipped through. With the pop
threshold correct the filler is wrong, and both the unit and integration
testbenches agree without it.
All suites pass: bba_register_file, bba_top, exi_capture, exi_j4,
spi_mode3_slave, tx_frame_drain, spram_arbiter, rx_frame_assembler,
w5100_parallel_master.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Replaces the invented 2-byte header (addr split across both bytes, 2-bit
length field) with the protocol libogc and Swiss actually speak:
command (bit7=0, 6-bit reg) [reg, 00, 00, data] one 4-byte transfer
register (bit7=1, 16-bit reg) [0x80|w<<6, ah, al, 00] + data until CS
This was not an optimisation: the old decoder read Swiss's NWAYS read
(80 00 31 00) as 'write 0x31 to NCRA', and bit 0 of 0x31 is RESET, so every
register read would have reset the device.
Changes:
* 16-bit address; no length field -- register reads and writes now stream
until CS, like the SPRAM path already did
* RHDR_LO starts the SPRAM fetch as soon as the address completes, using
the trailing header byte as the prefetch window the protocol provides
* command space serves EXI regs 0x02/0x03 (interrupt mask/status), the only
two Swiss uses
* EXI_GetID gets its own path: it offers just ONE header byte of warning, so
the first ID byte is pushed the instant byte 0 identifies the query
Two timing subtleties, both found by simulation and commented in place:
* MISO is driven LIVE from the tx FIFO head, so a byte pushed while the pad
byte is on the wire appears in the pad slot. A filler byte is parked
there (MISO is don't-care through the header) so the first real response
lands in data slot 0.
* the prefetch must be one-shot; re-firing while waiting for the pad byte
consumed the register stream early and shifted reads by a byte.
All testbenches updated to the real format and passing: bba_register_file
(7), bba_top (8 incl. the full W5100->SPRAM->GC loop and TX), exi_j4,
exi_capture, spi_mode3_slave, tx_frame_drain, spram_arbiter,
rx_frame_assembler, w5100_parallel_master. gc_bba_test rebuilt against the
new protocol.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
RX_RSR_CHK returned to IDLE when RSR==0 without clearing Sn_IR. SEND_OK
(bit 4) fires after every successful transmit and the RX path only ever
cleared RECV (bit 2), so INT_N stayed asserted; because IDLE tests
~w5100_int_n BEFORE tx_valid, the FSM span IDLE->RX_CHECK->RX_RSR_CHK->IDLE
and TX was starved permanently. Observed on hardware as the first frame
transmitting and every later one never setting TI while RX kept working.
Now clears every Sn_IR bit except RECV, so a frame arriving between the RSR
read and the write still raises INT_N and is served next pass.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
w5100_rst_n was declared and never driven anywhere -- the pin sat at its init
value of 1 forever. The only reset the chip ever got was the MR software
reset, which arrives over the very bus a wedged chip has stopped answering, so
a wedged W5100 could not be recovered at all: not by NCRA reset, not by
reflashing, not even by reconfiguring the FPGA (RSTn just floats to its R36
pull-up while unconfigured). Only a 12 V power cycle worked.
Init now pulses RSTn low (1 ms, datasheet minimum is ~2 us) and waits the full
post-reset settle before touching the bus. RSTn low time is clamped to
reset_cycles so the testbenches' small values stay reachable by wait_ctr.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The init does MR=0x80 (software reset) then waits reset_cycles before writing
NETLCKR/SHAR/SIPR/socket registers. That wait was 24_000 cycles = 1 ms at
24 MHz, but the W5100S needs the datasheet's 60.3 ms -- our own bus self-test
needed ~85 ms before the chip answered at all.
1 ms is marginal rather than simply wrong, which is why this presented as
intermittent: an NCRA reset sometimes left the chip still resetting while the
init writes landed on it and were silently dropped, leaving it misconfigured
and the bus master blocked. Symptom was a W5100 that wedged after an NCRA
reset -- TX never completing and the UART shell's own 'udp unicast' hanging --
recoverable only by a 12 V power cycle. Note RST_N is never asserted by the
gateware; only the software reset is used, so this wait is the only settle
time the chip gets.
Default is now 2_000_000 = 83 ms. Testbenches pass small values explicitly and
are unaffected.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
TXFrameDrain sat in IDLE until it had a length, and the length only reached
tx_ctrl when the NCRA ST bits were written -- after the GC had streamed the
whole frame. So the entire frame had to fit in the 16-deep tx_bytes FIFO, and
BBARegisterFile silently drops bytes when w_rdy is low (SPI cannot
back-pressure). On hardware this truncated frames at exactly 16 bytes: a
captured ARP had 16 correct bytes followed by the frame restarting over itself.
The length is now handed to the drain one cycle after the TWD_HI (0x35) write,
which the GC issues BEFORE streaming data. The drain starts early and blocks
on tx_bytes_r_rdy, consuming bytes as they arrive, so tx_bytes becomes a
rate-matching buffer instead of whole-frame storage. The ST write remains a
fallback trigger for software that skips TWD, guarded by tx_queued so one frame
is never queued twice.
