(old)Puppy Linux Discussion Forum

It includes only the modules that came with the Kernelsource of 2.6.27.
I removed all modules and firmware from 2.6.25.16.

wait... here is a list:
ftp://ks301128.kimsufi.com/testing/pupp ... odules.txt
Mark

Thanks.

It works fine on my desktop - ethernet works, sound works, video works, external usb harddisk works, and both my CPUs are detected. No glitch at all. I only tested live CD though. Thanks for this.

Hardinfo report attached.

EDIT: just to note that this post is made with your 2.6.27 iso.
EDIT: this kernel even enables me to do something i haven't been able to do in a very long time - watch a VCD !

MU wrote:I uploaded a test Iso.
ftp://ks301128.kimsufi.com/testing/puppy4.1-k2.6.27/



Mark

Hey Mark great to hear you got a test iso
so quickly

I am downloading
and will give it spin too


big_bass

hi mu,

to give you a short response on my testings:

Laptop Lenovo 3000 N200 Intel(R) Pentium(R) Dual CPU T2370 @ 1.73GH

- video => works out of the box (normal 4.10 too)
- sound => works out of the box after alsa_config (first time ever on linux no 'hand work'=> snd_hda_intel)
- wlan => does not work at all (normal 4.10 does) see attached pic. the failure occurs by running the connect-wizard. the loaded iwl3945 seems to be the correct one (same as in 4.10 'normal').

if it helps i have uploaded a hardinfo-report of both to

http://www.mediafire.com/aragon-puppy in test-2627

maybe we could fix the problem, as i'm too very interested on using dual-core.

cheers
aragon

aragon
It seems I forgot to update /lib/firmware.
Please install this attached pet.
ftp://ks301128.kimsufi.com/testing/pupp ... rmware.pet

I updated the iso, too, it now includes these updated firmware files.

Mark

ath5k_pci for wireless of the eeePC works out of the box, but the atl ethernet driver not. (This will be a good time to learn compiling of driver modules. Also, I wonder if fan control exists in the kernel?)

Audio works, including the microphone.

The Seamonkey page disappeared at full screen, reappeared when I restored the window (but this could be a seamonkey issue).

Hotpup does not automatically load the USB flash drive when inserted, but that is fine with me - I used pmount. EDIT: My SD card is not automatically loaded by hotpup, but USB flash is.

The line "MU sleep 4" appears at boot time. Booting was fast (I used Grub), the only noticeable delay was during the loading of kernel drivers.

EDIT re SD card only not being auto-loaded by hotpup.

I updated the Iso, it now includes the updated firmware files.

raffy

The line "MU sleep 4" appears at boot time.

yes, this is intended.
I hope someone will have a look at the delay I added.
I described that issue in detail in the former messages.
Maybe it could be replaced with some intelligent code to check, if all drives were detected.
Delays are just a workaround.

We might gain 3-4 seconds like that

Loading the kerneldrivers will be a bit slower, as there are more included now.
I tried to add as many as I could.

Mark

@MU:

Well done!

My dual-Xeon system does not suffer the "missing hyperthread" and "slowdown in zlib" problems which I reported here.
***EDITED Nov. 20: it _does_ show the 4 hyperthreads...which is as it should be, i.e. hardinfo considers each hyperthread as a separate CPU)***

Oddly, PMount shows my SCSI drives as usb devices (same as Barry's "stock 4.1").

Seamonkey runs extremely fast...now to try ttuuxx's Firepup-.02 with this kernel...

(***edited) Firepup is fast indeed...but, Seamonkey is fast too...

Here's the promised lzma patch for 2.6.27. The good news is taht it looks like this feature will soon be pulled into the main kernel branch. the bad nes is that the patch is till being reworked to please the kernel devs. It has already been updated 5 times, but the attached patch is an earlier version that works. the latest revisions are still having problems since they move lots of exiting kernel code around. to my credit, before I found these later patches, I got the old one working on my own. But I have attached the best alternative(for the moment) from the guy who is writing the patch. His work is based on an earlier idea which added bzip2 support to the 2.4 kernel. I use the original patch with my 2.4 kernels and it works fine.
If you want to try this patch, apply it after the other patches you are using with 2.6.27.

