(old)Puppy Linux Discussion Forum

Posted: **Fri 30 Sep 2011, 12:22**

Hi scijon,
thanks for reporting, that is for me...

you wrote:...
3/ First shutdown (yellow screens) ?have a problem? >on the First Shutdown : Get Ready Screen we have some code visable on the left side, it's "${T_yesno3}"...

Change /usr/sbin/shutdownconfig line 123
from

Code: Select all

  T_yesno="\Z1${T_yesno1}\Zn\n\n\Z5${T_yesno2}\Zn\n\n\${T_yesno3}"

to

Code: Select all

  T_yesno="\Z1${T_yesno1}\Zn\n\n\Z5${T_yesno2}Zn\n\n${T_yesno3}"

that is: delete the red backslash
T_yesno="\Z1${T_yesno1}\Zn\n\n\Z5${T_yesno2}\Zn\n\n\${T_yesno3}"
Hope that helps to display: Choose Continue button to continue

Thank you
L

Posted: **Fri 30 Sep 2011, 20:23**

Manual frugal install.
Racy Puppy, version 5.1.101 on Fri 30 Sep 2011

Chip description:
0.0 VGA compatible controller
ATI Technologies Inc Redwood PRO [Radeon HD 5500 Series]
oem: ATI ATOMBIOS
product: REDWOOD 01.00

X Server: Xorg
Driver used: radeon

X.Org version: 1.11.0
dimensions: 1920x1080 pixels (507x285 millimeters)
depth of root window: 24 planes

Computer
Processor 6x AMD Phenom(tm) II X6 1045T Processor
Memory 8310MB (157MB used)

I made a pet of midnight commander file manager, it includes slang so it is a little larger than normal.

Posted: **Fri 30 Sep 2011, 20:48**

Tried to compile the full udev 167 with extras from T2 with their patches.

Code: Select all

./configure --prefix=/usr --sysconfdir=/etc --sbindir=/sbin \
 --libdir=/usr/lib --with-rootlibdir=/lib --libexecdir=/lib/udev \
 --build=i486-t2-linux-gnu --disable-introspection
 
# but fails with
.....
checking for acl_init in -lacl... no
configure: error: libacl not found

So I try to compile acl-2.2.47 (T2-patched) with

Code: Select all

./configure --prefix=/usr \
  --libdir=/usr/lib --build=i486-t2-linux-gnu

#or more elaborate options  
#but it also fails with

FATAL ERROR: attr/xattr.h does not exist.
Install the extended attributes (attr) development package.
Alternatively, run "make install-dev" from the attr source

Though attr/xattr.h IS in /usr/include.

But then I try to compile attr-2.4.43 just the same with

Code: Select all

./configure --prefix=/usr --build=i486-t2-linux-gnu 
#also tried other config options but 
make #always fails with
=== attr ===
gcc -g -O2 -DDEBUG -funsigned-char -fno-strict-aliasing -Wall -DVERSION=\"2.4.43\" -DLOCALEDIR=\"/usr/share/locale\" -DPACKAGE=\"attr\" -I../include -D_GNU_SOURCE -D_FILE_OFFSET_BITS=64    -c -o attr.o attr.c
/usr/bin/libtool --mode=link --tag=CC gcc -o attr    attr.o   ../libattr/libattr.la  
/usr/bin/libtool: line 676: libtool_args+=: command not found
/usr/bin/libtool: line 676: compile_command+=: command not found
/usr/bin/libtool: line 676: finalize_command+=: command not found
/usr/bin/libtool: line 676: libtool_args+=: command not found
/usr/bin/libtool: line 676: compile_command+=: command not found
/usr/bin/libtool: line 676: finalize_command+=: command not found
/usr/bin/libtool: line 676: libtool_args+=: command not found
/usr/bin/libtool: line 676: compile_command+=: command not found
/usr/bin/libtool: line 676: finalize_command+=: command not found
/usr/bin/libtool: line 676: libtool_args+=: command not found
libtool: link: gcc  ../libattr/.libs/libattr.so
/usr/lib/gcc/i486-t2-linux-gnu/4.3.4/../../../crt1.o: In function `_start':
(.text+0x18): undefined reference to `main'
collect2: ld returned 1 exit status
make[1]: *** [attr] Error 1
make: *** [default] Error 2

Now, I'm far from expert compiler but is it me or something is missing from devx?
Does Racy uses the standard T2 toolchain?

Posted: **Sat 01 Oct 2011, 07:59**

Trying Racy on a VIA C7-M machine with the VIA VX855/VX875 chipset (OLPC XO-1.5) and a 2.6.35 (OLPC) kernel that inactivates the synaptic driver.
The result is that Racy boots to desktop but keyboard and mouse are dead (internal or external). X is working fine and the machine is not frozen.

