Precise Puppy RC2, October 20, 2012

Under development: PCMCIA, wireless, etc.
Message
Author
User avatar
rerwin
Posts: 2017
Joined: Wed 24 Aug 2005, 22:50
Location: Maine, USA

Another module-loader fix - INVALID, withdrawn!

#526 Post by rerwin »

Barry,
While testing my upgraded HSF modem support, I find that sometimes the HSF modem is discovered on first boot after installation or "pupdial erase", and other times not detected! I added trace logging to backend_modprobe to monitor its actions. I find that the trace log is identical (except for some of the ordering of unrelated events) whether the modem is detected or not. I suspect that the firmware package and extensive file changes made by the pinstall.hsfmodem script do not get made visible to other processes (i.e., /etc/init.d/hsfmodem) by the time they are needed.

I notice that the "backend_modprobe" script does not contain any 'sync' commands. I understand that command to force writing of file changes to the "device", where it becomes visible to all processes. My "rule of thumb" is to use 'sync' after writing a file that is intended to be used by other processes, on the assumption that the changes are visible only to the current process until the kernel decides to flush the buffers. Considering that several seconds appear to elapse before the initialization (init.d) script runs, it appears that the buffers do not get flushed even when the creating process (backend_modprobe's process) ends.

Because of the random nature of the detection/nondetection cases, it is difficult to verify absolutely that adding 'sync' results in success that could also happen by chance. But, since I added it, I have not seen a repeat of the detection failure.

My recommended fix is to add "sync" after the writing of the "lock1" and "protect1" files, and then after writing the "devpath" file. The latter would be effective for the files affected by a firmware package, which is where I suspect the HSF detection problem originates.

I attach a package containing only "backend_modprobe" with the 2 added 'sync' commands, in case anyone wants to try it. To test it, remove the files copied from the appropriate subdirectory in /lib/modules/all-firmware, from wherever they end up (usually /lib/firmware). Also remove /etc/modules/firmware.dep.inst.* (to cause reloading of the "firmware package"), then reboot.

Here is the unified difference listing:

Code: Select all

# diff -u /initrd/pup_ro2/sbin/pup_event_backend_modprobe /initrd/pup_rw/sbin/pup_event_backend_modprobe
--- /initrd/pup_ro2/sbin/pup_event_backend_modprobe	2012-08-28 03:25:20.000000000 -0400
+++ /initrd/pup_rw/sbin/pup_event_backend_modprobe	2012-09-06 12:03:01.000000000 -0400
@@ -16,6 +16,7 @@
 #120823 rerwin: $FIRMPKG always written. 
 #120823 rerwin: --use-blacklist to apply the blacklist commands in the configuration files (if any) to module names as well.
 #120828 rerwin: --use-blacklist again.
+#120906 rerwin: Add sync after writing lock/protect and devpath files to ensure files and firmware packages seen by all immediately. 
 
 export LANG=C
 . /etc/rc.d/PUPSTATE
@@ -182,6 +183,7 @@
 touch /tmp/pup_event_backend/lock1-${$} #start lock region.
 mREGEX=" ${MODULE} "
 echo "${$} ${MODULE} " > /tmp/pup_event_backend/protect1-${$}
+sync #120906
 for NUM in 1 2
 do
  SIMULT="`cat /tmp/pup_event_backend/protect1-* | grep "${mREGEX}"`"
@@ -216,6 +218,7 @@
 
 #log to file. rc.sysinit needs this info to find out if any modaliases missed (also above)...
 echo "MODULE=$MODULE DEVPATH=$DEVPATH MODALIAS=$MODALIAS" >> /tmp/pup_event_backend/pup_event_module_devpath_log${$}
+sync #120906
 
 cd /sbin #v408 rerwin thinks this is needed for slamr module.
 exec /sbin/modprobe $MODULE
UPDATE 9/10/12: The fix turns out to be associated with the problem, but not as a cure! The 'sync' command should not be used with module loading, at all. Please uninstall the "delta-3c" package if you used it. The problem appears to be caused by 'sync' commands within the HSF package, which are being removed.
Last edited by rerwin on Tue 11 Sep 2012, 13:44, edited 1 time in total.

tiangeng
Posts: 73
Joined: Thu 23 Jul 2009, 04:23
Location: æ²³å￾— China

#527 Post by tiangeng »

rcrsn51 wrote:
Yozo Office maybe a bug.
I have one more suggestion. In Racy, upgrade Ghostscript from here..

Then try the Gutenprint driver again.
Address Not Found

www.datafilehost.com could not be found.

User avatar
01micko
Posts: 8741
Joined: Sat 11 Oct 2008, 13:39
Location: qld
Contact:

#528 Post by 01micko »

tiangeng wrote:
rcrsn51 wrote:
Yozo Office maybe a bug.
I have one more suggestion. In Racy, upgrade Ghostscript from here..

Then try the Gutenprint driver again.
Address Not Found

www.datafilehost.com could not be found.
Probably blocked in China.

Try here.
Puppy Linux Blog - contact me for access

drblock2
Posts: 153
Joined: Mon 19 Jan 2009, 12:38

nvidia and radeon

#529 Post by drblock2 »

Tried Beta 5 with the Nvidia FX 5200 graphics card. Same results as before with Beta 4 (see also bigpup's report on page 34). Beta 5 boots into a desktop with broken menus and icons due to the nouveau driver.

Banning nouveau by adding nouveau.modeset=0 to the kernel line in grub boots into a perfect desktop with vesa, but too slow (glxgears = ca. 60 fps).

The Nvidia-173 driver compiles, but cannot be installed.

---------

Tried Beta 5 with my ThinkPad R60 (ATI Mobility Radeon X1400). The radeon driver was selected. Once again, very slow (glxgears 60 fps). After about 3 minutes, the display begins to disintegrate (pixels change color), leading to paralysis of the mouse and keyboard. Resurrection with hardware reboot.

Attempt to ban radeon driver with radeon.modeset=0 fails, radeon selected anyway. Attempt to opt out to vesa via xorgwizard also unsuccessful.

Since vesa really does do a reasonably good job for tasks that are not video intensive, assuring that the vesa driver can be installed is rather important. No distribution can be all things for all hardware, but I would be gratified to see performance equal to my main dog, Slacko-5.3.2.7 (kernel 3.1.10), which runs at 350 fps.

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

#530 Post by BarryK »

oldyeller wrote:Beta 5 OK here, booted just fine on my IBM X60

There is still one problem with the puppy-precise-official repo will not let you download from ibiblio, it only gives you the chose of local folder.
Thanks, I fixed it.
[url]https://bkhome.org/news/[/url]

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

Re: MOUSE FROZEN 90% of Time at boot

#531 Post by BarryK »

Ted Dog wrote:This BUG has existed for a while but it's now percise sub version .73 taking hours to get a working mouse. This really needs to be addressed. Tab reset only works once then disappears.
:cry:
I don't see how I can help. Seems to be a hardware incompatibility with the kernel. It may be that you will have no choice but to use some other Puppy for which your mouse works.
[url]https://bkhome.org/news/[/url]

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

Re: Precise Puppy beta5, Sept. 4, 2012

#532 Post by BarryK »

L18L wrote:I have loaded outdated langpack_de-20120205.pet just for testing new ppm

2 issues with ppm in this situation which can be normal in the future
- no package description in "installed packages"
- wrong (ALREADY INSTALLED) for not installed actual package
Well, I have fixed the missing icon and description in installed package.
Will now investigate the other bug.
[url]https://bkhome.org/news/[/url]

User avatar
davids45
Posts: 1326
Joined: Sun 26 Nov 2006, 23:33
Location: Chatswood, NSW

#533 Post by davids45 »

G'day,

Odd behaviour in a Full 5.3.91 made with the Universal Installer from a Frugal 5.3.91 using its home directory as the source of the installation files.
The Full to a freshly wiped partition.
The Full 5.3.91 install starts up and does all the right stuff and closes down as normal and without error messages.
The only new thing I've noticed is a change of display text size during booting and closing (text display starts as before but then goes coarser just before X opens), but other recent Pups do the same thing without the problem I get in this Full 5.3.91.

On re-booting the Full, after the first white text line, an orange-red text message appears almost immediately about checking its partition, and then with more white lines of text flowing down the screen, the computer freezes requiring a push-button restart.

When I check the dud Full 5.3.91 partition from another Pup or even from its OK 5.3.91 Frugal, I see in /mnt a ram0 directory listed. If I delete this ram0, the Full then starts up as normal. But on shutting down and attempting to reboot to this "fixed" Full, the same freeze occurs needing a reset to another Pup to get a good boot.

And the ram0 is back in the /mnt of the Full 5.3.91. Delete this and it will start again, but fail after a shut-down & reboot with the ram0 back again.

