Many hyperlinks are disabled.
Use anonymous login
to enable hyperlinks.
181 check-ins going from 2a11db3ffe41df6d to 1bec0d528f671d99
|
2017-12-09
| ||
| 01:25 | Rebuilt cc8.tu56 in case the buffer overrun in txt2ptp affected its content. check-in: 2a11db3ffe user: tangent tags: trunk | |
|
2017-12-08
| ||
| 23:38 | Double size of obuf. Worst case of "need one more output buffer character space to accomodate adding \r before \n" check-in: 18fb7d3bb4 user: poetnerd tags: trunk | |
| 20:21 | . check-in: 20e18746ab user: tangent tags: trunk | |
| 18:18 | Updated cc8.tu56 to contain latest n8.c changes. check-in: bbd9fc92a6 user: tangent tags: trunk | |
| 17:26 | Updated src/cc8/os8/n8.c with Ian's latest fixes. check-in: 0e01e6b563 user: tangent tags: trunk | |
|
2017-12-07
| ||
| 20:32 | . check-in: c901777e25 user: tangent tags: trunk | |
| 20:24 | . check-in: 1b3ae6c27b user: tangent tags: trunk | |
| 18:15 | Added FORTRAN II and FORTRAN IV versions of PEP001, correpsonding to the versions described in the current version of the PEP001.FT wiki article. check-in: 1dde81a5e4 user: tangent tags: trunk | |
| 18:14 | Updated pep001.c to the current version given in the wiki article check-in: 28e2ce8371 user: tangent tags: trunk | |
|
2017-12-06
| ||
| 19:01 | Assorted improvements to the description for the "./configure --prefix" option. check-in: 71f21fddc8 user: tangent tags: trunk | |
| 18:52 | Added a brief "Running the Software" section to the top-level README to cover the PATH modification issue. check-in: e8e6290de2 user: tangent tags: trunk | |
| 18:46 | Clarified the bits about "make reconfig" in the top-level README check-in: 9769d8929d user: tangent tags: trunk | |
| 15:27 | Initial implementation of pep001.fc check-in: 804af8e141 user: tangent tags: trunk | |
| 13:41 | Improved description of the DECUS submission in the U/W FOCAL manual supplement's References section. check-in: ac7600237c user: tangent tags: trunk | |
| 13:37 | Raised a few heading levels in the supplement check-in: 1b63be1563 user: tangent tags: trunk | |
| 13:37 | Reworked the U/W FOCAL manual supplement's rationale section somewhat. check-in: c41bab7c4a user: tangent tags: trunk | |
| 13:17 | Fixed a few more OCR errors in uwfocal-manual.md check-in: 182ac763c4 user: tangent tags: trunk | |
| 00:20 | Merged usb-scan-depth-increase experimental feature branch into trunk. check-in: ea7ee05b55 user: tangent tags: trunk | |
|
2017-12-05
| ||
| 22:54 | No longer rebuilding bin/*.rk05 just because the simulator got rebuilt. (Moved $(SIM) to the other side of the | in the rule to make it an exists-only dependency.) check-in: fd495f8920 user: tangent tags: trunk | |
| 13:53 | . check-in: 79f607a22e user: tangent tags: trunk | |
| 13:07 | Reworked pep001.c in terms of the new modulus operator support. (The only reason I checked in the previous ire0() based versions is because they're used to tell the story in the PEP001.C wiki article I'm composing, and I'd like to be able to pull those versions back out of source control again later.) check-in: 199a0dff7e user: tangent tags: trunk | |
| 12:53 | Simplified pep001.c a bit more. check-in: daf2ff9090 user: tangent tags: trunk | |
| 12:32 | More whitespace fixes to CC8 check-in: 04712bea7d user: tangent tags: trunk | |
| 12:30 | Whitespace fix check-in: 31fc8dd267 user: tangent tags: trunk | |
| 12:10 | Reworked examples/pep001.c in terms of a new subroutine, ire0(). check-in: 3458e3e939 user: tangent tags: trunk | |
| 11:48 | Removed an unnecessary #include from pep001.c check-in: b218a6484c user: tangent tags: trunk | |
| 11:47 | Small fixes to examples/pep001.c check-in: 70613e5eeb user: tangent tags: trunk | |
| 10:32 | Extracted the manual COMP and LOADER steps in tools/cc8-tu56-update out as an OS/8 batch script (bldcc8.bi) which we send along with all the other static text files. This eliminates the possibility of any differences in the back-and-forth between cc8-tu56-update, pexpect, and SIMH from causing differences in the resulting cc8.tu56 file, which we're also checkin in at this time. check-in: ab58d17e25 user: tangent tags: trunk | |
| 09:17 | Typo fix check-in: 42bc4c6bd5 user: tangent tags: trunk | |
| 09:15 | Applied GPLv3 license to CC8, per private email from Ian Schofield on 2017.07.15 and 16. check-in: f4827a481f user: tangent tags: trunk | |
| 09:12 | Assorted improvements to the ChangeLog check-in: 19b733c33d user: tangent tags: trunk | |
| 08:53 | Removed src/cc8/os8/os8.sb: apparently unused. check-in: 0f784fb71d user: tangent tags: trunk | |
| 08:52 | Removed another reference to the now-removed boot/cc8.script file. check-in: 6e0be06dc9 user: tangent tags: trunk | |
| 08:20 | Added a hyphen in "Small-C" to one of the uses in the CC8 README check-in: 59741762d2 user: tangent tags: trunk | |
