From jkrell at elego.de Fri Nov 2 22:33:14 2012 From: jkrell at elego.de (Jay Krell) Date: Fri, 2 Nov 2012 22:33:14 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121102213314.9D968CC9C0@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/02 22:33:14 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: fix, and work in progress From jkrell at elego.de Sat Nov 10 09:19:06 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:19:06 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110081906.3E36BCC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:19:06 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: work in progress migrating from stack of text to stack of expressions We need a higher level intermediate form in order to: nice: reduce casts more importantly: fix warnings, esp. around range checks against constants; and might as well fold constants it compiles, but probably doesn't work, temporarily M3CG.TypeUID now that Tony fixed it to remove packing requirement Use "M3CG." explicitly everywhere, so I might have a different local meaning of "Type"..or not, I'll probably use Type_t From jkrell at elego.de Sat Nov 10 09:20:16 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:20:16 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110082016.A8FC9CC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:20:16 Modified files: cm3/m3-sys/m3middle/src/: M3CG_MultiPass.i3 Log message: use M3CG.TypeUID now that Tony removed the bogus packing requirement From jkrell at elego.de Sat Nov 10 09:39:56 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:39:56 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110083956.573FACC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:39:56 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: better approach..more to do.. From jkrell at elego.de Sat Nov 10 09:43:17 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:43:17 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110084317.4FBF4CC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:43:17 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: work in progress From jkrell at elego.de Sat Nov 10 18:48:25 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 18:48:25 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110174826.20A6BCC9E3@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 18:48:25 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: safekeeping -- reuse multipass.op_t? No, stack vs. operands..doesn't work out.. From jkrell at elego.de Sat Nov 10 18:50:41 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 18:50:41 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110175041.12019CC9E3@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 18:50:41 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: go back, and progress just a bit From jkrell at elego.de Thu Nov 29 13:50:10 2012 From: jkrell at elego.de (Jay Krell) Date: Thu, 29 Nov 2012 13:50:10 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121129125010.7B4762474003@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/29 13:50:10 Modified files: cm3/m3-sys/m3cc/gcc/gcc/: Makefile.in passes.c ./: Makefile.in passes.c cm3/m3-sys/m3cc/gcc-4.5/gcc/: Makefile.in passes.c ./: Makefile.in passes.c cm3/m3-sys/m3cc/gcc-4.7/gcc/: Makefile.in passes.c ./: Makefile.in passes.c Removed files: cm3/m3-sys/m3cc/gcc/gcc/: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c ./: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c cm3/m3-sys/m3cc/gcc-4.5/gcc/: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c ./: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c cm3/m3-sys/m3cc/gcc-4.7/gcc/: matrix-reorg.c ./: matrix-reorg.c cm3/m3-sys/m3cc/gcc-4.6/gcc/: ipa-struct-reorg.h Log message: http://www.gnu.org/software/gcc/gcc-4.8/changes.html The struct reorg and matrix reorg optimizations (command-line options -fipa-struct-reorg and -fipa-matrix-reorg) have been removed. They did not work correctly nor with link-time optimization (LTO), hence were only applicable to programs consisting of a single translation unit. so remove them in our copies of older versions too. Always looking for an excuse to remove gnarly code that probably isn't used..that the owners removed it in newer versions is a good reason.. From jkrell at elego.de Thu Nov 29 14:36:15 2012 From: jkrell at elego.de (Jay Krell) Date: Thu, 29 Nov 2012 14:36:15 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121129133615.EB46A2474003@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/29 14:36:15 Modified files: cm3/m3-sys/cminstall/src/config-no-install/: cm3cfg.common Log message: use m3-sys/m3cc/I386_DARWIN to cross to AMD64_DARWIN, if it exists, and m3-sys/m3cc/I386_DARWIN-AMD64_DARWIN does not, and vice versa (actually, also should be willing to use *-I386_DARWIN to target AMD64_DARWIN, and vice versa, that is different, that is about chosing a cross compiler that has multiple targets, vs. this which is the native compiler, which has multiple targets) reuses the pattern of InternalTargetCanRunOnHost and establishes equivalency of many hosts/targets Note that the config file elsewhere says to use "-m64" so that cm3cg's defaulting to I386_DARWIN or AMD64_DARWIN is overriden. This is somewhat esoteric stuff (cross builds), and might not be fully correct, but probably is. From jkrell at