Thomas Harte
|
1349e85d83
|
[Mostly] fixed track write-back.
|
2016-12-25 19:19:22 -05:00 |
|
Thomas Harte
|
007c13ec16
|
Fixed: cycles_per_bit_ isn't a function of the rotational multiplier, it's absolute. Also made sure that exactly hitting the end of a bit counts.
|
2016-12-25 16:35:39 -05:00 |
|
Thomas Harte
|
98be6ede45
|
Shuffled a little to reduce risk of overflow, ensured writing is a loop, still seem to be writing too quickly for some reason.
|
2016-12-25 16:13:05 -05:00 |
|
Thomas Harte
|
d2ad2c756e
|
Added enough shovelling to write rubbish for an entire sector.
|
2016-12-25 15:46:49 -05:00 |
|
Thomas Harte
|
ec55a25620
|
It makes sense to simplify these ahead of time.
|
2016-12-25 12:32:25 -05:00 |
|
Thomas Harte
|
aceb7e3b6b
|
Started implementing write sector on the 1770, immediately deciding it would be useful to have a callback for end-of-queued-data-written from disk controller. So had a go at implementing that, naively. More investigation required.
|
2016-12-25 12:31:38 -05:00 |
|
Thomas Harte
|
901f19f89c
|
Added enough stuff that SSDs attached to a 1770 will now reach the entry point for writing.
|
2016-12-25 09:46:12 -05:00 |
|
Thomas Harte
|
9d555c4a02
|
Let's try just declining to pump the PLL while in write mode. Added documentation to explain.
|
2016-12-25 09:19:18 -05:00 |
|
Thomas Harte
|
b57038edc5
|
Actually, at least index holes will still be receivable while writing, so this wasn't entirely correct. Probably best to leave it in.
|
2016-12-25 09:16:09 -05:00 |
|
Thomas Harte
|
af69b21033
|
This is almost complete, except that it doesn't act appropriately if some bits are written but not enough to cover the entire writing period.
|
2016-12-24 22:51:26 -05:00 |
|
Thomas Harte
|
f601d796f5
|
Added documentation.
|
2016-12-24 22:37:20 -05:00 |
|
Thomas Harte
|
6e94d0c19f
|
Extended Storage::Disk::Disk to permit write-back of modified tracks, exposed some interface via Storage::Disk::Drive.
|
2016-12-24 22:11:31 -05:00 |
|
Thomas Harte
|
7f303cfceb
|
Continued the baby steps.
|
2016-12-24 21:54:43 -05:00 |
|
Thomas Harte
|
afc6f4129c
|
Withdrew unused tally.
|
2016-12-24 21:47:57 -05:00 |
|
Thomas Harte
|
1e416d4af0
|
Withdrew now-unused and never-implemented API from TimedEventLoop, and the redundant track time count from DiskController.
|
2016-12-24 21:02:10 -05:00 |
|
Thomas Harte
|
bedea48d03
|
This is a much better way of dealing with being partway into an incoming event. Subject to eliminating overruns, of course.
|
2016-12-24 20:54:27 -05:00 |
|
Thomas Harte
|
4cb17143ef
|
Messing around trying to lock down timing precisely. Which includes formal initial conditions.
|
2016-12-24 15:18:46 -05:00 |
|
Thomas Harte
|
1e970a9772
|
Started stepping slowly towards allowing writing on the disk controller, taking the opportunity to introduce self-simplifying behaviour to Storage::Time.
|
2016-12-24 13:07:23 -05:00 |
|
Thomas Harte
|
e4e0347638
|
Attempted to consolidate some of the repetition.
|
2016-12-21 22:17:00 -05:00 |
|
Thomas Harte
|
03579f33f1
|
Fixed multi-coverage insertion, via an appropriate test.
|
2016-12-20 21:38:32 -05:00 |
|
Thomas Harte
|
7eca910cc5
|
Fixed insertion location finding logic, working on the relevant test.
|
2016-12-20 21:14:05 -05:00 |
|
Thomas Harte
|
823ab9bc34
|
Completed initial non-trivial test, fixing revealed errors.
