Slacko64

classic Classic list List threaded Threaded
76 messages Options
1234
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

01micko
Administrator
588 is up there now

See main post of this topic

Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

aarf
just going to seed this torrent from archive.org

https://archive.org/download/Puppy_Linux_Slacko64/Puppy_Linux_Slacko64_archive.torrent

all 588 files are in it but no md5sums

also straight mirror at https://ia700601.us.archive.org/30/items/Puppy_Linux_Slacko64/

not my work. seems an auto mirror as it updates fast. think its ally but not sure
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

iguleder
Downloading, my UEFI Debian installation just ruined itself when I upgraded it from stable to testing. I'm going to install Slacko64 as the main OS of my workhorse laptop, so I can play with UEFI support in woof-CE.

I think I found the recipe for UEFI support -
1) Take the MBR from isolinux - both the regular one and the "hybrid" one.
2) Create a FAT12 image with boot/EFI/bootx64.efi, a boot loader configuration file, the initramfs and a kernel.
3) Use xorriso, not mkisofs (as I did here - https://github.com/iguleder/lazyux/blob/master/create_iso.sh).

It just works great - you get one image for optical media, flash drives, BIOS and UEFI (all combinations).

I already started working on conversion of 3builddistro from the old mkisofs to xorriso for this. We'll have to add a 64-bit boot loader and the MBR files to woof-CE.
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

Iguleder
01micko, can you merge the Slacko64 changes in your fork, please? I made the changes required for basic UEFI support (in my own fork), but I cannot test them. I want to merge them to the testing branch, so we can actually test the whole thing.

I used ELILO, which is fairly limited and no longer developed - I think I we can fork it and add support for a splash screen.
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

01micko
Administrator
Iguleder wrote
01micko, can you merge the Slacko64 changes in your fork, please? I made the changes required for basic UEFI support (in my own fork), but I cannot test them. I want to merge them to the testing branch, so we can actually test the whole thing.

I used ELILO, which is fairly limited and no longer developed - I think I we can fork it and add support for a splash screen.
The changes are at https://github.com/01micko/woof-CE/tree/latest

I'm not sure that they are ready for 'testing' just yet because I haven't had the time. At Easter I may (or may not) get the time.

Can you create a new branch in your tree and merge my 'latest' branch?

Cheers.
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

jamesbond3142
In reply to this post by Iguleder
On Sat, 22 Mar 2014 03:30:36 -0700 (PDT)
"Iguleder [via woof-CE]" <[hidden email]> wrote:

>
> I used ELILO, which is fairly limited and no longer developed - I think I we
> can fork it and add support for a splash screen.
>

Why not use rEFInd? It has good set of features and is actively supported.
It's not as small as ELILO but at least you *don't* have to maintain another fork.

cheers!
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

Iguleder
ELILO is simpler and easier to build.

Working on a rEFInd build script ;)
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

Karl Godt
Administrator
In reply to this post by 01micko
Have It running ATM

Compiled :

libVTE + sakura

Xosview

Busybox static ash-standalone  wo PAM, Selinux, inetd, ipcalc, nslookup

# ls -s ./busybox
2472 ./busybox

is almost 1MB bigger than in common 32bit .

# ./busybox
BusyBox v1.23.0.git (2014-06-03 23:37:26 GMT-8) multi-call binary.


Reply | Threaded
Open this post in threaded view
|

Re: Slacko64 5.8.8 k3.13.1 # CONFIG_IA32_EMULATION is not set

Karl Godt
Administrator
In reply to this post by 01micko
According to
arch/x86/Kconfig :

config IA32_EMULATION
	bool "IA32 Emulation"
	depends on X86_64
	select BINFMT_ELF
	select COMPAT_BINFMT_ELF
	select HAVE_UID16
	---help---
	  Include code to run legacy 32-bit programs under a
	  64-bit kernel. You should likely turn this on, unless you're
	  100% sure that you don't have any 32-bit programs left.

config IA32_AOUT
	tristate "IA32 a.out support"
	depends on IA32_EMULATION
	---help---
	  Support old a.out binaries in the 32bit emulation.

config X86_X32
	bool "x32 ABI for 64-bit mode"
	depends on X86_64 && IA32_EMULATION
	---help---
	  Include code to run binaries for the x32 native 32-bit ABI
	  for 64-bit processors.  An x32 process gets access to the
	  full 64-bit register file and wide data path while leaving
	  pointers at 32 bits for smaller memory footprint.

