LinuxQuestions.org
Visit Jeremy's Blog.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Distributions > Slackware
User Name
Password
Slackware This Forum is for the discussion of Slackware Linux.

Notices


Reply
  Search this Thread
Old 12-06-2017, 06:30 AM   #181
Darth Vader
Senior Member
 
Registered: May 2008
Location: Romania
Distribution: DARKSTAR Linux 2008.1
Posts: 2,014

Rep: Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717

Quote:
Originally Posted by chrisVV View Post
The linux kernel will need to use PAE to address more than about 3GB of memory in practice (try running 'free' with PAE turned off and you will see some of your 4GB disappear). That doesn't mean that a binary with 32-bit pointers might not run faster than a binary of the same program with 64-bit pointers, but using 4GB as your cut-off is somewhat arbitrary.
Well, usually I work with the range: 2GB -> 4GB -> 8GB, I have only two computers with 32, almost all has 8. Of course, I talk about my own computers.

Last edited by Darth Vader; 12-06-2017 at 09:26 AM.
 
Old 12-08-2017, 09:51 PM   #182
aaazen
Member
 
Registered: Dec 2009
Distribution: Slackware
Posts: 322

Rep: Reputation: 159Reputation: 159
Patrick's testing came to the attention of the kernel developers today.

Code:
kernels 4.13.*, 4.14.* 4.15-rc2 crash on occasion, especially
on x86-32 systems. To trigger the problem, run as root:

while true
do
        /sbin/udevadm trigger --type=subsystems --action=change
        /sbin/udevadm trigger --type=devices --action=change
        /sbin/udevadm settle --timeout=120
done

(Thanks to Patrick Volkerding for the reproducer).
...
https://marc.info/?l=linux-kernel&m=151277731921517&w=2

I can now crash my 64-bit AMD machine by installing a partition with slackware-current (32-bit) Linux 4-14.4 and running the Bash script loop above.

It crashes within a few minutes.
 
9 members found this post helpful.
Old 12-09-2017, 06:02 AM   #183
Darth Vader
Senior Member
 
Registered: May 2008
Location: Romania
Distribution: DARKSTAR Linux 2008.1
Posts: 2,014

Rep: Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717
Quote:
Originally Posted by aaazen View Post
Patrick's testing came to the attention of the kernel developers today.

Code:
kernels 4.13.*, 4.14.* 4.15-rc2 crash on occasion, especially
on x86-32 systems. To trigger the problem, run as root:

while true
do
        /sbin/udevadm trigger --type=subsystems --action=change
        /sbin/udevadm trigger --type=devices --action=change
        /sbin/udevadm settle --timeout=120
done

(Thanks to Patrick Volkerding for the reproducer).
...
https://marc.info/?l=linux-kernel&m=151277731921517&w=2

I can now crash my 64-bit AMD machine by installing a partition with slackware-current (32-bit) Linux 4-14.4 and running the Bash script loop above.

It crashes within a few minutes.
Congrats to our BDFL and the future looks brighter for us, as users.

Still, please do not forget also about those two poor souls who insisted to lament "But it crash!" while our BDFL responded to them a "Nope! Just happened a bit, and that was in the first release!"

In other hand, NEED_BOUNCE_POOL ? Could be this a thing to follow? Our BDFL already suspect that the issue is also config dependent...

Last edited by Darth Vader; 12-09-2017 at 08:44 AM.
 
Old 12-09-2017, 04:56 PM   #184
dracolich
Senior Member
 
Registered: Jul 2005
Distribution: Slackware
Posts: 1,188

Rep: Reputation: 50
Just to share my experience with 4.14.x. I have used .0, .1, .2, .3 and .4; all built using my 4.13.x config with no changes. .0 - .3 would crash during the first boot after installing the kernel, but never again until installing and booting a new kernel. .4 has not crashed for me in four boots, even during the first boot. The biggest difference I can think of between .3 and .4 is that I fixed all of my custom udev rules that it complained about.

I run Slackware 14.2 32-bit with 16GB DDR3 RAM and a Phenom II x2 555 unlocked to x4. For booting kernels I use GRUB and I don't use an initrd.

Just booted again, no problem, and I checked my udev version is 182-i486-7

Last edited by dracolich; 12-09-2017 at 06:55 PM. Reason: additional details
 
Old 12-10-2017, 10:36 AM   #185
nobodino
Member
 
Registered: Jul 2010
Location: in France
Distribution: slackware, slackware from scratch, LFS, linux Mint, Niresh (MacOS)...
Posts: 239

Original Poster
Rep: Reputation: 204Reputation: 204Reputation: 204
kernel-4.14.5 = "same punishment", segfault as usual.
No progess on that side.
 
