6502/CLK
1
0
Fork 0
mirror of https://github.com/TomHarte/CLK.git synced 2024-12-27 01:31:42 +00:00
Code Issues Projects Releases Wiki Activity
8,386 Commits 7 Branches 121 Tags 500 MiB
02c88e6826
Commit Graph

934 Commits

Author SHA1 Message Date
Thomas Harte
9a952c889f Fixes exit from random gain noise. 2020-07-15 22:44:54 -04:00
Thomas Harte
8da7806ee9 Liberalises segment parser not necessarily to require a standard epilogue. 
It seems that real disks don't always have them; I guess the boot ROM doesn't require one.
 2020-07-15 22:27:04 -04:00
Thomas Harte
aed61f6251 Implements latest advocated MC3470 behaviour. 2020-07-15 19:34:05 -04:00
Thomas Harte
d065d6d98f Adds read-only WOZ 2 support. 2020-07-15 19:15:03 -04:00
Thomas Harte
68645742f7 This is a deliberate fallthrough. 2020-06-20 00:12:08 -04:00
Thomas Harte
59458f6444 Resolves errant spaces. 2020-06-16 23:15:38 -04:00
Thomas Harte
495024d6fe Cleans up all redundant lock/unique_guard declarations. 2020-06-15 00:24:10 -04:00
Thomas Harte
73131735fa Further qmake warning corrections. 2020-05-30 19:31:17 -04:00
Thomas Harte
48afc54af6 Cuts down unused parameter warnings to just a few that may well indicate implementation errors. 2020-05-30 01:06:43 -04:00
Thomas Harte
0bf7de9d43 Advances to actually completing a build. 
Many more warnings to iron out, however.
 2020-05-30 00:47:43 -04:00
Thomas Harte
267006782f Starts to add Qt target; resolves many build warnings. 2020-05-30 00:37:06 -04:00
Thomas Harte
f7c10ef9e9 Replaces POSIX stpncpy with ANSI strlen, memcpy and memset. 2020-05-27 21:31:46 -04:00
Thomas Harte
512a52e88d Increases const correctness, marks some additional constructors as constexpr, switches std::atomic construction style. 2020-05-20 23:34:26 -04:00
Thomas Harte
25996ce180 Further doubles down on construction syntax for type conversions. 2020-05-09 23:00:39 -04:00
Thomas Harte
31c6faf3c8 Adds a bunch of consts. 2020-05-09 21:23:52 -04:00
Thomas Harte
f720a6201b Adds explicit type cast. 2020-03-29 18:36:57 -04:00
Thomas Harte
ea26f4f7bf Eliminates test code, adds a caveat. 2020-03-25 21:22:30 -04:00
Thomas Harte
5fd2be3c8e Makes a genuine attempt at five and three decoding. 2020-03-25 20:50:26 -04:00
Thomas Harte
2320b5c1fe Takes some steps towards five-and-three decoding. 
Now I 'just' need to figure out how bits are distributed within the decoded sector. The XORing and data checksum seem the same (?)
 2020-03-25 00:15:31 -04:00
Thomas Harte
e5cbdfc67c It turns out that 5-and-3 disks have a different header prologue. 2020-03-24 21:59:55 -04:00
Thomas Harte
894d196b64 Avoids massive overallocation where sync blocks overlap the index hole. 2020-03-24 21:34:33 -04:00
Thomas Harte
614032198e Ensures no divide by zero during initial construction. 2020-02-18 22:58:37 -05:00
Thomas Harte
3cb6bbf771 Uses the union of all drive statuses to determine Drive::Controller's preferred clocking. 2020-02-12 22:28:42 -05:00
Thomas Harte
0c2f77305f Eliminates dangling printf. 2020-02-11 22:12:30 -05:00
Thomas Harte
05bcd73f82 Attempts to pull drive ownership into DiskController. 
For the sake of being more intelligent as to drive clocking, hopefully. And, eventually, to support multiple drive selection.
 2020-02-11 21:59:13 -05:00
Thomas Harte
1b27eedf6b Ensure this can definitely never divide by 0. 2020-01-28 23:25:21 -05:00
Thomas Harte
294e09f275 All these 'override's can be 'final's. 
At least for the purpose of being communicative. I doubt there's much to gain in terms of compiler output — the DiskImageHolder can avoid some virtual lookups but nothing else leaps out.
