2008-11-17 09:11 jkrell * scripts/python/pylib.py: m3cc has to come first in case of bootstrapping rom just native cm3 (ok import-libs can come before it) 2008-11-17 08:14 jkrell * www/uploaded-archives/update_download_index.sh: more flexible naming, at the cost of hardcoding a list of TARGETS 2008-11-17 07:01 jkrell * www/uploaded-archives/update_download_index.sh: accept anything cm3-*.archive add lzma suffixes allow single files compressed -- .gz without .tar, etc. 2008-11-16 17:57 jkrell * m3-comm/netobj/src/netobj.tmpl, m3-libs/m3core/src/runtime/POSIX/RTOS.m3, m3-libs/m3core/src/runtime/POSIX/RTOSbrk.m3, m3-libs/m3core/src/runtime/POSIX/RTOSmmap.m3, m3-libs/m3core/src/runtime/POSIX/m3makefile, m3-libs/m3core/src/unix/linux-libc6/Unix.i3, m3-sys/cminstall/src/config-no-install/PPC_LINUX, m3-sys/cminstall/src/config-no-install/cm3cfg.common, scripts/python/pylib.py: use sbrk instead of mmap on AMD64_LINUX to allocate memory, so that it has a tendency to say contiguous, since we allocate bookkeeping per page and otherwise run out of memory netobj.tmp: go back a version, since it was working around out of memory from "mmap vs. sbrk" pylib.py: fix PPC_LINUX bootstrap, put more files in bootstrap cm3cfg.common: don't overwrite preceding GNU_MAKE 2008-11-16 16:28 jkrell * scripts/sysinfo.sh: detect MIPS64_OPENBSD (should later on revisit 'macppc' vs. something more generic, like this) 2008-11-16 02:13 rodney * m3-sys/m3gdb/gdb/gdb/: Makefile.in, m3-lang.c, m3-threads.c, m3-util.c, m3-util.h, m3-valprint.c: When libm3core is dynamically loaded and a subsequent 'run' command, (after the first) is done, the location of runtime info (e.g., the typecells) needed by m3gdb can (and usually does) change. Formerly, m3gdb did not re-lookup runtime information, leaving lots of things horribly broken after the second 'run' command. Fix this bug. Also, make m3gdb a bit more robust in reacting to inability to find runtime information. 2008-11-15 03:11 rodney * m3-sys/m3gdb/gdb/gdb/: m3-eval.c, m3-exp.c: Fix a bug in which ADR function, applied to a formal parameter passed by reference or a WITH-temp bound by reference failed to do the dereferencing. 2008-11-14 20:21 jkrell * m3-comm/netobj/src/netobj.tmpl: turn off gc on AMD64_LINUX host here until further debugged 2008-11-14 16:17 jkrell * m3-sys/m3middle/src/Target.m3, m3-libs/m3core/src/C/PPC_LINUX/Csetjmp.i3: restore PPC_LINUX to not crashing at startup fix jmpbuf size still doesn't write out any *.mc files and then m3cg hangs 2008-11-14 12:34 jkrell * m3-libs/m3core/src/unix/openbsd-common/1.cpp: print more clearly, and remove two lines that don't compile on several systems 2008-11-14 12:13 jkrell * m3-sys/cminstall/src/config-no-install/Solaris.common: fix newlines 2008-11-14 12:11 jkrell * m3-sys/cminstall/src/config-no-install/: Unix.common, cm3cfg.common: workaround strange/undebugged namespace? behavior (on PPC_LINUX 5.2.6) and fix other namespace oversight 2008-11-14 11:04 jkrell * m3-sys/cminstall/src/config-no-install/cm3cfg.common: fix for bootstrapping from older cm3 that doesn't define HOST, such as PPC 5.2.6; something else still wrong 2008-11-12 15:21 rodney * m3-libs/libm3/src/pickle/ver2/Pickle2.m3: Pickles in both PM3 and CM3 have some hard-coded fingerprints for common types, but the values are not consistent about byte order. Fix CM2 pickle-reading code to accept values of types NULL, ROOT, UNTRACED ROOT, ADDRESS, AND REFANY, written by PM3 pickle-writing code. Probably, the latter two can't even happen in a pickle, unless somebody wrote a custom special for them. This still won't handle a more complex type that depends on these types, e.g., an object that contains a field of one of them. This latter is probably not feasable, because it would require "decrypting" fingerprints, and the hash algorithm for fingerprints is lossy. 2008-11-12 15:05 rodney * doc/help/m3gdb/m3gdb.docbook: Insert one minor missing blank 2008-11-12 14:36 jkrell * scripts/python/pylib.py: turning off gc here isn't where it is 'needed' anyway, it is the invocations of 'stubgen' 2008-11-12 13:40 jkrell * m3-libs/m3core/src/unix/: linux-amd64/m3makefile, linux-sparc32/m3makefile, linux-sparc64/m3makefile: forgot to copy/commit m3makefile -- remove UpthreadMachine 2008-11-12 13:38 jkrell * m3-libs/m3core/src/unix/: linux-32/Upthreadtypes.i3, linux-32/m3makefile, linux-64/Upthreadtypes.i3, linux-64/m3makefile, linux-amd64/UpthreadMachine.i3, linux-i386/UpthreadMachine.i3, linux-i386/m3makefile, linux-libc6/Upthread.i3, linux-sparc32/UpthreadMachine.i3, linux-sparc64/UpthreadMachine.i3: fix alignment of pthread types array of char is not right This has not been seen to be a problem, but it is definitely highly suspicious. Unless perhaps record alignment is somehow always boosted beyond its constituents? (unlikely) perhaps these should be wrapped in arrays to avoid some undesired type equivalences? This is tedious/error prone, could use triple/quadruple checking. LINUXLIBC6 ./upgrade.py from a recent snapshot succeeded PPC_LINUX was already broken/crashing, not tested today (to be looked into) not likely any other Linux target in use (yet) only Linux platforms affected while at it, apparently fix SPARC32_LINUX, to use 32bit sizes instead of 64bit. Note that one of the types has "long long" alignment and therefore was never? properly implementable with an older compiler. Note also that bootstrapping will now require, if it didn't already, a compiler that supports LONGINT. (However this could be patched to an array of two INTEGERs for bootstrapping purposes.) note also this turned up a compiler internal error initializing an array of longint (later..) and some strange problems targeting I386_LINUX with an AMD64_LINUX front end -- the negative shifts in Unix.i3 are reported as not constant, and there are some errors in m3x86. (later..) (should be upthreadTypes.i3 or upthreadtypes.i3? I chose to follow what the C headers do) 2008-11-12 11:35 jkrell * m3-libs/m3core/src/C/MIPS64_OPENBSD/Csetjmp.i3, m3-sys/m3middle/src/Target.m3: It IS important to size the jmp_buf correctly, on pthreads systems, else large allocation (such as in M3ID's initializers) => garbage collector (or just via normal garbage accretion) => ProcessStacks => ProcessMe => RTMachine.SaveState => typicall setjmp, or getcontext Win32 doesn't use RTMachine.SaveState, it uses GetThreadContext. Ideally this size would not be repeated in both Csetjmp.i3 and Target.m3. Also change Target.m3 for this case to "look the same" -- use hex and integers/address instead of decimal/bytes (safe assumption that sizeof(integer)==sizeof(address)) 2008-11-12 09:52 jkrell * m3-sys/m3middle/src/Target.m3: fix size of MIPS64_OPENBSD jmp_buf, so M3ID main body doesn't corrupt stack; and alignment generally of 64 bit jump_bufs 2008-11-12 07:32 jkrell * m3-sys/m3middle/src/Target.m3: MIPS64 procedures are not aligned to the alignment an integer. This is actually a safe inefficient default, and very common on 64 bit non-x86 systems -- systems at which reading a code pointer as an integer needs to be careful about alignment. This is used by the generated code for the imho questionable sniffing to determine if a function pointer is a "closure" or not, by checking if it starts with the integer -1. 2008-11-12 04:58 jkrell * m3-sys/cminstall/src/config-no-install/SPARC64_SOLARIS: fix newlines and fix ignored error printed out when we go to probe the compiler 2008-11-12 04:41 jkrell * scripts/python/pylib.py: allow keep on command line 2008-11-12 04:27 jkrell * m3-libs/m3core/src/C/: AMD64_LINUX/Csetjmp.i3, SPARC32_LINUX/Csetjmp.i3, SPARC64_LINUX/Csetjmp.i3, SPARC64_OPENBSD/Csetjmp.i3, SPARC64_SOLARIS/Csetjmp.i3: fix jumpbuf sizes, but I again don't think it matters, and AMD64_LINUX was already the right size (comment only there) so alas that won't help with the intermittent but common problem running stubgen 2008-11-12 04:24 jkrell * m3-libs/m3core/src/C/NT386/Csetjmp.i3: actually I think it was ok -- the compiler allocates the jmpbuf for try blocks, so the runtime doesn't need to know the size...probably the others were ok but need to double double check 2008-11-12 04:21 jkrell * m3-libs/m3core/src/C/NT386/Csetjmp.i3: and then fix it too (huh/wow?) 2008-11-12 04:20 jkrell * m3-libs/m3core/src/C/NT386/Csetjmp.i3: go back a version here too 2008-11-11 14:27 jkrell * m3-libs/m3core/src/: C/AMD64_DARWIN/Csetjmp.i3, C/AMD64_LINUX/Csetjmp.i3, C/AMD64_LINUX/m3makefile, C/Common/Csetjmp.i3, C/Common/m3makefile, C/MIPS64_OPENBSD/Csetjmp.i3, C/MIPS64_OPENBSD/m3makefile, C/SPARC32_LINUX/Csetjmp.i3, C/SPARC32_LINUX/m3makefile, C/SPARC64_LINUX/Csetjmp.i3, C/SPARC64_LINUX/m3makefile, C/SPARC64_OPENBSD/Csetjmp.i3, C/SPARC64_OPENBSD/m3makefile, C/SPARC64_SOLARIS/Csetjmp.i3, C/SPARC64_SOLARIS/m3makefile, runtime/AMD64_LINUX/RTMachine.i3, runtime/SPARC32_LINUX/RTMachine.i3, runtime/SPARC64_LINUX/RTMachine.i3, runtime/SPARC64_SOLARIS/RTMachine.i3: back up a bunch -- Csetjmp is not dead and we need to adjust it perhaps 2008-11-11 14:05 jkrell * m3-libs/m3core/src/runtime/common/RTCollector.m3: fix warning 2008-11-11 13:58 jkrell * m3-libs/m3core/src/C/Common/m3makefile: oops 2008-11-11 13:57 jkrell * m3-libs/m3core/src/C/: AMD64_LINUX/Csetjmp.i3, AMD64_LINUX/Csignal.i3, AMD64_LINUX/m3makefile, Common/m3makefile, MIPS64_OPENBSD/Csetjmp.i3, MIPS64_OPENBSD/Csignal.i3, MIPS64_OPENBSD/Cstdio.i3, MIPS64_OPENBSD/Cstring.i3, MIPS64_OPENBSD/m3makefile, PPC32_OPENBSD/Csetjmp.i3, PPC32_OPENBSD/Csignal.i3, PPC32_OPENBSD/m3makefile, SPARC32_LINUX/Csetjmp.i3, SPARC32_LINUX/Csignal.i3, SPARC32_LINUX/m3makefile, SPARC64_LINUX/Csetjmp.i3, SPARC64_LINUX/Csignal.i3, SPARC64_LINUX/m3makefile, SPARC64_OPENBSD/Csetjmp.i3, SPARC64_OPENBSD/Csignal.i3, SPARC64_OPENBSD/m3makefile, SPARC64_SOLARIS/Csetjmp.i3, SPARC64_SOLARIS/Csignal.i3, SPARC64_SOLARIS/m3makefile: more common code, esp. on systems with no user threads (some of the previous commit was incomplete) 2008-11-11 13:41 jkrell * m3-libs/m3core/src/C/Common/Csetjmp.i3, m3-libs/m3core/src/C/Common/Cstdio.i3, m3-libs/m3core/src/C/Common/Cstring.i3, m3-libs/m3core/src/C/Common/m3makefile, m3-libs/m3core/src/C/NT386/Csetjmp.i3, m3-libs/m3core/src/C/NT386/Cstdio.i3, m3-libs/m3core/src/C/NT386/CstdioC.c, m3-libs/m3core/src/C/NT386/Cstring.i3, m3-libs/m3core/src/C/NT386/m3makefile, m3-libs/m3core/src/runtime/AMD64_LINUX/RTMachine.i3, m3-libs/m3core/src/runtime/NT386/RTMachine.i3, m3-libs/m3core/src/runtime/NT386GNU/RTMachine.i3, m3-libs/m3core/src/runtime/SPARC32_LINUX/RTMachine.i3, m3-libs/m3core/src/runtime/SPARC64_LINUX/RTMachine.i3, m3-libs/m3core/src/runtime/SPARC64_SOLARIS/RTMachine.i3, scripts/python/pylib.py: remove more residue of user thread support for systems on which it was never complete/working (and probably never will be) where possible -- which is fairly often -- share Csetjmp.i3 again, on systems without user thread support, so the size of jmp_buf isn't needed, and where longjmp == "_longjmp" (not NT386) move NT386 to common Cstdio.i3, Cstring.i3 use paranoidgc on AMD64_LINUX for now (pylib.py) 2008-11-11 12:47 jkrell * m3-libs/m3core/src/runtime/: AMD64_LINUX/RTMachine.i3, SPARC32_LINUX/RTMachine.i3, SPARC64_LINUX/RTMachine.i3: fix build-breakage from earlier cleanup 2008-11-10 14:41 jkrell * m3-libs/m3core/src/runtime/SPARC64_SOLARIS/RTMachine.i3, m3-sys/m3middle/src/Target.m3: fix two causes of alignment faults on SPARC64_SOLARIS, one checking if binders are closures (they never are, right? maybe this can be optimized?) and the second while collecting garbage 2008-11-10 14:40 jkrell * scripts/python/pylib.py: make bootstrap packaging less finicky -- the directory is often in use and can't be deleted, but all the files are and that suffices 2008-11-10 14:39 jkrell * m3-libs/m3core/src/text/TextLiteral.i3: let 32 bit hosts bootstrap 64 bit targets, until the real fix is made 2008-11-10 13:44 jkrell * m3-libs/libm3/src/os/POSIX/OSConfigPosix.m3, m3-libs/libm3/src/random/m3makefile, m3-libs/m3core/src/m3makefile, m3-libs/m3core/src/C/AMD64_LINUX/Cstdio.i3, m3-libs/m3core/src/C/AMD64_LINUX/Cstring.i3, m3-libs/m3core/src/C/AMD64_LINUX/m3makefile, m3-libs/m3core/src/C/Common/Cstdio.i3, m3-libs/m3core/src/C/Common/Cstring.i3, m3-libs/m3core/src/C/Common/m3makefile, m3-libs/m3core/src/C/PPC32_OPENBSD/Cstdio.i3, m3-libs/m3core/src/C/PPC32_OPENBSD/Cstring.i3, m3-libs/m3core/src/C/SPARC32_LINUX/Cstdio.i3, m3-libs/m3core/src/C/SPARC32_LINUX/Cstring.i3, m3-libs/m3core/src/C/SPARC64_LINUX/Cstdio.i3, m3-libs/m3core/src/C/SPARC64_LINUX/Cstring.i3, m3-libs/m3core/src/C/SPARC64_OPENBSD/Cstdio.i3, m3-libs/m3core/src/C/SPARC64_SOLARIS/Csetjmp.i3, m3-libs/m3core/src/C/SPARC64_SOLARIS/Csignal.i3, m3-libs/m3core/src/C/SPARC64_SOLARIS/m3makefile, m3-libs/m3core/src/Csupport/m3makefile, m3-libs/m3core/src/float/m3makefile, m3-libs/m3core/src/runtime/AMD64_LINUX/RTThread.m3, m3-libs/m3core/src/runtime/AMD64_LINUX/m3makefile, m3-libs/m3core/src/runtime/NT386GNU/m3makefile, m3-libs/m3core/src/runtime/POSIX/m3makefile, m3-libs/m3core/src/runtime/SPARC32_LINUX/RTThread.m3, m3-libs/m3core/src/runtime/SPARC32_LINUX/RTThreadC.c, m3-libs/m3core/src/runtime/SPARC32_LINUX/m3makefile, m3-libs/m3core/src/runtime/SPARC64_LINUX/RTThread.m3, m3-libs/m3core/src/runtime/SPARC64_LINUX/RTThreadC.c, m3-libs/m3core/src/runtime/SPARC64_LINUX/m3makefile, m3-libs/m3core/src/runtime/SPARC64_SOLARIS/RTMachine.i3, m3-libs/m3core/src/runtime/SPARC64_SOLARIS/RTSignal.m3, m3-libs/m3core/src/runtime/SPARC64_SOLARIS/m3makefile, m3-libs/m3core/src/runtime/common/Compiler.tmpl, m3-libs/m3core/src/thread/m3makefile, m3-libs/m3core/src/time/POSIX/m3makefile, m3-libs/m3core/src/unix/m3makefile, m3-sys/cminstall/src/config-no-install/SOLgnu, m3-sys/cminstall/src/config-no-install/SOLsun, m3-sys/cminstall/src/config-no-install/SPARC64_SOLARIS, m3-sys/cminstall/src/config-no-install/Solaris.common, m3-sys/m3cc/src/m3makefile, m3-sys/m3middle/src/Target.i3, m3-sys/m3middle/src/Target.m3, scripts/python/pylib.py: enough SPARC64_SOLARIS support to build a bootstrap package, build the bootstrap on the target, and startup cm3, which then hits an alignment fault in RTLinker__AddUnit. This shouldn't be hard to get past, it is doing ldx against an address that is not a multiple of 4. -bash-3.00$ dbx ./cm3 (dbx) run Running: cm3 (process id 3185) Reading libc_psr.so.1 t@1 (l@1) signal BUS (invalid address alignment) in RTLinker__AddUnit at 0x10036e028 0x000000010036e028: RTLinker__AddUnit+0x001c: ldx [%g1], %g1 Current function is main 12 RTLinker__InitRuntime (argc, argv, envp, 0); (dbx) where current thread: t@1 [1] RTLinker__AddUnit(0x10036d984, 0x0, 0x0, 0x0, 0x0, 0x0), at 0x10036e028 [2] RTLinker__InitRuntime(0x1, 0xffffffff7ffffb78, 0xffffffff7ffffb88, 0x0, 0x0, 0x0), at 0x10036db18 =>[3] main(argc = 1, argv = 0xffffffff7ffffb78, envp = 0xffffffff7ffffb88), line 12 in "_m3main.c" (dbx) print $g1 $g1 = 4298561924ULL This gives us another big endian 64 bit system, rather than only MIPS64_OPENBSD. Also remove a sh wrapper on NT in m3cc. Otherwise I was going to rename the file to include M3CC_TARGET, so concurrent builds don't step on each other. Significant working and reworking of config-no-install/*sol*, though ultimately this will probably be trimmed down substantially once I use multilibed gcc. (Note