|
2016-12-20 19:15:36 -05:00 |
|
Thomas Harte
|
5a508ea0df
|
Attempted properly to cover the exactly-equal starts and ends cases, and to improve meaning.
|
2016-12-20 18:32:49 -05:00 |
|
Thomas Harte
|
63d861a2f3
|
Switched from C-in-the-brain manual offset counting to using iterators like an ordinary C++ person.
|
2016-12-20 18:17:54 -05:00 |
|
Thomas Harte
|
6f17076003
|
Switched to much more logical shared_ptr ownership of PCMSegmentEventSources by Periods.
|
2016-12-20 18:13:10 -05:00 |
|
Thomas Harte
|
497b2ae4dd
|
Still by manual inspection: the time for the next event should be provisional until proven acceptable, allowing a proper measurement of time until exiting the period to be taken; also fixed the accumulated period error when seeking back onto the underlying track.
|
2016-12-20 08:14:16 -05:00 |
|
Thomas Harte
|
6bdde542c5
|
Edging towards functioning automatic tests, fixed right-period adjustment and slightly decreased searching cost while in the process of adding a test.
|
2016-12-20 07:52:14 -05:00 |
|
Thomas Harte
|
ec624eaab1
|
Made an attempt fully to implement PCMPatchedTrack. Which now requires tests.
|
2016-12-20 07:30:57 -05:00 |
|
Thomas Harte
|
1ef1f6ec69
|
Attempted to implemnt seek_to and to finish add_segment. Started doing a little of get_next_event but ran out of time for the day.
|
2016-12-19 21:46:02 -05:00 |
|
Thomas Harte
|
8f937ceac8
|
Made an attempt to come up with a data structure that actually makes sense (though perhaps this is textbook list rather than vector stuff? I guess it depends on the frequency I expect inserts to occur versus reads) and to implement inserts. Though the Periods aren't yet honoured.
|
2016-12-19 07:42:43 -05:00 |
|
Thomas Harte
|
e081f224b6
|
Implemented a very basic PCMTrack test, nevertheless revealing an oversight in PCMSegmentEventSource related to improperly counting to the index hole if the final bit is set. Took that as a message that I should comment and document the event source.
|
2016-12-18 22:53:24 -05:00 |
|
Thomas Harte
|
a6354ebb01
|
Reimplemented PCMTrack to use PCMSegmentEventSource, eliminating code duplication.
|
2016-12-18 21:37:05 -05:00 |
|
Thomas Harte
|
f9a5595dad
|
Added seeking tests, correcting such errors as uncovered.
|
2016-12-18 10:19:24 -05:00 |
|
Thomas Harte
|
3297f6d545
|
Made an attempt to implement seek_to on PCMSegmentEventSource, taking account of off-by-half counting.
|
2016-12-17 22:44:33 -05:00 |
|
Thomas Harte
|
3116a2cf4c
|
Realised I was actually testing PCMSegmentEventSource, not PCMSegment; implemented a spread of tests; hence fixed PCMSegmentEventSource.
|
2016-12-17 21:47:13 -05:00 |
|
Thomas Harte
|
254cc41fd6
|
Made an attempt to separate and isolate the stuff of creating flux events from a PCMSegment, eventually to factor that out of PCMTrack and make it available also to PCMPatchedTrack.
|
2016-12-17 21:13:57 -05:00 |
|
Thomas Harte
|
313db75303
|
Ensured the patchable track owns its underlying track.
|
2016-12-17 18:17:22 -05:00 |
|
Thomas Harte
|
f1a08b7ab5
|
Opted to pass times by reference and added enough to PCMPatchedTrack that it could start being used by the disk controller, albeit that it doesn't work.
|
2016-12-17 16:26:45 -05:00 |
|
Thomas Harte
|
dc08a23ceb
|
This is going to be a slow walk, I think. This class attempts to be the scratchpad which will hold in-memory track modifications.
|
2016-12-16 19:20:38 -05:00 |
|
Thomas Harte
|
c4041b06a8
|
This'll do as a write interface, won't it?