 2020-01-23 22:57:51 -05:00
Thomas Harte
98daad45c7 Removers Factors.hpp; now this is a C++17 project. 2020-01-19 23:18:59 -05:00
Thomas Harte
1b4b6b0aee Renames: NumberTheory -> Numeric. 2020-01-19 23:14:35 -05:00
Thomas Harte
357137918d Adds fuzzy but marking through the GetTrackWithSectors interface. 2020-01-19 21:41:10 -05:00
Thomas Harte
da3ee381f4 Attempts a full wiring up of fuzzy bits. 2020-01-19 21:20:21 -05:00
Thomas Harte
d27d14d2b0 Supplies fuzzy masks where specified. 2020-01-19 21:08:49 -05:00
Thomas Harte
b0326530d6 Allows fuzzy masks to be fed into the FM and MFM encoders. 2020-01-19 21:08:15 -05:00
Thomas Harte
c2bd5be51a This seems to be the proper interpretation of speeds? 2020-01-19 20:42:51 -05:00
Thomas Harte
84f5feab70 Properly flags up overloads. 2020-01-19 20:37:54 -05:00
Thomas Harte
5391a699a4 Adds the ability for a PCMSegment to maintain 'fuzzy' (i.e. random) bits. Implements an LFSR for bit generation. 
I'm not necessarily happy with the idea of just shoving in a [pseudo-]random number generator in rather than emulating the proper process underneath, but for now I throw my arms up.
 2020-01-19 20:09:11 -05:00
Thomas Harte
f3f8345e5e Corrects spelling mistake. 2020-01-19 20:05:52 -05:00
Thomas Harte
f34ddce28f Adds support for STX speed zones. 2020-01-19 12:38:33 -05:00
Thomas Harte
50348c9fe7 Adds the ability to substitute a target during encoding. 2020-01-19 12:11:56 -05:00
Thomas Harte
d4be052e76 Switch to matching fragments. 2020-01-18 14:18:59 -05:00
Thomas Harte
229b7b36ed Merge branch 'master' into ShifterSync 2020-01-18 13:38:56 -05:00
Thomas Harte
aeac6b5888 Allows the type of RDY signal to be specified. 2020-01-16 21:34:48 -05:00
Thomas Harte
99122efbbc Adds a slight cool-down period on end-of-rotation. 
Along with the corresponding inactive transition of the ready signal.
 2020-01-15 23:29:52 -05:00
Thomas Harte
30e856b9e4 Renames motor_is_on_ to motor_input_is_on_ to start to disambiguate the two things. 2020-01-15 23:16:25 -05:00
Thomas Harte
6b08239199 Adapts slightly; it would seem that BD-DOS disks really fill up space. 2020-01-14 23:16:06 -05:00
Thomas Harte
ecb5807ec0 Enssures STX interprets sector sizes correctly. 2020-01-14 21:35:37 -05:00
Thomas Harte
942986aadc Insures against badly-placed locations. 2020-01-13 22:49:12 -05:00
Thomas Harte
1f539822ee Adds better support for WD-esque false sync, improves STX track patching. 2020-01-13 22:19:48 -05:00
Thomas Harte
fab35b360a Ensure an encoder is created even if no sectors are placed. 2020-01-12 22:37:00 -05:00
Thomas Harte
b3b2e18c4b Ensures head and track counts are reported accurately. 2020-01-12 22:23:34 -05:00
Thomas Harte
2d233b6358 Makes a more concrete attempt at track/sector combination. 2020-01-12 22:18:31 -05:00
Thomas Harte
83ed36eb08 Add missing #include. 2020-01-12 17:56:04 -05:00
Thomas Harte
89f4032ffc Merge branch 'master' into STX2 2020-01-12 17:55:19 -05:00
Thomas Harte
514141f8c5 Eliminates the optionality of a DPLL receiver. 2020-01-12 17:45:02 -05:00
Thomas Harte
8e3a618619 Corrects Mac build, shrinks default history [back] to 3 slots. 2020-01-12 17:33:34 -05:00
Thomas Harte
6df6af09de Remove dead .cpp. 2020-01-12 17:25:59 -05:00
Thomas Harte
f42655a0fc Promote DigitalPhaseLockedLoop to a template, simplify to O(1) add_pulse. 2020-01-12 17:25:21 -05:00
Thomas Harte
f81a7f0faf Ensures prefixes are MFM encoded and decoded. 2020-01-11 22:10:41 -05:00
Thomas Harte
2b4c924399 Makes an effort to locate address and data bodies within track. 
"Not completely successful" would be the polite term.