that SOLsun is easy, if you dig past the gcc stuff, and that SPARC64_SOLARIS is using the Sun compiler also, but Solaris now factored like NT so that you can say C_COMPILER = "SUN" or C_COMPILER = "GNU" (at least my config files are; I didn't touch the others). Presumably fix the building of a few minor platforms from earlier tonight, and then some -- go ahead and remove the dummy RTThread.i3 and RTThread.m3 on platforms that just had stubs. (I broke them by removing RTMachine.StackTopPad or such, presumably; these are mostly "immature" platforms, except I386_CYGWIN) Note that I cheated in building my m3cg and hand patched auto-host.h rather than wait for the configure command line change to trickle through. It should be equiv. Move the common Cstdio.i3 and Cstring.i3 to common. This can be more all platforms later, but for now is only for some. Also the Solaris "unix" directory needs to be revisited, probably start over from a minimal Cygwin or OpenBSD base, at least for new Solaris targets. Also new files are likely to have carriage returns, but preexisting files shouldn't. 2008-11-10 08:03 jkrell * m3-libs/m3core/src/runtime/: AMD64_LINUX/RTMachine.i3, NT386/RTMachine.i3, NT386GNU/RTMachine.i3, SPARC32_LINUX/RTMachine.i3, SPARC64_LINUX/RTMachine.i3: FramePadBottom and FramePadTop are also unused on platforms that only have pthreads/kernel/NT threads 2008-11-10 07:59 jkrell * m3-libs/m3core/src/runtime/: AMD64_LINUX/RTMachine.i3, NT386/RTMachine.i3, NT386GNU/RTMachine.i3, SPARC32_LINUX/RTMachine.i3, SPARC64_LINUX/RTMachine.i3: remove unused StackFrameAlignment, it is only for runtimes that allocate their own stacks, like posix/user/vtalarm threads, not pthreads/kernel/NT threads 2008-11-09 23:22 jkrell * m3-sys/cminstall/src/: config/NT386.common, config-no-install/SOLgnu, config-no-install/SOLsun, config-no-install/Unix.common, config-no-install/cm3cfg.common: cleanup and combine config files linking gcc output with MS linker doesn't work, remove support because they differ in the name of the __ImageBase symbol "gcc output" desired was libcygwin.a, along with integrated backend output GNU ld throws out the integrated backend symbol The failed goal was to use integrated backend for I386_CYGWIN, and be debuggable with windbg/cdb. GNU ld is now on my path on Solaris, use /usr/ccs/bin/ld directly gcc is not working for me for linking on Solaris, use /usr/ccs/bin/ld directly (something simple, unresolved symbol "main", I must not be telling it correctly to build a shared object) gcc on my path is sparc64 targeting and not multilib enabled, so use {target}-gcc explicitly, where target is sparc-sun-solaris2.10; {target}-gcc is always reasonable, though {target} may have a non-canonical name like spar-solaris, or be very site specific like the "2.10" part (sparc64...) move portable parts to cm3cfg.common, even if they aren't used there 2008-11-09 23:13 jkrell * scripts/sysinfo.sh: let user set CM3_TARGET environment variable himself -- in order to let it be SOLsun instead of SOLgnu 2008-11-09 23:07 jkrell * m3-sys/cminstall/src/OS.i3: fix 2008-11-09 16:43 jkrell * m3-sys/m3cc/gcc/gcc/m3cg/parse.c: fix reading integers that don't fit in a "host wide int" since host wide int is usually 64 bits, this doesn't really fix anything, but the code is bothering to read two host wide ints, so make it appear to do so correctly (unless I'm wrong..) fix tracing of integers 2008-11-09 13:33 jkrell * m3-sys/cminstall/src/config-no-install/: MIPS64_OPENBSD, cm3cfg.common: fix small but fatal error -- setting m3back prevented the probe 2008-11-09 11:36 jkrell * m3-sys/cminstall/src/config-no-install/: AMD64_LINUX, cm3cfg.common: fix new probing functionality, and comment as to it needing more work to trim the probes 2008-11-09 11:05 jkrell * m3-sys/: cminstall/src/OS.i3, m3middle/src/M3File.m3, cminstall/src/Main.m3: use a more direct way to determine 'OnUnix' 2008-11-09 10:53 jkrell * m3-sys/m3quake/src/QMachine.m3, scripts/python/pylib.py: fix bootstrapping from older versions such as 5.5 -- older Win32 runtimes don't understand forward slashes in paths 2008-11-09 10:06 jkrell * m3-libs/libm3/src/os/POSIX/OSConfigPosix.m3, m3-libs/libm3/src/random/m3makefile, m3-libs/m3core/src/C/m3makefile, m3-libs/m3core/src/C/MIPS64_OPENBSD/Csetjmp.i3, m3-libs/m3core/src/C/MIPS64_OPENBSD/Csignal.i3, m3-libs/m3core/src/C/MIPS64_OPENBSD/Cstdio.i3, m3-libs/m3core/src/C/MIPS64_OPENBSD/Cstring.i3, m3-libs/m3core/src/C/MIPS64_OPENBSD/m3makefile, m3-libs/m3core/src/C/SPARC32_LINUX/Cstring.i3, m3-libs/m3core/src/C/SPARC64_LINUX/Cstring.i3, m3-libs/m3core/src/C/SPARC64_OPENBSD/Csetjmp.i3, m3-libs/m3core/src/C/SPARC64_OPENBSD/Csignal.i3, m3-libs/m3core/src/C/SPARC64_OPENBSD/Cstdio.i3, m3-libs/m3core/src/C/SPARC64_OPENBSD/m3makefile, m3-libs/m3core/src/Csupport/m3makefile, m3-libs/m3core/src/float/m3makefile, m3-libs/m3core/src/runtime/m3makefile, m3-libs/m3core/src/runtime/common/Compiler.tmpl, m3-libs/m3core/src/thread/m3makefile, m3-libs/m3core/src/time/POSIX/m3makefile, m3-libs/m3core/src/unix/m3makefile, m3-sys/m3cc/src/m3makefile, m3-sys/m3middle/src/Target.i3, m3-sys/m3middle/src/Target.m3: initial MIPS64_OPENBSD support There are least two problems, progress is made by switching #if's in parse.c, but there is still a crash after that. hack on gcc a bit less use : for null command instead of echo for now, multiply out some small files instead of sharing in an unusual way later will move these to a "common" directory 2008-11-09 09:56 jkrell * m3-libs/libm3/src/os/POSIX/OSConfigPosix.m3: fix newlines 2008-11-09 09:17 jkrell * m3-sys/m3middle/src/: Target.i3, Target.m3: remove Checks_integer_ops, it isn't used remove bitfields_can_overlap, it isn't used factor out commonalities PCC_bitfield_type_matters is typicall true little endian is typical true address always is word, except for cg_type Structure_size_boundary is typically 8 Setjmp is typicall _setjmp This reduces the amount of target-specific code, but maybe is the wrong choice, since each new port does have to consider the values and just letting e.g. little endian default is not safe. The others are more safe. 2008-11-09 08:06 jkrell * scripts/python/pylib.py: fix 2008-11-09 08:05 jkrell * scripts/python/pylib.py: this is Python, not Quake 2008-11-09 07:57 jkrell * m3-sys/m3middle/src/: Target.i3, Target.m3: Guard_page_size and Fixed_frame_size are never used, remove them 2008-11-09 07:54 jkrell * m3-sys/m3middle/src/Target.i3: fix newlines 2008-11-09 07:42 jkrell * m3-sys/m3middle/src/Target.m3: set the default first readable address to 4k; anything smaller than a page, on systems with an mmu, is suspicious, the FreeBSD value is suspicious since it forgets to convert a byte size to a bit size; however 0 is a safe value for this, it just causes extra null checks 2008-11-09 07:37 jkrell * m3-sys/m3middle/src/Target.m3: set AMD64_LINUX first readable address to 4k, removes many checks for null 2008-11-09 07:34 jkrell * m3-sys/m3middle/src/Target.m3: fix newlines 2008-11-09 07:10 jkrell * m3-sys/cminstall/src/: config/cm3cfg.common, config-no-install/Unix.common, config-no-install/cm3cfg.common: more fix and factor, and add stub for NT386 I thought I had just added 2008-11-09 06:35 jkrell * m3-sys/cminstall/src/config-no-install/: PPC_DARWIN, SOLgnu, Unix.common: some small repair and cleanup 2008-11-09 06:24 jkrell * scripts/python/pylib.py: improved bootstrapping support - copy likely to be changed source files around - fix the output extension of .c files so incrementality works add cm3cfg.common 2008-11-09 06:20 jkrell * m3-sys/cminstall/src/: config/NT386.common, config/cm3.cfg, config-no-install/AMD64_LINUX, config-no-install/LINUXLIBC6, config-no-install/MIPS64_OPENBSD, config-no-install/PPC32_OPENBSD, config-no-install/PPC_DARWIN, config-no-install/PPC_LINUX, config-no-install/SOLgnu, config-no-install/SOLsun, config-no-install/SPARC32_LINUX, config-no-install/SPARC64_LINUX, config-no-install/SPARC64_OPENBSD, config-no-install/Unix.common, config-no-install/cm3cfg.common: begin combining nt386.common and unix.common (into cm3cg.common) mips64_openbsd support being migrating config-no-install/solgnu and solsun to Unix.common (had this sitting around, in intermediate state) go back to old formating because that is all the old C written m3build could handle add probing for host/target specific cm3cg (so I can stop copying it around as I switch targets) 2008-11-07 03:08 rodney * m3-sys/m3gdb/gdb/gdb/m3-util.c: Fix a bug in which, under certain conditioins, m3gdb could not find a global variable having a procedure type, e.g. RTAllocator.callback. 2008-11-06 07:10 wagner * www/: nav-doc.html, news.html, ship-cm3-www: add m3gdb links 2008-11-06 07:06 wagner * doc/: ship-cm3-www-doc, help/cm3/cm3-quickref.html: add cm3 quickref 2008-11-06 06:36 wagner * doc/help/: cm3/cm3.html, cm3/debugging.html, cm3/gc.html, m3gdb/m3gdb-onepage.html: update m3gdb references 2008-11-05 15:48 rodney * doc/help/m3gdb/m3gdb.html: needed stylesheet and better name 2008-11-05 15:45 rodney * doc/help/m3gdb/m3gdb-onepage.html: Initial commit of new m3gdb documentation. 2008-11-05 14:18 rodney * doc/help/m3gdb/: m3gdb.docbook, m3gdb.html: Initial commit of new m3gdb documentation. 2008-11-04 20:34 rodney * m3-sys/m3gdb/gdb/gdb/: m3-lang.c, m3-util.c, m3-util.h: Make some int results that are truly boolean into bool. 2008-11-04 02:02 rodney * m3-sys/m3gdb/gdb/gdb/: linespec.c, m3-eval.c, m3-exp.c, m3-lang.c, m3-util.c: 1) Fix start command in CM3 to use Main_I3 as its breakpoint. This fixes wrong behaviour. 2) Some consistency fixes to unqualified identifier lookup in M3 linespecs. 3) Misc. minor fixes to comments, messages, and formatting. 2008-11-02 12:13 jkrell * m3-libs/m3core/src/thread/m3makefile: fix newlines 2008-11-02 10:46 jkrell * m3-libs/m3core/src/unix/cygwin/: Uerror.i3, Uerror.i3.cpp: reduce error prone repitition, like in the OpenBSD case 2008-11-02 10:41 jkrell * m3-libs/m3core/src/unix/openbsd-common/: Uerror.i3, Uerror.i3.cpp: Make sure Uerror.i3 is correct -- generate it from C. 2008-10-31 13:41 jkrell * scripts/python/make-dist.py: stick with gzip, like OpenBSD, as bzip2 is slow and/or memory hungry (lzma is about as fast as gzip, and compresses even much better than bzip2, maybe use that, but it isn't very widespread) 2008-10-31 13:36 jkrell * scripts/python/pylib.py: leave garbage collection on now; it works! Thanks Tony. 2008-10-31 11:23 jkrell * m3-libs/m3core/src/runtime/: AMD64_LINUX/RTMachine.i3, SPARC32_LINUX/RTMachine.i3, SPARC64_LINUX/RTMachine.i3: some opportunistic vertical density 2008-10-31 11:20 jkrell * m3-libs/m3core/src/runtime/: AMD64_LINUX/RTMachine.i3, SPARC32_LINUX/RTMachine.i3, SPARC64_LINUX/RTMachine.i3: fix StackFrameAlignment and PointerAlignment per Tony (need to write some autoconf-ish tests to derive these) 2008-10-30 11:13 jkrell * scripts/python/pylib.py: add not yet active Irix support, turn off garbage collection on some new targets pending debugging, some of them have definitely had problems, though perhaps not all 2008-10-30 10:44 jkrell * m3-sys/m3cc/gcc/gcc/m3cg/parse.c: do not overwrite the global type nodes; this is why 64 bit host m3cg tends to not work 2008-10-20 15:46 hosking * m3-libs/m3core/src/runtime/common/RTCollector.m3: Use @M3nogc if you don't trust the garbage collector. Otherwise, fix the bug in your port. 2008-10-16 18:11 jkrell * m3-sys/m3cc/gcc/gcc/m3cg/parse.c: alter tracing slightly 2008-10-16 17:11 jkrell * m3-sys/m3cc/gcc/gcc/m3cg/parse.c: some predantic printf fixes -- int vs. long, and some logging that shows where 32 bit and 64 bit hosts diverge 2008-09-30 14:11 jkrell * m3-sys/m3cc/src/m3makefile: use M3CC_HOST instead of TARGET for the first part of a cross directory name; usually these are the same 2008-09-30 13:55 jkrell * m3-sys/cminstall/src/config-no-install/LINUXLIBC6: forgot to save file -- also remove unconditional -m32 from cm3cg invocation since 5.2.6 does not support it; later code adds it in if it is supported 2008-09-30 13:54 jkrell * m3-sys/cminstall/src/config-no-install/: LINUXLIBC6, PPC_LINUX: fix bootstrapping LINUX_PPC from 5.2.6 by using shared code; have Linux/x86 use ar -s instead of separate invocation of ranlib 2008-09-25 23:22 rodney * m3-sys/: m3cc/gcc/gcc/dbxout.c, m3cc/gcc/gcc/tree-nested.c, m3gdb/gdb/gdb/m3-eval.c, m3gdb/gdb/gdb/m3-util.c, m3gdb/gdb/gdb/m3-valprint.c: gcc 4.3.0, from which the latest m3cc is derived, raises debugger hostility to new heights as far as finding static links and their target activation records is concerned. It could even be necessary to rework m3cc and m3gdb to use dwarf2 or dwarf3 to fix this. This commit is a temporary, incomplete fix. As long as nested procedures are called only as procedure constants, it should work. If a nested procedure is called through a formal parameter, m3gdb might find the wrong instance of a statically containing procedure, when accessing the containing procedure's variables nonlocally. For now, m3gdb warns whenever accessing variables nonlocally. NOTE: There are additional, related problems with debug output as a result of gcc 4.3.0's handling of nonlocal variable access. If a formal parameter is accessed nonlocally anywhere in the program, accessing it locally in m3gdb by 'info arg', 'frame', or 'backtrace' gives a wrong value, while 'info loc' will give two values for a parameter, only one of which is right. The 'print' command appears to be right in such cases, as far as tested to date. 2008-09-15 16:21 jkrell * scripts/python/make-dist.py: don't delete cm3 here 2008-09-15 15:52 jkrell * scripts/python/make-dist.py: don't delete cm3cg here either 2008-09-15 14:17 jkrell * scripts/python/make-dist.py: don't delete cm3cg here 2008-09-15 09:15 jkrell * m3-sys/m3cc/src/m3makefile: mkdir fix for NT cd .. && mkdir foo instead of mkdir ../foo in order to avoid chosing a slash Modula-3 doesn't use the decimal type, so neuter that code in gcc and don't bother building libdecnumber, just a dummy don't build makedepend.exe 2008-09-03 21:05 wagner * m3-sys/cm3ide/src/m3overrides: add missing override for tcp 2008-08-22 14:32 rcoleburn * m3-ui/vbtkit/src/lego/WIN32/ScrollerVBTClass.m3: The following changes greatly improve appearance and functionality of scroll bars on Windows platforms. Internally, deal with all sizes in pixels, vs. millimeters. Provide conversions between millimeters and pixels as needed. Since FormsVBT scales pixmaps by default, scale the slider and trough to match the size of the button pixmaps. Change scroll behavior to more closely resemble user expectation in Windows. Formerly, scrolling was proportional to distance between mouse click and thumb. Now, clicking up/down buttons scrolls backward/forward by one line, and clicking in the slider trough above/below the thumb scrolls backward/forward by one window at a time. Unfortunately, the interface comments still reference the proportional behavior, but that behavior is still done in the POSIX implementation, so the interface comments are correct for POSIX. Perhaps we should add a comment about the Windows implementation; after all, the interface does say the implementation is free to change the behavior.