	  You will need a recent binutils (2.22 or later) with
	  elf32_x86_64 support enabled to compile a kernel with this
	  option set.

I remember a LH-Pup 64bit that was missing the chat binary .
I just took the 32bit one I could find from some
Puppy and was able to connect .

Please enable IA32_EMULATION whenever possible !!
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64 5.8.8 k3.13.1 # CONFIG_IA32_EMULATION is not set

01micko
Administrator
Karl Godt wrote
I remember a LH-Pup 64bit that was missing the chat binary .
I just took the 32bit one I could find from some
Puppy and was able to connect .

Please enable IA32_EMULATION whenever possible !!
Is enabled now and kernel built. I hope to get some time and bring 64 up to scratch.
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64 5.8.8 k3.13.1 # CONFIG_IA32_EMULATION is not set

01micko
Administrator
Just fired up woof.

Hopefully by tomorrow we get a usable slacko64-6.0 alpha.

Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

Will McDade
In reply to this post by 01micko
I like this! I always loved the Puppies but never really looked at the 64 bit ones. I have one problem though. GCC doesn't seem to work.

checking for a BSD-compatible install... /usr/bin/ginstall -c
checking whether build environment is sane... yes
checking for a thread-safe mkdir -p... ./install-sh -c -d
checking for gawk... gawk
checking whether make sets $(MAKE)... no
checking whether to enable maintainer-specific portions of Makefiles... no
checking for style of include used by make... none
checking for gcc... gcc
checking whether the C compiler works... no
configure: error: in `/tmp/SBo/libgnomecanvas-2.30.3':
configure: error: C compiler cannot create executables


(Trying to install Steam)
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

q5sys
Anyone have any luck in getting fluxbox to run in this?  I installed it from the slackware repo along with imlib2 which is a dep... but it doesnt seem to want to work.

Never had a problem starting it before in other Puppies.   Tried all the past methods:
exec startflubox (just returns me to the prompt with no error)
startx fluxbox (X crashes and it gives the standard xorgwizard response)
xwin fluxbox [after editing the xinitrc] (same as startx fluxbox)

Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

Karl Godt
Administrator
In reply to this post by 01micko
Slacko64-5.8.8 full installed to btrfs formatted partition .

Using an initrd.ext2 image on a separate ext4 formatted partition because grub and grub4dos are unable to access btrfs directly
with added file-system drivers and using DEV1PUP variable as PUPMODE 2 indicator in /init
I was able to boot it !
I took the PUPMODE 3 approach of UMOUNTRO="/pup_ro1" which worked , but I guess I should have used "/pup_rw" .
 2)  CREATEFOLDERS="$CREATEFOLDERS"
     CREATEPDEV1="/pup_ro1"
     CREATEPUPSAVE2FS="$CREATEPUPSAVE2FS"
     CREATEPUPXXXSFS="$CREATEPUPXXXSFS"
     CREATETMPFS="$CREATETMPFS";
     OLDFILESMNTPT="/pup_ro1";
     NEWFILESMNTPT="/pup_ro1";
     #unionlayer /mountpoint=mountoptions
     # traditional
     UMNTMAIN="/pup_ro1=rw"
     # new
     UMNTRO0="/pup_ro1=rw"
     # risky
     UPUP_SAVE="/pup_ro1=rw"
     ;;
 3)  CREATEFOLDERS="$CREATEFOLDERS"
     CREATEPDEV1="/pup_ro1"
     CREATEPUPSAVE2FS="$CREATEPUPSAVE2FS"
     CREATEPUPXXXSFS="$CREATEPUPXXXSFS"
     CREATETMPFS="/pup_rw";
     OLDFILESMNTPT="/pup_ro1";
     NEWFILESMNTPT="/pup_ro1";
     #unionlayer /mountpoint=mountoptions
     # traditional
     UMNTMAIN="/pup_rw=rw:/pup_ro1=ro+wh"
     # new
     UMNTRW="/pup_rw=rw:";
     UMNTRO0="/pup_ro1=ro+wh:"
     # risky
     UL_SAVE="/pup_ro1=ro+wh"
     UL_RW="/pup_rw=rw"
     ;;
Nevertheless worked .