 2020-01-09 23:28:07 -05:00
Thomas Harte
64517a02b7 Adds code to deal with sector-free tracks. 2020-01-09 21:50:32 -05:00
Thomas Harte
b4befd57a9 Advances to being able to cope with STXs with no special features whatsoever. 
Well, other than perhaps a broken data CRC. Fuzzy bits, timing differences and the stuff between sectors are all currently absent.
 2020-01-09 21:03:01 -05:00
Thomas Harte
985b36da73 Starts towards STX support. 2020-01-07 23:21:32 -05:00
Thomas Harte
c0b5bfe726 Ensure no possible return without value. 2020-01-02 23:43:53 -05:00
Thomas Harte
54fe01b532 Fixed: the final track is inclusive, not exclusive. 2019-12-24 23:08:16 -05:00
Thomas Harte
f633cf4c3f Adds a basic implementation of the non-instantaneous index pulse. 2019-12-24 21:05:17 -05:00
Thomas Harte
dfa6b11737 Adds responsibility for an ongoing index pulse to the drive. 2019-12-24 20:53:37 -05:00
Thomas Harte
adc2b77833 Enhances with constexpr. 2019-12-24 18:53:50 -05:00
Thomas Harte
0dae608da5 Embraces std::make_[unique/shared] in place of .reset(new . 2019-12-23 21:31:46 -05:00
Thomas Harte
34a3790e11 Minor static_cast clean-ups. 2019-12-22 17:56:59 -05:00
Thomas Harte
5456a4a39d Eliminates static where constexpra aren't class members; adds some if constexprs for clarity. 2019-12-22 13:42:24 -05:00
Thomas Harte
274867579b Deploys constexpr as a stricter const. 2019-12-22 00:22:17 -05:00
Thomas Harte
a847654ef2 Corrects various old-fashioned bits of indentation, plus the odd const. 2019-12-22 00:00:23 -05:00
Thomas Harte
05d77d3297 Also deploys make_unique/shared to avoid type repetition. 2019-12-21 23:52:04 -05:00
Thomas Harte
e9318efeb6 Switches to std::make_shared/make_unique in a bunch of applicable places. 
No doubt many more similar improvements are available, these are just the ones that were easy to find.
 2019-12-21 23:34:25 -05:00
Thomas Harte
416d68ab3a Installs some additional safety guards. 2019-12-19 22:27:50 -05:00
Thomas Harte
797abae4b3 Compacts gaps when necessary to fit more sectors. 2019-12-19 19:36:19 -05:00
Thomas Harte
332f0d6167 Ensures MSAs are explicitly read-only. 2019-12-08 11:52:15 -05:00
Thomas Harte
9d07765823 Ensures proper precedence of * over %. 2019-11-14 23:19:31 -05:00
Thomas Harte
2c4f372872 Adds support for the .ST file format. 2019-11-12 23:23:14 -05:00
Thomas Harte
74be876d72 Corrects track count calculation for DSD disks. 2019-11-12 23:22:56 -05:00
Thomas Harte
e8e166eec5 Ensures no out-of-disk-bounds mirroring. 2019-11-12 23:22:25 -05:00
Thomas Harte
0204003680 Resolves GCC warnings. 2019-11-09 16:12:37 -05:00
Thomas Harte
0971bbeebe Merge branch 'master' into AtariST 2019-11-05 23:24:26 -05:00
Thomas Harte
f3e2e88986 Update explanation of NIB processing. 2019-11-05 23:20:20 -05:00
Thomas Harte
46ec3510be Fixed: ST sectors are numbered from 1, not 0. 2019-11-03 21:18:29 -05:00
Thomas Harte
1c154131f9 Expands size of storage in Cycles/HalfCycles; adjusts widely to compensate. 2019-10-29 22:36:29 -04:00
Thomas Harte
909685d87d A drive with no disk is now happy to spin its motor. 
It'll continue to produce index-hole events, which might not be accurate for 5.25" drives. Research to do.