---Randy Coleburn 2008-08-22 14:06 rcoleburn * m3-ui/vbtkit/src/lego/: Image.i3, Image.m3: Add comment about Raw.x/yres referring to the resolution at which the original image was constructed. Add procedure ForceUnScaledBehavior(). ---Randy Coleburn 2008-08-14 15:05 jkrell * scripts/: find-packages.sh, win/find-packages.cmd: go back a version unless proven to be needed 2008-08-14 14:54 jkrell * scripts/: find-packages.sh, pkginfo.txt, python/make-dist.py, python/pylib.py, win/find-packages.cmd: add webcat, deskscape, webscape they might not be particularly useful, but they do build and startup 2008-08-14 14:26 jkrell * m3-ui/ui/src/winvbt/WinTrestle.m3: release locks when calling out to Win32 to prevent recursive lock acquisition. correct? 2008-08-13 11:10 jkrell * m3-ui/vbtkit/src/lego/WIN32/ScrollerVBTClass.m3: fix it to compile, at least and eliminate common subexpressions will see if we can get Windows to draw much more of this using DrawFrameControl 2008-08-13 09:41 jkrell * m3-ui/vbtkit/src/lego/: Image.i3, Image.m3, ImageInit.i3, m3makefile, POSIX/ImageInit.m3, POSIX/m3makefile, WIN32/ImageInit.m3, WIN32/m3makefile: go back one version -- back to hard coding Image xres/yres = 75 and leaving them publically accessible 2008-08-11 10:43 jkrell * m3-ui/vbtkit/src/lego/WIN32/ImageInit.m3: When I am copy/pasting code I don't really understand, it helps to look for more nearby code that should also probably be copied. Specifically: Call ReleaseDC after GetDC. 2008-08-09 09:08 wagner * m3-sys/cm3ide/src/m3overrides: fix overrides (adapt to cm3 build conventions) 2008-08-09 09:03 wagner * m3-ui/vbtkit/src/lego/POSIX/ImageInit.m3: make it compile (fix syntax error) 2008-08-09 08:50 rcoleburn * m3-ui/vbtkit/src/lego/WIN32/ScrollerVBTClass.m3: Revert to prior version. My idea about truncating the millimeters was wrong, even though I thought it tested out ok. Sorry. --R.Coleburn 2008-08-09 08:02 rcoleburn * m3-ui/vbtkit/src/lego/WIN32/ScrollerVBTClass.m3: Ensure millimeters are whole, not fractional.---R.Coleburn 2008-08-09 05:09 rcoleburn * m3-ui/vbtkit/src/lego/WIN32/ScrollerVBTClass.m3: Compute scroll bar slider width based on screen resolution rather than making it a fixed size in millimeters. Allows scroll bar to adapt to different screen resolutions. Used new procedures in ImageInit provided by Jay Krell.---Randy Coleburn 2008-08-08 13:46 jkrell * m3-ui/vbtkit/src/lego/ImageInit.i3: forgot to add file 2008-08-08 13:45 jkrell * m3-libs/m3core/src/win32/WinNT.i3: small Win64 fix (not relevant to anything at this time) 2008-08-08 13:42 jkrell * m3-libs/m3core/src/win32/m3makefile: include WinNTc.c in all platforms, I expect it is needed 2008-08-08 13:38 jkrell * m3-ui/vbtkit/src/lego/WIN32/ImageInit.m3: remove unused import (I was going to use InterlockedIncrement, but that is perhaps overkill, and unsafe apparently) 2008-08-08 13:34 jkrell * m3-ui/vbtkit/src/lego/: Image.i3, Image.m3, m3makefile, POSIX/ImageInit.m3, POSIX/m3makefile, WIN32/ImageInit.m3, WIN32/m3makefile: make all uses of Image.Raw.xres and yres go through accessor functions in fact, renaming xres/yres so any failure to catch uses will fail to compile On Windows, compute xres and yres from Devicecaps(desktop window) and GetSystemMetrics on X Windows, keep the same old 75 (but this looks not difficult to fix) this has a small dependency on WinNT* change from a minute ago 2008-08-08 13:29 jkrell * m3-libs/m3core/src/win32/: WinDef.i3, WinNT.i3, WinNTc.c, WinUser.i3, WinUser.m3, m3makefile: some header file updates WinNT* in particular is used by the upcoming pixmap change (I'd say "fix" but I can't test it, no high dpi system available, nor perhaps apps with noticable deformity) 2008-08-08 13:14 jkrell * m3-sys/cminstall/src/config/NT386.common: fix this: LINK : warning LNK4044: unrecognized option '/cm3\lib\hand.obj'; ignored 2008-08-05 10:47 jkrell * m3-sys/m3cc/src/m3makefile: dependency on libiconv.so still present, use the previous code 2008-08-03 09:44 wagner * m3-ui/vbtkit/src/: m3overrides, etext/MacModel.m3, etext/TextPort.m3: tentative fix for text selection and insertion in the Mac TextPortModel: The strange behaviour reported by Randy Coleburn (mouse click, left, left, insert --> first character inserted at mouse click point) should now be gone; cursor up and down at the right end of ragged paragraphs still looks a bit strange, but that seems to be not worse than before. It still should be fixed though. These changes should be reviewed and tested by others; they seem to do the right thing for me, but I haven't done extensive testing, especially in other modes. 2008-08-02 01:36 jkrell * m3-ui/vbtkit/src/vbtkitutils/VBTKitEnv.m3: allow getting the emacs text editing model via the environment variable 2008-08-01 13:46 jkrell * m3-ui/: anim3D/src/win-opengl/Win_OpenGL_Base.m3, ui/src/winvbt/WinMsg.m3: trace a little more, sure could use some code consolidation here.. 2008-08-01 13:43 jkrell * m3-libs/m3core/src/win32/WinUser.i3: a few more updates, still quite out of date but arguably that is just fine -- only declare what the m3 libs need 2008-08-01 13:17 jkrell * m3-libs/m3core/src/win32/: WinGDI.i3, WinGDI.m3, WinUser.i3, WinUser.m3: some header updates and repair not all of it compatible -- identifiers not in headers generally removed "max" identifiers that aren't constants, removed The Modula-3 headers appear to be from VERY old Windows headers, possibly NT 3.1 beta since the level of stuff the Modula-3 headers make up that isn't in headers is surprising. only added result of findstr #define on winuser.h did not check if things #if'ed out did not add new functions this triggered by alarm over unrecognized output of windows message tracing, which was merely a message that didn't exist long ago when this stuff was written ie: pre NT 4 -- by a lot I think. always normalize some booleans to true/false the extra check of is_chicago is not worthwhile not worth the cost to initialize and use the global though Modula-3 ought to define boolean as a full 32 bit integer that is 0 or non-zero like C... probably should just do this for all extern functions????? it is not really probably a bug for C to return other values for true note that Modula-3 doesn't have the "#if WINVER >= 0x4000" stuff. everything is unconditionally declared m3core.dll et. al. should probably take care to GetProcAddress anything they use that was added after some agreed upon lowest common denominator all this header duplication could really use some automation/verification none introduced here today.. 2008-08-01 09:46 jkrell * m3-ui/ui/src/winvbt/WinScrnColorMap.m3: fix warning 2008-08-01 06:02 jkrell * scripts/python/pylib.py: also the D: drive, probably should check SystemDrive environment variable and iterate a..z or c..z 2008-08-01 06:00 jkrell * scripts/python/pylib.py: use another popular (at least with me) guess for where the cm3 executable is -- /cm3/bin/cm3 2008-07-10 06:50 wagner * www/ship-cm3-www: add selective shipping 2008-07-10 06:26 wagner * www/: news.html, start.html: add some CM3IDE info 2008-07-05 15:26 rodney * m3-ui/vbtkit/src/vbtkitutils/Rsrc.i3: Fix minor typos in comments. 2008-07-04 22:58 rodney * m3-sys/cm3/src/: Builder.m3, Msg.m3: Add some helpful error messages to cm3. Previously, there were several places where separate executables (e.g. cm3cg, as, gcc, etc.) could fail silently. The only message given would be a "not building.." at the end, with no clue what unit or program failed. 2008-07-04 19:45 jkrell * m3-sys/m3cc/src/m3makefile: cut libiconv dependency a simpler way, albeit a bit unclear that this works or is meant to work 2008-07-01 20:46 wagner * scripts/pkginfo.txt: uniquely identify the web package to build 2008-06-21 14:57 rcoleburn * m3-sys/cm3ide/UsersGuide/CM3_IDE_UsersGuide.doc, m3-sys/cm3ide/UsersGuide/CM3_IDE_UsersGuide.pdf, m3-sys/cm3ide/UsersGuide/more-info.pdf, m3-sys/cm3ide/UsersGuide/user-guide.pdf, doc/help/CM3_IDE/more-info.pdf, doc/help/CM3_IDE/user-guide.pdf: Provide description of Tony's paper in chapter 8.--R.Coleburn 2008-06-21 05:40 rcoleburn * examples/browsing/src/index.html: [no log message] 2008-06-21 04:40 rcoleburn * scripts/pkginfo.txt: Add cm3ide to list of packages.--R.Coleburn 2008-06-21 04:30 rcoleburn * m3-sys/cm3ide/: UsersGuide/CM3_IDE_ComposedFigures.ppt, UsersGuide/CM3_IDE_Screenshots_Figures.doc, UsersGuide/CM3_IDE_UsersGuide.doc, UsersGuide/CM3_IDE_UsersGuide.pdf, UsersGuide/ReadMe.txt, UsersGuide/basics.pdf, UsersGuide/beyond-basics.pdf, UsersGuide/customization.pdf, UsersGuide/environment.pdf, UsersGuide/further-info.pdf, UsersGuide/interface-index.pdf, UsersGuide/intro.pdf, UsersGuide/more-info.pdf, UsersGuide/packages.pdf, UsersGuide/recipes.pdf, UsersGuide/user-guide.pdf, src/Main.m3, src/ReadMe.html, src/m3makefile, src/m3overrides, src/forms/Config.i3, src/forms/Config.m3, src/forms/Form.i3, src/forms/Form.m3, src/forms/NewPkg.i3, src/forms/NewPkg.m3, src/forms/Rescan.i3, src/forms/Rescan.m3, src/forms/m3makefile, src/markup/CMarkUp.i3, src/markup/CMarkUp.m3, src/markup/M3MarkUp.i3, src/markup/M3MarkUp.m3, src/markup/M3Sym.i3, src/markup/M3Sym.m3, src/markup/MarkUp.i3, src/markup/MarkUp.m3, src/markup/Marker.i3, src/markup/XFormat.i3, src/markup/XFormat.m3, src/markup/m3makefile, src/misc/BrowserDB.i3, src/misc/BrowserDB.m3, src/misc/Builder.i3, src/misc/Builder.m3, src/misc/ConfigItem.i3, src/misc/ConfigItem.m3, src/misc/Default.i3, src/misc/Default.m3, src/misc/Display.i3, src/misc/Display.m3, src/misc/Editor.i3, src/misc/Editor.m3, src/misc/HTML.i3, src/misc/HTML.m3, src/misc/LineWr.i3, src/misc/LineWr.m3, src/misc/MxExpire.i3, src/misc/MxExpire.m3, src/misc/UserState.i3, src/misc/UserState.m3, src/misc/WebServer.i3, src/misc/WebServer.m3, src/misc/m3makefile, src/nodes/BuildCache.i3, src/nodes/BuildCache.m3, src/nodes/ClassDir.i3, src/nodes/ClassDir.m3, src/nodes/Decl.i3, src/nodes/Decl.m3, src/nodes/Derived.i3, src/nodes/Derived.m3, src/nodes/Dir.i3, src/nodes/Dir.m3, src/nodes/FileDir.i3, src/nodes/FileDir.m3, src/nodes/FileNode.i3, src/nodes/FileNode.m3, src/nodes/Fixed.i3, src/nodes/Fixed.m3, src/nodes/Node.i3, src/nodes/Node.m3, src/nodes/Pkg.i3, src/nodes/Pkg.m3, src/nodes/PkgRoot.i3, src/nodes/PkgRoot.m3, src/nodes/Roots.i3, src/nodes/Roots.m3, src/nodes/Source.i3, src/nodes/Source.m3, src/nodes/Type.i3, src/nodes/Type.m3, src/nodes/m3makefile, src/rsrc/CM3_IDE.gif, src/rsrc/CM3_IDE.html, src/rsrc/CM3_IDE_Icons.ico, src/rsrc/CM3_IDE_iconSource.bmp, src/rsrc/CM3_IDE_iconSource.gif, src/rsrc/CM3_IDE_large.gif, src/rsrc/CM3_IDE_sm.bmp, src/rsrc/CM3_IDE_sm.gif, src/rsrc/CM3_IDE_small.gif, src/rsrc/CM3_IDE_xx.ico, src/rsrc/about.html, src/rsrc/arrow-down.gif, src/rsrc/arrow-left.gif, src/rsrc/arrow-right.gif, src/rsrc/arrow-up.gif, src/rsrc/authors.html, src/rsrc/browse-pkg.gif, src/rsrc/browse-root.gif, src/rsrc/browse.gif, src/rsrc/build-pkg.gif, src/rsrc/build-root.gif, src/rsrc/cache.gif, src/rsrc/class.gif, src/rsrc/cmass.gif, src/rsrc/com.gif, src/rsrc/config.gif, src/rsrc/confighelp.html, src/rsrc/const.gif, src/rsrc/debug.gif, src/rsrc/debug2.gif, src/rsrc/digital-license.html, src/rsrc/dir.gif, src/rsrc/doc.gif, src/rsrc/doc_0.gif, src/rsrc/edit.gif, src/rsrc/error.gif, src/rsrc/ex.gif, src/rsrc/except.gif, src/rsrc/expired.html, src/rsrc/expiring.html, src/rsrc/file.gif, src/rsrc/gformal.gif, src/rsrc/help.gif, src/rsrc/home.gif, src/rsrc/html.gif, src/rsrc/iconRes.rc, src/rsrc/info.gif, src/rsrc/info.html, src/rsrc/intro.gif, src/rsrc/lib.gif, src/rsrc/license.html, src/rsrc/m3makefile, src/rsrc/man.gif, src/rsrc/new-pkg.html, src/rsrc/pdf.gif, src/rsrc/pgm.gif, src/rsrc/pkg.gif, src/rsrc/proc-body.gif, src/rsrc/proc-decl.gif, src/rsrc/proc.gif, src/rsrc/ps.gif, src/rsrc/ref.gif, src/rsrc/roots.gif, src/rsrc/rsrc.gif, src/rsrc/run.gif, src/rsrc/run2.gif, src/rsrc/ship.gif, src/rsrc/start.html, src/rsrc/start1.html, src/rsrc/start2.html, src/rsrc/tut.gif, src/rsrc/type-decl.gif, src/rsrc/type.gif, src/rsrc/unknown.gif, src/rsrc/unknown2.gif, src/rsrc/util.gif, src/rsrc/util.html, src/rsrc/var.gif, src/rsrc/warn.gif, src/rsrc/what.gif, src/rsrc/x-c.gif, src/rsrc/x-h.gif, src/rsrc/x-i3.gif, src/rsrc/x-ig.gif, src/rsrc/x-m3.gif, src/rsrc/x-mg.gif, src/rsrc/x-misc.gif, src/rsrc/x-quake.gif, src/rsrc/_templates/circle-empty.gif, src/rsrc/_templates/circle.gif, src/rsrc/_templates/diamond.gif, src/rsrc/_templates/square.gif, src/rsrc/_templates/triangle.gif, src/rsrc/originalArtWork/CM3_IDE_OrigArtWork.ppt, src/rsrc/originalArtWork/CM3_IDE_icon_uncropped.TIF, src/rsrc/originalArtWork/CM3_IDE_icon_uncropped.gif, src/rsrc/originalArtWork/CM3_IDE_icon_uncropped.jpg, src/rsrc/originalArtWork/CM3_IDE_icon_uncropped.png, src/rsrc/originalArtWork/CM3_IDE_uncropped.gif, src/rsrc/originalArtWork/CM3_IDE_uncropped.jpg, src/rsrc/originalArtWork/CM3_IDE_uncropped.png, src/rsrc/originalArtWork/CM3_IDE_uncropped.tif, src/rsrc/originalArtWork/catalystOrigArtWork.ppt, src/server/TCPServer.i3, src/server/TCPServer.m3, src/server/m3makefile, src/utils/Buf.i3, src/utils/Buf.m3, src/utils/ErrLog.i3, src/utils/ErrLog.m3, src/utils/ID.i3, src/utils/ID.m3, src/utils/LexMisc.i3, src/utils/LexMisc.m3, src/utils/Loc.i3, src/utils/Loc.m3, src/utils/OS.i3, src/utils/OS.m3, src/utils/RegExpr.i3, src/utils/RegExpr.m3, src/utils/Text2.i3, src/utils/Text2.m3, src/utils/Wx.i3, src/utils/Wx.m3, src/utils/m3makefile: Open-source release of CM3-IDE. Enjoy!--R.Coleburn NOTE: Do not change/remove the copyright legend that appears at the bottom of each page served by CM3-IDE. This legend is a requirement for the open-source release. 2008-06-21 04:22 rcoleburn * examples/: README, index.html, browsing/src/index.html, browsing/src/m3makefile, c-calling-m3/src/CcallsM3.m3, c-calling-m3/src/Cstuff.c, c-calling-m3/src/Cstuff.i3, c-calling-m3/src/index.html, c-calling-m3/src/m3makefile, calling-c-unix/src/CallingC.m3, calling-c-unix/src/Lib.i3, calling-c-unix/src/Lib.m3, calling-c-unix/src/Ulib.i3, calling-c-unix/src/index.html, calling-c-unix/src/m3makefile, calling-c-win32/src/OK.m3, calling-c-win32/src/WinLib.m3, calling-c-win32/src/index.html, calling-c-win32/src/m3makefile, copy/src/Copy.m3, copy/src/FakeOS.i3, copy/src/FakeOS.m3, copy/src/index.html, copy/src/m3makefile, enum/src/ExtensionLookup.m3, enum/src/index.html, enum/src/m3makefile, exceptions/src/Copy.m3, exceptions/src/Exceptions.m3, exceptions/src/FakeOS.i3, exceptions/src/FakeOS.m3, exceptions/src/index.html, exceptions/src/m3makefile, files/src/DirectoryListing.m3, files/src/index.html, files/src/m3makefile, finger/src/Finger.m3, finger/src/index.html, finger/src/m3makefile, fingerprint/src/M3Compare.m3, fingerprint/src/index.html, fingerprint/src/m3makefile, generics/src/Action.i3, generics/src/Action.m3, generics/src/Generics.m3, generics/src/index.html, generics/src/m3makefile, hello/src/Hello.m3, hello/src/index.html, hello/src/m3makefile, httpd/src/CM3_IDE.gif, httpd/src/HTTPD.m3, httpd/src/index.html, httpd/src/m3makefile, httpd/src/next.html, httpd/src/welcome.html, m3sh/src/M3sh.m3, m3sh/src/index.html, m3sh/src/m3makefile, netobj-client/src/Bank.i3, netobj-client/src/Client.m3, netobj-client/src/NetObjClient.m3, netobj-client/src/index.html, netobj-client/src/m3makefile, netobj-interface/src/Bank.i3, netobj-interface/src/index.html, netobj-interface/src/m3makefile, netobj-server/src/NetObjServer.m3, netobj-server/src/index.html, netobj-server/src/m3makefile, objects/src/Objects.m3, objects/src/index.html, opaque/src/OpaqueExample.m3, opaque/src/Person.i3, opaque/src/Person.m3, opaque/src/index.html, opaque/src/m3makefile, partially-opaque/src/Employee.i3, partially-opaque/src/Employee.m3, partially-opaque/src/PartiallyOpaque.m3, partially-opaque/src/Person.i3, partially-opaque/src/Person.m3, partially-opaque/src/Util.m3, partially-opaque/src/index.html, partially-opaque/src/m3makefile, pickles/src/Action.i3, pickles/src/Action.m3, pickles/src/PickleExample.m3, pickles/src/Pickles.m3, pickles/src/index.html, pickles/src/m3makefile, refman/src/index.html, refman/src/m3makefile, sieve-1/src/Sieve.m3, sieve-1/src/index.html, sieve-1/src/m3makefile, sieve-2/src/Sieve.m3, sieve-2/src/index.html, sieve-2/src/m3makefile, threads/src/ThreadExample.m3, threads/src/index.html, threads/src/m3makefile, unsafe/src/Clib.i3, unsafe/src/Unsafe.m3, unsafe/src/index.html, unsafe/src/m3makefile, web/src/MainWeb.m3, web/src/WebContact.m3, web/src/index.html, web/src/m3makefile, win32-dll/src/RTMain.i3, win32-dll/src/RTMain.m3, win32-dll/src/call_io.cpp, win32-dll/src/m3main.c, win32-dll/src/m3makefile, win32-dll/src/use_m3dll.mak: Added for open-source release of CM3-IDE.