Repository navigation
Conversation
…ript geos128crt.py wraps a GEOS 128 v2 U36 ROM as a C128 .crt with the cartridge-port loader patch, plus the sd2iec 1581 hook for the English 1581 ROM. iecdecode.py and vicemon.py are how the 1581 restart loop was found: decode the drive commands from a VICE KERNAL trace, with sd2iec's fast-loader CRC. type.py types a line by DMA, in C64 or C128 mode.
…the first included
GEOS 128 handles years past 1999 on its own; what it lacks is a clock source, so every cold boot starts at 1:00 PM on 3/23/92. The loader's unpack hook now calls a routine in unused ROM filler that sets CIA1's TOD and GEOS's date from a tagged record, which the tool fills with the build time or --clock and the TR can keep current. The Am1581 sd2iec fix moves into the same ROM filler, since the stub's 34 free bytes no longer hold both.
…ard-coded addresses say
…T128 instead of the U36 ROM The cart unpacks GEOBOOT128 into RAM bank 1 and enters it as GEOS's ROM loader does, so GEOS runs the same kernal a disk boot loads: GeoRAM works with GEOS 2.0r's GEOBOOT, and the U36 ROM's own kernal and its sd2iec 1581 fix are gone. It breaks GEOS's RAM reset key, which crashed every later C128 crt until a power cycle, and masks IRQs in the cold start, which the C128 enters with the cart's top page mapped over the CPU vectors.
…to the cold start, and checks the rest of its baked-in operands An IRQ taken through the RTI at $FFF9 is still asserted when the RTI returns, and the 6502 takes it again before the next instruction, so one arriving between the KERNAL's call and the SEI hung the boot.
…nd CLOCK, and refuses a pack that puts CBM at $C007
… code covers all of BOOT4
This was referenced Oct 16, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stacked on #60 (C128 tools): the first commits here are #60's. Review the last five, starting at 62ca47c.
geos128crt.pynow builds the GEOS 128 crt from the GEOBOOT128 file on a GEOS 128 v2 boot disk instead of the U36 function ROM:The cart LZ-packs GEOBOOT128 into its 32 KB, unpacks it into RAM bank 1 at
$2B19where the disk's boot loader would put it, and enters it at$2F3Cas GEOS's ROM loader does. GEOS loads the rest from drive 8, so the cart runs the same kernal a disk boot does, and the boot disk picks the RAM expansion. That makes GeoRAM work (#61): GEOS 2.0r's GeoRAM GEOBOOT (GR81128) with128 CONFIG 2.0Rneeds the 2.0r kernal, which the U36 ROM's build is not. The U36 path, its patches and its sd2iec 1581 fix are gone. The clock record (#62) stays, at$81F0.Two things the cart has to do, both found on the TR+:
CBMand a vector to$03E4at bank 1$FFF5. The C128 KERNAL jumps through it on every C128-mode reset, before initializing anything, and it survives every reset. The TR menu runs in C64 mode, which reuses$03E4, so after one GEOS session every later C128 crt (BASIC 7 included) crashed at reset until a power cycle. The tool breaks theC, as Berkeley's ROM build does.RTIat$FFF9. Without it, a jiffy IRQ during the stub copy crashed about 1 launch in 3 (VICE's timing never hits it).Tested: TR+ in an NTSC C128, 40-column, TR GeoRAM 512 KB, SD2IEC with the GR81128 boot D81: 20 launches in a row reached GEOS with no power cycle (16 on the prototype, 4 on the final build, which differs from it only in the IRQ vector); a DMA read of the screen showed the DeskTop with a RAM 1571 as drive B and the TR-kept clock. VICE: the GeoRAM build reaches the DeskTop in 40 and 80 columns; the stock GEOBOOT128 build with a 512 KB REU reaches GEOS's screen (DeskTop not inspected). TR+ with its 512 KB REU, the stock GEOBOOT128 build and the stock 1581 boot disk (
GEOS128.D81): 6 launches in a row reached the DeskTop, and GEOS reports the REU (ramExpSize8,sysRAMFlg$50).Draft until the GEOS package is roughly complete.