Problems :
1) initrd does not like compressed modules . Failed to load squashfs and aufs drivers . Mount error handling could need some improvements then .
   Slacko64 has also depmod and that is fast on few modules instead of incompatible or missing modules.dep file (tahrpuppy-6.0) .
2) Slacko64 installer installed /lib but not /lib64 , so chroot failed for /sbin/init /bin/bash dynamically linked did not find /lib64/ld* and libc* ..

Compiled kernel 3.13.11 because I had it there from git .
Enabled IA32-Emulation.
With btrfs , xfs and jfs file-systems inside the kernel .
Git repo was on jfs formatted partition .

Now with IA32EMULATION I am able to chroot into 32-Bit full installs .

Not everything 32-Bit works .
My current internet connection is by cdc-ether.ko usb-modem driver : Prepaid by provider O2/Telefonica .
It has a pre-compiled binary named "mbbservice" and an installer that works ok in usual Puppies despite some missing commands .
It is from year 2008 by an employee of Huawei .
Cannot find source or 64-Bit binary .
That first said not found .
Then linking /lib/ld* libc* libm* libpthread* into /lib64/* it says segmentation fault .
 
So am posting from chroot'ed ifconfig eth1 up dhcpcd -d -I '' eth1 'ed firefox now .


jfs driver in my 3.13.11 has BUG for whatever cause does not mount jfs partition .
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64 rdev is x86_i386 only

Karl Godt
Administrator
rdev manpage says :

BUGS
       The rdev utility, when used other than to find a name for  the  current
       root  device,  is an ancient hack that works by patching a kernel image
       at a magic offset with magic numbers. It does not work on architectures
       other  than  i386.   Its use is strongly discouraged.

busybox rdev seems not to work on btrfs full install . Have not tested other file systems .
Output is "" like in frugal mode 12 or 13 .

df script to put the right icon on running OS partition :
#!/bin/ash
#(c) Barry Kauler 2009, licence GPL2
#w482 fix for /dev/root.

DF="busybox df"
[ "`which df-FULL`" != "" ] && DF="df-FULL"

#150420: df only works on entries in /proc/mounts
test -e /proc/mounts || busybox mount -t proc proc /proc >&2
#150420 Slacko64: manpage of rdev claims rdev only works on i386 arch ,
# but df-FULL now shows real root device when root=/dev/sdaX sdaX is mounted on another mountpoint once temporarily

_force_root_mount(){
    tMNTPT=$((RANDOM));
    mkdir "/etc/$tMNTPT" || return 1
    mount-FULL $ROOTD "/etc/$tMNTPT" || { rmdir "/etc/$tMNTPT"; return 2; }
    #(
    sleep 1 #&&
    busybox umount -lr "/etc/$tMNTPT"
    sleep 1 #&&
    rmdir "/etc/$tMNTPT"
    #) #&
}

ROOTD=`grep -o 'root=[^[:blank:]]\+' /proc/cmdline | cut -f2- -d'='`
if test -b "$ROOTD"; then

