precise Puppy 5.5beta with 3.8.0 kernel

I announced Precise 5.5beta yesterday:

Now I have uploaded the same thing, except rebuilt using the 3.8.0 kernel.

Well, almost the same, there is a PPM bug fix, quite an important one, that affects all Upups and Dpups:

This build has been assigned version number 5.4.93, but still referring to it as "5.5beta".

Why would you use it?
1. Experiment with f2fs.
2. Have an older CPU that can't handle PAE.
3. Need the drivers of the very latest kernel.
4. Just want to have fun.

Download from here:

Posted on 21 Feb 2013, 15:05


Posted on 21 Feb 2013, 18:31 by Sage
No.4 qualifier
Remarkably slow loading and initial setup, but everything works ootb including video, sound and Flash.

There is a very very strange feature that occasionally turns up in several mutts, including this one. A bog std. ball PS/2 mouse is not detected/inactive at first boot attempt but fires up at second??? Sometimes, if a tearing hurry, I plug in a USB rodent which invariably overrides resident one - and that does work.

Posted on 21 Feb 2013, 18:46 by aarf
deltas and torrent Puppy 5.5beta with 3.8.0 kernel
info here
files here
to get files *need to use
login: puppy@changed
password: linux1

Posted on 21 Feb 2013, 19:56 by Sage
sr0 icon doesn't show on desktop.

Posted on 21 Feb 2013, 21:25 by BarryK
re sr0
Is it a scsi drive?

Both 5.4.92 and 5.4.93 are built without any scsi drivers.

Posted on 21 Feb 2013, 22:06 by Sage
Uh - how are you defining 'scsi' ?! It's a regular IDE DVD drive on the IDE port.

{Theorectically, everything is still 'scsi' in Linux, isn't it? Built into the fabric of all the file systems? At least that's what I understood a couple of decades ago - everything is a chained as /dev/xyz.}

[As for 'real' scsi, had reluctantly to abandon those a couple of years ago as the drives were all rather small capacity. Still got a boxload of scsi bits, though, including a very expensive Adaptec PCI drive card. Nothing is 15K rpm, probably 10K max? Let me know if you'd like to receive any stuff. Small can come by air, boxload by ship.]

Otherwise, Precise is shaping up really well with good feedback all round. micko's 'Earth rise' wallpaper has really proved a winner with folks I'm coaching into a better way of doing things! First impressions and all that.

Posted on 21 Feb 2013, 22:09 by Raffy
Broadcom 4313
Broadcom 4313 wireless does not show up here, but everything else seems OK.

Posted on 21 Feb 2013, 24:08 by GCMartin
Both of your Beta versions
Hi BarryK

I am running your Betas, LIVE. Both PAE and non-PAE versions boot, and operate without issues on the several 32bit PCs.

Looking forward to your next beta with the bug fixes and PPM fixes you've discovered.

And I want to share that the PuppyLinux downlad site for your BETA is a much welcomed location, for community pulls, to assist in beta test. Thanks for this.

Here to help

Posted on 22 Feb 2013, 7:21 by scsijon
JWM Upgraded to V691
I have been working with Joe Wing to fix problems with JWM for puppy while he is upgrading it. Version 691 is now out, which I think has fixed everything outstanding I could find in our bb system.

The pet and url for the thread is[url][/url]

Posted on 22 Feb 2013, 10:42 by aarf
precise-5.4.93> chosing modsetting driver now boots to desktop but with fbdev video driver still reported as being used by Quicksetup. screen resolution stays at 640x480 pixels.

precise-5.4.92 refuses to go to desktop at all as before.

Posted on 22 Feb 2013, 10:44 by scsijon
maybe we should always PAE?
I was interested to see this comment from one of the Kernel Developers on the opensuse site has added into the version 3.1+ kernels.

They have a group that specifically handle this build function, so I think it would be safe to consider it as gospel.

PAE is not only more physical address space but also important for the

"no execute" feature which disables execution of code that is marked as
non-executable. Therefore, the PAE kernel should be used on any systems
that support it, regardless of the amount of main memory.

Posted on 23 Feb 2013, 5:49 by peebee
Driver compiling difficulties in 5.4.93
Hi Barry

I have comipiled the Broadcom wl driver for Precise 5.4.93

However I had 2 problems:

I downloaded the devx twice but in both cases attempting to load it with sfs-on-the-fly crashed Precise. Installing by the boot manager and rebooting seemed to work however.

I got the following errors when compiling
  MODPOST 1 modules

WARNING: modpost: Found 1 section mismatch(es).
To see full details build your kernel with:

but the compile completed and created a working wl.ko

Posted on 23 Feb 2013, 6:35 by GCMartin
HARDINFO for WOOF (Precise)
Hi BarryK
@Rodin.S has made an additional contribution for 32bit PUPs with this Hardinfo update.

Here to help

Posted on 23 Feb 2013, 7:55 by carolus
In contrast to earlier puppies (e.g. lupu 5.2.8 or racy 5.3, I have been unable to get a usable window when running precise puppies in virtualbox. The earlier puppies work fine when set to a resolution of 800x600 (otherwise the display is bigger than my screen size). In precise 5.42, I was unable to change the resolution setting. In the present 5.5 beta I can change the resolution setting, which changes the window size but leaves part of the display lies outside the window.

I doubt this is a very high priority problem, but since it was working before, maybe there is an obvious fix.

Posted on 23 Feb 2013, 8:44 by Ted Dog
VirtualBox issue not linux
Not a puppylinux issue but Vbox, this would help

C:\>cd "c:\Program Files\Oracle\VirtualBox"
C:\Program Files\Oracle\VirtualBox>VboxManage setextradata "<vmname>" "CustomVideoMode1" "1280x800x16"

looks like Vbox has been bent/broke trying to support new M$ window version.

Posted on 23 Feb 2013, 8:36 by aarf
wheezy via probe to modesetting

Posted on 23 Feb 2013, 17:25 by BarryK
Re JWM 691
It is great that you are working with Joe to improve JWM running in Puppy.

I will certainly get the latest and start using it, but probably not for Precise 5.5.

I want to starting looking at freezing Precise, with a view to getting 5.5 out asap.

I might even make the next release an "RC".

Posted on 23 Feb 2013, 17:44 by BarryK
I have just read through the Forum thread. It is great that you guys have made Joe aware of the "windows opening partly under the tray" problem, and you have taken a photo to show the problem.

I am not sure from reading latest posts in that thread whether Joe has finally fixed that one.

Posted on 23 Feb 2013, 18:39 by aarf
before it can be called fully fixed i hope jwm can address the font issues at
havent had time to look at the latest jwm but since it is hot attention is called.

Posted on 23 Feb 2013, 24:22 by darkcity
thanks for pae
@scsijon thanks for the PAE tip - added to Wiki

Posted on 24 Feb 2013, 5:58 by carolus
Precise Puppy in VirtualBox
In contrast to earlier puppies (e.g. lupu 5.2.8 or racy 5.3, I have been unable to get a usable window when running precise puppies in virtualbox.
Ted Dog wrote:
Not a puppylinux issue but Vbox, this would help

C:\>cd "c:\Program Files\Oracle\VirtualBox"

C:\Program Files\Oracle\VirtualBox>VboxManage setextradata "<vmname>" "CustomVideoMode1" "1280x800x16"
The display resolution seems to be governed by Puppy, as shown by vboxmanage showvminfo. vboxmanage setextradata has no effect (but returns no error messages); showvminfo shows the same resolution before and after an attempted change. Changing the resolution within puppy, however, does change the showvminfo resolution to match. The problem is that with 800x600x32 in both virtualbox (seen from showvminfo) and in Puppy (seen from xrandr), only the left upper portion of the Puppy desktop is visible.

Posted on 24 Feb 2013, 8:03 by scsijon
@BarryK, the new one (when i'm back home to build) will, 691 did for most 692 is suppose to have improve further for small screen matrix's. I didn't expect it to be automatically included, just wanted to let you know it's available, do you want me to do a build a jwm with precise for precise when I build a 'release' pet or is wary ok still as a build base for most?

@darkcity, thanks for adding to the wiki, I relly must look and update my pages their; when I read the PAE comment a light went on as it explained why a couple of applications failed with the PAE kernel but not with a non-PAE, I let the relevant app programmers know about it and they are fixing them. With us running as root I think this sort of extra protection a bonus, even if a nusance to chase down.

@aarf, if you haven't already, could you add the info into my thread please as joe is keeping an eye on it and it will both ensure it gets looked at and that I add the problem to the list to test against.

Posted on 24 Feb 2013, 9:28 by BarryK
JWM in wary
Yes, I prefer it to be compiled in Wary, for max. compatibility with all pups.

Posted on 26 Feb 2013, 24:36 by darkcity
I've just noticed a thread saying there could be performance penalties using PAE - so I've also noted that on the Wiki.

thread link at bottom of the page