From noreply at github.com Thu Apr 1 00:10:03 2021 From: noreply at github.com (Jay Krell) Date: Wed, 31 Mar 2021 15:10:03 -0700 Subject: [M3commit] [modula3/cm3] a0d56b: FreeBSD m3core: We have const Uerror.Max = 248 (#285) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: a0d56b5d311d35eec40ab1231d79c897027a1309 https://github.com/modula3/cm3/commit/a0d56b5d311d35eec40ab1231d79c897027a1309 Author: Jay Krell Date: 2021-03-31 (Wed, 31 Mar 2021) Changed paths: M m3-libs/m3core/src/unix/Common/Uconstants.c Log Message: ----------- FreeBSD m3core: We have const Uerror.Max = 248 (#285) and we verify that this is greater than all errno values. FreeBSD/12.2/amd64 has a few errnos in the 9900-9999 range. Skip those. The two duplicated users of Uerror.Max do not need an exact number, and growing their static arrays to 9999 is perhaps overkill. Granted, that code is never used, really. Having static arrays is perhaps not needed either. Having Uerror.Max is perhaps not needed either. The way the FreeBSD/libc++ code is stated, it could be additional errors, but this should be ok for now. We should really consider eliminating Uerror.Max, as a portability burden that provides near zero value. There are uses in the tree, they are not quite correct or needed. From noreply at github.com Thu Apr 1 00:17:38 2021 From: noreply at github.com (Jay Krell) Date: Wed, 31 Mar 2021 15:17:38 -0700 Subject: [M3commit] [modula3/cm3] 1a1fe6: m3core: Check for _LIBCPP_ERRNO_H instead of __cpl... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 1a1fe67110e9a246ab404aa75f8338bc342a46c5 https://github.com/modula3/cm3/commit/1a1fe67110e9a246ab404aa75f8338bc342a46c5 Author: Jay Krell Date: 2021-03-31 (Wed, 31 Mar 2021) Changed paths: M m3-libs/m3core/src/unix/Common/Uconstants.c Log Message: ----------- m3core: Check for _LIBCPP_ERRNO_H instead of __cplusplus && __clang__ __FreeBSD__. (#286) Either way is not great, but note that the following code makes it less bad, and perhaps suffices. i.e. checks for >9900 && <9999 From noreply at github.com Thu Apr 1 02:21:08 2021 From: noreply at github.com (Jay Krell) Date: Wed, 31 Mar 2021 17:21:08 -0700 Subject: [M3commit] [modula3/cm3] 495cd4: scripts: Remove mklib from Unix bootstraps. (#287) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 495cd4aeea74ef0e2f2ce81c7720ef01237b80f0 https://github.com/modula3/cm3/commit/495cd4aeea74ef0e2f2ce81c7720ef01237b80f0 Author: Jay Krell Date: 2021-03-31 (Wed, 31 Mar 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- scripts: Remove mklib from Unix bootstraps. (#287) It should work and is slightly useful, to quickly get a system that can cross to NT, but there is some problem with the Makefile, seen on Solaris and BSD and not super worth resolving for now. Eventually we should use cmake here. From noreply at github.com Thu Apr 1 03:57:05 2021 From: noreply at github.com (Jay Krell) Date: Wed, 31 Mar 2021 18:57:05 -0700 Subject: [M3commit] [modula3/cm3] c086ed: Scripts: Convert newlines to Unix format. (#288) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: c086ed5ef736ee63cdd20d2743d90883c4761731 https://github.com/modula3/cm3/commit/c086ed5ef736ee63cdd20d2743d90883c4761731 Author: Jay Krell Date: 2021-03-31 (Wed, 31 Mar 2021) Changed paths: M scripts/python/boot2.py M scripts/python/pylib.py Log Message: ----------- Scripts: Convert newlines to Unix format. (#288) Just a few laggards. From noreply at github.com Thu Apr 1 05:49:27 2021 From: noreply at github.com (Jay Krell) Date: Wed, 31 Mar 2021 20:49:27 -0700 Subject: [M3commit] [modula3/cm3] 492ec3: m3-sys: Consolidate strings for CG type names. (#290) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 492ec31b86da45fcf92fa70d28b3381fe08929ba https://github.com/modula3/cm3/commit/492ec31b86da45fcf92fa70d28b3381fe08929ba Author: Jay Krell Date: 2021-03-31 (Wed, 31 Mar 2021) Changed paths: M m3-sys/m3front/src/misc/CG.m3 M m3-sys/m3middle/src/M3CG_Rd.m3 M m3-sys/m3middle/src/M3CG_Wr.m3 M m3-sys/m3middle/src/Target.i3 M m3-sys/m3middle/src/Target.m3 Log Message: ----------- m3-sys: Consolidate strings for CG type names. (#290) PROCEDURE Target.TypeImage() => CONST Target.TypeNames[] CONST CG.TypeName[] => CONST Target.TypeNamesFixedWidth[] CONST Target.TypeNames[] => CONST Target.TypeNamesDotted[] Arguably should be VAR or PROCEDURE but the main point is they are all Target.TypeNames. Two were already CONST so stick with that. Arguably do not need three forms, but the point is consolidation not behavioral change, and exact printed strings are behavior. This is cleanup. Keep the duplication, but put it all near each other. From noreply at github.com Thu Apr 1 06:31:33 2021 From: noreply at github.com (Jay Krell) Date: Wed, 31 Mar 2021 21:31:33 -0700 Subject: [M3commit] [modula3/cm3] 8ff440: m3front m3c: More verbose errors in m3front and mo... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 8ff4408afb65879cedfcf2341f3508f028b79cf2 https://github.com/modula3/cm3/commit/8ff4408afb65879cedfcf2341f3508f028b79cf2 Author: Jay Krell Date: 2021-03-31 (Wed, 31 Mar 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 M m3-sys/m3front/src/misc/CG.m3 Log Message: ----------- m3front m3c: More verbose errors in m3front and more verbose (#291) tracing in m3c. This is a bit unfortunate, 'cause the tracing is really verbose. In time we might produce two files, verbose and not-verbose and compile the not-verbose one. Or save the binary form and write a dumper, which we really already have both. From noreply at github.com Fri Apr 2 03:44:47 2021 From: noreply at github.com (Jay Krell) Date: Thu, 01 Apr 2021 18:44:47 -0700 Subject: [M3commit] [modula3/cm3] ce15ea: m3c: Stronger typing. (#293) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: ce15ea52b7b7395b0176c1c2d9e99b6d3be3279e https://github.com/modula3/cm3/commit/ce15ea52b7b7395b0176c1c2d9e99b6d3be3279e Author: Jay Krell Date: 2021-04-01 (Thu, 01 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Stronger typing. (#293) This change does a much better job of typing, though there is more to do. Specifically before, pointers to anything tended to just be: typedef void* ADDRESS; typedef ADDRESS Twhatever; where whatever is the typeid. This included for example VAR parameters. Now we do a much better job of making Twhatever point to some other type. At least pointers to base types. Records might need completion still. From noreply at github.com Fri Apr 2 04:41:01 2021 From: noreply at github.com (Jay Krell) Date: Thu, 01 Apr 2021 19:41:01 -0700 Subject: [M3commit] [modula3/cm3] b7b3e8: m3c: More verbose load_indirect tracing. (#294) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: b7b3e821aaa710e82da02e68678a9f3fbad611f9 https://github.com/modula3/cm3/commit/b7b3e821aaa710e82da02e68678a9f3fbad611f9 Author: Jay Krell Date: 2021-04-01 (Thu, 01 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: More verbose load_indirect tracing. (#294) From noreply at github.com Fri Apr 2 07:21:09 2021 From: noreply at github.com (Jay Krell) Date: Thu, 01 Apr 2021 22:21:09 -0700 Subject: [M3commit] [modula3/cm3] c37034: m3c: More verbose tracing in store. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: c370349e3e647e63f18f83058da2a3f7b4f57366 https://github.com/modula3/cm3/commit/c370349e3e647e63f18f83058da2a3f7b4f57366 Author: JayKrell Date: 2021-04-01 (Thu, 01 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: More verbose tracing in store. From noreply at github.com Fri Apr 2 07:52:13 2021 From: noreply at github.com (Jay Krell) Date: Thu, 01 Apr 2021 22:52:13 -0700 Subject: [M3commit] [modula3/cm3] 2ef219: m3c:Fix calling conventions. It was very little code. Message-ID: Branch: refs/heads/working-32bit-c-backend Home: https://github.com/modula3/cm3 Commit: 2ef219a5b8ec20e01d39cb652d8f4e791a61da10 https://github.com/modula3/cm3/commit/2ef219a5b8ec20e01d39cb652d8f4e791a61da10 Author: JayKrell Date: 2021-04-01 (Thu, 01 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 M m3-sys/m3middle/src/Target.m3 Log Message: ----------- m3c:Fix calling conventions. It was very little code. This suppports I386_NT and hopefully no other platorm has multiple calling conventions. Commit: 2a016fe8d472ed699a7073fdbb060c572f8dca27 https://github.com/modula3/cm3/commit/2a016fe8d472ed699a7073fdbb060c572f8dca27 Author: JayKrell Date: 2021-04-01 (Thu, 01 Apr 2021) Changed paths: M m3-sys/m3middle/src/Target.m3 Log Message: ----------- m3middle: For C backend only, make Allow_packed_byte_aligned constant false. This does not affect m3cc, llvm, or integrated backend. Though that isn't a bad idea. Commit: d0c25b33d392646557e8d42c26ea0dd0e8be4f97 https://github.com/modula3/cm3/commit/d0c25b33d392646557e8d42c26ea0dd0e8be4f97 Author: Jay Krell Date: 2021-04-01 (Thu, 01 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: More verbose load_indirect tracing. (#294) Compare: https://github.com/modula3/cm3/compare/2ef219a5b8ec%5E...d0c25b33d392 From noreply at github.com Fri Apr 2 08:24:25 2021 From: noreply at github.com (Jay Krell) Date: Thu, 01 Apr 2021 23:24:25 -0700 Subject: [M3commit] [modula3/cm3] 6ed276: m3c: Somewhat synchronize tracing between working-... Message-ID: Branch: refs/heads/working-32bit-c-backend Home: https://github.com/modula3/cm3 Commit: 6ed2769f2e8b7af1493f3db9a57cb7d2e978dc54 https://github.com/modula3/cm3/commit/6ed2769f2e8b7af1493f3db9a57cb7d2e978dc54 Author: JayKrell Date: 2021-04-01 (Thu, 01 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Somewhat synchronize tracing between working-32bit-c-backend and master Commit: af6d2446e04cb0b67d459842d6263b3f92d633a8 https://github.com/modula3/cm3/commit/af6d2446e04cb0b67d459842d6263b3f92d633a8 Author: JayKrell Date: 2021-04-01 (Thu, 01 Apr 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- scripts: Remove C from end of build_dir. Commit: 878b6ae28000c0ce9b961c990b214fcf33859f89 https://github.com/modula3/cm3/commit/878b6ae28000c0ce9b961c990b214fcf33859f89 Author: JayKrell Date: 2021-04-01 (Thu, 01 Apr 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- scripts: add -DTARGET to cm3 invocations. Compare: https://github.com/modula3/cm3/compare/d0c25b33d392...878b6ae28000 From noreply at github.com Fri Apr 2 09:02:58 2021 From: noreply at github.com (Jay Krell) Date: Fri, 02 Apr 2021 00:02:58 -0700 Subject: [M3commit] [modula3/cm3] 53ddb0: scripts: Remove C from end of build_dir. (#295) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 53ddb08be8d6212c864205e4dd6efcb5aa75f5df https://github.com/modula3/cm3/commit/53ddb08be8d6212c864205e4dd6efcb5aa75f5df Author: Jay Krell Date: 2021-04-02 (Fri, 02 Apr 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- scripts: Remove C from end of build_dir. (#295) Yes it solves a problem, but it is too confusing to track when it is appended. The point is that C backend is not always call-compatible with gcc backend, specifically because of how the static link is passed. gcc backend can pass it in an ABI-defined extra register. C backend passes it as last parameter. We might be able to put an attribute on it to unify, in gcc/clang. Maybe declspec(thread) on NT. Gross but quite efficient really. Maybe some assembly for NT/x86. But probably nothing ever really. Eventually, we should have: /src/m3/libs/m3core/foo.m3 /src/m3/libs/libm3/bar.m3 /obj/m3/libs/m3core/foo.o /obj/m3/libs/libm3/bar.o where /obj/m3 is entirely user-specified and can contain arch-os. i.e. out of tree builds like autotools and cmake at the multi-package level, not individual packages mkdir -p /obj/m3 cd /obj/m3 cmake /src/cm3/min -Dtarget=arch ninja ninja install rm -rf * cmake /src/cm3/rest -Dtarget=arch ninja ninja install should be the goal. Having merely each package build out of its own source tree but not the overall source tree is not satisfying. In this model, where output root is generated less often, it is easier to manage this, if we really ever want to. From noreply at github.com Fri Apr 2 09:06:03 2021 From: noreply at github.com (Jay Krell) Date: Fri, 02 Apr 2021 00:06:03 -0700 Subject: [M3commit] [modula3/cm3] bb3188: scripts: add -DTARGET to cm3 invocations. (#296) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: bb31888c76dc294795578a1ecc3fb053feddc831 https://github.com/modula3/cm3/commit/bb31888c76dc294795578a1ecc3fb053feddc831 Author: Jay Krell Date: 2021-04-02 (Fri, 02 Apr 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- scripts: add -DTARGET to cm3 invocations. (#296) From noreply at github.com Fri Apr 2 18:23:38 2021 From: noreply at github.com (Jay Krell) Date: Fri, 02 Apr 2021 09:23:38 -0700 Subject: [M3commit] [modula3/cm3] 5069e5: cm3 version: add __cdecl and space (#297) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 5069e5e60ef9f6b2a79074b9dc619c6829a5d49f https://github.com/modula3/cm3/commit/5069e5e60ef9f6b2a79074b9dc619c6829a5d49f Author: Jay Krell Date: 2021-04-02 (Fri, 02 Apr 2021) Changed paths: M m3-sys/cm3/src/Version.c Log Message: ----------- cm3 version: add __cdecl and space (#297) From noreply at github.com Sat Apr 3 01:34:15 2021 From: noreply at github.com (Jay Krell) Date: Fri, 02 Apr 2021 16:34:15 -0700 Subject: [M3commit] [modula3/cm3] baccd4: m3c:A little less verbose. (#298) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: baccd43bed82f875be6127f130ed55760b4d8b59 https://github.com/modula3/cm3/commit/baccd43bed82f875be6127f130ed55760b4d8b59 Author: Jay Krell Date: 2021-04-02 (Fri, 02 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c:A little less verbose. (#298) From noreply at github.com Sat Apr 3 06:48:13 2021 From: noreply at github.com (Jay Krell) Date: Fri, 02 Apr 2021 21:48:13 -0700 Subject: [M3commit] [modula3/cm3] 7aefd9: m3c: Repair a little after baccd43bed82f875be6127f... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7aefd94ed41b090f596cee38326396e249444e47 https://github.com/modula3/cm3/commit/7aefd94ed41b090f596cee38326396e249444e47 Author: Jay Krell Date: 2021-04-02 (Fri, 02 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Repair a little after baccd43bed82f875be6127f130ed55760b4d8b59 (#299) From noreply at github.com Sat Apr 3 23:15:59 2021 From: noreply at github.com (Jay Krell) Date: Sat, 03 Apr 2021 14:15:59 -0700 Subject: [M3commit] [modula3/cm3] 9d1ea0: version:Add git info. (#300) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 9d1ea0157117e243fe345833aa2f346e20f6e186 https://github.com/modula3/cm3/commit/9d1ea0157117e243fe345833aa2f346e20f6e186 Author: Jay Krell Date: 2021-04-03 (Sat, 03 Apr 2021) Changed paths: M m3-sys/cm3/src/Makefile.m3 M m3-sys/cm3/src/version.quake Log Message: ----------- version:Add git info. (#300) Like this: C:\s\cm3.2\m3-sys\cm3>cm3 && AMD64_NT\cm3.exe --version --- building in AMD64_NT --- ignoring ..