 2019-10-26 22:57:05 -04:00
Thomas Harte
c66c484c54 Removes unused includes. 2019-10-20 20:38:54 -04:00
Thomas Harte
345b32d6e3 Implements read-only MSA support. 2019-10-20 20:38:54 -04:00
Thomas Harte
8b397626bf Adds a route through the static analyser to the Atari ST. 2019-10-20 20:38:54 -04:00
Thomas Harte
9d17d48bca Ensures cycles_per_revolution_ is populated for fixed-rate drives. 2019-09-28 23:23:15 -04:00
Thomas Harte
776b819a5a Adds the SCSI bus as an Activity::Source. 2019-09-19 19:31:22 -04:00
Thomas Harte
3d54d55dbb Adds missing #include for assert. 2019-09-18 22:06:13 -04:00
Thomas Harte
72c0a631f7 Moves includes to correct file. 2019-09-18 22:04:54 -04:00
Thomas Harte
1608a90d5d Takes another stab at finding ssize_t. 2019-09-18 22:03:51 -04:00
Thomas Harte
4f8a45a6ce Adds #include for ssize_t. 2019-09-18 22:02:59 -04:00
Thomas Harte
4f0f1dcf18 Corrects accidental use of #import. 2019-09-18 21:53:22 -04:00
Thomas Harte
8d4a96683a Reduces output noise. 2019-09-18 21:41:29 -04:00
Thomas Harte
128a1da626 Enables write support. 2019-09-18 20:18:02 -04:00
Thomas Harte
962275c22a Removes clock for NCR 5380. 
It doesn't have one in real life, and now can live off the time counting that occurs on the SCSI bus.
 2019-09-18 20:17:47 -04:00
Thomas Harte
3002ac8a4a Adds mapping of READ(8) size 0 to size 256. 2019-09-17 21:59:32 -04:00
Thomas Harte
ff43674638 Corrects partition map: string fields are 32 bytes long. 2019-09-17 21:46:14 -04:00
Thomas Harte
2f6c366668 Makes a concerted effort at properly wrapping a hard disk image. 2019-09-17 21:30:04 -04:00
Thomas Harte
2ce1f0a3b1 Implements multi-sector read/write. 
This once again unblocks Apple HD SC Setup. Progress!
 2019-09-16 22:20:42 -04:00
Thomas Harte
210129c3a1 Updated as per Inside Macintosh IV. 
Of which I now own a copy.
 2019-09-16 21:31:43 -04:00
Thomas Harte
934901447a Adds a temporary version of block access writing. 
Whatever I'm doing, it's still not correct. The Macintosh ostensibly appears to 0-fill the direct-access device, then reads a sector back and hangs.
 2019-09-15 22:06:45 -04:00
Thomas Harte
c849188016 Adds format and write to the SCSI target. 
Now I think I need to switch back to the 5380 to ensure proper DMA mode interactions when writing.
 2019-09-12 21:58:09 -04:00
Thomas Harte
87e8dade2f Implements READ BUFFER to do, you know, *something*. Plus READ CAPACITY. 
The HD SC utility now offers up drive 6 for formatting. That's progress.
 2019-09-11 21:52:02 -04:00
Thomas Harte
6fc5b4e825 Simplifies INQUIRY for future targets; implements enough of SENSE MODE to advance. 
The HD SC setup utility is now looking to read buffer.
 2019-09-08 21:59:56 -04:00
Thomas Harte
00ce7f8ae0 Takes a first shot at INQUIRY. 2019-09-07 22:04:44 -04:00
Thomas Harte
6e0e9afe2f Fixed: to post a message, I want message in, not message out. 2019-09-07 13:35:38 -04:00
Thomas Harte
cb0d994827 Provides empty data for the unimplemented sectors. 2019-09-07 13:17:53 -04:00
Thomas Harte
bee782234a Ensures no state transitions while acknowledge is still asserted. 2019-09-07 13:17:34 -04:00
Thomas Harte
cbd1a8cf78 Factors out command termination, adds a default implementation of test unit ready. 2019-09-03 22:40:18 -04:00
Thomas Harte
a4ab0afce3 Takes a shot at completing a full SCSI interaction. 2019-09-03 21:15:30 -04:00
Thomas Harte
1c7e0f3c9d Fixes control line modification by the 5380 and SCSI target command chaining. 
So now I'm back to trying to guess how a SCSI command terminates re: the relative meanings of a message phase and a status phase.
 2019-09-02 23:14:37 -04:00
Thomas Harte
318cdb41ea Adds SCSI bus clocking to the Macintosh, and fixes its internal counting. 2019-09-02 16:03:33 -04:00
Thomas Harte
2f8e31bc8b Takes a first bash at implementing the new SCSI::Bus timing infrastructure. 2019-09-02 13:00:01 -04:00
Thomas Harte
310c722cc0 Starts a transition to bus-level knowledge of SCSI-specific bus timing thresholds. 
The idea being that bus attachees need not all count time for themselves. They can be very plain finite state machines.

New semantics are not yet implemented within the Bus. The plan is to do that, remove the internal counting of time within the NCR, then adjust the Target to be more explicitly stateful.
