Get yourselves a grey hound
Saucy EmCE.iso
145mb iso !
stripped a lot of stuff out, quite minimal, too minimal for novices perhaps. Good for use as a building block.
kernel-3.15-rc6
oops!
https://drive.google.com/file/d/0B4GhZV ... edit?usp=1
stemsee
Saucy Salamander CE with kernel-3.15.0-rc6
-
stemsee
Saucy Salamander CE with kernel-3.15.0-rc6
Last edited by stemsee on Fri 27 Jun 2014, 09:55, edited 3 times in total.
mirrored here: http://archive.org/details/Puppy_Linux_Saucy
well done stemsee, uploading to the archive, will take her for a spin in a while
well done stemsee, uploading to the archive, will take her for a spin in a while
Wot, no terminal?!!
If desktop icons are unused, I think they should have been removed.
No sound either.
Can't find firefox yet either. Apart from that, running quite well from USB on a 2ghz Pentium lappie from 2005.
If desktop icons are unused, I think they should have been removed.
No sound either.
Can't find firefox yet either. Apart from that, running quite well from USB on a 2ghz Pentium lappie from 2005.
- Attachments
-
- rxvt.png
- (10.06 KiB) Downloaded 1637 times
-
stemsee
i told you
aha
use ppm to install one you like, lxterminal for example, or you can load as sfs your favourite puppy base or main .sfs trry it!!!
use ppm to install one you like, lxterminal for example, or you can load as sfs your favourite puppy base or main .sfs trry it!!!
Re: NEW ! ! Saucy Salamandar CE KERNEL 3.15.0-rc6 !
Shocking! It's up and running, kernel-3.15-rc6 and all on a pre-Dothan Pentium M. Those had PAE, but the flag was disabled. For the first time, I used forcepae as a kernel parameter in Grub4Dos...stemsee wrote:Get yourselves a grey hound
Saucy EmCE.iso
145mb iso !
stripped a lot of stuff out, quite minimal, too minimal for novices perhaps. Good for use as a building block.
kernel-3.15-rc6
firefox
https://drive.google.com/file/d/0B4GhZV ... edit?usp=1
stemsee
and up it came on...title Experimental currently saucy
find --set-root --ignore-floppies --ignore-cd /Experimental/initrd.gz
kernel /Experimental/vmlinuz forcepae psubdir=Experimental pmedia=ataflash pfix=copy
ipv6.disable=1
initrd /Experimental/initrd.gz
Wireless, video resolution flawless. Posting from it (Opera portable) now. A big leap for this old lappie.-Processor-
Name : Intel(R) Pentium(R) M processor 1600MHz
Family, model, stepping : 6, 9, 5 (Pentium M)
Vendor : Intel
More later on sound etc.
Edited: Processor name corrected. Thanks Keef.
Last edited by Marv on Sat 31 May 2014, 15:56, edited 1 time in total.
Pups currently in kennel :D Older LxPupSc and X-slacko-4.4 for my users; LxPupSc, LxPupSc64 and upupEF for me. All good pups indeed, and all running savefiles for look'n'feel only. Browsers, etc. solely from SFS.
-
stemsee
try this
If you got slacko.sfs or puppy_precise_5.7.1.sfs load it as an sfs load on the fly. that is the idea. That way you get a new kerne libs etc and your favourite old base system.
Happy for you Marv.
If you don't load some other base.sfs then use ppm to reinstall alsa base and alsa-common and deps. Or i can share a savefile with it already in there, what i did today. but it wouldn't remaster properly!!
ALLY
please rename the iso with EmCE in there somewhere! cheers!!
Happy for you Marv.
If you don't load some other base.sfs then use ppm to reinstall alsa base and alsa-common and deps. Or i can share a savefile with it already in there, what i did today. but it wouldn't remaster properly!!
ALLY
please rename the iso with EmCE in there somewhere! cheers!!
-
LateAdopter
- Posts: 361
- Joined: Fri 27 May 2011, 17:21
- Location: Reading UK
Hello stemsee
It's amazing: you can breed puppies faster than I can keep up! But I've tried this one.