|
2016-12-07 22:19:20 -05:00 |
|
Thomas Harte
|
0dc2aa6454
|
Commuted all of 'Storage' other than 'Tape' to postfix underscores.
|
2016-12-03 11:59:28 -05:00 |
|
Thomas Harte
|
81ee834530
|
As well as a bunch of logging, reinstated rotation position preservation across tracks.
|
2016-12-02 18:36:47 -05:00 |
|
Thomas Harte
|
93c573bfa9
|
Implemented missing status bits (other than the index hole), and a head loading delay for the Microdisc.
|
2016-12-01 21:13:16 -05:00 |
|
Thomas Harte
|
2222cb65d6
|
Split the status up into flags, assembled into a register upon demand. Attempted to implement some of the differences between the 1770/1772 and 1773/1793. Albeit with a motor fix still in place.
|
2016-11-30 22:26:02 -05:00 |
|
Thomas Harte
|
f44542c18c
|
Improved naming: this now explains what, not the mechanics of how.
|
2016-11-26 23:35:11 +08:00 |
|
Thomas Harte
|
5c8ecd3051
|
It probably needs a better name, but hastily implemented track caching at the Disk level.
|
2016-11-26 14:27:06 +08:00 |
|
Thomas Harte
|
2f86b07cfa
|
Added a parser for Oric-format MFM disks. Causing my first disk to load!
|
2016-11-26 13:40:10 +08:00 |
|
Thomas Harte
|
dcfdd73077
|
Switched to more conventional runtime polymorphism, also now allowing encoders (shifters as were) to be obtained, as a slightly lower-level interface than vector-of-sectors.
|
2016-11-26 13:39:20 +08:00 |
|
Thomas Harte
|
b180f04c87
|
Okay, so this file format wasn't what I hoped it was. It's another hack. Lots of work to do.
|
2016-11-26 10:19:10 +08:00 |
|
Thomas Harte
|
7613755f94
|
Fixed addressing: types are 1 and 2, not 0 and 1.
|
2016-11-26 10:13:12 +08:00 |
|
Thomas Harte
|
8499783b14
|
Dragged multibyte primitives and signature checks up to the base class. Implemented support for Oric MFM-style .DSK, at the file format level.
|
2016-11-21 20:47:16 +08:00 |
|
Thomas Harte
|
31c2548804
|
Created a base class for the boilerplate fopen stuff, switched as many classes as possible to its use, switched to postfix underscores and non-camelCase names.
|
2016-11-21 20:14:09 +08:00 |
|
Thomas Harte
|
ca53fac732
|
Switched to assuming a single-sided disk, moved out magic constants.
|
2016-09-26 21:20:30 -04:00 |
|
Thomas Harte
|
572d5587d9
|
Made a first stab at enabling multi-disk machines and thereby obeying (some of) the Plus 3's status register.
|
2016-09-25 21:24:16 -04:00 |
|
Thomas Harte
|
9bbcbd1001
|
Renamed class, intending to turn a Disk::Drive into literally just that, and have a thing with a PLL that consumes events be a Controller.
|
2016-09-25 20:05:56 -04:00 |
|
Thomas Harte
|
523dbb9678
|
This'll do for getting the ADF into the machine.
|
2016-09-25 18:32:26 -04:00 |
|
Thomas Harte
|
de863719d0
|
Made a first attempt at Acorn ADFS support plus the start of a suitable analyser.
|
2016-09-25 17:46:11 -04:00 |
|
Thomas Harte
|
a9e65e9b7a
|
Tweaked disk side density, added call-outs to a WD1770 if the Electron had one (albeit without run_for_cycles yet as I need to figure out the clock rate), added a shell of the basic functions of the WD1770. No implementation yet.
|
2016-09-19 22:06:56 -04:00 |
|
Thomas Harte
|
64f2538b1f
|
Added CRC checking to DFS comprehension; fixed a bunch of places where I'd used Objective-C's #import rather than #include.
|
2016-09-19 08:16:06 -04:00 |
|
Thomas Harte
|
8a1b805d11
|
Fixed file offset calculation for single-sided images.
