Created attachment 223575 [details] dps8m shar Archive DPS8M is a simh based simulator for the Honeywell h6180/dps8m Multics processor. One can run Multics MR12.5 and MR12.6 with it. The fastest start is to download Multics from here: https://multics-wiki.swenson.org/index.php/Main_Page https://s3.amazonaws.com/eswenson-multics/public/releases/MR12.6f/QuickStart_MR12.6f.zip Extract QuickStart_MR12.6f.zip, cd into it and execute "dps8 MR12.6f_boot.ini" Multics starts an you can login with telnet 127.0.0.1 6180.
Unless it's a static archive (this seems to be generated on the fly) please use the suggested way for Gitlab mentioned in the handbook. Is the current ICU license incompatible with dps8m's otherwise I'd recommend you to define the same? https://svnweb.freebsd.org/ports/head/devel/icu/Makefile?revision=558348&view=markup#l16
>Is the current ICU license incompatible with dps8m's otherwise I'd recommend you to define the same? I don't know, not a Lawyer, it is explicitly mentioned ICU 1.8.1 and later as you can read here: https://gitlab.com/dps8m/dps8m/-/blob/master/LICENSE Yes will change it to the "Gitlab-Way".
The dist-files are made "on the fly", and are not resolvable thru the commit-hash. Probably because Github to Gitlab migration, so for now i leave the shar as it is.
Hi, That sounds a bit odd, the hash listed here doesn't work? https://gitlab.com/dps8m/dps8m/-/tags/R2.0 --> https://gitlab.com/dps8m/dps8m/-/commit/2a56f38d5658458da870acee6d04e45c70b79ea5 Best regards, Daniel
Well try it, it's not found in the /archive directory.
Hi, Are you following the instructions properly? https://docs.freebsd.org/en/books/porters-handbook/book.html#makefile-master_sites-gitlab-description Do note that you do not specify MASTER_SITES Example: https://gitlab.com/ecodis/exhale/-/releases/v1.1.3 https://svnweb.freebsd.org/ports/head/audio/exhale/Makefile?revision=567863&view=markup Best regards, Daniel
>Are you following the instructions properly? Yes, both the simple and the "more" complete.
Created attachment 223823 [details] Patch for dps8m (WIP) Use USE_GITLAB framework
Fails to compile on FreeBSD 13.0-RC1 (arm64) ... LD dps8 ld: error: duplicate symbol: sim_vm_init >>> defined at dps8_sys.c:107 >>> ./dps8_sys.o:(sim_vm_init) >>> defined at scp.c:397 >>> scp.o:(.bss+0x120) in archive ../simh-master/simh.a clang: error: linker command failed with exit code 1 (use -v to see invocation) gmake[3]: *** [Makefile:304: dps8] Error 1 gmake[2]: *** [Makefile:344: locallibs_then_dps8] Error 2 gmake[2]: Leaving directory '/usr/ports/emulators/dps8m/work/dps8m-R2.0-2a56f38d5658458da870acee6d04e45c70b79ea5/src/dps8' gmake[1]: *** [Makefile:11: all] Error 2 gmake[1]: Leaving directory '/usr/ports/emulators/dps8m/work/dps8m-R2.0-2a56f38d5658458da870acee6d04e45c70b79ea5' ===> Compilation failed unexpectedly. Try to set MAKE_JOBS_UNSAFE=yes and rebuild before reporting the failure to the maintainer. *** Error code 1 Stop. make: stopped in /usr/ports/emulators/dps8m
A quick note regarding your patches, you shouldn't hardset compiler instead you should rely on the ports framework. Best regards, Daniel
Closing due to submitter timeout
I would like to keep it open, the new version of Multics 12.7 was released some weeks ago, and i just wait until the corresponding version (2.01) of dps8m is getting released. BTW the newest dps8m compiles now on FB13.
Looks like upstream got stuck at RC1? https://gitlab.com/dps8m/dps8m/-/tags
I meant RC2 sorry :-/
(In reply to Daniel Engberg from comment #14) It's a bit special, let me explain. First a new Multics Version is released (12.7), then the dps8m emulator gets optimized or extended for that release, but they are quite active i think: https://gitlab.com/dps8m/dps8m/-/commits/master 2.1 will be the version for 12.7, whereas 2.0 was for 12.6f (f is the patch-version, .7 is the feature-version)
Created attachment 229270 [details] Comilation on FB13 Fixed new Version -Removed patches (not needed anymore) -Fix Compilation on FB13 (-fcommon -pthread) -Added files (punch-card-reader and pdf-printer) -Added default options from makefile (-Wall -g3 -O3 -fno-strict-aliasing)
Hi Daniel, Could you test the new 'shar'?
Ping?
Created attachment 229966 [details] Patch for dps8m (WIP) v2 Compile tested on 13-STABLE, 12.2-RELEASE
Hi, Can you give this a try? I've adapted your patch to follow guidelines for ports and bumped it to latest commit upstream. I'm not sure about the DISTVERSION string but that's a minor issue that's easily to change prior to commit. I think we can stick with the same definition for ICU license as the icu port. Defined C/LDFLAGS are already set by upstream's Makefile so no need to have it in the ports Makefile.
I also got confirmation that the current style of DISTVERSION is the way to go :-)
(In reply to Daniel Engberg from comment #21) Hy Daniel, Sorry for the late response, it works great thanks for the minor cleanup's LGTM!
A commit in branch main references this bug: URL: https://cgit.FreeBSD.org/ports/commit/?id=e84b431b31bb3d65bb30934f2f6f42b3d80cc0f7 commit e84b431b31bb3d65bb30934f2f6f42b3d80cc0f7 Author: Gian-Simon Purkert <gspurki@gmail.com> AuthorDate: 2021-12-23 20:41:02 +0000 Commit: Daniel Engberg <diizzy@FreeBSD.org> CommitDate: 2021-12-23 21:20:59 +0000 emulators/dps8m: New port: 'SIMH' based simulator for the Honeywell h6180 / dps8m Multics processor PR: 254553 Approved by: mentors (implicit) emulators/Makefile | 1 + emulators/dps8m/Makefile (new) | 41 +++++++++++++++++++++++ emulators/dps8m/distinfo (new) | 3 ++ emulators/dps8m/files/patch-src_Makefile.mk (new) | 11 ++++++ emulators/dps8m/pkg-descr (new) | 9 +++++ 5 files changed, 65 insertions(+)
Committed, thanks!