Re: [Freedos-devel] SCANDISK vs DOSFSCK

2004-03-29 Thread Luchezar Georgiev
I disagree here: the important thing is to get the job done. After that if there is a nice interface, it can be great, but not essencial. In DR-DOS you only have one CHKDSK without UI. Why is the focus of scandisk on the interface I cannot imagine, even to the point of someone making an empty i

[Freedos-devel] Proposed EMM386 changes and status

2004-03-29 Thread Michael Devore
I am going to change the way EMM386 works a final time and since EMM386 with VCPI support is close to signing off as a generally stable release as far the VCPI feature, I'm posting the status so people know where we're at and have a last chance for input on VCPI-related mods. NIOS has problems

Re: [Freedos-devel] FD DEBUG issues

2004-03-29 Thread Matthias Paul
On 2004-03-27, Arkady V.Belousov wrote: > Also, this will be fine if DEBUG implements support for different > radix: say, "18h" for hex, "10t" for decimal and "101010i" for > binary. Also, may be used ADA-like syntax 2$10101, 10$123, 16$FF, > $FF, where radix is a decimal value. Just another sugg

Re: [Freedos-devel] SCANDISK vs DOSFSCK

2004-03-29 Thread Aitor Santamaría Merino
Alain escribió: I disagree here: the important thing is to get the job done. After that if there is a nice interface, it can be great, but not essencial. In DR-DOS you only have one CHKDSK without UI. Why is the focus of scandisk on the interface I cannot imagine, even to the point of someone

Re: [Freedos-devel] UDMA

2004-03-29 Thread Michael Devore
At 09:21 PM 3/29/2004 +0200, Bernd Blaauw wrote: >>Not yet, looks like it should be. I'll see about it, read what MS docs say about >>the option. Should be a reasonably quick add, since it's awfully close to NOEMS in >>a lot of behaviors. > >suggested using FRAME=NONE to Erwin, since he needs

Re: [Freedos-devel] SCANDISK vs DOSFSCK

2004-03-29 Thread Alain
Aitor Santamari'a Merino escreveu: Luchezar Georgiev escribio': On Thu, 25 Mar 2004 20:08:32 -0300, Alain wrote: about fat32 testing: I believe a working DOSFSCK 2.10 just what is needed (not what is whished for). Actually, I agree! If Eric can say "FreeDOS SMARTDRV is LBACACHE", why not s

Re: [Freedos-devel] UDMA

2004-03-29 Thread Bernd Blaauw
Hi Michael, Not yet, looks like it should be. I'll see about it, read what MS docs say about the option. Should be a reasonably quick add, since it's awfully close to NOEMS in a lot of behaviors. suggested using FRAME=NONE to Erwin, since he needs to limit VCPI to 2MB, but NOEMS makes 32MB on

Re: [Freedos-devel] BIOS

2004-03-29 Thread Luchezar Georgiev
On Mon, 29 Mar 2004 19:00:06 +0200, Bernd Blaauw wrote: does not imply the manufacturers or even the creators of the generic BIOS know the BIOS code very good. BIOS code developed in same way as Windows: ...and DOS keep compatibility and attach some really ugly extensions and work-arounds. Ind

Re: [Freedos-devel] UDMA

2004-03-29 Thread Michael Devore
At 04:58 PM 3/29/2004 +0200, Bernd Blaauw wrote: >btw, is FRAME=NONE supported in EMM386.EXE rc2 ? Not yet, looks like it should be. I'll see about it, read what MS docs say about the option. Should be a reasonably quick add, since it's awfully close to NOEMS in a lot of behaviors. --

Re: [Freedos-devel] BIOS

2004-03-29 Thread Bernd Blaauw
Luchezar Georgiev schreef: As far as I see, LinuxBIOS *replaces* the built-in BIOS, whereas I'm talking about fitting an [D]OS kernel into the free space of *existing* BIOS. No one should know better their board than the manufacturer itself who has bought the generic BIOS from Award (Phoenix),

Re: [Freedos-devel] BIOS

2004-03-29 Thread Luchezar Georgiev
On Mon, 29 Mar 2004 15:43:26 +0200, Bernd Blaauw wrote: Luchezar Georgiev schreef: I think that each OS has its own niche. Neiether Linux nor DOS can oust each other. But try to fit a Linux in the BIOS flash ROM chip when you have only several tens of kilobytes free there! www.linuxbios.org As

Re: [Freedos-devel] UDMA

2004-03-29 Thread tom ehlert
Hello Bernd, BB> can Lucho, Michael or Tom confirm VDS is only needed when EMS is provided? VDS is required, if logical adresses (as seen from CPU) != physical addresses (as seen from UDMA controller). In praxis, with our current EMM386, addresses <= 0xa are mapped identical, so everything w

Re: [Freedos-devel] UDMA

2004-03-29 Thread Bernd Blaauw
that's nice. So VDS is required when EMS is served, not when VCPI and/or UMBs are provided. NOEMS and EMS settings use different blockdevice-names: one more optimizing thing for me then: @echo off rem load UDMA from commandline using DEVLOAD program. /H = DEVICEHIGH behaviour. rem try NOEMS first,

Re: [Freedos-devel] UDMA

2004-03-29 Thread Johnson Lam
Hello, >>I'll try DEVICEHIGH with EMM386. >>didn't Tom implement a basic VDS option? >>DEVICE=EMM386.EXE VDS Even Lucho said UDMA need VDS, I simply use the DEVICE=EMM386.EXE NOEMS it works fine. It's more stable than FDDXMS+UMBPCI in my ALI Aladdin chipset, using FDDXMS+UMBPCI sometimes UDMA cl

Re: [Freedos-devel] UDMA

2004-03-29 Thread Bernd Blaauw
Luchezar Georgiev schreef: I think that each OS has its own niche. Neiether Linux nor DOS can oust each other. But try to fit a Linux in the BIOS flash ROM chip when you have only several tens of kilobytes free there! www.linuxbios.org ---

Re: [Freedos-devel] UDMA

2004-03-29 Thread Luchezar Georgiev
On Mon, 29 Mar 2004 20:50:39 +0800, Johnson Lam wrote: Though the driver may not bug free, but it bring the technology back to DOS. Still have a lot of people including my friend think that DOS is out, why not Linux? I think that each OS has its own niche. Neiether Linux nor DOS can oust each o

Re: [Freedos-devel] UDMA

2004-03-29 Thread Johnson Lam
On Sun, 28 Mar 2004 17:34:56 +0300, you wrote: Hi Lucho, >It's good to have so experienced programmers contribute to FreeDOS. One >thing could prevent them from doing so though - the possibility of being Though the driver may not bug free, but it bring the technology back to DOS. Still have a