Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.os.cpm > #1645 > unrolled thread
| Started by | Philippe <pas.de.spam@ici.fr> |
|---|---|
| First post | 2012-08-15 18:37 +0200 |
| Last post | 2012-08-18 21:50 +0700 |
| Articles | 18 — 7 participants |
Back to article view | Back to comp.os.cpm
cpm-86 disk image Philippe <pas.de.spam@ici.fr> - 2012-08-15 18:37 +0200
Re: cpm-86 disk image Peter Dassow <z80eu@arcor.de> - 2012-08-17 12:37 +0200
Re: cpm-86 disk image with CBASIC-86 Peter Dassow <z80eu@arcor.de> - 2012-08-17 13:11 +0200
Re: cpm-86 disk image with CBASIC-86 Philippe <pas.de.spam@ici.fr> - 2012-08-17 18:26 +0200
Re: cpm-86 disk image with CBASIC-86 Peter Dassow <z80eu@arcor.de> - 2012-08-17 21:04 +0200
running CP/M-86 images Ivan Shmakov <oneingray@gmail.com> - 2012-08-18 01:12 +0700
Re: running CP/M-86 images Steve Nickolas <usotsuki@buric.co> - 2012-08-17 20:36 +0200
Re: running CP/M-86 images Ivan Shmakov <oneingray@gmail.com> - 2012-08-18 02:08 +0700
Re: running CP/M-86 images Ivan Shmakov <oneingray@gmail.com> - 2012-08-18 02:42 +0700
network boot: CP/M-86, FreeDOS, GNU/Linux Ivan Shmakov <oneingray@gmail.com> - 2012-08-18 18:46 +0700
Re: network boot: CP/M-86, FreeDOS, GNU/Linux "Harald Peters" <haj.peters@home.nl> - 2014-04-23 00:25 +0200
Re: running CP/M-86 images - Yes it runs also with modern real PC hardware Peter Dassow <z80eu@arcor.de> - 2012-08-18 20:25 +0200
Re: running CP/M-86 images - Yes it runs also with modern real PC hardware Tarkin <tarkin000@gmail.com> - 2012-09-19 11:32 -0700
Re: running CP/M-86 images Peter Dassow <z80eu@arcor.de> - 2012-08-17 20:58 +0200
Re: running CP/M-86 images Ivan Shmakov <oneingray@gmail.com> - 2012-08-18 02:23 +0700
Re: running CP/M-86 images Peter Dassow <z80eu@arcor.de> - 2012-08-17 21:47 +0200
Re: running CP/M-86 images s_dubrovich@yahoo.com - 2012-08-17 20:43 -0700
Re: running CP/M-86 images Ivan Shmakov <oneingray@gmail.com> - 2012-08-18 21:50 +0700
| From | Philippe <pas.de.spam@ici.fr> |
|---|---|
| Date | 2012-08-15 18:37 +0200 |
| Subject | cpm-86 disk image |
| Message-ID | <502bd041$0$6184$ba4acef3@reader.news.orange.fr> |
Hello there, I have built a working cpm-86 emulator with virtualbox (with these guide : http://www.z80.eu/blog/index.php?m=08&y=12&d=29&entry=entry120807-180000). In order to do that, I run the included MAKEDISK.BAT in a virtual WindowsXP, writing in a virtual floppy. Then I rename the resulting image file from img to dsk and use it as a boot floppy for the virtual cpm-86 machine. Now I try to add a second floppy to the cpm-86 system with another image (made from this file : http://www.cpm.z80.de/download/cbasic86.zip) but now the question : how can I generate a cpm-86 readable floppy image (readable by virtualbox and by cpm-86) ? Many thanks for helping. Philippe
[toc] | [next] | [standalone]
| From | Peter Dassow <z80eu@arcor.de> |
|---|---|
| Date | 2012-08-17 12:37 +0200 |
| Message-ID | <502e1ee3$0$6553$9b4e6d93@newsspool4.arcor-online.net> |
| In reply to | #1645 |
On 15.08.2012 18:37, Philippe wrote: > I have built a working cpm-86 emulator with virtualbox (with these guide > : > http://www.z80.eu/blog/index.php?m=08&y=12&d=29&entry=entry120807-180000). > > In order to do that, I run the included MAKEDISK.BAT in a virtual > WindowsXP, writing in a virtual floppy. Then I rename the resulting > image file from img to dsk and use it as a boot floppy for the virtual > cpm-86 machine. That's because you need a non-compressed image file. > Now I try to add a second floppy to the cpm-86 system with another image > (made from this file : http://www.cpm.z80.de/download/cbasic86.zip) but > now the question : > how can I generate a cpm-86 readable floppy image (readable by > virtualbox and by cpm-86) ? Your mentioned cbasic86.zip contains only CP/M program and data files, not a floppy image file. You will be able to generate a blank floppy disk (image) inside the emulator / while running the emulator with CP/M-86. Then you should convert the VirtualBox image file with my small utility (I added a usage hint in my blog, too). After the image file is "cpmtools" compatible, you can manipulate the image file with cpmtools commands. When you're done, convert the "cpmtools" image back to a VirtualBox compatible image file again with my small utility. You should be able now to work with the "new" image/floppy. Regards Peter
