Page 10 of 13

Posted: Wed 14 Apr 2010, 12:34
by Ray MK
Hi

Using Wary019 now with Opera10 sfs. Used SM2 earlier and all seems very good.

Initial impression is that Wary feels a little faster than Quirky018.

Using a little less ram memory to run as more is now available.
293 as opposed to 289 in quirky.

Like the new memory meter. Acpid works fine.
Nice to see abiword 2.8.3. - Gnumeric 1.8.4 (maybe 1.10.xx when time allows)

Overall IMHO - Superb.

As always - mega thanks for a smashing new Puppy and
Very best regards - Ray

Posted: Wed 14 Apr 2010, 17:09
by Sit Heel Speak
The /filesystem, /misc, and /wireless links at
http://bkhome.org/sources/kernel-2.6.33.2/3rd-party/
don't work. Are they unpopulated?

Posted: Wed 14 Apr 2010, 17:55
by `f00
my updated q018 test-article on p1

@capoverde on p6 (ref scrolling desks) - blame it on rox's pinboard (scrollwheel deskswitching in openbox is restricted a bit by using a roxpin for a drag'n'drop-featured desktop), openbox can be made more efficient but possibly it may 'break' in the process of trimming various features as integrated in q018.

@BarryK on p7 (ref fbpanel) - tpyos :), actually I like request and center on the left for those options, docking and setpartialstrut is another story..

confirm geany works fine in q018 (and other tes as well)

mmm, 'full' (poweroff) shutdown - actually I like power to remain on sometimes (using a liveCD 'bootdisc' since my old y2k BIOS won't properly boot from usb-stick and often I like the option to have power still 'on' so I can eject the liveCD 'bootdisc' at end of session) - btw, "system halted" makes sense to me as the final message .. maybe the "It is now safe to turn your computer off" for a retro look :)

retrovol (advanced) r-click menu works oki after the initial session - on initial session it poofs the advanced window and traycon.

addenda to my testing report - the usual (for me) edit of xorg got it working properly on my twin-vidcard usual machine, xorg accepted the edit and good to go (black fg on blue bg and all .. some xorgs are even harder to see the selected choice in an out-of-box console :lol: )

wary019 and trackpad

Posted: Wed 14 Apr 2010, 20:08
by broomdodger
wary019 can not turn off tap on trackpad
so sensitive just getting close will open items

tried mouse wizard -- no good
tried editing xorg.conf -- no good
replaced xorg.conf with one from quirky018 -- no good

quirky018 I can turn off tap and add side scroll
wary019 neither can be done

Sharp PC-PJ2-S2 Celeron 300MHz

Otherwise seems to work.

8.0rc (X 7.5) .1download error: zombie control panel

Posted: Wed 14 Apr 2010, 21:11
by Sit Heel Speak
The 1download control panel does not close when 1download finishes and is exited by pressing <Enter>.

Posted: Wed 14 Apr 2010, 21:34
by maddox
Quirky 0.18 suspend -> S3 state
One link : articles.techrepublic.com.com/5100-10878_11-6138911.htm S3 State
An excerpt :
Windozs XP, for example, has a default registry key that prevents PCs from dropping down to S3 sleep mode when USB KB/Mouse devices are detected,
because the motherboard might fail to wake up.
S3 is a state in which a PC is using nearly zero power and using barely enough to keep the contents of RAM intact.
My BIOS only has Wake on Lan, so it now makes sense that either opening the lid or moving the mouse (touchpad and USB ext mouse) can't initiate resume
I can see the lid switch state before suspend, but after S3/suspend there is no way to read the file contents
Hit any key : "/var/log/messages" contains lines "entering S3 state" and resuming from "S3 state" with additional info about restoring usb and pci

BTW, "/var/log/messages" gets truncated after subsequent resumes, initial value :1772 lines, next resume 39 & next resume 736 lines
also noticed a TZ diff in the Time of msgs in this file, -2 Hours after resume
My initial live-cd TZ choice was "Europe/Paris"

8.0rc (X 7.5) .2createpackages and .3builddistro errors

Posted: Wed 14 Apr 2010, 21:39
by Sit Heel Speak
Hope I'm not being a needless pest, but...
(***edited --figured it out***)

Posted: Wed 14 Apr 2010, 21:53
by Sit Heel Speak
(***edited out -- silly fool me***)

Thanks for the scsi!

Posted: Wed 14 Apr 2010, 22:04
by Sit Heel Speak
On a positive note: I am thrilled to see the option to include old-style scsi drivers in .3builddistro!

Posted: Wed 14 Apr 2010, 22:19
by Sit Heel Speak
(deleted -- good lord sometimes I'm thick..)

Posted: Wed 14 Apr 2010, 22:34
by Sit Heel Speak
(deleted...one learns by doing...)

Quirky-018 8.0rc (X-7.5) 3builddistro --the saga continues

Posted: Wed 14 Apr 2010, 23:35
by Sit Heel Speak
***edited:
a compile-from-source of gcolor2 places gcolor2.png in /usr/share/pixmaps
not
/usr/share/pixmaps/gcolor2
It is a symlink to /usr/share/pixmaps/gcolor2/icon.png
***

making a copy of /usr/share/pixmaps/gcolor2/icon.png
called /usr/share/pixmaps/gcolor2/gcolor2.png
solves the dependency

Wary 019

Posted: Thu 15 Apr 2010, 02:08
by BarryK
For Wary 019 only

Leon tried the rolled-back Ati driver, 6.7.197, reported it didn't work.

Ok, attached is version 6.7.195, same as used in Puppy 431.

Wary 019

Posted: Thu 15 Apr 2010, 02:10
by BarryK
On the other hand, going the other way, here is the very latest Ati driver (April 5, 2010).

Leon, try this also...

Posted: Thu 15 Apr 2010, 02:40
by Sit Heel Speak
(deleted by poster, who is smarter than he was yesterday)

Nvidia Closed Source drivers for Quirky

Posted: Thu 15 Apr 2010, 05:30
by Hesse James
Good morning Barry
Sorry for the late response. Generally I am interested in having the Closed Source Nvidia graphic drivers on board (or at least installable).
Reason is: I owe a HTPC with Soundgraph Imon VFD/RC and a Nvidia Chipset 9300. My intention is to build finally an alternative for my XP operation system which is too slow at the moment for HDTV despite CUDA-accelerated H.264 codec. I expect that VDPAU ist necessary to keep the used CPU resources as low as possible. VDPAU acceleration is only possible with the original NVIDIA drivers, patched mplayer, additional VDPAU packages available for Ubuntu and others...
My experience so far:
gray's PETs and the NVIDIA istaller work on 4.31 or earlier.
WOOF based Puplets: Nearly all newer WOOF based Puplets have problems with the NVIDIA-installer because of a gcc version mismatch the kernel was compiled with and the development system is equiped with. Quirky and LUPU6 refuse the installation because of that. UPUP and DPUP allow a continuation of the installation but crash if switching to 24 bit graphics (16bit works). All PET installations fail because of an interference with my Imon. USB/HID driver missing message despite installed...
So far only the attempt of Iguleder worked for me.
Kernel recompilation sometimes fail so that I didn't follow that because of lack of time.
Latest version so far is NVIDIA-Linux-x86-195.36.15-pkg1.run
Christian

Booting quirky-018 from an IDE drive on the Edubook

Posted: Thu 15 Apr 2010, 15:25
by pakt
Barry, I tested the RDC IDE driver for the Edubook that you actuated in the kernel.

I bought a short ribbon cable and had no problem booting quirky-018 from a Toshiba 1GB CF card in a CF-to-IDE adapter mounted in the Edubook.