|
2017-12-04
| ||
| 13:09 | Reworked the "build PS.CC" example in src/cc8/README.md to use CC0 instead of CC1. I find it considerably easier. check-in: 1d737f0325 user: tangent tags: trunk | |
| 13:06 | Removed cc8.rk05 and its boot script now that os8v3d-*.rk05 is a functional superset of it. check-in: 882b8a05b7 user: tangent tags: trunk | |
| 12:52 | bosi prepare now sets up auto-expanding root filesystem check-in: 666c3dbe59 user: tangent tags: trunk | |
| 11:43 | Moved the "zero core memory" step up a bit within cc8-tu56-update: it was between the mag tape and disk pack attachment commands to SIMH, and there's no good reason to split those two. Since it's easier to do it before attaching media than to slide it in between the attachments and the boot command, I've done it that way. check-in: 0a87f19eed user: tangent tags: trunk | |
| 11:34 | Updated cc8.tu56 after the zero-core fix. check-in: bd1d639081 user: tangent tags: trunk | |
| 11:32 | cc8-tu56-update now zeroes core before beginning work. Since it uses the OS/8 SAVE command, it really needs known prior core contents before we do the SAVE, else the output is nondeterministic. check-in: 5b75dd49a0 user: tangent tags: trunk | |
| 08:52 | Re-generated cc8.tu56 after previous updates check-in: 1fe3644a71 user: tangent tags: trunk | |
| 08:51 | Fixed escaping of a * prompt in a regex string sent to os8_send_cmd(). check-in: 9c59dd5a2d user: tangent tags: trunk | |
| 08:36 | Quadrupled the size of the files that test-os8-send-file transfers to try and tickle bugs. No dice, at least on my iMac. check-in: 2cb6f721a5 user: tangent tags: trunk | |
| 08:31 | Merged comment fix in check-in: 3ae5eaf9e7 user: tangent tags: trunk | |
| 08:18 | Now that test-os8-send-file is transferring a single file reliably, reworked its main() to try a loop of 1000 send/get cycles. check-in: d192d2bcc0 user: tangent tags: trunk | |
| 08:01 | Removed ASCII 127 from the set of characters we send via tools/test-os8-send-file: it's clearly being interpreted, causing needless differences in the test output. check-in: 94811a17f0 user: tangent tags: trunk | |
| 07:54 | Removed double quotes from "attach ptp FILENAME" command sent by simh.os8_send_file(). SIMH doesn't let you quote a file name; implicit in that is that we're dependent on Python's temp file naming scheme not to give us a string with spaces or other problematic characters. check-in: a78fadf377 user: tangent tags: trunk | |
| 07:09 | Jim Tcl compatibility fix in recent auto.def changes. check-in: 7ad143c18b user: tangent tags: trunk | |
|
2017-11-30
| ||
| 16:16 | Added several things to "make clean". check-in: cc50ffe3d1 user: tangent tags: trunk | |
| 15:23 | Assorted fixes to allow the cc8 cross-compiler to build in Xcode: added "return 0" to functions without an explicit return value which don't end in a return statement in all code paths; renamed kill() to kill_line() to avoid conflicting with the POSIX syscall; and removed link(), which also conflicts with POSIX syscall set, but isn't actually used here. Also fixed line ending styles and re-removed the getchar() at the end of main(). check-in: 27e81d863c user: tangent tags: trunk | |
|
2017-11-28
| ||
| 14:16 | Added simh.os8_get_file() and tools/test-os8-send-file to test it and its inverse, simh.os8_send_file(). It isn't working yet, but it won't interfere with anything, and someone else may be able to finish it. check-in: 71ddf33004 user: tangent tags: trunk | |
|
2017-11-27
| ||
| 19:41 | A few typo fixes in previous check-in: c9b341cd23 user: tangent tags: trunk | |
| 19:36 | Reworked the "Government Health Warning" section of the CC8 README to split the limitations apart from the list of known-working features. Also spread the recently added "Nov 2018" (sic) stuff through the document rather than just have it tacked onto the end. check-in: be43971b09 user: tangent tags: trunk | |
| 17:07 | Healed Ian Schofield's inadvertent trunk fork. check-in: 41b8fa50af user: tangent tags: trunk | |
| 16:38 | Update of include file names. check-in: f12365e05c user: isysxp tags: trunk | |
| 16:04 | This update contains a correction for the expression parser and the addition of 2 operators ...% and ~. In addition, the spacing issue has been resolved to a degree. This is very difficult to test. Overall, the native compiler is for a bit of fun. If you want to run a serious c programme, try the hosted compiler (cc8) as this will compile the (native) compiler and library. check-in: 05b40e1d4c user: isysxp tags: trunk | |
| 07:21 | "make clean" no longer removes Makefiles generated from Makefile.ins. You now have to say "make distclean" to nuke those, along with all the other files that bring us back closer to a "fossil clean" state. check-in: 70dcfbc33d user: tangent tags: trunk | |
