From wagner at elego.de Wed Sep 3 23:05:02 2008 From: wagner at elego.de (Olaf Wagner) Date: Wed, 3 Sep 2008 23:05:02 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080903210502.CE4F310D4FD7@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: wagner at birch. 08/09/03 23:05:02 Modified files: cm3/m3-sys/cm3ide/src/: m3overrides Log message: add missing override for tcp From jkrell at elego.de Mon Sep 15 11:15:30 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 11:15:30 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915091530.37BA910D5298@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 11:15:30 Modified files: cm3/m3-sys/m3cc/src/: m3makefile Log message: 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 From jkrell at elego.de Mon Sep 15 16:17:35 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 16:17:35 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915141735.B5B8210D52A8@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 16:17:35 Modified files: cm3/scripts/python/: make-dist.py Log message: don't delete cm3cg here From jkrell at elego.de Mon Sep 15 17:52:09 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 17:52:09 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915155209.4177410D52AA@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 17:52:09 Modified files: cm3/scripts/python/: make-dist.py Log message: don't delete cm3cg here either From jkrell at elego.de Mon Sep 15 18:21:27 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 18:21:27 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915162127.9DACF10D52A4@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 18:21:27 Modified files: cm3/scripts/python/: make-dist.py Log message: don't delete cm3 here From rodney at elego.de Fri Sep 26 01:22:11 2008 From: rodney at elego.de (Rodney M. Bates) Date: Fri, 26 Sep 2008 1:22:11 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080925232211.920A410D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: rodney at birch. 08/09/26 01:22:11 Modified files: cm3/m3-sys/m3cc/gcc/gcc/: dbxout.c tree-nested.c Log message: 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. From rodney at elego.de Fri Sep 26 01:23:20 2008 From: rodney at elego.de (Rodney M. Bates) Date: Fri, 26 Sep 2008 1:23:20 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080925232320.A8EDC10D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: rodney at birch. 08/09/26 01:23:20 Modified files: cm3/m3-sys/m3gdb/gdb/gdb/: m3-eval.c m3-util.c m3-valprint.c Log message: 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. From jkrell at elego.de Tue Sep 30 15:54:25 2008 From: jkrell at elego.de (Jay Krell) Date: Tue, 30 Sep 2008 15:54:25 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080930135425.AC43610D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/30 15:54:25 Modified files: cm3/m3-sys/cminstall/src/config-no-install/: LINUXLIBC6 PPC_LINUX Log message: fix bootstrapping LINUX_PPC from 5.2.6 by using shared code; have Linux/x86 use ar -s instead of separate invocation of ranlib From jkrell at elego.de Tue Sep 30 15:55:51 2008 From: jkrell at elego.de (Jay Krell) Date: Tue, 30 Sep 2008 15:55:51 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080930135551.8E6BD10D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/30 15:55:51 Modified files: cm3/m3-sys/cminstall/src/config-no-install/: LINUXLIBC6 Log message: 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 From jkrell at elego.de Tue Sep 30 16:11:58 2008 From: jkrell at elego.de (Jay Krell) Date: Tue, 30 Sep 2008 16:11:58 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080930141158.6144410D5423@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/30 16:11:58 Modified files: cm3/m3-sys/m3cc/src/: m3makefile Log message: use M3CC_HOST instead of TARGET for the first part of a cross directory name; usually these are the same From wagner at elego.de Wed Sep 3 23:05:02 2008 From: wagner at elego.de (Olaf Wagner) Date: Wed, 3 Sep 2008 23:05:02 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080903210502.CE4F310D4FD7@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: wagner at birch. 08/09/03 23:05:02 Modified files: cm3/m3-sys/cm3ide/src/: m3overrides Log message: add missing override for tcp From jkrell at elego.de Mon Sep 15 11:15:30 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 11:15:30 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915091530.37BA910D5298@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 11:15:30 Modified files: cm3/m3-sys/m3cc/src/: m3makefile Log message: 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 From jkrell at elego.de Mon Sep 15 16:17:35 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 16:17:35 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915141735.B5B8210D52A8@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 16:17:35 Modified files: cm3/scripts/python/: make-dist.py Log message: don't delete cm3cg here From jkrell at elego.de Mon Sep 15 17:52:09 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 17:52:09 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915155209.4177410D52AA@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 17:52:09 Modified files: cm3/scripts/python/: make-dist.py Log message: don't delete cm3cg here either From jkrell at elego.de Mon Sep 15 18:21:27 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 18:21:27 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915162127.9DACF10D52A4@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 18:21:27 Modified files: cm3/scripts/python/: make-dist.py Log message: don't delete cm3 here From rodney at elego.de Fri Sep 26 01:22:11 2008 From: rodney at elego.de (Rodney M. Bates) Date: Fri, 26 Sep 2008 1:22:11 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080925232211.920A410D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: rodney at birch. 