[toc] | [prev] | [next] | [standalone]
| From | Peter Dassow <z80eu@arcor.de> |
|---|---|
| Date | 2012-08-17 13:11 +0200 |
| Subject | Re: cpm-86 disk image with CBASIC-86 |
| Message-ID | <502e26ec$0$6564$9b4e6d93@newsspool4.arcor-online.net> |
| In reply to | #1656 |
On 17.08.2012 12:37, Peter Dassow wrote: > [...] > When you're done, convert the "cpmtools" image back to a VirtualBox > compatible image file again with my small utility. > You should be able now to work with the "new" image/floppy. For your convenience and as a proof, I added even a disk image with added CBASIC-86 to my blog entry at www.z80.eu/blog/ ... Regards Peter
[toc] | [prev] | [next] | [standalone]
| From | Philippe <pas.de.spam@ici.fr> |
|---|---|
| Date | 2012-08-17 18:26 +0200 |
| Subject | Re: cpm-86 disk image with CBASIC-86 |
| Message-ID | <502e70b3$0$6191$ba4acef3@reader.news.orange.fr> |
| In reply to | #1657 |
Le 17/08/2012 13:11, Peter Dassow a écrit : > On 17.08.2012 12:37, Peter Dassow wrote: >> [...] >> When you're done, convert the "cpmtools" image back to a VirtualBox >> compatible image file again with my small utility. >> You should be able now to work with the "new" image/floppy. > For your convenience and as a proof, I added even a disk image with > added CBASIC-86 to my blog entry at www.z80.eu/blog/ ... > > Regards > Peter > I appreciate your help but things are not very clear for me. Can you explain step by step how you have created the empty floppy in cpm86 (running in virtualbox), move them to a cpmtools writable format, put the files inside and then move them back to virtualbox ? Again I really appreciate your help. Regards. Philippe
[toc] | [prev] | [next] | [standalone]
| From | Peter Dassow <z80eu@arcor.de> |
|---|---|
| Date | 2012-08-17 21:04 +0200 |
| Subject | Re: cpm-86 disk image with CBASIC-86 |
| Message-ID | <502e95db$0$6557$9b4e6d93@newsspool4.arcor-online.net> |
| In reply to | #1658 |
On 17.08.2012 18:26, Philippe wrote: >>> [...] >>> When you're done, convert the "cpmtools" image back to a VirtualBox >>> compatible image file again with my small utility. >>> You should be able now to work with the "new" image/floppy. >> For your convenience and as a proof, I added even a disk image with >> added CBASIC-86 to my blog entry at www.z80.eu/blog/ ... >> >> Regards >> Peter >> > I appreciate your help but things are not very clear for me. Can you > explain step by step how you have created the empty floppy in cpm86 > (running in virtualbox), move them to a cpmtools writable format, put > the files inside and then move them back to virtualbox ? > > Again I really appreciate your help. When you managed it to boot CP/M-86 (having two drives ready configured), you can use the CP/M command "DSKMAINT". More easy is you make a copy of the boot disk image, mount this with the second floppy drive and after booting CP/M-86 from A:, delete everything from B: (e.g. with command "era b:*.*"). As a result, you get a blank floppy. Instead of using a second floppy drive, try to use the harddisk (as described in my blog again). Regards Peter
[toc] | [prev] | [next] | [standalone]
| From | Ivan Shmakov <oneingray@gmail.com> |
|---|---|
| Date | 2012-08-18 01:12 +0700 |
| Subject | running CP/M-86 images |
| Message-ID | <868vdd2xrj.fsf_-_@gray.siamics.net> |
| In reply to | #1657 |
>>>>> Peter Dassow <z80eu@arcor.de> writes:
[...]
> For your convenience and as a proof, I added even a disk image with
> added CBASIC-86 to my blog entry at www.z80.eu/blog/ ...
BTW, out of curiosity (I vaguely recall running a Unix-based
CP/M + Z80 emulator a few times, having no experience with
either CP/M proper or CP/M-86) I've downloaded and extracted the
image (floppy.flp of cpm86fl2.zip), and... It didn't work.
So, the first question: is floppy.flp a "raw" floppy disk image
(as could be guessed by its size, which is exactly 1440 KiB)?
If it indeed is, it was my understanding that PC BIOS expects
the first sector to be marked with 0x55, 0xAA in its two last
bytes for the media to be considered "bootable":
--cut: http://en.wikipedia.org/wiki/Boot_sector --
[...] The BIOS merely passes control to whatever exists there, as
long as the sector meets the very simple qualification of having the
boot record signature of 0x55, 0xAA in its last two bytes. [...]