| 07:12 | Replaced hard-coded -DHAVE_DLOPEN=so CFLAGS option with one based on SH_SOEXT variable exported by autosetup, which allows SIMH to dynamically link to optional libraries such as libreadline on platforms that use a different dlopen(2) file name extension. Tested as not changing anything on Raspbian and as allowing command completion in SCP on macOS. Theoretically, it should allow use of DLLs on Windows, but this has not been tested; it probably only does so under Cygwin. check-in: 161b1c1b6b user: tangent tags: trunk | |
| 05:43 | Minor corrections of typos and formatting problems to docs amended in previous update. check-in: bc636c64e7 user: poetnerd tags: trunk | |
| 05:18 | After much testing I provide this new integration of MACREL: v2. Additionally I have updated os-patching.md to reflect current reality, and spun two sections from that doc into their own docs that cover macrel and the OS/8 V3D Device Extensions kit. MACREL v2 contains an update version of FUTIL that matches MACREL. That drove hand-integration of FUTIL patch 32.13.1M to get it working again under batch. The test program, RESORC, from the OS/8 V3D Device Extensions kit runs just fine when installed on SYS: but not anywhere else. I wailed on this problem for a while but could not find root cause. That version if RESORC is NOT installed. I'm just trying to understand what it is about that epoch of building that behaves in that unexpected way. check-in: c0e2d1ce56 user: poetnerd tags: trunk | |
| 04:46 | Regularized os8-* option help texts in auto.def and fixed tenses in verbs so they match the sense of the option's *inverse*, since that is how the help text is used in ./configure --help output. check-in: 09f9503302 user: tangent tags: trunk | |
| 04:39 | Hand integrated patch 41.5.1 M which is a source change to OVRDRV.MA. check-in: a0c3e3490f user: poetnerd tags: trunk | |
|
2017-11-25
| ||
| 20:22 | Implementing rename. check-in: 6f9077faef user: poetnerd tags: trunk | |
| 20:20 | FUTIL V8A hangs under batch, and so it breaks our auto-build process as soon as we copy it in. I have hand-applied patch 35.13.1M which fixes the problem and upgrades FUTIL to v8B. I have documented work in README.TX, and I'm going to rename this image al-5642c-ba, a version never shipped, but embodying a critical fix. check-in: 86934ed2b1 user: poetnerd tags: trunk | |
| 17:20 | Adding a hand-tooled MACREL v2 DECtape image. Contains FUTIL.SV version 8a that should replace v7D to match new Core Control Block format used by the new MACREL. When FUTIL is replaced, we need to re-test and probably disable patch FUTIL-31.21.2M. check-in: 1d6f691bbc user: poetnerd tags: trunk | |
|
2017-11-23
| ||
| 00:16 | Reworked simh.os8_send_file() so that it gets EOF from txt2ptp rather than inserting it into the simulator's stdin itself. check-in: 816c827edb user: tangent tags: trunk | |
| 00:15 | Another "PIP input from paper tape" documentation fix tracking the addition of Ctrl-Z at EOF in txt2ptp. check-in: 0c2b3b2035 user: tangent tags: trunk | |
| 00:02 | Added a missing "hit Escape" in another PIP example check-in: e3054b879f user: tangent tags: trunk | |
| 00:01 | Changed a recommendation of Ctrl-C to leave PIP to Escape. check-in: 4cd111528d user: tangent tags: trunk | |
|
2017-11-21
| ||
| 04:41 | Further documentation around the patch FRTS-51.3.3-O. It might be OK. Or it might be yet another "only works with the Devices Extension Kit" patch. check-in: c68ec2b288 user: poetnerd tags: trunk | |
| 03:30 | When making an ASCII paper tape, if the input file does not already end with ^Z, add it before emitting the leader/trailer. check-in: 6c4272cbbc user: poetnerd tags: trunk | |
| 02:02 | Pushed mkos8 componentization to next release; didn't get to it this weekened, and won't have time this next weekend. check-in: 6f4c6ebd39 user: tangent tags: trunk | |
|
2017-11-18
| ||
| 15:01 | Assorted improvements to src/cc8/examples/README.md check-in: 644ac70416 user: tangent tags: trunk | |
|
2017-11-16
| ||
| 20:16 | Thought up a workaround for the lack of toupper() and || in CC8 in the calc.c example. Also added a bunch of comments explaining why it is written the way it is. check-in: 784f8842cd user: tangent tags: trunk | |
| 20:01 | Removed the test for lowercase 'y' in the reworked CC8 example calc.c: CC8 doesn't yet understand the C || operator. None of my other attempts to work around this work, either. (cupper, restructuring as nested if/elseif/else tests, etc.) check-in: 156b2b69e9 user: tangent tags: trunk | |
| 19:51 | Removed a pointless "return 0" in pep001.c check-in: 1657a72a20 user: tangent tags: trunk | |