When I first found this problem fix (i.e. delete the ram0 directory), there were two ram directories in the Full's /mnt (ram0 and ram1), both of which I deleted and the Full seemed to re-start OK (until re-booted :roll: ).

I think 5.2.73 had the same freezing on re-boot problem but I overwrote that Full with this next version (5.3.91) so don't know if it also had this ram0 oddity.

Any advice or simple fix?

David S.
.

User avatar
Karl Godt
Posts: 4199
Joined: Sun 20 Jun 2010, 13:52
Location: Kiel,Germany

#534 Post by Karl Godt »

davids45 wrote:G'day,
On re-booting the Full, after the first white text line, an orange-red text message appears almost immediately about checking its partition, and then with more white lines of text flowing down the screen, the computer freezes requiring a push-button restart.

When I check the dud Full 5.3.91 partition from another Pup or even from its OK 5.3.91 Frugal, I see in /mnt a ram0 directory listed. If I delete this ram0, the Full then starts up as normal. But on shutting down and attempting to reboot to this "fixed" Full, the same freeze occurs needing a reset to another Pup to get a good boot.

I think 5.2.73 had the same freezing on re-boot problem but I overwrote that Full with this next version (5.3.91) so don't know if it also had this ram0 oddity.

Any advice or simple fix?

David S.
.
Dave, I have no precise idea, but /sbin/init does the fs check . Barry had incorporated some of my code form here recently http://murga-linux.com/puppy/viewtopic.php?t=67844 . I would suggest to

A)
* revert to an older /sbin/init from a Puppy 5 series year 2010 or 2011to see if that works better
B)
* try link /sbin/init to /bin/busybox
C)
* try one of my /sbin/init from the above link which should give you the possibility to select two debugging modes and post your experiences .

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

Re: Precise Puppy beta5, Sept. 4, 2012

#535 Post by BarryK »

L18L wrote:I have loaded outdated langpack_de-20120205.pet just for testing new ppm

2 issues with ppm in this situation which can be normal in the future
- no package description in "installed packages"
- wrong (ALREADY INSTALLED) for not installed actual package
I have implemented a fix for the "ALREADY INSTALLED" problem.
[url]https://bkhome.org/news/[/url]

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

#536 Post by BarryK »

LateAdopter wrote:Hello BarryK

Thanks for the latest Precise Puppy 5.3.91

I have tried some experiments with the package manager.

It hasn't gone to sleep on me yet.


The precise update packages still get downloaded and create separate databases, but they do not merge as it says it will do.

The result is too many Ubuntu repos for the PPM to handle at once.

I have tried deleting the update databases from the .Packages directory and updating just the precise updates on their own, but the outcome is the same.
I have fixed the merging.
[url]https://bkhome.org/news/[/url]

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

#537 Post by BarryK »

LateAdopter wrote:The symbolic link to GNU linux i386 is missing from /usr/bin so python failed to install there. I copied the link from /usr/lib and then python did install
Thanks for the report, fixed.
[url]https://bkhome.org/news/[/url]

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

#538 Post by BarryK »

01micko wrote:I have just compiled the nv driver and am about to test on this machine.

------------------------------------------------------------

Mixed success with nv
While it does work I couldn't get my desired resolution without manual intervention, however, mine is an obscure monitor with 1360x768 res, norm is 1366x768. Nouveau_unload worked as expected and rebooted to a vesa desktop. xorgwizard wouldn't select nv so I deleted xorg.conf and ran xorgwizard-automatic from the prompt.

# report-video
Precise Puppy, version 5.2.71 on Thu 30 Aug 2012

Chip description:
0.0 VGA compatible controller
NVIDIA Corporation NV6 [Vanta/Vanta LT] (rev 15)
oem: NVidia
product: Riva TNT Chip Rev B1

X Server: Xorg
Driver used: nv

X.Org version: 1.11.3
dimensions: 1360x768 pixels (342x191 millimeters)
depth of root window: 16 planes

...the above also recorded in /tmp/root/ as report-video,
and archived with xorg.conf and Xorg.0.log as report-video-full.gz


Here's nv compiled, --prefix=/usr --enable-xaa. Obviously no mesa counterpart so will use swrast.
I have also compiled 'nv', version 2.1.20. The PET will be uploaded soon.

Ubuntu has discontinued providing the nv driver, so we have to provide it.

