Retro/pcjs
Watch
1
0
Fork
You've already forked pcjs
 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
2,881 commits 4 branches 83 tags 557 MiB
Commit graph pcjs/versions
Author SHA1 Message Date
Jeff Parsons
0bd1be0df9 When a 'state' file is passed as a URL parameter, that's supposed to override a machine's 'resume' setting, but if a 'state' file is passed as EITHER a computer parameter (the original XML-based method) OR as a machine parameter (a newer method that allows a web page to override certain XML settings), that is NOT supposed to override the 'resume' setting. This restores the ability of the DONKEY.BAS demo to BOTH define a state file AND allow the machine to be suspended and resumed. Additionally, a new-ish Video card property (offStartAddr), which was originally added for the EGA/VGA, was not properly initialized for other cards, so when we restored the display of a suspended machine with a CGA, nothing would be displayed, due to lots of NaN buffer address calculations. 2017-06-22 17:45:36 -07:00
Jeff Parsons
a16f0586f4 More alert suppression changes 2017-06-22 16:36:51 -07:00
Jeff Parsons
3ba6f0e047 Changed the alert style for restore failures 2017-06-22 16:06:41 -07:00
Jeff Parsons
1ce03810d3 Added support for (attempting to) suppress alerts that are meaningless if the user has aborted a machine by exiting the page before it was done loading 2017-06-22 15:53:34 -07:00
Jeff Parsons
9626fed7c2 Got rid of more app-specific demo machines (eg, DONKEY and WIN101); demos like that are more properly located in either the /apps or /disks folders (not /devices) 2017-06-22 12:03:52 -07:00
Jeff Parsons
6fd2a3bcb6 Started work on v1.35.1 2017-06-21 11:18:02 -07:00
Jeff Parsons
b954bed327 Source maps should be checked in now, too 2017-06-21 10:51:26 -07:00
Jeff Parsons
921fe75d7d autoType sequences no longer require double backslashes 2017-06-21 10:48:31 -07:00
Jeff Parsons
07b330bc66 The Disk component needed improved handling of disk images containing sectors with no 'data' or 'byte' arrays, the Disk restore code needed to stop complaining about disk mismatches when there was no previous disk, and Disk notice() and println() calls (including those made on its behalf; eg, from assert() calls) needed to be passed through to the controller that called it. Also, since the PC-DOS 1.x disk images were changed recently to include BPBs, machines that use those disks and have a machine state file needed their state file updated (eg, the DONKEY.BAS demo). 2017-06-20 10:24:51 -07:00
Jeff Parsons
d527e1bbe3 Updated the VisiCalc demo with an autoType sequence, a fresh disk image, new state file, and reference cards; also, updated autoType support to cancel autoType if a saved machine state is loaded 2017-06-19 08:48:42 -07:00
Jeff Parsons
6d019e7d9d Fixed autoMount overrides (formerly, machines that wanted to override the diskette autoMount settings for a machine would set "autoMount: ''", but that was a sledgehammer solution, especially when the goal was merely to prevent a diskette from being loaded in drive A; the preferred solution now is to set "autoMount: A: None", allowing the rest of the machine's autoMount settings, if any, to remain intact) 2017-06-17 21:30:03 -07:00
Jeff Parsons
cb43a6b360 Fixed cancellation of key injection whenever a soft key is pressed 2017-06-17 17:43:43 -07:00
Jeff Parsons
fd3348f907 Added an autoType sequence to the RatBas demo 2017-06-17 10:56:22 -07:00
Jeff Parsons
8f8082d170 Stop keyboard injection as soon as any real keyboard activity is detected 2017-06-15 16:48:12 -07:00
Jeff Parsons
21e252b2b9 Updated v1.35.0 files 2017-06-15 16:04:10 -07:00
Jeff Parsons
d5b21be782 Updated v1.35.0 files 2017-06-15 16:03:05 -07:00
Jeff Parsons
f0f3d8fa42 Added v1.35.0 files 2017-06-15 15:12:30 -07:00
Jeff Parsons
3e1aa196bd Added an empty 10Mb disk image formatted by PC-DOS 2.0 (currently used by one machine: 5160/cga/640kb); also fixed some issues uncovered in the process of creating the disk image (eg, a bug in the HDC's save logic when no image has been loaded), and made some fixes and optimizations in DiskDump 2017-06-01 15:32:17 -07:00
Jeff
ea95de4939 Fixed (or rather, band-aided) a MACRO-10 regression assembling the nested macro test (/apps/pdp10/tests/macro10/TEST.MAC) 2017-04-05 20:42:44 -07:00
Jeff
5f72f1627f Updated the first PDP-10 reliability test (DAKBA) 2017-04-05 16:33:16 -07:00
Jeff
6bd24c5347 Fixed how index operands are assembled into PDP-10 instructions, which was an unexpectedly convoluted process (they are NOT masked to 4 bits, shifted left 18 bits, and OR'ed into the opcode; instead, they are computed as a 36-bit value whose left and right 18-bit halves are swapped and then ADDED to the opcode, which means that an instruction encoded with indexing may not end up using indexing at all, depending on where all those bits fall) 2017-04-05 12:38:36 -07:00