| 17:38 | The "SAVE SYS BUILD" did the wrong thing. It carefully saved the updated setup on SYS which had been changed to the RK05, and then, moments later, overwrote it with the original from the DECTAPE. So we do "SAVE DTA0 BUILD" to put our revised setup where it will then be copied along with everything else to the RK05 SYS. check-in: 40a3f17c58 user: poetnerd tags: trunk | |
| 07:59 | First attempt at pep001.c. It builds and runs correctly under Clang, but it fails under CC8. check-in: 176f2cabcd user: tangent tags: trunk | |
| 07:31 | Reordered the commands in cc8-tu56-update to collect all of the "send file" calls together. There is some kind of timing sensitivity in all of this which caused the previous method to truncate some of the files or even leave them empty. check-in: 308b654b57 user: tangent tags: trunk | |
| 06:42 | All of the users of class simh which were using "expect" regexes like '\.' to wait for an OS/8 prompt had to be changed to '\\.' because the string is double-interpreted, first by Python in a string formatting construct (printf(), kinda) and then again in a regex construct, so that without the doubled backslash it was just looking for ., meaning any character in RE-speak. Did the same for '\*' because that was reducing to *, meaning "zero or more of the previous", and without a "previous", it matched anything. This doesn't seem to fix anything; I found it trying to fix something else, which I hope to check in soon. check-in: f8d7dc89bf user: tangent tags: trunk | |
| 04:31 | Backed out Ian Schofield's improvements to src/cc8/os8/n8.c accidentally checked in as part of [d2f66e76b9]. Will check it back in on a branch, since we're waiting for Ian to check it in himself, both to ensure credit is given where due and so that he can describe what the change actually does in his own words. check-in: 51ecbfa05d user: tangent tags: trunk | |
| 03:25 | Documented the LF -> CR+LF handling in the header comment of ptp2txt.c and made other minor improvements to it. check-in: 0487e8ce1e user: tangent tags: trunk | |
|
2017-11-15
| ||
| 17:29 | Added Fetch LED item to the ChangeLog check-in: 69ae00927e user: tangent tags: trunk | |
| 17:25 | Turning off Fetch LED when in STOP or single-step state as we had been doing with Execute prior to this. Investigation of the PDP-8/I maintenance manual by Bill Cattey shows this to be correct. Closes [347ae454039]. check-in: 9ff851e459 user: tangent tags: trunk | |
| 16:41 | Clarified comment about the meaning of the optional "set remote telnet 3141" SIMH setup command in teco-pi-demo. check-in: 4ab764f2cf user: tangent tags: trunk | |
| 16:35 | Gave up on trying to fix the %lld/%zu issue in d8tape/main.c portably. Just going with the one that builds without warnings on the supported version of Raspbian, then turning off the warning so it doesn't appear on platforms that really want the other form. We're not seriously going to be stressing the local integer size anyway; the compiler is just being pedantic here. check-in: cd094984d4 user: tangent tags: trunk | |
| 04:00 | Subtle nomenclature changes to comments. Totally cosmetic, but slightly more accurate. check-in: ecb612b614 user: poetnerd tags: trunk | |
|
2017-11-14
| ||
| 18:24 | More ChangeLog.md updates check-in: f14a91f551 user: tangent tags: trunk | |
| 18:13 | Assorted improvements to the new lowercase I/O section of the ChangeLog check-in: b46a2c515c user: tangent tags: trunk | |
| 17:52 | Another attempt at fixing the printf() complaint in d8tape/main.c. This one should be more portable than the last. check-in: 750c63b0a1 user: tangent tags: trunk | |
| 09:34 | cc8-tu56-update tells class simh it's okay to run without front panel control. check-in: 9b34e6443c user: tangent tags: trunk | |
| 09:13 | Fixed a crash in one of mkos8's exception handlers. check-in: b6acd6291c user: tangent tags: trunk | |
| 09:12 | "mkos8 patch" no longer tries to apply patches to FORTRAN II and FORTRAN IV files if the corresponding --disable-os8-fortran-* options were given to the configure script. check-in: f4d3cf118a user: tangent tags: trunk | |
| 08:54 | Reworked the CC[0-2].SV copy command in cc8-tu56-update so that it skips CCL.SV, thus avoiding the hole in the resulting DECtape image. check-in: d2f66e76b9 user: tangent tags: trunk | |
| 08:36 | Rewrote the code in mkos8 emitting DEL foo.SV commands as part of --disable-fortran-ii processing so that only one command goes to the simulator, which runs faster and logs nicer. check-in: ff1e1c0545 user: tangent tags: trunk | |
| 08:27 | Small Python style tweak check-in: 0db3c5fc38 user: tangent tags: trunk | |
| 08:13 | Giving --disable-os8-fortran-ii now leaves out LOADER, LIBSET, and SABR. Because the OS/8 version of CC8 depends on LOADER and SABR, it also implicitly sets --disable-os8-cc8. check-in: 648ea9921c user: tangent tags: trunk | |