What I have found:
It boots to a desktop
The Radeon driver is missing so it says its using ATI but the X.org log looks like vesa.
Pfind in not included, and I'm lost without that. So I can't say how much of the Radeon drive is missing.
My AMD RS880G chipset can operate in UMS or KMS mode, so it will almost always work. But for newer AMD graphics KMS is required.
I tried what you said, and loaded the base SFS from Pemasu's Raring 3.9.9.2, at startup, being the nearest Ubuntu puppy.
When I booted it obviously found the Radeon driver, from there, because it modeswitched during the text mode so it had loaded the KMS mode. But then X.org would not start. I would think a "version mismatch" error, between the radeon kernel module and the radeon driver from Raring.
One other problem is an error message while booting after saucy has been run: "Drive was not properly unmounted"
That was in the scrolling onscreen messages. Win2k complains too.
I expect it is a Woof CE bug, because 01micko's Slacko64 does that as well.
Note: I am running from a FAT32 partition.
Thanks for your efforts. Very interesting.
EDIT:
After some investigation...
The X.org radeon driver won't load because libglamor in missing
The radeon kernel module won't load because the firmware r600_rlc.bin is missing.
There are only 27 items in /lib/firmware/radeon compared with 59 in Pemasu's Precise.
I'll fix that and try again
EDIT 2
Those two items fixed it. So now it loads the correct radeon driver with KMS and 2D acceleration working.
AIGLX 3D acceleration fails, as usual, with the spurious dependency on LLVM. But you don't want that because its HUGE and can easily be loaded from the package manager.
Thanks again.
It's amazing: you can breed puppies faster than I can keep up! But I've tried this one.
What I have found:
It boots to a desktop
The Radeon driver is missing so it says its using ATI but the X.org log looks like vesa.
Pfind in not included, and I'm lost without that. So I can't say how much of the Radeon drive is missing.
My AMD RS880G chipset can operate in UMS or KMS mode, so it will almost always work. But for newer AMD graphics KMS is required.
I tried what you said, and loaded the base SFS from Pemasu's Raring 3.9.9.2, at startup, being the nearest Ubuntu puppy.
When I booted it obviously found the Radeon driver, from there, because it modeswitched during the text mode so it had loaded the KMS mode. But then X.org would not start. I would think a "version mismatch" error, between the radeon kernel module and the radeon driver from Raring.
One other problem is an error message while booting after saucy has been run: "Drive was not properly unmounted"
That was in the scrolling onscreen messages. Win2k complains too.
I expect it is a Woof CE bug, because 01micko's Slacko64 does that as well.
Note: I am running from a FAT32 partition.
Thanks for your efforts. Very interesting.
EDIT:
After some investigation...
The X.org radeon driver won't load because libglamor in missing
The radeon kernel module won't load because the firmware r600_rlc.bin is missing.
There are only 27 items in /lib/firmware/radeon compared with 59 in Pemasu's Precise.
I'll fix that and try again
EDIT 2
Those two items fixed it. So now it loads the correct radeon driver with KMS and 2D acceleration working.
AIGLX 3D acceleration fails, as usual, with the spurious dependency on LLVM. But you don't want that because its HUGE and can easily be loaded from the package manager.
Thanks again.
Last edited by LateAdopter on Wed 28 May 2014, 14:02, edited 2 times in total.
Thanks. This old dog learned a new trick or two from this pup already. By the way, in the process I unsquashed the sfs and when resquashing used the alternate compression parameters (still xz) in dir2sfs and the resultant sfs goes from 139 Mb to 103 Mb. The relevant lines showing the parameters are 29 to 34 or so in the attached dir2sfs.stemsee wrote:
Anyway I just recompiled the same kernel and I will post a remaster shortly, with functioning sound and firefox and a merlin mixer and unused desktop icons gone.