elego.de Fri Nov 2 22:33:14 2012 From: jkrell at elego.de (Jay Krell) Date: Fri, 2 Nov 2012 22:33:14 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121102213314.9D968CC9C0@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/02 22:33:14 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: fix, and work in progress From jkrell at elego.de Sat Nov 10 09:19:06 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:19:06 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110081906.3E36BCC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:19:06 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: work in progress migrating from stack of text to stack of expressions We need a higher level intermediate form in order to: nice: reduce casts more importantly: fix warnings, esp. around range checks against constants; and might as well fold constants it compiles, but probably doesn't work, temporarily M3CG.TypeUID now that Tony fixed it to remove packing requirement Use "M3CG." explicitly everywhere, so I might have a different local meaning of "Type"..or not, I'll probably use Type_t From jkrell at elego.de Sat Nov 10 09:20:16 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:20:16 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110082016.A8FC9CC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:20:16 Modified files: cm3/m3-sys/m3middle/src/: M3CG_MultiPass.i3 Log message: use M3CG.TypeUID now that Tony removed the bogus packing requirement From jkrell at elego.de Sat Nov 10 09:39:56 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:39:56 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110083956.573FACC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:39:56 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: better approach..more to do.. From jkrell at elego.de Sat Nov 10 09:43:17 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:43:17 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110084317.4FBF4CC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:43:17 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: work in progress From jkrell at elego.de Sat Nov 10 18:48:25 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 18:48:25 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110174826.20A6BCC9E3@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 18:48:25 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: safekeeping -- reuse multipass.op_t? No, stack vs. operands..doesn't work out.. From jkrell at elego.de Sat Nov 10 18:50:41 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 18:50:41 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110175041.12019CC9E3@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 18:50:41 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: go back, and progress just a bit From jkrell at elego.de Thu Nov 29 13:50:10 2012 From: jkrell at elego.de (Jay Krell) Date: Thu, 29 Nov 2012 13:50:10 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121129125010.7B4762474003@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/29 13:50:10 Modified files: cm3/m3-sys/m3cc/gcc/gcc/: Makefile.in passes.c ./: Makefile.in passes.c cm3/m3-sys/m3cc/gcc-4.5/gcc/: Makefile.in passes.c ./: Makefile.in passes.c cm3/m3-sys/m3cc/gcc-4.7/gcc/: Makefile.in passes.c ./: Makefile.in passes.c Removed files: cm3/m3-sys/m3cc/gcc/gcc/: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c ./: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c cm3/m3-sys/m3cc/gcc-4.5/gcc/: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c ./: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c cm3/m3-sys/m3cc/gcc-4.7/gcc/: matrix-reorg.c ./: matrix-reorg.c cm3/m3-sys/m3cc/gcc-4.6/gcc/: ipa-struct-reorg.h Log message: http://www.gnu.org/software/gcc/gcc-4.8/changes.html The struct reorg and matrix reorg optimizations (command-line options -fipa-struct-reorg and -fipa-matrix-reorg) have been removed. They did not work correctly nor with link-time optimization (LTO), hence were only applicable to programs consisting of a single translation unit. so remove them in our copies of older versions too. Always looking for an excuse to remove gnarly code that probably isn't used..that the owners removed it in newer versions is a good reason.. From jkrell at elego.de Thu Nov 29 14:36:15 2012 From: jkrell at elego.de (Jay Krell) Date: Thu, 29 Nov 2012 14:36:15 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121129133615.EB46A2474003@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/29 14:36:15 Modified files: cm3/m3-sys/cminstall/src/config-no-install/: cm3cfg.common Log message: use m3-sys/m3cc/I386_DARWIN to cross to AMD64_DARWIN, if it exists, and m3-sys/m3cc/I386_DARWIN-AMD64_DARWIN does not, and vice versa (actually, also should be willing to use *-I386_DARWIN to target AMD64_DARWIN, and vice versa, that is different, that is about chosing a cross compiler that has multiple targets, vs. this which is the native compiler, which has multiple targets) reuses the pattern of InternalTargetCanRunOnHost and establishes equivalency of many hosts/targets Note that the config file elsewhere says to use "-m64" so that cm3cg's defaulting to I386_DARWIN or AMD64_DARWIN is overriden. This is somewhat esoteric stuff (cross builds), and might not be fully correct, but probably is. From jkrell at elego.de Fri Nov 2 