|
|
0eecfdcf40
|
Tools/Gen_litedram: Major improvements and cleanup including work in ensuring it is reproducible
|
2023-12-04 21:36:48 +00:00 |
|
|
|
8c921380bc
|
Peripherals/BuiltinRam: Fixed high impedance warning in yosys
|
2023-12-04 17:04:22 +00:00 |
|
|
|
374b1946e9
|
Added .keep file for intentionally empty directory
|
2023-12-03 19:30:59 +00:00 |
|
|
|
63ea29e399
|
Peripherals/Memory: Added support for the litedram DDR memory controller, created a new memory map and updated all relevant code and files including the addition of rudimentary wait state support for the cpu (BIU)
|
2023-12-03 19:24:39 +00:00 |
|
|
|
f1dc9d8a59
|
Processor/Instructions: Fixed a bug where if IN executed after a microcoded instruction the cpu would go into undefined behavior
|
2023-11-26 00:18:15 +00:00 |
|
|
|
f07e0e7c1f
|
Processor/Instructions: Added support for the IN <immediate> instruction. Also changed some stuff in system.v to add more devices in the IO/Memory space
|
2023-11-25 04:12:05 +00:00 |
|
|
|
17638d5cbd
|
Build system: Slight improvements, randomised nextpnr rng seed and printed it to the terminal. In case timing fails running it a bunch of times can yield one value that passes.
|
2023-11-23 23:27:19 +00:00 |
|
|
|
aedefddb5d
|
Project: The cpu finally works perfectly now even at full speed as far as i can tell! I made the ram and register writes synchronous which fixed the weird issues I had, then I added -abc9 to yosys so that nextpnr can actually route the cpu at full speed and increased the display fifo since the cpu is so fast now!!
|
2023-11-15 18:43:56 +00:00 |
|
|
|
29bc2e6d96
|
Project: Cleaned up some code and run the project through aspell
|
2023-11-15 14:37:46 +00:00 |
|
|
|
09ccce5f30
|
Peripherals/HD44780: Rewrote and cleaned up a lot of the driver code. Unfortunately what i think is a very weird bug in yosys is still affecting the codebase
|
2023-11-15 00:26:46 +00:00 |
|
|
|
2c8e8a9d9c
|
Added simple support for \n and \r on the HD44780 driver, increased the synthesised mem to fit brainfuck_compiled.asm and made it the default.
|
2023-11-12 21:39:27 +00:00 |
|
|
|
189b037bdf
|
Added proper line wrapping for HD44780 LCDs and rewrote half the driver to make it more flexible
|
2023-11-12 17:30:52 +00:00 |
|
|
|
618c3102d8
|
Fixed some dependencies on the makefiles
|
2023-11-12 13:30:12 +00:00 |
|
|
|
7d2cb5672f
|
Reduced numbers to be sorted in gnome_sort.asm to fit in lcd, fixed hlt on real hardware, slowed down cpu, increased lcd fifo and with that I almost got gnome_sort.asm working perfectly on real hardware
|
2023-11-12 07:31:05 +00:00 |
|
|
|
e06c0eeaa0
|
Made the build system simplify the microcode so that yosys understands and synthesises it! Now gnome_sort.asm almost works!
|
2023-11-12 04:04:56 +00:00 |
|
|
|
fa62b07c14
|
Removed probably unnecessary high impedance case yosys was complaining about in registers.v
|
2023-11-12 03:13:22 +00:00 |
|
|
|
f471b305d8
|
Switched some assignments in decode.v to non-blocking which fixed a seemingly unrelated bug with incrementing the accumulator, added some more working test code in colored_led.asm and did some semantic changes as per yosys suggestions
|
2023-11-12 02:54:41 +00:00 |
|
|
|
4c130a8d63
|
Added back removed warnings to verilator since we have now fixed those issues
|
2023-11-12 00:07:33 +00:00 |
|
|
|
09b3d51015
|
Added statistics to place&route
|
2023-11-09 23:10:06 +00:00 |
|
|
|
a88c420ca5
|
Added an I2C driver, a PCF8574 driver and an HD44780 display driver. Unfortunately this shows that even fibonacci doesn't run correctly. Nonetheless, I made colored_led.asm output text to the display!