It will be in the next Precise release, but in /usr/lib/x/drivers-alternate, so that by default it won't conflict with nouveau driver, but xorgwizard can get it back.
[url]https://bkhome.org/news/[/url]

User avatar
davids45
Posts: 1326
Joined: Sun 26 Nov 2006, 23:33
Location: Chatswood, NSW

#539 Post by davids45 »

G'day Karl Godt,
Thank you for your suggestions in regard to my odd problem with a Full 5.3.91. Not much luck so far.

A)
* revert to an older /sbin/init from a Puppy 5 series year 2010 or 2011to see if that works better

I tried pemasu's Upup Precise init then the init from the latest 3-headed dog. Both would run 5.3.91 initially but both gave the problem causing ram0 so a re-boot failed. The messages as it freezes are saying something about not enough space for ram0.

B)
* try link /sbin/init to /bin/busybox

I think I did this correctly, but no booting just a prompt with something like #sh3. Keyboard was working so it hadn't froze.

C)
* try one of my /sbin/init from the above link which should give you the possibility to select two debugging modes and post your experiences .

I shall try that next.
I was wondering if my problem relates to 5.3.91 failing to wipe/delete the ram drive during closing down on this desk-top - I have problems with Ubuntu-based ati drivers stopping too soon on close-down. Using the proprietary ati driver has fixed that problem in the past.
As no one else has reported anything like my ram0 "hangover", and I don't think I'm the only one setting up a Full 5.3.91 directly from a Frugal 5.3.91, it looks like it is a hardware issue, at least in part.

Thanks for your time and thoughts,

David S

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

#540 Post by BarryK »

feelsub wrote:Talking about PPM.

I suggest that the confirmation buttons, such as for trim are placed at the top of the screen and not down. :idea:
Indeed when there is a long list, the buttons are out of screen, and it's only by KNOWING how many tabs and then hitting enter, that it works.
I have fixed this, with scrollbars.
[url]https://bkhome.org/news/[/url]

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

#541 Post by BarryK »

01micko wrote:I had no problem switching to 24 bit colours using xorgwizard in beta5. I chose the "probe" option. I don't know why 16 is the default, in the current Slacko beta we are defaulting to 24. Same kernel series, similar config.
I set it to 16 bits as my ARM boards require it.
[url]https://bkhome.org/news/[/url]

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

#542 Post by BarryK »

don570 wrote:There's a missing GTK icon in the distribution
Both in pmirror and puppy clock

The line of script is

Code: Select all

 <input file icon="gtk-open"></input>
Image

___________________________________________
These apps are zigbert's babies. I have sent him a pm.
[url]https://bkhome.org/news/[/url]

User avatar
pemasu
Posts: 5474
Joined: Wed 08 Jul 2009, 12:26
Location: Finland

#543 Post by pemasu »

I have a collection of those gtk* png icons. gtk-open.png is among them. In the attached tarbal is those gtk* png icons. You can pick the gtk-open.png from there.
Attachments
gtk-png-icons-0.0.1.tar.gz
(77.01 KiB) Downloaded 253 times

User avatar
rjbrewer
Posts: 4405
Joined: Tue 22 Jan 2008, 21:41
Location: merriam, kansas

#544 Post by rjbrewer »

davids45 wrote: As no one else has reported anything like my ram0 "hangover", and I don't think I'm the only one setting up a Full 5.3.91 directly from a Frugal 5.3.91, it looks like it is a hardware issue, at least in part.

Thanks for your time and thoughts,

David S
Not a hardware issue.
Same problem on my MSI netbook with full install from usb.
Reboot from Precise or any other OS ends with that
kernel panic.
Boot from power off works.
Did not have that problem with the previous (non pae)
version.

Inspiron 700m, Pent.M 1.6Ghz, 1Gb ram.
Msi Wind U100, N270 1.6>2.0Ghz, 1.5Gb ram.
Eeepc 8g 701, 900Mhz, 1Gb ram.
Full installs

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

Re: Puppy Backup

#545 Post by BarryK »

don570 wrote:When I tried to install puppy backup
I found that the .desktop file had been changed. The category had
been changed from 'Archiving' to 'X-Archiving' .

This prevented the app from displaying in the Start menu. :cry:

_______________________________________________________
I tested, installed Puppy Backup 1.9, and did not get that problem.

But then, I am running a build later than beta5.
[url]https://bkhome.org/news/[/url]

Post Reply