Jeff
d4c12c8c7a Fixed the PDP-10 BLT instruction 2017-04-05 11:32:16 -07:00
Jeff
7e1fc5586e Added a minor improvement to the simplistic MACRO-10 fixup support in PDPjs 2017-04-04 16:34:03 -07:00
Jeff
427e6afdb8 Fixed OPDEF opcode generation (improper use of bitwise operators) 2017-04-04 11:26:49 -07:00
Jeff
d28d1ee940 Added a few PDP-10 byte pointer fixes, along with the ability to pass MACRO-10 text directly into the MACRO-10 assembler (use the Debugger's "as" command) 2017-04-04 10:42:41 -07:00
Jeff
38f709bb7a Corrected a few expression parser changes that had adversely affected other debuggers (eg, PCx86) 2017-04-03 16:27:52 -07:00
Jeff
09dc4dcb9a Lots of MACRO-10 improvements, DAKAM now assembles (but does not pass) 2017-04-03 16:10:08 -07:00
Jeff
3fb3f44384 Made a tiny 72-bit division optimization 2017-04-02 11:43:54 -07:00
Jeff
01550cc95e Well, that was brutal, but ALL the PDP-10 DAKAL multiplication and division tests FINALLY pass 2017-04-02 10:35:59 -07:00
Jeff
493ae6c310 I *think* all the PDP-10 DAKAL multiplication tests pass now. Division tests, alas, are a different story. 2017-04-01 23:06:41 -07:00
Jeff
470bd1d69e Fixed unary operators applied to bracketed expressions 2017-04-01 21:11:01 -07:00
Jeff
1a8c1306d4 Fixed 36-bit multiplication results in cases of overflow 2017-04-01 15:14:27 -07:00
Jeff
75b9a9c2ba Fixed the Debugger's arithmetic truncation function (truncate()) 2017-04-01 14:36:39 -07:00
Jeff
1711393755 Fixed 36-bit multiplication in the expression parser so that, within the PDP-10 environment, it produces the same 36-bit results that a PDP-10 would produce (important for MACRO-10 compatibility) 2017-04-01 00:43:44 -07:00
Jeff
26de8bbb1a Added another PDP-10 KA10 diagnostic (DAKAL) 2017-03-31 19:52:11 -07:00
Jeff
1786b3bf31 Fixed some regressions in MACRO-10 support (all the KA10 diagnostics archived so far assemble and pass again) 2017-03-31 18:43:13 -07:00
Jeff
afd621dc79 Updated the PDP-10 DAKAK diagnostic page, and verified the ability to assemble (but not run) all the original DAKAK source files, after making further improvements to the built-in MACRO-10 Mini-Assembler 2017-03-31 16:32:37 -07:00
Jeff
e7988597c5 Changed the remainder operator from '%' to '^/', mainly because '%' must be allowed in MACRO-10 symbols instead, but also because '%' led one to believe it was a "mod" operator (it wasn't); added IFDEF support to the MACRO-10 Mini-Assembler; updated the Debugger's assemble command to recognize filenames with "KLM" extensions, too. 2017-03-31 09:37:38 -07:00
Jeff
71e2ecb5cb PDP-10 "DAKAK" multiplication diagnostic passes 2017-03-31 08:14:56 -07:00
Jeff
307517fd76 PDP-10 multiply diagnostic testing finally underway; fixed one bug related to the KA10 (not that anyone really cares anymore) 2017-03-30 23:46:01 -07:00
Jeff
90ae3f1db8 Fixed the signed 8-bit calculation of all PDP-10 shift/rotate counts; diagnostic DAKAJ now passes 2017-03-30 17:31:13 -07:00
Jeff
881da5b19c Added another PDP-10 diagnostic (DAKAJ), which in turn required some more MACRO-10 support (eg, macro parameters with backslashes) 2017-03-30 16:04:00 -07:00
Jeff
3f06667597 Fixed/simplified opASH() a bit; it wasn't preserving the sign bit in all cases, and converting the input from unsigned to signed and back again was more confusing than helpful 2017-03-30 14:49:49 -07:00
Jeff
c6de12287c Documented a (mildly) interesting MACRO-10 bug 2017-03-30 12:43:24 -07:00
Jeff
1d1ac0cb7b The shared expression parser now adapts to the (critically) different precedence of MACRO-10 operators 2017-03-30 11:53:21 -07:00
Jeff
ab41fdef3c More robust handling of PDP-10 memory errors (so that when a file is assembled and loaded into non-existent memory, an error is generated); bumped the PDP-10 test machine to 32K words 2017-03-30 09:57:53 -07:00
Jeff
ec642a8c5b Made the Debugger's assemble command more flexible (the target address is optional now; the Debugger will use whatever load and/or start addresses the file(s) specified instead) 2017-03-30 09:04:56 -07:00
Jeff
15417ca686 DAKAG PDP-10 diagnostic passes now (fixed the relative addressing location used within literal scopes) 2017-03-29 13:59:50 -07:00
Jeff
c045e78395 Added the DAKAG PDP-10 diagnostics, implemented PUSHJ and POPJ instructions, and improved the debugging of XCT sequences 2017-03-29 13:26:23 -07:00
Jeff
d0ff759684 Added MACRO-10 support for OPDEFs, which are macro-like statements used to generate words which can also be combined with addressing modifiers 2017-03-29 10:12:45 -07:00