From noreply at github.com Sun Feb 21 10:55:57 2021 From: noreply at github.com (Jay Krell) Date: Sun, 21 Feb 2021 01:55:57 -0800 Subject: [M3commit] [modula3/cm3] 129229: Start fixing C backend. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 129229fbee7963303547f10debae104d91bc4125 https://github.com/modula3/cm3/commit/129229fbee7963303547f10debae104d91bc4125 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- Start fixing C backend. _setjmp is declared incorrectly. This particularly breaks NT targets such as AMD64_NT. You must include setjmp.h. You cannot declare setjmp yourself. From noreply at github.com Sun Feb 21 11:08:15 2021 From: noreply at github.com (Jay Krell) Date: Sun, 21 Feb 2021 02:08:15 -0800 Subject: [M3commit] [modula3/cm3] 2bf136: Fix compilation due to bad comment. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 2bf1363d4f9ba477c8483ea1cc7c9f03ecb30d29 https://github.com/modula3/cm3/commit/2bf1363d4f9ba477c8483ea1cc7c9f03ecb30d29 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- Fix compilation due to bad comment. From noreply at github.com Sun Feb 21 12:00:54 2021 From: noreply at github.com (Jay Krell) Date: Sun, 21 Feb 2021 03:00:54 -0800 Subject: [M3commit] [modula3/cm3] 43a4d0: Skip building caltech-other/voronoi on Win32 since... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 43a4d01f561c5248a87897e12ebaca561108e127 https://github.com/modula3/cm3/commit/43a4d01f561c5248a87897e12ebaca561108e127 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M caltech-other/voronoi/src/m3makefile Log Message: ----------- Skip building caltech-other/voronoi on Win32 since it fails to build. (#64) * Skip building caltech-other/voronoi on Win32 since it fails to build. From noreply at github.com Sun Feb 21 12:05:29 2021 From: noreply at github.com (Jay Krell) Date: Sun, 21 Feb 2021 03:05:29 -0800 Subject: [M3commit] [modula3/cm3] 7e74b2: caltech-other/newuoa/m3makefile: Move library() to... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7e74b22e8c543ba12581f0d6507237f5bc4d002a https://github.com/modula3/cm3/commit/7e74b22e8c543ba12581f0d6507237f5bc4d002a Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M caltech-other/newuoa/src/m3makefile Log Message: ----------- caltech-other/newuoa/m3makefile: Move library() to within if amd64_linux so it builds/ships (#65) without error (does nothing) on other platforms, such as amd64_nt. From noreply at github.com Sun Feb 21 12:12:16 2021 From: noreply at github.com (Jay Krell) Date: Sun, 21 Feb 2021 03:12:16 -0800 Subject: [M3commit] [modula3/cm3] 41e2e7: Skip m3-scheme/mscheme on Win32 where it fails to ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 41e2e793123d1482d52ef9c713fb026f906d0b2a https://github.com/modula3/cm3/commit/41e2e793123d1482d52ef9c713fb026f906d0b2a Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-scheme/mscheme/src/m3makefile Log Message: ----------- Skip m3-scheme/mscheme on Win32 where it fails to build. (#66) From noreply at github.com Sun Feb 21 12:34:46 2021 From: noreply at github.com (Jay Krell) Date: Sun, 21 Feb 2021 03:34:46 -0800 Subject: [M3commit] [modula3/cm3] c0c388: Skip more mscheme builds on Win32 and fix path to ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: c0c388f3f2590f3b4339b1a25d8c02e80fb59e0a https://github.com/modula3/cm3/commit/c0c388f3f2590f3b4339b1a25d8c02e80fb59e0a Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: 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 Log Message: ----------- Skip more mscheme builds on Win32 and fix path to root m3overrides. (#67) From noreply at github.com Sun Feb 21 12:43:44 2021 From: noreply at github.com (Jay Krell) Date: Sun, 21 Feb 2021 03:43:44 -0800 Subject: [M3commit] [modula3/cm3] dedaf7: Disable Grisu for all Win32 targets. (#68) Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: dedaf7210d4249e3a83adf6cec0dff57e0080fbb https://github.com/modula3/cm3/commit/dedaf7210d4249e3a83adf6cec0dff57e0080fbb Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-libs/m3core/src/float/Common/grisu/m3makefile Log Message: ----------- Disable Grisu for all Win32 targets. (#68) This merits further investigation, but problems have been seen and it was ok before, right? From noreply at github.com Sun Feb 21 19:17:19 2021 From: noreply at github.com (Jay Krell) Date: Sun, 21 Feb 2021 10:17:19 -0800 Subject: [M3commit] [modula3/cm3] 57e97e: WinConstants: ifdef conditionalize RTL_CRITSECT_TY... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 57e97e0f2394d42d092a5e16214a4b7687227f07 https://github.com/modula3/cm3/commit/57e97e0f2394d42d092a5e16214a4b7687227f07 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-libs/m3core/src/win32/WinConstants.c Log Message: ----------- WinConstants: ifdef conditionalize RTL_CRITSECT_TYPE and RTL_RESOURCE_TYPE. They must have been removed from later SDKs. Commit: c6c1e4388e42f3eb65e3f97ad0e4d28a900c9de7 https://github.com/modula3/cm3/commit/c6c1e4388e42f3eb65e3f97ad0e4d28a900c9de7 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-libs/m3core/src/win32/m3makefile Log Message: ----------- Compile WinConstants.c. Commit: 90bfe0d46a045cc0cdbd496ef46fe452f7e726f3 https://github.com/modula3/cm3/commit/90bfe0d46a045cc0cdbd496ef46fe452f7e726f3 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/NT.common Log Message: ----------- Do statically link to Uconstants.obj as intended, it is important to not crashing. Otherwise we crash because m3_jmpbuf_size is wrong/huge. Commit: 0b9ec155fa3678093003241fddb2e15d06954c76 https://github.com/modula3/cm3/commit/0b9ec155fa3678093003241fddb2e15d06954c76 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/NT.common Log Message: ----------- Do ship WinConstants.obj as intended, maybe not important. Commit: 58c0b72413e1bb74298270aeef6637b73532126d https://github.com/modula3/cm3/commit/58c0b72413e1bb74298270aeef6637b73532126d Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M scripts/python/make-dist.py Log Message: ----------- make-dist: Copy the .objs for all stages. This is required to avoid crashing on AMD64_NT with stack overflow, when running functions with try. This used to work somehow? The overall problem here is that when dynamic linking on NT, dynamically linked symbols must be accessed through a pointer, the name is prefixed with __imp__. For functions, the linker will create the function for you: foo: jmp [__imp__foo] but it cannot do that for data. it is generally undesirable to micromanage, even in a compiler, which symbols are imported or not. There are choices to avoid such micromanagement: 1 No dynamic linking. Not really acceptable. 2 No dynamic linking of data. This is done here. By putting the data in a few known objs that get special treatment. It could be better. Another way would really be to dynamically link no data. Use accessor functions. That is tempting. 3 Reference all symbols via a pointer, which might just point back to the referencing module. Somehow need to manage the defaulting of those. I think this is the Unix way. This is tempting, but has not been attempted. It might get linker warnings but they are ignorable. Likely advantage of option 3 is pointer identity. The reason for the stack overflow is we use alloca(m3_jmpbuf_size) and m3_jmpbuf_size, data, is messed up. One would think the mess up is it is a pointer to size, but that didn't seem to be the case. Compare: https://github.com/modula3/cm3/compare/dedaf7210d42...58c0b72413e1 From noreply at github.com Mon Feb 22 04:13:54 2021 From: noreply at github.com (Jay Krell) Date: Sun, 21 Feb 2021 19:13:54 -0800 Subject: [M3commit] [modula3/cm3] aabd11: Win64: Fix some unsigned/signed problems around Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: aabd1140f5c7dd3ea91ff0c53500df311d4e866f https://github.com/modula3/cm3/commit/aabd1140f5c7dd3ea91ff0c53500df311d4e866f Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-libs/libm3/src/os/WIN32/LazyConsole.i3 M m3-libs/libm3/src/os/WIN32/LazyConsole.m3 M m3-libs/m3core/src/win32/WinSock.i3 Log Message: ----------- Win64: Fix some unsigned/signed problems around GetStdHandle, SetStdHandle, ioctlsocket that cause fatal runtime conditions. Commit: c1b2de31a72606e63e6d2122e270bc8310f1c535 https://github.com/modula3/cm3/commit/c1b2de31a72606e63e6d2122e270bc8310f1c535 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-libs/m3core/src/m3makefile M m3-sys/cminstall/src/config-no-install/NT.common Log Message: ----------- Win32: Fix static instead of dynamic linking to data. This is crucial, i.e. due to alloca(m3_jmpbuf_size). Otherwise stack overflows running "try" using the C backend targeting NT. It used to work? Commit: 4ea76accf5fd5f1c269aadf247e76337fc032c61 https://github.com/modula3/cm3/commit/4ea76accf5fd5f1c269aadf247e76337fc032c61 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M m3-libs/m3core/src/win32/WinBase.i3 Log Message: ----------- Remove a newline. Commit: 2b8eee3a76b2fa969b9ab1479f69dd74d5382ea3 https://github.com/modula3/cm3/commit/2b8eee3a76b2fa969b9ab1479f69dd74d5382ea3 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M scripts/python/make-dist.py Log Message: ----------- make-dist: Stop differentiating old vs. new Visual C++. Make all and msi for all, not just min zip. If anyone really tries pre-80, well, not sure what to say about that. Undoing roughly: commit b848e3bbecfe1adceed1c54c87ff22f2e0b743dc Author: Jay Michael Krell Date: Sun Apr 4 06:04:20 2010 +0000 for Visual C++ older than 8.0, only make "min" "zip" -- not "all", not "msi" restore replacement of "NT386" with "x86" that I seem to have totally lost replace "pre-" with "pre" everywhere some vertical compression Problem became, having to always add newer versions. Commit: 49b42bec0a361956974bc728197c059486ac5f01 https://github.com/modula3/cm3/commit/49b42bec0a361956974bc728197c059486ac5f01 Author: Jay Krell Date: 2021-02-21 (Sun, 21 Feb 2021) Changed paths: M scripts/python/pylib.py Log Message: ----------- make-dist/pylib: Look for WiX in a current place. Compare: https://github.com/modula3/cm3/compare/58c0b72413e1...49b42bec0a36 From noreply at github.com Mon Feb 22 04:26:44 2021 From: noreply at github.com (Jay Krell) Date: Sun, 21 Feb 2021 19:26:44 -0800 Subject: [M3commit] [modula3/cm3] Message-ID: Branch: refs/tags/AMD64_NT-d5.10.0-VC2019-20210221 Home: https://github.com/modula3/cm3 From noreply at github.com Tue Feb 23 07:28:32 2021 From: noreply at github.com (Jay Krell) Date: Mon, 22 Feb 2021 22:28:32 -0800 Subject: [M3commit] [modula3/cm3] 1a6a2d: scripts: Stop building import-libs, it had an error Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 1a6a2d40e1e74867b8f4b071657d3d7a6c1a5a53 https://github.com/modula3/cm3/commit/1a6a2d40e1e74867b8f4b071657d3d7a6c1a5a53 Author: Jay Krell Date: 2021-02-22 (Mon, 22 Feb 2021) Changed paths: M scripts/python/pylib.py M scripts/python/upgrade.py Log Message: ----------- scripts: Stop building import-libs, it had an error and isn't really needed, hardly ever was. The main thing was that old releases shipped bad ones. Remove newlines from ends of lines. Start being case insensitive about argv. Start consdering if "c" is appended to BuildDir, as a separate factor, and considering if a target only has the C backend. Commit: ce2d23fdaf6743b2bc63db15a3bd68f48664d380 https://github.com/modula3/cm3/commit/ce2d23fdaf6743b2bc63db15a3bd68f48664d380 Author: Jay Krell Date: 2021-02-22 (Mon, 22 Feb 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/AMD64_NT Log Message: ----------- config: Make AMD64_NT parallel like AMD64.common. M3_PARALLEL_BACK = 20 % AMD64 machines usually have lots of cores Ideally this would be more detected at runtime for all hosts. Commit: 645fa71cc58497bcce7244189521ee13b0e4c5a3 https://github.com/modula3/cm3/commit/645fa71cc58497bcce7244189521ee13b0e4c5a3 Author: Jay Krell Date: 2021-02-22 (Mon, 22 Feb 2021) Changed paths: M m3-libs/m3core/src/m3makefile Log Message: ----------- Fix build break related to `Obj` on non-Windows. Commit: 146bbe7d82f5069e3e805e19ff8c7cc691a20d71 https://github.com/modula3/cm3/commit/146bbe7d82f5069e3e805e19ff8c7cc691a20d71 Author: Jay Krell Date: 2021-02-22 (Mon, 22 Feb 2021) Changed paths: M .gitignore M m3-libs/m3core/src/runtime/m3makefile M m3-sys/m3gdb/src/m3makefile M m3-sys/m3middle/src/Target.i3 M scripts/python/targets.txt M www/uploaded-archives/targets.txt Log Message: ----------- Add new targets: ARM64_{DARWIN, LINUX, NT}, RISCV64_LINUX. This adds them not yet in all places. e.g. no config files yet. They are not implemented yet, but there should not be much to it, given existing ports and the C backend. Compare: https://github.com/modula3/cm3/compare/49b42bec0a36...146bbe7d82f5 From noreply at github.com Tue Feb 23 07:55:34 2021 From: noreply at github.com (Jay Krell) Date: Mon, 22 Feb 2021 22:55:34 -0800 Subject: [M3commit] [modula3/cm3] 273b04: Fix Unix build: Ship WinConstants only in win32. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 273b04bb216509e89f0ea8c02f4da86dbdf9f90d https://github.com/modula3/cm3/commit/273b04bb216509e89f0ea8c02f4da86dbdf9f90d Author: Jay Krell Date: 2021-02-22 (Mon, 22 Feb 2021) Changed paths: M m3-libs/m3core/src/m3makefile M m3-libs/m3core/src/win32/m3makefile Log Message: ----------- Fix Unix build: Ship WinConstants only in win32. From noreply at github.com Tue Feb 23 17:51:44 2021 From: noreply at github.com (Jay Krell) Date: Tue, 23 Feb 2021 08:51:44 -0800 Subject: [M3commit] [modula3/cm3] 9fe816: m3cc: Remove -pedantic. It causes many warnings, a... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 9fe8163eee66c834d0a907aaecbffa16a9315d54 https://github.com/modula3/cm3/commit/9fe8163eee66c834d0a907aaecbffa16a9315d54 Author: JayKrell Date: 2021-02-23 (Tue, 23 Feb 2021) Changed paths: M m3-sys/m3cc/gcc-4.7/gcc/configure Log Message: ----------- m3cc: Remove -pedantic. It causes many warnings, about __FUNCTION__. From noreply at github.com Tue Feb 23 22:40:08 2021 From: noreply at github.com (Jay Krell) Date: Tue, 23 Feb 2021 13:40:08 -0800 Subject: [M3commit] [modula3/cm3] b86bf2: m3core: Build Win32 for all targets. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: b86bf2f6cb81c1c23fd067c331688bbaa0c08219 https://github.com/modula3/cm3/commit/b86bf2f6cb81c1c23fd067c331688bbaa0c08219 Author: JayKrell Date: 2021-02-23 (Tue, 23 Feb 2021) Changed paths: M m3-libs/m3core/src/m3makefile Log Message: ----------- m3core: Build Win32 for all targets. The .i3 there can be useful, for example to build mklib on other hosts. This effectively undoes commit 997e499081 from year 2014. This originally went in commit 907515e97f in year 2008. Frontend accepts all the Win32 calling conventions, silently ignoring them on most platforms. From noreply at github.com Tue Feb 23 22:41:13 2021 From: noreply at github.com (Jay Krell) Date: Tue, 23 Feb 2021 13:41:13 -0800 Subject: [M3commit] [modula3/cm3] 0257cb: m3cc: Do not optimize gcc register allocator as th... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 0257cbf56b8e72f1293d2ee20200b9eec4ecce70 https://github.com/modula3/cm3/commit/0257cbf56b8e72f1293d2ee20200b9eec4ecce70 Author: JayKrell Date: 2021-02-23 (Tue, 23 Feb 2021) Changed paths: M m3-sys/m3cc/gcc-4.7/gcc/ira-build.c M m3-sys/m3cc/gcc-4.7/gcc/ira-color.c M m3-sys/m3cc/gcc-4.7/gcc/ira-conflicts.c M m3-sys/m3cc/gcc-4.7/gcc/ira-costs.c M m3-sys/m3cc/gcc-4.7/gcc/ira-emit.c M m3-sys/m3cc/gcc-4.7/gcc/ira-lives.c M m3-sys/m3cc/gcc-4.7/gcc/ira.c Log Message: ----------- m3cc: Do not optimize gcc register allocator as that causes cm3cg to crash when building for Linux/x86. More debugging would be good. From noreply at github.com Wed Feb 24 00:55:35 2021 From: noreply at github.com (Jay Krell) Date: Tue, 23 Feb 2021 15:55:35 -0800 Subject: [M3commit] [modula3/cm3] 552270: caltech-voronoi: Make it valid C++ and valid ANSI ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 552270556130d92cc39ae14f470cb1147c73f0a9 https://github.com/modula3/cm3/commit/552270556130d92cc39ae14f470cb1147c73f0a9 Author: JayKrell Date: 2021-02-23 (Tue, 23 Feb 2021) Changed paths: M caltech-other/voronoi/src/defs.c M caltech-other/voronoi/src/defs.h M caltech-other/voronoi/src/edgelist.c M caltech-other/voronoi/src/geometry.c M caltech-other/voronoi/src/heap.c M caltech-other/voronoi/src/m3makefile M caltech-other/voronoi/src/main.c M caltech-other/voronoi/src/memory.c M caltech-other/voronoi/src/mod3_main.c M caltech-other/voronoi/src/mymalloc.c M caltech-other/voronoi/src/mymalloc.h M caltech-other/voronoi/src/output.c M caltech-other/voronoi/src/voronoi.c Log Message: ----------- caltech-voronoi: Make it valid C++ and valid ANSI C89. i.e. not K&R C. Also remove casts from NULL, but it was ok. Also sometimes remove parens on return, ok either way. ANSI C made return a statement not an expression, so there is no precedence concern. Also add return instead of falling off non-void. Also fix some Windows newlines. From noreply at github.com Wed Feb 24 00:55:49 2021 From: noreply at github.com (Jay Krell) Date: Tue, 23 Feb 2021 15:55:49 -0800 Subject: [M3commit] [modula3/cm3] 77e686: caltech-bool: Convert to valid C++. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 77e6868786b3cc054483a33f97f5adf608161c5e https://github.com/modula3/cm3/commit/77e6868786b3cc054483a33f97f5adf608161c5e Author: JayKrell Date: 2021-02-23 (Tue, 23 Feb 2021) Changed paths: M caltech-other/bool/src/bool.c M caltech-other/bool/src/bool.h M caltech-other/bool/src/misc.c M caltech-other/bool/src/misc.h Log Message: ----------- caltech-bool: Convert to valid C++. - Add extern "C". - Rename true and false to True and False. From noreply at github.com Wed Feb 24 00:55:57 2021 From: noreply at github.com (Jay Krell) Date: Tue, 23 Feb 2021 15:55:57 -0800 Subject: [M3commit] [modula3/cm3] e12df9: m3-scheme: Add extern "C". Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: e12df920a3daca5bfab9cbf64f677b1d67fe631f https://github.com/modula3/cm3/commit/e12df920a3daca5bfab9cbf64f677b1d67fe631f Author: JayKrell Date: 2021-02-23 (Tue, 23 Feb 2021) Changed paths: M m3-scheme/mscheme/src/schemeUnixDeps.c M m3-scheme/schemesig/src/sighandler.c Log Message: ----------- m3-scheme: Add extern "C". From noreply at github.com Wed Feb 24 01:24:36 2021 From: noreply at github.com (Jay Krell) Date: Tue, 23 Feb 2021 16:24:36 -0800 Subject: [M3commit] [modula3/cm3] Message-ID: Branch: refs/tags/20210223 Home: https://github.com/modula3/cm3 From noreply at github.com Wed Feb 24 01:29:52 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 00:29:52 +0000 (UTC) Subject: [M3commit] [modula3/cm3] Message-ID: Branch: refs/tags/20210223-Linux-x8-c Home: https://github.com/modula3/cm3 From noreply at github.com Wed Feb 24 01:41:48 2021 From: noreply at github.com (Jay Krell) Date: Tue, 23 Feb 2021 16:41:48 -0800 Subject: [M3commit] [modula3/cm3] Message-ID: Branch: refs/tags/20210223-amd64linuxbootc Home: https://github.com/modula3/cm3 From noreply at github.com Wed Feb 24 03:56:48 2021 From: noreply at github.com (Jay Krell) Date: Tue, 23 Feb 2021 18:56:48 -0800 Subject: [M3commit] [modula3/cm3] fb0142: m3cc: Attempt to reduce warnings building m3cc. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: fb0142b501b9fcefb630ad569ed04175f6b8ae29 https://github.com/modula3/cm3/commit/fb0142b501b9fcefb630ad569ed04175f6b8ae29 Author: JayKrell Date: 2021-02-23 (Tue, 23 Feb 2021) Changed paths: M m3-sys/m3cc/gcc-4.7/gcc/builtins.c M m3-sys/m3cc/gcc-4.7/gcc/configure M m3-sys/m3cc/gcc-4.7/gcc/configure.ac M m3-sys/m3cc/gcc-4.7/gcc/gengtype.c M m3-sys/m3cc/gcc-4.7/gcc/gengtype.h M m3-sys/m3cc/gcc-4.7/gcc/system.h M m3-sys/m3cc/gcc-4.7/include/ansidecl.h M m3-sys/m3cc/gcc-4.7/libcpp/configure M m3-sys/m3cc/gcc-4.7/libiberty/configure Log Message: ----------- m3cc: Attempt to reduce warnings building m3cc. - Remove -pedantic more/better/cleaner (probably same amount). - Convert some if undefined to if 0 or 1. - Remove unused static function. From noreply at github.com Wed Feb 24 12:29:59 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 03:29:59 -0800 Subject: [M3commit] [modula3/cm3] f5eff1: m3core: Convert coroutine code to valid C++. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: f5eff1a7541478420f14f2d9553168f20d709a13 https://github.com/modula3/cm3/commit/f5eff1a7541478420f14f2d9553168f20d709a13 Author: JayKrell Date: 2021-02-24 (Wed, 24 Feb 2021) Changed paths: M m3-libs/m3core/src/coroutine/UCONTEXT/ContextC.c M m3-libs/m3core/src/coroutine/UCONTEXT/ContextC.i3 M m3-libs/m3core/src/coroutine/UCONTEXT/CoroutineUcontext.m3 Log Message: ----------- m3core: Convert coroutine code to valid C++. The function ContextC__PushContext is not convincing. The memset is inefficient, and can be completely optimized away. Intent is probably to do alloca(256) here. Clean it up a little. Try to accomodate stacks that grow up. i.e. HPPA, but that is not tested (I might have tested it long ago with the Posix threads). include m3core.h Use size_t instead of unsigned long; or rather, whatever m3core.h does Fix mild printf problems %p vs. %lx. Cast to correct pointer type instead of void*. Do not use auto, at least under if cplusplus. Add extern C (maybe some day remove it) From noreply at github.com Wed Feb 24 12:33:00 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 03:33:00 -0800 Subject: [M3commit] [modula3/cm3] 2f8746: config: Undo AMD64_NT parallel back. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 2f874639848066369531cff50202b3e0cd2ea55f https://github.com/modula3/cm3/commit/2f874639848066369531cff50202b3e0cd2ea55f Author: JayKrell Date: 2021-02-24 (Wed, 24 Feb 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/AMD64_NT Log Message: ----------- config: Undo AMD64_NT parallel back. I and another are getting sharing violations. Investigate at a much later time or never. (bulk cl will help, possibly with /MP) This reverts ce2d23fdaf6743b2bc63db15a3bd68f48664d380 From noreply at github.com Wed Feb 24 13:14:15 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 04:14:15 -0800 Subject: [M3commit] [modula3/cm3] cd7ceb: m3core: Disable Grisu for C backend until debugged. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: cd7ceba2670340a1feabe2e97e81b998c969bda1 https://github.com/modula3/cm3/commit/cd7ceba2670340a1feabe2e97e81b998c969bda1 Author: JayKrell Date: 2021-02-24 (Wed, 24 Feb 2021) Changed paths: M m3-libs/m3core/src/float/Common/grisu/m3makefile Log Message: ----------- m3core: Disable Grisu for C backend until debugged. From noreply at github.com Wed Feb 24 13:33:35 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 04:33:35 -0800 Subject: [M3commit] [modula3/cm3] 156396: c backend setjmp work: Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 1563967d40e35c1aabe97d8a3e6c37de77acf27c https://github.com/modula3/cm3/commit/1563967d40e35c1aabe97d8a3e6c37de77acf27c Author: JayKrell Date: 2021-02-24 (Wed, 24 Feb 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 M m3-sys/m3middle/src/Target.m3 Log Message: ----------- c backend setjmp work: - Avoid include setjmp.h unless call seen to setjmp, or less likely, longjmp. This should be a little faster. - Handle the VMS setjmp name. - Cast first parameter to setjmp tp *(jmpbuf*) to fix C++ compilation. - setjmp instead of _setjmp on nascent ARM64_NT - some comments From noreply at github.com Wed Feb 24 13:44:56 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 04:44:56 -0800 Subject: [M3commit] [modula3/cm3] ae59df: C backend: Remove longjmp support. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: ae59df42e7868c5fedf557dd30bbdd8305c78b42 https://github.com/modula3/cm3/commit/ae59df42e7868c5fedf557dd30bbdd8305c78b42 Author: JayKrell Date: 2021-02-24 (Wed, 24 Feb 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- C backend: Remove longjmp support. I just regressed it (void vs. not_void), but nobody references it this way in Modula-3. The uses all go via C wrappers. From noreply at github.com Wed Feb 24 17:01:49 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 08:01:49 -0800 Subject: [M3commit] [modula3/cm3] 76b404: caltech: More C++ fixes. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 76b404e98454e48d906b3998c027530ecca58242 https://github.com/modula3/cm3/commit/76b404e98454e48d906b3998c027530ecca58242 Author: JayKrell Date: 2021-02-24 (Wed, 24 Feb 2021) Changed paths: M caltech-other/voronoi/src/defs.h Log Message: ----------- caltech: More C++ fixes. Forward declare structs so that function declarations do not introduce private structs. From noreply at github.com Wed Feb 24 19:04:03 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 10:04:03 -0800 Subject: [M3commit] [modula3/cm3] 7b7dc0: arm64_darwin support Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 7b7dc01250fbdfe7ff1f6fd095c30f33bb2510f8 https://github.com/modula3/cm3/commit/7b7dc01250fbdfe7ff1f6fd095c30f33bb2510f8 Author: JayKrell Date: 2021-02-24 (Wed, 24 Feb 2021) Changed paths: M caltech-parser/drawcontext/dcpane/src/m3makefile M caltech-parser/drawcontext/kgv/src/m3makefile M caltech-parser/drawcontext/src/m3makefile M caltech-parser/drawcontext/test/src/m3makefile M caltech-parser/paneman/kemacs/src/m3makefile M caltech-parser/paneman/src/m3makefile M m3-comm/sharedobj/tests/obsequence/src/m3makefile M m3-demo/mentor/src/minimax/m3makefile M m3-demo/sharedboard/board/src/m3makefile M m3-demo/sharedboard/calendar/src/m3makefile M m3-games/badbricks/src/m3makefile M m3-games/columns/src/m3makefile M m3-games/fours/src/m3makefile M m3-games/solitaire/src/m3makefile M m3-libs/libm3/src/os/POSIX/m3makefile M m3-libs/m3core/src/runtime/POSIX/RTSignalC.c M m3-libs/m3core/src/thread/PTHREAD/ThreadApple.c M m3-obliq/obliqbin3D/src/m3makefile M m3-obliq/obliqbinstd/src/m3makefile M m3-obliq/obliqlib3D/src/m3makefile M m3-obliq/obliqlibanim/src/m3makefile M m3-obliq/obliqlibui/src/m3makefile A m3-sys/cminstall/src/config-no-install/ARM64.common A m3-sys/cminstall/src/config-no-install/ARM64_DARWIN M m3-sys/cminstall/src/config-no-install/Darwin.common M m3-sys/m3middle/src/Target.m3 M m3-ui/anim3D/src/m3makefile M m3-ui/ui-tests/pixtest/src/m3makefile M m3-ui/webvbt/src/oblet/m3makefile M m3-www/proxy/src/m3makefile A scripts/dos2unix-all.cmd M scripts/python/pylib.py Log Message: ----------- arm64_darwin support Bootstrapped from AMD64_NT and then self builds. using C backend. From noreply at github.com Wed Feb 24 19:49:41 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 10:49:41 -0800 Subject: [M3commit] [modula3/cm3] Message-ID: Branch: refs/tags/ARM64_DARWIN-20210224 Home: https://github.com/modula3/cm3 From noreply at github.com Wed Feb 24 19:51:48 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 10:51:48 -0800 Subject: [M3commit] [modula3/cm3] Message-ID: Branch: refs/tags/MD64_DARWIN-d5.10.0-Darwin16.7.0-20180512 Home: https://github.com/modula3/cm3 From noreply at github.com Wed Feb 24 19:55:04 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 10:55:04 -0800 Subject: [M3commit] [modula3/cm3] Message-ID: Branch: refs/tags/MD64_DARWIN-d5.10.0-Darwin16.7.0-20180512 Home: https://github.com/modula3/cm3 From noreply at github.com Thu Feb 25 07:53:33 2021 From: noreply at github.com (Jay Krell) Date: Wed, 24 Feb 2021 22:53:33 -0800 Subject: [M3commit] [modula3/cm3] f39e8f: Win32: Widen some parameters from UINT32 to SIZE_T. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: f39e8f66f94e2cefec08dc909abc41f34dfec96c https://github.com/modula3/cm3/commit/f39e8f66f94e2cefec08dc909abc41f34dfec96c Author: JayKrell Date: 2021-02-24 (Wed, 24 Feb 2021) Changed paths: M m3-libs/m3core/src/win32/WinBase.i3 Log Message: ----------- Win32: Widen some parameters from UINT32 to SIZE_T. This is not the -1 vs. UINT32 problem, this is just other mistakes I noticed, i.e. HeapAlloc and affinity. From noreply at github.com Thu Feb 25 20:04:01 2021 From: noreply at github.com (Jay Krell) Date: Thu, 25 Feb 2021 11:04:01 -0800 Subject: [M3commit] [modula3/cm3] 90b829: Win32: Widen some UINT32 parameters to UINT32_ARG ... Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 90b8299e25fb66692a21ee2167323569f336efae https://github.com/modula3/cm3/commit/90b8299e25fb66692a21ee2167323569f336efae Author: JayKrell Date: 2021-02-25 (Thu, 25 Feb 2021) Changed paths: M m3-libs/m3core/src/win32/WinBase.i3 M m3-libs/m3core/src/win32/WinSock.i3 Log Message: ----------- Win32: Widen some UINT32 parameters to UINT32_ARG = INTEGER so that 64bit systems can pass -1 or other seemingly invalid but acceptable values. ExitProcess because people pass -1 TerminateProcess because it is like ExitProcess And ioctlsocket because it has been a problem. On NT/amd64 parameters are never passed in half registers, or half stack positions. Always 64bits per parameter. So this is safe. It does not work for out pointer/var parameters however. From noreply at github.com Fri Feb 26 07:59:08 2021 From: noreply at github.com (Jay Krell) Date: Thu, 25 Feb 2021 22:59:08 -0800 Subject: [M3commit] [modula3/cm3] afd597: C backend misc. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: afd5974d1999b9991489584691a7201578236af0 https://github.com/modula3/cm3/commit/afd5974d1999b9991489584691a7201578236af0 Author: JayKrell Date: 2021-02-25 (Thu, 25 Feb 2021) Changed paths: M m3-sys/m3back/src/M3C.m3 Log Message: ----------- C backend misc. - fix typo in comment - dial up debug level (should be command line parameter) - TIntLiteral: We had: commented out assert(x < TInt.Max8); commented out assert(x < TInt.Max16); commented out assert(x < TInt.Max32); I guess they failed for me in the past. not commented out assert(x < TInt.Max64); It fails on big endian like PA32_HPUX, in LongRealRep.i3. Change these all to: not commented out assert(x < TWord.Max8); not commented out assert(x < TWord.Max16); not commented out assert(x < TWord.Max32); not commented out assert(x < TWord.Max64); This merits more attention in the frontend. Frontend I think plays fast and loose with signedness, and likely all literals should be considered signed, but I don't know. When asserts fail in TIntLiteral, be sure to flush the C file. To do that, move TIntLiteral to be a method, and change all callers correspondingly. - Remove some newlines by style. From noreply at github.com Fri Feb 26 08:09:55 2021 From: noreply at github.com (Jay Krell) Date: Thu, 25 Feb 2021 23:09:55 -0800 Subject: [M3commit] [modula3/cm3] d3e2e7: Config: Fix Alpha VMS builds. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: d3e2e72605dcb4a5972b01cb6216c9c8a6ff4639 https://github.com/modula3/cm3/commit/d3e2e72605dcb4a5972b01cb6216c9c8a6ff4639 Author: JayKrell Date: 2021-02-25 (Thu, 25 Feb 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/VMS.common Log Message: ----------- Config: Fix Alpha VMS builds. cd C:\s\cm3\m3-sys\cminstall\src\config-no-install findstr TARGET_ENDIAN AL* vms* Alpha32.common:readonly TARGET_ENDIAN = "LITTLE" % { "BIG" OR "LITTLE" } Alpha64.common:readonly TARGET_ENDIAN = "LITTLE" % { "BIG" OR "LITTLE" } ALPHA_OSF:readonly TARGET_ENDIAN = "LITTLE" % { "BIG" OR "LITTLE" } VMS.common:readonly TARGET_ENDIAN = "LITTLE" This fails because cannot write to readonly variable. Quake should allow it, maybe, since writing the same value it already has. But for now, just remove the last line. Arguably it doesn't belong there, or even where it is. Arguably it is a function of the root config, in case of architectures like ARM, PowerPC, MIPS whose endianness varies. Ideally we remove endian sensitivity from the system, really. From noreply at github.com Fri Feb 26 10:54:10 2021 From: noreply at github.com (Jay Krell) Date: Fri, 26 Feb 2021 01:54:10 -0800 Subject: [M3commit] [modula3/cm3] Message-ID: Branch: refs/tags/AMD64_DARWIN-20210226 Home: https://github.com/modula3/cm3 From noreply at github.com Sat Feb 27 12:13:50 2021 From: noreply at github.com (Jay Krell) Date: Sat, 27 Feb 2021 03:13:50 -0800 Subject: [M3commit] [modula3/cm3] 5eca40: RISCV64_LINUX support. Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: 5eca40e93a3d7f2eb5edd2b7864902af76aa77ba https://github.com/modula3/cm3/commit/5eca40e93a3d7f2eb5edd2b7864902af76aa77ba Author: JayKrell Date: 2021-02-27 (Sat, 27 Feb 2021) Changed paths: M m3-libs/libm3/src/os/POSIX/m3makefile M m3-libs/m3core/src/runtime/POSIX/RTSignalC.c A m3-sys/cminstall/src/config-no-install/RISCV64.common A m3-sys/cminstall/src/config-no-install/RISCV64_LINUX M m3-sys/m3middle/src/Target.m3 M scripts/python/pylib.py Log Message: ----------- RISCV64_LINUX support. From noreply at github.com Sat Feb 27 23:39:40 2021 From: noreply at github.com (Jay Krell) Date: Sat, 27 Feb 2021 14:39:40 -0800 Subject: [M3commit] [modula3/cm3] Message-ID: Branch: refs/tags/RISCV64_LINUX-20210227 Home: https://github.com/modula3/cm3 From noreply at github.com Sun Feb 28 02:25:52 2021 From: noreply at github.com (peter mckinna) Date: Sat, 27 Feb 2021 17:25:52 -0800 Subject: [M3commit] [modula3/cm3] b571d8: Added IP6 socket structs Message-ID: Branch: refs/heads/master Home: https://github.com/modula3/cm3 Commit: b571d812e4daeb8f4f5906b1d1a9aab49af88d48 https://github.com/modula3/cm3/commit/b571d812e4daeb8f4f5906b1d1a9aab49af88d48 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-libs/m3core/src/unix/Common/Utypes.i3 M m3-libs/m3core/src/unix/uin-common/Uin.i3 Log Message: ----------- Added IP6 socket structs Commit: ebe2ad18358e6ed43043b3e095619911d5eee6c7 https://github.com/modula3/cm3/commit/ebe2ad18358e6ed43043b3e095619911d5eee6c7 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-comm/tcp/src/POSIX/IP.m3 M m3-comm/tcp/src/common/IP.i3 M m3-libs/m3core/src/unix/Common/Unetdb.i3 M m3-libs/m3core/src/unix/Common/Usocket.i3 Log Message: ----------- Added getaddrinfo and getnameinfo support to cope with gethostname being deprecated. Commit: 0b952877e5abc31829006e12d036834322168ee4 https://github.com/modula3/cm3/commit/0b952877e5abc31829006e12d036834322168ee4 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: R m3-ui/vbtkit/src/vbtkitutils/Rsrc.i3 R m3-ui/vbtkit/src/vbtkitutils/Rsrc.m3 M m3-ui/vbtkit/src/vbtkitutils/m3makefile Log Message: ----------- Move Rsrc from vbtkit to libm3 first delete from vbtkit Commit: 2a071c6d559d95c39b16a12f70ca4a2046687e90 https://github.com/modula3/cm3/commit/2a071c6d559d95c39b16a12f70ca4a2046687e90 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-libs/libm3/src/m3makefile A m3-libs/libm3/src/resource/Rsrc.i3 A m3-libs/libm3/src/resource/Rsrc.m3 A m3-libs/libm3/src/resource/m3makefile Log Message: ----------- Finish moving Rsrc from vbtkit to libm3 Commit: c4a07d98db2d25329859a5215d37ab5253ebcc31 https://github.com/modula3/cm3/commit/c4a07d98db2d25329859a5215d37ab5253ebcc31 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-tools/pp/src/m3makefile Log Message: ----------- Remove dependency on vbtkit for pp. Rsrc is now in libm3. Also add g++ as system_cc so the link succeeds. Commit: ac61f634c7cf87409d1bd2c6bfb2ceaf84abb98c https://github.com/modula3/cm3/commit/ac61f634c7cf87409d1bd2c6bfb2ceaf84abb98c Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-libs/m3core/src/unix/Common/Unetdb.i3 M m3-libs/m3core/src/unix/Common/Utypes.i3 Log Message: ----------- Fix socklen_t - revert to previous definition. There seems some confusion regarding the C type of socklen_t. If it's not 32 bits it breaks structure definitions sockaddr and if its not word size it asserts in function parameters. Linus absolutely says it should be type int or else it breaks NetBSD sockets. Commit: cd850a45af22264b18b74bc24ad5349227508ba4 https://github.com/modula3/cm3/commit/cd850a45af22264b18b74bc24ad5349227508ba4 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-sys/m3middle/src/M3Timers.i3 M m3-sys/m3middle/src/M3Timers.m3 Log Message: ----------- Add llvm timers for llc and opt backend passes Commit: 51b28b4134f283254750b73e1fa36baea0ec7038 https://github.com/modula3/cm3/commit/51b28b4134f283254750b73e1fa36baea0ec7038 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-sys/cm3/src/Builder.m3 Log Message: ----------- Add call to llvm optimiser opt Commit: 04707630112aea4c06271684de3aba592607fd13 https://github.com/modula3/cm3/commit/04707630112aea4c06271684de3aba592607fd13 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/Unix.common Log Message: ----------- Updates for llvm llc and opt Commit: a02dafeaafe7e5d24e1ff3fada74fb66abc95c06 https://github.com/modula3/cm3/commit/a02dafeaafe7e5d24e1ff3fada74fb66abc95c06 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-sys/cminstall/src/config-no-install/Linux.common Log Message: ----------- Add gold as default linker for Linux Commit: 8a5c71e1ab17e40a72cb325aadc30e8c029a4f2b https://github.com/modula3/cm3/commit/8a5c71e1ab17e40a72cb325aadc30e8c029a4f2b Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-tools/pp/src/m3overrides Log Message: ----------- pp overrides - forgot to commit Commit: c62c53ddfc204b0623db343469db6418d37e527d https://github.com/modula3/cm3/commit/c62c53ddfc204b0623db343469db6418d37e527d Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-libs/libm3/src/etimer/ETimer.i3 M m3-libs/libm3/src/etimer/ETimer.m3 Log Message: ----------- Add append operation to timers. Possible to change timer due to external circumstances. Commit: c13c81eedf4d2ded881e289c3f8d7c9155dc1318 https://github.com/modula3/cm3/commit/c13c81eedf4d2ded881e289c3f8d7c9155dc1318 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-sys/m3quake/src/QMachine.i3 M m3-sys/m3quake/src/QMachine.m3 Log Message: ----------- Add wall clock timing to parallel build threads. Commit: acdd4116738e0caa39c325c0d2d2c106627518b1 https://github.com/modula3/cm3/commit/acdd4116738e0caa39c325c0d2d2c106627518b1 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-sys/cm3/src/Builder.m3 Log Message: ----------- Ensure the parallel build timers get processed Commit: e2304234f74357239e332cd9d57294cbfd794129 https://github.com/modula3/cm3/commit/e2304234f74357239e332cd9d57294cbfd794129 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-libs/m3core/src/thread/POSIX/ThreadPosixC.c Log Message: ----------- _BSD_SOURCE deprecated - use _DEFAULT_SOURCE Commit: 0da8bd73f5ff711535e3a1a56ed97d588ff7c5ef https://github.com/modula3/cm3/commit/0da8bd73f5ff711535e3a1a56ed97d588ff7c5ef Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-libs/m3core/src/m3makefile Log Message: ----------- coroutines depend on PTHREADS making it impossible to build POSIX threads Commit: 8eb8dd3e36b9a36d9e4c35fbb20c7b846f4cd0e5 https://github.com/modula3/cm3/commit/8eb8dd3e36b9a36d9e4c35fbb20c7b846f4cd0e5 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-libs/libm3/src/pickle/ver2/ConvertPacking.m3 M m3-libs/libm3/src/pickle/ver2/PklFpMap.m3 Log Message: ----------- Add nowarn to clean up compiler messages for unused declarations Commit: debc5e8ac4788281d159083cd7ea3109d6acb800 https://github.com/modula3/cm3/commit/debc5e8ac4788281d159083cd7ea3109d6acb800 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-libs/libm3/src/pickle/ver2/Pickle2.m3 Log Message: ----------- Remove the Blair tags scattered throughout. He is still noted as the author and holds the copyright but the time is long past development notes littering the code Commit: a6a92a11aa704be43996563b8811a05105a1c9f2 https://github.com/modula3/cm3/commit/a6a92a11aa704be43996563b8811a05105a1c9f2 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-ui/X11R4/src/Common/X.i3 Log Message: ----------- Add XInitThreads support and lock display Commit: 12295e95c5b64356aab6cf2f2347125a4bb9c0ac https://github.com/modula3/cm3/commit/12295e95c5b64356aab6cf2f2347125a4bb9c0ac Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: A m3-ui/X11R4/src/Common/XShape.i3 M m3-ui/X11R4/src/Common/m3makefile Log Message: ----------- Add XShape support for rounded corner windows among other enhancements Commit: 33e1ed68efe2a8b0450491f18538400a162e2828 https://github.com/modula3/cm3/commit/33e1ed68efe2a8b0450491f18538400a162e2828 Author: peter mckinna Date: 2021-02-28 (Sun, 28 Feb 2021) Changed paths: M m3-ui/ui/src/xvbt/XClientF.m3 Log Message: ----------- Call XInitThreads to fix threading problems Compare: https://github.com/modula3/cm3/compare/5eca40e93a3d...33e1ed68efe2