#grep $Q "^$ROOTD " /proc/mounts >&2 || { (
#    tMNTPT=$((RANDOM));
#    mkdir "/etc/$tMNTPT";
#    mount-FULL -o ro $ROOTD "/etc/$tMNTPT"
#    ( sleep 1 && umount -lr "/etc/$tMNTPT" && sleep 3 && rmdir "/etc/$tMNTPT" ) &
#    ) >&2; }

grep $Q "^$ROOTD " /proc/mounts >&2 || _force_root_mount >&2
else true
fi

RETSTUFF="`$DF "$@"`" #quotes added
RETVAL=$?

RETSTUFF="`echo "$RETSTUFF" | grep -v '^rootfs '`"

__old_real_rootdevice__(){
#replace /dev/root with correct root partition...
#rdev is a busybox applet...
ROOTPARTITION="`rdev | grep ' /$' | cut -f 1 -d ' ' | grep -E '/dev/sd|/dev/hd|/dev/mmc'`"
nPATTERN="s%^/dev/root %${ROOTPARTITION} %"
[ "$ROOTPARTITION" ] && RETSTUFF="`echo "$RETSTUFF" | sed -e "$nPATTERN"`"
}

echo "$RETSTUFF"
exit $RETVAL

###END###

Forcing the mount of OS partition grep'ed from /proc/cmdline root= parameter another time temporarily somehow triggers something into the kernel,
that makes the kernel change the /dev/root entry in /proc/mounts to the rdev real root_device .
mount -o ro does error if already mounted on '/' rw .

df needs /proc/mounts and shows /proc/mounts entries according to parameter switches ( -a | -T )

# cat /proc/mounts
rootfs / rootfs rw 0 0
/dev/sda10 / btrfs rw,relatime,space_cache 0 0
devtmpfs /dev devtmpfs rw,relatime,size=2021512k,nr_inodes=505378,mode=755 0 0
proc /proc proc rw,relatime 0 0
tmpfs /tmp tmpfs rw,relatime,size=0k 0 0
none /sys sysfs rw,relatime 0 0
none /dev/pts devpts rw,relatime,gid=2,mode=620 0 0
shmfs /dev/shm tmpfs rw,relatime,size=1004420k 0 0
fusectl /sys/fs/fuse/connections fusectl rw,relatime 0 0
# df-FULL -T
Filesystem     Type     1K-blocks     Used Available Use% Mounted on
rootfs         rootfs    15998976  2689036  11373580  20% /
/dev/sda10     btrfs     15998976  2689036  11373580  20% /
devtmpfs       devtmpfs   2021512        0   2021512   0% /dev
shmfs          tmpfs      1004420        0   1004420   0% /dev/shm
/dev/sda6      ext3      15616412 11329092   3487372  77% /mnt/sda6
/dev/sda1      ext3       9434456  7666140   1282404  86% /mnt/sda1
devtmpfs       devtmpfs   2021512        0   2021512   0% /usr/local/MobileBrServ/dev
devtmpfs       devtmpfs   2021512        0   2021512   0% /mnt/sda6/dev
/dev/sda2      ext4      14909436  6809116   7319904  49% /mnt/sda2
/dev/loop0     ext2         12887     9581      2641  79% /mnt/initrd.ext2.27040
Reply | Threaded
Open this post in threaded view
|

Re: Slacko64

Karl Godt
Administrator
In reply to this post by Karl Godt
PUPMODE 2 using initrd on "/pup_rw" : Works better than using "/pup_ro1" .

Shutdown shows behaviour to kill itself somewhere in rc.shutdown .
Puppy Linux PUPMODE 2 using initrd
Attached screenshot shows content of /proc/mounts, /etc/rc.d/PUPSTATE and /proc/cmdline .


Have implemented the pupmode=N kernel parameter to override auto-detection of PUPMODE in /init to force pupmode=3 on HDD full installation .
This works without problems .
1234