Looking at logs I can see in dmesg

Code: Select all

input: PS/2 Synaptics TouchPad as /devices/platform/i8042/serio1/input/input5

and in Xorg.log

Code: Select all

(**) Synaptics Mouse: always reports core events
(**) Option "Device" "/dev/psaux"
(EE) xf86OpenSerial: Cannot open device /dev/psaux
	No such device.
(EE) Synaptics driver unable to open device
(EE) PreInit returned 11 for "Synaptics Mouse"
(II) UnloadModule: "synaptics"
(II) Unloading synaptics
DEBUG: Stalled LOOP: 314
…

However the same exact messages are showing also with the same kernel and the same machine in puppies that work fine on this laptop. Namely, puppy-5xx, Slacko, dpup etc. There Xorg.log has (from puppy-5xx)

Code: Select all

(II) Synaptics touchpad driver version 1.2.2
Synaptics Mouse no synaptics event device found
(**) Option "Device" "/dev/psaux"
(EE) xf86OpenSerial: Cannot open device /dev/psaux
	No such file or directory.
(EE) Synaptics driver unable to open device
(EE) PreInit failed for input device "Synaptics Mouse"
(II) UnloadModule: "synaptics"
(**) Option "Protocol" "IMPS/2"
(**) Mouse0: Device: "/dev/input/mice"
(**) Mouse0: Protocol: "IMPS/2"
…..

So I guess the question is, is xserver/synaptics configured/compiled in some specific way distinct from the other woof-based puppies?

It's a petty because Racy looks faster than the other puppies at least on the OLPC XO-1 (that works OK with the exception of power management dues to the missing extras in udev-see above post)

For the record the xorg.cong is attached.

Posted: **Sat 01 Oct 2011, 08:28**

... and since I'm at it, though rather a woof issue.
in all puppies "modbrobe --ignore-install/remove" is an "unrecognized command" even when the "FULL" version is used.
Is this intentional?

Posted: **Sat 01 Oct 2011, 11:54**

... and since I'm at it, though rather a woof issue.
in all puppies "modbrobe --ignore-install/remove" is an "unrecognized command" even when the "FULL" version is used.
Is this intentional?

I know a little bit of the ^FULL^ modprobe source code :

1.: there are some -options missing in the options macro , at least for modprobe -V 3.6 and 3.11 .

[edit]
ERRATA: after testing I got a similar message

Code: Select all

modbrobe --ignore-install aes

bash: modbrobe: Kommando nicht gefunden.
bash: modprobe : Command not found

Then I looked inside the modprobe source , and I found the macros :
{ "ignore-install", 0, NULL, 'i' },
{ "ignore-remove", 0, NULL, 'i' },

Then I tried

Code: Select all

modprobe --ignore-install aes

a second time and then it worked :

lsmod | grep aes
aes_i586 7098 0
aes_generic 26376 1 aes_i586

So point 2 is partly wrong despite the '-i' option :
2.: [strike-through]there are no '--ignore-install|--ignore-remove' macros defined as options , but [/strike-through] they both are defined as '-i' ; passing the binary internal 'mit_ignore_commands' flag to the appropriate internal functions .
[/edit]

3.: as far as I know , Puppy uses the modprobe ^-FULL^ commands by default .
The first Racy still seems to use the /sbin/modprobe instead of /sbin/modprobe-BUSYBOX-NOT-USED ,
but for tinyRacy I can not check this from 431 , because the dotSFS are made with features , only a kernel 2.6.38+ seems to be able to access because of the new XZ/LZMA compression .

Posted: **Sat 01 Oct 2011, 13:16**

Karl Godt wrote: Then I looked inside the modprobe source , and I found the macros :
{ "ignore-install", 0, NULL, 'i' },
{ "ignore-remove", 0, NULL, 'i' },

You are right.
"modprobe --ignore-install" does not work but "modprobe -i" does.
Thanks

Posted: **Sun 02 Oct 2011, 11:32**

Got the 'disappearing CD icon' problem - right click overlaps underlying 'delete'. Problem discussed somewhere but don't remember solution. Suggested moving the position of the underlying pop-up?

Posted: **Sun 02 Oct 2011, 19:09**

Sage wrote:Got the 'disappearing CD icon' problem - right click overlaps underlying 'delete'. Problem discussed somewhere but don't remember solution. Suggested moving the position of the underlying pop-up?

Dependent on your screen resolution , the 'delete' option appears directly under the mouse or not .

