[Freedos-devel] French in FreeDOS-1.3 (the problem of accented letters in package descriptions)

Was thinking of writing this for a few days, but did not came to do it. First, congrats for the 25 years of the project! I have seen the Lunduke video about it, and this convinced me to come back, see if I can contribute a bit for 1.3 release. Sadly, my contribution is much about telling

Re: [Freedos-devel] Proposition for what a DOS package should be

Well, honestly, while I was writing the first messages about packages, I did not read yet: http://wiki.freedos.org/wiki/index.php/Package but I was more thinking about what I would expect, and I still believe there is good stuff in what I propose that is missing in most packaging solutions (for

Re: [Freedos-devel] Proposition for what a DOS package should be

Beside, this is not just a a matter of trust. This is code janitoring. By recompiling programs, we make sure that the programs needed for them are still available. We can make small changes if possible, to make the programs built from more open implementations of C or assembler. Briefly,

Re: [Freedos-devel] Proposition for what a DOS package should be

Thanks Tom Ehlert, to confirm my feeling that most DOS packages have not been rebuilt from source code. We used to trust people because we knew no better way in the past. Well, I was not expecting in any way that the DOS kernel would in any way verify binaries. And I am/was finally aware

[Freedos-devel] Re : Proposition for what a DOS package should be

Well, thinking a bit more... often "packages" will be on web sites. And the user might well not care about let's say source code, even if builder wish to include it in the package. Why force the user to download source code inside the package if he/she does not want it? So I would propose

[Freedos-devel] Proposition for what a DOS package should be

I am thinking about how software is/should be distributed. I believe, source code is for programmers. I believe, executable code is for users. I believe a intermediary between the two should exist. I guess it is a package. I believe this intermediary is for installer(s). I believe this

[Freedos-devel] Re : Proposition for what a DOS package should be

This is clearly work in progress. Publishing for comments, even if rather incomplete. I am realizing I am ignoring the project current package format(s). I have the feeling the package format for 1.3 is not documented yet. Although one could look at it inside the iso image. I believe we

Re: [Freedos-devel] Re : Re : Could the installer use gcdrom to get SATA cdrom working without IDE emulation?

I should have describe a little bit more why I was suggesting it might be necessary to include AHCI driver in DOS kernel. That's because on a AHCI only computer (without PATA or SATA IDE),  you cannot access the sectors with INT 13h. Well, I have tried in Qemu with '-M Q35' machine

Re: [Freedos-devel] Re : Re : Could the installer use gcdrom to get SATA cdrom working without IDE emulation?

Le lun., 02 nov. 2020 15:32:43 -0500 Eric Auer écrit Hi Paul, > That's because on a AHCI only computer (without PATA or SATA IDE), >  you cannot access the sectors with INT 13h. Yes you can. For harddisk/SSD. I am glad to report that I was wrong. Indeed INT 13h continue to

Re: [Freedos-devel] Obsolete DJGPP in upcomming FreeDos 1.3

Le mer., 28 oct. 2020 16:13:13 -0400 Jerome Shidel écrit There are about 12 DJGPP packages mirrored on ibiblio and in the software repository. I would assume they all are outdated. It would be nice if someone provided updates for these packages. However,I

[Freedos-devel] began to list problems I see with FD 1.3-rc3

P120: 'keyb cf' works, but there is no French (Canada) in the keyboard list of the installer P121: I have chosen 'French' as language, French(alternative?) as keyboard, install essential... when rebooting the system I get 'Bienvenue au systemeParameter Error' followed by working C: prompt

[Freedos-devel] 1.3-rc3: should give meaningfull error message on AHCI (SATA) only system

If I had: '-M q35' to qemu to enable a SATA only machine, when booting in live mode 1.3-rc3: UDVD2: Nothing to use; UDVD2 not loaded. I hope UDVD return an error code we could use to give a more meaninful message like: Your computer is running in SATA only mode, so I cannot access the CD-ROM.

[Freedos-devel] I am sad less seems to be an alias on more

I wish | less would allows PageUp to go back in the text!___ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-devel

Re: [Freedos-devel] I am sad less seems to be an alias on more

Le ven., 30 oct. 2020 12:22:45 -0400 Jim Hall écrit >However, you can use u to go up by half a page. Or b to go back by a full page. Oh thanks! I kind like never learn those keys. But a newer version in 1.3 would be

[Freedos-devel] PCI BIOS

Well... I just would like to share info about PCI-BIOS, that I recently learn about. This is interrupt calls for using PCI devices. Can see them at: http://www.delorie.com/djgpp/doc/rbinter/ix/1A/B1.html I don't know how much they are implemented in real life... but I have seen MACH

[Freedos-devel] Re : PCI BIOS

Well, I was able to verify the presence of PCI-BIOS easily with debug (under Qemu): -a 100 mov ax,b101 sub di,di int 1a int 20 g=100 107 AH=00 PCIBIOS present AL=01 configuration space access mechanism 1 supported BX=0210 version 2.1 supported EDX = 20494350h (' ICP') (well I only see 4350

[Freedos-devel] Smaller C

Well, after having a problem with IA-16 gcc (which seems to requires DJGPP gcc installed and working), I have search for an other free C compiler for FreeDOS. I have found Smaller C: https://github.com/alexfru/SmallerC/blob/master/v0100/doc/smlrc.md I note: Smaller C is not an optimizing

Re: [Freedos-devel] Smaller C

I have just scan all the .exe from /v0100/bind/ directory (16 bits version) and /v0100/bindp/ directory (32 bits DPMI) and none were suspicious. Was also able to build test.exe with 32 bits DPMI version. I guess would be better to package 32 bits DPMI version because we have cwsdpmi by

[Freedos-devel] Re : Open Watcom: Install, setup, compile and link your 'first' program

Ok, since my previous message I discover: wcl and wcl386: wcl:  C/C++ compile and link utility for 16-bit targets wcl386:  C/C++ compile and link utility for 32-bit targets taken from: https://flaterco.com/kb/ow.html So rather than compile, then link in previous post, you can just do: wcl

[Freedos-devel] Open Watcom: Install, setup, compile and link your 'first' program

Hi! It is said on: https://www.freedos.org/contribute/ that: "Port FreeDOS utilities to OpenWatcom C and NASM—our preferred C compiler and Assembler for FreeDOS." I kind of just discovered the OpenWatcom C,C++ compiler yesterday, and just was able to compile and link my first program this

Re: [Freedos-devel] began to list problems I see with FD 1.3-rc3

Le sam., 31 oct. 2020 06:26:59 -0400 Jerome Shidel écrit Hi Paul, Have you listed the issues you have found on the FreeDOS "feature requests” and “bug tracker” lists on http://SourceForge.net? https://sourceforge.net/p/freedos/feature-requests/

[Freedos-devel] Priority of ticket, 1 should be highest priority

I was misinterpreting the priority of tickets on SourceForge. I thought 9 was the highest, but it seems to be the lowest, according to: https://sourceforge.net/p/forge/site-support/18975/#cc07___ Freedos-devel mailing list

[Freedos-devel] Re : Could the installer use gcdrom to get SATA cdrom working without IDE emulation?

I said: "but why don't we use gcdrom rather than UDVD2?" Obviously because a lot of DOS computers does not have a SATA "CDROM" (probably more DVD disk). Ideally I would say try gcdrom, if it does not work, use old UDVD2. But, to be more prudent I propose: try UDVD2, if it does not work,

[Freedos-devel] To boot FreeDOS from GPT partition table...

It seems not so hard in theory: Define a GUID for FreeDOS. Create a BIOS Boot partition. Write at the beginning of BIOS Boot partition a special "MBR" that search FreeDOS GUID partition in the GPT, load it and jump to it.___ Freedos-devel

[Freedos-devel] Could the installer use gcdrom to get SATA cdrom working without IDE emulation?

Hey! We already have a AHCI driver (SATA)!? I see it at: http://www.ibiblio.org/pub/micro/pc-stuff/freedos/files/distributions/1.2/repos/pkg-html/gcdrom.html I know the installer now fail with UDVD2 for SATA... but why don't we use gcdrom rather than

[Freedos-devel] Re : Re : Could the installer use gcdrom to get SATA cdrom working without IDE emulation?

Sorry I was confused about gcdrom.sys. (Re)reading http://www.bootablecd.de/fdhelp-internet/en/hhstndrd/base/gcdrom.htm I now realize it is still an IDE driver, not an AHCI driver like I thought. Anyway, I wonder if the kernel itself would be needed to support AHCI mode without IDE

[Freedos-devel] FD13LIVE.ISO: does not install on a new (unpartitioned) disk

Hi! I have tried http://www.ibiblio.org/pub/micro/pc-stuff/freedos/files/distributions/1.3/previews/1.3-rc3/FD13-LiveCD.zip with qemu, and I have come to the conclusion that I cannot install (with install from the main menu, or with setup.exe in live mode) on a newly created disk image. I

Re: [Freedos-devel] Problems on 1.3rc4 on VirtualBox and QEMU - Paku Paku 160x100 mode

Le lun., 14 juin 2021 06:33:06 -0400 Eric Auer écrit >Does the game fill the whole window with BOTH >virtualbox and qemu for you? Because it only >fills the upper half in dosemu2 at the moment. Testing on Ubuntu 21.04. With virtualbox, it is full-windows. Going full-screen

[Freedos-devel] The confusing fdisk releases

I check about 1h ago, and concluded that there was no new releases of fdisk since 1.3RC4. I went to: https://www.ibiblio.org/pub/micro/pc-stuff/freedos/files/distributions/1.3/previews/1.3-rc4/report.html checked:

[Freedos-devel] 2 diffrent index of ibiblio?

I am a bit confused... I used to see: https://www.ibiblio.org/pub/micro/pc-stuff/freedos/files/repositories/latest/pkg-html/group-devel.html but searching ibiblio from myself found: http://www.ibiblio.org/pub/micro/pc-stuff/freedos/files/devel/?C=N;O=A The first seems to be more the

Re: [Freedos-devel] Problems on 1.3rc4 on VirtualBox

Le ven., 11 juin 2021 14:14:58 -0400 Eric Auer écrit [citating only "pertinent" info] So apparently VirtualBox 6.1.22 has a surprising amount of problems with DOS games and I wonder why that is the case. I guess all games work equally well or unwell for

Re: [Freedos-devel] Problems on 1.3rc4 on VirtualBox

Resending to the list, had responded to Tom privately (thought it had sent it to me privately) Le ven., 11 juin 2021 11:14:43 -0400 Tom Ehlert Software écrit >> French: I think I have not seen any accented characters, like all replaced >> by 2 chars

Re: [Freedos-devel] Problems on 1.3rc4 on VirtualBox

After many many hours of fighting with the installer to install in dosemu... then realized that it was dosemu1 not 2, I finally managed to have FreeDOS mostly running under Dosemu2. But when I ran a program using cwsdpmi, it complains I should run cwsdpmi. And then, cwsdpmi complains it

Re: [Freedos-devel] Problems on 1.3rc4 on VirtualBox

Fixed problem of hanging after partitioning by adding -boot d to the script, that make it always boot virtual Live CD. Changed it back to -boot c after installation to boot the installed system. fmines ok freedoom, some text scrolling fast, then illegal instruction, return to command

[Freedos-devel] a color bug in VirtualBox for VGA

I have tried to find a bug report about color not working on VGA in VirtualBox: the most relevant I see seems: https://www.virtualbox.org/ticket/4777 VESA palette support broken (Closed: Obsolete)___ Freedos-devel mailing list

Re: [Freedos-devel] Problems on 1.3rc4 on VirtualBox

Yesterday, the first Linux I installed, to test outside of NixOS was Debian 11 (same computer, I make different partitions under GPT, UEFI). I was surprised that Debian got rid of VirtualBox in Debian 10 (because they don't like how the company handle CVE for older versions), and also got rid

Re: [Freedos-devel] Problems on 1.3rc4 on QEMU

Le dim., 13 juin 2021 00:11:11 -0400 Jim Hall écrit >I hadn't tried the games yet, so verifying your bug report was my first time. Then you may get the wrong impression that many games does not work. That's not really the case, I have been playing many of

Re: [Freedos-devel] FreeDOS kernel and FreeCOM command.com updates

Le dim., 13 juin 2021 19:13:45 -0400 écrit On Sun, Jun 13, 2021, 4:44 PM tom ehlert wrote: > Unfortunately, ALL files in the kernel source download have the same > timestamp, making it impossible to see what was last changed when :-o ALL times of

Re: [Freedos-devel] Problems on 1.3rc4 on VirtualBox and QEMU

Eric asked: what have change? (game was working for me, not anymore) The previous was with under VirtualBox [that I believe does not handle color palette right] And the newer results were with QEMU. I have observed that bolitare works under QEMU if I remove the option that add sb16

Re: [Freedos-devel] Why games have bad color in VirtualBox

That said, I am at least partially wrong. Changing colors works in VirtualBox, QEMU and DOSEMU2. color 0 is black, this is the default color for text mode. In Debug: -o 3c8,0 -o 3c9,0 -o 3c9,0 -o 3c9,ff Change the black to blue. (out 3c8,0 choose color 0, then first value at 3c9 is

Re: [Freedos-devel] Why games have bad color in VirtualBox

vgapal works in VirtualBox ... and DOSEMU2, not tested under QEMU. I think problematic games run under DPMI (protected mode). Might be linked to the problem for VirtualBox not handling color palettes correctly... sometime. Will have to look the source code of the games to see how they do...

Re: [Freedos-devel] Why games have bad color in VirtualBox

I was thinking... I think DOSEMU2 execute program in VM86 mode. That way, direct IN/OUT generate an interrupt that can be intercepted to emulate hardware. It is a bit unclear how QEMU/VirtualBox works... I have read, that newer CPU emulate a full CPU. But I don't know the details... and how

Re: [Freedos-devel] Why games have bad color in VirtualBox

Ok, a few minutes ago, I had no idea about which I/O ports were related to colors... and now, I begin to suspect what could be wrong in VirtualBox code, that is it seems to ignore port 0x3c6 (mask register)... which is said: "in normal use, should be 0xff"... In:

Re: [Freedos-devel] Why games have bad color in VirtualBox

More precisely, the default DAC Mask ( i 3c6 in debug) is: QEMU: FF DOSEMU2: FF VirtualBox: 00 VirtualBox default mask of 0 and the fact that it cannot be changed, could well explains changing colors directly does not work. (as I believe a mask of 0 make it not change any colors when

Re: [Freedos-devel] Additional notes on Games

I have taken a short look at: (gnuchess source for showing board) http://git.savannah.gnu.org/cgit/chess.git/tree/src/frontend/output.cc and did not have "the feeling" it was using ANSI anymore. But then according to:

Re: [Freedos-devel] Why games have bad color in VirtualBox

DOSEMU2 seems to report itself almost exactly as VirtualBOX, that is: VBE 2.0, D1=0:VGA compatible D1=0, D0=1:DAC is fixed with 6 bit by primary color This destroy my hypothesis about games USING VGA Set palette function... rather than the VBE one (AL=9... had wrongly written AH=9 in previous

Re: [Freedos-devel] Why games have bad color in VirtualBox

I think I understand what is happening. First, an extract from: https://www.phatcode.net/res/224/files/html/ch34/34-05.html The DAC Mask "There’s one register in the DAC that I haven’t mentioned yet, the DAC Mask register at 03C6H. The operation of this register is simple but powerful; it can

[Freedos-devel] Re : Wirth compilers (was: 2 diffrent index of ibiblio?)

Thanks a lot Rugxulo for your email about Wirth compilers. Some projects, I did began to look, some I was not aware. And your opinion and info about FPC now supporting 16 bit DOS is very appreciated!___ Freedos-devel mailing list

Re: [Freedos-devel] Why games have bad color in VirtualBox

QEMU, AX=4F00 INT 10h (Return VBE controller information) reports itself as version 3.0. So I use http://www.petesqbsite.com/sections/tutorials/tuts/vbe3.pdf Capabilities: D0=0 DAC is fixed width, with 6 bits per primary color D1=0 Controller is VGA compatible D2=0Normal RAMDAC operations D3=0

[Freedos-devel] Re : Wirth compilers (was: 2 diffrent index of ibiblio?)

Looks like Modula-2 could come part of GCC (GNU compiler collection): https://www.phoronix.com/scan.php?page=news_item=Modula-2-GCC-Mainline-Hope-2021 That would be great!___ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net

[Freedos-devel] easily create a gcc cross compiler for compiling C on Linux making exe to be run on DJGPP

I'd like to mention: https://github.com/jwt27/build-gcc Have worked great for me. Made an hello world! in Linux for DJGPP.___ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-devel

[Freedos-devel] easily create a gcc cross compiler for compiling C on Linux making exe to be run on DJGPP

And when cross-compiling on Linux for running on DJGPP you probably want: ./configure --build=i686-pc-linux-gnu --host=i386-pc-msdosdjgpp when calling configure. Details at:

[Freedos-devel] problems cross-compiling gnuchess

I guess I need to mention what I did before  and why: Ok... I directly plugged http://www.delorie.com/pub/djgpp/current/v2gnu/pth207b.zip inside: /home/paul/opt/cross/lib/gcc/i386-pc-msdosdjgpp/10.2.0/ which I feels is a bit wrong... but working cowboy approach. [I now think

[Freedos-devel] Problems cross-compiling gnuchess

Frankly, this is in part a note to myselft... but could be considered a follow-up to me trying to compile gnuchess. So after "plugging" http://www.delorie.com/pub/djgpp/current/v2gnu/gpp1020b.zip and http://www.delorie.com/pub/djgpp/current/v2tk/ls080b.zip The problem is linked to the

[Freedos-devel] Re : Problems cross-compiling gnuchess

Le dim., 20 juin 2021 20:20:17 -0400 Paul Dufresne via Freedos-devel <mailto:freedos-devel@lists.sourceforge.net> écrit >In fact I have tried: >#include >//#include > >int main (){ >  return 0; >} >which does not find cstdllib.h. yes,

[Freedos-devel] kraptor in VirtualBox

I now have realized that kraptor colors were completely bad in kraptor... until I select some video mode in configuration menu (first page). It seems whatever the mode I choose, colors are fixed by choosing a video mode in the configuration (there are a lot of choice). So I begin to

Re: [Freedos-devel] allegro 4.2 branch, many problems

Le mar., 22 juin 2021 15:30:29 -0400 Ladislav Lacina écrit >The last working Allegro for DOS is the version 4.2.3 not 4.2.4 >Download the working package from here: http://www.mrdictionary.net/allegro/ >or from here: http://www.doshaven.eu/programming-tools/ Thanks! I

[Freedos-devel] Got a working gnuchess 5.05

I have been able to cross-compile gnuchess 5.05. Version 5 was released in 1999: Announce: https://www.tim-mann.org/gnuchess/5.00.txt According to: https://www.tim-mann.org/gnuchess.html version 6 was released in 2011 So that version, 5.05 does not use pthread, and is easier to

[Freedos-devel] Re : Got a working gnuchess 5.05

made a package for GNU CHESS 5.05 called GCHES, at: https://gitlab.com/dufresnep/gches/-/tree/main/GCHES1_0 Feel free to test for virus ( I did not ), and try.___ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net

Re: [Freedos-devel] Got a working gnuchess 5.05

The "board" only shows after your had enter your first move: Try "e2e4" Enter. But you have to know a bit chess notation: https://en.wikipedia.org/wiki/Algebraic_notation_(chess) K=king, Q=Queen, R=Root, B=Bishop N=kNight, P=Pawn___ Freedos-devel

Re: [Freedos-devel] Got a working gnuchess 5.05

So: 1) Show the board before asking the first move 2) Add coordinates (A-H, 1-8) when showing board. 3) Allows entering move by moving cursor... this would probably need pdcurses.h. ... will look at first two later... maybe tomorrow. I have doubts I will ever do 3) Thanks for the first

Re: [Freedos-devel] Almost compiled kraptor game

Le lun., 21 juin 2021 18:40:33 -0400 Eric Auer écrit [...] >> Explicitly set -std=gnu89 and -std=gnu++98 for DJGPP compilation >> >> Versions 5.00 and above of GCC default to -std=gnu11 instead of -std=gnu89. >> This breaks linking of Allegro 4.2 on

[Freedos-devel] Almost compiled kraptor game

kraptor game has evident problems with palette of images in VirtualBox (only) and I tought compiling it would help do some experiments to figure out what the problem is. Fun fact: the author of the game (released in 2004) have made kfixpal tool, to change a bunch of pictures, so as they

Re: [Freedos-devel] allegro 4.2 branch, many problems

I fixed the missing rule to build allegro.so by adding the file: paul@kasparno:~/Téléchargements/allegro5-4.2$ cat makefile.dep lib/unix/liballeg-4.2.4.so: $(ALLEGRO_LIBRARY_SOURCES) It is not present, but included in makefile (ignoring any errors). but next:

Re: [Freedos-devel] Going toward FreeDOS using GPT

Le jeu., 24 juin 2021 11:40:37 -0400 Eric Auer écrit >> I think it is where I should start: write a FreeDOS utility to create and >> manipulae a GPT. >I disagree, you can use Linux or Windows to create GPT >FAT partitions for DOS. So you would want to start by >DOS being able

Re: [Freedos-devel] Going toward FreeDOS using GPT

Le jeu., 24 juin 2021 09:15:08 -0400 tom ehlert écrit >> And so, it seems to me that there is not much reason to create MBR partition >> tables today. >except the fact the DOS utilities create MBR partitions, and DOS >kernels only detect MBR partitions. I think it is where I should

[Freedos-devel] The GPL version issue with the NET thing

I don't understand clearly what it is about. So I will try to give an idea of what I believe it is, so that you can correct easily my bad understanding. X have made a program A that let you find what is the packet driver you need for your network card, and make sure it is booted either in

Re: [Freedos-devel] Going toward FreeDOS using GPT

Le jeu., 24 juin 2021 17:35:46 -0400 Bret Johnson écrit >Paul: [..] >> ... And so, it seems to me that there is not much reason to create >> MBR partition tables today. >Not true at all. Even a GPT-formatted disk is REQUIRED to have an MBR in >sector 0 (called a Protective

Re: [Freedos-devel] Why games have bad color in VirtualBox

I have checked, https://www.virtualbox.org/ticket/4777 and it seems there is no bug... today. (bug was reported 12 years ago) I am using doc at: https://web.archive.org/web/20081211174813/http://docs.ruudkoot.nl/vbe20.txt VBox, return VBE 2.0, so it does support AH=9 to set/get palette.

[Freedos-devel] Problems on 1.3rc4 on VirtualBox

VirtualBox 6.1.22 r144080 128 Mb memroy Sound disabled: Get an error on host when launching with it, so I disabled it Processior: i3-8100 French: I think I have not seen any accented characters, like all replaced by 2 chars (Unicode) mkeyb fr: m letter seems undefined  *I think* it is

[Freedos-devel] LSM: Entered-date means modified date

I stumble on: http://www.ibiblio.org/pub/micro/pc-stuff/freedos/files/repositories/1.3/games/senet.zip where LSM file contains: Begin3 Title:  Senet Version:    1.0a Entered-date:   2020-07-07 Modified-date:  2021-03-18.0 I believed this is wrong. One tend to believe

[Freedos-devel] mkeyb us: don't need a driver for US... BUT I AM NOT IN US!

Trying FC4 in QEMU on Linux with: $ qemu-system-i386 -enable-kvm -m 64M -cdrom FD13LIVE.iso  -hda hd.img -soundhw sb16 -boot c I have install to hard disk. Tried to choose Alternate-French for keyboard layout... I was expecting to choose Québec from a menu, but it continued to install.

[Freedos-devel] Some stuff at the beginning of Apps\adplay\readme.txt ?

I have some stuff at the beginning of Apps\adplay\readme.txt file. Chacracters like ... etc. which make less think it might be a binary file. It make me realize I did not check my download. Oh well, too lazy to do it... mentioning it anyway. ;-)___

[Freedos-devel] Re : mkeyb us: don't need a driver for US... BUT I AM NOT IN US!

I have given wrong info in the too much info I have given: $ qemu-system-i386 -enable-kvm -m 64M -cdrom FD13LIVE.iso  -hda hd.img -soundhw sb16 -boot c that's what I use now that it is intalled... to boot C: but first time for installation, I used -boot d to boot from the cdrom

[Freedos-devel] Re : Some stuff at the beginning of Apps\adplay\readme.txt ?

Will try to refrain from posting too fast... I opened it with edit and realized it is "ADPLAY" in a banner style. So it is fine.___ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-devel

[Freedos-devel] Re : Numerous problems with games and reboot (1.3RC4) on real hardware too

I took the hard disk where I had installed FreeDOS on my i3-8100, and tranfer the disk in a E6300 computer. All the games that were not working on i3-8100 worked on the older E6300 CPU (with default option 1). Well... except the USB mouse that was obviously not working... and gnuchess that

[Freedos-devel] 1.3 RC4 works on my hardware: I am surprised!

I'd like to report that the livecd.iso works for me on my hardware: Prime H310M-C R2.0, with I3-8100 CPU. I am impress and surprise, because it have a SATA chip that does not emulate IDE. It does fail on first cdrom driver: UDVD2?, but works with fallback: El-Torito. I was expecting it

[Freedos-devel] Numerous problems with games and reboot (1.3RC4) on real hardware too

I first installed on a 500 Gb disk I found that was labeled: Read errors. After finding many errors, I installed on an other disk of 1000 Gb... did not test as much, bug got same errors. Here what I have noted: reboot: JemmEx exception 0d at CS:EIP=E800:6DA, ErrC=0 Esc to abort, worked

Re: [Freedos-devel] Numerous problems with games and reboot (1.3RC4) on real hardware too

Le mar., 18 mai 2021 14:23:43 -0400 Eric Auer écrit >Or can you access CD/DVD even after booting the installed FreeDOS from >harddisk? No I cannot access DVD on installed system. >> reboot: >> >> JemmEx exception 0d at CS:EIP=E800:6DA, ErrC=0 >> >> Esc to abort, worked >You mean

Re: [Freedos-devel] Numerous problems with games and reboot (1.3RC4) on real hardware too

I have discovered that the fact of "loading" jemm386 or jemmex, breaks my int 13h interrupt. I had renamed fdauto.bat for it not to loads... so as better isolate the problem with fdconfig.sys. I have no reason to believe that it load it in my 4kb reserved word around segment 5800 (it loads

Re: [Freedos-devel] Which assembley syntax is this?

Tom wrote: >end of your support budget for today. Ok... I found the answer by the author of cowgol2: https://github.com/davidgiven/cowgol/issues/117#issuecomment-848595904 So it has to be called with: nasm -f obj hello.s (works for me in FreeDOS) and then linked with djlink which it

Re: [Freedos-devel] Which assembley syntax is this?

As it is, I get: hello.s:17: error: parser: instruction expected and if I comment out (add ; at the beginning of) line 17: group dgroup data bss More context:     12        segment stack stack noprogbits align=16     13        resb 128     14    stacktop:     15        segment data

[Freedos-devel] Which assembley syntax is this?

I have seen in FreeDOS news about cowgol2 programming language: http://cowlark.com/cowgol/ I have built in on Linux Mint... main problem was to get package: m68k-atari-mint-binutils Got it after: sudo add-apt-repository ppa:vriviere/ppa apt-update taken from:

Re: [Freedos-devel] Numerous problems with games and reboot (1.3RC4) on real hardware too

Le dim., 23 mai 2021 10:11:02 -0400 Eric Auer écrit >yes, EMM386 could switch the entire system back to >real mode for certain things, but that would make >the performance loss significantly worse than what >you get with any EMM386 anyway unless you use the >VME (V86 mode

Re: [Freedos-devel] Numerous problems with games and reboot (1.3RC4) on real hardware too

hum... looking back at the problem of calling int 13h with ax=0 and dl=80 give an exception ... on lidt instruction I think... Jemmex only do mov byte cs:[1d],2 before calling original BIOS interrupt. I begin to expect it to: save idtr and gdtr switch to real mode (from protected mode

Re: [Freedos-devel] Numerous problems with games and reboot (1.3RC4) on real hardware too

Le dim., 23 mai 2021 09:47:32 -0400 Paul Dufresne écrit >hum... looking back at the problem of calling int 13h with ax=0 and dl=80 give >an exception ... on lidt instruction I think... > >Jemmex only do mov byte cs:[1d],2 before calling original BIOS interrupt. > >I begin to

Re: [Freedos-devel] Numerous problems with games and reboot (1.3RC4) on real hardware too

Le jeu., 20 mai 2021 07:58:38 -0400 Eric Auer écrit >but the actual problem is that Paul runs into - apparently BIOS >related - attempts to switch to protected mode in spite of >EMM386 already using that. The problematic code - Paul writes Yeah, it make sense... it must need to

Re: [Freedos-devel] Numerous problems with games and reboot (1.3RC4) on real hardware too

Oh!... Sorry. The bug is in my head: You add to the segment... which automatically multiply by the number of paragraphs. ... because each segment begin at a paragraph.___ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net

Re: [Freedos-devel] weird memory map explained (was numerous problems...)

