Thomas Harte
b62f484d93
Start scaffolding a 65816 test generator.
2022-06-18 13:28:15 -04:00
Thomas Harte
6cc41d6dda
Restore 1000 test count.
2022-06-14 22:02:53 -04:00
Thomas Harte
d91f8a264e
Flip presumption, reenabling most tests.
2022-06-14 21:57:14 -04:00
Thomas Harte
e066546c13
Resolve PEA timing errors.
2022-06-13 14:08:42 -04:00
Thomas Harte
7dc66128c2
Fix strobe output.
2022-06-13 10:49:47 -04:00
Thomas Harte
e484e4c9d7
Expand test to make sure that correct data strobes are active.
2022-06-13 10:39:06 -04:00
Thomas Harte
f316cbcf94
The old implementation was correct.
2022-06-11 21:15:08 -04:00
Thomas Harte
0a6b2b7d32
Verify newer CMPA.l, RTE, TRAP[V] and CHK.
2022-06-11 11:17:18 -04:00
Thomas Harte
c3345dd839
Fix MOVEM timing.
2022-06-10 21:52:07 -04:00
Thomas Harte
917b7fbf80
Notarise won't fix status of CLR, NEGX, NEG, NOT.
2022-06-10 16:50:38 -04:00
Thomas Harte
97715e7ccc
Expand test set to include those with timing discrepancies.
2022-06-10 16:34:05 -04:00
Thomas Harte
43c0dea1bd
With the difference in RESET times now factored out, test timing too.
2022-06-10 16:12:54 -04:00
Thomas Harte
2e4652209b
Remove entire RESET sequence, move to testing PEA.
2022-06-10 15:57:54 -04:00
Thomas Harte
e2d811a7a0
Notarise digressions that appear to be correct, remove now-working RTE/RTR.
2022-06-09 21:48:15 -04:00
Thomas Harte
dd5c903fd6
DIVS also appears sometimes to differ.
2022-06-09 20:19:39 -04:00
Thomas Harte
2e1675066d
Reinstate address error non-testing.
2022-06-09 16:59:06 -04:00
Thomas Harte
be84ce657b
Add an optional testing whitelist.
2022-06-09 16:18:04 -04:00
Thomas Harte
64053d697f
Take improved guess at address error stacking order.
2022-06-09 16:17:09 -04:00
Thomas Harte
a59ad06438
Print out summary of failure.
2022-06-09 13:13:33 -04:00
Thomas Harte
5af03d74ec
Add note to self about first diagnosis.
2022-06-09 12:21:39 -04:00
Thomas Harte
ba2803c807
Include all bus activity after the split.
2022-06-09 11:30:22 -04:00
Thomas Harte
fdcbf617d8
Avoid STOP.
2022-06-09 08:42:31 -04:00
Thomas Harte
2e42bda0a3
Permit instructions that end in an address error to differ in transactions.
2022-06-08 16:15:33 -04:00
Thomas Harte
168dc12e27
Avoid spurious mismatches.
2022-06-08 16:03:02 -04:00
Thomas Harte
fd1955e15b
Attempt to randomise and test register contents.
2022-06-08 15:12:47 -04:00
Thomas Harte
f4f93f4836
Test a single, whole instruction; record read/write.
2022-06-08 14:53:04 -04:00
Thomas Harte
dd0a7533ab
Randomise all parts of memory other than the opcode.
2022-06-08 14:43:51 -04:00
Thomas Harte
50130b7004
Minor layout tweak.
2022-06-08 11:42:42 -04:00
Thomas Harte
ab52c5cef2
Pass first all-zeroes test, establishing that processors aren't being fully reset.
2022-06-08 10:56:54 -04:00
Thomas Harte
c7fa93a5bc
Attempt human-legible explanation of differences encountered.
2022-06-08 10:51:05 -04:00
Thomas Harte
400b73b5a2
Allow capture to be limited; retain timestamps.
2022-06-08 09:49:27 -04:00
Thomas Harte
788b026cf5
Log and attempt to compare some activity. Sort of.
2022-06-07 16:56:05 -04:00
Thomas Harte
c4ae5d4c8d
Establishes at least that both 68000s can run.
2022-06-06 21:47:10 -04:00
Thomas Harte
ca8dd61045
Start sketching out an old vs new 68000 test.
2022-06-06 21:19:57 -04:00
Thomas Harte
7b3cf6e747
Add missing instruction: RESET.
2022-06-03 11:15:39 -04:00
Thomas Harte
640b04e59e
Test only well-defined flags.
...
Albeit that timing is still off.
2022-06-03 10:18:46 -04:00
Thomas Harte
10b9b13673
Disable divide-by-zero PC test in lieu of better documentation.
2022-06-03 08:27:20 -04:00
Thomas Harte
90d720ca28
Don't test undocumented flags.
2022-06-02 12:30:39 -04:00
Thomas Harte
6dd89eb0d7
Adjust my expectation as to length.
2022-06-02 12:11:54 -04:00
Thomas Harte
e1abf431cb
Don't test undefined flags.
2022-05-30 16:23:51 -04:00
Thomas Harte
8e0fa3bb5f
DIV # with a divide by zero should be 44 cycles.
2022-05-29 21:22:45 -04:00
Thomas Harte
9eea471e72
Resolve infinite recursion.
2022-05-29 20:39:22 -04:00
Thomas Harte
2a40e419fc
Fix CHK tests: timing and expected flags.
2022-05-29 15:26:56 -04:00
Thomas Harte
5f030edea4
Simplify transaction.
2022-05-26 19:37:30 -04:00
Thomas Harte
88e33353a1
Fix instruction and time counting, and initial state.
2022-05-26 09:17:37 -04:00
Thomas Harte
f3c0c62c79
Switch register-setting interface.
2022-05-26 07:52:14 -04:00
Thomas Harte
866787c5d3
Make an effort to withdraw from the high-circuitous stuff of working around the reset sequence.
2022-05-25 20:22:38 -04:00
Thomas Harte
64491525b4
Work further to guess at caller's intention for set_state.
...
Probably I should just eliminate the initial reset, somehow.
2022-05-25 17:01:18 -04:00
Thomas Harte
68b184885f
Reapply only the status.
2022-05-25 16:54:25 -04:00
Thomas Harte
06f3c716f5
Make better effort to establish initial state.
2022-05-25 16:47:41 -04:00
Thomas Harte
22714b8c7f
Capture state at instruction end, for potential inspection.
2022-05-25 16:32:26 -04:00
Thomas Harte
f9d1c554b7
Fix for the actual number of cycles in a standard reset.
2022-05-25 16:05:28 -04:00
Thomas Harte
f2a7660390
Merge branch 'master' into 68000Mk2
2022-05-25 15:40:10 -04:00
Thomas Harte
4961e39fb6
Mention DIVU/DIVS flags.
2022-05-25 15:39:00 -04:00
Thomas Harte
0bedf608c0
Add details on gaps in coverage.
2022-05-25 15:36:27 -04:00
Thomas Harte
1ab831f571
Add the option to log a list of all untested instructions.
2022-05-25 13:17:01 -04:00
Thomas Harte
2c6b9b4c9d
Switch comparative trace tests to 68000 Mk2.
2022-05-25 11:32:00 -04:00
Thomas Harte
463fbb07f9
Adapt remaining 68000 tests to use Mk2.
2022-05-25 10:55:17 -04:00
Thomas Harte
4b07c41df9
Ensure alignment of storage.
2022-05-24 11:29:28 -04:00
Thomas Harte
a87f6a28c9
Fix LINK A7.
2022-05-23 10:43:17 -04:00
Thomas Harte
98325325b1
Fix UNLINK A7.
2022-05-23 10:27:44 -04:00
Thomas Harte
26bf66e3f8
Fix shifts and rolls.