- Attachments
-
- dir2sfs.gz
- (671 Bytes) Downloaded 377 times
Pups currently in kennel :D Older LxPupSc and X-slacko-4.4 for my users; LxPupSc, LxPupSc64 and upupEF for me. All good pups indeed, and all running savefiles for look'n'feel only. Browsers, etc. solely from SFS.
Regarding the size this command will make it even smaller than the one from dir2sfs ( I guess under 100Mb):Marv wrote:...By the way, in the process I unsquashed the sfs and when resquashing used the alternate compression parameters (still xz) in dir2sfs and the resultant sfs goes from 139 Mb to 103 Mb. The relevant lines showing the parameters are 29 to 34 or so in the attached dir2sfs.
Code: Select all
mksquashfs /path-to-dir /path-to/some-name.sfs -b 1048576 -comp xz -Xbcj x86Code: Select all
"-comp xz -b 1048576 -Xbcj x86"Hello, stemsee.musher0 wrote:The computer just froze. No keyboard access.
I had to reboot manually by pushing the boot button on the computer face.
I tried again, this time with all files at the same level.*
Same thing, it freezes at "loading kernel modules".
I am attaching a hard info report, if it can be of any help. As an afterthought, would
your PuppySaucy be allergic to certain grub4dos settings?
Best regards.
musher0
~~~~~~~~~~~
* See my folder listing. The first time, initrd and vmlinuz were in
/mnt/sda9/PuppySaucy-5.8/launch and the sfs was in /mnt/sda9/PuppySaucy-5.8.
- Attachments
-
- my_PuppySaucy-5.8_dir.txt.zip
- (446 Bytes) Downloaded 395 times
-
- hardinfo_report.html.zip
- (5.7 KiB) Downloaded 318 times
musher0
~~~~~~~~~~
"You want it darker? We kill the flame." (L. Cohen)
~~~~~~~~~~
"You want it darker? We kill the flame." (L. Cohen)
-
stemsee
WoW !
Excellent feedback everyone.
I am going to try these optimisations from marv and saintless!
I rebuilt the thing early this morning with a recompiled kernel for Highmem=4gb, so now the kernel sees 3GB instead of 1GB yesterday !! Even though pae is set it doesn't seem to work on this bleeding edge kernel! But it doesn't make any difference to the wonderful performance it brings to my pup.
Any missing libs apps firmware can be downloaded from the fully functional PPM (puppy package manager).
I am trying an experiment that came after using debiandog, ie putting a *.squashfs file in the live directory gets it loaded at boot time. I imagine it uses a wildcard * with extension .squashfs to find such files. If my experiment works then musher0 wil be able to prepare an sfs from some other pup with whatever is missing at the moment and it will load with the base.sfs and any other sfs in the root folder.
I still have to strip much more stuff away though --- if only I knew which files were needed to boot to gui and load an sfs and ppm, that would be all that is needed. Everything else would be brought to the party by the guest base.sfs
I actually build about two/day but i don't post iso experiments which i think are not useful generally. But I will persue this new idea, because i need it. Therefore i tend not to maintain my own distros, it is more interesting to build new ones. But this one is interesting!
I am going to try these optimisations from marv and saintless!
I rebuilt the thing early this morning with a recompiled kernel for Highmem=4gb, so now the kernel sees 3GB instead of 1GB yesterday !! Even though pae is set it doesn't seem to work on this bleeding edge kernel! But it doesn't make any difference to the wonderful performance it brings to my pup.
Any missing libs apps firmware can be downloaded from the fully functional PPM (puppy package manager).
I am trying an experiment that came after using debiandog, ie putting a *.squashfs file in the live directory gets it loaded at boot time. I imagine it uses a wildcard * with extension .squashfs to find such files. If my experiment works then musher0 wil be able to prepare an sfs from some other pup with whatever is missing at the moment and it will load with the base.sfs and any other sfs in the root folder.
I still have to strip much more stuff away though --- if only I knew which files were needed to boot to gui and load an sfs and ppm, that would be all that is needed. Everything else would be brought to the party by the guest base.sfs
I actually build about two/day but i don't post iso experiments which i think are not useful generally. But I will persue this new idea, because i need it. Therefore i tend not to maintain my own distros, it is more interesting to build new ones. But this one is interesting!