....
If you want to try this patch, apply it after the other patches you are using with 2.6.27.
Sorry, I had trouble attaching the patch. Here it is:

Hi Mark,

I tested your updated iso (updated in response to aragon) on my work laptop, a toshiba with dual-core Core CPUs. Everything works fine except the wireless - which uses iwl3945 driver, just like aragon, same error message.

I did a little bit of investigation, the /lib/firmware doesn't contain anything that sounds like iwl or 3945, so I mounted the original pup_410.sfs from 4.1 release, copy all the /lib/modules/all-firmware directory from there to the same location.

Then I tried to bring the wlan0 up, still the same error. I did a quick look at the this newly-copied /lib/modules/all-firmware, I found iwlwifi.tgz which sounds like the correct driver, and I extracted it to /lib/firmware. Then I tried to bring wlan0 up again. This time it works !!!

So I guess we need to copy the all-firmware directory from 4.1 release, but I wonder why the auto-loading doesn't work ... why it has to be manually expanded. Or perhaps we should just expand it ... since it's contained within pup_410.sfs, which is already compressed.

cheers!

EDIT: I tried to boot from original 4.1, and I confirm the firmware is automatically detected and extracted into /lib/firmware.

Hey Mark

I was doing a lot of downloads yesterday
so today I had a chance to boot into your new test iso

all booted fine and seems to be working fine

I included my bootkernel.log so you can look over it
and see what was detected

P.S after looking at all you had to do to compile on puppy
its good you got it working so quickly

it took me two days to apply it and getting it working on Slackware using a different kernel (slackware 12.1 )
so I understand the effort you invested

great work

big_bass

jamesbond

I found:
/etc/modules/

It includes a file called
firmware.dep.2.6.25.16

I wonder how this file is created?
I think it had to be renamed to
firmware.dep.2.6.27

It also should be updated to the firmware that comes with the new kernel, but for a first test, we could ignore that.

I wonder, if your firmware now gets extracted correctly?

Who knows, how this file is created?
And what is /etc/modules/DOTconfig-K2.6.25.16 needed for?
Is it there for documentation, or is it scanned by some scripts?
Mark

Amigo,
thanks for the lzma patch, I'll try to appy it tonight.
Mark

I'm not sure about this, but I suspect that these two files:
firmware.dep.2.6.25.16
/etc/modules/DOTconfig-K2.6.25.16
are part of Puppy's frimware and event-handling setup. They are not reguklar Linux files. the DOTconfig is most likely a copy of the kernel config file (.config).

yes, so we had to copy /lib/modules/2.6.27/.config to
/etc/modules/DOTconfig-K2.6.27

For those who did not download the source sfs, here is the config:
http://www.murga-linux.com/puppy/viewto ... 215#239215

Now I must find from which files the firmware.dep is created.
Maybe I have to write a small script, that generates it.

Mark

MU wrote:jamesbond

I found:
/etc/modules/

It includes a file called
firmware.dep.2.6.25.16

I wonder how this file is created?
I think it had to be renamed to
firmware.dep.2.6.27

It also should be updated to the firmware that comes with the new kernel, but for a first test, we could ignore that.

I wonder, if your firmware now gets extracted correctly?

Who knows, how this file is created?
And what is /etc/modules/DOTconfig-K2.6.25.16 needed for?
Is it there for documentation, or is it scanned by some scripts?
Mark

Hey Mark

in the unleashed createpuppy script there is a reference to /lib/modules
not etc/modules


this all gets built in unleashed

********************************cut and pasted relevant lines of createpuppy script************
--------------------------------not the complete script ----------------------------------------


cp -a kernels/$USEKERNEL/firmware.dep.$USEKERNEL
echo "" >> modules-complete/lib/modules/firmware.dep.$USEKERNEL





echo '#These are the directories and files in each firmware pkg...' >> modules-complete/lib/modules/firmware.dep.$USEKERNEL