3 members found this post helpful.
Old 12-10-2017, 12:11 PM   #186
Darth Vader
Senior Member
 
Registered: May 2008
Location: Romania
Distribution: DARKSTAR Linux 2008.1
Posts: 2,014

Rep: Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717
Same here, the 4.14.5, built following religious the original ricette from /testing, given on my mini-PC: "veni, vidi, ... crash" (it had gone crash on the third boot)

Last edited by Darth Vader; 12-10-2017 at 09:03 PM.
 
Old 12-15-2017, 11:28 AM   #187
dracolich
Senior Member
 
Registered: Jul 2005
Distribution: Slackware
Posts: 1,188

Rep: Reputation: 50
I had another kernel panic this morning. Second time booting a new 4.14.6.
I think I've traced mine to a 4-port usb hub. When the crash happened, before rebooting I disconnected the hub and error -71 messages began appearing on the screen. After a reboot the same crash occurred. I left the hub unplugged during the next boot and there was no crash.
After the successful boot and logging in I plugged in the hub and dmesg showed the -71 errors again on all three devices connected to the hub - an hdd, a webcam and a APC Back-UPS. I checked all of the connections between the hub and the back of the computer (it's a powered hub that passes through a usb switch box), reseated the power cord and all cables between it and the back of the computer and it's working again...for now. Probably the hub or switch is breaking down.
It's annoying that the 4.14 kernels are crashing when attempting to mount a usb hdd that produces i/o errors. I had these errors during 4.12 and 4.13. Those kernels did not crash, though; they just did not mount the device and continued.
 
1 members found this post helpful.
Old 12-15-2017, 11:46 AM   #188
orbea
Member
 
Registered: Feb 2015
Distribution: Slackware64-current
Posts: 899

Rep: Reputation: Disabled
Its possible that your usb hub is just slowly degrading. If you go back to 4.12 or 4.13 does it avoid the crash? If so then it should be reported upstream to the kernel devs.
 
Old 12-15-2017, 01:01 PM   #189
Drakeo
Senior Member
 
Registered: Jan 2008
Location: Urbana IL
Distribution: Slackware, Slacko,
Posts: 3,398
Blog Entries: 3

Rep: Reputation: 403Reputation: 403Reputation: 403Reputation: 403Reputation: 403
Quote:
Originally Posted by dracolich View Post
I had another kernel panic this morning. Second time booting a new 4.14.6.
I think I've traced mine to a 4-port usb hub. When the crash happened, before rebooting I disconnected the hub and error -71 messages began appearing on the screen. After a reboot the same crash occurred. I left the hub unplugged during the next boot and there was no crash.
After the successful boot and logging in I plugged in the hub and dmesg showed the -71 errors again on all three devices connected to the hub - an hdd, a webcam and a APC Back-UPS. I checked all of the connections between the hub and the back of the computer (it's a powered hub that passes through a usb switch box), reseated the power cord and all cables between it and the back of the computer and it's working again...for now. Probably the hub or switch is breaking down.
It's annoying that the 4.14 kernels are crashing when attempting to mount a usb hdd that produces i/o errors. I had these errors during 4.12 and 4.13. Those kernels did not crash, though; they just did not mount the device and continued.
You may need to blacklist a module. that happens when 2 are loaded. try it out.
Quote:
-71 errors
Feed back from /etc/udev/rules.d
Do you have any special rules for some devices like I do.

Last edited by Drakeo; 12-15-2017 at 01:03 PM.
 
Old 12-16-2017, 03:09 AM   #190
nobodino
Member
 
Registered: Jul 2010
Location: in France
Distribution: slackware, slackware from scratch, LFS, linux Mint, Niresh (MacOS)...
Posts: 239

Original Poster
Rep: Reputation: 204Reputation: 204Reputation: 204
when you look a the message from https://marc.info/?l=linux-kernel&m=151277731921517&w=2
There are some tracks to be followed.
Until we won't see in the changelog of the kernel something modifying:
-block/blk-map.c
-block/bio.c
-include/linux/blkdev.h
-drivers/target/target_core_pscsi.c
and probably some more.

I think it's not worth testing any new kernel-4.14.x of that series.
Personaly I won't.
Let's the kernel developpers do their job and wait.

Last edited by nobodino; 12-16-2017 at 09:18 AM.
 