If you have a mouse , that tends to send dobble-click impulses thought you felt you clicked the drive icon only once ,
it sends the second signal to the pop-up menu and activates the menu-point under the mouse .

I would suggest an older or newer mouse , replace the batteries , clean or dirty the internal contacts of the mouse .

Posted: **Sun 02 Oct 2011, 20:44**

Nothing wrong with this (HP-badged Logitech) mouse! There's no contact bounce due to the opto-couplers. There's some wrong setting in the coding because all other Puppies, new & old, used on new & old machines do not exhibit this feature. There was some discussion, but I forget the outcome - if there was one. Perhaps BK was testing it on a flaky serial mouse just before he wrapped up his development cycle?! In short, it's unique.

Posted: **Sun 02 Oct 2011, 23:04**

Running Live media system. Selecting FULLSCREEN yields the following.

On boot, system did not properly discover video and produced a blank screen. I Ctrl-Alt-Bjso and xorgwizard for VESA to get to desktop.

I opened a folder of video files and system does not drive the actual video in Full screen even though the application has expanded to fullscreen.

Not sure if this is the video player or the video sybsystem.

Code: Select all

Racy Puppy, version 5.1.101 on Sun 2 Oct 2011

Chip description:
 d.0 VGA compatible controller
 nVidia Corporation C61 [GeForce 6150SE nForce 430] (rev a2)  
oem: NVIDIA
product: MCP61 - mcp61-80 Chip Rev
 
X Server: Xorg
Driver used: vesa

X.Org version: 1.11.0
  dimensions:    1280x1024 pixels (339x271 millimeters)
  depth of root window:    24 planes

I am not sure if the two items mentioned are related.

Hope this helps

Posted: **Mon 03 Oct 2011, 03:19**

From my little knowledge of MPlayer, the option "xv" enables full screen while the default produces the full screen effect that you said. However, "xv" requires Xorg and will not work in Xvesa.

Since my install can only use Xvesa, that left me no hope for "xv".

That was in MPlayer. I installed vlc and qt4 and got video full screen working.

ADDITIONAL: Racy can't use the old OpenOffice sfs4 that Barry uploaded with Puppy 4.31. It is working in other sfs4 puppies, so there must be a problem somewhere.

Posted: **Mon 03 Oct 2011, 13:29**

@raffy. Thanks for your help. You are right about the video driver. I had to revert to VESA when the system started desktop with black screen. But, I just xorgwizard'ed Noveau and now the player's video is as expected. Thanks again.

I saw the OOo problem as well. But, it may be due to some missing libs.

I sent a PM to Barry on potential missing libs requesting consideration.

But, RACY is fast, very fast!
Here's my current video report

Code: Select all

Racy Puppy, version 5.1.101 on Mon 3 Oct 2011

Chip description:
 d.0 VGA compatible controller
 nVidia Corporation C61 [GeForce 6150SE nForce 430] (rev a2)  
oem: NVIDIA
product: MCP61 - mcp61-80 Chip Rev
 
X Server: Xorg
Driver used: nouveau

X.Org version: 1.11.0
  dimensions:    1024x768 pixels (270x203 millimeters)
  depth of root window:    24 planes

Hope this helps.

Posted: **Tue 04 Oct 2011, 06:27**

This is probably not a problem which will concern many people but Racy has broken Mailwasher for Linux for me. It no longer saves the registration code on reboot and reverts to unregistered/trial mode. Apart from this it works as expected. I wonder if this is a kernel issue or issue with updated libraries?

Posted: **Tue 04 Oct 2011, 11:52**

Hello ; I delved deeper into the xwin code , and maybe the described loop by me and 01micko could have been left with code like this :

Code: Select all

@@ -537,6 +537,19 @@
 echo -n "false" > /etc/.XLOADED #see note above.
 #...if PC hung, run xorgwizard at next bootup (see further up).
 
+fatalServerError=`grep -i 'Fatal server error' /tmp/xerrs.log`
+if [ "$fatalServerError" ] ; then
+echo -e "\e[1;31m"'To start the Xserver='"$HAVEX"' apparently failed !'"\e[0;39m"
+echo -e "\e[0;31m"'Now dumping the first 10 and last 6 lines of /tmp/xerrs.log :'"\e[0;39m"
+head -n 10 /tmp/xerrs.log
+echo -e "\e[0;33m"'-------------------------'"\e[0;39m"
+tail -n 6 /tmp/xerrs.log
+echo -e "\e[0;33m"'-------------------------'"\e[0;39m"
+[ "$HAVEX" = "Xorg" ] && echo "For more information , type 'mp /var/log/Xorg.0.log'"
+echo -n 'exit' > /tmp/wmexitmode.txt
+rm -f /tmp/videomode  ##edit changed '-r' to '-f' :Ooops:
+fi
+

Not sure about the right grep syntax for 'Fatal server error' at the moment , though .
[edit]
For the Puppy-43 xorg-server 1-3-0 it looks like that :

Code: Select all

grep -i -A1 'Fatal server error' xerrs.Xnvidia.log
Fatal server error:
No matching cards found!