Stemsee, you might want to look here:
http://murga-linux.com/puppy/viewtopic. ... 921#764921
I had the same idea but Puppy init modification is too complicated for me yet. The opposite approach works with any separate kernel module taken from DebianDog but it is not real Puppy after you boot it this way. You will find many useful instructions from Mikeb in the above link.
Puppy already has option to load one extra sfs on boot in case you give the right name:
http://murga-linux.com/puppy/viewtopic. ... 837#779837
http://murga-linux.com/puppy/viewtopic. ... 921#764921
I had the same idea but Puppy init modification is too complicated for me yet. The opposite approach works with any separate kernel module taken from DebianDog but it is not real Puppy after you boot it this way. You will find many useful instructions from Mikeb in the above link.
Puppy already has option to load one extra sfs on boot in case you give the right name:
http://murga-linux.com/puppy/viewtopic. ... 837#779837
-
stemsee
as usual
Saintless
After reading all of that I have no clear idea.
I don't know too much. Maybe i only need to have all modules built into kernel and firmware packed into initrd to be mounted on a layer above that of the guest base.sfs using zdrv renamed mechanism or wildcard mechanism. Then the kernel and initrd might load any base.sfs and work! How to implement the wildcard mechanism from kernel argument and initrd init? Just using a variable in distro specs base sfs field, or wildcard ... filter the real name to zdrv.
I prefer to experiment and then read!
Musher0 - I don't know!! Wish I did. I am going to scrap this remaster as it is pointless in comparison to wht I have just learnt. I have to rebuild the kernel aneww and the initrd too.
After reading all of that I have no clear idea.
I don't know too much. Maybe i only need to have all modules built into kernel and firmware packed into initrd to be mounted on a layer above that of the guest base.sfs using zdrv renamed mechanism or wildcard mechanism. Then the kernel and initrd might load any base.sfs and work! How to implement the wildcard mechanism from kernel argument and initrd init? Just using a variable in distro specs base sfs field, or wildcard ... filter the real name to zdrv.
I prefer to experiment and then read!
Musher0 - I don't know!! Wish I did. I am going to scrap this remaster as it is pointless in comparison to wht I have just learnt. I have to rebuild the kernel aneww and the initrd too.
So stemsee does not have an answer to my question ?... I can keep at it until I get one,
you know !
If you're busy, a simple yes or a no will do fine.
I am quite sure that you don't want people to start a rumor to the effect that Saucy
Puppy-5.8 won't work on 7-year old mono-core Acer computers. Or perhaps you're
working on it?
An answer, please. Thanks in advance.
Best regards.
musher0
I can keep at it until I get one, you know !
If you're busy, a simple yes or a no will do fine.I am quite sure that you don't want people to start a rumor to the effect that Saucy
Puppy-5.8 won't work on 7-year old mono-core Acer computers. Or perhaps you're
working on it?
An answer, please. Thanks in advance.
Best regards.
musher0
musher0
~~~~~~~~~~
"You want it darker? We kill the flame." (L. Cohen)
~~~~~~~~~~
"You want it darker? We kill the flame." (L. Cohen)
-
stemsee
musher0
How to answer ?
I suggest experimenting with kernel arguments, as everyone else doesn't share this problem it is machine specific!, I drew a blank at this stage Are you booting the iso or frugally? As I wrote I am not trying to 'fix' this i am going to start again - just wait for the next iteration ... I will also build a full featured version of Saucy Salamandar in good time.
I had a look at the info you posted, but no alarrm bells went off in my head!
I suggest experimenting with kernel arguments, as everyone else doesn't share this problem it is machine specific!, I drew a blank at this stage Are you booting the iso or frugally? As I wrote I am not trying to 'fix' this i am going to start again - just wait for the next iteration ... I will also build a full featured version of Saucy Salamandar in good time.
I had a look at the info you posted, but no alarrm bells went off in my head!