4.8 kernel, under heavy load desktop unusable

[Gerry Boland]

I've updated from Xenial to Yakkety on my Desktop - Sandybridge Macbook Pro 8,1

With Xenial, on 4.4 kernel, doing a compile (with ninja for example), I could use all CPU cores and still have a usable desktop: app switching was relatively quick, I could (slowly) surf the web while the compile was happening.

Now with Yakkety 4.8 kernel, doing a compile using all cores totally seizes up my desktop - app switching is incredibly slow, I see redraw errors, mouse has poor response.

As a workaround, I have to manually leave a CPU core free when compiling.

ProblemType: Bug
DistroRelease: Ubuntu 17.04
Package: linux-image-4.8.0-26-generic 4.8.0-26.28
ProcVersionSignature: Ubuntu 4.8.0-26.28-generic 4.8.0
Uname: Linux 4.8.0-26-generic x86_64
NonfreeKernelModules: wl
ApportVersion: 2.20.3-0ubuntu8
Architecture: amd64
AudioDevicesInUse:
 USER PID ACCESS COMMAND
 /dev/snd/pcmC0D0p: gerry 4929 F...m pulseaudio
 /dev/snd/controlC0: gerry 4929 F.... pulseaudio
CurrentDesktop: Unity
Date: Tue Nov 1 12:02:02 2016
HibernationDevice: RESUME=UUID=2cba34fc-e62e-44c2-904a-1f3f96fbc5eb
InstallationDate: Installed on 2016-05-05 (179 days ago)
InstallationMedia: Ubuntu 16.04 LTS "Xenial Xerus" - Release amd64 (20160420.1)
MachineType: Apple Inc. MacBookPro8,1
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-4.8.0-26-generic.efi.signed root=UUID=a9e3a76e-0502-4f4e-a8d9-132fb83f44b1 ro noprompt quiet splash vt.handoff=7
RelatedPackageVersions:
 linux-restricted-modules-4.8.0-26-generic N/A
 linux-backports-modules-4.8.0-26-generic N/A
 linux-firmware 1.161
SourcePackage: linux
UpgradeStatus: Upgraded to zesty on 2016-10-27 (4 days ago)
dmi.bios.date: 10/26/15
dmi.bios.vendor: Apple Inc.
dmi.bios.version: MBP81.88Z.0047.B2C.1510261540
dmi.board.asset.tag: Base Board Asset Tag#
dmi.board.name: Mac-94245B3640C91C81
dmi.board.vendor: Apple Inc.
dmi.board.version: MacBookPro8,1
dmi.chassis.type: 10
dmi.chassis.vendor: Apple Inc.
dmi.chassis.version: Mac-94245B3640C91C81
dmi.modalias: dmi:bvnAppleInc.:bvrMBP81.88Z.0047.B2C.1510261540:bd10/26/15:svnAppleInc.:pnMacBookPro8,1:pvr1.0:rvnAppleInc.:rnMac-94245B3640C91C81:rvrMacBookPro8,1:cvnAppleInc.:ct10:cvrMac-94245B3640C91C81:
dmi.product.name: MacBookPro8,1
dmi.product.version: 1.0
dmi.sys.vendor: Apple Inc.
klp-taint: 12289

[Brad Figg (brad-figg)]

This change was made by a bot.

[Brandon Schaefer (brandontschaefer)]

This happens for me as well. More then enough Mem/swap left. Easy to reproduce when compiling -j4.

Testing with 4.4 and no issues.

If you need more hardware information let me know!

[Joseph Salisbury (jsalisbury)]

Did this issue start happening after an update/upgrade? Was there a prior kernel version where you were not having this particular problem?

Would it be possible for you to test the latest upstream kernel? Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Please test the latest v4.9 kernel[0].

If this bug is fixed in the mainline kernel, please add the following tag 'kernel-fixed-upstream'.

If the mainline kernel does not fix this bug, please add the tag: 'kernel-bug-exists-upstream'.

Once testing of the upstream kernel is complete, please mark this bug as "Confirmed".

Thanks in advance.

[0] http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.9-rc3

[Daniel van Vugt (vanvugt)]

This could be related to IO scheduling. In the past Unity7 has been guilty of doing IO during graphics operations. So graphics performance was impacted by heavy IO.

I haven't noticed this regression with kernel 4.8 myself. Perhaps because I've been using reasonably efficient SSDs?

[Daniel van Vugt (vanvugt)]

Maybe related: bug 1637666

[Brandon Schaefer (brandontschaefer)]

Tested with 4.9rc3 and things are working again (sweet and thanks!). I went back to 4.8.0.11 and it was still an issue there. As well as checking 4.4.0.34 and zero issues with that. So seems to have happened when going from 4.4 --> 4.8

[Gerry Boland (gerboland)]

Installing kernel from http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.9-rc3 fixes the issue

[Joseph Salisbury (jsalisbury)]

Can you also give this kernel a test to see if it's the same sched bug:
http://kernel.ubuntu.com/~jsalisbury/lp1627108/upstream/

[Daniel van Vugt (vanvugt)]

Indeed bug 1627108 sounds similar

[Gerry Boland (gerboland)]

@jsalisbury - I tried your kernel, it fixes the issue for me.

[Daniel van Vugt (vanvugt)]

Is it really a good idea to mark this well-described bug as a duplicate of something that just describes a solution?

I find one of the most valuable things we can do to propel the community forward is to have accurately described bugs so people can get together and discuss the same issue (even after a fix becomes available). But by marking this as a duplicate of bug 1637520, that can't happen.