| 07:11 | Running cc8-tu56-update on a Pi 3 gives a different result than running it on a macOS box using the *.s files produced on the Pi 3. Not sure why, but of the two, the result on the Pi 3 is more trustworthy, so using that one. check-in: 7d73c6e8e5 user: tangent tags: trunk | |
| 06:54 | Removed the getchar() call near the end of main() in the cc8 cross-compiler. It blocks cc8-tu56-update, and we don't want to add a "send CR" bit to the script just to placate it. check-in: 96b6cef0c3 user: tangent tags: trunk | |
| 06:50 | Fixed COPY commands near the end of cc8-tu56-update: they weren't grabbing everything we thought they should. Also removed CCL.SV, inadvertently copied to the tape by the CC?.SV pattern. check-in: 0eaba4e478 user: tangent tags: trunk | |
| 06:33 | Initial version of tools/cc8-tu56-update complete, which updates media/os8/subsys/cc8.tu56 from the contents of src/cc8. The latter is now much smaller than before, since we no longer include the intermediate files: all *.SB and most *.RL. The only exception is LIBC.RL, which the linking step needs when building end-user programs. check-in: 4e67bbdfaf user: tangent tags: trunk | |
| 06:30 | Rewrote src/cc8/examples/calc.c for clarity. It could be better, but it doesn't build and run properly as-is, so I've stopped here, waiting for improvements to the C compiler. check-in: 832db28e22 user: tangent tags: trunk | |
| 06:29 | Fixed a bug in src/cc8/examples/fib.c which caused junk output check-in: 18ebdc34e4 user: tangent tags: trunk | |
| 06:28 | Several small tweaks to src/cc8/examples/README.md check-in: c6ea800ac1 user: tangent tags: trunk | |
| 06:26 | URL fix in previous check-in: 0981e043d7 user: tangent tags: trunk | |
| 06:25 | Many improvements to the new src/cc8/examples/README.md file. check-in: d454056d80 user: tangent tags: trunk | |
| 06:24 | Reverted the LF -> CRLF change in media/os8/init.tx.in, since autosetup just reverts it when creating init.tx. Instead, modified txt2ptp to turn LF input into CRLF output. I chose this instead of running a "dos2unix" type program on the files from simh.os8_send_file() because there is no standard for the name of such a program. Since our txt2ptp filter has to be involved anyway, we might as well give it the job. check-in: 6653eac20e user: tangent tags: trunk | |
| 06:06 | Reworked the stock init.tx.in file to use lowercase, now that it's allowed, and to have DOS line endings now that it's being copied in as-is, rather than filtered through mkos8, which implicitly converts the line ending style. check-in: 93f3beed61 user: tangent tags: trunk | |
| 06:01 | Rewrote simh.os8_send_file in terms of txt2ptp and SIMH paper tape attachment and read-in. It's faster, it works with much larger files, and it doesn't fold lowercase input. check-in: 69766295e7 user: tangent tags: trunk | |
| 04:35 | The top-level "run" target no longer depends on the build RK05 media because we need to be able to use it to run whatever's built, even if it's technically out of date. The thing we're working on now modifies the RK05 disk images, and we need to then be able to say "make run" to examine what it did without having the work overwritten. check-in: b4e684c40a user: tangent tags: trunk | |
| 03:49 | Moved another part of the "send file to OS/8" process from mkos8 to simh.os8_send_file(). check-in: 12dd946f7a user: tangent tags: trunk | |
| 02:14 | Collected the CC8 example and test programs into src/cc8/examples and wrote a more extensive README.md for them based on the old src/cc8/examples/README.md file. check-in: 9cf664a25e user: tangent tags: trunk | |
| 01:44 | Added another libc.h symlink check-in: c787c0ca0a user: tangent tags: trunk | |
| 00:17 | Factored the "send INIT.TX" step of mkos8 out as simh.os8_send_file() so we can reuse it in another script. check-in: b14acea3cd user: tangent tags: trunk | |
|
2017-11-13
| ||
| 23:28 | Fixed some code in the #include handling of CC8 which relied on undefined behavior. The replacement is actually simpler than the original, because the function using UB reduces to a no-op except on CP/M, which of course we don't care about. check-in: c548671ea0 user: tangent tags: trunk | |
| 23:18 | Added a redirect-to-top[ Makefile.in to src/cc8/os8, since you have to be cd'd into that directory to build the OS/8 compiler, so you want to be able to say "make" from there. check-in: a388b49124 user: tangent tags: trunk | |
| 23:12 | Added libc.h symlink to src/cc8/os8 directory pointing at ../include/libc.h since CC8 assumes a flat filesystem. check-in: 5fe720a649 user: tangent tags: trunk | |
|
2017-11-12
| ||
| 07:40 | Small refinements to test-mkos8 check-in: ce3acfd9bd user: tangent tags: trunk | |