--R.Coleburn 2008-06-21 04:20 rcoleburn * doc/src_reports/: doc.gif, html.gif, index.html, license.html, pdf.gif, ps.gif: Updates/additions for open-source release of CM3-IDE.--R.Coleburn 2008-06-21 04:19 rcoleburn * doc/: reference/intfs.html, tutorial/index.html, tutorial/m3/m3_65.html, tutorial/m3/m3_66.html, tutorial/m3/m3_toc.html, tutorial/m3/old-index.html: Updated for open-source release of CM3-IDE.--R.Coleburn 2008-06-21 04:15 rcoleburn * doc/help/CM3_IDE/: basics.pdf, beyond-basics.pdf, customization.pdf, environment.pdf, index.html, interface-index.pdf, intro.pdf, more-info.pdf, packages.pdf, recipes.pdf, user-guide.pdf: Added for open-source release of CM3-IDE.--R.Coleburn 2008-06-21 04:12 rcoleburn * doc/help/: getting-started.html, index.html, cm3/cm3.html, cm3/debugging.html, cm3/example.html, cm3/history.html: Updated for open-source release of CM3-IDE.--R.Coleburn 2008-06-21 04:08 rcoleburn * doc/index.html: Updated for open-source release of CM3-IDE.--R.Coleburn 2008-06-15 10:52 jkrell * m3-sys/cminstall/src/config-no-install/SOLsun: stock Solaris seems to have no ODBC or Postgres, and the packages that use them check this table instead of the other 2008-06-15 10:45 jkrell * www/installation-windows.html: NT386 needs to be NT386* and then take the time to review most of the document, editing here and there, encouraging people to use Python instead of cmd, providing a newer example with the yet newer distribution format (toplevel directory is versioned) shorten wording about nearly every Visual C++ and Windows works strike incorrect comment about Cygwin/MinGWin not supported (but leaving further information out for now) 2008-06-15 10:34 jkrell * scripts/win/: backup-pkgs.cmd, upgrade.cmd: sh and py no longer backup packages (found reviewing the installation notes) 2008-06-15 09:57 jkrell * scripts/python/pylib.py: far more than just Linux needs Unix.common, so always include it in distributions (found on SOLsun, but also true for Darwin and *BSD) 2008-06-15 09:40 jkrell * m3-sys/m3cc/src/m3makefile: cleanup: don't change the files more than once, use more portable command lines where possible (&& instead of ;) only about two commands need a sh wrapper 2008-06-15 00:24 jkrell * m3-sys/m3cc/src/m3makefile: remove libiconv dependency, such as for Solaris (before it was using it if it was present, as a result, cm3cg from one person's machine wouldn't work on another), and use single slashes, they work 2008-06-14 18:35 jkrell * m3-libs/m3core/src/runtime/common/RTCollector.m3: fix newlines 2008-06-12 21:04 rcoleburn * scripts/cm3SetupCmdEnv.cmd: minor change to display cm3 version number in window title.--R.Coleburn 2008-06-12 12:55 jkrell * m3-libs/m3core/src/float/m3makefile, m3-sys/cminstall/src/config-no-install/SOLsun, scripts/python/pylib.py: main in C prevents: Text relocation remains referenced against symbol offset in file __fsr_init_value 0x48 /opt/SUNWspro/prod/lib/crt1.o __fsr_init_value 0x4c /opt/SUNWspro/prod/lib/crt1.o ld: fatal: relocations remain against allocatable but non-writable sections I don't know why, it was just a guess, reasonable.. Trade the a rare working FloatMode for more efficient linking options: -z text -z defs -B direct -xldscope=symbolic Still TBD: generate map files to avoid exporting functions not in interfaces (unless needed for derived class vtables?) Let pylib.py know that SOLsun and SOLgnu are platforms. 2008-06-09 16:24 jkrell * m3-libs/m3core/src/Csupport/libgcc/libgcc.c, m3-sys/cminstall/src/config-no-install/SOLsun, m3-sys/cminstall/src/config-no-install/Unix.common, m3-libs/m3core/src/Csupport/m3makefile, m3-libs/m3core/src/Csupport/libgcc/m3makefile: not entirely satisfactory fixes for SOLsun in particular, gcc backend outputs function calls to 64 bit math (at least with optimization off, revisit this variable) those functions are in libgcc, but SOLsun doesn't have libgcc the functions are trivial one line of C each, if the C compiler implements them inline, or calls yet other-named functions furthermore, -z defs and -z text both sound good, but using them both is seemingly not possible, not as long as -lsunmath is needed -z defs resolves all symbols in a shared object at link time instead of leaving undefined functions for resolution at load/run time well, they are resolved at load/run time, but a lib must be provided at link time corresponding to load/run time -z text checks that there are no relocations in the text section -- that all code is built position independent essentially it seems that libsunmath is the offender here -z defs is the opposite of historical and current defaults for shared objects, but is encouraged by the man page One of these must presently be omitted and which one is the lesser evil is uncertain. furthermore: the trivial one line functions in libgcc.a are GPLed and attempts to static link them for dependency-management failed, though that wouldn't help, possibly hurt, the GPL tbd: see if optimization or particular flags force inlining of 64 bit math, even though in general I'd rather not optimize since it hurts debugability 2008-06-09 14:13 jkrell * m3-sys/cminstall/src/config-no-install/: SOLsun, Unix.common: Solaris support 2008-06-09 12:36 jkrell * m3-sys/m3tests/src/m3makefile: make -clean work; note that realclean still does not work; further note that regular does do clean as they go, but control-c in flight, or cm3 -keep leaves stuff around 2008-06-09 12:06 jkrell * m3-sys/cminstall/src/config-no-install/Unix.common: just cleanup -- remove tabs, use four space indent 2008-06-09 11:59 jkrell * m3-sys/cminstall/src/config-no-install/Unix.common: again, prepare for more code sharing, using Unix.common for more platforms 2008-06-09 11:54 jkrell * m3-sys/cminstall/src/config-no-install/Unix.common: go back a version, checked in wrong file (didn't mean to combine cleanup with the rest) 2008-06-09 10:58 jkrell * m3-sys/cminstall/src/config-no-install/Unix.common: prepare for more code sharing, using Unix.common for more platforms 2008-06-09 05:05 jkrell * m3-sys/cm3/src/version.quake: fix copy/paste errors 2008-06-09 04:48 jkrell * m3-sys/cm3/src/: Main.m3, version.quake: ferry along what GNU_MAKE is so it can be configured for the HOST instead for the TARGET; of course, it all seems circular since it is the cm3 TARGET when cm3 is built that determines the later cm3 HOST when cm3 is later run, but this is in fact correct, or at least has higher tendency to be by default than the old way; if GNU_MAKE is a full path, then it will not necessarily be correct from machine to machine, but "make in $PATH" vs. "gmake in $PATH" will tend to be correct; e.g. try cross-building platforms that have BSD make as make and GNU make as gmake 2008-06-09 02:01 jkrell * m3-sys/m3cc/src/m3makefile: which make to use is a host configuration option, not a target one, use it here; needs to be exposed by cm3 2008-06-09 01:29 jkrell * m3-sys/m3cc/gcc/gcc/m3cg/parse.c: 'rewrite' scan_float to store floats in longs, correctly aligned and with only up to 32 bits per long (even if they are larger). This should fix alignment/bus error on OpenBSD/sparc64 and help on other 64 bit host platforms. 2008-06-09 01:17 rcoleburn * scripts/: ArcDir.CMD, UnArc.CMD: Provide a couple of useful scripts for Windows environments that archive/unarchive folders using TAR and GZIP utilities. (The TAR and GZIP utilities can be obtained from the CYGWIN distribution.)--R.Coleburn 2008-06-09 01:10 rcoleburn * scripts/cm3Proj.cmd: Allowed flexibility in location of GZIP and TAR utilities; Added date/time stamp to archive filenames. 2008-06-08 09:58 jkrell * m3-sys/m3tests/src/: m3makefile, p2/p211/A.i3, p2/p211/A.m3, p2/p211/C.c, p2/p211/m3makefile, p2/p211/stderr.pgm, p2/p211/stdout.pgm: add test case that checks that float and double constants are written and read correctly by the compiler, so that the code can be fixed to work correctly a) on systems where long holds more than 32 bits b) on systems where long requires a larger alignment than double e.g. OpenBSD/sparc64 The code passes floats and doubles along with "matching" strings from Modula-3 to C and the C code checks that things match up. Notice that Modula-3 turns "-0" into the same as "0", but C does not. That test case is therefore skipped. 2008-06-07 22:26 jkrell * scripts/python/pylib.py: fix Target 2008-06-07 10:48 jkrell * m3-sys/m3cc/src/m3makefile: check for .exe extension for some reason doesn't work, maybe it is before the file exists, loosen the check 2008-06-07 09:18 jkrell * scripts/python/pylib.py: more cosmetics -- comments and whitespace 2008-06-07 09:07 jkrell * scripts/python/pylib.py: Do a lot of unteasing of 'host' and 'target. Previously everything was about 'target' and native builds were mostly assumed. A bunch of cleanup/movement and adding of comments and vertical space. Soon we should be able to choose 'cm3cg' more often automatically correctly to ease cross builds and building a bunch of 'boot' archives. 2008-06-07 08:10 jkrell * scripts/python/pylib.py: some case insensitivity and add windowsResources in more places 2008-06-07 08:05 jkrell * m3-sys/windowsResources/src/cm3-41-hack.c: Use correct function signatures (take nothing, not anything), and be sure not to return garbage. 2008-06-06 11:43 jkrell * scripts/python/make-dist.py: only ship front when not building back 2008-06-06 11:42 jkrell * scripts/python/make-dist.py: check if m3cc is present before removing, it often isn't, and errors otherwise 2008-06-06 10:43 jkrell * scripts/python/make-dist.py: make NT386GNU std archive too 2008-06-05 11:25 jkrell * m3-libs/m3core/src/unix/openbsd-common/: Utypes.i3, Uuio.i3: ssize_t in Utypes more typically, maybe shouldn't be in Cstddef.i3 2008-06-03 11:03 jkrell * m3-sys/m3cc/gcc/mpfr/configure: take minor change to older boilerplate from other nearby configure files, in order to let it work under "nohup", on OpenBSD otherwise there is an error about an invalid file handle 2008-06-03 11:01 jkrell * m3-libs/m3core/src/C/Common/Cstdint.i3, m3-libs/m3core/src/C/Common/Ctypes.i3, m3-libs/m3core/src/win32/WinBase.i3, m3-libs/m3core/src/win32/WinBaseTypes.i3, m3-libs/m3core/src/win32/WinGDI.i3, m3-libs/m3core/src/win32/WinIoctl.i3, m3-libs/m3core/src/win32/WinListView.i3, m3-libs/m3core/src/win32/WinNT.i3, m3-libs/m3core/src/win32/WinSock.i3, m3-libs/m3core/src/win32/WinTabCon.i3, m3-libs/m3core/src/win32/WinUser.i3, m3-ui/ui/src/winvbt/WinScrnColorMap.m3: make Cstdint more correct by specifying bits and remove values that can't be portably defined cascade from this: repair Windows headers: 1) add explicit padding; these were ok before 2) add a field that was randomly missing, was probably ok as padding would have been inserted 3) "unalign" some fields; this is not satisfactory but probably more important to get the containing records sized and aligned correctly than to expose every field; some way to expose them as one field instead of broken up, while preserving the unalignment, would be good; as well, platforms that have alignment faults (aka bus errors) must or should be able to use these types these were very seemingly incorrect before more work is needed on reducing header cloning and/or ensuring it is done correctly and/or automating it, and regularly checking it 4) a bit field-for-padding had the wrong size, probably a copy/paste error, and probably ok as it was, as further padding would have been inserted by compiler alignment seems like a big mess in the broader context, too easy to get wrong This might break code using the Windows types BYTE, WORD, DWORD, UINT, ULONG, etc. but NOT silently -- at compile-time. 2008-06-03 10:49 jkrell * m3-sys/m3cc/src/m3makefile: fix line endings (use Unix format) 2008-06-03 10:39 jkrell * m3-libs/m3core/src/float/m3makefile: add missing OS/2 2008-06-02 11:21 jkrell * m3-libs/m3core/src/C/Common/Cstddef.i3: wide char is 16 bits on some platforms, and 32 bits on some platforms, so just remove it 2008-06-02 11:17 jkrell * m3-libs/m3core/src/C/Common/: Cstdint.i3, m3makefile: expose clearly named fixed size integer types like the updated C standard 2008-06-02 06:25 jkrell * m3-sys/m3cc/gcc/gcc/m3cg/parse.c: The comment says typeids are unsigned 32bit numbers but on 64 bit hosts they get signed extended. This causes the output of cm3cg -y to vary between 32 bit and 64 bit hosts. Eliminate this red herring by treating the data as actually unsigned 32 bit numbers. (red herring because NT386GNU-hosted SPARC64_OPENBSD-targeted cross cm3cg works, but native SPARC64_OPENBSD does not, exhibiting the same behavior as native AMD64_LINUX) 2008-06-02 05:39 jkrell * m3-libs/m3core/src/unix/: darwin-amd64/Upthread.i3, darwin-amd64/m3makefile, darwin-generic/Upthread.i3, darwin-generic/m3makefile, darwin-i386/Upthread.i3, darwin-i386/m3makefile, darwin-ppc/Upthread.i3, darwin-ppc/m3makefile: eliminate three unnecessary header clones 2008-06-02 03:58 jkrell * scripts/python/make-dist.py: endeavor to only build m3cc once, and make building one of the other package lists clearer 2008-06-02 02:22 jkrell * m3-sys/cminstall/src/config-no-install/Unix.common: add path to X libs for OpenBSD 2008-06-01 11:37 jkrell * scripts/python/pylib.py: OpenBSD doesn't let you set 'all' bits to make a file writable/deletable; sysinfo.sh no long ever run by cm3.cfg, cm3's builtin HOST variable is used instead 2008-06-01 06:25 jkrell * scripts/python/: make-dist.py, pylib.py: fixes for make-dist -- /tmp/cm3 is taken, just use mkdtemp, Linux.common renamed to Unix.common since it used on OpenBSD and presumably in future others 2008-06-01 06:06 jkrell * m3-sys/cminstall/src/config-no-install/: PPC32_OPENBSD, SPARC64_OPENBSD: BSD needs to use gmake to build cm3cg, not surprising 2008-06-01 05:38 jkrell * m3-sys/cm3/src/version.quake: just whitespace and comments 2008-06-01 05:30 jkrell * m3-sys/cm3/src/version.quake: fix and cleanup, I thought I had already fixed this 2008-06-01 05:15 jkrell * m3-sys/cminstall/src/config-no-install/Unix.common: fix 2008-06-01 05:10 jkrell * m3-libs/m3core/src/unix/irix-5.2/Usignal.i3: undo accidental commit 2008-06-01 04:51 jkrell * m3-libs/m3core/src/unix/: m3makefile, aix-3-2/Uin.i3, aix-3-2/Uin.m3, aix-3-2/m3makefile, aix-ps2-1-2/Uin.i3, aix-ps2-1-2/Uin.m3, aix-ps2-1-2/m3makefile, big-endian/Uin.m3, cygwin/Uin.i3, cygwin/Uin.m3, cygwin/m3makefile, darwin-amd64/Uin.m3, darwin-amd64/m3makefile, darwin-generic/Uin.i3, darwin-generic/Utypes.i3, darwin-generic/m3makefile, darwin-i386/Uin.m3, darwin-i386/m3makefile, darwin-ppc/Uin.m3, darwin-ppc/m3makefile, freebsd-1/Uin.i3, freebsd-1/Uin.m3, freebsd-1/m3makefile, freebsd-2/Uin.i3, freebsd-2/Uin.m3, freebsd-2/m3makefile, freebsd-3/Uin.i3, freebsd-3/Uin.m3, freebsd-3/m3makefile, freebsd-4/Uin.i3, freebsd-4/Uin.m3, freebsd-4/Utypes.i3, freebsd-4/m3makefile, hpux-7-0/Uin.i3, hpux-7-0/Uin.m3, hpux-7-0/m3makefile, ibm-4-3/Uin.i3, ibm-4-3/Uin.m3, ibm-4-3/m3makefile, irix-5.2/Uin.i3, irix-5.2/Uin.m3, irix-5.2/Usignal.i3, irix-5.2/m3makefile, linux/Uin.i3, linux/Uin.m3, linux/m3makefile, linux-libc6/Uin.i3, linux-libc6/Utypes.i3, linux-libc6/m3makefile, linux-ppc/Uin.m3, linux-ppc/m3makefile, little-endian/Uin.m3, netbsd2-i386/Uin.i3, netbsd2-i386/Uin.m3, netbsd2-i386/m3makefile, openbsd-common/Uin.i3, openbsd-common/Utypes.i3, openbsd-common/m3makefile, os2/Uin.i3, os2/Uin.m3, os2/m3makefile, osf-1.ALPHA_OSF/Uin.i3, osf-1.ALPHA_OSF/Uin.m3, osf-1.ALPHA_OSF/m3makefile, osf-1.DS3100/Uin.i3, osf-1.DS3100/Uin.m3, osf-1.DS3100/m3makefile, solaris-2-x/Uin.i3, solaris-2-x/Uin.m3, solaris-2-x/m3makefile, sunos-4-x/Uin.i3, sunos-4-x/Uin.m3, sunos-4-x/m3makefile, sysv-4.0/Uin.i3, sysv-4.0/Uin.m3, sysv-4.0/m3makefile, uin-common/Uin.i3, uin-common/m3makefile, uin-len/Uin.i3, uin-len/m3makefile, ultrix-3-1.AP3000/Uin.i3, ultrix-3-1.AP3000/Uin.m3, ultrix-3-1.AP3000/m3makefile, ultrix-3-1.ARM/Uin.i3, ultrix-3-1.ARM/Uin.m3, ultrix-3-1.ARM/m3makefile, ultrix-3-1.DS3100/Uin.i3, ultrix-3-1.DS3100/Uin.m3, ultrix-3-1.DS3100/m3makefile, ultrix-3-1.NEXT/Uin.i3, ultrix-3-1.NEXT/Uin.m3, ultrix-3-1.NEXT/m3makefile, ultrix-3-1.SEQUENT/Uin.i3, ultrix-3-1.SEQUENT/Uin.m3, ultrix-3-1.SEQUENT/m3makefile, ultrix-3-1.SUN3/Uin.i3, ultrix-3-1.SUN3/Uin.m3, ultrix-3-1.SUN3/m3makefile, ultrix-3-1.SUN386/Uin.i3, ultrix-3-1.SUN386/Uin.m3, ultrix-3-1.SUN386/m3makefile, ultrix-3-1.UMAX/Uin.i3, ultrix-3-1.UMAX/Uin.m3, ultrix-3-1.UMAX/m3makefile, ultrix-3-1.VAX/Uin.i3, ultrix-3-1.VAX/Uin.m3, ultrix-3-1.VAX/m3makefile: reduce around 62 files to 4 files Uin.i3 has two variants -- with sin_len or without. FreeBSD (except -1), OpenBSD, and Darwin have sin_len. Where sin_len is absent, sin_family expands to 16 bits to take its space. There were also variant swhere sin_family was short, but just make it always unsigned_short, it doesn't matter. Uin.m3 has two variants -- little-endian and big-endian network order == host order or not (network order is big endian, like SPARC and unlike x86) Previously each of these files was duplicated for every platform. Recently there was some reduction, now much more. 2008-05-31 16:39 mika * m3-comm/netobj/tests/luca/src/m3makefile: Added quotation marks to update m3makefile so that "Luca's test" of Network Objects compiles using newer compilers. 2008-05-31 08:04 jkrell * m3-comm/tcp/src/WIN32/WinSock.m3: like Winsock.i3, this Winsock.m3 is also dead, and also in m3-libs/m3core/src/win32 2008-05-30 11:57 jkrell * m3-sys/cminstall/src/: config/NT386MINGNU, config-no-install/AMD64_LINUX, config-no-install/PPC32_OPENBSD, config-no-install/PPC_LINUX, config-no-install/SPARC32_LINUX, config-no-install/SPARC64_LINUX, config-no-install/SPARC64_OPENBSD, config-no-install/Unix.common: adjust line endings 2008-05-30 11:55 jkrell * m3-sys/: cm3/src/Builder.m3, cm3/src/M3Path.i3, cm3/src/Main.m3, cm3/src/Version.i3, cm3/src/m3makefile, cm3/src/version.quake, cminstall/src/config/NT386, cminstall/src/config/NT386.common, cminstall/src/config/NT386.main, cminstall/src/config/NT386MINGNU, cminstall/src/config/cm3.cfg, m3middle/src/Target.i3: allow strings for M3_BACKEND_MODE and NAMING_CONVENTIONS allow BACKEND_MODE instead of M3_BACKEND_MODE expose aspects of the host to Quake -- HOST, HOST_OS_TYPE and use host aspects to default to a native build so the config files can just a *little* less this does change the "if defined" paths as part of this, make Version.i3 the generated file instead of Version.m3, with string constants 2008-05-30 08:20 jkrell * m3-comm/tcp/src/WIN32/WinSock.i3: attack the clones: delete dead clone of cloned header the one in m3core suffices 2008-05-30 08:14 jkrell * m3-libs/m3core/src/unix/openbsd-common/Ustat.i3: remove S_IFPIPE which is always zero and doesn't really exist use S_IFIFO directly instead of S_IFPORT and remove S_IFPORT S_IFPORT perhaps should exist and have a different value on Solaris, but nobody ever noticed, so err toward both reduction and preserving semantics 2008-05-30 08:09 jkrell * m3-libs/: libm3/src/os/POSIX/FilePosix.i3, libm3/src/os/POSIX/FilePosix.m3, m3core/src/unix/aix-3-2/Ustat.i3, m3core/src/unix/aix-ps2-1-2/Ustat.i3, m3core/src/unix/cygwin/Ustat.i3, m3core/src/unix/darwin-generic/Ustat.i3, m3core/src/unix/freebsd-1/Ustat.i3, m3core/src/unix/freebsd-2/Ustat.i3, m3core/src/unix/freebsd-3/Ustat.i3, m3core/src/unix/freebsd-4/Ustat.i3, m3core/src/unix/hpux-7-0/Ustat.i3, m3core/src/unix/ibm-4-3/Ustat.i3, m3core/src/unix/irix-5.2/Ustat.i3, m3core/src/unix/linux/Ustat.i3, m3core/src/unix/linux-32/Ustat.i3, m3core/src/unix/linux-64/Ustat.i3, m3core/src/unix/netbsd2-i386/Ustat.i3, m3core/src/unix/os2/Ustat.i3, m3core/src/unix/osf-1.generic/Ustat.i3, m3core/src/unix/solaris-2-x/Ustat.i3, m3core/src/unix/sunos-4-x/Ustat.i3, m3core/src/unix/sysv-4.0/Ustat.i3, m3core/src/unix/ultrix-3-1.AP3000/Ustat.i3, m3core/src/unix/ultrix-3-1.ARM/Ustat.i3, m3core/src/unix/ultrix-3-1.DS3100/Ustat.i3, m3core/src/unix/ultrix-3-1.NEXT/Ustat.i3, m3core/src/unix/ultrix-3-1.SEQUENT/Ustat.i3, m3core/src/unix/ultrix-3-1.SUN3/Ustat.i3, m3core/src/unix/ultrix-3-1.SUN386/Ustat.i3, m3core/src/unix/ultrix-3-1.UMAX/Ustat.i3, m3core/src/unix/ultrix-3-1.VAX/Ustat.i3: remove S_IFPIPE which is always zero and doesn't really exist use S_IFIFO directly instead of S_IFPORT and remove S_IFPORT S_IFPORT perhaps should exist and have a different value on Solaris, but nobody every noticed, so err toward both reduction and preserving semantics 2008-05-30 08:03 jkrell * m3-libs/libm3/src/os/POSIX/ProcessCygwin.m3: remove unused import to fix warning 2008-05-30 07:56 jkrell * m3-libs/m3core/src/unix/linux-amd64/: UpthreadMachine.i3, Uucontext.i3: slight textual reduction of cloned headers, more to do here 2008-05-30 07:48 jkrell * m3-libs/m3core/src/unix/: cygwin/1.cmd, cygwin/Udir.i3, cygwin/Uerror.i3, cygwin/Uerror.i3.cpp, cygwin/Uexec.i3, cygwin/Uin.i3, cygwin/Umman.i3, cygwin/Unetdb.i3, cygwin/Unix.i3, cygwin/Unix.i3.cpp, cygwin/Upwd.i3, cygwin/Uresource.i3, cygwin/Usignal.i3, cygwin/Usignal.i3.c, cygwin/Usocket.i3, cygwin/Utime.i3, cygwin/UtimeC.c, cygwin/Utypes.i3, cygwin/Utypes.i3.cpp, cygwin/Uuio.i3, cygwin/Uutsname.i3, openbsd-common/Udir.i3, openbsd-common/Uexec.i3, openbsd-common/Upwd.i3, openbsd-common/Uresource.i3, openbsd-common/Usched.i3, openbsd-common/Usem.i3: continue to reduce cloned headers down to only what is needed, and generate parts of them, since cloned headers are bad and bloated 2008-05-30 07:07 jkrell * scripts/python/pylib.py: add windowsResource package; don't require X, though this isn't right, if it is there, we should maybe put it on path, but if not there, don't error out 2008-05-29 19:04 jkrell * m3-sys/cminstall/src/config-no-install/: AMD64_LINUX, LINUXLIBC6, Linux.common, PPC32_OPENBSD, PPC_DARWIN, PPC_LINUX, SPARC32_LINUX, SPARC64_LINUX, SPARC64_OPENBSD, Unix.common: PPC32_OPENBSD/SPARC64_OPENBSD support, cleanup delayed from previous due to some cvs/lock issue 2008-05-29 12:43 jkrell * m3-libs/libm3/src/os/POSIX/OSConfigPosix.m3, m3-libs/libm3/src/random/m3makefile, m3-libs/m3core/src/C/m3makefile, m3-libs/m3core/src/C/AMD64_LINUX/Cstdio.i3, m3-libs/m3core/src/C/AMD64_LINUX/m3makefile, m3-libs/m3core/src/C/Common/Cstddef.i3, m3-libs/m3core/src/C/Common/Ctypes.i3, m3-libs/m3core/src/C/NT386/Cstdio.i3, m3-libs/m3core/src/C/PPC32_OPENBSD/Csetjmp.i3, m3-libs/m3core/src/C/PPC32_OPENBSD/Csignal.i3, m3-libs/m3core/src/C/PPC32_OPENBSD/Cstdio.i3, m3-libs/m3core/src/C/PPC32_OPENBSD/Cstring.i3, m3-libs/m3core/src/C/PPC32_OPENBSD/m3makefile, m3-libs/m3core/src/C/PPC_LINUX/m3makefile, m3-libs/m3core/src/float/m3makefile, m3-libs/m3core/src/runtime/m3makefile, m3-libs/m3core/src/runtime/common/Compiler.tmpl, m3-libs/m3core/src/runtime/common/RTCollector.m3, m3-libs/m3core/src/runtime/ex_frame/m3makefile, m3-libs/m3core/src/thread/m3makefile, m3-libs/m3core/src/time/POSIX/m3makefile, m3-libs/m3core/src/unix/m3makefile, m3-libs/m3core/src/unix/big-endian/Uin.m3, m3-libs/m3core/src/unix/cygwin/Uerror.i3.cpp, m3-libs/m3core/src/unix/cygwin/Uexec.i3, m3-libs/m3core/src/unix/cygwin/Unetdb.i3, m3-libs/m3core/src/unix/cygwin/Unix.i3, m3-libs/m3core/src/unix/cygwin/Uprocess.i3, m3-libs/m3core/src/unix/cygwin/Utypes.i3, m3-libs/m3core/src/unix/cygwin/Utypes.i3.cpp, m3-libs/m3core/src/unix/openbsd-common/1.cpp, m3-libs/m3core/src/unix/openbsd-common/1.sh, m3-libs/m3core/src/unix/openbsd-common/Udir.i3, m3-libs/m3core/src/unix/openbsd-common/Uerror.i3, m3-libs/m3core/src/unix/openbsd-common/Uexec.i3, m3-libs/m3core/src/unix/openbsd-common/Uin.i3, m3-libs/m3core/src/unix/openbsd-common/Umman.i3, m3-libs/m3core/src/unix/openbsd-common/Unetdb.i3, m3-libs/m3core/src/unix/openbsd-common/Unix.i3, m3-libs/m3core/src/unix/openbsd-common/Uprocess.i3, m3-libs/m3core/src/unix/openbsd-common/Upthread.i3, m3-libs/m3core/src/unix/openbsd-common/Upwd.i3, m3-libs/m3core/src/unix/openbsd-common/Uresource.i3, m3-libs/m3core/src/unix/openbsd-common/Usched.i3, m3-libs/m3core/src/unix/openbsd-common/Usem.i3, m3-libs/m3core/src/unix/openbsd-common/Usignal.i3, m3-libs/m3core/src/unix/openbsd-common/Usocket.i3, m3-libs/m3core/src/unix/openbsd-common/Ustat.i3, m3-libs/m3core/src/unix/openbsd-common/Utime.i3, m3-libs/m3core/src/unix/openbsd-common/Utypes.i3, m3-libs/m3core/src/unix/openbsd-common/Uucontext.i3, m3-libs/m3core/src/unix/openbsd-common/Uugid.i3, m3-libs/m3core/src/unix/openbsd-common/Uuio.i3, m3-libs/m3core/src/unix/openbsd-common/Uutmp.i3, m3-libs/m3core/src/unix/openbsd-common/Uutsname.i3, m3-libs/m3core/src/unix/openbsd-common/m3makefile, m3-sys/cminstall/src/config/cm3.cfg, m3-sys/m3cc/gcc/gcc/m3cg/parse.c, m3-sys/m3middle/src/Target.i3, m3-sys/m3middle/src/Target.m3, scripts/sysinfo.sh, scripts/python/pylib.py: enough support for PPC32_OPENBSD and SPARC64_OPENBSD to build bootstrap archives and then cm3 for each and to run cm3 on at least one of them and various small cleanup/reduction having trouble commiting in cminstall/src/config-no-install 2008-05-28 01:19 rcoleburn * m3-ui/vbtkit/src/: lego/ScrollerVBTClass.m3, lego/ZChassisVBT.m3, vbtkitutils/VBTKitEnv.i3: These files have been moved to the platform-specific subfolders, e.g. WIN32 or POSIX. 2008-05-28 01:10 rcoleburn * m3-comm/netobjd/src/NetObjD.ICO: add 48-bit pattern in addition to the 32-bit pattern 2008-05-27 04:35 jkrell * m3-libs/m3core/src/unix/linux-32/Ustat.i3: dos2unix (convert line endings to unix format) 2008-05-27 02:29 rcoleburn * m3-ui/vbtkit/src/: lego/m3makefile, vbtkitresources/ScrollBg, vbtkitresources/close.ppm, vbtkitresources/down.ppm, vbtkitresources/grow.ppm, vbtkitresources/left.ppm, vbtkitresources/m3makefile, vbtkitresources/right.ppm, vbtkitresources/up.ppm, lego/POSIX/ScrollerVBTClass.m3, lego/POSIX/ZChassisVBT.m3, lego/POSIX/m3makefile, vbtkitutils/m3makefile, lego/WIN32/ScrollerVBTClass.m3, lego/WIN32/ZChassisVBT.m3, lego/WIN32/m3makefile, vbtkitutils/POSIX/VBTKitEnv.i3, vbtkitutils/POSIX/m3makefile, vbtkitutils/WIN32/VBTKitEnv.i3, vbtkitutils/WIN32/m3makefile: Incorporate changes made to vbtkit by Critical Mass circa September 1997 that make the GUI appearance more like that of Windows. The m3makefiles have been adjusted so that these changes affect only Windows (WIN32) platforms. These changes were part of work sponsored by Randy Coleburn at Scientific Research Corporation (SRC). The main files affected are VBTKitEnv.i3, ScrollerVBTClass.m3, and ZChassisVBT.m3, plus some new resource files have been added for WIN32 only. 2008-05-26 23:36 rcoleburn * m3-ui/cmvbt/src/README.html: Add note that the installation instructions apply only to v4.1 of Reactor. The changes listed in this readme file have already been made to subsequent cm3 releases. 2008-05-26 22:51 rcoleburn * m3-ui/cmvbt/src/IPTypeinVBT.m3: Integrate my changes from 03/13/2005 that replaced buggy IsComplete, Get, and Key procedures. --R.Coleburn 2008-05-26 22:38 rcoleburn * m3-ui/cmvbt/src/: ClockVBT.i3, ClockVBT.m3: Integrated my change from 05/16/2001 that fixes bug that caused T.proc not to be used and that prevented specification of the time zone. -- R.C.Coleburn 2008-05-25 20:06 jkrell * m3-libs/m3core/src/unix/cygwin/: Uin.i3, Usocket.i3, Ustat.i3, Utime.i3, Utypes.i3, Utypes.i3.cpp, Uugid.i3, Uuio.i3, Uutmp.i3: cleanup and reduce, esp. in preparation for future merging of common code including a lot of simple line count or text size reduction comments copied from /usr/include/*.h can be found be reading /usr/include/*.h or manual pages favor clearly named fixed size integer types instead of the funny names function declarations on one compact line Notice how all function declarations essentially have to match across Posix in order for the callers to have source compatibility, though they can vary *slightly* like int vs. long. include only what is used within the cm3 tree 2008-05-25 19:56 jkrell * m3-libs/m3core/src/win32/: WinBase.i3, WinGDI.i3, WinTabCon.m3, WinUser.i3: dos2unix (change small number of carriagereturn-newline sequences to just newline) 2008-05-25 19:48 jkrell * scripts/win/boot1.cmd: favor Python 2008-05-25 19:23 jkrell * m3-sys/m3cc/src/: m3makefile, patches/openbsd/patch-gcc_config_exec-stack_h, patches/openbsd/patch-gcc_config_gcc, patches/openbsd/patch-gcc_config_host, patches/openbsd/patch-gcc_config_host-openbsd_c, patches/openbsd/patch-gcc_config_i386_openbsd64_h, patches/openbsd/patch-gcc_config_i386_openbsd_h, patches/openbsd/patch-gcc_config_i386_openbsdelf_h, patches/openbsd/patch-gcc_config_m68k_openbsd_h, patches/openbsd/patch-gcc_config_mips_openbsd_h, patches/openbsd/patch-gcc_config_openbsd-libpthread_h, patches/openbsd/patch-gcc_config_openbsd_h, patches/openbsd/patch-gcc_config_rs6000_openbsd_h, patches/openbsd/patch-gcc_config_t-openbsd, patches/openbsd/patch-gcc_config_vax_openbsd_h, patches/openbsd/patch-gcc_config_x-openbsd, patches/openbsd/patch-gcc_tree-ssa-pre_c, patches/openbsd/patch-libtool_m4, patches/openbsd/readme.txt: support for targeting OpenBSD these files are from the OpenBSD "port" of gcc 4.3 as described in the readme There are many more patches in the OpenBSD port but many of them are clearly not needed here, some of them are perhaps not needed, and many of them appear to be stated against 4.2 or at least a very old 4.3, so pruning was favored over repairing; what is left is fairly uninvasive 2008-05-25 03:24 jkrell * m3-sys/cminstall/src/config-no-install/: Linux.common, PPC_DARWIN: fix -- current PPC_DARWIN backends do not understand -m32 These files will be merging and changing names, but ok for now. (Unix.common -- Linux, Darwin, *BSD) 2008-05-25 03:08 jkrell * m3-libs/m3core/src/Csupport/: NEXT/dtoa.c, big-endian/dtoa.c, little-endian/dtoa.c: forgot to delete one, and put in the paranoia that this one alone had 2008-05-25 03:01 jkrell * m3-libs/m3core/src/Csupport/: m3makefile, ALPHA_OSF/dtoa.c, ALPHA_OSF/m3makefile, Common/dtoa.h, Common/m3makefile, HP300/dtoa.c, HP300/m3makefile, HPPA/dtoa.c, HPPA/m3makefile, LINUX/dtoa.c, LINUX/m3makefile, LINUXELF/dtoa.c, LINUXELF/m3makefile, NEXT/m3makefile, OS2/dtoa.c, OS2/m3makefile, SPARC/dtoa.c, SPARC/float.h, SPARC/m3makefile, SUN3/dtoa.c, SUN3/float.h, SUN3/m3makefile, SUN386/dtoa.c, SUN386/float.h, SUN386/m3makefile, UMAX/dtoa.c, UMAX/float.h, UMAX/m3makefile, VAX/dtoa.c, VAX/float.h, VAX/m3makefile, big-endian/dtoa.c, big-endian/m3makefile, little-endian/dtoa.c, little-endian/m3makefile: cleanup a lot of duplication and dead code (float.h were dead) 32bit vs. 64bit can be probed via limits.h big-endian and little-endian are the resulting two versions of dtoa.c plus VAX 2008-05-25 01:16 jkrell * scripts/python/: boot1.py, do-cm3-all.py, do-cm3-base.py, do-cm3-caltech-parser.py, do-cm3-comm.py, do-cm3-core.py, do-cm3-front.py, do-cm3-gui.py, do-cm3-min.py, do-cm3-std.py, do-pkg.py, install-back.py, install-cm3-compiler.py, pylib.py: allow command line parameters such as boot boot makes a new-fangled boot archive switch to gzip since OpenBSD doesn't include bzip2 by default, for perf reasons boot1.py is basically same as do-cm3-front boot switch from strings to function pointers a little bit of reformating some BSD support reorder some imports (that's what most of the files are) 2008-05-24 09:24 rcoleburn * scripts/pkginfo.txt: Add "windowsResources" to package list. I marked it as "std". Not sure if it should be marked with any other tags. BTW, are the meanings of the various tags defined anywhere, e.g., "core", "min", "front", "base", "comm" ? 