--cut: http://en.wikipedia.org/wiki/Boot_sector --
For floppy.flp, the last two bytes are as follows:
$ od -Ax -j 510 -N 2 -t x1 < floppy.flp
0001fe 01 90
000200
$
Naturally, QEMU refuses to boot from such a media.
Also, I've tried a few images from http://cpm.z80.de/.
Unfortunately, these seem to be in a vendor-specific CopyQM
format, which I know of no utility (beside CopyQM) can support.
Unfortunately, CopyQM fails with "Drive not ready" under
FreeDOS/QEMU for me, and I'm reluctant to try it on some "real"
hardware (with a FDD.)
Any suggestions?
TIA.
--
FSF associate member #7257 http://sf-day.org/
[toc] | [prev] | [next] | [standalone]
| From | Steve Nickolas <usotsuki@buric.co> |
|---|---|
| Date | 2012-08-17 20:36 +0200 |
| Subject | Re: running CP/M-86 images |
| Message-ID | <alpine.DEB.2.00.1208172036210.25096@ns383864.ovh.net> |
| In reply to | #1660 |
On Sat, 18 Aug 2012, Ivan Shmakov wrote: > Naturally, QEMU refuses to boot from such a media. qemu -no-fd-bootchk -uso.
[toc] | [prev] | [next] | [standalone]
| From | Ivan Shmakov <oneingray@gmail.com> |
|---|---|
| Date | 2012-08-18 02:08 +0700 |
| Subject | Re: running CP/M-86 images |
| Message-ID | <864no12v64.fsf@gray.siamics.net> |
| In reply to | #1661 |
>>>>> Steve Nickolas <usotsuki@buric.co> writes: >>>>> On Sat, 18 Aug 2012, Ivan Shmakov wrote: >> Naturally, QEMU refuses to boot from such a media. > qemu -no-fd-bootchk Indeed, it works that way, thanks! However, I'm still curious on whether a "typical" PC BIOS (as of the time cheaper PC's still had FDC's) forces this check or not? -- FSF associate member #7257 http://sf-day.org/
[toc] | [prev] | [next] | [standalone]
| From | Ivan Shmakov <oneingray@gmail.com> |
|---|---|
| Date | 2012-08-18 02:42 +0700 |
| Subject | Re: running CP/M-86 images |
| Message-ID | <86vcgh1f1s.fsf@gray.siamics.net> |
| In reply to | #1664 |
To note is that I've added the CP/M-86 1.44" image to my
iPXE-based "Network boot" menu, too. Now, virtually any x86
machine (or an "emulation" thereof) in my LAN, may boot a
CP/M-86 system given a suitable iPXE command sequence or script.
The relevant parts of the iPXE "menu" script I've used are as
follows:
#!ipxe
:menu
menu Boot menu
# ...
item cpm86 Boot CP/M-86 [cpm.z80.de/download/144cpm86.zip; memdisk]
# ...
item shell Run iPXE Shell
choose boot && goto ${boot}
# ...
:cpm86
kernel /diskless/memdisk.lkrn
initrd ../bin/144cpm86.img
boot || exit 1
# ...
PS. I'm quite sure that this configuration will work over Internet,
too, but I'm somewhat reluctant to share the server's URI to
point iPXE at, as this version of CP/M-86 doesn't seem like a
free software (as in "free speech".)
--
FSF associate member #7257 http://sf-day.org/
[toc] | [prev] | [next] | [standalone]
| From | Ivan Shmakov <oneingray@gmail.com> |
|---|---|
| Date | 2012-08-18 18:46 +0700 |
| Subject | network boot: CP/M-86, FreeDOS, GNU/Linux |
| Message-ID | <868vdc1kyb.fsf_-_@gray.siamics.net> |
| In reply to | #1667 |
>>>>> Ivan Shmakov <oneingray@gmail.com> writes:
[Cross-posting to news:comp.os.linux.networking and
news:comp.os.msdos.misc, for an example setup for booting the
Debian GNU/Linux 6.0.4 Live and FreeDOS Balder images over a
network is provided. I'm looking for both a better-suited
newsgroup, and a configuration to boot a ZX Spectrum emulator.]
> To note is that I've added the CP/M-86 1.44" image to my iPXE-based
> "Network boot" menu, too. Now, virtually any x86 machine (or an
> "emulation" thereof) in my LAN, may boot a CP/M-86 system given a
> suitable iPXE command sequence or script.
[...]
> PS. I'm quite sure that this configuration will work over Internet,
> too, but I'm somewhat reluctant to share the server's URI to point
> iPXE at, as this version of CP/M-86 doesn't seem like a free software
> (as in "free speech".)