Tom said: > [    0.00] BIOS-e820: [mem 0x00058000-0x00058fff] reserved >if this is real (that's a BIG if), then this machine is not DOS >compatible. Well... Tom was right about the BIG if. First here what I now believe is the real memory map of my computer: [   

Re: [Freedos-devel] Which assembley syntax is this?

I said: >which it seems can be found: http://www.delorie.com/djgpp/16bit/djlink/ >but the author says he lost interest before finishing these projects >(not tried yet) I was not able to compile djlink with djgpp 10.3. So, I did: nasm -f obj hello.s wlink file hello.obj generate

Re: [Freedos-devel] weird memory map explained (was numerous problems...)

I was looking in BIOS interrupts info from: http://www2.ift.ulaval.ca/~marchand/ift17583/dosints.pdf When I saw: INT 15 - AH = 89h SYSTEM - SWITCH TO VIRTUAL MODE (AT,XT286,PS50+) BL = interrupt number of IRQ0 (IRQ1-7 use next 7 interrupts) BH = interrupt number of IRQ8 (IRQ9-F use next 7

[Freedos-devel] Re : GPT processing code from Bret

I am unsure why FreeDOS kernel need to know about GPT. I means, once the kernel have the LBA of the first sector of the FAT partition, I guess it find the info for the last LBA accessible from that partition. I may be wrong, but in my head, I expect FreeDOS not to have to know if the partition

[Freedos-devel] Re : Re : how to handle GPT partitions

Thinking a bit more about it... I would suggest the following design: config.sys would load partmgr.sys, that at load time, would scan each hard drive. partmgr would know about MBR partitions and GPT partitions. It would scan each partition, and if it is a known partition type (FAT), would

Re: [Freedos-devel] Numerous problems with games and reboot (1.3RC4) on real hardware too

Le ven., 21 mai 2021 10:47:31 -0400 Eric Auer écrit >> I can compile a small program to open a file in writing, write a char, >> then close. In 32 bit mode (wcl386) and it works. >No matter on which disk you put the file, even with JEMMEX loaded? Well... on my hard disk...

[Freedos-devel] What a good MBR should do

I was thinking what a modern MBR should do and come to the following pseudo-code: If GPT exist If BIOS_Boot_Partition exist Give control to it (probably second stage GRUB) else if a GPT bootable partition exist Give control to it (could be many things) else Error 'No GPT bootable

Re: [Freedos-devel] What a good MBR should do

I used to think that older BIOS cannot boot GPT partition, but I now believe the work to handle GPT is done by the MasterBootRecord. And so, it seems to me that there is not much reason to create MBR partition tables today. I think FreeDOS could do a GPT, without BIOS boot partition, just

[Freedos-devel] cpp -remap Enable special code to work around file systems which only permit very short file names, such as MS-DOS.

While reading by mistake https://gcc.gnu.org/onlinedocs/cpp/Invocation.html (I had searched for gcc invocation) So this is the preprocessor (to gcc). I noted these parameters that can seems usefull for FreeDOS and less for DJGPP: -remap Enable special code to work around file systems

Re: [Freedos-devel] Allegro library would need to be recompiled with -fgnu89-inline ?

Ok I did encounter again a multiple definition error with ld (linking) in a different project today. The idea seems that if you have multiple files that include some.h, and in it you have functions or variables not using the static keyword (at least for C), then they will be defined as

Re: [Freedos-devel] Allegro library would need to be recompiled with -fgnu89-inline ?

I now think that adding "external" before declarations in .h is the correct solution. If they were not mark static already, it is that the value is expected to be shared (being global). The solution is *I now think* to add external, and have the real declaration in a unique place, like in

Re: [Freedos-devel] Allegro library would need to be recompiled with -fgnu89-inline ?

Le sam., 26 juin 2021 18:20:36 -0400 Paul Dufresne I wrote >I now think that adding "external" before declarations in .h is the correct >solution. >If they were not mark static already, it is that the value is expected to be >shared (being global). >The

  1   2   3   >