echo "Adding $ONEFIRM filelist to modules-complete/lib/modules/firmware.dep.$USEKERNEL..."
cp -af kernels/$USEKERNEL/all-firmware/$ONEFIRM/* modules-complete/
#need a list of all files in the package...
echo -n "${ONEFIRM}|" >> modules-complete/lib/modules/firmware.dep.$USEKERNEL
cd kernels/$USEKERNEL/all-firmware/$ONEFIRM
find ./ -mount | sed -e 's/^\.//g' | grep -v '^/$' | tr ' ' '_' | tr '\n' ' ' >> $PUPPYDIR/modules-complete/lib/modules/firmware.dep.$USEKERNEL
sync
echo >> $PUPPYDIR/modules-complete/lib/modules/firmware.dep.$USEKERNEL
cd $PUPPYDIR
done
sync
******************************************************************************************************

hope that helps btw that was a code snipit from 3.01 unleashed I dont know if in version
4.00 that got changed

big_bass

ok thanks, that might help...

I encountered other issues in this context.
1.) I could not find out, where the firmware comes from.
I could not find it in the Kernelsource, that has much less firmware files.
So I think the firmware files are something, that Barry got somewhere else.

For example iwl3945.ko IS part of the Kernelsource 2.6.27.
But /lib/firmware/iwlwifi-3945-1.ucode is NOT.
That file is only available in Puppy 4.1.
I also could not find it in the 3rd party modules at:
http://puptrix.org/sources/kernel-2.6.25.16/

So I then wanted to use the /lib/modules/all-firmware/ from Puppy 4.1.
But I encountered the same problem as jamesbond.

When I type:
modprobe iwl3945
then the
/lib/modules/all-firmware/iwlwifi.tar.gz
is NOT extracted.

I then thought it might have to do with a new kerneloption
CONFIG_FIRMWARE_IN_KERNEL = y
http://cateee.net/lkddb/web-lkddb/FIRMW ... ERNEL.html

So I deactivated it, and recompiled the Kernel.
But without success.

2.) Then I ran Puppy 4.1 original again (2.6.25.16).
Here I had the same problem:
When I type:
modprobe iwl3945
then the
/lib/modules/all-firmware/iwlwifi.tar.gz
is NOT extracted.

Same is with
modprobe dvb-usb
That problem was reported by linuxcbon here:
http://www.murga-linux.com/puppy/viewtopic.php?p=237657

So it seems, there is something wrong with the extraction of the firmware modules.

I then thought, that I might extract them all.
But that would work just in Puppy 3, that has a modified modprobe script.
It executes installation scripts, that are included in the firmware packages.
Like pinstall.slmodem.sh.
In Puppy 4, this mechanism is no longer used, instead, it is executed by the new (for me nonworking) event based extraction.

Fortunately, only some modems seem to have these scripts:
pinstall.dgcmodem.sh
pinstall.ltmodem.sh
pinstall.mwavem-1.0.2.sh
pinstall.pgprs.sh
pinstall.slmodem.sh

So I think I will rebuild the iso with the extracted firmware, and accept, that these modems will not work, unless the scripts are executed by hand.

This seems the best compromize, until a solution/bugfix is found.
The disadvantage is, that some firmware packages put scripts in /etc/init.d/ , what might slow down startup.
But I will test it, Muppy Live uses this solution, too.

Mark

I updated the Iso.
The Kernel now uses the lzma patch provided by Amigo.

I included the firmware from the original Puppy.
It is extracted, except these modem firmware packages:
dgcmodem.tar.gz
ltmodem.tar.gz
mwavem-1.0.2.tar.gz
pgprs.tar.gz
slmodem.tar.gz
Those have these pinstallscripts described above, so simply extracting them would make no sense.
Instead, the extraction script should be enhanced to work on every computer.

So Puppys firmware also should work, if on your computer the automatic extraction does not work.
The files in /etc/modules were updated/renamed.

I also upgraded the networkwizard to oct-11th.
And added the Glint xorg driver.

And I modified init again.
I removed the 4 seconds delay (and also the 3 seconds for classmate), and instead added a check, if usb-storage had finished to scan for drives.
Barry already had such code there, but for me, it did not work. I think, that is also, why he had added the 3 seconds for the classmate.
So I replaced it with own one, I hope this works for everybody.

Mark

hi mark,

i will test this evening.

one more question: did you compile the kernel with the "tickless idle"?

cheers
aragon