|
2016-09-19 07:34:10 -04:00 |
|
Thomas Harte
|
b1e7f2dfd0
|
There's no cache and no CRC checking yet, but this is probably a rough outline of an FM parser.
|
2016-09-18 22:03:06 -04:00 |
|
Thomas Harte
|
d1c861d3a5
|
That should be that, I hope.
|
2016-09-18 21:09:32 -04:00 |
|
Thomas Harte
|
02c9a82cb5
|
Edging towards SSD/DSD support. Hold on!
|
2016-09-18 19:32:08 -04:00 |
|
Thomas Harte
|
91cd7e143b
|
Started on the SSD/DSD support. Realised I had ommitted multiple head support from my disk class. Fixed that.
|
2016-09-18 19:21:02 -04:00 |
|
Thomas Harte
|
5409c8ec54
|
Switched PCMSegments to std::vector; ensured generated [M]FM tracks are correctly sized, thereby making sure the individual flux windows will be correctly sized.
|
2016-09-18 18:56:35 -04:00 |
|
Thomas Harte
|
d9aaf456f0
|
Adapted pervasively to MSB-first output. Which seems to be correct.
|
2016-09-18 18:46:58 -04:00 |
|
Thomas Harte
|
180c3df2d4
|
Calling it 'number theory' probably isn't accurate but extracted the CRC stuff and started using it for [M]FM encoding.
|
2016-09-18 18:33:26 -04:00 |
|
Thomas Harte
|
55a7418cbf
|
Parameterised to perform FM and MFM encodings, at least subject to CRCs still being missing.
|
2016-09-18 17:16:20 -04:00 |
|
Thomas Harte
|
22eed60d2b
|
Started sketching out basic MFM track encoding. Which possibly even means that the shifters don't need to be public?
|
2016-09-18 16:53:21 -04:00 |
|
Thomas Harte
|
0089c830c6
|
This possibly correctly encapsultes the lowest level of FM and MFM rules.
|
2016-09-18 14:17:06 -04:00 |
|
Thomas Harte
|
bcf91de7e9
|
Declared support for the Acorn disk files, started hammering out an encoder.
|
2016-09-18 13:35:54 -04:00 |
|
Thomas Harte
|
79ef38b123
|
Attempted to use the new get_cycles_until_next_event method to take some repetition outside of the PLL and drive event loops.
|
2016-09-17 22:02:38 -04:00 |
|
Thomas Harte
|
9d6dcb80a7
|
Started work on a GCR parser and the helper functions that lie behind that.
|
2016-09-13 21:53:36 -04:00 |
|
Thomas Harte
|
82c8459055
|
Minor tidying of namespace usage declarations.
|
2016-08-27 17:18:12 -04:00 |
|
Thomas Harte
|
56c0d70c1f
|
Gave disks their own namespace.
|
2016-08-27 17:15:09 -04:00 |
|
Thomas Harte
|
9d76cd9b60
|
Removed dead instance storage.
|
2016-08-04 21:41:12 -04:00 |
|
Thomas Harte
|
3e90b85ff8
|
Made an attempt to insert proper conversions to/from rotational speed. The Time class is now really turning into a full-on quotient. Might need refactoring.
|
2016-08-04 21:36:39 -04:00 |
|
Thomas Harte
|
e57ab1025d
|
Fixed location calculation, up to a point.
|
2016-08-03 22:33:00 -04:00 |
|
Thomas Harte
|
9af9b28baf
|
Made very first attempt at closing the loop on this. But it's time for work.
|
2016-08-03 08:16:23 -04:00 |
|
Thomas Harte
|
e15241dc3c
|
Added ability to query how long since the new interval was set to the timed event loop. Discovered that LCM will returning the net effect of the common factors only. Otherwise continued iterating towards time preservation.
|
2016-08-03 07:49:00 -04:00 |
|
Thomas Harte
|
a3a3486f54
|
Okay, it's becoming more apparent where the (very mild) complexity will be here. But started moving towards retaining rotation between tracks.