2022-05-23 10:09:46 -04:00
Thomas Harte
c6b3281274
Attempt the shifts and rolls.
2022-05-23 09:29:19 -04:00
Thomas Harte
1e8adc2bd9
Fix MOVEP to R.
2022-05-23 09:00:37 -04:00
Thomas Harte
c73021cf3c
Implement MOVE.
2022-05-23 08:46:06 -04:00
Federico Berti
1a26d4e409
Update nbcd_pea.json
...
Add missing bracket
2022-05-23 12:14:00 +01:00
Thomas Harte
269263eecf
Implement RTE, RTS, RTR.
2022-05-22 21:16:38 -04:00
Thomas Harte
4e21cdfc63
Enable NEGX/CLR tests.
2022-05-22 20:55:21 -04:00
Thomas Harte
faef5633f8
Ensure MOVE from SR has an effective address to write to.
2022-05-22 20:52:00 -04:00
Thomas Harte
7d1f1a3175
Implement MOVE [to/from] [CCR/SR].
2022-05-22 19:45:22 -04:00
Thomas Harte
4e34727195
Fully implement TAS.
2022-05-22 16:14:03 -04:00
Thomas Harte
1dd6ed6ae3
Implement TAS Dn, with detour for other TASes.
2022-05-22 16:08:30 -04:00
Thomas Harte
cb4d6710df
Switch to a more direct indication of progress.
2022-05-22 11:27:58 -04:00
Thomas Harte
284f23c6ea
Implement JMP.
2022-05-22 07:16:38 -04:00
Thomas Harte
4b35899a12
Bcc: properly establish offset.
2022-05-21 20:59:34 -04:00
Thomas Harte
94288d5a94
Excludes DBcc from standard operand fetch.
2022-05-21 19:53:28 -04:00
Thomas Harte
c869eb1eec
Correct omission: wasn't testing the final PC.
...
Plenty of new errors incoming.
2022-05-21 15:56:27 -04:00
Thomas Harte
176c8355cb
The tests in chk.json now pass.
2022-05-21 14:32:58 -04:00
Thomas Harte
e46a3c4046
Implement JSR.
2022-05-21 10:29:36 -04:00
Thomas Harte
256da43fe5
Fix MOVEM other than postinc and predec.
2022-05-20 20:47:54 -04:00
Thomas Harte
a818650027
Add a faulty attempt at MOVEM.
2022-05-20 18:48:19 -04:00
Thomas Harte
c7c12f9638
After a quick check, eori_andi_ori also now passes.
2022-05-20 14:47:11 -04:00
Thomas Harte
ee942c5c17
Fix PC-relative fetches.
2022-05-20 14:42:51 -04:00
Thomas Harte
d157819c49
Implement the various to-[SR/CCR] actions, which do a 'repeat' prefetch.
...
(which isn't exactly a repeat, at least in the SR cases, because the function code might have changed)
2022-05-20 14:29:14 -04:00
Thomas Harte
2d91fb5441
Implement MOVEP.
2022-05-20 14:22:32 -04:00
Thomas Harte
81431a5453
Attempt BTST, BCHG, BCLR and BSET.
2022-05-20 12:58:45 -04:00
Thomas Harte
6d7ec07216
Uncover another three already-working test files.
2022-05-20 12:44:57 -04:00
Thomas Harte
b4978d1452
Implement BSR, adding one more test file to the working set.
2022-05-20 12:40:35 -04:00
Thomas Harte
45e9648b8c
Implement Bcc.
2022-05-20 12:04:43 -04:00
Thomas Harte
ce32957d9d
Shuffle two more into the working column.
2022-05-20 11:53:12 -04:00
Thomas Harte
452dd3ccfd
Add a performer call-out for Scc; use it to implement proper timing in the mk2 68000.
2022-05-20 11:20:23 -04:00
Thomas Harte
e5c1621382
Add missing fallthrough
, patterns for all ADDs and SUBs.
2022-05-20 07:02:02 -04:00
Thomas Harte
af3518dc1f
Implement various ADD, SUB patterns.
2022-05-19 20:50:37 -04:00
Thomas Harte
6cfc0e80d9
Don't test the unrecognised instruction exception.
2022-05-19 19:45:38 -04:00
Thomas Harte
334e3ec529
Add privilege and instruction error exceptions; permit two operands to be stored.
2022-05-19 16:55:16 -04:00
Thomas Harte
84c165459f
ext.json now passes.
2022-05-19 16:32:40 -04:00
Thomas Harte
22b63fe1f8
Add EXT, and notes to self.
2022-05-19 15:41:02 -04:00
Thomas Harte
c6c6213460
Bifurcate the fetch-operand flow.
...
Address calculation will be the same, but the fetch will differ. I don't think there's a neat costless way to factor out the address calculations, alas, but I'll see whether macros can save the day.
2022-05-19 10:27:51 -04:00
Thomas Harte
29f6b02c04
Factor out register setup/testing, generalising the DIVU/DIVS flag check.
2022-05-18 21:13:34 -04:00
Thomas Harte
1bf7c0ae5f
Attempt better to avoid entering a second instruction.
2022-05-18 21:00:34 -04:00
Thomas Harte
44ae084794
Avoid the repeated .fill; reduces debug-build executor test time to 1.5s.
...
i.e. eliminates about 95% of costs.
2022-05-18 17:10:23 -04:00
Thomas Harte
13a1809101
Avoid memset.
2022-05-18 17:00:35 -04:00
Thomas Harte
c35200fbd0
Shuffle mildly, primarily to avoid repeated 16mb allocations.
2022-05-18 16:59:37 -04:00
Thomas Harte
4a40581deb
Completes performance of NBCD D0.
2022-05-17 16:10:20 -04:00
Thomas Harte
3db2de7478
Works 68000 mk2 into the comparative tests.
...
... revealing that I've leant a little too hard on __LINE__.
2022-05-16 20:04:13 -04:00
Thomas Harte
20a191f144
Switch to same tests, run through a more modern emulator.
2022-05-15 16:33:08 -04:00
Thomas Harte
f60f1932f2
Restrict DIVU and DIVS tests to those which are well-defined.
2022-05-14 20:28:54 -04:00
Thomas Harte
7f704fdae1
Improve README.
2022-05-13 16:28:56 -04:00
Thomas Harte
dd63a6b61e
Correct all [A/S/N]BCD tests.
2022-05-13 16:18:58 -04:00
Thomas Harte
1935d968c5
Add ability to suggest solutions.
2022-05-13 15:27:11 -04:00
Thomas Harte
84cfbaa0a4
Remove manual test count, now that all are being performed.
2022-05-13 11:00:26 -04:00
Thomas Harte
0d81992f6a
Move object creation.
2022-05-13 10:50:16 -04:00
Thomas Harte
6594b38567
Tidy up, and reduce for now to a summary report.
2022-05-13 08:02:20 -04:00
Thomas Harte
2e796f31d4
Support interrupts; documentation to come.
2022-05-12 20:52:24 -04:00
Thomas Harte
3d8f5d4302
Improve failure logging.
...
This confirms that it's only the *BCDs and DIVU/DIVS in which I do not match the tests.
2022-05-12 20:23:32 -04:00
Thomas Harte
2fa6b2301b
Move string logic into Preinstruction
.
2022-05-12 19:46:08 -04:00
Thomas Harte
4ba20132b9
Avoid repeated allocations on the new path, reducing total runtime by almost two thirds.
2022-05-12 16:35:41 -04:00
Thomas Harte
192513656a
After much guesswork, fix SBCD and thereby pass flamewing tests.
2022-05-12 11:39:01 -04:00
Thomas Harte
f3c1b1f052
Name flags, remove closing underscores on exposed data fields.
2022-05-12 08:19:41 -04:00
Thomas Harte
56ce1ec6e8
No need to subclass.