 2019-08-31 21:44:22 -04:00
Thomas Harte
25956bd90f Mildly improves temporary logging. 
A deckchair shuffle, at best.
 2019-08-26 22:35:11 -04:00
Thomas Harte
1a60ced61b Starts trying to deal with creating a whole volume from merely a partition. 2019-08-25 23:03:54 -04:00
Thomas Harte
081316c071 Adds some additional commentary as this takes shape. 2019-08-25 17:46:05 -04:00
Thomas Harte
eafbc12cc1 Ensures a clean entry into the command phase. 2019-08-25 17:43:14 -04:00
Thomas Harte
ca08716c52 Introduces real hard disk images to the nascent world of SCSI. 2019-08-25 17:03:41 -04:00
Thomas Harte
30cef1ee22 Adds write support. 2019-08-25 15:16:35 -04:00
Thomas Harte
404b088199 Adds a trivial mass-storage device, for Macintosh HFV volumes. 2019-08-25 15:09:27 -04:00
Thomas Harte
7d61df238a Localises #include. 2019-08-25 15:09:04 -04:00
Thomas Harte
c86db12f1c Starts implementing DMA support on the 5380. 
The Macintosh doesn't actually use the DMA signals, but uses pseudo-DMA mode so they nevertheless need to be appropriate.
 2019-08-24 22:47:11 -04:00
Thomas Harte
ce2e85af8b Adds missing bus state callouts. 2019-08-22 23:27:00 -04:00
Thomas Harte
2d82855f26 Attempts to provide a data out phase. 2019-08-22 23:16:58 -04:00
Thomas Harte
faec516a2c Starts pushing towards figuring out a proper infrastructure for mass storage. 2019-08-21 23:22:58 -04:00
Thomas Harte
f668e4a54c Makes an attempt at getting the 5380 past arbitration. 
Not entirely successful. Also gets a bit smarter with `final` on ClockingHint::Sources.
 2019-08-15 23:14:40 -04:00
Thomas Harte
b1f04ed96d Eliminates the crutch of PlusToo BIN files. 
Thereby returning the .bin extension to the various consoles.
 2019-07-28 16:07:16 -04:00
Thomas Harte
786b26d23e Adds an admission of incompleteness. 
I really need to figure out the 5-and-3 encoding.
 2019-07-26 23:23:40 -04:00
Thomas Harte
d08d8ed22c Adds a further drift safeguard. 2019-07-26 23:23:01 -04:00
Thomas Harte
39d7e3c62c Ensures relative_exponents less than or equal to -32 don't overflow. 2019-07-26 22:19:40 -04:00
Thomas Harte
572e8b52e1 At the cost of extra storage, attempts to simplify away potential rounding issues around the index hole. 2019-07-26 17:20:32 -04:00
Thomas Harte
9b634472c6 Gets a little more rigorous on "high resolution". 2019-07-26 15:26:51 -04:00
Thomas Harte
d2bfd59953 Ensures positional continuity across rotation speed changes. 2019-07-25 22:29:54 -04:00
Thomas Harte
3d20ae47ea Ensures cycles_since_index_hole_ is wrapped to track length. 2019-07-25 21:48:01 -04:00
Thomas Harte
77333ff9f7 It appears that file checksums are not reliable. 2019-07-17 14:56:50 -04:00
Thomas Harte
79a624e696 Applies more rigorous logic to deciding when to stop parsing. 2019-07-16 18:06:54 -04:00
Thomas Harte
c123ca1054 Slightly improves syntax. 2019-07-16 18:05:58 -04:00
Thomas Harte
9f0f35033d Introduces sector interleaving. 2019-07-16 18:05:40 -04:00
Thomas Harte
3633285aaa Ensures a trailing zero bit isn't dropped. 2019-07-16 16:36:00 -04:00
Thomas Harte
cb16790330 Improves qualifiers. 2019-07-15 22:40:45 -04:00
Thomas Harte
ca37fd8f4c Corrects tag preservation. 2019-07-15 17:15:06 -04:00
Thomas Harte
46b98dab70 Bumps up the amount of reserved storage. 
To avoid a reallocation when reading Mac data.
 2019-07-15 17:12:31 -04:00
Thomas Harte
0568996264 Fixes a couple of data arrangement issues on output. 2019-07-15 17:11:58 -04:00
Thomas Harte
5b05a9bc61 Extends Drive to report is_writing and so that writing works as the first action. 2019-07-12 18:53:41 -04:00