| 07:22 | Added output reporting to test-mkos8. check-in: 3243380301 user: tangent tags: trunk | |
| 06:18 | Briefer output from test-mkos8. check-in: 97d4065304 user: tangent tags: trunk | |
| 04:43 | The method we were using for comparing lz4-compressed RK05 disk images in test-mkos8 only works if lz4 on the generating system and the test system give the same output for the same input, but this appears not to be the case between macOS 10.12 and Raspbian Stretch, likely due to improvements in the compression algorithm or compression defaults between the two versions in use. Therefore, instead of doing a binary comparison of the rklz files, we now decompress the rklz files and compare those outputs using a new tools/cmplz script which test-mkos8 calls. check-in: 8d51be13c7 user: tangent tags: trunk | |
| 04:13 | Still more test-mkos8 sanitize_log() RE fixes check-in: 5b9d507ff0 user: tangent tags: trunk | |
| 04:06 | More tweaks to regexes in test-mkos8 check-in: 33a1b89e49 user: tangent tags: trunk | |
| 03:34 | Fixed regexes in sanitize_log() in test-mkos8 and rewrote them as commented Extended REs. check-in: 42c31eadd0 user: tangent tags: trunk | |
| 01:41 | Improvements to sanitize_log() in test-mkos8. check-in: aa6e92da63 user: tangent tags: trunk | |
| 01:22 | Small adjustments to new test-mkos8 color output. check-in: f2bd9d0a7a user: tangent tags: trunk | |
| 01:14 | Colorized the test-mkos8 pass/fail result line so it's easier to see. check-in: 0ab8609fa6 user: tangent tags: trunk | |
| 00:35 | Merged trecent changes from test-mkos8-pinned into trunk. check-in: 5160d2a336 user: tangent tags: trunk | |
| 00:26 | mkos8 was reading init.tx from dirs.os8mi, which is in the source tree, not the build tree, so the recent change to test-mkos8 to write a neutral init.tx file was not doing anything useful. This removes the reason for this branch, so I'm closing it now. Closed-Leaf check-in: 5111887856 user: tangent tags: test-mkos8-pinned | |
| 00:20 | Assorted minor fixes to test-mkos8 check-in: 6377e01ab5 user: tangent tags: test-mkos8-pinned | |
|
2017-11-11
| ||
| 23:59 | test-mkos8 now overwrites init.tx with a neutral replacement so that tests made on one machine can be compared to those performed on another. check-in: 68de63f2c8 user: tangent tags: test-mkos8-pinned | |
| 06:21 | Merged in trunk changes. The original reason for this branch just disappeared, to be replaced by a new one: the test results can only be compared between two identical checkins because of the INIT.TX file, but that just means we now need to use this branch to pin a particular version in place. Thus, we might as well bring the recent trunk changes in and start again. With test-mkos8 -es, we don't need to regenerate the entire test corpus to learn useful things. check-in: 111c244601 user: tangent tags: test-mkos8-pinned | |
| 05:55 | test-mkos8 saves the mismatched rklz file if the test doesn't pass check-in: 2147ab88c0 user: tangent tags: test-mkos8-pinned | |
| 05:07 | Renamed test-mkos8 -s option to -1 and implemented -e and -s, which together allow Monte Carlo testing on a subset of the test corpus. This is on a branch because the current test corpus was generated before the change to 3-column DIRECT output in generated OS/8 disk images, which changes the disk contents and log contents. Thus, we need to do this development on a branch stemming from the last checkin before that change and any possible later ones, until we're ready to regenerate the test corpus. check-in: 47217f7552 user: tangent tags: test-mkos8-pinned | |
| 00:44 | Class simh now throws an exception when the GPIO is locked at simulator start time. If this happens, teco-pi-demo exits with an error and mkos8 just grouses a bit; the difference being, teco-pi-demo really needs the display and mkos8 really doesn't. Applying tag "test-mkos8-mark" to this release because a change a bit down the timeline breaks all of the test-mkos8 outputs so far, so we need to be able to return to this checkin symbolically until we're done using that test corpus. check-in: fe441b85b4 user: tangent tags: trunk | |
|
2017-11-01
| ||
| 06:46 | Including a comparison to our guess at the average IPS rate for a PDP-8/I in the teco-pi-demo -b output. check-in: 4c68402d93 user: tangent tags: trunk | |
| 05:58 | Updated hardcoded RPi B+ constant in the pidp8i.ips module based on actual test results. check-in: ab68fa4e64 user: tangent tags: trunk | |
| 05:04 | ChangeLog updates check-in: 9ec0b6f43f user: tangent tags: trunk | |
