Linux 4.18 Problem Status

Under development: PCMCIA, wireless, etc.
Post Reply
Message
Author
scsijon
Posts: 1596
Joined: Thu 24 May 2007, 03:59
Location: the australian mallee
Contact:

Linux 4.18 Problem Status

#1 Post by scsijon »

EDIT: see bottom posts for any latest info

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

Linux 4.18 Is Appearing To Cause Problems For Those Running Older CPUs

https://www.phoronix.com/scan.php?page= ... -CPU-Issue to give some history

some fixes are appearing in subversions from 4.18.1, but there are still ongoing issues at .5

and it's not just the core2duo that has problems, some earlier ones that lead to the core2duo, or those based on the core2duo framework also have them.

regards
scsijon

EDIT1: Latest message, seems to track cause back a 'few minor code versions', although only visably appearing in 4.18.x with code tracking in progress, 'someone' is very unhappy with it only being found now.
Last edited by scsijon on Thu 27 Sep 2018, 09:31, edited 4 times in total.

musher0
Posts: 14629
Joined: Mon 05 Jan 2009, 00:54
Location: Gatineau (Qc), Canada

#2 Post by musher0 »

Thanks for the info, scsijon.

Any complaints about kernel 4.18 being slow on those older CPUs? TIA.
musher0
~~~~~~~~~~
"You want it darker? We kill the flame." (L. Cohen)

musher0
Posts: 14629
Joined: Mon 05 Jan 2009, 00:54
Location: Gatineau (Qc), Canada

#3 Post by musher0 »

Double. Sorry.
musher0
~~~~~~~~~~
"You want it darker? We kill the flame." (L. Cohen)

User avatar
Billtoo
Posts: 3720
Joined: Tue 07 Apr 2009, 13:47
Location: Ontario Canada

Linux 4.18 Is Appearing To Cause Problems with older cpu's

#4 Post by Billtoo »

Been running on my older i3 for the past 3 hours without problems:

System: Host: bill-pc Kernel: 4.18.5-1-MANJARO x86_64 bits: 64 Desktop: N/A Distro: Manjaro Linux
Machine: Type: Desktop Mobo: Acer model: Aspire X3950 serial: <root required> BIOS: American Megatrends
v: P01-A3 date: 05/05/2010
CPU: Dual Core: Intel Core i3 540 type: MT MCP speed: 1799 MHz min/max: 1200/3067 MHz
Graphics: Device-1: Intel Core Processor Integrated Graphics driver: i915 v: kernel
Display: x11 server: X.Org 1.20.1 driver: intel unloaded: modesetting
resolution: 1920x1080~60Hz, 1920x1080~60Hz
OpenGL: renderer: Mesa DRI Intel Ironlake Desktop v: 2.1 Mesa 18.1.7
Network: Device-1: Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet driver: r8168
Drives: Local Storage: total: 465.76 GiB used: 16.83 GiB (3.6%)
Weather: Temperature: 22 C (71 F) Conditions: Mostly Cloudy
Current Time: Fri 31 Aug 2018 10:29:35 PM EDT (America/Toronto)
Info: Processes: 178 Uptime: 3h 03m Memory: 5.75 GiB used: 1.27 GiB (22.0%) Shell: bash inxi: 3.0.21

scsijon
Posts: 1596
Joined: Thu 24 May 2007, 03:59
Location: the australian mallee
Contact:

#5 Post by scsijon »

@Billtoo, 4.18.5 is suppose to have it mostly worked out and the causial problems left 'bypassed' as I read it.

The next 'info meeting' will be in ~4 hours after all testing across platforms/hardware mixes are suppose to have completed and reported back.

regards
EDIT1: Apparently nothing came of it, other than everything seems to be ok.

On the other hand a couple of codeing problems that track back to 2004 and linux version 2.x were found and are now fixed

Also, as they are now working on 4.19, rc1 is out, only fixes are happening to 4.18. I suspect anything further will appear in 4.19.

scsijon
Posts: 1596
Joined: Thu 24 May 2007, 03:59
Location: the australian mallee
Contact:

is now 4 18.6

#6 Post by scsijon »

Just a note that 4.18.6 has been released containing fixes and tests ok.

If your using 4.18 series please update as soon as possible!

4.19 is at rc2.

scsijon
Posts: 1596
Joined: Thu 24 May 2007, 03:59
Location: the australian mallee
Contact:

now at 4.18.8

#7 Post by scsijon »

A few of the team are going back through code and fixing things and replacing code removed that in error.

If your using the 4.18.x kernel, can I sugest you don't be too far behind at present.

The latest sets include quite a few fixes for drm, i915 and amd processors and a couple of x86 ones that are also being inserted in 3.x.x longterm kernels. There's also fixes for that parallel building across processors (O=? or -j? option) we've had to put up with failing for a long time, ip6, tunnels and the usual cleanup after sucessfull code changes.

scsijon
Posts: 1596
Joined: Thu 24 May 2007, 03:59
Location: the australian mallee
Contact:

CVE-2018-17182 vunerability in 4.18.x pre .9

#8 Post by scsijon »

it's just been found a problem with 4.18 to .8 please update to .9 asap and don't release any earlier.

scsijon
Posts: 1596
Joined: Thu 24 May 2007, 03:59
Location: the australian mallee
Contact:

4.18.10

#9 Post by scsijon »

it's now out to .10

fixes include a timeout against a firmware problem (bug), 'stuff' for qualcomm drivers (including apparently a typo that got through qc), virtio-scsi to complete disk probe patch (fixes and drops a problem from ~20sec to 100ms), more drm fixes, nvme panic fix if port offline (unplugged) while live, selftests, multiport connections, f2fs fixes and a pile of new hardware additions and the normal fixes when code wrong found.

Includes a kbuild: do not update config when running install targets fix (one of my build hates as I never knew if something was changed for some reason or not).

This should be the last message in this thread unless a serious problem is found, although I will monitor 4.18.x until we change to 4.19.

Post Reply