2008-05-24 09:15 rcoleburn * m3-comm/netobjd/src/: NetObjD.ICO, iconRes.rc, m3makefile: Add a program icon for use on Windows platforms. This change has no effect on non-windows platforms. If anyone doesn't like the icon, you are welcome to change it to something better. If you are using Windows, don't forget to build and ship the "windowsResources" package located in m3-sys because this change depends on that package. 2008-05-19 13:09 jkrell * m3-sys/m3cc/src/: boot.py, m3makefile: m3makefile: clear out 32/64 stuff, just use native do things more consistently across platforms now that has been more tested measurement should be done to see if lack of -O2 makes m3cg noticably slower; maybe it is worth the doubled build time and should be left only configured on the command line (for me) boot.py: emulate m3makefile on new targets that don't yet have cm3 2008-05-14 08:30 jkrell * m3-sys/cminstall/src/config-no-install/: SPARC32_LINUX, SPARC64_LINUX: I forgot to add these. 2008-05-13 18:11 jkrell * scripts/win/boot1.cmd: produce three options -- make.sh, makeverbose.sh, Makefile, instead of the previous 1.sh, equiv. to the now makeverbose.sh 2008-05-13 07:44 jkrell * m3-sys/m3middle/src/: Target.i3, Target.m3: let's consistently use bit sizes/offets/addresses and not byte sizes/offsets/addresses CAVEATS: I don't really know what was intended here, but it seems that a mix of byte and bit addresses were used bit addresses seem better, they are the more general case, they can represent byte addresses, but byte addresses cannot represent bit addresses however that is not entirely true -- a bit address in an INTEGER has an eighth of the range of a byte address in an INTEGER; the multiplication/overflow was happening anyway, in NilChkExpr Look at this way -- the multiplications are optimized out. I remember thinking that multiplication by byte.size should omitted, since it is obviously 1, but it is not, it is bits not bytes verifying the correctness of all these numbers would be good setting the values lower than "correct" is ok, but less efficient 2008-05-13 06:50 jkrell * m3-libs/m3core/src/C/ALPHA_OSF/Cstring.i3: missed one 2008-05-13 06:43 jkrell * m3-libs/m3core/src/C/: AIX386/Cstring.i3, ALPHA_OSF/Cstring.i3, AMD64_DARWIN/Cstring.i3, AMD64_LINUX/Cstring.i3, AP3000/Cstring.i3, ARM/Cstring.i3, DS3100/Cstring.i3, FreeBSD/Cstring.i3, FreeBSD2/Cstring.i3, FreeBSD3/Cstring.i3, FreeBSD4/Cstring.i3, HP300/Cstring.i3, HPPA/Cstring.i3, I386_DARWIN/Cstring.i3, IBMR2/Cstring.i3, IBMRT/Cstring.i3, IRIX5/Cstring.i3, LINUX/Cstring.i3, LINUXELF/Cstring.i3, LINUXLIBC6/Cstring.i3, NEXT/Cstring.i3, NT386/Cstring.i3, NetBSD2_i386/Cstring.i3, OKI/Cstring.i3, OS2/Cstring.i3, PPC_DARWIN/Cstring.i3, PPC_LINUX/Cstring.i3, SEQUENT/Cstring.i3, SOLgnu/Cstring.i3, SOLsun/Cstring.i3, SPARC/Cstring.i3, SUN3/Cstring.i3, SUN386/Cstring.i3, UMAX/Cstring.i3, VAX/Cstring.i3: further distill down these 35 nearly identical files put all functions on one line each, as some of them already were merge two of the import lines is now easy to see that the files are really nearly identical save for a few functions added/removed like strxform, strerror 2008-05-13 06:26 jkrell * m3-libs/m3core/src/C/: AIX386/Cstring.i3, ALPHA_OSF/Cstring.i3, AMD64_DARWIN/Cstring.i3, AMD64_LINUX/Cstring.i3, AP3000/Cstring.i3, ARM/Cstring.i3, DS3100/Cstring.i3, FreeBSD/Cstring.i3, FreeBSD2/Cstring.i3, FreeBSD3/Cstring.i3, FreeBSD4/Cstring.i3, HP300/Cstring.i3, HPPA/Cstring.i3, I386_DARWIN/Cstring.i3, IBMR2/Cstring.i3, IBMRT/Cstring.i3, IRIX5/Cstring.i3, LINUX/Cstring.i3, LINUXELF/Cstring.i3, LINUXLIBC6/Cstring.i3, NEXT/Cstring.i3, NT386/Cstring.i3, NetBSD2_i386/Cstring.i3, OKI/Cstring.i3, OS2/Cstring.i3, PPC_DARWIN/Cstring.i3, PPC_LINUX/Cstring.i3, SEQUENT/Cstring.i3, SOLgnu/Cstring.i3, SOLsun/Cstring.i3, SPARC/Cstring.i3, SUN3/Cstring.i3, SUN386/Cstring.i3, UMAX/Cstring.i3, VAX/Cstring.i3: distill these 35 nearly identical files down so that their near identicalness can be better seen and removed; it is bad to clone headers, and it is worse to clone clones of headers get size_t from Cstddef, remove all comments these functions are all well known and well documented elsewhere 2008-05-13 06:24 jkrell * m3-libs/m3core/src/C/Common/Cstddef.i3: size_t = unsigned_long; ptrdiff_t = long; no actual change 2008-05-13 05:44 jkrell * m3-libs/m3core/src/C/m3makefile: remove platform switch, at least for now these files are currently identical (except for whitespace) I can see there being setjmp/longjmp changes though, as NT386GNU as a "signal state" and NT386 does not, so NT386GNU has two varieties of setjmp/longjmp, like other Posix platforms, I think, probably.. 2008-05-13 04:45 jkrell * m3-sys/: m3front/src/misc/Marker.m3, m3middle/src/Target.i3, m3middle/src/Target.m3: remove Global_handler_stack, just assume it is FALSE; it being TRUE is an optimization on less interesting platforms that never use kernel threads 2008-05-12 20:32 hosking * m3-sys/m3cc/gcc/gcc/m3cg/parse.c: These are always t_int = Target.Integer.cg_type as per structural constraints in CG.m3. If they are not then there is something wrong about the way the backend was compiled (wrong target!), since t_int depends on BITS_WORD_SIZE of the target. 2008-05-12 20:15 hosking * m3-sys/m3middle/src/Target.m3: Cleanup weirdo DOS formatting. 2008-05-12 19:31 jkrell * www/uploaded-archives/update_download_index.sh: allow much broader names -- allow 'boot', allow omitting the redundant OSTYPE 2008-05-12 19:24 jkrell * m3-libs/libm3/src/random/m3makefile, m3-libs/m3core/src/C/SPARC32_LINUX/Csetjmp.i3, m3-libs/m3core/src/C/SPARC32_LINUX/Csignal.i3, m3-libs/m3core/src/C/SPARC32_LINUX/Cstdio.i3, m3-libs/m3core/src/C/SPARC32_LINUX/m3makefile, m3-libs/m3core/src/C/SPARC64_LINUX/Csetjmp.i3, m3-libs/m3core/src/C/SPARC64_LINUX/Csignal.i3, m3-libs/m3core/src/C/SPARC64_LINUX/Cstdio.i3, m3-libs/m3core/src/C/SPARC64_LINUX/m3makefile, m3-libs/m3core/src/float/m3makefile, m3-libs/m3core/src/runtime/m3makefile, m3-libs/m3core/src/runtime/SPARC32_LINUX/RTMachine.i3, m3-libs/m3core/src/runtime/SPARC32_LINUX/RTSignal.m3, m3-libs/m3core/src/runtime/SPARC32_LINUX/RTThread.m3, m3-libs/m3core/src/runtime/SPARC32_LINUX/RTThreadC.c, m3-libs/m3core/src/runtime/SPARC32_LINUX/m3makefile, m3-libs/m3core/src/runtime/SPARC64_LINUX/RTMachine.i3, m3-libs/m3core/src/runtime/SPARC64_LINUX/RTSignal.m3, m3-libs/m3core/src/runtime/SPARC64_LINUX/RTThread.m3, m3-libs/m3core/src/runtime/SPARC64_LINUX/RTThreadC.c, m3-libs/m3core/src/runtime/SPARC64_LINUX/m3makefile, m3-libs/m3core/src/runtime/common/Compiler.tmpl, m3-libs/m3core/src/runtime/common/RTCollector.m3, m3-libs/m3core/src/thread/m3makefile, m3-libs/m3core/src/time/POSIX/m3makefile, m3-libs/m3core/src/unix/m3makefile, m3-libs/m3core/src/unix/big-endian/Uin.m3, m3-libs/m3core/src/unix/big-endian/m3makefile, m3-libs/m3core/src/unix/linux-32/Ustat.i3, m3-libs/m3core/src/unix/linux-sparc32/UpthreadMachine.i3, m3-libs/m3core/src/unix/linux-sparc32/Uucontext.i3, m3-libs/m3core/src/unix/linux-sparc32/m3makefile, m3-libs/m3core/src/unix/linux-sparc64/UpthreadMachine.i3, m3-libs/m3core/src/unix/linux-sparc64/Uucontext.i3, m3-libs/m3core/src/unix/linux-sparc64/m3makefile, m3-sys/cminstall/src/config/NT386.common, m3-sys/cminstall/src/config-no-install/AMD64_LINUX, m3-sys/cminstall/src/config-no-install/LINUXLIBC6, m3-sys/cminstall/src/config-no-install/Linux.common, m3-sys/cminstall/src/config-no-install/PPC_DARWIN, m3-sys/cminstall/src/config-no-install/PPC_LINUX, m3-sys/cminstall/src/config-no-install/SOLgnu, m3-sys/m3cc/gcc/gcc/m3cg/parse.c, m3-sys/m3cc/src/m3makefile, m3-sys/m3middle/src/Target.i3, m3-sys/m3middle/src/Target.m3, scripts/python/pylib.py, scripts/win/boot1.cmd, scripts/win/upgrade.cmd: SPARC32_LINUX up and working bootstrapped from NT386GNU almost bootstrappable from NT386, but not quite due to stat alignment and then built natively -fPIC off could be artifact of cross build, will try with it on with -fPIC on, RTLinker__InitRuntime crashes on the second or so line needs further investigation garbage collection off possibly just paranoia, or maybe enabled getting further needs further investigation -keep produces a broken m3make.args that breaks the subsequent -ship omitting -keep works needs further investigation Tony, please review parse.c add some SPARC64_LINUX support while at it cleanup my config files move constants to constants -- -gstabs+, -fPIC, etc. may need to move some back, depending on what I bootstrap from, we'll see later always use -gstabs+, -g and -ggdb usually crash cm3cg (but probably not gcc) parse.c: enable some more tracing loosen assertions on SPARC, needs review/investigation, probably not right, either they should always be loosened, or they need to be fixed, shouldn't be sparc-specific Target.m3, .i3 Add SPARC32_LINUX, SPARC64_LINUX move Global_handler_stack to individual targetes -- we should probably just get rid of this and assume it is always false; all active targets have it false, and making it true merely enables an optimization -- manipulate data inline instead of making a function call remove pattern matching from upgrade.cmd, it was only to build cminstall, which isn't needed in upgrade, and that I never use pylib.py support SPARC, remove quotes, dos2unix bootstrap and/or min/std distributions of SPARC32_LINUX to be available shortly 2008-05-12 17:26 jkrell * m3-sys/cminstall/src/config/cm3.cfg: fix for when config file not found next to compiler, duh 2008-05-12 16:21 jkrell * m3-sys/cminstall/src/config/cm3.cfg: fix probing -- a) include the found file, not the file named 'TRUE', two of the files were the same twice in a row due to failure to edit after copy/paste 2008-05-12 15:46 jkrell * m3-sys/cm3/src/: Builder.m3, M3Build.m3, M3Path.i3, M3Path.m3, Makefile.m3: do what was probably really intended here -- enable .m3ship files to be used on the target of a bootstrap build, where host/target vary in path syntax I believe this hypothetically works IF the roots are the same on the two system, easy and quake were still written in C, not easy. And it depends on Win32 CM3_INSTALL being set like /cm3 instead of c:\cm3, which works fine. bootstrapping can be achieved without this 2008-05-11 21:34 jkrell * scripts/python/pylib.py: more and less experimental changes for bootstrapping CM3_FLAGS can be used to pass -boot yet another attempt for one unified path syntax, will probably just end munging all the .m3ship files generalize check for "recent Linux" platform ie: ones that never had gcvm 2008-05-11 21:29 jkrell * m3-libs/m3core/src/runtime/common/RTLinkerC.c: remove dependency on header missing from a small number of build environments 2008-05-09 22:30 jkrell * scripts/pkginfo.txt: sysutils needed in front for win/make-dist.cmd to work 2008-05-09 22:21 jkrell * scripts/win/make-dist.cmd: put ostype back in archive names for now until/unless limit is removed from make-uploaded-archive.sh 2008-05-09 21:34 jkrell * scripts/python/make-dist.py: put ostype back in archive names for now until/unless limit is removed from make-uploaded-archive.sh 2008-05-09 20:59 jkrell * scripts/python/pylib.py: don't require cm3cg for NT386 2008-05-09 08:11 jkrell * m3-libs/m3core/src/runtime/: WIN32/RTArgs.m3, common/RTLinker.i3, common/RTLinker.m3, common/RTLinkerC.c, common/m3makefile: fix RTArgs environment variable functionality that has always been broken in some form or another, at least back to 3.6, and which changed but didn't get fixed by Olaf in 2003 with: http://dcvs.elegosoft.com/cgi-bin/cvsweb.cgi/cm3/m3-libs/m3core/src/runtime/WIN32/RTArgs.m3.diff?r1=1.1;r2=1.2;f=u I knew! nothing had actually changed here, just that it never really worked, but was slow to get around to looking into it. This change is "carefully" written to avoid a compiler/runtime coupling. However a better change is probably to change the compiler, when targeting NT386, to always call GetEnvironmentStrings and never use the envp parameter to main. Or perhaps something sleazy like setting the low bit to indicate "type". Or perhaps for main/WinMain to set the "useFlatEnv" variable (through a function call, to avoid dynamically linked data). 2008-05-09 05:49 jkrell * m3-libs/libm3/src/os/WIN32/ProcessWin32.m3: remove tabs 2008-05-08 18:30 jkrell * m3-sys/m3tests/src/p2/p210/Main.m3: I messed up trying to add this a few days ago, here it is now; note that this successfully compiles and then fails at runtime; needs further investigation 2008-05-08 18:21 jkrell * scripts/win/: make-dist.cmd, sysinfo.cmd: some cleanup, restructuring, and accounting for other changes including but not necessarily limited to: remove 'core' and 'base' distributions, leaving just 'min' and 'std' min is cm3, mklib, m3core, libm3 std is basically everything rename archives to cm3-min-NT386-version.zip OSTYPE is implied, as well as 'WIN32' being sort of wrong for AMD64/IA64 make one symbols archive that is the union of what they all have copy NT386* config files and create one line cm3.cfg that just has include this is what scripts/python/make-dist.py was already doing in sysinfo.cmd, don't do everything twice; it mostly doesn't matter but the echoed stuff was being printed repeatedly; more than twice actually, since sysinfo is called by make-dist repeatedly, but now at least it is somwhat reduced no longer have make-dist do the careful build compiler with previous runtime step; usually that isn't needed, and when it is, user should run upgrade to accomplish it trim out *.lib.sa files except for m3core, libm3, sysutils *.lib.sa are generally large and dead; they are for building 'standalone' sa == standalone however the three that are preserved are important for bootstrapping in the event that an old compiler cannot build a current runtime, as happened for example when LONGINT was introduced in the runtime (and the compiler is careful not to use LONGINT) anyone that really wants to build standalone with more than just those .libs, can build from source. reasonable? 