Note however that in order to fit the bottom plate, I had to cut away the plastic lock from the ribbon cable's standard IDC connector (on the motherboard end), otherwise it was too high! Figures... :roll:

Anyway, the IDE interface seems to work perfectly.

The output of 'cat /proc/scsi/scsi':

Code: Select all

Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
  Vendor: ATA      Model: SA04G     RDC SD Rev: 0100
  Type:   Direct-Access                    ANSI  SCSI revision: 05
Host: scsi1 Channel: 00 Id: 00 Lun: 00
  Vendor: ATA      Model: TOSHIBA THNCF1G0 Rev: 3.00
  Type:   Direct-Access                    ANSI  SCSI revision: 05
Also, I didn't have to choose the boot order - it booted directly from the CF card, ignoring the inserted (bootable) SD card.

Patch needed to kernel 2.6.33.2 source code

Posted: Thu 15 Apr 2010, 15:53
by Sit Heel Speak
Overnight I've been compiling an rt version of the Puppy 2.6.33.2 kernel (still going), and have found an error in the aufs source code.

In file fs/aufs/file.h, line 102, aufs_release_nondir is defined, but in the two sourcefiles which include file.h and use aufs_release_nondir (fs/aufs/f_op.c and f_op_sp.c) aufs_release_nondir is redefined.

This stops the make. By commenting-out line 102 of file.h, i.e. making the snippet look like

Code: Select all

/* f_op.c */
extern const struct file_operations aufs_file_fop;
int aufs_flush(struct file *file, fl_owner_t id);
int au_do_open_nondir(struct file *file, int flags);
/* int aufs_release_nondir(struct inode *inode __maybe_unused, struct file *file); */
the make proceeds.

I will report back in a few hours on whether Quirky-018 with this corrected aufs will frugal-boot or not. It might require rebuilding the initrd.gz by hand as well, we'll see.

Posted: Fri 16 Apr 2010, 13:25
by fyujj
I don't know if these were reported (one of the two must have been).
First issue is with the Brazilian keyboard layout set-up in Puppy 5 series (Quirky 018, Lupu 004).
First, at the keyboard configuration page there are some odd options: br-abnt, br-abnt2, br-latin1-abnt2(?) and br-latin1-us(?). I'm no expert but as far as I know abnt2 is a layout, not a model. Usually the options would be (from the earliest to the most recent): br-latin1, br-abnt and br-abnt2 (default keyboard layout in Brazil for years now).
Up to Puppy 4.2, there was only br-latin1 and br-abnt options and one would have to open keyboard setup and choose abnt2 layout to setup this (most common) keyboard. This was fixed in Puppy 4.3.
*Important* Now any of the offered setups even work with the abnt2 keyboard (not even br-abnt2). It must be set through Setup > Change the mouse and keyboard and even this tool must be fixed in this respect: first there's apparently a major and general bug (this one that most probably was reported already) when one changes the keyboard from Choose keyboard layout for your country... option. This goes to a blue and gray screen of death, that one can get out with C+A+Bsp but then X wont load anymore (needs a reboot for an ignorant like me, even running Xorgwizard doesn't solve it). The other bug, this related to the BR keyboard setup, is that if one opens Advanced Xorg keyboard configuration... > Layouts and chooses br it doesn't work (this should enable by default the br-abnt2 layout). For it to work properly one must open Keyboard Model and choose Brazilian abnt2.

Regards and thank you.

This "bones+woof" thing-y is fabulous...

Posted: Fri 16 Apr 2010, 15:21
by Sit Heel Speak
8) 8) 8)

This bones + woof buildsystem, once you figure out 0setup and the DISTRO_PKGS_SPECS-{distro}-{version} syntax, and get the hang of turning .deb packages into .pets, is a fabulous work of absolute genius...

One rt-smp 2.6.33.2 version of Lucid, based on April 15 Bones and Woof, coming up soon...presently on "cups" in 2createpackages...