08/09/26 01:22:11 Modified files: cm3/m3-sys/m3cc/gcc/gcc/: dbxout.c tree-nested.c Log message: 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. From rodney at elego.de Fri Sep 26 01:23:20 2008 From: rodney at elego.de (Rodney M. Bates) Date: Fri, 26 Sep 2008 1:23:20 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080925232320.A8EDC10D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: rodney at birch. 08/09/26 01:23:20 Modified files: cm3/m3-sys/m3gdb/gdb/gdb/: m3-eval.c m3-util.c m3-valprint.c Log message: 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. From jkrell at elego.de Tue Sep 30 15:54:25 2008 From: jkrell at elego.de (Jay Krell) Date: Tue, 30 Sep 2008 15:54:25 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080930135425.AC43610D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/30 15:54:25 Modified files: cm3/m3-sys/cminstall/src/config-no-install/: LINUXLIBC6 PPC_LINUX Log message: fix bootstrapping LINUX_PPC from 5.2.6 by using shared code; have Linux/x86 use ar -s instead of separate invocation of ranlib From jkrell at elego.de Tue Sep 30 15:55:51 2008 From: jkrell at elego.de (Jay Krell) Date: Tue, 30 Sep 2008 15:55:51 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080930135551.8E6BD10D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/30 15:55:51 Modified files: cm3/m3-sys/cminstall/src/config-no-install/: LINUXLIBC6 Log message: 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 From jkrell at elego.de Tue Sep 30 16:11:58 2008 From: jkrell at elego.de (Jay Krell) Date: Tue, 30 Sep 2008 16:11:58 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080930141158.6144410D5423@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/30 16:11:58 Modified files: cm3/m3-sys/m3cc/src/: m3makefile Log message: use M3CC_HOST instead of TARGET for the first part of a cross directory name; usually these are the same From wagner at elego.de Wed Sep 3 23:05:02 2008 From: wagner at elego.de (Olaf Wagner) Date: Wed, 3 Sep 2008 23:05:02 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080903210502.CE4F310D4FD7@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: wagner at birch. 08/09/03 23:05:02 Modified files: cm3/m3-sys/cm3ide/src/: m3overrides Log message: add missing override for tcp From jkrell at elego.de Mon Sep 15 11:15:30 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 11:15:30 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915091530.37BA910D5298@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 11:15:30 Modified files: cm3/m3-sys/m3cc/src/: m3makefile Log message: 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 From jkrell at elego.de Mon Sep 15 16:17:35 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 16:17:35 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915141735.B5B8210D52A8@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 16:17:35 Modified files: cm3/scripts/python/: make-dist.py Log message: don't delete cm3cg here From jkrell at elego.de Mon Sep 15 17:52:09 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 17:52:09 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915155209.4177410D52AA@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 17:52:09 Modified files: cm3/scripts/python/: make-dist.py Log message: don't delete cm3cg here either From jkrell at elego.de Mon Sep 15 18:21:27 2008 From: jkrell at elego.de (Jay Krell) Date: Mon, 15 Sep 2008 18:21:27 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080915162127.9DACF10D52A4@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/15 18:21:27 Modified files: cm3/scripts/python/: make-dist.py Log message: don't delete cm3 here From rodney at elego.de Fri Sep 26 01:22:11 2008 From: rodney at elego.de (Rodney M. Bates) Date: Fri, 26 Sep 2008 1:22:11 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080925232211.920A410D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: rodney at birch. 08/09/26 01:22:11 Modified files: cm3/m3-sys/m3cc/gcc/gcc/: dbxout.c tree-nested.c Log message: 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. From rodney at elego.de Fri Sep 26 01:23:20 2008 From: rodney at elego.de (Rodney M. Bates) Date: Fri, 26 Sep 2008 1:23:20 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080925232320.A8EDC10D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: rodney at birch. 08/09/26 01:23:20 Modified files: cm3/m3-sys/m3gdb/gdb/gdb/: m3-eval.c m3-util.c m3-valprint.c Log message: 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. From jkrell at elego.de Tue Sep 30 15:54:25 2008 From: jkrell at elego.de (Jay Krell) Date: Tue, 30 Sep 2008 15:54:25 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080930135425.AC43610D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/30 15:54:25 Modified files: cm3/m3-sys/cminstall/src/config-no-install/: LINUXLIBC6 PPC_LINUX Log message: fix bootstrapping LINUX_PPC from 5.2.6 by using shared code; have Linux/x86 use ar -s instead of separate invocation of ranlib From jkrell at elego.de Tue Sep 30 15:55:51 2008 From: jkrell at elego.de (Jay Krell) Date: Tue, 30 Sep 2008 15:55:51 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080930135551.8E6BD10D539B@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/30 15:55:51 Modified files: cm3/m3-sys/cminstall/src/config-no-install/: LINUXLIBC6 Log message: 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 From jkrell at elego.de Tue Sep 30 16:11:58 2008 From: jkrell at elego.de (Jay Krell) Date: Tue, 30 Sep 2008 16:11:58 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20080930141158.6144410D5423@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 08/09/30 16:11:58 Modified files: cm3/m3-sys/m3cc/src/: m3makefile Log message: use M3CC_HOST instead of TARGET for the first part of a cross directory name; usually these are the same