Newsgroups: comp.archives Path: utzoo!utgpu!news-server.csri.toronto.edu!ox.com!msen.com!emv From: delorie@ctron.COM Subject: [gnu-msdos] patch #2 for G++/386/DOS (djgpp) Message-ID: <1991May15.042916.5498@ox.com> Followup-To: list.gnu-msdos Sender: emv@msen.com (Edward Vielmetti, MSEN) Reply-To: info-gnu-msdos@wugate.wustl.edu Organization: (none) Date: Wed, 15 May 1991 04:29:16 GMT Approved: emv@msen.com (Edward Vielmetti, MSEN) X-Original-Newsgroups: list.gnu-msdos Archive-name: gnu/g++/djgpp/0-- Archive-directory: grape.ecs.clarkson.edu:/pub/msdos/djgpp/ [128.153.28.129] Original-posting-by: delorie@ctron.COM Original-subject: patch #2 for G++/386/DOS (djgpp) Reposted-by: emv@msen.com (Edward Vielmetti, MSEN) Patch #2 has been uploaded to: grape.ecs.clarkson.edu 128.153.28.129 FTP pub/msdos/djgpp For convenience to BITFTP users, I have uploaded everything EXCEPT djgpp.zip to: sun.soe.clarkson.edu 128.153.12.3 BITFTP pub/msdos/djgpp Sun.soe.clarkson.edu has an archive server that can mail you the files. To get help on the archive server, send mail to archive-server@sun.soe.clarkson.edu, with the word "help" as the message (not subject). The archive name is msdos/djgpp. --- DJGPP - G++ for MSDOS/386 --- System requirements: 80386, 512K memory Supports: 80387, Extended memory (up to 128M), XMS, VGA, SuperVGA, paging to disk (up to 128M). total 2483 -rw-r--r-- 1 dj 2134610 May 10 09:41 djgpp.zip -rw-r--r-- 1 dj 21623 May 10 08:41 readme -rw-r--r-- 1 dj 336422 May 10 08:41 patch2.zip -rw-r--r-- 1 dj 2144 May 10 08:41 patches2.doc -rw-r--r-- 1 dj 976 May 10 08:41 merge.c -rw-r--r-- 1 dj 6924 May 10 08:41 merge.exe -rw-r--r-- 1 dj 1244 May 10 08:41 split.c -rw-r--r-- 1 dj 7474 May 10 08:41 split.exe -rw-r--r-- 1 dj 32768 May 10 08:41 djgppzip.000 through djgppzip.065 ("merge djgppzip djgpp.zip" builds the big .zip from these) A copy of patches2.doc is attached. Please do not ask me to make this available anywhere other than grape; I've tried before and never had any luck. If you'd like it available elsewhere, contact that site's administrator and ask them to transfer it. DJ dj@ctron.com ================================ patches2.doc ================================ ** Patch #2 for djgpp ** Summary: * Trident drivers updated * More info on GNU GPL restrictions * Updated libgr documentation * Replaced GNU malloc with BSD malloc * Fixed cosmetic bugs in debug32 * Patch program for cc1plus to compile inline methods * Experimental 80387 emulator NOTE: The replacement of malloc in libc.a is to allow developers to compile most C and C++ programs without linking in GNU code. If you link in GNU code, your work must be distributed under the GNU GPL. The C++ classes in libc.a, and obstacks, are GNU code. See the source files for official copyright information. DRIVERS/TRIDENT.ASM DRIVERS/TRIDENT.GRD DRIVERS/TRIDNT89.ASM DRIVERS/TRIDNT89.GRD Fixed to set "new" mode properly. NOTE: The Trident chips can't do reads properly, so you can't to blits, XOR colors, or use point(). These drivers are designed for writing to the screen ONLY. If you don't like this, call Trident and complain. DOCS/README Added more information about conforming to the GNU General Public License. DOCS/LIBGR.DOC Described the GR_default_graphics and GR_default_text modes better. DOCS/VERSION Updated to 1.02 GO32/NPX.H Values larger than 2 show up properly in the 'npx' command of debug32 GO32/UNASSMBL.C The 'bsr' instruction appears properly BIN/GO32.EXE BIN/DEBUG32.EXE LIBSRC/C/LIB/MALLOC.C Replaced GNU's malloc with BSD's malloc. DIFFS/GPP2.DIF Replace /dev/null with nul UTILS/NULFIX.C UTILS/NULFIX.EXE Utility to change /dev/null to nul in cc1plus. Run this in the bin directory - it's automatic. This will allow you to compile code like this: class A { int m() { return 2; } }; LIB/LIBC.A Version with BSD's malloc GO32/CONTROL.C GO32/PAGING.C GO32/PAGING.H EMU387/*.* TEST version of the 80387 emulator. To use, append "emu /djgpp/emu387/emu387" (substitute proper path) to the GO32 environment variable. I don't guarantee this will work; I'm still working on it myself. It's not done but should handle most common floating point math (+ - * / sqrt).