2008-05-08 15:54 jkrell * m3-sys/cminstall/src/config/NT386.common: whitespace 2008-05-08 15:52 jkrell * m3-sys/cminstall/src/config/NT386.common: reduce output 2008-05-08 15:04 jkrell * m3-sys/cminstall/src/config/NT386.main: This should work for Olaf and has less duplication of code. 2008-05-08 14:58 jkrell * m3-win/import-libs/src/m3makefile: always export, so that rebuilding and reshipping will honor the current CM3_INSTALL environment variable, rather than the one that was current when a file was last out of date; this way making a distribution need not rebuild files for both min and std, only reship them 2008-05-08 11:37 jkrell * scripts/win/upgrade.cmd: a more consistent order, but ok either way 2008-05-08 11:36 jkrell * scripts/win/: backup-pkgs.cmd, clearenv.cmd, do-cm3-all.cmd, do-cm3-base.cmd, do-cm3-core.cmd, do-cm3-front.cmd, do-cm3-min.cmd, do-cm3-std.cmd, do-pkg.cmd, find-packages.cmd, install-cm3-compiler.cmd, make-dist.cmd, pkgcmds.cmd, pkggnu_clean.cmd, pkginfo.cmd, sysinfo.cmd, upgrade.cmd: unix2dos, since I had a nonsensical error that this fixed 2008-05-08 06:00 jkrell * scripts/python/pylib.py: include Linux.common config file in Linux distributions 2008-05-07 16:41 jkrell * m3-libs/m3core/src/runtime/common/RTCollector.m3: temporarily turn off garbage collection on AMD64_LINUX; this enables very much progress, and therefore shows itself as the 'one' thing left 2008-05-07 16:07 jkrell * scripts/python/: make-dist.py, pylib.py: remove ostype from archive names, platform suffices; restore make-dist to working by restoring ShipCompiler; enable cleaning in some places; minor cleanup 2008-05-07 14:51 jkrell * m3-libs/m3core/src/runtime/common/Compiler.tmpl: expose ThisOS and ThisPlatform as strings (or 'text'?) and not just enums; exposure in quake however is going to have to clone this file in order for building a new compiler to NOT require a new runtime; in Quake these shall be available as HOST and HOST_OSTYPE, analogous to TARGET and OSTYPE and for example can, at least temporarily, guide my proxy cm3.cfg to a default -- default to a native build (until such time as more code is merged into cm3 itself) 2008-05-05 22:34 rcoleburn * scripts/win/upgrade.cmd: Add "sysutils" to "p_compiler" just before "m3quake". This seems to solve the problem I was having, but Jay Krell should double-check this fix and make sure I've done this right. --Randy Coleburn 2008-05-05 05:23 wagner * scripts/regression/defs.sh: use m3-sys/m3tests/TARGET/res.ko for error reporting 2008-05-05 04:48 wagner * m3-sys/m3tests/: PkgTags, src/m3makefile: keep lists of test results in res.ok and res.ko modified: m3tests/src/m3makefile 2008-05-04 20:37 wagner * scripts/regression/defs.sh: try to use the combined stdout/stderr for m3tests evaluation; at least it cannot make the situation worse (stderr is currently empty) 2008-05-04 15:33 jkrell * m3-sys/cminstall/src/config/NT386.common: better -- use hand.obj by full path, except when building m3core.dll, and check if it exists, to enable bootstrapping against old runtime..which is ok for the compiler since it is standalone but otherwise dubious 2008-05-04 15:24 jkrell * m3-sys/cminstall/src/config/NT386.common: quick hack to statically link hand.obj always rather than figure out how to turn _lowbits and _highbits into pointers this 'hack' is actually pretty darn acceptable issues to maybe improve are: - don't statically link the rest of hand.obj, just the data - don't depend on a loose .obj, merge it into m3core.lib - use hand.o for GNU ld - link to it by full path instead of relying on %LIB% This is probably the biggest. Works for me via scripts/python, but probably won't otherwise, might be so bad as do break Olaf. will fix that in a few minutes 2008-05-04 13:55 jkrell * m3-sys/m3tests/src/p2/p206/Main.m3: somehow lost this change too -- this test fails to compile, but I know think this is the correct behavior; other problems discovered with similar code have been moved to p209 and p210; if indeed this should not compile, it should be an 'e' test instead of a 'p' test -- 'e' tests are compiled to verify they error well, 'p' tests are compiled to be run to see if they output ok, not a big deal probably, some 'p' tests just print 'ok' 2008-05-04 13:52 jkrell * m3-sys/m3tests/src/: m3makefile, p2/p209/Main.m3, p2/p210/m3makefile, p2/p210/stderr.build, p2/p210/stderr.pgm, p2/p210/stdout.build: somehow p210 were not added, despite showing as added in the commit could be due to copying p209 to p210? also had it backwards which is a compile problem, which is a runtime problem, minor also dos2unix just in case 2008-05-04 13:47 jkrell * m3-sys/m3tests/src/: m3makefile, p2/p209/Main.m3, p2/p209/m3makefile, p2/p209/stderr.build, p2/p209/stderr.pgm, p2/p209/stdout.build, p2/p209/stdout.pgm: upon further investigation variables with no type but an open array initializer seem rightfully invalid a key point is that constants set to open arrays really do have a type that is open array, not fixed size based on the number of elements in the initializer Therefore break out from p206 what was discovered to hit an assertion failure in the compiler, as well as a similar variant that compiles but hits an assertion failure at startup. ie: revert p206 to its old self that fails to compile and is I think correct -- should be an "e" test not a "p" test, and add two new "p" tests p209 and p210 that both presently fail but I believe should succeed, pending compiler and possibly runtime fixes (perhaps the runtime problem in the case that successfully compiles is due to a mis-compile, perhaps the same problem that causes the other case to fail to compile) 2008-05-04 12:54 jkrell * m3-sys/m3quake/src/extensions-2008-1-29.txt: fix documentation -- contains vs. tcontains, del_chars takes just two parameters 2008-05-04 12:03 jkrell * m3-sys/m3tests/src/: Test.common, r0/r001/stdout.build: get r0/r001 to have the same output at least on NT386 and LINUXLIBC6; it has always worked, but the output varies across platforms, this might need more work, and might be 'impossible' to fix, and might just not be worth running in Tinderbox, even this fix is very not great 2008-05-04 11:31 jkrell * m3-sys/m3tests/src/m3makefile: p208 passes ok, remove it from list of failing tests 2008-05-04 11:27 jkrell * m3-sys/m3tests/src/Test.common: turns out none of the tests are expected to issue uncaught assertion failures, so we don't need this; if we do need it, LINUXLIBC6 prints either 'aborted' or nothing, no stack trace 2008-05-04 11:03 jkrell * m3-sys/m3front/src/: builtinLong/LongPlus.m3, builtinOps/Dispose.m3, builtinWord/WordPlus.m3, exprs/AddExpr.m3, exprs/AddressExpr.m3, exprs/AndExpr.m3, exprs/CallExpr.m3, exprs/CastExpr.m3, exprs/CheckExpr.m3, exprs/CompareExpr.m3, exprs/ConcatExpr.m3, exprs/DerefExpr.m3, exprs/DivExpr.m3, exprs/DivideExpr.m3, exprs/EnumExpr.m3, exprs/EqualExpr.m3, exprs/ExprParse.m3, exprs/InExpr.m3, exprs/IntegerExpr.m3, exprs/KeywordExpr.m3, exprs/MethodExpr.m3, exprs/ModExpr.m3, exprs/MultiplyExpr.m3, exprs/NamedExpr.m3, exprs/NarrowExpr.m3, exprs/NegateExpr.m3, exprs/NilChkExpr.m3, exprs/NotExpr.m3, exprs/OrExpr.m3, exprs/PlusExpr.m3, exprs/ProcExpr.m3, exprs/QualifyExpr.m3, exprs/RangeExpr.m3, exprs/RecordExpr.m3, exprs/ReelExpr.m3, exprs/SubscriptExpr.m3, exprs/SubtractExpr.m3, exprs/TextExpr.m3, exprs/TypeExpr.m3, exprs/VarExpr.m3, misc/CG.m3, misc/Coverage.m3, misc/Scanner.m3, misc/Scope.m3, misc/Token.i3, stmts/AssertStmt.m3, stmts/AssignStmt.m3, stmts/BlockStmt.m3, stmts/CallStmt.m3, stmts/CaseStmt.m3, stmts/DebugStmt.m3, stmts/EvalStmt.m3, stmts/ExitStmt.m3, stmts/ForStmt.m3, stmts/IfStmt.m3, stmts/LockStmt.m3, stmts/LoopStmt.m3, stmts/RaiseStmt.m3, stmts/RepeatStmt.m3, stmts/ReturnStmt.m3, stmts/TryFinStmt.m3, stmts/TryStmt.m3, stmts/TypeCaseStmt.m3, stmts/WhileStmt.m3, stmts/WithStmt.m3, types/ArrayType.m3, types/EnumType.m3, types/NamedType.m3, types/ObjectType.m3, types/OpaqueType.m3, types/RefType.m3, types/SetType.m3, types/Type.m3, types/UserProc.m3, values/Constant.i3, values/Constant.m3, values/EnumElt.m3, values/Exceptionz.m3, values/Field.m3, values/Formal.m3, values/Method.m3, values/Module.m3, values/Procedure.m3, values/Revelation.m3, values/Tipe.m3, values/Variable.m3: add more information printed for some assertion failures move some assertion failures to run more often/earlier They will tend to run twice now; that isn't the point; the point is to run them earlier, when the work is queued, rather than later when it is dequed. Said assertions occur nowhere building 'std' on NT386, but do occur building testcase p206; this might be somewhat platform specific since it is alignment related. Remove a bunch of tabs. Fix warning leftover from Tony replacing my fix related to value vs. lvalue of constant arrays being used for subarray. 2008-05-04 10:57 jkrell * m3-sys/m3tests/src/: Test.common, m3makefile: oops fix the previous.. 2008-05-04 10:33 jkrell * m3-sys/m3tests/src/Test.common: use the abstraction 2008-05-04 10:32 jkrell * m3-sys/m3tests/src/m3makefile: oops, mind the extra echoing, which is stable enough, but I don't want to update all the stdout.build files 2008-05-04 10:28 jkrell * m3-sys/m3tests/src/: Test.common, m3makefile, p0/p010/Main.m3: output to foo.raw and then create foo that is "normalized" move the normalization code to Test.common so it happens when building in leaves share Test.common into the root, and sleazily avoid program("pgm") via stale("Test.common") (with a clear variable name to explain the trickery) add to the "normalization" truncation of stack dumps before they become "random" we have to fully remove stack dumps, as there is a file that just says "aborted" in place of a stack dump, perhaps stack dumps don't work on most platforms? Either way, this is progress along the right lines. for example, test r0\r001 prints: C:\dev2\cm3\m3-sys\m3tests\src\r0\r001>type NT386\stderr.pgm.raw *** *** runtime error: *** Unhandled exception: Main.a *** file "..\Main.m3", line 13 *** Stack trace: FP PC Procedure --------- --------- ------------------------------- 0x12fee0 0x401084 Main_M3 + 0x34 in ..\Main.m3 C:\dev2\cm3\m3-sys\m3tests\src\r0\r001>type NT386\stderr.pgm *** *** runtime error: *** Unhandled exception: Main.a *** file "../Main.m3", line 13 *** Stack trace: FP PC Procedure --------- --------- ------------------------------- 0x Clearly the non-raw output is more viable. The unhandled exception is expected. remove tabs in p0/p010/Main.m3 2008-05-04 09:26 jkrell * m3-sys/m3tests/src/m3makefile: simpler 2008-05-04 09:25 jkrell * m3-sys/m3tests/src/m3makefile: oops 2008-05-04 07:44 jkrell * m3-sys/m3tests/src/p2/p206/Main.m3: modify test case to reveal an assertion failure in the compiler, while leaving the previous code commented out for now 2008-05-04 07:34 jkrell * m3-sys/m3tests/src/p2/p206/Main.m3: extend broken test case to suggest that maybe it is in fact reasonable -- if open arrays can be constants, with an implied size, then an open array initializer seems reasonable too 2008-05-04 07:10 jkrell * m3-sys/m3tests/src/e0/e026/stdout.build: the output is slightly different here now and that seems definitely ok 2008-05-04 07:03 jkrell * m3-sys/m3tests/src/: src.pl, r0/r001/stderr.pgm, r0/r003/stderr.pgm: remove src from pgm output too, not just build output 2008-05-04 06:56 jkrell * scripts/regression/defs.sh, m3-sys/m3tests/src/m3makefile: always use diff and head if they work, and not if they don't 2008-05-04 06:29 jkrell * m3-sys/m3tests/src/p2/p205/stdout.pgm: update output 2008-05-04 06:28 jkrell * m3-sys/m3front/src/types/Brand.m3: fix test case e026 -- compile-time detection of duplicate brands gotta love the constant reinvention of linked lists and hash tables.. more code => more bugs => reuse code instead 2008-05-04 03:34 jkrell * m3-sys/: m3tests/src/m3makefile, cminstall/src/config/NT386.common: adjust echoing for tests, unfortunately along with the name of the file compiled, we'll lose any C compilation errors here 2008-05-04 02:48 jkrell * m3-sys/m3tests/src/: src.pl, e0/e001/stdout.build, e0/e005/stdout.build, e0/e006/stdout.build, e0/e008/stdout.build, e0/e009/stdout.build, e0/e010/stdout.build, e0/e013/stdout.build, e0/e014/stdout.build, e0/e015/stdout.build, e0/e017/stdout.build, e0/e018/stdout.build, e0/e019/stdout.build, e0/e021/stdout.build, e0/e023/stdout.build, e0/e024/stdout.build, e0/e025/stdout.build, e0/e026/stdout.build, e0/e028/stdout.build, e0/e029/stdout.build, e0/e031/stdout.build, e0/e032/stdout.build, e0/e035/stdout.build, p0/p004/stdout.build, p0/p019/stdout.build, p0/p020/stdout.build, p0/p023/stdout.build, p0/p066/stdout.build, p0/p076/stdout.build, p0/p083/stdout.build, p0/p085/stdout.build, p0/p098/stdout.build, p1/p104/stdout.build, p1/p116/stdout.build, p1/p122/stdout.build, p1/p140/stdout.build, r0/r003/stdout.build: remove "src" from expected output (since removing copying/linking) 2008-05-04 02:45 jkrell * m3-sys/m3tests/src/: src.pl, e0/e001/stdout.build, e0/e002/stdout.build, e0/e003/stdout.build, e0/e004/stdout.build, e0/e005/stdout.build, e0/e006/stdout.build, e0/e008/stdout.build, e0/e009/stdout.build, e0/e010/stdout.build, e0/e011/stdout.build, e0/e012/stdout.build, e0/e013/stdout.build, e0/e014/stdout.build, e0/e015/stdout.build, e0/e016/stdout.build, e0/e017/stdout.build, e0/e018/stdout.build, e0/e019/stdout.build, e0/e021/stdout.build, e0/e022/stdout.build, e0/e023/stdout.build, e0/e024/stdout.build, e0/e025/stdout.build, e0/e026/stdout.build, e0/e027/stdout.build, e0/e028/stdout.build, e0/e029/stdout.build, e0/e031/stdout.build, e0/e032/stdout.build, e0/e033/stdout.build, e0/e034/stdout.build, e0/e035/stdout.build, p0/p004/stdout.build, p0/p017/stdout.build, p0/p019/stdout.build, p0/p020/stdout.build, p0/p023/stdout.build, p0/p050/stdout.build, p0/p066/stdout.build, p0/p076/stdout.build, p0/p083/stdout.build, p0/p085/stdout.build, p0/p098/stdout.build, p1/p104/stdout.build, p1/p116/stdout.build, p1/p122/stdout.build, p1/p135/stdout.build, p1/p140/stdout.build, p1/p168/stdout.build, r0/r001/stdout.build, r0/r003/stdout.build: remove "src" from output paths 2008-05-03 21:00 wagner * scripts/regression/defs.sh: always use posix tools for m3tests during regression 2008-05-03 20:23 jkrell * m3-sys/m3tests/src/: e0/e001/stdout.build, e0/e002/stdout.build, e0/e003/stdout.build, e0/e004/stdout.build, e0/e005/stdout.build, e0/e006/stdout.build, e0/e008/stdout.build, e0/e009/stdout.build, e0/e010/stdout.build, e0/e011/stdout.build, e0/e012/stdout.build, e0/e013/stdout.build, e0/e014/stdout.build, e0/e015/stdout.build, e0/e016/stdout.build, e0/e017/stdout.build, e0/e018/stdout.build, e0/e019/stdout.build, e0/e020/stdout.build, e0/e021/stdout.build, e0/e022/stdout.build, e0/e023/stdout.build, e0/e024/stdout.build, e0/e025/stdout.build, e0/e026/stdout.build, e0/e027/stdout.build, e0/e028/stdout.build, e0/e029/stdout.build, e0/e031/stdout.build, e0/e032/stdout.build, e0/e033/stdout.build, e0/e034/stdout.build, e0/e035/stdout.build, p0/p004/stdout.build, p0/p017/stdout.build, p0/p019/stdout.build, p0/p020/stdout.build, p0/p023/stdout.build, p0/p050/stdout.build, p0/p066/stdout.build, p0/p076/stdout.build, p0/p083/stdout.build, p0/p085/stdout.build, p0/p098/stdout.build, p1/p104/stdout.build, p1/p116/stdout.build, p1/p122/stdout.build, p1/p140/stdout.build, p1/p168/stdout.build, p2/p204/stdout.build, p2/p205/stdout.pgm, p2/p206/stdout.build, p2/p207/stdout.build, r0/r001/stderr.pgm, r0/r001/stdout.build, r0/r002/stdout.build, r0/r003/stderr.pgm, r0/r003/stdout.build, r0/r004/stdout.build: This should bring back all the expected outputs to what they were one week ago. Rather than merely replace them with current output, let's manually edit the paths in them. 2008-05-03 19:51 jkrell * m3-sys/m3tests/src/p2/p205/I.i3: forgot to check this in 2008-05-03 14:41 hosking * m3-sys/m3tests/src/p2/p205/Main.m3: Eliminate ^M at end of line (dos2unix). 