There is another syntax for the second line with something like
"Failed to map the device" ; but to check this I have to boot another kernel with maybe different drivers loaded .
I have a
vendor: NVIDIA Corporation
product: NV18 Board Chip Rev A2
card and "Xorg (1-3-0,1-7-X,1-11-X)" and the "Xchips" and "Xvesa" (1-3-0) do run it , but not "Xnvidia" and almost all other kdrives [ with or without the appropriate vendorFramebuffer.ko 's inserted ] .
Funny , choosing 'chips' as Xorg driver , did start 'X' with max resolution (1600x1200) as the ^Lupu Xorg boot directly to desktop^ did - with the whole device section missing in xorg.conf . Racy Xorg or xorgwizard gives me 1024x768 in PUPMODE=5 .
That's it for now .
[/edit]
________________________________________

Another [ unproblematic ] occurrence is , that if

echo $LANG
de_DE@euro

is given ,
dc returns numbers like 123,456
comma instead of a dot for floats !

There are quite some scripts , that use "cut -f X -d '.'" to format the output of the 'dc' command to be useable for the 'expr' command ; example from /tmp/xerrs.log :

expr: non-numeric argument
/usr/sbin/background_reshape: line 56: [: -ge: unary operator expected

could need a LANG=C to prevent this .

"grep 'dc ' /usr/sbin/* | grep 'cut'" in /usr/*/*[/*] and /etc/*/*
did not throw much and the scripts that show additionally have already LANG=C set . So background_reshape may be [ one of ] the last leftovers , that did not get updated for that until now .

Anyway , reading

grep LANG=C /usr/sbin/re*
/usr/sbin/remasterpup2:LANG=C #faster, plus Xdialog happier.

may be the solution for the Xdialog warnings described in one of my former posts :

I recompiled Xdialog from debian lenny , because there is no Xdialog for later debians , including the patch on my machine "uname -m 'i686'" with "./configure --with-gtk2" which installs to /usr/local and the Xdialog of bootmanager is working correctly now . The size of the binary is still 'ls -s 69KB" .
The only thing is
"(Xdialog:26076): Pango-WARNING **: Invalid UTF-8 string passed to pango_layout_set_text()"
that spits out .

Posted: **Wed 05 Oct 2011, 02:48**

Sage wrote:Got the 'disappearing CD icon' problem - right click overlaps underlying 'delete'. Problem discussed somewhere but don't remember solution. Suggested moving the position of the underlying pop-up?

I have the same problem. Such a small design oversight, but such a nuisance!

I'm still using an early Wary .... can't spare even 100MB download on my 3G plan at the moment

.... and I experience no other problems with mouse clicks in any other use. So the problem seems unique here.

Posted: **Wed 05 Oct 2011, 12:57**

It's a bit different with Wary 103. Depending on the detente of one's finger, either the options box pops up or a new panel appears at top of screen listing the contents of sr0.
Clearly BK is aware of the issue, maybe not quite there yet? Mouse characteristics are adjustable, of course, but playing with those consumes much time with uncertain results.

Posted: **Wed 05 Oct 2011, 23:29**

Running Live media RACY.

After running couple days I've been watching a desktop "save" button that I had never tried before. As there never was any apparent need. But for the first time, I decided to try it. I had expected to save to Live media my current session without have to reboot.

Click "save"
system dropped to a terminal session (F1) indicating automatic login in prompt
Nothing???
No messages, no physical indication of anything...nothing

.So I typed "xwin"; I'm back to desktop to report to you. I have attached a system report if you want some idea of what the system is seeing.

Question
Should save button have saved my session to its Live media?

Hope this helps

Posted: **Fri 07 Oct 2011, 19:38**

Barry.. can I request a .pet for Seamonkey 2.4.1 please? The reason is 'drastically improved memory usage' quoted from Mozilla site.
Clearly Racy requires a particular compile.

Posted: **Sun 09 Oct 2011, 06:58**

This battery tray icon works

(old)Puppy Linux Discussion Forum

Racy Puppy 5.2.2 Final, 18 Nov. 2011

Re: racy101

Racy Puppy 5.1.101 (5.2 beta1)

devx issue?

Synaptics

modprobe

Problem with Video playback

full screen

Maybe Security Fix for /usr/bin/xwin