2022-05-11 21:25:38 -04:00
Thomas Harte
de168956e4
Fix tested operand order.
2022-05-11 16:44:39 -04:00
Thomas Harte
5b80844d81
Add a sanity test count, temporarily.
2022-05-11 16:34:28 -04:00
Thomas Harte
17add4b585
Introduce and overwhelmingly fail the flamewing BCD tests.
2022-05-11 15:19:39 -04:00
Thomas Harte
943c924382
Add missing: MOVE to/from USP, RESET.
2022-05-11 07:52:23 -04:00
Thomas Harte
f2a6a12f79
Remove further vestiges of timing.
2022-05-09 20:58:51 -04:00
Thomas Harte
0af8660181
Remove add_pc
and decline_branch
in favour of operation-specific signals.
2022-05-09 16:19:25 -04:00
Thomas Harte
330ec1b848
TODO is done.
2022-05-09 11:52:33 -04:00
Thomas Harte
539932dc56
Provide function codes. TODO: optionally.
2022-05-09 09:18:02 -04:00
Thomas Harte
5ab5e1270e
Fix test for new MOVEM semantics.
2022-05-09 09:17:48 -04:00
Thomas Harte
98cb9cc1eb
Fix CHK operand size.
2022-05-07 21:16:44 -04:00
Thomas Harte
bf8c97abbb
Permit TRAP, TRAPV and CHK to push the next PC rather than the current.
2022-05-07 20:32:39 -04:00
Thomas Harte
2b3900fd14
Fix LINK A7.
2022-05-07 08:15:26 -04:00
Thomas Harte
1defeca1ad
Implement RTS, RTR, RTE.
2022-05-06 12:30:49 -04:00
Thomas Harte
ac6a9ab631
Fix TAS Dn.
2022-05-06 12:23:04 -04:00
Thomas Harte
8176bb6f79
Expose issues with TST and TAS.
2022-05-06 12:18:56 -04:00
Thomas Harte
9c266d4316
Proceed to unimplemented TST.
2022-05-06 11:33:57 -04:00
Thomas Harte
d478a1b448
Proceed to next failure: PEA.
2022-05-06 10:04:20 -04:00
Thomas Harte
607ddd2f78
Preserve MOVEM order in Operation
.
2022-05-06 09:45:06 -04:00
Thomas Harte
06fe320cc0
Correct source counting, but this leaves the operands still being the wrong way around.
2022-05-05 21:06:53 -04:00
Thomas Harte
d7d0a5c15e
Implement MOVEM to memory.
2022-05-05 18:51:29 -04:00
Thomas Harte
47f4bbeec6
Switch to a contiguous block of 16 registers.
2022-05-05 15:31:59 -04:00
Thomas Harte
70cdc2ca9f
Fix MOVEP to register.
...
Advance to lack of MOVEM.
2022-05-05 12:37:47 -04:00
Thomas Harte
f63a872387
BTST does not write back.
2022-05-05 12:32:15 -04:00
Thomas Harte
46686b4b9c
Start testing move.
2022-05-04 20:38:56 -04:00
Thomas Harte
15c90e546f
Fix rotates and shifts to memory.
2022-05-04 19:44:59 -04:00
Thomas Harte
5aabe01b6d
Mostly fix LINK and UNLK.
2022-05-04 08:41:55 -04:00
Thomas Harte
d3b55a74a5
Fix LEA, proceed to non-functional LINK and UNLK.
2022-05-03 20:45:36 -04:00
Thomas Harte
de58ec71fd
Fix EXT, SWAP.
2022-05-03 20:17:36 -04:00
Thomas Harte
052ba80fd7
Add enough wiring to complete but fail EXT and JMP/JSR.
2022-05-03 15:49:55 -04:00
Thomas Harte
39f0ec7536
Get far enough through CHK to realise that MOVEM probably needs to be divided by direction.
2022-05-03 15:40:04 -04:00