Adds BBATop T8: streams a 34-byte frame (> the 16-deep FIFO) over EXI and
checks the W5100 model received it byte-for-byte. Verified to FAIL without the
fix with exactly the hardware symptom -- 16/34 bytes, first bad byte at index
16. The gap that let this ship: TXFrameDrain's own tests drive the FIFO
signals directly, and the BBATop test only covered the RX loop.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Factors the EXI_J4 pad merge out of synth.py so the synthesis path and the
test exercise the same code. Verifies the device ID (04 02 02 00) enumerates
over SP1 with J4 idle, over J4 with SP1 idle, and over SP1 again afterwards,
plus a register read via J4 and the full cs_n truth table.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Lets a bench host drive EXI without the SP1 edge connector, for bring-up.
Merges a second set of EXI pads onto the existing ones rather than repinning,
so the SP1 connector stays live and the default build is untouched. The three
inputs are AND-merged, which is correct because every EXI input is inactive
HIGH (CPOL=1 clk, active-low cs_n, mosi only matters while cs_n is low): the
unplugged side idles at 1 through its SB_IO PULLUP and passes the other side
through, so no select signal is needed. Outputs drive both pads.
All five J4 pads are on the same bottom die edge as the EXI pads (x=13..18 vs
x=5..9), so the capture domain keeps its anchor. It still adds 3 LUTs ahead of
the input synchronisers, so EXI_J4 builds must sweep seeds.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The W5100S ethernet back-end now works end-to-end on real hardware: a complete,
correctly-checksummed UDP datagram (10.0.0.99:40000 -> 10.0.0.50:6464, 32-byte
payload) was transmitted by the FPGA-driven W5100S and captured on a PC, after a
full bidirectional ARP handshake.
Records the three W5100S-specific findings in CLAUDE.md (NETLCKR unlock, the
15-bit IDM address space putting the buffers at 0x4000/0x6000, and the 60.3 ms
reset-to-ready time), and notes that UDP unicast is the reliable bring-up path
while a 255.255.255.255 send sets SEND_OK without appearing on the wire.
w5100_udptest.py is the self-contained configure+send+readback probe that
isolated all of this.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Reverts the 0x8000/0xC000 change from 162dbf7, which was WRONG and actively
broke transmission. The W5100S datasheet pseudo-code quotes 0x8000 (TX) /
0xC000 (RX) for the memory blocks, but the PARALLEL/indirect (IDM) interface
only decodes a 15-bit offset. Bench-proven with an address-aliasing probe:
writing distinct values to each candidate base and reading them back showed
0xC000 and 0x4000 are the SAME cell (0xC000 & 0x7FFF == 0x4000) while 0x6000
holds its own value.
Consequence of the bad base: every payload byte written to 0x9800+i aliased
down onto 0x0000+i, landing on SHAR (0x09-0x0E) and SIPR (0x0F-0x12) and
overwriting them with the payload. The chip then transmitted frames with a
garbage source MAC/IP, which the peer NIC drops without counting -- the
"SEND_OK but rx_packets=0" symptom, and the bogus SIPR0 readback (0x50 was
literally payload byte 15, 'P').
The NETLCKR unlock from 9c4bc7a is a genuine W5100S requirement and stays.
Also:
- w5100_udptest.py: self-contained configure+send+readback probe (and the
address-aliasing probe that found this), reports over UART.
- scripts/build-flash.sh: sweep N seeds and flash the best (capture timing is
seed-dependent, so the sweep is mandatory).
- scripts/uart-console.py: interactive UART console, plus --send/--read
one-shot modes for scripting.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Reads socket-3 / status registers over the indirect bus and streams them on the
UART, without resetting the chip, to inspect what a UDP send left behind. First
hardware run was inconclusive (socket regs read 0x00 but SIPR/PHYSR didn't match
either "state persisted" or "clean reset") -- the FPGA reflash appears to glitch
RST_N, so a self-contained configure+send+readback is the reliable next step.
Committed as a starting point for that.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Second W5100S-vs-W5100 gotcha found at bring-up: the W5100S has a Network Lock
Register (NETLCKR, 0x0071) the original W5100 lacks. It defaults to 0x00 (locked),
and while locked, ALL writes to GWR/SUBR/SHAR/SIPR are silently dropped -- so the
chip has no source IP/MAC and the UDP/MACRAW send builds nothing (chip reports
SEND_OK but transmits nothing; socket registers aren't locked, which is why an
indirect write/readback of Sn_PORT succeeded). Fix: write the unlock value 0x3A
to NETLCKR right after MR enables indirect mode, before the network-config writes.
On the bench this flipped the board from silent to actually keying frames onto
the wire (magjack activity LED now blinks).