2008-05-03 14:21 hosking * m3-sys/m3front/src/exprs/QualifyExpr.m3: Let's try this instead (cribbed from NamedExpr.CompileLV). 2008-05-03 11:17 jkrell * m3-sys/m3middle/src/Target.m3: fix and unify NT386 jmpbuf/setjmp it APPEARS jmpbuf was understated for Visual C++ though it was probably ok it appears if you compile C/C++, the compiler generates a call to _setjmp3, which indeed uses more of the declared-16 jmpbuf but that if we call _setjmp directly, it only uses 8 Cygwin was overstated because their setjmp.h appears to confuse bytes and ints, it only uses 13. So unify the former 8 and 13 to 16. As well, Cygwin provides aliased setjmp and _setjmp, so unify on _setjmp NOTE that using _setjmp3 for Visual C++ is probably desirable but cross that bridge another time, perhaps we'll just stop using setjmp entirely therefore making the three NT386 flavors much more similar 2008-05-03 09:01 jkrell * m3-sys/m3tests/src/p2/p205/Main.m3: same thing 2008-05-03 09:00 jkrell * m3-sys/m3tests/src/p2/p205/: Main.m3, stdout.pgm: extend test case, seems to reveal more bugs 2008-05-03 08:28 jkrell * m3-sys/m3tests/src/p1/p130/Main.m3: inline asserts so the line number is useful 2008-05-03 08:25 jkrell * m3-sys/m3tests/src/p1/p130/Main.m3: adapt it to compile 2008-05-03 08:19 jkrell * m3-sys/m3tests/src/: p1/p116/stdout.build, p2/p205/stderr.build, r0/r001/stderr.pgm: more updates 2008-05-03 08:14 jkrell * m3-sys/m3cc/gcc/gcc/m3cg/parse.c: sacrifice some maintainability for debuggability -- put enum values in comments next to the enums; and print most numbers in hex, as it generally more natural in all things computer related (except line numbers, and generated identifiers to avoid churn) 2008-05-03 08:11 jkrell * m3-sys/cminstall/src/config/NT386.common: comment only, essentially 2008-05-03 08:04 jkrell * m3-sys/: m3front/src/exprs/QualifyExpr.m3, m3front/src/values/Constant.i3, m3front/src/values/Constant.m3, m3tests/src/m3makefile: enough to fix test p2/p205, but needs more attention as the fix seems to be at the wrong abstraction level and not cover every case in particular, confusion around lvalues vs. non-lvalues in particular, read the comments in QualifyExpr.m3 2008-05-03 05:16 jkrell * m3-sys/m3tests/src/m3makefile: cleanup -- make names clearer and remove print that isn't all that useful, esp. perhaps now that this is all works more simply (? or do I just understand it now?) 2008-05-03 05:08 jkrell * m3-sys/m3tests/src/r0/r003/stderr.pgm: update for LINUXLIBC6 on birch 2008-05-02 19:12 jkrell * m3-sys/m3tests/src/m3makefile: cleanup 2008-05-02 19:05 jkrell * m3-sys/m3tests/src/m3makefile: restore minus sign instead of try_exec, and remove annoying looking double space; this did used to have an at too but avoid that if possible so we can see and reveal/teach what is going on 2008-05-02 18:51 jkrell * m3-sys/m3tests/src/: e0/e029/stderr.build, e0/e029/stdout.build, p0/p004/stdout.build, p0/p051/stdout.build, p2/p204/stderr.build, p2/p204/stdout.build, p2/p205/stderr.build, p2/p207/stdout.build: where posix and win32 vary for now, take posix, from ppc_darwin 2008-05-02 18:49 jkrell * m3-sys/m3tests/src/m3makefile: always standalone, doesn't make much bad difference and the good difference is it the tests can run on PPC_DARWIN 2008-05-02 18:48 jkrell * m3-sys/m3tests/src/m3makefile: dos2unix 2008-05-02 18:42 jkrell * m3-sys/m3tests/src/: m3makefile, e0/e001/stdout.build, e0/e002/stdout.build, e0/e003/stdout.build, e0/e004/stdout.build, e0/e005/stdout.build, e0/e006/stdout.build, e0/e008/stdout.build, e0/e009/stdout.build, e0/e010/stdout.build, e0/e011/stdout.build, e0/e012/stdout.build, e0/e013/stdout.build, e0/e014/stdout.build, e0/e015/stdout.build, e0/e016/stdout.build, e0/e017/stdout.build, e0/e018/stdout.build, e0/e019/stdout.build, e0/e020/stdout.build, e0/e021/stdout.build, e0/e022/stdout.build, e0/e023/stdout.build, e0/e024/stdout.build, e0/e025/stdout.build, e0/e026/stdout.build, e0/e027/stdout.build, e0/e028/stdout.build, e0/e029/stderr.build, e0/e029/stdout.build, e0/e031/stdout.build, e0/e032/stdout.build, e0/e033/stdout.build, e0/e034/stdout.build, e0/e035/stdout.build, p0/p004/stdout.build, p0/p017/stdout.build, p0/p019/stdout.build, p0/p020/stdout.build, p0/p023/stdout.build, p0/p050/stdout.build, p0/p051/stdout.build, p0/p066/stdout.build, p0/p076/stdout.build, p0/p083/stdout.build, p0/p085/stdout.build, p0/p098/stdout.build, p1/p104/stdout.build, p1/p122/stdout.build, p1/p140/stdout.build, p1/p168/stdout.build, p2/p204/stderr.build, p2/p205/stderr.build, p2/p206/stdout.build, p2/p207/stdout.build, r0/r001/stdout.build, r0/r002/stdout.build, r0/r003/stdout.build, r0/r004/stdout.build: regenerate expected output, at least for Win32, a few will be redone for Posix, where the backend doesn't fail 2008-05-02 18:21 jkrell * m3-sys/m3tests/src/m3makefile: don't stop after first failure and don't fail at end due to over-applied systematic change 2008-05-02 18:17 jkrell * m3-sys/m3tests/src/m3makefile: rename variables for clarity; prepare to regenerate expected build output 2008-05-02 15:22 jkrell * m3-sys/m3tests/src/m3makefile: adapt to new structure 2008-05-02 15:10 jkrell * m3-sys/m3tests/src/m3makefile: fix paths to Cygwin diff 2008-05-02 12:42 wagner * m3-sys/m3tests/src/m3makefile: fix cmp_file: adapt to Jay's structural changes All file names in error messages and warning are now wrong and reported as an error :-/ 2008-05-02 09:03 jkrell * m3-sys/m3tests/src/: Test.common, append_include.cmd, build_all.cmd, m3makefile, c0/c001/m3makefile, c0/c002/m3makefile, c0/c003/m3makefile, c0/c004/m3makefile, c0/c005/m3makefile, c0/c006/m3makefile, c0/c007/m3makefile, c0/c008/m3makefile, c0/c009/m3makefile, c0/c010/m3makefile, c0/c011/m3makefile, c0/c012/m3makefile, c0/c013/m3makefile, c0/c014/m3makefile, c0/c015/m3makefile, c0/c016/m3makefile, c0/c017/m3makefile, c0/c018/m3makefile, c0/c019/m3makefile, c0/c020/m3makefile, c0/c021/m3makefile, c0/c022/m3makefile, c0/c023/m3makefile, c0/c024/m3makefile, c0/c025/m3makefile, c0/c026/m3makefile, c0/c027/m3makefile, c0/c028/m3makefile, c0/c029/m3makefile, c0/c030/m3makefile, c0/c031/m3makefile, c0/c032/m3makefile, c0/c033/m3makefile, c0/c034/m3makefile, c0/c035/m3makefile, c0/c036/m3makefile, c0/c037/m3makefile, c0/c038/m3makefile, c0/c039/m3makefile, c0/c040/m3makefile, c0/c041/m3makefile, c0/c042/m3makefile, c0/c043/m3makefile, c0/c044/m3makefile, c0/c045/m3makefile, c0/c046/m3makefile, c0/c047/m3makefile, c0/c048/m3makefile, c0/c049/m3makefile, c0/c050/m3makefile, c0/c051/m3makefile, c0/c052/m3makefile, c0/c053/m3makefile, c0/c054/m3makefile, c0/c055/m3makefile, c0/c056/m3makefile, c0/c057/m3makefile, c0/c058/m3makefile, c0/c059/m3makefile, c0/c060/m3makefile, c0/c061/m3makefile, c0/c062/m3makefile, c0/c063/m3makefile, c0/c064/m3makefile, c0/c065/m3makefile, c0/c066/m3makefile, c0/c067/m3makefile, c0/c068/m3makefile, c0/c069/m3makefile, c0/c070/m3makefile, c0/c071/m3makefile, c0/c072/m3makefile, c0/c073/m3makefile, c0/c074/m3makefile, c0/c075/m3makefile, c0/c076/m3makefile, c0/c077/m3makefile, c0/c078/m3makefile, c0/c079/m3makefile, c0/c080/m3makefile, c0/c081/m3makefile, c0/c082/m3makefile, c0/c083/m3makefile, c0/c084/m3makefile, c0/c085/m3makefile, c0/c086/m3makefile, c0/c087/m3makefile, c0/c088/m3makefile, c0/c089/m3makefile, c0/c090/m3makefile, c0/c091/m3makefile, c0/c092/m3makefile, c0/c093/m3makefile, c0/c094/m3makefile, c0/c095/m3makefile, c0/c096/m3makefile, c0/c097/m3makefile, c0/c098/m3makefile, c0/c099/m3makefile, c1/c100/m3makefile, c1/c101/m3makefile, c1/c102/m3makefile, c1/c103/m3makefile, c1/c104/m3makefile, c1/c105/m3makefile, c1/c106/m3makefile, c1/c107/m3makefile, c1/c108/m3makefile, c1/c109/m3makefile, c1/c110/m3makefile, c1/c111/m3makefile, c1/c112/m3makefile, c1/c113/m3makefile, c1/c114/m3makefile, c1/c115/m3makefile, c1/c116/m3makefile, c1/c117/m3makefile, c1/c118/m3makefile, c1/c119/m3makefile, c1/c120/m3makefile, c1/c121/m3makefile, c1/c122/m3makefile, c1/c123/m3makefile, c1/c124/m3makefile, c1/c125/m3makefile, c1/c126/m3makefile, c1/c127/m3makefile, c1/c128/m3makefile, c1/c129/m3makefile, c1/c130/m3makefile, c1/c131/m3makefile, c1/c132/m3makefile, c1/c133/m3makefile, c1/c134/m3makefile, e0/e001/m3makefile, e0/e002/m3makefile, e0/e003/m3makefile, e0/e004/m3makefile, e0/e005/m3makefile, e0/e006/m3makefile, e0/e007/m3makefile, e0/e008/m3makefile, e0/e009/m3makefile, e0/e010/m3makefile, e0/e011/m3makefile, e0/e012/m3makefile, e0/e013/m3makefile, e0/e014/m3makefile, e0/e015/m3makefile, e0/e016/m3makefile, e0/e017/m3makefile, e0/e018/m3makefile, e0/e019/m3makefile, e0/e020/m3makefile, e0/e021/m3makefile, e0/e022/m3makefile, e0/e023/m3makefile, e0/e024/m3makefile, e0/e025/m3makefile, e0/e026/m3makefile, e0/e027/m3makefile, e0/e028/m3makefile, e0/e029/m3makefile, e0/e030/m3makefile, e0/e031/m3makefile, e0/e032/m3makefile, e0/e033/m3makefile, e0/e034/m3makefile, e0/e035/m3makefile, e0/e036/m3makefile, e0/e037/m3makefile, e0/e038/m3makefile, e0/e039/m3makefile, p0/p001/m3makefile, p0/p002/m3makefile, p0/p003/m3makefile, p0/p004/m3makefile, p0/p005/m3makefile, p0/p006/m3makefile, p0/p007/m3makefile, p0/p008/m3makefile, p0/p009/m3makefile, p0/p010/m3makefile, p0/p011/m3makefile, p0/p014/m3makefile, p0/p015/m3makefile, p0/p016/m3makefile, p0/p017/m3makefile, p0/p018/m3makefile, p0/p019/m3makefile, p0/p020/m3makefile, p0/p021/m3makefile, p0/p022/m3makefile, p0/p023/m3makefile, p0/p024/m3makefile, p0/p025/m3makefile, p0/p026/m3makefile, p0/p027/m3makefile, p0/p028/m3makefile, p0/p029/m3makefile, p0/p030/m3makefile, p0/p034/m3makefile, p0/p035/m3makefile, p0/p036/m3makefile, p0/p037/m3makefile, p0/p038/m3makefile, p0/p039/m3makefile, p0/p040/m3makefile, p0/p041/m3makefile, p0/p042/m3makefile, p0/p043/m3makefile, p0/p044/m3makefile, p0/p045/m3makefile, p0/p046/m3makefile, p0/p048/m3makefile, p0/p049/m3makefile, p0/p050/m3makefile, p0/p051/m3makefile, p0/p052/m3makefile, p0/p053/m3makefile, p0/p054/m3makefile, p0/p055/m3makefile, p0/p056/m3makefile, p0/p057/m3makefile, p0/p058/m3makefile, p0/p059/m3makefile, p0/p060/m3makefile, p0/p061/m3makefile, p0/p062/m3makefile, p0/p063/m3makefile, p0/p064/m3makefile, p0/p065/m3makefile, p0/p066/m3makefile, p0/p067/m3makefile, p0/p068/m3makefile, p0/p069/m3makefile, p0/p070/m3makefile, p0/p071/m3makefile, p0/p072/m3makefile, p0/p073/m3makefile, p0/p074/m3makefile, p0/p075/m3makefile, p0/p076/m3makefile, p0/p077/m3makefile, p0/p078/m3makefile, p0/p079/m3makefile, p0/p080/m3makefile, p0/p081/m3makefile, p0/p082/m3makefile, p0/p083/m3makefile, p0/p084/m3makefile, p0/p085/m3makefile, p0/p086/m3makefile, p0/p087/m3makefile, p0/p088/m3makefile, p0/p089/m3makefile, p0/p090/m3makefile, p0/p091/m3makefile, p0/p092/m3makefile, p0/p093/m3makefile, p0/p094/m3makefile, p0/p095/m3makefile, p0/p096/m3makefile, p0/p097/m3makefile, p0/p098/m3makefile, p0/p099/m3makefile, p1/p100/m3makefile, p1/p101/m3makefile, p1/p102/m3makefile, p1/p103/m3makefile, p1/p104/m3makefile, p1/p105/m3makefile, p1/p106/m3makefile, p1/p107/m3makefile, p1/p108/m3makefile, p1/p109/m3makefile, p1/p110/m3makefile, p1/p111/m3makefile, p1/p112/m3makefile, p1/p113/m3makefile, p1/p114/m3makefile, p1/p115/m3makefile, p1/p116/m3makefile, p1/p116b/m3makefile, p1/p117/m3makefile, p1/p118/m3makefile, p1/p119/m3makefile, p1/p120/m3makefile, p1/p121/m3makefile, p1/p122/m3makefile, p1/p123/m3makefile, p1/p124/m3makefile, p1/p125/m3makefile, p1/p126/m3makefile, p1/p127/m3makefile, p1/p128/m3makefile, p1/p129/m3makefile, p1/p130/m3makefile, p1/p131/m3makefile, p1/p132/m3makefile, p1/p133/m3makefile, p1/p134/m3makefile, p1/p135/m3makefile, p1/p136/m3makefile, p1/p137/m3makefile, p1/p138/m3makefile, p1/p139/m3makefile, p1/p140/m3makefile, p1/p141/m3makefile, p1/p142/m3makefile, p1/p143/m3makefile, p1/p144/m3makefile, p1/p145/m3makefile, p1/p146/m3makefile, p1/p147/m3makefile, p1/p148/m3makefile, p1/p149/m3makefile, p1/p150/m3makefile, p1/p151/m3makefile, p1/p152/m3makefile, p1/p153/m3makefile, p1/p154/m3makefile, p1/p155/m3makefile, p1/p156/m3makefile, p1/p157/m3makefile, p1/p158/m3makefile, p1/p159/m3makefile, p1/p160/m3makefile, p1/p161/m3makefile, p1/p162/m3makefile, p1/p163/m3makefile, p1/p164/m3makefile, p1/p165/m3makefile, p1/p166/m3makefile, p1/p167/m3makefile, p1/p168/m3makefile, p1/p169/m3makefile, p1/p170/m3makefile, p1/p171/m3makefile, p1/p172/m3makefile, p1/p173/m3makefile, p1/p174/m3makefile, p1/p175/m3makefile, p1/p176/m3makefile, p1/p177/m3makefile, p1/p178/m3makefile, p1/p179/m3makefile, p1/p180/m3makefile, p1/p181/m3makefile, p1/p182/m3makefile, p1/p183/m3makefile, p1/p184/m3makefile, p1/p185/m3makefile, p1/p186/m3makefile, p1/p187/m3makefile, p1/p188/m3makefile, p1/p189/m3makefile, p1/p190/m3makefile, p1/p191/m3makefile, p1/p192/m3makefile, p1/p193/m3makefile, p1/p194/m3makefile, p1/p195/m3makefile, p1/p196/m3makefile, p1/p197/m3makefile, p1/p198/m3makefile, p1/p199/m3makefile, p2/p200/m3makefile, p2/p201/m3makefile, p2/p202/m3makefile, p2/p203/m3makefile, p2/p204/m3makefile, p2/p205/m3makefile, p2/p206/m3makefile, p2/p207/m3makefile, p2/p208/m3makefile, r0/r001/m3makefile, r0/r002/m3makefile, r0/r003/m3makefile, r0/r004/m3makefile, x0/x001/m3makefile, x0/x002/m3makefile, x0/x003/m3makefile, x0/x004/m3makefile, x0/x005/m3makefile, x0/x006/m3makefile, x0/x007/m3makefile: no more annoying copy/linking of files or wierdo -FA switch to cm3 now just cd into directories and run cm3 results in more portable code and very easy to build and test individual tests just cd to their directory and run cm3, nice (after first having built the library and control-c) checkin instead of generate the essentially constant file, moving conditionals regardings it generation to be conditionals within it 2008-05-02 07:33 jkrell * m3-sys/m3tests/src/m3makefile: fully standalone test .exes seem like overkill, just make test.lib standalone instead and still dynamically link to m3core.dll and m3.dll; I assume that will work (it works for me, assuming wrt Olaf and Tinderbox) 2008-05-02