... On a second though, here it is:
http://ritic.am-1.org/diskless/ipxe/world-default
One can run it with QEMU (or QEMU/KVM) like:
$ qemu \
-kernel virtio-net.lkrn \
-net nic,model=virtio \
-net user,bootfile=http://173.0.50.100/diskless/ipxe/world-default
It seems that the version of QEMU I use fails to provide DNS to
the child, so I had to use "plain" IPv4 addresses within both
the command line above and the "world-default" iPXE script, as
well as providing a copy of balder10.img on my server (as
http://ibiblio.org/ apparently requires the proper Host: HTTP
header.)
For booting "real" hardware, one'd need a DHCP server, and also
a TFTP one (unless a recent version of iPXE is burned into the
boot ROM.)
E. g., the ISC DHCP server may be configured roughly as follows
(assuming a static MAC to IPv4 mapping.)
### dhcpd.conf -*- Default-Generic -*-
default-lease-time 600;
max-lease-time 7200;
subnet 192.0.2.0 netmask 255.255.255.0 {
use-host-decl-names on;
option subnet-mask 255.255.255.0;
option routers 192.0.2.1;
option domain-name-servers
192.0.2.53;
# option ntp-servers 192.0.2.123;
if exists user-class and option user-class = "iPXE" {
filename "http://173.0.50.100/diskless/ipxe/world-default";
} else {
## download iPXE from a TFTP server
next-server 192.0.2.69;
filename "undionly.kpxe";
}
host diskless.example.org {
hardware ethernet 00:27:0e:05:e5:9e;
fixed-address 192.0.2.97;
}
}
### dhcpd.conf ends here
Also note that the Debian 6.0.4 Live "standard" image is some
177 MiB in total, and thus takes a while to load from
http://cdimage.debian.org/ (AKA http://130.239.18.163/.) It's
provided as an example only; one should really consider putting
it somewhere within his or her own network! (There's also a
section for loading the "Xfce desktop" image, but its menu entry
was commented-out, so not to abuse the cdimage server.)
The files provided (below /bin/) are as follows. (I'd like to
thank Aioe for its wonderful "441 Lines longer than 79 chars".)
Image: 144cpm86.img
Copy-of: http://z80.eu/downloads/cpm86fl2.zip (floppy.flp)
Description: CP/M-86 1.44" bootable floppy
Originally from http://cpm.z80.de/download/144cpm86.zip.
Image: balder10.img
Copy-of: http://ibiblio.org/pub/micro/pc-stuff/freedos/files/distributions/
unofficial/balder/balder10.img
Description: Balder 1.0 (FreeDOS) 1.44" bootable floppy
Image: memdisk.lkrn
Copy-of: http://snapshot.debian.org/archive/debian/20120630T153537Z/pool/
main/s/syslinux/syslinux-common_4.05%2Bdfsg-6_all.deb
(./usr/lib/syslinux/memdisk)
Description: MEMDISK implements a RAM disk by hooking INT 13h
This one comes from the SYSLINUX project.
Image: memtest86+.0
Source: https://git.ipxe.org/people/mcb30/memtest.git/commit/f9d3679f3770
Description: Memtest86+ memory tester (PXE)
This version of Memtest86+ was built with an experimental PXE
support.
Image: memtest86.bin
Copy-of: http://snapshot.debian.org/archive/debian/20120514T041508Z/pool/
main/m/memtest86/memtest86_4.0s-1_amd64.deb (./boot/memtest86.bin)
Description: Memtest86 memory tester
Image: pxelinux.0
Copy-of: http://snapshot.debian.org/archive/debian/20120630T153537Z/pool/
main/s/syslinux/syslinux-common_4.05%2Bdfsg-6_all.deb
(./usr/lib/syslinux/pxelinux.0)
Description: PXELINfUX loads Linux from a network using PXE
--
FSF associate member #7257 http://sf-day.org/
[toc] | [prev] | [next] | [standalone]
| From | "Harald Peters" <haj.peters@home.nl> |
|---|---|
| Date | 2014-04-23 00:25 +0200 |
| Subject | Re: network boot: CP/M-86, FreeDOS, GNU/Linux |
| Message-ID | <op.xeqoj6knngizc1@kargamel> |
| In reply to | #1678 |
is that image Public Domain?
Op Sat, 18 Aug 2012 13:46:52 +0200 schreef Ivan Shmakov
<oneingray@gmail.com>:
>>>>>> Ivan Shmakov <oneingray@gmail.com> writes:
>
> [Cross-posting to news:comp.os.linux.networking and
> news:comp.os.msdos.misc, for an example setup for booting the
> Debian GNU/Linux 6.0.4 Live and FreeDOS Balder images over a
> network is provided. I'm looking for both a better-suited
> newsgroup, and a configuration to boot a ZX Spectrum emulator.]
>
> > To note is that I've added the CP/M-86 1.44" image to my iPXE-based
> > "Network boot" menu, too. Now, virtually any x86 machine (or an
> > "emulation" thereof) in my LAN, may boot a CP/M-86 system given a
> > suitable iPXE command sequence or script.
>
> [...]