Also:
- bba_top.py test model: RX buffer base 0x6000 -> 0xC000 to match the W5100S
buffer-base fix (the model hardcoded the old W5100 address).
- w5100_selftest.py: added an indirect-mode (IDM) write/readback probe of a
socket register + IND= field in the UART report, to isolate direct-vs-indirect
access (confirmed indirect works on the real chip).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
From the physical V2 bring-up session (2026-09-20):
- exi_bba/w5100_selftest.py (NEW): standalone FPGA<->W5100 bus diagnostic.
Clock-independent data-bus walk (direct-mode GAR0 read-back, all 8 lines) +
MR software-reset self-clear (needs the 25 MHz crystal). Reports 4 verdicts on
D6/D7 LEDs AND over UART (115200): GOOD / data-line-N / clock-dead / no-response.
Boot wait is 85 ms (> the W5100S 60.3 ms init). On this board: GOOD -- bus +
all 8 data lines + 25 MHz clock all verified (unlike V1's broken D3).
- REVIEW.md / TODO.md: eFuse U13 OVLO is tied to GND, which on the TPS25961
selects the internal FIXED ~6 V overvoltage cutoff -- so it blocks the 12 V
rail entirely (bench-confirmed: 12V_EXI=0, GC_ON=0, W5100 unpowered). FAB-
BLOCKER for V2.1: fix with an OVLO divider (~1.1M/100k -> trip ~15 V). Bench
workaround: lift U13 pin 2 (OVLO) off GND. Also closed the eFuse-placement,
GND-via/3V3-zone, and CC-ESD TODO items.
- hardware/re-bba-rb/exi_devid_spi.py (NEW): Pi hardware-SPI (~25 MHz) EXI
device-ID host, companion to the bit-bang exi_devid_rpi.py.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The W5100ParallelMaster used the original W5100's TX (0x4000) / RX (0x6000)
memory-block base addresses. This board's chip is a W5100S, whose TX/RX memory
blocks live at 0x8000 / 0xC000 (datasheet: TX 0x8000-0xBFFF, RX 0xC000-0xFFFF).
Consequence found at hardware bring-up (2026-09-20): every frame was written to
dead address space, so SEND transmitted the (empty) real TX buffer -- the chip
reported "sent" but nothing valid reached the wire (peer rx_packets stayed 0,
link up). The sim bus-model used these same constants, so it never caught it --
exactly the "register addresses are datasheet-from-memory, confirm at bring-up"
risk. All socket register offsets were verified correct against the W5100S
datasheet; only the two memory bases were wrong. Fixes both the socket-N UDP
test and the socket-0 MACRAW (GC) path.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
An interactive UART command shell drives a UDP send/receive test on the
W5100's socket 3, running ALONGSIDE the live BBA (socket-0 MACRAW) with
EXI keeping bus priority. MACRAW+UDP coexistence is W5100S-datasheet
confirmed (S4.6 + "4 independent SOCKETs"). Now the default flash build
(--console selects the old event-log console).
New / changed gateware:
- uart_shell.py (new): rebbarb> shell over FT2232H channel B. Commands:
help; udp unicast <ip> [msg]; udp broadcast [msg]. After each send it
waits (bounded, else "timeout") for a reply and prints "rx <payload>".
Line buffer + message ROM live in block RAM with a sequential parser
(LC-efficient); 1-deep RX holding reg keeps pastes intact. 8 sim tests.
- w5100_parallel_master.py: configurable UDP socket (default 3) with UDP
send AND receive (IP-stack init, runtime dest IP, WIZnet UDP RX header
+ payload). Gated by enable_udp_test so the MACRAW path is unchanged
when off. Tests U1-U4 + MACRAW T1-T5.
- exi_capture.py: CAPTURE-DOMAIN CRITICAL-PATH FIX. The TX byte-FIFO
read-enable was gated by its own gray-coded ready
(r_en = ... | (flushing & r_rdy)), forming a consume_ptr -> gray ->
r_rdy -> flush -> r_en -> consume_ptr loop that capped capture_clk.
Replaced the r_rdy-based "drain until empty" flush with a fixed-length
drain counter (FIFO is only tx_depth deep), removing the pointer
feedback from r_en. Path 24.3 -> 19.6 ns; flush behavior preserved.
- bba_top.py: wire shell <-> W5100 UDP (send + rx); shell additive.
- synth.py: shell default build; env-var UDP network config; documents a
reverted PNR-timing-priority experiment.
Timing (--seeds 8, default shell build, 67% LC): capture closes on 4/8
seeds (best seed 4 = 58.36 MHz, +8%), clk passes on all. This is BETTER
than the pre-shell 2/8 baseline because the flush fix improved the
capture domain intrinsically. Flash build/seed4/top.bin.
Bring-up caveats (unchanged): W5100 socket register addresses / UDP
header format are datasheet-derived (confirm on hardware); UDP_SRC_IP /
subnet / gateway must match the LAN for unicast ARP.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Roflin
Dennis Brentjes