22:33:14 2012 From: jkrell at elego.de (Jay Krell) Date: Fri, 2 Nov 2012 22:33:14 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121102213314.9D968CC9C0@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/02 22:33:14 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: fix, and work in progress From jkrell at elego.de Sat Nov 10 09:19:06 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:19:06 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110081906.3E36BCC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:19:06 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: work in progress migrating from stack of text to stack of expressions We need a higher level intermediate form in order to: nice: reduce casts more importantly: fix warnings, esp. around range checks against constants; and might as well fold constants it compiles, but probably doesn't work, temporarily M3CG.TypeUID now that Tony fixed it to remove packing requirement Use "M3CG." explicitly everywhere, so I might have a different local meaning of "Type"..or not, I'll probably use Type_t From jkrell at elego.de Sat Nov 10 09:20:16 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:20:16 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110082016.A8FC9CC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:20:16 Modified files: cm3/m3-sys/m3middle/src/: M3CG_MultiPass.i3 Log message: use M3CG.TypeUID now that Tony removed the bogus packing requirement From jkrell at elego.de Sat Nov 10 09:39:56 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:39:56 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110083956.573FACC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:39:56 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: better approach..more to do.. From jkrell at elego.de Sat Nov 10 09:43:17 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 9:43:17 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110084317.4FBF4CC9DB@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 09:43:17 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: work in progress From jkrell at elego.de Sat Nov 10 18:48:25 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 18:48:25 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110174826.20A6BCC9E3@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 18:48:25 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: safekeeping -- reuse multipass.op_t? No, stack vs. operands..doesn't work out.. From jkrell at elego.de Sat Nov 10 18:50:41 2012 From: jkrell at elego.de (Jay Krell) Date: Sat, 10 Nov 2012 18:50:41 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121110175041.12019CC9E3@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/10 18:50:41 Modified files: cm3/m3-sys/m3back/src/: M3C.m3 Log message: go back, and progress just a bit From jkrell at elego.de Thu Nov 29 13:50:10 2012 From: jkrell at elego.de (Jay Krell) Date: Thu, 29 Nov 2012 13:50:10 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121129125010.7B4762474003@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/29 13:50:10 Modified files: cm3/m3-sys/m3cc/gcc/gcc/: Makefile.in passes.c ./: Makefile.in passes.c cm3/m3-sys/m3cc/gcc-4.5/gcc/: Makefile.in passes.c ./: Makefile.in passes.c cm3/m3-sys/m3cc/gcc-4.7/gcc/: Makefile.in passes.c ./: Makefile.in passes.c Removed files: cm3/m3-sys/m3cc/gcc/gcc/: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c ./: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c cm3/m3-sys/m3cc/gcc-4.5/gcc/: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c ./: ipa-struct-reorg.c ipa-struct-reorg.h matrix-reorg.c cm3/m3-sys/m3cc/gcc-4.7/gcc/: matrix-reorg.c ./: matrix-reorg.c cm3/m3-sys/m3cc/gcc-4.6/gcc/: ipa-struct-reorg.h Log message: http://www.gnu.org/software/gcc/gcc-4.8/changes.html The struct reorg and matrix reorg optimizations (command-line options -fipa-struct-reorg and -fipa-matrix-reorg) have been removed. They did not work correctly nor with link-time optimization (LTO), hence were only applicable to programs consisting of a single translation unit. so remove them in our copies of older versions too. Always looking for an excuse to remove gnarly code that probably isn't used..that the owners removed it in newer versions is a good reason.. From jkrell at elego.de Thu Nov 29 14:36:15 2012 From: jkrell at elego.de (Jay Krell) Date: Thu, 29 Nov 2012 14:36:15 () Subject: [M3commit] CVS Update: cm3 Message-ID: <20121129133615.EB46A2474003@birch.elegosoft.com> CVSROOT: /usr/cvs Changes by: jkrell at birch. 12/11/29 14:36:15 Modified files: cm3/m3-sys/cminstall/src/config-no-install/: cm3cfg.common Log message: use m3-sys/m3cc/I386_DARWIN to cross to AMD64_DARWIN, if it exists, and m3-sys/m3cc/I386_DARWIN-AMD64_DARWIN does not, and vice versa (actually, also should be willing to use *-I386_DARWIN to target AMD64_DARWIN, and vice versa, that is different, that is about chosing a cross compiler that has multiple targets, vs. this which is the native compiler, which has multiple targets) reuses the pattern of InternalTargetCanRunOnHost and establishes equivalency of many hosts/targets Note that the config file elsewhere says to use "-m64" so that cm3cg's defaulting to I386_DARWIN or AMD64_DARWIN is overriden. This is somewhat esoteric stuff (cross builds), and might not be fully correct, but probably is.