>
> > PS. I'm quite sure that this configuration will work over Internet,
> > too, but I'm somewhat reluctant to share the server's URI to point
> > iPXE at, as this version of CP/M-86 doesn't seem like a free software
> > (as in "free speech".)
>
> ... On a second though, here it is:
>
> http://ritic.am-1.org/diskless/ipxe/world-default
>
> One can run it with QEMU (or QEMU/KVM) like:
>
> $ qemu \
> -kernel virtio-net.lkrn \
> -net nic,model=virtio \
> -net user,bootfile=http://173.0.50.100/diskless/ipxe/world-default
>
> It seems that the version of QEMU I use fails to provide DNS to
> the child, so I had to use "plain" IPv4 addresses within both
> the command line above and the "world-default" iPXE script, as
> well as providing a copy of balder10.img on my server (as
> http://ibiblio.org/ apparently requires the proper Host: HTTP
> header.)
>
> For booting "real" hardware, one'd need a DHCP server, and also
> a TFTP one (unless a recent version of iPXE is burned into the
> boot ROM.)
>
> E. g., the ISC DHCP server may be configured roughly as follows
> (assuming a static MAC to IPv4 mapping.)
>
> ### dhcpd.conf -*- Default-Generic -*-
>
> default-lease-time 600;
> max-lease-time 7200;
>
> subnet 192.0.2.0 netmask 255.255.255.0 {
> use-host-decl-names on;
> option subnet-mask 255.255.255.0;
> option routers 192.0.2.1;
> option domain-name-servers
> 192.0.2.53;
> # option ntp-servers 192.0.2.123;
>
> if exists user-class and option user-class = "iPXE" {
> filename
> "http://173.0.50.100/diskless/ipxe/world-default";
> } else {
> ## download iPXE from a TFTP server
> next-server 192.0.2.69;
> filename "undionly.kpxe";
> }
> host diskless.example.org {
> hardware ethernet 00:27:0e:05:e5:9e;
> fixed-address 192.0.2.97;
> }
> }
>
> ### dhcpd.conf ends here
>
> Also note that the Debian 6.0.4 Live "standard" image is some
> 177 MiB in total, and thus takes a while to load from
> http://cdimage.debian.org/ (AKA http://130.239.18.163/.) It's
> provided as an example only; one should really consider putting
> it somewhere within his or her own network! (There's also a
> section for loading the "Xfce desktop" image, but its menu entry
> was commented-out, so not to abuse the cdimage server.)
>
> The files provided (below /bin/) are as follows. (I'd like to
> thank Aioe for its wonderful "441 Lines longer than 79 chars".)
>
> Image: 144cpm86.img
> Copy-of: http://z80.eu/downloads/cpm86fl2.zip (floppy.flp)
> Description: CP/M-86 1.44" bootable floppy
> Originally from http://cpm.z80.de/download/144cpm86.zip.
>
> Image: balder10.img
> Copy-of:
> http://ibiblio.org/pub/micro/pc-stuff/freedos/files/distributions/
> unofficial/balder/balder10.img
> Description: Balder 1.0 (FreeDOS) 1.44" bootable floppy
>
> Image: memdisk.lkrn
> Copy-of:
> http://snapshot.debian.org/archive/debian/20120630T153537Z/pool/
> main/s/syslinux/syslinux-common_4.05%2Bdfsg-6_all.deb
> (./usr/lib/syslinux/memdisk)
> Description: MEMDISK implements a RAM disk by hooking INT 13h
> This one comes from the SYSLINUX project.
>
> Image: memtest86+.0
> Source:
> https://git.ipxe.org/people/mcb30/memtest.git/commit/f9d3679f3770
> Description: Memtest86+ memory tester (PXE)
> This version of Memtest86+ was built with an experimental PXE
> support.
