AMIX installation made easy...no tape drive!

[Failure]

Well, I'm probably more pleased with myself than I really should be but...I have installed AMIX 2.01 without a tape drive and without leaving the install script.  I used a variation of a method I found in a Usenet post, using dd to copy the install tape "image" to a hard drive.  Here is the short, short version (it's late):

* I made a big cpio archive of all the individual cpio files on the tape
* I dd'd this file using the old 1.1 installation (could have used x86 Linux, or AmigaOS...AmigaOS has dd available?) onto a spare HD
* Modified the 2.03 root floppy install script to skip the tape drive, and un-cpio my kludge instead
* Booted the test A3000 into installation with the pseudo tape drive attached externally...went off without a hitch.
* Rebooted into AMIX 2.01 without having done any crap on the CLI to install the kernel, etc, that I noticed had to be done in all these other kludges people came up with...

I actually installed 2.01 by mistake, I'll redo the big ol' cpio using 2.03 files.  Should work identically.  It's a nasty kludge, what I did, but it works and since most people won't have the A3070 tape drive, it's a nice solution.  Should be able to be done using only AmigaOS to prep for it.

Anyway I'll document more on it in the next day or so, I was just so happy I wanted to post something right away ;-)

Here's some pics just for fun:

The rig, ugly but works...I used the A3070 case to mount the HD inside
My kludge working, I had it echo filenames so I could tell if it hung
Installation completed, time to reboot!
And here it is, with a shot of some of the post-install config

[Failure]

I was planning on doing a guide on getting AMIX installed via this method, I want to clean up one or two things in the script first and get 2.03 installed.  Mostly I want to get rid of a couple error messages that might be alarming to new users, since they are unimportant with the new installation method.

I've been going through the 2.03 archives and, unfortunately, a number of the cpio files are corrupt :-(  These are the sysadm package, emacs, X extras, and all the source code except the Amiga-specific.  I plan on using the 2.01 versions of all of these, since there should only be minor changes and the source code (except Amiga-specific!) is likely of dubious importance anyway since it's really, really old.  Everything else seems to be good.

I can't find 2.1 media anywhere; 2.1c is, I believe, the last release of AMIX ever (obtained by installing base 2.1, followed by the patch disk to 2.1c).  I have come across references of people downloading it off the internet, but I certainly can't find it now.

2.01 is already way way better than 1.1.  For example, "w" works properly :-)

[Failure]

Hmm.  I'm not sure what you mean.  I'm using unzip on Linux to uncompress the archive, AMIX203.ZIP, which contains a bunch of cpio files (i_01.cpio.gz - i_27.cpio.gz) and several of these are bad when I attempt to extract them with cpio.  After uncompressing them of course.

Do you maybe have a different archive, with maybe files that work?  :-D

*edit* the files I have are the ones from the mmhart.com site

[Failure]

I fixed the error messages in the install script, and installed 2.03 last night, so I wrote up how to do it:

AMIX installation instructions

The only thing I wasn't sure of was how to write the cpio file to the drive in AmigaOS, hopefully someone here can help with that.  From reading the install script, it seems that AMIX can also run on a machine with 1.3 ROMs and 2620/2630 card.

Best of luck to those that try this!

As a side note, when the 3000UX with the large HD failed to boot (described in the instructions) it came up with a weird boot screen, it said "Version 1.4 Beta, do not distribute" with the animated floppy drive.  I had never seen that before.

[Failure]

I've been busy :-)

I made a new version of the install disk that allows you to select the SCSI ID and partition to get the install media from.  Now it is possible to install using only one drive!  I haven't updated the documentation yet, but it's pretty easy, basically make a partition (I recommend at the front of the drive) as FFS but do not format it.  AMIX ignores AmigaOS partitions, so it won't get zapped.  dd the cpio archive to the partition, reboot using the AMIX floppies, and tell the installer where you put it...and off it goes.  When the install finishes, you can wipe out that partition and install AmigaOS on it if you like, or use it for UNIX.  I also had the script copy cpio to the hard disk and use that for extraction, as constantly going back to the floppy for it slows the install down.

I stuck 3.1 ROMs in the 3000UX and AMIX still booted fine, so that works.  The 3000 the ROMs came out of is apart for soldering...

And LocalH (pack up the cats!) is right, the window gadgets were...weird for that 1.4 beta.  I dumped a ROM image for posterity.

Finally, I did locate dd for AmigaOS on Aminet, as dd.lzh.  So it should definitely be possible now to install AMIX using only AmigaOS to prep for it.

I'll update the docs time permitting.  I'd love to hear about someone doing this with just AmigaOS.