|
2023-11-09 22:10:55 +00:00 |
|
|
|
1a1634c673
|
Updated README, improved fpga-specific makefile options and updated the version number
|
2023-11-07 14:37:22 +00:00 |
|
|
|
01dcbfa7a1
|
The CPU works on real hardware for the first time! I added an adjustment for ram size, added control for a led and a test program for it. On the fpga board there is an actual led there that I used to verify functionality
|
2023-11-06 08:13:36 +00:00 |
|
|
|
30ffa1b00c
|
Fixed a "combinatorial loop" and now if the build-in memory is reduced the design can be synthesized!
|
2023-11-06 05:36:04 +00:00 |
|
|
|
5ebd53b11c
|
fixed more driver conflicts
|
2023-11-06 01:35:48 +00:00 |
|
|
|
ae16c79b0a
|
Fixed another driver conflict
|
2023-11-05 20:18:11 +00:00 |
|
|
|
9947517693
|
Fixed simulation with icarus verilog and removed another driver conflict
|
2023-11-05 19:43:49 +00:00 |
|
|
|
4a5df9c74e
|
Fixed another driver conflict
|
2023-11-05 16:23:05 +00:00 |
|
|
|
aa9b7c0a50
|
Removed more "conflicting driver" issues with yet more performance penalties...
|
2023-11-04 15:33:23 +00:00 |
|
|
|
df2975fa09
|
Fixed a lot of "conflicting driver" issues but I had to roll back an optimisation
|
2023-11-04 11:04:22 +00:00 |
|
|
|
c7ddf3fa9e
|
More small fixes
|
2023-11-04 08:31:05 +00:00 |
|
|
|
694f708a32
|
Fixed some relatively low hanging fruit
|
2023-11-04 08:08:22 +00:00 |
|
|
|
934e2f5a36
|
Fixed a bunch of things wrong with fpga_top.v and gated off some more simulation-only code
|
2023-11-02 23:46:12 +00:00 |
|
|
|
08aac5c7b6
|
Removed some code that wasn't meant for synthesis and fixed important bug in Makefile
|
2023-11-02 22:19:15 +00:00 |
|
|
|
601397b7f0
|
Properly added fpga_top.v stuff in the build system and fixed some syntax errors
|
2023-11-02 22:00:07 +00:00 |
|
|
|
43f3e16ca4
|
Removed all instances of inout since from what i understand it's mostly synthesisable
|
2023-11-02 21:48:12 +00:00 |
|
|
|
36bf8f9c7a
|
Added OrangeCrab board-specific code to connect the cpu to the outside world
|
2023-11-02 20:40:04 +00:00 |
|
|
|
5feee9de57
|
Added support to the build system for synthesising, place and routing, serialising and uploading the design to a Lattice ECP5 OrangeCrab FPGA
|
2023-11-02 00:29:14 +00:00 |
|
|
|
85512d5ace
|
Last minute fix of dependencies in Makefile before release
|
2023-11-01 19:09:59 +00:00 |
|
|
|
3ec90b1843
|
Added version stamp and last commit to json log
|
2023-11-01 06:03:53 +00:00 |
|
|
|
557d160be6
|
did some cleanup relating to the generation of the VALID_INSTRUCTION signal
|
2023-11-01 05:00:09 +00:00 |
|
|
|
3a63e916f5
|
Added cycles to waveform capture for icarus verilog
|
2023-10-31 19:32:35 +00:00 |
|
|
|
49335a2c2f
|
Fixed a small bug in log generation and did some cleanup
|
2023-10-31 19:01:34 +00:00 |
|
|
|
8a62b89a13
|
Fixed small bug with json data reporting and improved slightly the graphs
|
2023-10-30 08:01:03 +00:00 |
|
|
|
d151435ac1
|
Removed redundant checks for enabled early instruction detection in BIU
|
2023-10-23 02:30:25 +01:00 |
|
|
|
ced03c48d6
|
Added cache flush after write, potentially fixing support for self modifying code
|
2023-10-23 02:09:13 +01:00 |
|
|
|
8d3b54b812
|
Small change from when I last worked on this and an update to the versions on the README
|
2023-10-21 18:38:50 +01:00 |
|
|
|
42c319d55d
|
Lots of cleanup mainly on processor.v
|
2023-06-01 02:13:55 +01:00 |
|
|
|
a693b87e96
|
General cleanup and moved the reading of instruction parameters from a separate stage in execute to circuitry during the decode
|
2023-05-29 02:29:15 +01:00 |
|
|
|
af63ef1d68
|
Moved the decoder logic to decoder.v Now processor.v only connects the different modules
|
2023-05-27 23:35:00 +01:00 |
|
|
|
d2a98c02ff
|
Removed duplicate code and improved microcode entry
|
2023-05-27 17:59:52 +01:00 |
|