>
> Image: memtest86.bin
> Copy-of:
> http://snapshot.debian.org/archive/debian/20120514T041508Z/pool/
> main/m/memtest86/memtest86_4.0s-1_amd64.deb (./boot/memtest86.bin)
> Description: Memtest86 memory tester
>
> Image: pxelinux.0
> Copy-of:
> http://snapshot.debian.org/archive/debian/20120630T153537Z/pool/
> main/s/syslinux/syslinux-common_4.05%2Bdfsg-6_all.deb
> (./usr/lib/syslinux/pxelinux.0)
> Description: PXELINfUX loads Linux from a network using PXE
>
--
Gemaakt met Opera's e-mailprogramma: http://www.opera.com/mail/
[toc] | [prev] | [next] | [standalone]
| From | Peter Dassow <z80eu@arcor.de> |
|---|---|
| Date | 2012-08-18 20:25 +0200 |
| Subject | Re: running CP/M-86 images - Yes it runs also with modern real PC hardware |
| Message-ID | <502fde11$0$6548$9b4e6d93@newsspool4.arcor-online.net> |
| In reply to | #1664 |
On 17.08.2012 21:08, Ivan Shmakov wrote: >>>>>> Steve Nickolas <usotsuki@buric.co> writes: >>>>>> On Sat, 18 Aug 2012, Ivan Shmakov wrote: > > >> Naturally, QEMU refuses to boot from such a media. > > > qemu -no-fd-bootchk > > Indeed, it works that way, thanks! > > However, I'm still curious on whether a "typical" PC BIOS (as of > the time cheaper PC's still had FDC's) forces this check or not? I tried to discuss this further at the mentioned Wikipedia boot sector page. Then I stopped and thought, why not trying it in reality with a more modern Thinkpad notebook which still has a build-in floppy drive ? So I added a new blog entry at http://www.z80.eu/blog/ for a photo as a proof that it does work also with much more modern PCs. Btw. VMWare's PhoenixBIOS is really new enough for another proof that it works with newer hardware (although this "hardware" is virtual). Regards Peter
[toc] | [prev] | [next] | [standalone]
| From | Tarkin <tarkin000@gmail.com> |
|---|---|
| Date | 2012-09-19 11:32 -0700 |
| Subject | Re: running CP/M-86 images - Yes it runs also with modern real PC hardware |
| Message-ID | <ad00172e-c5be-4777-8214-90225a05c804@k1g2000vbp.googlegroups.com> |
| In reply to | #1683 |
On Aug 18, 2:25 pm, Peter Dassow <z8...@arcor.de> wrote: > On 17.08.2012 21:08, Ivan Shmakov wrote: > > >>>>>> Steve Nickolas <usots...@buric.co> writes: > >>>>>> On Sat, 18 Aug 2012, Ivan Shmakov wrote: > > > >> Naturally, QEMU refuses to boot from such a media. > > > > qemu -no-fd-bootchk > > > Indeed, it works that way, thanks! > > > However, I'm still curious on whether a "typical" PC BIOS (as of > > the time cheaper PC's still had FDC's) forces this check or not? > > I tried to discuss this further at the mentioned Wikipedia boot sector page. > Then I stopped and thought, why not trying it in reality with a more > modern Thinkpad notebook which still has a build-in floppy drive ? > So I added a new blog entry athttp://www.z80.eu/blog/for a photo as a > proof that it does work also with much more modern PCs. > Btw. VMWare's PhoenixBIOS is really new enough for another proof that it > works with newer hardware (although this "hardware" is virtual). > > Regards > Peter Given the age of CP/M-86, I am curious to know why the floppy boot signature was ever enforced. Could it be an artifact of a specific period of PC history? TTFN, Tarkin
[toc] | [prev] | [next] | [standalone]
| From | Peter Dassow <z80eu@arcor.de> |
|---|---|
| Date | 2012-08-17 20:58 +0200 |
| Subject | Re: running CP/M-86 images |
| Message-ID | <502e9461$0$6570$9b4e6d93@newsspool4.arcor-online.net> |
| In reply to | #1660 |
On 17.08.2012 20:12, Ivan Shmakov wrote: > [...] > > > For your convenience and as a proof, I added even a disk image with > > added CBASIC-86 to my blog entry at www.z80.eu/blog/ ... > > BTW, out of curiosity (I vaguely recall running a Unix-based > CP/M + Z80 emulator a few times, having no experience with > either CP/M proper or CP/M-86) I've downloaded and extracted the > image (floppy.flp of cpm86fl2.zip), and... It didn't work. What exactly did not work ? I documented my success with screenshots, but I didn't wanted to make a step-by-step manual. You can make more than one mistake even in configuration of the VM itself. > So, the first question: is floppy.flp a "raw" floppy disk image > (as could be guessed by its size, which is exactly 1440 KiB)? Yes. But it's track/head order differs from other floppy image formats. > If it indeed is, it was my understanding that PC BIOS expects > the first sector to be marked with 0x55, 0xAA in its two last > bytes for the media to be considered "bootable": No. BIOS is not related with DOS. BIOS will try to load the first sector, regardless what's inside this sector. If loaded in memory (typically at 0000:7C00), IP is set to the first byte of the loaded sector. "Magic Numbers" like 55 AA are only important for Microsoft operating systems. > --cut: http://en.wikipedia.org/wiki/Boot_sector -- > [...] The BIOS merely passes control to whatever exists there, as > long as the sector meets the very simple qualification of having the > boot record