|
2016-08-03 07:33:26 -04:00 |
|
Thomas Harte
|
21f1fa37a4
|
Fixed Time addition, added accumulation of distance into track into the disk drive, added a short circuit for LCM.
|
2016-08-03 07:26:05 -04:00 |
|
Thomas Harte
|
30f8b6baa4
|
Made an attempt to add the necessary extension to Track, with a concrete implementation in PCMTrack, to support time-based seeking, the intended mechanism for not magically spinning the disk back to the index hole upon every head step.
|
2016-08-03 06:59:45 -04:00 |
|
Thomas Harte
|
d832e5e10d
|
Reduced 1540 PLL to running at 4Mhz. Which is possibly correct (?) Made minor change to avoid divide if possible.
|
2016-08-02 21:28:50 -04:00 |
|
Thomas Harte
|
bc10b3ee9a
|
It appears the problem is as simple as sectors being counted from zero.
|
2016-08-01 10:08:38 -04:00 |
|
Thomas Harte
|
f5e4ea3351
|
Some minor tidying, lots more of the caveman stuff as I try to determine what I'm doing wrong.
|
2016-08-01 09:43:08 -04:00 |
|
Thomas Harte
|
a00f9adba3
|
Made a first attempt at D64 support. Made an error somewhere but this should be 90% of it.
|
2016-08-01 08:41:16 -04:00 |
|
Thomas Harte
|
58297f1baf
|
Performed the basic metadata and routing for opening D64 files. Realised that I wasn't actually necessarily catching exceptions properly for all file opens, and fixed.
|
2016-08-01 07:09:15 -04:00 |
|
Thomas Harte
|
2799a87218
|
Reduced possibility of overflow on LCM, improved commenting widely, removed one stale piece of G64 bootstrapping caveman stuff.
|
2016-08-01 06:04:55 -04:00 |
|
Thomas Harte
|
198fbbedc7
|
Reeled back all appropriate pieces of caveman debugging.
|
2016-07-31 13:42:34 -04:00 |
|
Thomas Harte
|
2332f72875
|
Formalised clock-rate multiplication within disk drives, discovered that the stepper didn't have ideal behaviour for my timed event loop and hence nailed down the semantics a ilttle more.
(obiter: the 1540 now appears to discern the correct sequence of bits. Framing is off in my test printfs but that's neither here nor there).
|
2016-07-31 13:32:30 -04:00 |
|
Thomas Harte
|
6ee784a893
|
Increased logging; discovered I was packing events together too closely. Now getting some zeros.
|
2016-07-30 04:30:55 -04:00 |
|
Thomas Harte
|
fead524eb5
|
Attempted to give the PLL a litte extra leeway, and fixed PCMTrack length test.
|
2016-07-29 18:52:50 -04:00 |
|
Thomas Harte
|
8f62211f5e
|
Wired up the 1540 as a PLL delegate. Which prima facie means it should start receiving a bit stream. Except that I clearly have something in the timing way off — either my flux transitions are far too short or I need to significantly increase the clock rate on the PLL.
|
2016-07-29 12:08:18 -04:00 |
|
Thomas Harte
|
89a1881fef
|
Started turning the 1540 into an actual disk drive.
|
2016-07-29 11:03:09 -04:00 |
|
Thomas Harte
|
6532b3a8c8
|
Filled in a calculation of clocks per bit that'll do.
|
2016-07-29 07:51:07 -04:00 |
|
Thomas Harte
|
f984de42a3
|
Took some small steps towards having a disk drive that at least can select a track and pump relevant events into a PLL.
|
2016-07-29 07:31:02 -04:00 |
|
Thomas Harte
|
e55db0cfe8
|
Made an attempt to eliminate creeping tape processing accuracy misses, which implied factoring out the GCM and LCD functions, which I then felt didn't really amount to signal processing.
|
2016-07-29 05:19:01 -04:00 |
|
Thomas Harte
|
5c1614ce7b
|
Attempted to simplify, very slightly.
|
2016-07-28 14:35:39 -04:00 |
|
Thomas Harte
|
015cea494d
|
Switched to a much-more straightforward PLL. I think I'm just fiddling now rather than moving forwards. Probably time to move on?
