- sim.v: the coco3fpga ports changed (RED_O/GREEN_O/BLUE_O, PROBE_O/PROBE_I,
Display_Debug) and so did the switch word (rs_swap, UART_INPUT, and the cart
interrupt option on SWITCH[9]); mirror CoCo3.sv. SNAC inputs idle high.
- rtl/ovo.v: Verilog translation of the new rtl/OVO.vhd overlay; the
Makefile drops the removed Mister_Video_GIMEX.v.
- PIX_CLK is now a clk_sys enable, high 2 of every 4 cycles, fed to the
mixer's ce_pix. Capture video as video_mixer.sv does: on a clk_sys rising
edge, one pixel per rising edge of the enable.
- New reference screenshots: the release's video timing moves the picture
about 2 pixels; text checks unchanged.
- Commit tests/dw_boot.sh, which the dw_boot* ignore pattern was hiding.
Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
The 20260927 release reused status[16] / SWITCH[4] for "Swap MiSTer Uart"
(rs_swap) and dropped the "Cart Interrupt Disabled" item that the cart
start-up change wires into the slot 3 CART interrupt. Bring the item back in
the Debug menu on a free status bit (OJ, status[19]) and pass it through the
unused SWITCH[9].
Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
coco_mem_fetch.sv compared read_count against WORD_END_COUNT on the same
114 MHz edge that took SDRAM_VID_ACK and updated VID_READ_OFFSET, the fetch
address; that path missed timing (-0.7 ns). read_count only changes on the
ack, so the compare is now registered in state_start with identical results.
Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
A cart loaded after boot in Multi-Pak slot 3 made CART toggle every CPU cycle
forever -- the autostart of a program pak, whose CART pin is tied to Q. A
disk-style ROM (HDB-DOS, Disk BASIC) never drives CART; BASIC starts it from
the "DK" signature at power-on. Under the toggling, FujiNet's CFGLOAD.BIN
(which re-enables the CART interrupt) hung in an endless FIRQ after 17
DriveWire reads; an OSD reset cleared it.
- A cart whose first two bytes are "DK" now gets a cold boot (the existing
AMW/EE_Cold_Bt path) instead of the autostart interrupt.
- OSD "Cart Interrupt Disabled" (status[16] -> SWITCH[4]) was not connected;
it now suppresses the slot-3 autostart interrupt.
- Program paks are unchanged: Arkanoid loaded after boot still autostarts.
Verified in the Verilator sim with the hardware sequence (cart loaded after
boot, slot 3, no reset): the FujiNet S3 cart now reads all 98 sectors of
CONFIG, matching the MiSTer after a manual reset. Regression tests unchanged.
The sim gains --cart-frame N to load the cart after boot, as the OSD and MGL
do; loading it before reset had hidden the problem.
Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
verilator/ runs the core on the host with the shared FM-7 / Apple-IIgs sim
framework: windowed (ImGui/SDL2) or --headless, typing on a frame schedule,
screenshots, text-screen dump, 6809 disassembly trace, floppy/SDC/cart/cassette,
and RS-232 Pak serial debugging (TXD decoder, RXD driver, pty bridge, ACIA
trace) with a minimal DriveWire server and a script to run the real FujiNet
daemon against the sim. See verilator/README.md.
- The VHDL the build uses (ROMs, RAMs, font) is converted to Verilog in
verilator/rtl/; the VHDL CPU is not instantiated (CoCo3_CYC_ACC_6809).
- sdram2.sv is simulated with its own controller logic and an SDR chip model.
- run_tests.sh: boot, keyboard, disk, WIDTH 80 and HSCREEN tests, judged on
the text screen and byte-for-byte against test-ref/ screenshots.
rtl/wd1793.sv: spt_addr and buff_wr are declared at module scope instead of
inside generate blocks. They are assigned outside those blocks; Quartus
accepted it, Verilator rejected spt_addr and silently split buff_wr into a
separate net. No change on the FPGA (FM-7 made the same fix).
Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>
Two bugs kept the FujiNet HDB-DOS SY6551 carts from working:
- $FF68-$FF6B were decoded only when the Multi-Pak I/O select was slot 1
(MPI_SCS == 0), which the OSD cannot select. With the cart in slot 3 the
6551 was unreachable: HDB-DOS polled status forever and sent nothing. On a
real CoCo the SCS line covers only $FF40-$FF5F and the RS-232 Pak decodes
$FF68 itself, so it now answers in any slot.
- With CTL bit 4 = 0 ("external receive clock") the receiver ran off the fixed
1.8475 MHz enable, i.e. 115,200 baud. HDBDW3L3_19200.CCC writes CTL=$0F, so
it transmitted at 19,200 and received at 115,200 -- every DriveWire sector
failed its checksum. The receiver now uses the baud generator (RX_CLK =
TX_CLK); unchanged for CTL=$1x and for baud select 0.
Verified in the Verilator sim against the real FujiNet daemon (FujiNet-PC):
the L3 cart at 19,200 loads FujiNet CONFIG with no checksum errors.
Co-Authored-By: Claude Opus 5.5 (1M context) <[email protected]>