Bug 215548
| Summary: | High cpu usage caused by kernel process when upgraded to linux 5.16 | ||
|---|---|---|---|
| Product: | Process Management | Reporter: | William Tang (galaxyking0419) |
| Component: | Other | Assignee: | process_other |
| Status: | NEEDINFO --- | ||
| Severity: | high | CC: | regressions |
| Priority: | P1 | ||
| Hardware: | Intel | ||
| OS: | Linux | ||
| Kernel Version: | 5.16.0 | Subsystem: | |
| Regression: | Yes | Bisected commit-id: | |
| Attachments: |
htop screenshot
5.15 interrupts 5.16 interrupts |
||
|
Description
William Tang
2022-01-30 09:00:57 UTC
Created attachment 300350 [details]
htop screenshot
Hardware Model: Dell Inspiron 13 (7380) Not my area of expertise, but another bug about high CPU usage was filed today? https://bugzilla.kernel.org/show_bug.cgi?id=215546 Dup? Could you check /proc/interrupts? Your best chance of getting this fixed is to run git bisect: https://ldpreload.com/blog/git-bisect-run I've tried googling for this issue and it can be caused by a variety of different reasons, so it's hard to say what's broken for you. (In reply to The Linux kernel's regression tracker (Thorsten Leemhuis) from comment #3) > Not my area of expertise, but another bug about high CPU usage was filed > today? > https://bugzilla.kernel.org/show_bug.cgi?id=215546 > > Dup? Could you check /proc/interrupts? (In reply to Artem S. Tashkinov from comment #4) > Your best chance of getting this fixed is to run git bisect: > https://ldpreload.com/blog/git-bisect-run > > I've tried googling for this issue and it can be caused by a variety of > different reasons, so it's hard to say what's broken for you. Hi, thanks for the replies. I'm afraid I'm currently do not have the time to do a full bisect on my laptop. However, I turned of the my DE and take a snapshot of /proc/interrupts (see the attachments for the logs). It looks like there are substantially more "Local timer interrupts", "Rescheduling interrupts" and "Function call interrupts" on 5.16. Hope this helps. Created attachment 300372 [details]
5.15 interrupts
Created attachment 300373 [details]
5.16 interrupts
(In reply to William Tang from comment #5) > It looks > like there are substantially more "Local timer interrupts", "Rescheduling > interrupts" and "Function call interrupts" on 5.16. Took a quick looks, to me it looks a lot like a dupe of https://bugzilla.kernel.org/show_bug.cgi?id=215533 which discussed a patch that seems to help. Maybe give it a try. > Hope this helps. Note, I'm not involved with the code, I'm just someone that keeps an eye out for regressions and wanted to help, as I noticed the similarity. (In reply to The Linux kernel's regression tracker (Thorsten Leemhuis) from comment #8) > (In reply to William Tang from comment #5) > > It looks > > like there are substantially more "Local timer interrupts", "Rescheduling > > interrupts" and "Function call interrupts" on 5.16. > > Took a quick looks, to me it looks a lot like a dupe of > https://bugzilla.kernel.org/show_bug.cgi?id=215533 > which discussed a patch that seems to help. Maybe give it a try. > > > Hope this helps. > > Note, I'm not involved with the code, I'm just someone that keeps an eye out > for regressions and wanted to help, as I noticed the similarity. I tried to change irq_services from 0 to PCIE_PORT_SERVICE_BWNOTIF. Unfortunately, it does not fix the issue for my laptop. |