[Failure]

Congratulations Dalamar, I am very happy that this can be accomplished with AmigaOS :-)

Someone had requested pics of X running...getting xdm going was tricky/archaic.  Once I figured out what was wrong, finding the solution didn't take too terribly long.  But look at this user-friendly way to enable xdm:

pmadm -d -p screens -s con10
sacadm -a -p xdm -t xdm -v1 -c /usr/X/bin/xdm

pmadm is "port monitor" admin tool...I found a lot of posts from bitter Solaris users about this gem.  sacadm, couldn't tell you.  In addition, the /usr/X/lib/xdm/Xsession file was not set executable, so logins would give you a crappy xterm and no window management.  Making the file executable got twm going, but without the never-gonna-find-one-A2410 card the display is so small, and in 1-bit color, so that the experience is fairly "meh".

Behold the xdm login screen in all its 1-bit glory:

Ta-daah!

Yeah not so exciting.  But this is X, so we don't have to be limited by that video hardware since we have the network.

xdm in 24 bit color!

Well, that's only really 4 colors but it *could* be displaying millions if it really wanted to!  That's an X server on a Linux box talking to the xdm on the Amiga.  I was actually surprised how responsive this was.  For graphical applications (to use the term loosely, I'm talking about xeyes here) the network was faster at rendering than on the Amiga locally.

X session over the network

For some reason applications that generate color do not display at all, but color is working since I set the background color in AMIX and you can see the windows are a different color.  I think it might work if I run at 8-bit depth.  I had similar problems with old Solaris apps.

Fun stuff.  You have to respect open standards like X11.  These two machines talking to each other, it would be like Windows 3.11 displaying applications natively on a Windows Server 2003 machine, in terms of the age difference.

[Failure]

Yeah I am trying to go about that now.  I had to get the screen utility first so that I can log out during the few days this process is likely to take ;-)  Screen didn't compile under 1.1 but it did fine with 2.03, happily.  There is actually one included with AMIX but it doesn't work, at least from a telnet session.

I figure on attempting gcc 2.95-ish since that is widely regarded as "stable", and if that works maybe try the 3.x.x.  I wonder though if I have enough HD space, I might need to reinstall on that 4GB drive with the 1GB partition that you found works.

Matt_H gave me some more info to chew on regarding installation from AmigaOS, including a possibly easier dd alternative.  Maybe he will chime in later :-)

For now I am considering creating some kind of mechanism on the AMIX site to prevent duplication of effort on compiling things, since it takes such a long time on these machines.  Not sure what to do yet.  But I made a new file gallery and stuck screen in there, pre-compiled.

*edit* it goes without saying, but really don't forget to strip the binaries...it shaved 1.4MB off of screen, which goes a long way with 8MB memory!

[Failure]

gcc 3.4 series will be the last that possibly compiles on AMIX:

http://gcc.gnu.org/gcc-3.4/changes.html#obsolete_systems

:-(

Struggling to get 2.95 compiled; the version of make with AMIX segfaults and dumps core so I am trying to get a newer version of make compiled.  It's like a kind of version hop-scotch here.

*edit* got GNU make 3.70 to compile (none newer I tried would compile), and it handles the gcc makefile...it's on its way...

*edit2* gcc 2.4.5 compiled, is now compiling itself again :-)

[Failure]

Hmm.  I dunno, my time stays correct between reboots, but I don't turn the thing off (My 3000UX has no battery).  However, if it *is* the setclk utility, take a look in /usr/amiga/src/cmd/setclk  *edit* setclk appears to be the equivalent of hwclock in Linux (ie, hwclock --systohc), so it's certainly the cause of the problem.

I get the "bad date conversion" but again, my time stays correct.  I will reboot the machine in just a minute because now I am curious if it shifts at all.

I got gcc 2.4.5 to compile itself, so I uploaded it as a cpio package to my AMIX site.  Next is 2.95 which I will let go overnight, I don't hold much hope for it compiling all the way but it should at least get farther than it did with...uh...gcc 1.4.0 or whatever it is.  It's so old, the version number is faded.

*badum-ching!*

Thank you, I'll be here all week.  Try the tuna.

[Failure]

Argh.  I appear to have hit a roadblock compiling gcc 2.95: running into a memory limit.  I only have 8 meg and 30 meg swap, and for some reason it fails on this step:

gcc  -DIN_GCC     -g -O2  -DHAVE_CONFIG_H  -o cc1 toplev.o version.o tree.o print-tree.o stor-layout.o fold-const.o function.o stmt.o except.o expr.o calls.o expmed.o explow.o optabs.o intl.o varasm.o rtl.o print-rtl.o rtlanal.o emit-rtl.o genrtl.o real.o dbxout.o sdbout.o dwarfout.o dwarf2out.o xcoffout.o bitmap.o alias.o gcse.o integrate.o jump.o cse.o loop.o unroll.o flow.o stupid.o combine.o varray.o regclass.o regmove.o local-alloc.o global.o reload.o reload1.o caller-save.o insn-peep.o reorg.o sched.o final.o recog.o reg-stack.o insn-opinit.o insn-recog.o insn-extract.o insn-output.o insn-emit.o lcm.o profile.o insn-attrtab.o m68k.o getpwd.o  convert.o mbchar.o dyn-string.o splay-tree.o graph.o sbitmap.o resource.o hash.o c-parse.o c-lang.o c-lex.o c-pragma.o c-decl.o c-typeck.o c-convert.o c-aux-info.o c-common.o c-iterate.o  obstack.o        ../libiberty/libiberty.a
ld: cc1: libelf error:  Memory error: output file space elf_update:
make[1]: *** [cc1] Error 1
make[1]: Leaving directory `/usr/src/gcc-2.95.3/gcc'
make: *** [all-gcc] Error 1

Uh...heh.  Yeah, so I am trying to figure out how to remove the limits (viewable by ulimit -a) so this will work.  This is a bit dangerous since I have such limited memory.  I'd like to buy more ZIPs but they're not on pricewatch :-)  Well if I crash it doing this, is anybody running with 16MB or > 30MB swap?

And I need to reinstall on a bigger drive, I'm running out of HD space too.

*snif* :-(

The good news is, gcc 2.95 is more or less compiled!

*edit* hmm probably need a new thread for this, since it's nothing to do with installation

[Failure]

I have not, but I am planning on doing that as soon as I get as far up the compiler version chain as I can...it's not a trivial process (crossing both arch *and* OS), so I want to avoid doing it more than once.  Reading the docs to do it with 2.4.5 makes me sweat a little but I think I could do it, although I don't want to.  I'd like one of the guys with 16MB to try it first!

If it continues to not work, I'm not sure whether to just stop at 2.4.5, try something less than 2.95.3, or go through the hassle of the cross compiler.

From this post, it clearly continues to fail even after I removed all system limits...

[Failure]

Yeah I'll do that if my next attempt doesn't work.  I noticed the ld that the system was using isn't GNU, and is also in excess of 14 years old.  So I am compiling the latest GNU binutils, and will try with that...if I am right, it would have failed with 16MB anyway.

[Failure]

It might be useful to note that AMIX doesn't depend on ROM 1.4, but works fine even on 3.1 ROMs.  Not that I'm trying to say getting 1.4 working in WinUAE is pointless, just that it isn't necessary for AMIX.

I played around a little with the UAE MMU patch, but wasn't able to do anything with it.  But I am a really poor programmer so I am pretty sure the hacks I was doing to get the MMU to "work" on 68020 were completely bogus.  IIRC, the MMU in that patch is for 68040 which is simpler than the MMU in the 68030 although I do not know the specifics of it.  There is a thread I started around here somewhere on that topic.

Back to my gcc 2.95.3 problems, GNU binutils won't compile since there is no m68k-cbm-sysv4 target for ld or gas.  So I gave up on that.  Further Usenet searching reveals that the problem really might just be memory, like I originally thought, so I'm gonna put the partially compiled sources online and hopefully Dalamar (with perhaps more than 30MB swap) will have better luck.

I am still having a good time but I wish they made a 300MHz 68030 :-)

[Failure]

After a bit of futzing around, I am rewarded with this:

# ./xgcc -v
Using builtin specs.
gcc version 2.95.3 20010315 (release)

Still not done compiling all the other stuff, 25MHz is slooooow for all this code.  One of the unfortunate aspects of working with AMIX is these lengthy compiles...I start the compilation in a screen session, log out, come back some hours later and it has errored out on something.  On the bright side, I replaced the HD in the machine when I reinstalled, so I have 100MB swap and everything is faster since the HD is a better one.  In particular, multitasking was helped.  It used to take several seconds to get a login via telnet if the machine was compiling, now it is very fast.

This compile is using m68k-cbm-sysv4 as a target and at least for gcc, it works fine.

Hey I know what I need!  distcc!  :-)

[Failure]

I have just finished a rewrite of the installation documentation incorporating the new information re: installing using AmigaOS to prepare (thanks to Matt and Dal), along with installation using a single HD and more detailed information regarding pitfalls in the installation process.

I am really tired so I probably made some mistakes or weird statements, it's a wiki so feel free to fix them!

http://amix.failsure.net/tiki-index.php?page=dd+Install

Enjoy and as usual, best of luck to anybody who gives this a shot.

Brief status update on gcc.  My AMIX box has been compiling for about 5 straight days.  gcc-2.95.3 is going to take more work, so I am jumping from version to version up from 2.4.5 until it breaks.  I am compiling 2.7.2.3 now, 2.5.8 and 2.6.3 compiled fine as well.