\src\m3overrides **** PARALLEL BACK-END BUILD, M3_PARALLEL_BACK = 20 Critical Mass Modula-3 version d5.11.0 GitInfo: https://github.com/jaykrell/cm3/commit/fb4448d43a0283855fe362d1206c79726966b19a https://github.com/jaykrell/cm3/commits/vd remote:https://github.com/jaykrell/cm3 revision:fb4448d43a0283855fe362d1206c79726966b19a branch:vd last updated: 2021-02-24 compiled: 2021-04- 3 13:57:27 configuration: C:\cm3\bin\cm3.cfg host: AMD64_NT target: AMD64_NT Several notes: - It can be changed. - If git fails, it prints "unknown", like: Critical Mass Modula-3 version d5.11.0 GitInfo: unknown last updated: 2021-02-24 compiled: 2021-04- 3 13:57:27 configuration: C:\cm3\bin\cm3.cfg host: AMD64_NT target: AMD64_NT - The url does not necessarily work. In this case the commit is only local. That is always a problem with git, being distributed. A hash can feed into a binary, the binary distributed and the hash not. - I didn't know what to call the first line. It seems most important. - Nor the second line. - Hash alone is useless without a repository, so this tries to be more complete. - Again, it can be changed. - You can see above "compiled:" has a bug, space vs. 0. - This will not work, as written, with the eventual goal of out of tree builds. - The code should be moved/copied to .cmd/.py/.sh and written out somewhere, like next to version.quake as gitversion.quake, but then ultimately into the output tree as part of a configure step. Maybe. That isn't up to date enough. So I guess an out of tree build should still cd over to the source tree. When git info is available, we no longer recompile and relink for every build of cm3. Only if the git info or other inputs changes. Not time alone. From noreply at github.com Sun Apr 4 03:24:14 2021 From: noreply at github.com (Jay Krell) Date: Sat, 03 Apr 2021 18:24:14 -0700 Subject: [M3commit] [modula3/cm3] 6b301e: NT config: Remove listing file code. (#301) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 6b301e6bf50a0b36cca1a7c6432dced46975fbba https://github.com/modula3/cm3/commit/6b301e6bf50a0b36cca1a7c6432dced46975fbba Author: Jay Krell Date: 2021-04-03 (Sat, 03 Apr 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/NT.common M m3-sys/cminstall/src/config-no-install/cm3cfg.common Log Message: ----------- NT config: Remove listing file code. (#301) It seems like too much for too little. And it hides warnings from C++ compiler that I would like to see and maybe deal with. Hopefully tests can be usefully run without this. Perhaps by having expected output appear, in order, but allow extra interspersed lines. Or perhaps keep all of this, perhaps.btt As I recall, I wrote all this but nobody cared much either way. Reduce overall build system code? From noreply at github.com Sun Apr 4 03:29:18 2021 From: noreply at github.com (Jay Krell) Date: Sat, 03 Apr 2021 18:29:18 -0700 Subject: [M3commit] [modula3/cm3] 42b8f6: m3c:Suppress warnings for Sun CC (C++) the same as... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 42b8f69aececff8023bf2e20cb095c1ab66ce7c4 https://github.com/modula3/cm3/commit/42b8f69aececff8023bf2e20cb095c1ab66ce7c4 Author: Jay Krell Date: 2021-04-03 (Sat, 03 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c:Suppress warnings for Sun CC (C++) the same as Sun cc (C). (#302) From noreply at github.com Sun Apr 4 04:34:24 2021 From: noreply at github.com (Jay Krell) Date: Sat, 03 Apr 2021 19:34:24 -0700 Subject: [M3commit] [modula3/cm3] ecb6b0: NT config: Leave stdout and stderr separate. (#303) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: ecb6b0858798ce85c8086373f4c60175db4fde4a https://github.com/modula3/cm3/commit/ecb6b0858798ce85c8086373f4c60175db4fde4a Author: Jay Krell Date: 2021-04-03 (Sat, 03 Apr 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/NT.common Log Message: ----------- NT config: Leave stdout and stderr separate. (#303) This is really probably from the listing file support (https://github.com/modula3/cm3/commit/6b301e6bf50a0b36cca1a7c6432dced46975fbba), so they'd both go to the listing file. From noreply at github.com Sun Apr 4 04:36:27 2021 From: noreply at github.com (Jay Krell) Date: Sat, 03 Apr 2021 19:36:27 -0700 Subject: [M3commit] [modula3/cm3] a677be: config: Normalize to Unix newlines. (#304) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: a677be1440f2455e3200b6c2469aa81327d85be2 https://github.com/modula3/cm3/commit/a677be1440f2455e3200b6c2469aa81327d85be2 Author: Jay Krell Date: 2021-04-03 (Sat, 03 Apr 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/NT.common M m3-sys/cminstall/src/config-no-install/RISCV64.common Log Message: ----------- config: Normalize to Unix newlines. (#304) There were only two. From noreply at github.com Sun Apr 4 05:55:42 2021 From: noreply at github.com (Jay Krell) Date: Sat, 03 Apr 2021 20:55:42 -0700 Subject: [M3commit] [modula3/cm3] d12ac5: Convert some to Unix newlines. (#305) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: d12ac5b251ab2d2da83288f436116cf15a7168b7 https://github.com/modula3/cm3/commit/d12ac5b251ab2d2da83288f436116cf15a7168b7 Author: Jay Krell Date: 2021-04-03 (Sat, 03 Apr 2021) Changed paths: M m3-libs/m3core/src/win32/WinConstants.c M m3-libs/m3core/src/win32/m3makefile Log Message: ----------- Convert some to Unix newlines. (#305) From noreply at github.com Sun Apr 4 08:12:16 2021 From: noreply at github.com (Jay Krell) Date: Sat, 03 Apr 2021 23:12:16 -0700 Subject: [M3commit] [modula3/cm3] 5c91cd: config: Do not append c to BUILD_DIR for C backend... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 5c91cdaf16ee71b5a7df5c36a97cfffc6669e675 https://github.com/modula3/cm3/commit/5c91cdaf16ee71b5a7df5c36a97cfffc6669e675 Author: Jay Krell Date: 2021-04-03 (Sat, 03 Apr 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/cm3cfg.common Log Message: ----------- config: Do not append c to BUILD_DIR for C backend. (#306) Like earlier, but I missed this spot. It serves a useful purpose, but is confusing and eventually eventually the output directory will just be the current directory, like for normal cmake/automake out of tree builds. From noreply at github.com Sun Apr 4 18:30:45 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 09:30:45 -0700 Subject: [M3commit] [modula3/cm3] 4f0232: Convert some newlines to Unix. (#307) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 4f0232fd7677d79f668d2128a15b4c3b797f50f3 https://github.com/modula3/cm3/commit/4f0232fd7677d79f668d2128a15b4c3b797f50f3 Author: Jay Krell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M m3-libs/m3core/src/atomic/m3makefile M m3-libs/m3core/src/coroutine/m3makefile M m3-libs/m3core/src/float/Common/no_grisu/m3makefile M m3-libs/m3core/src/m3makefile M m3-libs/m3core/src/runtime/AMD64_NT/m3makefile M m3-libs/m3core/src/runtime/I386_NT/m3makefile M m3-sys/m3back/src/m3makefile Log Message: ----------- Convert some newlines to Unix. (#307) From noreply at github.com Sun Apr 4 19:06:58 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 10:06:58 -0700 Subject: [M3commit] [modula3/cm3] 7e908c: scripts: Switch the bootstraps generally to C++ co... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7e908c3d1dc5f0eb3d0e349075bcb7171099561a https://github.com/modula3/cm3/commit/7e908c3d1dc5f0eb3d0e349075bcb7171099561a Author: Jay Krell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M scripts/python/c_compiler M scripts/python/pylib.py Log Message: ----------- scripts: Switch the bootstraps generally to C++ compiler driver. (#308) i.e. g++, c++, C++ Favor c++ over CC to be case insensitive but Solaris installs only CC in places. Not all systems tested, e.g. VMS and OSF1. From noreply at github.com Sun Apr 4 19:11:28 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 10:11:28 -0700 Subject: [M3commit] [modula3/cm3] d014d0: Bootstrap: -fPIC not needed 'for now' (only builds... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: d014d023600a59711d98a6210c3ee5d44ef84b8c https://github.com/modula3/cm3/commit/d014d023600a59711d98a6210c3ee5d44ef84b8c Author: Jay Krell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- Bootstrap: -fPIC not needed 'for now' (only builds an executable, no shared libraries). (#309) From noreply at github.com Sun Apr 4 19:19:15 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 10:19:15 -0700 Subject: [M3commit] [modula3/cm3] d499f2: Bootstrap: Add -pthread to most compiler/linker co... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: d499f21b036539188a116ee58d213a7de6d195de https://github.com/modula3/cm3/commit/d499f21b036539188a116ee58d213a7de6d195de Author: Jay Krell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- Bootstrap: Add -pthread to most compiler/linker commands. (#310) From noreply at github.com Sun Apr 4 19:29:31 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 10:29:31 -0700 Subject: [M3commit] [modula3/cm3] a08bee: Comment out redundant code. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: a08beea4fde58df111096213a80d29f0e6807818 https://github.com/modula3/cm3/commit/a08beea4fde58df111096213a80d29f0e6807818 Author: JayKrell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- Comment out redundant code. Commit: 42a53c5514ae5492048496123536b7df584b0762 https://github.com/modula3/cm3/commit/42a53c5514ae5492048496123536b7df584b0762 Author: JayKrell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- Bootstrap: Remove the -o and -Fo./ flags as we are just compiling foo.c to foo.o in current directory, it is default. Compare: https://github.com/modula3/cm3/compare/d499f21b0365...42a53c5514ae From noreply at github.com Sun Apr 4 21:45:55 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 12:45:55 -0700 Subject: [M3commit] [modula3/cm3] e1248a: version.c: Fix after 5069e5e60ef9f6b2a79074b9dc619... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: e1248a6c40d4fcd08f1f3cdc04d4183e53920bd1 https://github.com/modula3/cm3/commit/e1248a6c40d4fcd08f1f3cdc04d4183e53920bd1 Author: JayKrell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M m3-sys/cm3/src/Version.c Log Message: ----------- version.c: Fix after 5069e5e60ef9f6b2a79074b9dc619c6829a5d49f From noreply at github.com Sun Apr 4 21:51:33 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 12:51:33 -0700 Subject: [M3commit] [modula3/cm3] 6e7cec: m3core: Fix warning about defining X again. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 6e7cec676601eb90b4ef0363e69f86f4838f566e https://github.com/modula3/cm3/commit/6e7cec676601eb90b4ef0363e69f86f4838f566e Author: JayKrell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M m3-libs/m3core/src/unix/Common/Uconstants.c Log Message: ----------- m3core: Fix warning about defining X again. From noreply at github.com Sun Apr 4 22:04:53 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 13:04:53 -0700 Subject: [M3commit] [modula3/cm3] e45955: NT config/bootstrap: Compile with -Gy so linker ca... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: e459556ec63e2fb8441c49a8b947ea3e8909ea8d https://github.com/modula3/cm3/commit/e459556ec63e2fb8441c49a8b947ea3e8909ea8d Author: JayKrell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/NT.common M scripts/python/pylib.py Log Message: ----------- NT config/bootstrap: Compile with -Gy so linker can remove unused functions. -Gy is ancient and good. Newer toolsets also have -Gw for unused data. From noreply at github.com Mon Apr 5 04:04:09 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 19:04:09 -0700 Subject: [M3commit] [modula3/cm3] cb5526: main: Add __cdecl and tweak newlines. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: cb55268ab98ef26a374a3b06c3f286d0a63b4bba https://github.com/modula3/cm3/commit/cb55268ab98ef26a374a3b06c3f286d0a63b4bba Author: JayKrell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M m3-sys/m3linker/src/MxGen.m3 Log Message: ----------- main: Add __cdecl and tweak newlines. From noreply at github.com Mon Apr 5 04:18:34 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 19:18:34 -0700 Subject: [M3commit] [modula3/cm3] 7a07c7: main: Add __cdecl and tweak newlines. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7a07c78b863ab89e821c552c5f7cf25e8680ecf5 https://github.com/modula3/cm3/commit/7a07c78b863ab89e821c552c5f7cf25e8680ecf5 Author: JayKrell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M m3-sys/m3linker/src/MxGen.m3 Log Message: ----------- main: Add __cdecl and tweak newlines. From noreply at github.com Mon Apr 5 07:08:57 2021 From: noreply at github.com (peter mckinna) Date: Sun, 04 Apr 2021 22:08:57 -0700 Subject: [M3commit] [modula3/cm3] 003741: update m3gdb to compile on debian Message-ID: Branch: refs/heads/m3gdb Home: https://github.com/modula3/cm3 Commit: 0037410c6a682277c93c02b1f6cbbe53e8d47f89 https://github.com/modula3/cm3/commit/0037410c6a682277c93c02b1f6cbbe53e8d47f89 Author: peter mckinna Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-sys/m3gdb/gdb/configure M m3-sys/m3gdb/gdb/gdb/amd64-linux-nat.c M m3-sys/m3gdb/gdb/gdb/config/nm-linux.h M m3-sys/m3gdb/gdb/gdb/exceptions.h M m3-sys/m3gdb/gdb/gdb/gdbserver/gdbreplay.c M m3-sys/m3gdb/gdb/gdb/gdbserver/linux-low.c M m3-sys/m3gdb/gdb/gdb/gdbserver/linux-x86-64-low.c M m3-sys/m3gdb/gdb/gdb/gdbserver/proc-service.c M m3-sys/m3gdb/gdb/gdb/gdbserver/remote-utils.c M m3-sys/m3gdb/gdb/gdb/gdbserver/utils.c M m3-sys/m3gdb/gdb/gdb/linux-thread-db.c M m3-sys/m3gdb/gdb/gdb/proc-service.c Log Message: ----------- update m3gdb to compile on debian From noreply at github.com Mon Apr 5 08:01:08 2021 From: noreply at github.com (Jay Krell) Date: Sun, 04 Apr 2021 23:01:08 -0700 Subject: [M3commit] [modula3/cm3] 7cb2fa: m3gdb: Some reading of Dwarf symbols so that g++ -... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7cb2fa6127a8929265b17c6d7e6302830a93d30d https://github.com/modula3/cm3/commit/7cb2fa6127a8929265b17c6d7e6302830a93d30d Author: JayKrell Date: 2021-04-04 (Sun, 04 Apr 2021) Changed paths: M m3-sys/m3gdb/gdb/gdb/dwarf2read.c Log Message: ----------- m3gdb: Some reading of Dwarf symbols so that g++ -g _m3main.c works. It is not quite right. We do see small null and non-null pointers and just ignore them. "Small" enough? This is because we have such an old fork of gdb it does not handle current Dwarf. It would be great to update or unfork. From noreply at github.com Mon Apr 5 13:43:03 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 04:43:03 -0700 Subject: [M3commit] [modula3/cm3] bc53ad: Solaris: Sometimes mmap/mprotect/munmap expect cha... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: bc53ad5be35a90364af1a55247862bc329f49877 https://github.com/modula3/cm3/commit/bc53ad5be35a90364af1a55247862bc329f49877 Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/unix/Common/Umman.c Log Message: ----------- Solaris: Sometimes mmap/mprotect/munmap expect char* not void* and it fails to compile. It happened to me just now. From noreply at github.com Mon Apr 5 13:54:36 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 04:54:36 -0700 Subject: [M3commit] [modula3/cm3] 57d774: scripts:fix boot1.py to Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 57d77493395a55494a8713b6b63f208b0bb4106b https://github.com/modula3/cm3/commit/57d77493395a55494a8713b6b63f208b0bb4106b Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- scripts:fix boot1.py to From noreply at github.com Mon Apr 5 21:52:15 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 12:52:15 -0700 Subject: [M3commit] [modula3/cm3] 642d21: Revert "m3c:Suppress warnings for Sun CC (C++) the... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 642d21a02a5c40d2f37d0cb450644622eb91e960 https://github.com/modula3/cm3/commit/642d21a02a5c40d2f37d0cb450644622eb91e960 Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- Revert "m3c:Suppress warnings for Sun CC (C++) the same as Sun cc (C). (#302)" This reverts commit 42b8f69aececff8023bf2e20cb095c1ab66ce7c4. This only adds warnings, not reduces them. From noreply at github.com Tue Apr 6 04:02:25 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 19:02:25 -0700 Subject: [M3commit] [modula3/cm3] 7fc2bb: Convert to Unix newlines. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7fc2bb26c32908c5e042766d56a7fcc16755c31d https://github.com/modula3/cm3/commit/7fc2bb26c32908c5e042766d56a7fcc16755c31d Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-comm/sharedobj/src/sharedobj.tmpl M m3-scheme/modula3scheme/src/m3makefile M m3-scheme/mscheme-interactive/src/m3makefile M m3-scheme/mscheme-interactive_r/src/m3makefile M m3-scheme/mscheme/src/m3makefile M m3-scheme/mscheme/src/m3overrides M m3-scheme/schemereadline/src/m3makefile M m3-sys/cm3/src/Version.c M m3-sys/m3back/src/M3C.m3 M m3-sys/m3middle/src/CoffTime.i3 M m3-sys/m3quake/src/MxConfigC.i3 M m3-sys/windowsResources/src/m3overrides Log Message: ----------- Convert to Unix newlines. Very much content visited, very little to change. From noreply at github.com Tue Apr 6 04:09:00 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 19:09:00 -0700 Subject: [M3commit] [modula3/cm3] 96ca64: MxConfig: Replace HOST_WORD_SIZE with BITSIZE(INTE... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 96ca64bc99ba113c52ba407e6802d4faea5c3c3d https://github.com/modula3/cm3/commit/96ca64bc99ba113c52ba407e6802d4faea5c3c3d Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-sys/cm3/src/Main.m3 M m3-sys/m3quake/src/MxConfig.i3 M m3-sys/m3quake/src/MxConfig.m3 M m3-sys/m3quake/src/MxConfigC.c M m3-sys/m3quake/src/MxConfigC.i3 Log Message: ----------- MxConfig: Replace HOST_WORD_SIZE with BITSIZE(INTEGER) and HOST_WORD_SIZE_TEXT with Fmt.Int() thereof. There are vanishingly few users (one). This is arguably a problem, but part of a much larger one. From noreply at github.com Tue Apr 6 04:43:09 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 19:43:09 -0700 Subject: [M3commit] [modula3/cm3] f73614: mklib: Skip exporting what we also skip in .def file. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: f736147547f03d51f8c423b7de7da3af6b667913 https://github.com/modula3/cm3/commit/f736147547f03d51f8c423b7de7da3af6b667913 Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-sys/mklib/src/Main.m3 Log Message: ----------- mklib: Skip exporting what we also skip in .def file. Next step is to stop writing import libraries here at all. This addresses warnings seen, since listing file code removed. And try to write the .def file from cm3 instead of mklib. From noreply at github.com Tue Apr 6 06:28:20 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 21:28:20 -0700 Subject: [M3commit] [modula3/cm3] e7edbd: m3core: Refactor GetPC into a GET_PC and usually f... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: e7edbd782b46e07f086461cad336c0ba6c23abbb https://github.com/modula3/cm3/commit/e7edbd782b46e07f086461cad336c0ba6c23abbb Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-libs/m3core/src/runtime/POSIX/RTSignalC.c Log Message: ----------- m3core: Refactor GetPC into a GET_PC and usually function GetPC. The reason for this is speculation that we should compute HOST amidst the same code. If we don't just use uname. A full static host is fairly tempting, except, that on Debian uname.machine factors in the endian vagaries. Which I don't want to figure out, or worse, name. Therefore HOST should be computed statically in many cases, but not all. i.e. when GetPC had good clear support, or mostly (sometimes word size is ambiguous), but not in the cases where endian can vary. i.e. endian is unambiguous for x86, amd64, mac/ppc, ia64, sparc, arm64, riscv, m68k, hppa, alpha. Endian is ambiguous for mips, ppc, arm. Support is missing for m68k. Also anything we don't know we will fallback to uname. We might fallback to uname for a lot. Also the way NetBSD is handled here, nicely, we should fallback to uname. Or maybe just uname mostly, but this PR is ok. From noreply at github.com Tue Apr 6 06:39:47 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 21:39:47 -0700 Subject: [M3commit] [modula3/cm3] ca463b: Move chunk of RTSignalC.h to m3core.h, for sharing... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: ca463b0f04eaf6d8d7715297dae1fe882a323fae https://github.com/modula3/cm3/commit/ca463b0f04eaf6d8d7715297dae1fe882a323fae Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/runtime/POSIX/RTSignalC.c Log Message: ----------- Move chunk of RTSignalC.h to m3core.h, for sharing to MxConfig. Remove the trailing error case. The macro will be defined or not and the consumers will be prepapred for that or not. For MxConfig.HOST, we will intersperse static arch/os strings, but sometimes fail, and fallback to uname. From noreply at github.com Tue Apr 6 06:45:09 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 21:45:09 -0700 Subject: [M3commit] [modula3/cm3] 0eb320: m3core.h: Join lines and add parens, to GET_PC mac... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 0eb3205da3c06ecc12228833c092dc2d45e293d2 https://github.com/modula3/cm3/commit/0eb3205da3c06ecc12228833c092dc2d45e293d2 Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Join lines and add parens, to GET_PC macros. From noreply at github.com Tue Apr 6 07:55:31 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 22:55:31 -0700 Subject: [M3commit] [modula3/cm3] 2777ea: m3core.h: Add __WORDSIZE == 64 to comment. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 2777ea9134d25e61259d558be37bdc16bffacf6b https://github.com/modula3/cm3/commit/2777ea9134d25e61259d558be37bdc16bffacf6b Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Add __WORDSIZE == 64 to comment. From noreply at github.com Tue Apr 6 08:32:21 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 23:32:21 -0700 Subject: [M3commit] [modula3/cm3] 83f6ed: m3core.h More parens on GET_PC. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 83f6ed004356d4d186f248b6ffe55d151c12bd77 https://github.com/modula3/cm3/commit/83f6ed004356d4d186f248b6ffe55d151c12bd77 Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h More parens on GET_PC. From noreply at github.com Tue Apr 6 08:48:20 2021 From: noreply at github.com (Jay Krell) Date: Mon, 05 Apr 2021 23:48:20 -0700 Subject: [M3commit] [modula3/cm3] 042e54: MxConfig: Rewrite OS_TYPE as ifdef_win32. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 042e54a4f642020c321429ab88f302eef89916fb https://github.com/modula3/cm3/commit/042e54a4f642020c321429ab88f302eef89916fb Author: JayKrell Date: 2021-04-05 (Mon, 05 Apr 2021) Changed paths: M m3-sys/m3quake/src/MxConfig.m3 M m3-sys/m3quake/src/MxConfigC.c M m3-sys/m3quake/src/MxConfigC.i3 Log Message: ----------- MxConfig: Rewrite OS_TYPE as ifdef_win32. From noreply at github.com Tue Apr 6 09:21:49 2021 From: noreply at github.com (Jay Krell) Date: Tue, 06 Apr 2021 00:21:49 -0700 Subject: [M3commit] [modula3/cm3] 24fff5: Move MxConfig.HOST to C. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 24fff5dc01815c18ced9a61fe1dbd38321d82a1c https://github.com/modula3/cm3/commit/24fff5dc01815c18ced9a61fe1dbd38321d82a1c Author: JayKrell Date: 2021-04-06 (Tue, 06 Apr 2021) Changed paths: M m3-libs/rtinfo/src/RTInfo.m3 M m3-obliq/obliqrt/src/Obliq.m3 M m3-sys/cm3/src/Main.m3 M m3-sys/cm3/src/Makefile.m3 M m3-sys/m3quake/src/MxConfig.i3 M m3-sys/m3quake/src/MxConfig.m3 M m3-sys/m3quake/src/MxConfigC.c M m3-sys/m3quake/src/MxConfigC.i3 M m3-sys/m3quake/src/m3makefile Log Message: ----------- Move MxConfig.HOST to C. This is the last known divergence compiling C for same ostype/wordsize/endian in cm3 and its pieces (m3core, libm3, m3quake, etc.) This also further removes, like, a fixed list of targets from the system. While there is some special casing here, implying a list of known targets, there are few mitigating factors: - processor architecture and OS are separate - We really can just about use the uname strings asis, depending on what they look like and historical names. We do not special case every host, by far. From noreply at github.com Tue Apr 6 09:40:11 2021 From: noreply at github.com (Jay Krell) Date: Tue, 06 Apr 2021 00:40:11 -0700 Subject: [M3commit] [modula3/cm3] 780012: m3core.h: Add some constant M3_HOST. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 78001274dcb2f5e8599f0bf0c407adb5a9100775 https://github.com/modula3/cm3/commit/78001274dcb2f5e8599f0bf0c407adb5a9100775 Author: JayKrell Date: 2021-04-06 (Tue, 06 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Add some constant M3_HOST. From noreply at github.com Tue Apr 6 23:46:57 2021 From: noreply at github.com (Rodney M. Bates) Date: Tue, 06 Apr 2021 14:46:57 -0700 Subject: [M3commit] [modula3/cm3] 143484: Uncommited local changes, committed for travel. Message-ID: Branch: refs/heads/truckee Home: https://github.com/modula3/cm3 Commit: 1434847bfb6700b230624da33ab30dc030a36083 https://github.com/modula3/cm3/commit/1434847bfb6700b230624da33ab30dc030a36083 Author: Rodney Bates Date: 2021-04-06 (Tue, 06 Apr 2021) Changed paths: M m3-sys/m3front/src/exprs/ArrayExpr.m3 Log Message: ----------- Uncommited local changes, committed for travel. From noreply at github.com Wed Apr 7 03:01:24 2021 From: noreply at github.com (Jay Krell) Date: Tue, 06 Apr 2021 18:01:24 -0700 Subject: [M3commit] [modula3/cm3] 7679c7: m3c: Put ifndef/define around struct definitions, ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7679c73155d76ab306d01d1926eede1a0344ab76 https://github.com/modula3/cm3/commit/7679c73155d76ab306d01d1926eede1a0344ab76 Author: JayKrell Date: 2021-04-06 (Tue, 06 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Put ifndef/define around struct definitions, to facilitate concatenating all the .c files for a package into one. Otherwise struct definitions get duplicated and this is an error. From noreply at github.com Wed Apr 7 07:34:02 2021 From: noreply at github.com (Jay Krell) Date: Tue, 06 Apr 2021 22:34:02 -0700 Subject: [M3commit] [modula3/cm3] de1ae3: MxConfig: Cross check against m3core.h M3_HOST. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: de1ae3dd9415c44252de364b90578c5ddcf4f035 https://github.com/modula3/cm3/commit/de1ae3dd9415c44252de364b90578c5ddcf4f035 Author: JayKrell Date: 2021-04-06 (Tue, 06 Apr 2021) Changed paths: M m3-sys/m3quake/src/MxConfigC.c Log Message: ----------- MxConfig: Cross check against m3core.h M3_HOST. Consider just using M3_HOST. Commit: d81b4ae50946a1257faed7c08086517328925213 https://github.com/modula3/cm3/commit/d81b4ae50946a1257faed7c08086517328925213 Author: JayKrell Date: 2021-04-06 (Tue, 06 Apr 2021) Changed paths: M m3-sys/cm3/src/Builder.m3 Log Message: ----------- cm3: Fix nil dereference in cm3 -debug. Compare: https://github.com/modula3/cm3/compare/7679c73155d7...d81b4ae50946 From noreply at github.com Sun Apr 11 08:25:44 2021 From: noreply at github.com (Jay Krell) Date: Sat, 10 Apr 2021 23:25:44 -0700 Subject: [M3commit] [modula3/cm3] 96ae04: m3c: Uniquify segment names by package ("library o... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 96ae045ab585e4d8cfb1c867c0d3569837a60253 https://github.com/modula3/cm3/commit/96ae045ab585e4d8cfb1c867c0d3569837a60253 Author: JayKrell Date: 2021-04-10 (Sat, 10 Apr 2021) Changed paths: M m3-sys/cm3/src/Builder.m3 M m3-sys/cm3/src/M3Backend.i3 M m3-sys/cm3/src/M3Backend.m3 M m3-sys/m3back/src/M3C.i3 M m3-sys/m3back/src/M3C.m3 M m3-sys/m3cgcat/src/Main.m3 Log Message: ----------- m3c: Uniquify segment names by package ("library or program") and module This is an area where mfront/m3cg could use improvement. Specifically m3cg should be directly given a unique identifier prefix per source, that must be unique across the static link. i.e. if appended with a recurring sequence, like empty or 1, 2, 3, etc. And/or specifically m3front should be producing these names, for all backends. No nulls, no stars, no duplicates, no assumed file level static linkage. i.e. for segments. For any backend, append before link should be trivial. I'm a bit reluctant to mess with m3front though. While this change does not change m3front, it does tunnel through package/library/program name, and kinda source name to the backend. There is also a notion of unit_name, but it does not differentiate interface from module, which is perhaps unfortunate (I admit, I don't know what all goes into interfaces; one might imagine they have no code, but that is false). From noreply at github.com Mon Apr 12 00:06:04 2021 From: noreply at github.com (Rodney M. Bates) Date: Sun, 11 Apr 2021 15:06:04 -0700 Subject: [M3commit] [modula3/cm3] 6b9380: Fix long-standing bug in New, uncovered by better ... Message-ID: Branch: refs/heads/truckee Home: https://github.com/modula3/cm3 Commit: 6b9380b1bd6fe64cf324cbbd134f40a1f50e04e1 https://github.com/modula3/cm3/commit/6b9380b1bd6fe64cf324cbbd134f40a1f50e04e1 Author: ?Rodney Date: 2021-04-11 (Sun, 11 Apr 2021) Changed paths: M m3-sys/m3front/src/builtinOps/New.m3 Log Message: ----------- Fix long-standing bug in New, uncovered by better type checking in ArrayExpr. This is a bit complicated. The bug in New.m3 occurred when allocating an object with field initializatons. It was checking type assignability of an initializing expression against the wrong object field. This was originally covered up by incomplete checking when the expression was an array constructor. Recent completion of this checking uncovered the bug, with a symptom of false error messages, which then were not recovered from, resulting in massive internal CG failures. Note: The code being compiled does not have type assignabilty errors. The newly emitted messages were bogus, as result of the field mismatch. It is possible that the same error message, when produced legitimately, still fails to recover. From noreply at github.com Mon Apr 12 20:39:59 2021 From: noreply at github.com (Rodney M. Bates) Date: Mon, 12 Apr 2021 11:39:59 -0700 Subject: [M3commit] [modula3/cm3] 0244a9: Fix assertion failure compiling array constructor ... Message-ID: Branch: refs/heads/truckee Home: https://github.com/modula3/cm3 Commit: 0244a9c447a176493eb4677da8b0fae5a28da3a4 https://github.com/modula3/cm3/commit/0244a9c447a176493eb4677da8b0fae5a28da3a4 Author: ?Rodney Date: 2021-04-12 (Mon, 12 Apr 2021) Changed paths: M m3-sys/m3front/src/exprs/ArrayExpr.m3 Log Message: ----------- Fix assertion failure compiling array constructor '..' as inner constructor. Outer array constructor has '..', and its last real element is an inner array constructor. Needed to handle this case through direct array constructor recursion, rather than general expression recursion. This was already being done with non-'..' cases. Note: There was a booby trap here, with both a local varible and a WITH-binding of the same name 'argExpr'. This initially led me to think they were the same, leading to a severely broken compiler after recompiling itself. Renamed these to remove the confusion. From noreply at github.com Tue Apr 13 08:58:25 2021 From: noreply at github.com (Jay Krell) Date: Mon, 12 Apr 2021 23:58:25 -0700 Subject: [M3commit] [modula3/cm3] 04d813: Remove most or all of the accomodation for gcc pri... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 04d8137e31eced94edec664ef811196af753bc1b https://github.com/modula3/cm3/commit/04d8137e31eced94edec664ef811196af753bc1b Author: JayKrell Date: 2021-04-12 (Mon, 12 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- Remove most or all of the accomodation for gcc prior to gcc 4.3. Everything should still work with old gcc, but will likely warn, like for unsigned < 0. The problem with this code is not merely that it is large or ugly, but that it is conditional and therefore untested, until/unless such time as there is CI using gcc before 4.3. From noreply at github.com Tue Apr 13 09:42:28 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 00:42:28 -0700 Subject: [M3commit] [modula3/cm3] 4463b1: m3c: ifndef around struct types. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 4463b1b243248949f5fc53dc30dcd1658a8215e6 https://github.com/modula3/cm3/commit/4463b1b243248949f5fc53dc30dcd1658a8215e6 Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: ifndef around struct types. This is with the goal of concatenating all the files from a library/package, or across multiple library/packages, for a crude form of LTCG/LTO, or easier to bootstrap because cm3 will be just one file. From noreply at github.com Tue Apr 13 09:44:02 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 00:44:02 -0700 Subject: [M3commit] [modula3/cm3] be58ba: m3c: Add ifndef around helper function instantiati... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: be58ba4e00a14e6c6761068a0c496ffdd0a5563c https://github.com/modula3/cm3/commit/be58ba4e00a14e6c6761068a0c496ffdd0a5563c Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Add ifndef around helper function instantiations. They are template-like macros. This is in the interest of concatening all the files of a package or program, so cm3 redist is one source file. Rename ifndef_type to ifndef. This defines things to just be themselves. ifndef foo define foo foo void foo(...) { ... } endif From noreply at github.com Tue Apr 13 10:12:44 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 01:12:44 -0700 Subject: [M3commit] [modula3/cm3] cdaaec: m3c: Disable some warnings that are not currently ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: cdaaece76b6e5a3856670807d99d00015e28e570 https://github.com/modula3/cm3/commit/cdaaece76b6e5a3856670807d99d00015e28e570 Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Disable some warnings that are not currently enabled. From noreply at github.com Tue Apr 13 10:17:32 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 01:17:32 -0700 Subject: [M3commit] [modula3/cm3] Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 From noreply at github.com Tue Apr 13 10:34:33 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 01:34:33 -0700 Subject: [M3commit] [modula3/cm3] 59e8ca: NT config: Remove the delayload support. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 59e8ca3ff079626e296107860cde2527ae22d686 https://github.com/modula3/cm3/commit/59e8ca3ff079626e296107860cde2527ae22d686 Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/NT.common Log Message: ----------- NT config: Remove the delayload support. I added it. Probably nobody noticed or cares. It produces warnings, now that I removed the listing file stuff. It is not a bad idea, but perhaps this is not the way. Really the output should be factored and formed so that cm3 does not actually import this unused stuff. cm3 is not really CreateProcess/LoadLibrary intensive, as it runs per package not per source, and even this we might reduce. ws2_32.dll (commented out) wsock32.dll advapi32.dll gdi32.dll netapi32.dll user32.dll comctl32.dll rpcrt4.dll iphlpapid.dll (typo) It also doesn't work on quite old no-longer-untested toolsets (Visual C++ 2003 Express and pre-Visual++5.0sp?) From noreply at github.com Wed Apr 14 01:38:52 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 16:38:52 -0700 Subject: [M3commit] [modula3/cm3] 9e6e14: m3cggen: Fill in the missing opcodes. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 9e6e1418ee36da6d91207fe57664968aba66d87c https://github.com/modula3/cm3/commit/9e6e1418ee36da6d91207fe57664968aba66d87c Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3cggen/src/Main.m3 Log Message: ----------- m3cggen: Fill in the missing opcodes. They do not appear in the binary/file, but are useful. From noreply at github.com Wed Apr 14 01:57:03 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 16:57:03 -0700 Subject: [M3commit] [modula3/cm3] 15e7fb: Move m3cggen.Map to M3CG_Binary, exposed via M3CGB... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 15e7fb006af250c94581db61cd873d77ee5c1293 https://github.com/modula3/cm3/commit/15e7fb006af250c94581db61cd873d77ee5c1293 Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3cggen/src/Main.m3 M m3-sys/m3middle/src/M3CG_Binary.i3 M m3-sys/m3middle/src/M3CG_Binary.m3 Log Message: ----------- Move m3cggen.Map to M3CG_Binary, exposed via M3CGBinary.OpText(Op):TEXT. This is useful beyond m3ccgen, will be shortly. From noreply at github.com Wed Apr 14 02:20:54 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 17:20:54 -0700 Subject: [M3commit] [modula3/cm3] 6fbcdf: m3c: ifndef around set helpers. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 6fbcdf0ab3a4c9778030f756d29850bcb96fa636 https://github.com/modula3/cm3/commit/6fbcdf0ab3a4c9778030f756d29850bcb96fa636 Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: ifndef around set helpers. This is so files can concatenated. ifndef: Rename "type" to "id". Prepend "m3once_". Remove m3_ prepend at call sites. Add newlines. From noreply at github.com Wed Apr 14 02:39:37 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 17:39:37 -0700 Subject: [M3commit] [modula3/cm3] 8c37c9: m3c: Ifndef guards around conversion helpers. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 8c37c9c97918b5db754771589bc4e6192b2c1ac3 https://github.com/modula3/cm3/commit/8c37c9c97918b5db754771589bc4e6192b2c1ac3 Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Ifndef guards around conversion helpers. Or more generally around the type-independent prefixes. Add some comments and rename some identifiers for clarity, e.g. "already_printed" and "types_already_printed". The goal is to concatenate files and compile with one/fewer compiler invocations esp. so that cm3 is one redistributable source file. M3c is not significantly altered for this goal, nor is common code lifted out to a library. There are tradeoffs to every approach. From noreply at github.com Wed Apr 14 03:17:44 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 18:17:44 -0700 Subject: [M3commit] [modula3/cm3] e29d4b: m3front: Declare binders correctly. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: e29d4b5fb0b964fe94c64c34a974e27c18c8cc70 https://github.com/modula3/cm3/commit/e29d4b5fb0b964fe94c64c34a974e27c18c8cc70 Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3front/src/values/Module.m3 Log Message: ----------- m3front: Declare binders correctly. before: ADDRESS Foo_I3(void); after: ADDRESS Foo_I3(INTEGER mode); They were incorrect in .i3 files, correct in .m3 files. The mismatch leads to errors when you concatenate m3c output, and there is no reason for it. Move the implementation declaration to common code for reuse by the import. From noreply at github.com Wed Apr 14 06:25:49 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 21:25:49 -0700 Subject: [M3commit] [modula3/cm3] 9e90a4: m3c: More ifndef helper guards (fence, convert_int... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 9e90a43d39f9b63b6bfeb081620d95a61ccadea3 https://github.com/modula3/cm3/commit/9e90a43d39f9b63b6bfeb081620d95a61ccadea3 Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: More ifndef helper guards (fence, convert_int, etc.) shortn define m3once_foo m3once_foo to define m3once_foo From noreply at github.com Wed Apr 14 06:30:38 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 21:30:38 -0700 Subject: [M3commit] [modula3/cm3] 72c522: m3c: More ifndef helper guards (fence, convert_int... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 72c522cb59f07fc71128b36b60ce729bae27b0d9 https://github.com/modula3/cm3/commit/72c522cb59f07fc71128b36b60ce729bae27b0d9 Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: More ifndef helper guards (fence, convert_int, etc.) shortn define m3once_foo m3once_foo to define m3once_foo From noreply at github.com Wed Apr 14 07:38:45 2021 From: noreply at github.com (Jay Krell) Date: Tue, 13 Apr 2021 22:38:45 -0700 Subject: [M3commit] [modula3/cm3] 7d3465: m3c: Initialize some record fields to nil. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7d3465760eb1b7f8404f9a286316e9ae2f00decf https://github.com/modula3/cm3/commit/7d3465760eb1b7f8404f9a286316e9ae2f00decf Author: JayKrell Date: 2021-04-13 (Tue, 13 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Initialize some record fields to nil. From noreply at github.com Wed Apr 14 09:02:23 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 00:02:23 -0700 Subject: [M3commit] [modula3/cm3] b8ddfd: m3c: Fix typo declaring pointers and resulting fal... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: b8ddfd42d12175d8fbea530b60b5f5483665bc1f https://github.com/modula3/cm3/commit/b8ddfd42d12175d8fbea530b60b5f5483665bc1f Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Fix typo declaring pointers and resulting fallout. There are typoed recursive types in the system: TYPE FOO = REF FOO; We reluctantly accept these since the rest of the system does. When we see this, we attempt to break the loop by just treating it as TYPE FOO = ADDRESS. Luckily they are always self-referential. We do not have: TYPE FOO = REF BAR; BAR = REF FOO; which would take more work to detect. We do in some sense detect these, by erroring, due to a lack of forward progress, described below. The detection was supposed be like: if type.typeid == type.points_to_typeid fallback to address but was accidentally: if type.typeid == type.typeid which is always true. This was in type.canBeDefined, not type.define. This lead to calling type.define prematurely. This lead to type.define falling back and typedefing types to ADDRESS instead of something strong (pointer is correct, but pointer to what?). type.define really should have errored in this case, and now it does. This all worked ok. Just that debugging might suffer. However, if you go and concatenate files, this occurs only sometimes. So we have a mix of: struct foo; typedef foo* pfoo; // desired and typedef void* pfoo; // bogus fallback which is fortunately an error in C. Once you fix this (typeid == typeid vs. typeid == points_to_typeid), you then run into slight misimplementation of open arrays and forward declaration. Open arrays require a definition or at least forward declaration of their constituent elements. So they can declare a pointer to them. The implementation of forward declare ignored if a type can be forward declared. For example records/structs can be forward declared and pointers cannot be. As a result, you could get an open array with a pointer to a completely undeclared type. This is also an error. The fix for this is to call canBeForwardDeclared before calling forwardDeclare and setting the type's state to forward declared. Calling forwardDeclare does nothing for types that cannot be forward declared, but setting the state to forward declared, enables the open array declaration to proceed. It should be noted/reminded that the way all the type declarations works: - Build up in memory a representation of all of m3front's type declarations. - Iterate through the types repeatedly, forward declaring and defining what/when can be. Loop until no forward progress is made or until all types are defined. If no forward progress, error. - m3front is not presumed to output in any particular order. From noreply at github.com Wed Apr 14 11:07:35 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 02:07:35 -0700 Subject: [M3commit] [modula3/cm3] 531e64: m3c: Suppress more warnings. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 531e64160716887a58b6b0f4009ef73a8738a425 https://github.com/modula3/cm3/commit/531e64160716887a58b6b0f4009ef73a8738a425 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Suppress more warnings. 4715 not all control paths return a value 4716 must return a value From noreply at github.com Wed Apr 14 11:08:54 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 02:08:54 -0700 Subject: [M3commit] [modula3/cm3] abe12f: m3core: Unifity ADDRESS with m3c. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: abe12f3a74ac51bda14299370413b18c3dfa4c63 https://github.com/modula3/cm3/commit/abe12f3a74ac51bda14299370413b18c3dfa4c63 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/runtime/common/RTLinkerC.c Log Message: ----------- m3core: Unifity ADDRESS with m3c. This is maybe unfortunate, maybe ok. m3core.h had: if defined(__cplusplus) || __STDC__ typedef void* ADDRESS; else typedef char* ADDRESS; endif but m3c outputs: (* TODO ideally these are char* for K&R or ideally absent when strong typing and setjmp work done *) typedef char* ADDRESS; typedef char* STRUCT; and the resulting code adds integers to ADDRESS. Which is fine for char* but non-standard/illegal for void*. The strong typing thing might happen eventually, esp. for structs, but it also requires the offseting to be done by name/index in m3back instead of m3front layout, so pretty far out. If you concatenate the m3c output with the hand written C you get approximately one error, about RTLinker__GetEnvironmen We could alternatively rename one of them, or just have m3c output "char*" everywhere (more tokens, fewer symbols). However this change changes m3core.h to char* and RTLinker__GetEnvironmentStrings to use ADDRESS instead of void*. char* works with pre-ANSI, if that matters (we require ANSI C currently and intend to move up to C++). char* tends to show garbage in debuggers vs. void* shows nothing. So maybe revisit this. From noreply at github.com Wed Apr 14 11:10:05 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 02:10:05 -0700 Subject: [M3commit] [modula3/cm3] 6f78d8: m3c: Tolerate duplicate typeids. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 6f78d8fed7d3a97c53c1ea48d43dc948cadfcefe https://github.com/modula3/cm3/commit/6f78d8fed7d3a97c53c1ea48d43dc948cadfcefe Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Tolerate duplicate typeids. Subranges have colliding typeids. m3front bug? For example: libm3\Formatter.m3: Int = REF INTEGER; ints: ARRAY [-256..256] OF Int; declare_subrange typeid:T69A2A904 domain_type:T195C2A74 min:-256 max:256 bit_size:16 libm3\Sx.m3 MinBoxedInt = -100; MaxBoxedInt = 100; BoxedInts := ARRAY [MinBoxedInt .. MaxBoxedInt] OF REF INTEGER {NIL, ..}; declare_subrange typeid:T69A2A904 domain_type:T195C2A74 min:-100 max:100 bit_size:8 Partial workaround it here by appending the size to the type text. We could further workaround by deferring the typedef until use, as in this case, neither type is used. Just defined. This shows up if you concatenate all the files that make up cm3, which is a goal for easier redist and build. From noreply at github.com Wed Apr 14 21:51:55 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 12:51:55 -0700 Subject: [M3commit] [modula3/cm3] b44488: m3c: Replace CHAR with UCHAR due to mild conflict ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: b444881d3239654870377f0755ff6202d061ca6c https://github.com/modula3/cm3/commit/b444881d3239654870377f0755ff6202d061ca6c Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Replace CHAR with UCHAR due to mild conflict with windows.h when concating m3c output with hand written Win32 C. From noreply at github.com Thu Apr 15 02:53:58 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 17:53:58 -0700 Subject: [M3commit] [modula3/cm3] e531d7: m3core.h: Delete extra old/subsetted copies. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: e531d793c7be9c561fa2bdc1c10d20b080fc7584 https://github.com/modula3/cm3/commit/e531d793c7be9c561fa2bdc1c10d20b080fc7584 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: R m3-sys/m3back/src/m3core.h R m3-sys/m3middle/src/POSIX/m3core.h Log Message: ----------- m3core.h: Delete extra old/subsetted copies. There were to avoid depending on m3core long ago, and it was dubious then. Duplicate file names like this impede builds that combine files into fewer directories or files and I'm surprised they weren't a problem before. Well, becuase they aren't mentioned in m3makefile/h_source, that helps explain that. From noreply at github.com Thu Apr 15 05:20:36 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 20:20:36 -0700 Subject: [M3commit] [modula3/cm3] 4995b7: m3core.h: Hypothetically extend Solaris 9 4/03 use... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 4995b75626ec5039a9ca63ce2908911083a803d2 https://github.com/modula3/cm3/commit/4995b75626ec5039a9ca63ce2908911083a803d2 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Hypothetically extend Solaris 9 4/03 user thread support to x86. Really we should just drop user threads for most systems. From noreply at github.com Thu Apr 15 06:42:13 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 21:42:13 -0700 Subject: [M3commit] [modula3/cm3] a2cfaf: m3core.h: Put ifndef INCLUDED_M3CORE_H/endif aroun... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: a2cfaf538edd2b53eb44787f2f60c2aae4266c55 https://github.com/modula3/cm3/commit/a2cfaf538edd2b53eb44787f2f60c2aae4266c55 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-comm/tcp/src/common/IP_c.c M m3-comm/udp/src/Common/UDP_c.c M m3-demo/mentor/src/sorting/audio/MidiLineServerC.c M m3-libs/libm3/src/os/POSIX/FSPosixC.c M m3-libs/libm3/src/os/POSIX/FilePosixC.c M m3-libs/libm3/src/os/POSIX/OSConfigPosixC.c M m3-libs/libm3/src/os/POSIX/SocketPosixC.c M m3-libs/libm3/src/uid/POSIX/MachineIDPosixC.c M m3-libs/m3core/src/C/Common/CerrnoC.c M m3-libs/m3core/src/C/Common/CsetjmpC.c M m3-libs/m3core/src/C/Common/CsignalC.c M m3-libs/m3core/src/C/Common/CstdioC.c M m3-libs/m3core/src/C/Common/CstdlibC.c M m3-libs/m3core/src/C/Common/CstringC.c M m3-libs/m3core/src/Csupport/Common/hand.c M m3-libs/m3core/src/Csupport/big-endian/dtoa.c M m3-libs/m3core/src/Csupport/little-endian/dtoa.c M m3-libs/m3core/src/coroutine/UCONTEXT/ContextC.c M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/runtime/POSIX/RTOSc.c M m3-libs/m3core/src/runtime/POSIX/RTSignalC.c M m3-libs/m3core/src/runtime/POSIX/test_signal.c M m3-libs/m3core/src/runtime/SOLsun/RTStackC.c M m3-libs/m3core/src/runtime/WIN32/RTOSc.c M m3-libs/m3core/src/runtime/common/RTBuiltin.c M m3-libs/m3core/src/runtime/common/RTIOc.c M m3-libs/m3core/src/runtime/common/RTLinkerC.c M m3-libs/m3core/src/runtime/common/RTMiscC.c M m3-libs/m3core/src/runtime/common/RTProcessC.c M m3-libs/m3core/src/runtime/common/RTUntracedMemoryC.c M m3-libs/m3core/src/runtime/ex_frame/RTStackC.c M m3-libs/m3core/src/thread/Common/ThreadDebug.c M m3-libs/m3core/src/thread/Common/ThreadInternal.c M m3-libs/m3core/src/thread/POSIX/ThreadPosixC.c M m3-libs/m3core/src/thread/POSIX/test_timer.c M m3-libs/m3core/src/thread/PTHREAD/ThreadApple.c M m3-libs/m3core/src/thread/PTHREAD/ThreadFreeBSD.c M m3-libs/m3core/src/thread/PTHREAD/ThreadOpenBSD.c M m3-libs/m3core/src/thread/PTHREAD/ThreadPThreadC.c M m3-libs/m3core/src/thread/PTHREAD/interix_context_on_signal_stack.c M m3-libs/m3core/src/thread/PTHREAD/test_interix.c M m3-libs/m3core/src/time/POSIX/DatePosixC.c M m3-libs/m3core/src/time/POSIX/TimePosixC.c M m3-libs/m3core/src/unix/Common/Uconstants.c M m3-libs/m3core/src/unix/Common/UdirC.c M m3-libs/m3core/src/unix/Common/Uexec.c M m3-libs/m3core/src/unix/Common/Ugrp.c M m3-libs/m3core/src/unix/Common/Uin.c M m3-libs/m3core/src/unix/Common/Umman.c M m3-libs/m3core/src/unix/Common/Unetdb.c M m3-libs/m3core/src/unix/Common/UnixC.c M m3-libs/m3core/src/unix/Common/UnixLink.c M m3-libs/m3core/src/unix/Common/Uprocess.c M m3-libs/m3core/src/unix/Common/Upwd.c M m3-libs/m3core/src/unix/Common/Usignal.c M m3-libs/m3core/src/unix/Common/Usocket.c M m3-libs/m3core/src/unix/Common/UstatC.c M m3-libs/m3core/src/unix/Common/UtimeC.c M m3-libs/m3core/src/unix/Common/Uugid.c M m3-libs/m3core/src/unix/Common/Uuio.c M m3-libs/m3core/src/unix/Common/Uutmp.c M m3-sys/fix_nl/src/OS.c M m3-sys/m3back/src/M3CC.c M m3-sys/m3middle/src/CoffTime.c M m3-sys/m3middle/src/POSIX/CoffTime.c M m3-sys/m3quake/src/MxConfigC.c M m3-tools/cvsup/suplib/src/FileAttrC.c M m3-tools/cvsup/suplib/src/FreeBSD/UProcTitle.c M m3-tools/cvsup/suplib/src/UnixMiscC.c M m3-tools/cvsup/suplib/src/libglob/fnmatch.c M m3-tools/cvsup/suplib/src/libmd/md5c.c M m3-tools/cvsup/suplib/src/libmd/md5hl.c M m3-tools/cvsup/suptcp/src/common/SupErrnoC.c Log Message: ----------- m3core.h: Put ifndef INCLUDED_M3CORE_H/endif around include m3core.h. Perspective here vary: - Nice build optimization. On the assumption that it is preventing multiple include. (not likely the case here) On the assumpion that the compiler does not notice and optimize the pattern (they do). - Layer violation. The symbol's exact name is m3core.h's internal business. I tend toward the second and don't see people make this optimization much. Many code bases have no such symbols and just use pragma once or require single include. However in this case it is with the goal of concatenability. Eventually some of these files, m3core.h, and other files will in some context (redistribution) be concatenated, and m3core.h nonexistant. This is a step toward enabling that. Note that all the content before m3core.h needs to be reconciled. Not all of these files are likely to be concatenated. i.e. only what makes up cm3. From noreply at github.com Thu Apr 15 07:56:02 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 22:56:02 -0700 Subject: [M3commit] [modula3/cm3] fac3f3: m3core.h: Consolidation. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: fac3f337c40a733460cc827ed6d1ff0a6d1e3d7e https://github.com/modula3/cm3/commit/fac3f337c40a733460cc827ed6d1ff0a6d1e3d7e Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-comm/tcp/src/common/IP_c.c M m3-comm/udp/src/Common/UDP_c.c M m3-libs/libm3/src/os/POSIX/FilePosixC.c M m3-libs/libm3/src/os/POSIX/SocketPosixC.c M m3-libs/libm3/src/uid/POSIX/MachineIDPosixC.c M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Consolidation. Do not include stuff that m3core.h includes. Move stuff out of ifdef if ok. Combine pragma warning dispatch. General idea is to let these files be concatenated and keep correct and same meaning. Sort warning pragmas. From noreply at github.com Thu Apr 15 08:15:27 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 23:15:27 -0700 Subject: [M3commit] [modula3/cm3] 820e6b: m3core/libm3: Move from winsock.h to winsock2.h. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 820e6bb4a7e4c5f39a2dfc13d0108ad049420088 https://github.com/modula3/cm3/commit/820e6bb4a7e4c5f39a2dfc13d0108ad049420088 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-libs/libm3/src/os/POSIX/SocketPosixC.c M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core/libm3: Move from winsock.h to winsock2.h. From noreply at github.com Thu Apr 15 08:20:28 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 23:20:28 -0700 Subject: [M3commit] [modula3/cm3] 9de9e3: Remove include of winsock.h, replaced with winsock... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 9de9e3accf4540205c12d7ccab9ef64706569338 https://github.com/modula3/cm3/commit/9de9e3accf4540205c12d7ccab9ef64706569338 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- Remove include of winsock.h, replaced with winsock2.h. Commit: 3e7def0761a129e7567a987fe15275a131fc58c2 https://github.com/modula3/cm3/commit/3e7def0761a129e7567a987fe15275a131fc58c2 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Define WIN32_LEAN_AND_MEAN. Compare: https://github.com/modula3/cm3/compare/820e6bb4a7e4...3e7def0761a1 From noreply at github.com Thu Apr 15 08:24:21 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 23:24:21 -0700 Subject: [M3commit] [modula3/cm3] e72ea2: NT config: Use ws2_32 instead of wsock32. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: e72ea24dda22071ac3f903880921717f59a8abc0 https://github.com/modula3/cm3/commit/e72ea24dda22071ac3f903880921717f59a8abc0 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/NT.common Log Message: ----------- NT config: Use ws2_32 instead of wsock32. This moves required toolset up to around 1996. From noreply at github.com Thu Apr 15 08:29:21 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 23:29:21 -0700 Subject: [M3commit] [modula3/cm3] 2add87: NT config: Link to iphlpapi.lib. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 2add875ed74741a7a7d4e9a74a4938608e93bec9 https://github.com/modula3/cm3/commit/2add875ed74741a7a7d4e9a74a4938608e93bec9 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/NT.common Log Message: ----------- NT config: Link to iphlpapi.lib. This increased the required toolset, to something still very old. From noreply at github.com Thu Apr 15 08:30:38 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 23:30:38 -0700 Subject: [M3commit] [modula3/cm3] c2eccd: libm3: Bind to iphlpapi.dll more directly. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: c2eccd430e6adde39f46e1c74b40e78f72f6cd92 https://github.com/modula3/cm3/commit/c2eccd430e6adde39f46e1c74b40e78f72f6cd92 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-libs/libm3/src/uid/WIN32/MachineID.c Log Message: ----------- libm3: Bind to iphlpapi.dll more directly. (Consider linker delayload). From noreply at github.com Thu Apr 15 08:41:36 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 23:41:36 -0700 Subject: [M3commit] [modula3/cm3] b1be67: libm3/MachineID Win32: Use the Windows headers Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: b1be6731e0ddd965bd81ed9f133f57e44e00b1c9 https://github.com/modula3/cm3/commit/b1be6731e0ddd965bd81ed9f133f57e44e00b1c9 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-libs/libm3/src/uid/WIN32/MachineID.c Log Message: ----------- libm3/MachineID Win32: Use the Windows headers instead of duplicating their content. This requires a "newer" SDK, very old works. Commit: e6438999fea9ca81ca6ad65ca2fd7315d166415f https://github.com/modula3/cm3/commit/e6438999fea9ca81ca6ad65ca2fd7315d166415f Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-libs/libm3/src/uid/WIN32/MachineID.c Log Message: ----------- libm3 Win32: Make disabled test code easier to build. Compare: https://github.com/modula3/cm3/compare/c2eccd430e6a...e6438999fea9 From noreply at github.com Thu Apr 15 08:42:09 2021 From: noreply at github.com (Jay Krell) Date: Wed, 14 Apr 2021 23:42:09 -0700 Subject: [M3commit] [modula3/cm3] 02292b: libm3 Win32: Fix indent. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 02292bed356802e5f3809ce8448799b823b262f9 https://github.com/modula3/cm3/commit/02292bed356802e5f3809ce8448799b823b262f9 Author: JayKrell Date: 2021-04-14 (Wed, 14 Apr 2021) Changed paths: M m3-libs/libm3/src/uid/WIN32/MachineID.c Log Message: ----------- libm3 Win32: Fix indent. From noreply at github.com Thu Apr 15 23:51:54 2021 From: noreply at github.com (Jay Krell) Date: Thu, 15 Apr 2021 14:51:54 -0700 Subject: [M3commit] [modula3/cm3] ae97f0: scripts: Replace wsock32.lib with ws2_32.lib and a... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: ae97f0e995282cd58966ed467be71d4b2e6a2fe2 https://github.com/modula3/cm3/commit/ae97f0e995282cd58966ed467be71d4b2e6a2fe2 Author: JayKrell Date: 2021-04-15 (Thu, 15 Apr 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- scripts: Replace wsock32.lib with ws2_32.lib and add iphlpapi.lib. From noreply at github.com Sat Apr 17 00:06:52 2021 From: noreply at github.com (Jay Krell) Date: Fri, 16 Apr 2021 15:06:52 -0700 Subject: [M3commit] [modula3/cm3] d3fd54: m3core: Fix compiling Umman.c on most systems due ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: d3fd54281346be52dc7b95f7fa18f1f941ebd7cc https://github.com/modula3/cm3/commit/d3fd54281346be52dc7b95f7fa18f1f941ebd7cc Author: JayKrell Date: 2021-04-16 (Fri, 16 Apr 2021) Changed paths: M m3-libs/m3core/src/unix/Common/Umman.c Log Message: ----------- m3core: Fix compiling Umman.c on most systems due to Solaris fix. ../src/unix/Common/Umman.c: In function cchar* Umman__mmap(caddr_t, WORD_T, int, int, int, m3_off_t): ../src/m3core.h:315:106: error: invalid conversion from void* to ADDRESS (char*) Keep the parameters char*, which converts to void*. But make the return void*, which converts from char*. From noreply at github.com Sat Apr 17 00:33:31 2021 From: noreply at github.com (Jay Krell) Date: Fri, 16 Apr 2021 15:33:31 -0700 Subject: [M3commit] [modula3/cm3] 79610c: m3core.h: Fix .h file for mmap. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 79610c4ca6c7049e8f045550a38214721dd78dcc https://github.com/modula3/cm3/commit/79610c4ca6c7049e8f045550a38214721dd78dcc Author: JayKrell Date: 2021-04-16 (Fri, 16 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Fix .h file for mmap. From noreply at github.com Sat Apr 17 07:24:21 2021 From: noreply at github.com (Jay Krell) Date: Fri, 16 Apr 2021 22:24:21 -0700 Subject: [M3commit] [modula3/cm3] 0be04a: x-opengl: Fix calling convention of callbacks. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 0be04a6852860d03a0f197d91e7bf13843b5b3b1 https://github.com/modula3/cm3/commit/0be04a6852860d03a0f197d91e7bf13843b5b3b1 Author: JayKrell Date: 2021-04-16 (Fri, 16 Apr 2021) Changed paths: M m3-ui/anim3D/src/x-opengl/X_OpenGL_Base.m3 Log Message: ----------- x-opengl: Fix calling convention of callbacks. This fixes compiling x-opengl using the C backend. Generally calling conventions, having more than one, is an anomoly. It is only for NT/x86. However when using the C backend, m3middle/m3front/backend deliberately do not know this. So for example, I386_NT is the same as ARM32_NT. So all code needs to be written as if it is targeting NT/x86, regarding calling conventions. The fix is trivial and in good style. Instead of casting to PROCEDURE(), cast to GLu.GLUtessAnyProc. From noreply at github.com Sat Apr 17 07:45:54 2021 From: noreply at github.com (Jay Krell) Date: Fri, 16 Apr 2021 22:45:54 -0700 Subject: [M3commit] [modula3/cm3] 149d9c: sysutils/m3core consolidation Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 149d9cf2228056ca240e96b8cf41faf960408553 https://github.com/modula3/cm3/commit/149d9cf2228056ca240e96b8cf41faf960408553 Author: JayKrell Date: 2021-04-16 (Fri, 16 Apr 2021) Changed paths: M m3-libs/m3core/src/unix/Common/Uconstants.c M m3-libs/sysutils/src/FSUtils.i3 R m3-libs/sysutils/src/POSIX/FSUnixC.c M m3-libs/sysutils/src/POSIX/FSUnix_cm3.m3 M m3-libs/sysutils/src/POSIX/FSUnix_pm3.m3 M m3-libs/sysutils/src/POSIX/SystemPosix.m3 R m3-libs/sysutils/src/POSIX/SystemPosixC.c M m3-libs/sysutils/src/POSIX/m3makefile M m3-libs/sysutils/src/System.i3 R m3-libs/sysutils/src/SystemC.c M m3-libs/sysutils/src/WIN32/FSysWin32.m3 M m3-libs/sysutils/src/m3makefile Log Message: ----------- sysutils/m3core consolidation Sysutils was duplicating m3core for compatibility with old m3core which is not needed or worthwhile. From noreply at github.com Sat Apr 17 07:57:18 2021 From: noreply at github.com (Jay Krell) Date: Fri, 16 Apr 2021 22:57:18 -0700 Subject: [M3commit] [modula3/cm3] 89ee43: sysutil: Remove disabled workaround for old mklib. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 89ee4304451ae21fd19204b7affac1a2f023cb97 https://github.com/modula3/cm3/commit/89ee4304451ae21fd19204b7affac1a2f023cb97 Author: JayKrell Date: 2021-04-16 (Fri, 16 Apr 2021) Changed paths: M m3-libs/sysutils/src/m3makefile Log Message: ----------- sysutil: Remove disabled workaround for old mklib. From noreply at github.com Sat Apr 17 08:22:57 2021 From: noreply at github.com (Jay Krell) Date: Fri, 16 Apr 2021 23:22:57 -0700 Subject: [M3commit] [modula3/cm3] 7e060b: m3core: Remove FreeBSD 4 special case. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7e060b30bee80d97241e40c4d910036167120eda https://github.com/modula3/cm3/commit/7e060b30bee80d97241e40c4d910036167120eda Author: JayKrell Date: 2021-04-16 (Fri, 16 Apr 2021) Changed paths: M m3-libs/m3core/src/thread/m3makefile Log Message: ----------- m3core: Remove FreeBSD 4 special case. Simplifiy selection of Win32 threads. Some of the conditions were redundant. Require TARGET_OS to be set, such as to NT, CYGWIN, LINUX, DARWIN, etc. Generally TARGET = TARGET_ARCH + "_" + TARGET_OS. TARGET_OS is a further refinement of Posix generally (or NT vs. MINGW, though I think MINGW is really just a toolset, not a host or target). TARGET_ARCH isn't really interesting at all. Nothing outside of m3cc is architecture-specific, except for GET_PC and M3_HOST, which are optional. The actual factors are WORD_SIZE and TARGET_ENDIAN, which do closely correlate with TARGET_ARCH, but such decomposition hypothetically makes for fewer (four) total combinations. (if you consider sparc, mips, ppc alpha etc. 'cause actually there are only a few interesting processors these days: amd64, arm64, riscv64, wasm32, wasm64, decomposition into wordsize+endian isn't a huge help perhaps) If FreeBSD 4 and earlier pthreads are not adequate, this can be factored into the toplevel config. FreeBSD 4 is out of support since early 2007, and might still work. From noreply at github.com Sat Apr 17 08:31:32 2021 From: noreply at github.com (Jay Krell) Date: Fri, 16 Apr 2021 23:31:32 -0700 Subject: [M3commit] [modula3/cm3] e3debf: Revert "I had this change around: expose if we are... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: e3debff3f274af8c5ae74cbdfa15e702e5ae6277 https://github.com/modula3/cm3/commit/e3debff3f274af8c5ae74cbdfa15e702e5ae6277 Author: JayKrell Date: 2021-04-16 (Fri, 16 Apr 2021) Changed paths: M m3-libs/sysutils/src/POSIX/m3makefile Log Message: ----------- Revert "I had this change around: expose if we are using user threads, i.e. if we are OpenBSD" This reverts commit 5f41d677686c8f58eac0391390e50359427be3ed. Hopefully OpenBSD pthreads work, and otherwise we'd need multiple distributions, or, really we should make this decision possible at runtime not just build time. Unlikely anyone ever used this. From noreply at github.com Sat Apr 17 08:45:15 2021 From: noreply at github.com (Jay Krell) Date: Fri, 16 Apr 2021 23:45:15 -0700 Subject: [M3commit] [modula3/cm3] 659ef7: mklib: Remove m3core duplication. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 659ef7f918a41cfafe1285bcda9ca234394f2c4d https://github.com/modula3/cm3/commit/659ef7f918a41cfafe1285bcda9ca234394f2c4d Author: JayKrell Date: 2021-04-16 (Fri, 16 Apr 2021) Changed paths: M m3-sys/mklib/src/Main.m3 Log Message: ----------- mklib: Remove m3core duplication. m3core now provides Win32 types/constants for all systems, allowing some cross scenarios (the .lib file format working here should go away, just write .def files, in m3front) From noreply at github.com Sat Apr 17 09:05:41 2021 From: noreply at github.com (Jay Krell) Date: Sat, 17 Apr 2021 00:05:41 -0700 Subject: [M3commit] [modula3/cm3] 6e649f: m3core/mklib: Add support for "all" Windows platfo... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 6e649f984c654dc334bc53a1abe1e266361050c9 https://github.com/modula3/cm3/commit/6e649f984c654dc334bc53a1abe1e266361050c9 Author: JayKrell Date: 2021-04-17 (Sat, 17 Apr 2021) Changed paths: M m3-libs/m3core/src/win32/WinNT.i3 M m3-sys/mklib/src/Main.m3 Log Message: ----------- m3core/mklib: Add support for "all" Windows platforms. Esp. ARMNT and ARM64 should be interesting. Though this code should become dormant soon. Assume x86 is the only one with the underscores. From noreply at github.com Sat Apr 17 19:26:46 2021 From: noreply at github.com (Jay Krell) Date: Sat, 17 Apr 2021 10:26:46 -0700 Subject: [M3commit] [modula3/cm3] a2d778: m3c: Fill in ProcTypes. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: a2d778acee07538f21c4a61a85eba9fc8b218d51 https://github.com/modula3/cm3/commit/a2d778acee07538f21c4a61a85eba9fc8b218d51 Author: JayKrell Date: 2021-04-17 (Sat, 17 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Fill in ProcTypes. This does not help all that much, i.e. for debugging. It might help us remove casts on non-closure function pointer calls. It is a step toward being able to concatenate m3c output with all the .c files in the system, and distributing cm3 as one .c file. Otherwise there are errors e.g. regarding RTProcess__RegisterExitor. From noreply at github.com Sun Apr 18 03:19:51 2021 From: noreply at github.com (Rodney M. Bates) Date: Sat, 17 Apr 2021 18:19:51 -0700 Subject: [M3commit] [modula3/cm3] 143484: Uncommited local changes, committed for travel. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 1434847bfb6700b230624da33ab30dc030a36083 https://github.com/modula3/cm3/commit/1434847bfb6700b230624da33ab30dc030a36083 Author: Rodney Bates Date: 2021-04-06 (Tue, 06 Apr 2021) Changed paths: M m3-sys/m3front/src/exprs/ArrayExpr.m3 Log Message: ----------- Uncommited local changes, committed for travel. Commit: 6b9380b1bd6fe64cf324cbbd134f40a1f50e04e1 https://github.com/modula3/cm3/commit/6b9380b1bd6fe64cf324cbbd134f40a1f50e04e1 Author: ?Rodney Date: 2021-04-11 (Sun, 11 Apr 2021) Changed paths: M m3-sys/m3front/src/builtinOps/New.m3 Log Message: ----------- Fix long-standing bug in New, uncovered by better type checking in ArrayExpr. This is a bit complicated. The bug in New.m3 occurred when allocating an object with field initializatons. It was checking type assignability of an initializing expression against the wrong object field. This was originally covered up by incomplete checking when the expression was an array constructor. Recent completion of this checking uncovered the bug, with a symptom of false error messages, which then were not recovered from, resulting in massive internal CG failures. Note: The code being compiled does not have type assignabilty errors. The newly emitted messages were bogus, as result of the field mismatch. It is possible that the same error message, when produced legitimately, still fails to recover. Commit: 0244a9c447a176493eb4677da8b0fae5a28da3a4 https://github.com/modula3/cm3/commit/0244a9c447a176493eb4677da8b0fae5a28da3a4 Author: ?Rodney Date: 2021-04-12 (Mon, 12 Apr 2021) Changed paths: M m3-sys/m3front/src/exprs/ArrayExpr.m3 Log Message: ----------- Fix assertion failure compiling array constructor '..' as inner constructor. Outer array constructor has '..', and its last real element is an inner array constructor. Needed to handle this case through direct array constructor recursion, rather than general expression recursion. This was already being done with non-'..' cases. Note: There was a booby trap here, with both a local varible and a WITH-binding of the same name 'argExpr'. This initially led me to think they were the same, leading to a severely broken compiler after recompiling itself. Renamed these to remove the confusion. Commit: 0aafad098228d796b181cc013f0b0953a8842e1c https://github.com/modula3/cm3/commit/0aafad098228d796b181cc013f0b0953a8842e1c Author: Rodney Bates Date: 2021-04-16 (Fri, 16 Apr 2021) Changed paths: M m3-comm/sharedobj/src/sharedobj.tmpl M m3-comm/tcp/src/common/IP_c.c M m3-comm/udp/src/Common/UDP_c.c M m3-demo/mentor/src/sorting/audio/MidiLineServerC.c M m3-libs/libm3/src/os/POSIX/FSPosixC.c M m3-libs/libm3/src/os/POSIX/FilePosixC.c M m3-libs/libm3/src/os/POSIX/OSConfigPosixC.c M m3-libs/libm3/src/os/POSIX/SocketPosixC.c M m3-libs/libm3/src/uid/POSIX/MachineIDPosixC.c M m3-libs/libm3/src/uid/WIN32/MachineID.c M m3-libs/m3core/src/C/Common/CerrnoC.c M m3-libs/m3core/src/C/Common/CsetjmpC.c M m3-libs/m3core/src/C/Common/CsignalC.c M m3-libs/m3core/src/C/Common/CstdioC.c M m3-libs/m3core/src/C/Common/CstdlibC.c M m3-libs/m3core/src/C/Common/CstringC.c M m3-libs/m3core/src/Csupport/Common/hand.c M m3-libs/m3core/src/Csupport/big-endian/dtoa.c M m3-libs/m3core/src/Csupport/little-endian/dtoa.c M m3-libs/m3core/src/coroutine/UCONTEXT/ContextC.c M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/runtime/POSIX/RTOSc.c M m3-libs/m3core/src/runtime/POSIX/RTSignalC.c M m3-libs/m3core/src/runtime/POSIX/test_signal.c M m3-libs/m3core/src/runtime/SOLsun/RTStackC.c M m3-libs/m3core/src/runtime/WIN32/RTOSc.c M m3-libs/m3core/src/runtime/common/RTBuiltin.c M m3-libs/m3core/src/runtime/common/RTIOc.c M m3-libs/m3core/src/runtime/common/RTLinkerC.c M m3-libs/m3core/src/runtime/common/RTMiscC.c M m3-libs/m3core/src/runtime/common/RTProcessC.c M m3-libs/m3core/src/runtime/common/RTUntracedMemoryC.c M m3-libs/m3core/src/runtime/ex_frame/RTStackC.c M m3-libs/m3core/src/thread/Common/ThreadDebug.c M m3-libs/m3core/src/thread/Common/ThreadInternal.c M m3-libs/m3core/src/thread/POSIX/ThreadPosixC.c M m3-libs/m3core/src/thread/POSIX/test_timer.c M m3-libs/m3core/src/thread/PTHREAD/ThreadApple.c M m3-libs/m3core/src/thread/PTHREAD/ThreadFreeBSD.c M m3-libs/m3core/src/thread/PTHREAD/ThreadOpenBSD.c M m3-libs/m3core/src/thread/PTHREAD/ThreadPThreadC.c M m3-libs/m3core/src/thread/PTHREAD/interix_context_on_signal_stack.c M m3-libs/m3core/src/thread/PTHREAD/test_interix.c M m3-libs/m3core/src/time/POSIX/DatePosixC.c M m3-libs/m3core/src/time/POSIX/TimePosixC.c M m3-libs/m3core/src/unix/Common/Uconstants.c M m3-libs/m3core/src/unix/Common/UdirC.c M m3-libs/m3core/src/unix/Common/Uexec.c M m3-libs/m3core/src/unix/Common/Ugrp.c M m3-libs/m3core/src/unix/Common/Uin.c M m3-libs/m3core/src/unix/Common/Umman.c M m3-libs/m3core/src/unix/Common/Unetdb.c M m3-libs/m3core/src/unix/Common/UnixC.c M m3-libs/m3core/src/unix/Common/UnixLink.c M m3-libs/m3core/src/unix/Common/Uprocess.c M m3-libs/m3core/src/unix/Common/Upwd.c M m3-libs/m3core/src/unix/Common/Usignal.c M m3-libs/m3core/src/unix/Common/Usocket.c M m3-libs/m3core/src/unix/Common/UstatC.c M m3-libs/m3core/src/unix/Common/UtimeC.c M m3-libs/m3core/src/unix/Common/Uugid.c M m3-libs/m3core/src/unix/Common/Uuio.c M m3-libs/m3core/src/unix/Common/Uutmp.c M m3-libs/rtinfo/src/RTInfo.m3 M m3-obliq/obliqrt/src/Obliq.m3 M m3-scheme/modula3scheme/src/m3makefile M m3-scheme/mscheme-interactive/src/m3makefile M m3-scheme/mscheme-interactive_r/src/m3makefile M m3-scheme/mscheme/src/m3makefile M m3-scheme/mscheme/src/m3overrides M m3-scheme/schemereadline/src/m3makefile M m3-sys/cm3/src/Builder.m3 M m3-sys/cm3/src/M3Backend.i3 M m3-sys/cm3/src/M3Backend.m3 M m3-sys/cm3/src/Main.m3 M m3-sys/cm3/src/Makefile.m3 M m3-sys/cm3/src/Version.c M m3-sys/cminstall/src/config-no-install/NT.common M m3-sys/fix_nl/src/OS.c M m3-sys/m3back/src/M3C.i3 M m3-sys/m3back/src/M3C.m3 M m3-sys/m3back/src/M3CC.c R m3-sys/m3back/src/m3core.h M m3-sys/m3cgcat/src/Main.m3 M m3-sys/m3cggen/src/Main.m3 M m3-sys/m3front/src/values/Module.m3 M m3-sys/m3middle/src/CoffTime.c M m3-sys/m3middle/src/CoffTime.i3 M m3-sys/m3middle/src/M3CG_Binary.i3 M m3-sys/m3middle/src/M3CG_Binary.m3 M m3-sys/m3middle/src/POSIX/CoffTime.c R m3-sys/m3middle/src/POSIX/m3core.h M m3-sys/m3quake/src/MxConfig.i3 M m3-sys/m3quake/src/MxConfig.m3 M m3-sys/m3quake/src/MxConfigC.c M m3-sys/m3quake/src/MxConfigC.i3 M m3-sys/m3quake/src/m3makefile M m3-sys/mklib/src/Main.m3 M m3-sys/windowsResources/src/m3overrides M m3-tools/cvsup/suplib/src/FileAttrC.c M m3-tools/cvsup/suplib/src/FreeBSD/UProcTitle.c M m3-tools/cvsup/suplib/src/UnixMiscC.c M m3-tools/cvsup/suplib/src/libglob/fnmatch.c M m3-tools/cvsup/suplib/src/libmd/md5c.c M m3-tools/cvsup/suplib/src/libmd/md5hl.c M m3-tools/cvsup/suptcp/src/common/SupErrnoC.c M scripts/python/pylib.py Log Message: ----------- Merge branch 'master' of https://github.com/modula3/cm3 Bring in unrelated changes after fixes to New.m3 and ArrayExpr.m3 Commit: f6a85fd5865cb46e970b4b3ba06dbe285dc71455 https://github.com/modula3/cm3/commit/f6a85fd5865cb46e970b4b3ba06dbe285dc71455 Author: Rodney Bates Date: 2021-04-16 (Fri, 16 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/unix/Common/Umman.c Log Message: ----------- Merge branch 'master' of https://github.com/modula3/cm3 Commit: 40e9f88700b3971a22c8a63272195ac11174e1cd https://github.com/modula3/cm3/commit/40e9f88700b3971a22c8a63272195ac11174e1cd Author: Rodney Bates Date: 2021-04-17 (Sat, 17 Apr 2021) Changed paths: M m3-sys/m3front/src/misc/CG.i3 M m3-sys/m3front/src/misc/CG.m3 M m3-sys/m3front/src/values/Variable.m3 Log Message: ----------- Further unravel value alignments vs. address alignments. Fix cases failing when variables are accessed indirectly, and alignment of addresses is smaller than that of operands. e.g., LONGINT, LONGCARD, REAL, etc., on 32-bit hosts. The address's alignment was leaking back into the operands. Commit: 1e7faa9705f0101dd6253477eee3d955ac19c05d https://github.com/modula3/cm3/commit/1e7faa9705f0101dd6253477eee3d955ac19c05d Author: Rodney Bates Date: 2021-04-17 (Sat, 17 Apr 2021) Changed paths: M m3-libs/m3core/src/thread/m3makefile M m3-libs/m3core/src/unix/Common/Uconstants.c M m3-libs/m3core/src/win32/WinNT.i3 M m3-libs/sysutils/src/FSUtils.i3 R m3-libs/sysutils/src/POSIX/FSUnixC.c M m3-libs/sysutils/src/POSIX/FSUnix_cm3.m3 M m3-libs/sysutils/src/POSIX/FSUnix_pm3.m3 M m3-libs/sysutils/src/POSIX/SystemPosix.m3 R m3-libs/sysutils/src/POSIX/SystemPosixC.c M m3-libs/sysutils/src/POSIX/m3makefile M m3-libs/sysutils/src/System.i3 R m3-libs/sysutils/src/SystemC.c M m3-libs/sysutils/src/WIN32/FSysWin32.m3 M m3-libs/sysutils/src/m3makefile M m3-sys/m3back/src/M3C.m3 M m3-sys/mklib/src/Main.m3 M m3-ui/anim3D/src/x-opengl/X_OpenGL_Base.m3 Log Message: ----------- Merge branch 'master' of https://github.com/modula3/cm3 Compare: https://github.com/modula3/cm3/compare/a2d778acee07...1e7faa9705f0 From noreply at github.com Sun Apr 18 09:20:32 2021 From: noreply at github.com (Jay Krell) Date: Sun, 18 Apr 2021 00:20:32 -0700 Subject: [M3commit] [modula3/cm3] 1c23de: m3c: Remove some debugprint. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 1c23de8a6c2660e6cfd5414965cc2b685d54f5f2 https://github.com/modula3/cm3/commit/1c23de8a6c2660e6cfd5414965cc2b685d54f5f2 Author: JayKrell Date: 2021-04-18 (Sun, 18 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Remove some debugprint. From noreply at github.com Tue Apr 20 08:30:14 2021 From: noreply at github.com (Jay Krell) Date: Mon, 19 Apr 2021 23:30:14 -0700 Subject: [M3commit] [modula3/cm3] 6ddd4a: m3core: undef M3MODULE before defining it, Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 6ddd4aa424baa9a132de4b0df65db0fb4a734c39 https://github.com/modula3/cm3/commit/6ddd4aa424baa9a132de4b0df65db0fb4a734c39 Author: JayKrell Date: 2021-04-19 (Mon, 19 Apr 2021) Changed paths: M m3-libs/m3core/src/C/Common/CsignalC.c M m3-libs/m3core/src/C/Common/CstdioC.c M m3-libs/m3core/src/C/Common/CstdlibC.c M m3-libs/m3core/src/C/Common/CstringC.c M m3-libs/m3core/src/thread/PTHREAD/ThreadPThreadC.c M m3-libs/m3core/src/unix/Common/UdirC.c M m3-libs/m3core/src/unix/Common/Uexec.c M m3-libs/m3core/src/unix/Common/Umman.c M m3-libs/m3core/src/unix/Common/UnixC.c M m3-libs/m3core/src/unix/Common/Uprocess.c M m3-libs/m3core/src/unix/Common/Upwd.c M m3-libs/m3core/src/unix/Common/Usignal.c M m3-libs/m3core/src/unix/Common/Usocket.c M m3-libs/m3core/src/unix/Common/Uugid.c M m3-libs/m3core/src/unix/Common/Uuio.c M m3-libs/m3core/src/unix/Common/Uutmp.c Log Message: ----------- m3core: undef M3MODULE before defining it, so that multiple .c files can be concatenated without warning/error From noreply at github.com Tue Apr 20 08:44:09 2021 From: noreply at github.com (Jay Krell) Date: Mon, 19 Apr 2021 23:44:09 -0700 Subject: [M3commit] [modula3/cm3] fe1c78: m3core: undef THOUSAND to support concat Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: fe1c78ac4ccb185df7e904e8be4bd5be8e8e1336 https://github.com/modula3/cm3/commit/fe1c78ac4ccb185df7e904e8be4bd5be8e8e1336 Author: JayKrell Date: 2021-04-19 (Mon, 19 Apr 2021) Changed paths: M m3-libs/m3core/src/thread/Common/ThreadInternal.c M m3-libs/m3core/src/time/POSIX/TimePosixC.c Log Message: ----------- m3core: undef THOUSAND to support concat This is maybe not best solution -- try to find a shared definition or unshare the name. From noreply at github.com Tue Apr 20 11:26:24 2021 From: noreply at github.com (Jay Krell) Date: Tue, 20 Apr 2021 02:26:24 -0700 Subject: [M3commit] [modula3/cm3] 90a927: cm3: Use Unix.link instead of local Utils.link that Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 90a927f69fdb1a1f303d82c428d60b0a4ad75c19 https://github.com/modula3/cm3/commit/90a927f69fdb1a1f303d82c428d60b0a4ad75c19 Author: JayKrell Date: 2021-04-20 (Tue, 20 Apr 2021) Changed paths: M m3-sys/cm3/src/Utils.i3 M m3-sys/cm3/src/UtilsWin32.m3 R m3-sys/cm3/src/cm3unix.c M m3-sys/cm3/src/m3makefile Log Message: ----------- cm3: Use Unix.link instead of local Utils.link that was an automated copy of it. This is an old bootstrapping hack that inhibits concating files. From noreply at github.com Tue Apr 20 11:31:55 2021 From: noreply at github.com (Jay Krell) Date: Tue, 20 Apr 2021 02:31:55 -0700 Subject: [M3commit] [modula3/cm3] 32b792: m3core: Require Windows 2000. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 32b7925c16bbde2d7db2475e41c92877980812e4 https://github.com/modula3/cm3/commit/32b7925c16bbde2d7db2475e41c92877980812e4 Author: JayKrell Date: 2021-04-20 (Tue, 20 Apr 2021) Changed paths: M m3-libs/m3core/src/unix/Common/UnixLink.c Log Message: ----------- m3core: Require Windows 2000. Call CreateHardLink more directly, requiring Windows 2000 or newer. From noreply at github.com Tue Apr 20 11:42:07 2021 From: noreply at github.com (Jay Krell) Date: Tue, 20 Apr 2021 02:42:07 -0700 Subject: [M3commit] [modula3/cm3] 74a5e4: m3core: Add type names to static native_to_m3 to a... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 74a5e4734c3fa11baa5291eba3e4520a91710027 https://github.com/modula3/cm3/commit/74a5e4734c3fa11baa5291eba3e4520a91710027 Author: JayKrell Date: 2021-04-20 (Tue, 20 Apr 2021) Changed paths: M m3-libs/m3core/src/unix/Common/Ugrp.c M m3-libs/m3core/src/unix/Common/Unetdb.c Log Message: ----------- m3core: Add type names to static native_to_m3 to allow concat files. From noreply at github.com Thu Apr 22 08:38:49 2021 From: noreply at github.com (Jay Krell) Date: Wed, 21 Apr 2021 23:38:49 -0700 Subject: [M3commit] [modula3/cm3] 259fc0: m3core.h: Add comment confirming sizeof(size_t) !=... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 259fc0842f8d2d9c09aa21cf107f4c2ca2e5fa64 https://github.com/modula3/cm3/commit/259fc0842f8d2d9c09aa21cf107f4c2ca2e5fa64 Author: JayKrell Date: 2021-04-21 (Wed, 21 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Add comment confirming sizeof(size_t) != sizeof(void*) on VMS. From noreply at github.com Thu Apr 22 08:41:11 2021 From: noreply at github.com (Jay Krell) Date: Wed, 21 Apr 2021 23:41:11 -0700 Subject: [M3commit] [modula3/cm3] 4b1b7a: m3core.h: Add todo comment to use C99 Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 4b1b7a7c1caec20cffb71f597092ff1660fb8280 https://github.com/modula3/cm3/commit/4b1b7a7c1caec20cffb71f597092ff1660fb8280 Author: JayKrell Date: 2021-04-21 (Wed, 21 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Add todo comment to use C99 Commit: e0f75d12af6c4fc8fc1bf83e7ce9d5f3344ad7ca https://github.com/modula3/cm3/commit/e0f75d12af6c4fc8fc1bf83e7ce9d5f3344ad7ca Author: JayKrell Date: 2021-04-21 (Wed, 21 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Fix to Unix newline. Compare: https://github.com/modula3/cm3/compare/259fc0842f8d...e0f75d12af6c From noreply at github.com Thu Apr 22 08:42:25 2021 From: noreply at github.com (Jay Krell) Date: Wed, 21 Apr 2021 23:42:25 -0700 Subject: [M3commit] [modula3/cm3] 23323c: m3c: Add comment about integer types vs. m3core.h, ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 23323c7feca1a78799878b4935d49e172e65d1d1 https://github.com/modula3/cm3/commit/23323c7feca1a78799878b4935d49e172e65d1d1 Author: JayKrell Date: 2021-04-21 (Wed, 21 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Add comment about integer types vs. m3core.h, limits.h, C99 From noreply at github.com Thu Apr 22 12:43:54 2021 From: noreply at github.com (peter mckinna) Date: Thu, 22 Apr 2021 03:43:54 -0700 Subject: [M3commit] [modula3/cm3] 28ccb8: Add nowarn to unreachable Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 28ccb8f17c8da3e51a773400f80f21f2eca39bbf https://github.com/modula3/cm3/commit/28ccb8f17c8da3e51a773400f80f21f2eca39bbf Author: peter mckinna Date: 2021-04-22 (Thu, 22 Apr 2021) Changed paths: M m3-libs/m3core/src/runtime/common/RTHeapInfo.m3 Log Message: ----------- Add nowarn to unreachable Commit: 8426e3830edfa5cf9e9dddb2d614f9b7af7b3e7e https://github.com/modula3/cm3/commit/8426e3830edfa5cf9e9dddb2d614f9b7af7b3e7e Author: peter mckinna Date: 2021-04-22 (Thu, 22 Apr 2021) Changed paths: M m3-ui/mg/src/MGPaintOp.m3 Log Message: ----------- Add xrgb - part of fontconfig upgrade Commit: 6a57c90f23a832198d25ef3381a59bd89558f6ed https://github.com/modula3/cm3/commit/6a57c90f23a832198d25ef3381a59bd89558f6ed Author: peter mckinna Date: 2021-04-22 (Thu, 22 Apr 2021) Changed paths: M m3-db/mysql/src/MySQL.i3 M m3-db/mysql/src/MySQL.m3 M m3-db/mysql/src/MySQLRaw.i3 M m3-db/mysql/src/MySQLRaw.m3 A m3-db/mysql/src/build M m3-db/mysql/src/m3makefile A m3-db/mysql/src/mysql.i A m3-db/mysql/src/mysql_wrap.c A m3-db/mysql/src/mysqlmaps.i M m3-db/mysql/test/src/Main.m3 M m3-db/mysql/test/src/m3makefile Log Message: ----------- Update mysql bindings to latest(sic) version Compare: https://github.com/modula3/cm3/compare/23323c7feca1...6a57c90f23a8 From noreply at github.com Fri Apr 23 01:25:11 2021 From: noreply at github.com (Jay Krell) Date: Thu, 22 Apr 2021 16:25:11 -0700 Subject: [M3commit] [modula3/cm3] c07e3c: m3c/m3core.h: ifndef guards around INTEGER and WORD_T Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: c07e3cd0ea52c2c6def6643bd69d93c42f0b8526 https://github.com/modula3/cm3/commit/c07e3cd0ea52c2c6def6643bd69d93c42f0b8526 Author: JayKrell Date: 2021-04-22 (Thu, 22 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c/m3core.h: ifndef guards around INTEGER and WORD_T so m3c output can be concatented with hand written .c and .h m3c and m3core.h do not necessarily agree on the exact type of INTEGER and WORD_T. m3core.h pays to get ptrdiff_t/size_t defined and uses them, except in VMS corner case. m3c provides int32 and int64 and lets m3front/m3middle decide. These should always end up the same size, but could be different types, like int vs. 32bit long, or long long vs. 64bit long. From noreply at github.com Fri Apr 23 01:36:46 2021 From: noreply at github.com (Jay Krell) Date: Thu, 22 Apr 2021 16:36:46 -0700 Subject: [M3commit] [modula3/cm3] 552d61: m3core.h: Add more comments about VMS. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 552d6117c30bbb77a1f7018493d69c6319dc9471 https://github.com/modula3/cm3/commit/552d6117c30bbb77a1f7018493d69c6319dc9471 Author: JayKrell Date: 2021-04-22 (Thu, 22 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h Log Message: ----------- m3core.h: Add more comments about VMS. From noreply at github.com Sun Apr 25 08:14:41 2021 From: noreply at github.com (Jay Krell) Date: Sat, 24 Apr 2021 23:14:41 -0700 Subject: [M3commit] [modula3/cm3] b5be20: m3c: Tweak the ifndefs. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: b5be209ef5eb30fa61923d8a194fcf3027562aab https://github.com/modula3/cm3/commit/b5be209ef5eb30fa61923d8a194fcf3027562aab Author: JayKrell Date: 2021-04-24 (Sat, 24 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Tweak the ifndefs. - Use supplies the identifier, instead of putting m3once on it. - Remove some unnecessary. - Converge some symbols m3_foo vs. ifdef m3foo. More ifndefs are likely coming, in particular because all the hand written functions are difficult to replicate with matching signature in Modula-3 CG, as Modula-3 CG does not have equivalent of C char. So everyone will probably be defined and import will wrap declarations with ifndef. From noreply at github.com Sun Apr 25 08:17:23 2021 From: noreply at github.com (Jay Krell) Date: Sat, 24 Apr 2021 23:17:23 -0700 Subject: [M3commit] [modula3/cm3] ef8ff3: m3front: When reference types have a user name, pa... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: ef8ff3e2028ba20eff8fb37cbf8d4214873a61eb https://github.com/modula3/cm3/commit/ef8ff3e2028ba20eff8fb37cbf8d4214873a61eb Author: JayKrell Date: 2021-04-24 (Sat, 24 Apr 2021) Changed paths: M m3-sys/m3front/src/types/RefType.m3 Log Message: ----------- m3front: When reference types have a user name, pass it to declare_typename. This should greatly help let m3c and C prototypes match. I don't want type hashes in C. From noreply at github.com Sun Apr 25 18:39:31 2021 From: noreply at github.com (Jay Krell) Date: Sun, 25 Apr 2021 09:39:31 -0700 Subject: [M3commit] [modula3/cm3] e3b113: Remove m3middle from netobj due to duplicate TextU... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: e3b1133694ac27fad286b006dcbe1823ed06b60e https://github.com/modula3/cm3/commit/e3b1133694ac27fad286b006dcbe1823ed06b60e Author: JayKrell Date: 2021-04-25 (Sun, 25 Apr 2021) Changed paths: A m3-comm/netobj/src/Target.i3 M m3-comm/netobj/src/m3makefile Log Message: ----------- Remove m3middle from netobj due to duplicate TextUtils. From noreply at github.com Mon Apr 26 02:30:25 2021 From: noreply at github.com (=?UTF-8?B?TWlrYSBOeXN0csO2bQ==?=) Date: Sun, 25 Apr 2021 17:30:25 -0700 Subject: [M3commit] [modula3/cm3] 26c13c: rename Target to NetObjTarget to avoid name clash ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 26c13c5545053f65e13169aa23ff010e8b07f69f https://github.com/modula3/cm3/commit/26c13c5545053f65e13169aa23ff010e8b07f69f Author: mikanystrom Date: 2021-04-25 (Sun, 25 Apr 2021) Changed paths: A m3-comm/netobj/src/NetObjTarget.i3 R m3-comm/netobj/src/Target.i3 M m3-comm/netobj/src/m3makefile M m3-comm/netobj/src/netobjrt/StubLib.m3 Log Message: ----------- rename Target to NetObjTarget to avoid name clash in Obliq From noreply at github.com Mon Apr 26 02:31:10 2021 From: noreply at github.com (=?UTF-8?B?TWlrYSBOeXN0csO2bQ==?=) Date: Sun, 25 Apr 2021 17:31:10 -0700 Subject: [M3commit] [modula3/cm3] cf4b01: take out csv and csvalign until compiler is repaired Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: cf4b0142db40bcd7b38ca392695f5e83e2126186 https://github.com/modula3/cm3/commit/cf4b0142db40bcd7b38ca392695f5e83e2126186 Author: mikanystrom Date: 2021-04-25 (Sun, 25 Apr 2021) Changed paths: M scripts/pkginfo.txt Log Message: ----------- take out csv and csvalign until compiler is repaired From noreply at github.com Mon Apr 26 08:33:49 2021 From: noreply at github.com (Jay Krell) Date: Sun, 25 Apr 2021 23:33:49 -0700 Subject: [M3commit] [modula3/cm3] d7dc29: netobj: Move two lines from NetObjTarget to m3core... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: d7dc2902e46287e620b2051a9a2da16138087365 https://github.com/modula3/cm3/commit/d7dc2902e46287e620b2051a9a2da16138087365 Author: JayKrell Date: 2021-04-25 (Sun, 25 Apr 2021) Changed paths: R m3-comm/netobj/src/NetObjTarget.i3 M m3-comm/netobj/src/m3makefile M m3-comm/netobj/src/netobjrt/StubLib.m3 M m3-libs/m3core/src/types/WideChar.i3 Log Message: ----------- netobj: Move two lines from NetObjTarget to m3core\src\types\WideChar.i3. todo: Remove from Target. From noreply at github.com Mon Apr 26 09:05:16 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 00:05:16 -0700 Subject: [M3commit] [modula3/cm3] 0b7caa: netobj: Target. to WideChar. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 0b7caa2001b7a985442466dcff468f6b4a0e072f https://github.com/modula3/cm3/commit/0b7caa2001b7a985442466dcff468f6b4a0e072f Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-comm/netobj/src/netobjrt/StubLib.m3 Log Message: ----------- netobj: Target. to WideChar. From noreply at github.com Mon Apr 26 19:48:26 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 10:48:26 -0700 Subject: [M3commit] [modula3/cm3] 30371b: DatePosix: 0 vs. {0} Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 30371bb2be5f029b0e133dab257ea088b0f245e1 https://github.com/modula3/cm3/commit/30371bb2be5f029b0e133dab257ea088b0f245e1 Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: 0 vs. {0} Maybe {0} is not worth genericness and confusing people. From noreply at github.com Mon Apr 26 19:49:01 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 10:49:01 -0700 Subject: [M3commit] [modula3/cm3] 7cba2b: DatePosix: Tabs to spaces. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7cba2b8d19b186e4bc3892cc9a3ab4aad94eed40 https://github.com/modula3/cm3/commit/7cba2b8d19b186e4bc3892cc9a3ab4aad94eed40 Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: Tabs to spaces. From noreply at github.com Mon Apr 26 19:50:01 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 10:50:01 -0700 Subject: [M3commit] [modula3/cm3] 77b72d: DatePosix: Shorter comment. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 77b72d1d0bf306c79685f0a23bbca0e59d09f18f https://github.com/modula3/cm3/commit/77b72d1d0bf306c79685f0a23bbca0e59d09f18f Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: Shorter comment. From noreply at github.com Mon Apr 26 19:50:55 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 10:50:55 -0700 Subject: [M3commit] [modula3/cm3] 21746c: DatePosix: ptrdiff_t should be INTEGER. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 21746cbcf1f0bc2cc5444460e9c3d040e3029a03 https://github.com/modula3/cm3/commit/21746cbcf1f0bc2cc5444460e9c3d040e3029a03 Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: ptrdiff_t should be INTEGER. (Maybe int in future). From noreply at github.com Mon Apr 26 19:52:55 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 10:52:55 -0700 Subject: [M3commit] [modula3/cm3] b7fb91: DatePosix: Remove const for now. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: b7fb91c4c423c302cc05997518b6b814d35a997a https://github.com/modula3/cm3/commit/b7fb91c4c423c302cc05997518b6b814d35a997a Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: Remove const for now. Initial typename support will not support it. From noreply at github.com Mon Apr 26 20:11:47 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 11:11:47 -0700 Subject: [M3commit] [modula3/cm3] 41731f: m3c: Add declare_typename implementation. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 41731fc566bfc06f61e133d7d839e37165b6896b https://github.com/modula3/cm3/commit/41731fc566bfc06f61e133d7d839e37165b6896b Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Add declare_typename implementation. This is so C and m3c output can be concatenated. Tweak how text is turned into valid types. Simpler but perhaps too fragile/ambiguous. We'll see. From noreply at github.com Mon Apr 26 20:13:23 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 11:13:23 -0700 Subject: [M3commit] [modula3/cm3] 8336eb: m3c: Tweak how text is turned into valid types. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 8336eb1c41fcc703085c1d6b05a47a7d0bfcb0d7 https://github.com/modula3/cm3/commit/8336eb1c41fcc703085c1d6b05a47a7d0bfcb0d7 Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Tweak how text is turned into valid types. Simpler but perhaps too fragile/ambiguous. We'll see. From noreply at github.com Mon Apr 26 20:15:13 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 11:15:13 -0700 Subject: [M3commit] [modula3/cm3] 673114: m3c: Fix some newlines. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 6731147ff606124025d130be4ebf318dcb0e6394 https://github.com/modula3/cm3/commit/6731147ff606124025d130be4ebf318dcb0e6394 Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Fix some newlines. From noreply at github.com Mon Apr 26 20:18:55 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 11:18:55 -0700 Subject: [M3commit] [modula3/cm3] 7f7715: m3c: Change a double underscore to single. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7f77155c94001b65f020a861f66cb106cb7fbc0d https://github.com/modula3/cm3/commit/7f77155c94001b65f020a861f66cb106cb7fbc0d Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Change a double underscore to single. From noreply at github.com Mon Apr 26 20:30:47 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 11:30:47 -0700 Subject: [M3commit] [modula3/cm3] eaf1bf: m3c: Add more "address" typedefs. e.g. MUTEX. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: eaf1bfc81616dfae0794ab9a2198816645cf7ffd https://github.com/modula3/cm3/commit/eaf1bfc81616dfae0794ab9a2198816645cf7ffd Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Add more "address" typedefs. e.g. MUTEX. From noreply at github.com Mon Apr 26 22:36:13 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 13:36:13 -0700 Subject: [M3commit] [modula3/cm3] a06348: DatePosix: Remove branded pointer from C interface. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: a06348cbcf4c92fdaa435e92ac18825f05e70f67 https://github.com/modula3/cm3/commit/a06348cbcf4c92fdaa435e92ac18825f05e70f67 Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/time/POSIX/DatePosix.i3 M m3-libs/m3core/src/time/POSIX/DatePosix.m3 M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: Remove branded pointer from C interface. Out of ignorance of the representation. It does not even need to be a pointer, make it INTEGER. From noreply at github.com Mon Apr 26 22:38:32 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 13:38:32 -0700 Subject: [M3commit] [modula3/cm3] a24a2e: DatePosix: Change define Local and UTC to local co... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: a24a2e4853db20432d6901c472fbb7841c0707ee https://github.com/modula3/cm3/commit/a24a2e4853db20432d6901c472fbb7841c0707ee Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: Change define Local and UTC to local const INTEGER. From noreply at github.com Mon Apr 26 22:45:32 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 13:45:32 -0700 Subject: [M3commit] [modula3/cm3] bfd4e5: m3core: Rename m3_time_t to m3_time64_t. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: bfd4e5aec76daa0539b577cadffed39f0cb2ca0a https://github.com/modula3/cm3/commit/bfd4e5aec76daa0539b577cadffed39f0cb2ca0a Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/unix/Common/UtimeC.c Log Message: ----------- m3core: Rename m3_time_t to m3_time64_t. m3_time64_t is always 64bits. m3_time64_t is always at least the size of time_t. m3_time_t shall be repurposed to mean time_t or time64_t, whichever the C system supports. Many 32bit systems started with 32bit time_t and then later added 64bit time64_t or such. Sometimes both are available at the same time, sometimes defines silently changes time_t to time64_t, which is easier to program, if you do not need a mix. From the looks of it, this is all for OSF/1. From noreply at github.com Mon Apr 26 22:54:26 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 13:54:26 -0700 Subject: [M3commit] [modula3/cm3] 5ba0f4: m3core: Factor _TIME64_T somewhat via m3_time_t. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 5ba0f460ec112770d58f59e7d9ab41e01267e98c https://github.com/modula3/cm3/commit/5ba0f460ec112770d58f59e7d9ab41e01267e98c Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/time/POSIX/DatePosixC.c M m3-libs/m3core/src/unix/Common/UtimeC.c Log Message: ----------- m3core: Factor _TIME64_T somewhat via m3_time_t. From noreply at github.com Mon Apr 26 23:01:18 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 14:01:18 -0700 Subject: [M3commit] [modula3/cm3] 427574: DatePosix: Spell out "sec" as "seconds". Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 42757407dcba24ac6bd9f804799f833fc07774f6 https://github.com/modula3/cm3/commit/42757407dcba24ac6bd9f804799f833fc07774f6 Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: Spell out "sec" as "seconds". From noreply at github.com Mon Apr 26 23:10:19 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 14:10:19 -0700 Subject: [M3commit] [modula3/cm3] 799b27: DatePosix: Spell out SecsPerHour as SecondsPerHour. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 799b27d9204a576b2b301f91c8d5ead458abdf13 https://github.com/modula3/cm3/commit/799b27d9204a576b2b301f91c8d5ead458abdf13 Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: Spell out SecsPerHour as SecondsPerHour. Commit: 1c528db84b3752d9040b12ca39db4980f9d7b6ba https://github.com/modula3/cm3/commit/1c528db84b3752d9040b12ca39db4980f9d7b6ba Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: Shorten comment. Compare: https://github.com/modula3/cm3/compare/42757407dcba...1c528db84b37 From noreply at github.com Tue Apr 27 00:20:14 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 15:20:14 -0700 Subject: [M3commit] [modula3/cm3] 88adea: DatePosix: Remove managed types from C. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 88adeabd415fc80eca25fd63acc52e90257bb7dc https://github.com/modula3/cm3/commit/88adeabd415fc80eca25fd63acc52e90257bb7dc Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/time/POSIX/DatePosix.i3 M m3-libs/m3core/src/time/POSIX/DatePosix.m3 M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- DatePosix: Remove managed types from C. This is ultimately needed for precise GC with cooperative suspend (unless, something unspecified around finding managed values in C locals) This work will be approximately finished when TEXT and MtoC is removed from m3core.h, since those are the mainy/only infractions. From noreply at github.com Tue Apr 27 00:23:46 2021 From: noreply at github.com (Jay Krell) Date: Mon, 26 Apr 2021 15:23:46 -0700 Subject: [M3commit] [modula3/cm3] 7d81fc: m3c: Turn dashes into underscores (lost this recen... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7d81fc6d7be85a17c6231c0a775213e0492197f3 https://github.com/modula3/cm3/commit/7d81fc6d7be85a17c6231c0a775213e0492197f3 Author: JayKrell Date: 2021-04-26 (Mon, 26 Apr 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- m3c: Turn dashes into underscores (lost this recently). From noreply at github.com Wed Apr 28 08:33:02 2021 From: noreply at github.com (Jay Krell) Date: Tue, 27 Apr 2021 23:33:02 -0700 Subject: [M3commit] [modula3/cm3] bec43b: m3core: Rename Date_T to m3core_DatePosix_T. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: bec43bf0c19be159ad61d53f916f47d7109d914e https://github.com/modula3/cm3/commit/bec43bf0c19be159ad61d53f916f47d7109d914e Author: JayKrell Date: 2021-04-27 (Tue, 27 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- m3core: Rename Date_T to m3core_DatePosix_T. This should improve concatenating m3c output with C. From noreply at github.com Wed Apr 28 08:34:41 2021 From: noreply at github.com (Jay Krell) Date: Tue, 27 Apr 2021 23:34:41 -0700 Subject: [M3commit] [modula3/cm3] cd597e: DatePosix: Narrow some fields from INTEGER to int. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: cd597e9706889464ebc5d64db62da0fcdd838220 https://github.com/modula3/cm3/commit/cd597e9706889464ebc5d64db62da0fcdd838220 Author: JayKrell Date: 2021-04-27 (Tue, 27 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/time/POSIX/DatePosix.i3 Log Message: ----------- DatePosix: Narrow some fields from INTEGER to int. INTEGER is overkill. From noreply at github.com Wed Apr 28 21:02:46 2021 From: noreply at github.com (Rodney M. Bates) Date: Wed, 28 Apr 2021 12:02:46 -0700 Subject: [M3commit] [modula3/cm3] 4d1894: Remove 2 overzealous assertions involving multi-D ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 4d1894efbb5ac565d59b4b5e8dfa825a5250caed https://github.com/modula3/cm3/commit/4d1894efbb5ac565d59b4b5e8dfa825a5250caed Author: Rodney Bates Date: 2021-04-28 (Wed, 28 Apr 2021) Changed paths: M m3-sys/m3front/src/exprs/ArrayExpr.m3 M m3-sys/m3front/src/misc/CG.i3 Log Message: ----------- Remove 2 overzealous assertions involving multi-D open array constructors. Makes csv and csvalign compile. The failing case was an array constructor with at last two open dimensions but empty between the braces. The assertions expected no dynamic levels when compiling a CONST array, but in this case, the deeper open dimensions are such levels. But they are irrelevant, because compiling the array value will not reach them. In all the other cases, the assertions are both true and needed, but other assertions will assure this in other ways. Commit: cb495e9fc33df516a132cf24c48ea268a489633e https://github.com/modula3/cm3/commit/cb495e9fc33df516a132cf24c48ea268a489633e Author: Rodney Bates Date: 2021-04-28 (Wed, 28 Apr 2021) Changed paths: M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/time/POSIX/DatePosix.i3 M m3-libs/m3core/src/time/POSIX/DatePosixC.c Log Message: ----------- Merge branch 'master' of https://github.com/modula3/cm3 Compare: https://github.com/modula3/cm3/compare/cd597e970688...cb495e9fc33d From noreply at github.com Thu Apr 29 08:14:00 2021 From: noreply at github.com (=?UTF-8?B?TWlrYSBOeXN0csO2bQ==?=) Date: Wed, 28 Apr 2021 23:14:00 -0700 Subject: [M3commit] [modula3/cm3] 7362cc: temporary commit for MySQL fixes, really waiting f... Message-ID: Branch: refs/heads/mika-mysql Home: https://github.com/modula3/cm3 Commit: 7362cc5faa0a395df1b4a6ff5e5d997181ddace2 https://github.com/modula3/cm3/commit/7362cc5faa0a395df1b4a6ff5e5d997181ddace2 Author: mikanystrom Date: 2021-04-28 (Wed, 28 Apr 2021) Changed paths: M m3-db/mysql/src/MySQL.i3 M m3-db/mysql/src/MySQL.m3 M m3-db/mysql/src/MySQLRaw.i3 M m3-db/mysql/src/m3makefile M m3-db/mysql/src/mysql_wrap.c Log Message: ----------- temporary commit for MySQL fixes, really waiting for Peter McK. to make the "real" commit From noreply at github.com Thu Apr 29 08:44:24 2021 From: noreply at github.com (=?UTF-8?B?TWlrYSBOeXN0csO2bQ==?=) Date: Wed, 28 Apr 2021 23:44:24 -0700 Subject: [M3commit] [modula3/cm3] f0ee7a: MPFR interfaces for Modula-3 Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: f0ee7a197b93fedfd62e467d26a6ec48f46adfff https://github.com/modula3/cm3/commit/f0ee7a197b93fedfd62e467d26a6ec48f46adfff Author: mikanystrom Date: 2021-04-28 (Wed, 28 Apr 2021) Changed paths: A caltech-other/mpfr/src/Mpfr.i3 A caltech-other/mpfr/src/Mpfr.m3 A caltech-other/mpfr/src/MpfrC.c A caltech-other/mpfr/src/MpfrP.i3 A caltech-other/mpfr/src/MpfrRoundingMode.i3 A caltech-other/mpfr/src/m3makefile A caltech-other/mpfr/src/m3overrides Log Message: ----------- MPFR interfaces for Modula-3 Commit: aac35f75faf210c1af385963d684f1fd00ec3e9f https://github.com/modula3/cm3/commit/aac35f75faf210c1af385963d684f1fd00ec3e9f Author: mikanystrom Date: 2021-04-28 (Wed, 28 Apr 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/Unix.common Log Message: ----------- add MPFR (multi-precision floating point) bindings Commit: 7b6738bcc7f786e4065d81da1a54596d83314d27 https://github.com/modula3/cm3/commit/7b6738bcc7f786e4065d81da1a54596d83314d27 Author: mikanystrom Date: 2021-04-28 (Wed, 28 Apr 2021) Changed paths: R m3-comm/netobj/src/NetObjTarget.i3 M m3-comm/netobj/src/m3makefile M m3-comm/netobj/src/netobjrt/StubLib.m3 M m3-libs/m3core/src/m3core.h M m3-libs/m3core/src/time/POSIX/DatePosix.i3 M m3-libs/m3core/src/time/POSIX/DatePosix.m3 M m3-libs/m3core/src/time/POSIX/DatePosixC.c M m3-libs/m3core/src/types/WideChar.i3 M m3-libs/m3core/src/unix/Common/UtimeC.c M m3-sys/m3back/src/M3C.m3 M m3-sys/m3front/src/exprs/ArrayExpr.m3 M m3-sys/m3front/src/misc/CG.i3 Log Message: ----------- Merge branch 'master' of https://github.com/modula3/cm3 Compare: https://github.com/modula3/cm3/compare/cb495e9fc33d...7b6738bcc7f7 From noreply at github.com Thu Apr 29 17:08:54 2021 From: noreply at github.com (=?UTF-8?B?TWlrYSBOeXN0csO2bQ==?=) Date: Thu, 29 Apr 2021 08:08:54 -0700 Subject: [M3commit] [modula3/cm3] 29bec5: Revert "take out csv and csvalign until compiler i... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 29bec53dff8a3368be51ae18e49ecde75b19b8de https://github.com/modula3/cm3/commit/29bec53dff8a3368be51ae18e49ecde75b19b8de Author: mikanystrom Date: 2021-04-28 (Wed, 28 Apr 2021) Changed paths: M scripts/pkginfo.txt Log Message: ----------- Revert "take out csv and csvalign until compiler is repaired" This reverts commit cf4b0142db40bcd7b38ca392695f5e83e2126186. Commit: f0a720597c9221efba8ee6831722628492d2fbae https://github.com/modula3/cm3/commit/f0a720597c9221efba8ee6831722628492d2fbae Author: mikanystrom Date: 2021-04-29 (Thu, 29 Apr 2021) Changed paths: R caltech-other/mpfr/src/m3overrides Log Message: ----------- we don't need this file Compare: https://github.com/modula3/cm3/compare/7b6738bcc7f7...f0a720597c92 From noreply at github.com Fri Apr 30 08:22:16 2021 From: noreply at github.com (peter mckinna) Date: Thu, 29 Apr 2021 23:22:16 -0700 Subject: [M3commit] [modula3/cm3] b234cd: Moving maps to m3-db first step Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: b234cd0bf89bcfed010aebb2caa3cbf2e7fc67a7 https://github.com/modula3/cm3/commit/b234cd0bf89bcfed010aebb2caa3cbf2e7fc67a7 Author: peter mckinna Date: 2021-04-30 (Fri, 30 Apr 2021) Changed paths: A m3-db/mysql/src/class/MySQLMaps.i3 A m3-db/mysql/src/class/MySQLMaps.m3 A m3-db/mysql/src/class/m3makefile R m3-db/mysql/swig/mysql.i R m3-db/mysql/swig/mysqlapi.i R m3-db/mysql/swig/mysqlmaps.i R m3-db/mysql/swig/mysqltypes.i Log Message: ----------- Moving maps to m3-db first step Commit: b6e50a6dff91ffddd697292583c279fa9739266c https://github.com/modula3/cm3/commit/b6e50a6dff91ffddd697292583c279fa9739266c Author: peter mckinna Date: 2021-04-30 (Fri, 30 Apr 2021) Changed paths: M m3-db/db/src/mysqldb/MySQLDB.m3 M m3-db/mysql/src/MySQL.i3 M m3-db/mysql/src/m3makefile M m3-db/mysql/src/mysql.i M m3-db/mysql/src/mysqlmaps.i M m3-db/mysql/test/src/Main.m3 M m3-sys/cminstall/src/config-no-install/Linux.common M m3-sys/cminstall/src/config-no-install/Unix.common Log Message: ----------- Fix MySQLDB.m3 to work with upgraded mysql. Add field type definitions to mysql and update the system-libs to use mysql only for linux at moment Compare: https://github.com/modula3/cm3/compare/f0a720597c92...b6e50a6dff91