signature of 0x55, 0xAA in its last two bytes. [...] > --cut: http://en.wikipedia.org/wiki/Boot_sector -- Wikipedia Bullshit. There is no quality control what is published. That's the proof, CP/M-86 needs no "magic numbers". > For floppy.flp, the last two bytes are as follows: > > $ od -Ax -j 510 -N 2 -t x1 < floppy.flp > 0001fe 01 90 > 000200 > $ You're totally wrong with your assumptions. It doesn't matter what is stored at the end of the sector. > > Naturally, QEMU refuses to boot from such a media. Naturally ? I didn't try QEMU, but you can't speak about nature in conjunction with QEMU. I don't like QEMU, it's too specific in its intended use. VMWare has a far better quality and is much more compatible. > Also, I've tried a few images from http://cpm.z80.de/. > Unfortunately, these seem to be in a vendor-specific CopyQM > format, which I know of no utility (beside CopyQM) can support. > Unfortunately, CopyQM fails with "Drive not ready" under > FreeDOS/QEMU for me, and I'm reluctant to try it on some "real" > hardware (with a FDD.) You don't need to expand the compressed CopyQM image because I already saved a decompressed (but not otherwise modified version!) version. I don't know what you have tried, but it seems to be wrong. I used CopyQM with VMWare and an installed MS-DOS 7.1 without any problems. Regards Peter
[toc] | [prev] | [next] | [standalone]
| From | Ivan Shmakov <oneingray@gmail.com> |
|---|---|
| Date | 2012-08-18 02:23 +0700 |
| Subject | Re: running CP/M-86 images |
| Message-ID | <86zk5t1fwh.fsf@gray.siamics.net> |
| In reply to | #1662 |
>>>>> Peter Dassow <z80eu@arcor.de> writes: >>>>> On 17.08.2012 20:12, Ivan Shmakov wrote: [...] >> So, the first question: is floppy.flp a "raw" floppy disk image (as >> could be guessed by its size, which is exactly 1440 KiB)? > Yes. But it's track/head order differs from other floppy image > formats. ACK, thanks. [...] >> --cut: http://en.wikipedia.org/wiki/Boot_sector -- >> [...] The BIOS merely passes control to whatever exists there, as >> long as the sector meets the very simple qualification of having the >> boot record signature of 0x55, 0xAA in its last two bytes. [...] >> --cut: http://en.wikipedia.org/wiki/Boot_sector -- > Wikipedia Bullshit. There is no quality control what is published. If there's a reliable source supporting a different point, feel free to edit the page accordingly. My edition of Norton's "Programmer's guide to the IBM PC & PS/2" (1988, translated in 1994) states (Chapter 5, "Boot disks" section) that the BIOS indeed checks the boot media signature. > That's the proof, CP/M-86 needs no "magic numbers". There's no CP/M-86 until the BIOS bootloader passes control to it. Which is unlikely to happen without having the aforementioned signature in the first logical sector of the boot media. [...] > I don't like QEMU, it's too specific in its intended use. VMWare has > a far better quality and is much more compatible. Given a piece of code which I'm free to study and improve, for what reason should I consider surrendering those freedoms and use a non-free alternative instead? >> Also, I've tried a few images from http://cpm.z80.de/. >> Unfortunately, these seem to be in a vendor-specific CopyQM format, >> which I know of no utility (beside CopyQM) can support. >> Unfortunately, CopyQM fails with "Drive not ready" under >> FreeDOS/QEMU for me, and I'm reluctant to try it on some "real" >> hardware (with a FDD.) > You don't need to expand the compressed CopyQM image because I > already saved a decompressed (but not otherwise modified version!) > version. That's just a single image; there're quite a few more at http://cpm.z80.de/. > I don't know what you have tried, but it seems to be wrong. I used > CopyQM with VMWare and an installed MS-DOS 7.1 without any problems. I'm curious, where can one order a copy of MS-DOS 7.1 nowadays? -- FSF associate member #7257 http://sf-day.org/
[toc] | [prev] | [next] | [standalone]
| From | Peter Dassow <z80eu@arcor.de> |
|---|---|
| Date | 2012-08-17 21:47 +0200 |
| Subject | Re: running CP/M-86 images |
| Message-ID | <502e9ff0$0$6560$9b4e6d93@newsspool4.arcor-online.net> |
| In reply to | #1666 |
On 17.08.2012 21:23, Ivan Shmakov wrote: > > > Wikipedia Bullshit. There is no quality control what is published. > > If there's a reliable source supporting a different point, feel > free to edit the page accordingly. I did it a few minutes ago (see Wikipedia's "Talk" section). > My edition of Norton's "Programmer's guide to the IBM PC & PS/2" > (1988, translated in 1994) states (Chapter 5, "Boot disks" > section) that the BIOS indeed checks the boot media signature. > > > That's the proof, CP/M-86 needs no "magic numbers". > > There's no CP/M-86 until the BIOS bootloader passes control to > it. Which is unlikely to happen without having the > aforementioned signature in the first logical sector of the boot > media. You're right, it's not CP/M-86 which needs no "magic numbers", it's the BIOS itself which needs no magic numbers. Regardless of this, CP/M-86 does not need any magic number in the boot sector, but MS-DOS will check this. > > I don't know what you have tried, but it seems to be wrong. I used > > CopyQM with VMWare and an installed MS-DOS 7.1 without any problems. > > I'm curious, where can one order a copy of MS-DOS 7.1 nowadays? I guess FreeDOS will do it also, but anyway, I have still an original Windows 98 SE CD, so I can use MS-DOS 7.1 (which is included). If you want to discuss legal or illegal use of an old microsoft operating system, try another newsgroup like alt.lawyers or even alt.philosophy.law ... And btw. ... CopyQM was designed to run with MS-DOS, not FreeDOS. Regards Peter