| 04:53 | Replaced the broken calculation for simh._kbd_delay with a new one based on the new pidp8i.ips module, potentially modified by teco-pi-demo -b. The previous version had a Python bug in it that caused no delay at all to be applied between OS/8 keystrokes, which probably means it would only work reliably on sufficiently fast host machines. This one doesn't change the per-test build time all that much, but it will add increasing amounts of keyboard delays on slow hosts, as is appropriate. check-in: 2211503b72 user: tangent tags: trunk | |
| 04:51 | More refinement to the INFILES and OUTFILES stuff in Makefile.in check-in: fe265b4e07 user: tangent tags: trunk | |
| 03:50 | Better solution to the problem of parallel make after touching *.in causing parallel reconfigures, one copy for each file in INFILES. check-in: 27766013ab user: tangent tags: trunk | |
| 03:04 | Added error diagnostics around making the rk05 copy to be patched. check-in: 2f669b050f user: poetnerd tags: trunk | |
| 02:48 | Fixed an error in Python namespacing introduced in [b80f275e25] check-in: 73fd0da382 user: tangent tags: trunk | |
| 02:45 | Python's File.write() apparently doesn't include a newline, though print() does. Grrrr. check-in: 40f1d966f5 user: tangent tags: trunk | |
| 02:43 | Comment improvement check-in: e2ece94c5d user: tangent tags: trunk | |
| 02:41 | Improved clarity of benchmarking results from teco-pi-demo -b and that of the underlying constants in the new pidp8i.ips module. check-in: 8b7f53c352 user: tangent tags: trunk | |
| 02:37 | Add error detection around the creation of the bootable OS/8 DECtape image. Sometimes the creation fails and blows something else out unexpectedly. check-in: 892c109447 user: poetnerd tags: trunk | |
| 02:26 | Added teco-pi-demo -b mode, which runs the demo for 10 seconds, then gets the IPS value from SIMH and appends it to a new Python module — pidp8i.ips — which initally gives a hardcoded value for a Pi 1, with the update giving the speed of the current hardware. check-in: 7a35b0cb16 user: tangent tags: trunk | |
| 02:14 | A short document providing help with our FORTRAN IV stuff. check-in: a4a06ba896 user: poetnerd tags: trunk | |
| 02:00 | Subsetted the INFILES list in Makefile.in to segregate the set that requires a mkos8 re-call from those that merely cause a reconfigure and build. As the INFILES list grew, more and more files irrelevant to mkos8 were included, increasing the chances of a pointless mkos8 build. check-in: c0987c9ee3 user: tangent tags: trunk | |
|
2017-10-31
| ||
| 15:23 | Updated TODO section in ChangeLog.md check-in: 1e64c48994 user: tangent tags: trunk | |
| 14:20 | test-mkos8 now builds and keeps os8v3d-patched.rk05 when not configuring with --disable-os8-patches. Previously, it was building and keeping the -bin disk regardless of this configuration option. check-in: e1a78b7930 user: tangent tags: trunk | |
| 04:22 | mkos8 contains in-line code to replace FORLIB.RL with the patched one on local.tu56. os8-patching.md documents this new state of affairs. check-in: 1fbbf7a508 user: poetnerd tags: trunk | |
| 03:19 | Added the 3 FORTRAN IV Source DECtapes. They came from Dave Gesswein's Archive, pdp8.net under misc_dectapes. The sources I've built seem reasonable. The DECtape part numbers match those for OS/8 V3D FORTRAN IV from the DEC Software Catalog. TODO: Write up what we have for FORTRAN IV in a doc file. check-in: a8ba68ef88 user: poetnerd tags: trunk | |
| 03:14 | Updated Fortran IV install: Move the tapes to the subsys directory. Tape #2 is not needed. It contains the source files for the FORLIB.RL run time library. TODO: Add in the source DECtape images to the repository. check-in: 16d14955b5 user: poetnerd tags: trunk | |
| 03:12 | Added files necessary for mkos8 to integrate patch FORLIB-51.10.1M.patch8. As per instructions in that patch, a new DLOG.RA has been made with one source line change. DLOG.RA was assembled with RALF into DLOG.RL. LIBRA was used to make an updated FORLIB.RL DLOG.RA and FORLIB.RL are on local.tu56. DLOG.RA in the patch directory is a redundant copy of that file. TODO: Add logic in mkos8 under patches to replace FORLIB.RL with the updated one if BOTH with-os8-fortran-iv AND with-os8-patches are set. check-in: 87aa1be986 user: poetnerd tags: trunk | |
| 02:34 | Discovered a typo in the patch, AND that some OCR mistakes had been missed. Both are fixed now. TODO: Codify this patch into something mkos8 can work with. check-in: 69ce06acaf user: poetnerd tags: trunk | |
|
2017-10-28
| ||
| 21:52 | Uncommented the OS/8 SQUISH commands at the end of the make_patch function in mkos8. check-in: 887632c998 user: tangent tags: trunk | |
| 13:57 | Added "*Can* You Release Yet?" section to the release process doc check-in: b7b312b222 user: tangent tags: trunk | |
|
2017-10-27
| ||
| 09:17 | With the final (?) bugs squished in the parallel version of test-mkos8, merging it into trunk. check-in: 4fc0e64aff user: tangent tags: trunk | |