|
2016-07-28 11:32:14 -04:00 |
|
Thomas Harte
|
e061e849d4
|
Had a second bash at the PLL. Probably I should read some of the literature.
|
2016-07-27 16:24:24 -04:00 |
|
Thomas Harte
|
63f39608a6
|
Added just enough to get back to a working build.
|
2016-07-15 20:35:19 -04:00 |
|
Thomas Harte
|
165dbd9651
|
Started fleshing this out a bit. Hopefully.
|
2016-07-15 08:28:34 -04:00 |
|
Thomas Harte
|
0aa90b943b
|
Switched to specifying bit length as a quotient for the purposes of a PCMSegment and verified that I had the logic for picking a Commodore time zone backwards: bigger numbers are faster, not slower.
Started sketching out a DiskDrive class.
|
2016-07-15 06:51:11 -04:00 |
|
Thomas Harte
|
6afd619791
|
Eliminated floating point arithmetic.
|
2016-07-14 19:47:00 -04:00 |
|
Thomas Harte
|
6b4fec37ff
|
Moved down to a single divide.
|
2016-07-14 19:45:08 -04:00 |
|
Thomas Harte
|
481475a0f4
|
Switched to a full-on linear regression. Which causes the current tests to pass.
|
2016-07-14 19:42:01 -04:00 |
|
Thomas Harte
|
6d6b26b99f
|
Actually made things worse.
|
2016-07-14 07:32:27 -04:00 |
|
Thomas Harte
|
d8d3464c56
|
Made a quick-hack attempt at PLL synchronisation. Which doesn't work.
|
2016-07-14 07:31:23 -04:00 |
|
Thomas Harte
|
d1fe07f14d
|
Added test of perfect DPLL input timing.
|
2016-07-12 21:42:23 -04:00 |
|
Thomas Harte
|
94db45456e
|
Started sketching out the basic form here, albeit that it doesn't yet do _the only thing it advertises itself as useful for_.
|
2016-07-12 20:23:56 -04:00 |
|
Thomas Harte
|
75d95c0bc0
|
Sketched out an interface for a digial PLL. Not persuaded yet. Baby steps.
|
2016-07-11 22:12:58 -04:00 |
|
Thomas Harte
|
1e9eedc314
|
Factored out the PCM track since it's going to be a useful construct for almost every file format. Documented it a little better.
|
2016-07-10 18:36:52 -04:00 |
|
Thomas Harte
|
66895d3ac7
|
Actually, I think this is the correct conversion from received speed to clock rate. It'll become obvious if it's not when I get back to working on the 1541 itself.
|
2016-07-10 18:24:12 -04:00 |
|
Thomas Harte
|
19ee430d4a
|
Made an effort to support zoned tracks, at least.
|
2016-07-10 18:07:53 -04:00 |
|
Thomas Harte
|
6593caca93
|
Switched to a probably more helpful way around of expecting bits in bytes.
|
2016-07-10 16:21:52 -04:00 |
|
Thomas Harte
|
845a00ccef
|
Attempted via linear search to implement PCMTrack::get_next_event.
|
2016-07-10 16:17:25 -04:00 |
|
Thomas Harte
|
f9510c1b67
|
Put sufficiently much of PCMTrack into place to get to a stored list of segments, having determined a common clock rate between them and therefore a complete track length.
|
2016-07-10 16:10:05 -04:00 |
|
Thomas Harte
|
4ae9f5ad5d
|
Added preliminaries of reading the speed zone information.
|
2016-07-10 13:42:45 -04:00 |
|
Thomas Harte
|
8ae78ba4e0
|
Fixed signature check and fseek parameter order.
|
2016-07-10 13:32:59 -04:00 |
|
Thomas Harte
|
ff49857f5c
|
Started sketching out support for the G64 file format.
|
2016-07-10 10:17:53 -04:00 |
|
Thomas Harte
|
9e3d6b762b
|
Sketched out the generic interface for a disk, documenting it and the tape interface while I'm here.
|
2016-07-10 08:54:39 -04:00 |
|