[toc] | [prev] | [next] | [standalone]
| From | s_dubrovich@yahoo.com |
|---|---|
| Date | 2012-08-17 20:43 -0700 |
| Subject | Re: running CP/M-86 images |
| Message-ID | <89460837-679c-4daa-92ef-4426c8120099@googlegroups.com> |
| In reply to | #1668 |
On Friday, August 17, 2012 2:47:48 PM UTC-5, Peter Dassow wrote: > On 17.08.2012 21:23, Ivan Shmakov wrote: > [snip] > > > You're right, it's not CP/M-86 which needs no "magic numbers", it's the > > BIOS itself which needs no magic numbers. Regardless of this, CP/M-86 > > does not need any magic number in the boot sector, but MS-DOS will check > > this. > You forgot; the last byte is 00h for single sided, 01h for double sided media for cp/m-86 (this is an early ibm convention). Also, the hard drive controller 'option rom' -- the seagate scsi st-01 controller is one which revectors int13h thru its own int40h, and does in fact enforce the AA55h signature word, and it will not boot cp/m-86, but a Future Domain scsi controller doesn't do that and it will boot cp/m-86 from A:. IOW an option rom can modify the rom bios because the option roms are initialized later, or last, in the POST code. Of the more modern pentium class systems with floppy drives, there are a rumored few which enforce the floppy boot sector signature, but they are rare. Steve > > > > > > > I don't know what you have tried, but it seems to be wrong. I used > > > > CopyQM with VMWare and an installed MS-DOS 7.1 without any problems. > > > > > > I'm curious, where can one order a copy of MS-DOS 7.1 nowadays? > > > > I guess FreeDOS will do it also, but anyway, I have still an original > > Windows 98 SE CD, so I can use MS-DOS 7.1 (which is included). > > If you want to discuss legal or illegal use of an old microsoft > > operating system, try another newsgroup like alt.lawyers or even > > alt.philosophy.law ... > > And btw. ... CopyQM was designed to run with MS-DOS, not FreeDOS. > > > > Regards > > Peter
[toc] | [prev] | [next] | [standalone]
| From | Ivan Shmakov <oneingray@gmail.com> |
|---|---|
| Date | 2012-08-18 21:50 +0700 |
| Subject | Re: running CP/M-86 images |
| Message-ID | <864no01cfp.fsf@gray.siamics.net> |
| In reply to | #1668 |
>>>>> Peter Dassow <z80eu@arcor.de> writes:
>>>>> On 17.08.2012 21:23, Ivan Shmakov wrote:
[...]
> (see Wikipedia's "Talk" section).
ACK, I see. (Thanks to all the participants for an interesting
discussion, BTW.)
JFTR:
--cut: http://en.wikipedia.org/wiki/Talk:Boot_sector#Boot_signature --
[...] Consider for example this Microsft source [1] which makes clear
that it is the BIOS that first checks for the signature.
Crispmuncher 02:48, 18 August 2012 (UTC).
[1] http://support.microsoft.com/kb/149877
There is BIOS source code online e. g. at [2] which enforces these
rules: for a hard disk the 0xAA55 signature check is made, while
for a floppy disk the rule is that the first byte of the boot
sector must be less [greater or equal -- Matthiaspaul] than 0x06
and the first 9 words must not all be identical. While that
doesn't reveal the distribution of signature checks in modern
BIOSes, it does provide a latest date for their introduction.
Ewx (talk) 08:16, 18 August 2012 (UTC)
[2] http://www.intel-assembler.it/portale/5/ibm-at-286-souce-code/ibm-at-286-souce-code.asp
--cut: http://en.wikipedia.org/wiki/Talk:Boot_sector#Boot_signature --
[...]
>>> I don't know what you have tried, but it seems to be wrong. I used
>>> CopyQM with VMWare and an installed MS-DOS 7.1 without any
>>> problems.
>> I'm curious, where can one order a copy of MS-DOS 7.1 nowadays?
> I guess FreeDOS will do it also,
Actually, I believe it was a QEMU's fault. I'm reluctant to try
it on a "real" hardware, so I'd probably try other versions of
QEMU, or perhaps Bochs.
> but anyway, I have still an original Windows 98 SE CD, so I can use
> MS-DOS 7.1 (which is included).
While I've certainly used some version of DOS until early 1999,
I was then eager to switch to GNU/Linux, which I deem a
fortunate choice to this day.
I've never had a version of Windows installed on my computers,
and bought no license, either.
[...]
--
FSF associate member #7257 http://sf-day.org/
[toc] | [prev] | [standalone]
Back to top | Article view | comp.os.cpm
csiph-web