| Summary: | counter 0 used by nmi_watchdog | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Severin Gehwolf <sgehwolf> |
| Component: | oprofile | Assignee: | William Cohen <wcohen> |
| Status: | CLOSED WONTFIX | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 15 | CC: | jjohnstn, jreiser, patrickm, sgehwolf, tgl, wcohen |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2012-08-07 19:11:54 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
|
Description
Severin Gehwolf
2011-04-07 19:40:24 UTC
The problem is due to changes in how Linux kernel's watchdog timer reserves the performance counter hardware. It use to be that the watchdog timer, perf, and systemtap each used a different software mechanism. The watchdog timer has been merged into the perf framework. The downside of this is perf takes all the performance hardware when any of the performance monitoring hardware is being used by perf. Thus, oprofile, gets no counters when the nmi_watchdog is active. What would be the drawback of doing: opcontrol --deinit echo 0 > /proc/sys/kernel/nmi_watchdog before any profiling is done on F15? Or perhaps check if this perf. counter is used by something else, if so offer a better suggestion than just fail? Severin, it might be best if we throw a dialogue saying that the user needs to do that. Will suggested on IRC that some sysadmins may want to keep the nmi watchdog on so it's probably best to not turn it off under the covers. I'm running into this on F16 as well. It's fairly annoying that oprofile is unusable by default because somebody arbitrarily decided they could break it. I think the use of these hardware resources needs to be renegotiated. This message is a notice that Fedora 15 is now at end of life. Fedora has stopped maintaining and issuing updates for Fedora 15. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At this time, all open bugs with a Fedora 'version' of '15' have been closed as WONTFIX. (Please note: Our normal process is to give advanced warning of this occurring, but we forgot to do that. A thousand apologies.) Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, feel free to reopen this bug and simply change the 'version' to a later Fedora version. Bug Reporter: Thank you for reporting this issue and we are sorry that we were unable to fix it before Fedora 15 reached end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora, you are encouraged to click on "Clone This Bug" (top right of this page) and open it against that version of Fedora. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete. The process we are following is described here: http://fedoraproject.org/wiki/BugZappers/HouseKeeping The problem persists in Fedora 18 Alpha. "Perf" should have an interface to [re-]negotiate the use of the 4 HPET timers, and oprofile should use it. -----syslog (/var/log/messages) kernel: [ 1.310896] HPET: 4 timers in total, 0 timers will be used for per-cpu timer kernel: [ 1.317809] hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0, 0 kernel: [ 1.323076] hpet0: 4 comparators, 64-bit 14.318180 MHz counter ----- |