| 09:16 | Fixed a path bug that caused test-mkos8 to write finished test results to the wrong directory. That in turn caused subsequent runs to re-do tests already completed because the results weren't where it expected to find them. Closed-Leaf check-in: b513069b4c user: tangent tags: parallel-test-mkos8 | |
| 09:15 | Added test-mkos8 -v flag, enabling several new messages that explain what it's about to do. check-in: dad526a6ff user: tangent tags: parallel-test-mkos8 | |
| 08:54 | Fixed the last few problems preventing test-mkos8 from running multiple builds in parallel. check-in: db37af582e user: tangent tags: parallel-test-mkos8 | |
| 08:47 | Though we hacked tools/version to make it print a dummy version string when running under tools/test-mkos8, it didn't exit afterward, so it continued to run the code that was causing the problem we added the workaround to avoid! check-in: 3213505810 user: tangent tags: parallel-test-mkos8 | |
| 08:41 | The bootable-*.tu56 copy of the OS/8 boot tape created by mkos8 to avoid modifying input media is now being created in dirs.os8mo instead of os8mi, since the latter is shared among the children of test-mkos8, which means that when one of them deletes their temporary copy of this file, the copy used by all of them gets deleted. The file, being volatile, must go into each child's build directory. check-in: 0e8e159cf5 user: tangent tags: parallel-test-mkos8 | |
| 08:08 | Merged in trunk changes check-in: ded363caa7 user: tangent tags: parallel-test-mkos8 | |
| 05:20 | Merged in trunk changes check-in: 16f6e36f5c user: tangent tags: parallel-test-mkos8 | |
| 01:49 | Merged in trunk changes check-in: 1d21cdd108 user: tangent tags: parallel-test-mkos8 | |
|
2017-10-25
| ||
| 23:18 | Merged in trunk changes check-in: 78467c8474 user: tangent tags: parallel-test-mkos8 | |
|
2017-10-19
| ||
| 21:25 | 44 Patches, all but one verified against source. The one un-verified one is for FRTS.SV, which is recommended. I just cant figure out how to build FRTS from source to verify it. mkos8 successfully applies them all. patch_list.txt names all patches, comments out non-recommended ones and says why. check-in: 1bec0d528f user: poetnerd tags: trunk | |
|
2017-10-18
| ||
| 20:41 | Reverted Unix -> DOS line ending change for src/*.in: autosetup blindly assumes Unix line endings and emits mixed line endings when *.in has DOS line endings as it passes input CRs through on lines it doesn't touch and fails to include a CR on the lines it rewrites. The symptom is that you get a confusing compiler error in gpio-common.c. check-in: 860fd189d5 user: tangent tags: trunk | |
|
2017-10-15
| ||
| 23:46 | Refined patcher. Successfully applies about 40 of the 56 patches. TODO: Verify the successful patches at source code level. Get the rest of the patches to apply. check-in: 7278e6aceb user: poetnerd tags: trunk | |
| 22:30 | Backed out [1e62781bcd0]: it added a guard for a condition that apparently cannot fail, so it's useless at best and may be masking the real problem now at worst. check-in: f396fbb736 user: tangent tags: trunk | |
| 22:18 | Whitespace tweaks in Makefile.in check-in: 26f03607d2 user: tangent tags: trunk | |
| 22:11 | The dependency of bin/*-focal.pt on txt2ptp wasn't expressed clearly enough, so whether they got built or not depended purely on whether make(1) decided to build ptp2txt and hard link it to txt2ptp before attempting it. check-in: 7f25f66583 user: tangent tags: trunk | |
| 22:06 | Suppressed the harmless warning about missing setcap on non-Linux hosts during "make install". check-in: b0790a69cb user: tangent tags: trunk | |
| 17:22 | ODT Patches now check old value and abort the patch if the actual contents do not match the expected contents. There is detection of "dont care" old contents as well. Added ability to comment out lines in patch_list.txt manifest file. TODO: 1. Add in a bunch of patches and try them out. 2. Add in support for other patches than ODT. 3. Make patching conditional in auto.defs. check-in: a5d1d41858 user: poetnerd tags: trunk | |
| 16:53 | Assorted fixes to parallel test-mkos8 check-in: 5797acc1f3 user: tangent tags: parallel-test-mkos8 | |
| 16:32 | Made the "get version string" mechanisms more durable in the face of multiple parallel instances, as now happens with parallel test-mkos8. check-in: c34a426eed user: tangent tags: parallel-test-mkos8 | |
| 15:49 | First pass at reworking mkos8-test to run $cores tests in parallel unless -s is given. Each test is built out-of-tree in test/tmp/$PID, so they don't interfere with each other. check-in: f5cad8d9f6 user: tangent tags: parallel-test-mkos8 | |
| 15:24 | Switched compression in test-mkos8 from lzma to lz4. check-in: 6addfa56e1 user: tangent tags: trunk | |