3 members found this post helpful.
Old 12-16-2017, 07:45 PM   #191
USUARIONUEVO
Member
 
Registered: Apr 2015
Posts: 757

Rep: Reputation: 225Reputation: 225Reputation: 225
check next 4.14.7 release ,some of comented files get a patch

https://lkml.org/lkml/2017/12/15/287
 
Old 12-17-2017, 03:43 AM   #192
nobodino
Member
 
Registered: Jul 2010
Location: in France
Distribution: slackware, slackware from scratch, LFS, linux Mint, Niresh (MacOS)...
Posts: 239

Original Poster
Rep: Reputation: 204Reputation: 204Reputation: 204
None of the listed files appears on https://lkml.org/lkml/2017/12/15/287.
Don't hope too much for kernel-4.14.7.
But some have changes in the kernel git repository linux-stable tree, but not all.

Last edited by nobodino; 12-17-2017 at 03:44 AM.
 
Old 12-17-2017, 11:48 AM   #193
davleema
Member
 
Registered: Jun 2015
Location: Belgium
Distribution: Slackware x64
Posts: 31

Rep: Reputation: Disabled
Quote:
Originally Posted by dracolich View Post
Just to share my experience with 4.14.x. I have used .0, .1, .2, .3 and .4; all built using my 4.13.x config with no changes. .0 - .3 would crash during the first boot after installing the kernel, but never again until installing and booting a new kernel. .4 has not crashed for me in four boots, even during the first boot. The biggest difference I can think of between .3 and .4 is that I fixed all of my custom udev rules that it complained about.

I run Slackware 14.2 32-bit with 16GB DDR3 RAM and a Phenom II x2 555 unlocked to x4. For booting kernels I use GRUB and I don't use an initrd.

Just booted again, no problem, and I checked my udev version is 182-i486-7
Why are you running 32-bit Linux on a machine with 16GB ram ? Can you even use all the 16GB ?
 
Old 12-17-2017, 11:58 AM   #194
Darth Vader
Senior Member
 
Registered: May 2008
Location: Romania
Distribution: DARKSTAR Linux 2008.1
Posts: 2,014

Rep: Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717Reputation: 717
Yep, he can use his 16GB RAM all the way, under 32bits, via PAE.

The real limit is 64GB, and I tested practically that limit in the past. IF your computer has 128GB memory, the 32 bits PAE kernels will see and use only (the first?) 64GB.

And some caveats: no program can ask for more than 4GB for itself.

This limitation could be seen also as an huge advantage, specially when you play with recidivist hogs like Firefox.

PS. The 4.14.7 generic SMP build given me crash on every boot on that mini-PC and 32 bits. They improved it so much, that now it does not work anymore.

Last edited by Darth Vader; 12-17-2017 at 12:31 PM.
 
2 members found this post helpful.
Old 12-17-2017, 02:24 PM   #195
dracolich
Senior Member
 
Registered: Jul 2005
Distribution: Slackware
Posts: 1,188

Rep: Reputation: 50
Quote:
Originally Posted by orbea
Its possible that your usb hub is just slowly degrading. If you go back to 4.12 or 4.13 does it avoid the crash? If so then it should be reported upstream to the kernel devs.
I haven't tried going back to one of those. However, I remember well that I did not have such crashes when I last used a 4.13 kernel about a month ago. My 4.14 config is almost entirely based on what I used in 4.13

This Belkin 4-port hub is pretty darn old, so I wouldn't be surprised

@Drakeo
Yeah, maybe. The other things on this hub are an APC UPS interface cable and a Creative webcam that uses gspca_zc3xx

@davleema
I've had this Slackware installation since 9.1. I've just been too lazy to switch to 64-bit :P

@Darth Vader
Yep, no problem seeing and using 16GB. I mainly run Fluxbox as my de. Dang, that sounds discouraging about .7. I might just go back to .5 and wait a while

Last edited by dracolich; 12-17-2017 at 02:30 PM.
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off



Similar Threads
Thread Thread Starter Forum Replies Last Post
[SOLVED] Strange behavior battles Linux - Newbie 5 07-10-2014 09:59 AM
dcgui-qt's strange behavior harken Linux - Software 2 02-24-2005 03:10 PM
Very Strange Behavior raysr Mandriva 4 08-31-2004 03:06 PM
Strange Behavior andrewb758 Linux - Hardware 5 08-31-2003 03:42 PM
strange behavior abhijit Linux - General 3 07-10-2003 12:25 AM

LinuxQuestions.org > Forums > Linux Forums > Linux - Distributions > Slackware

All times are GMT -5. The time now is 09:33 PM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration