Red Hat Bugzilla – Bug 172909
Kernel 2.6.14-1.1637_FC4smp breaks irqbalance
Last modified: 2015-01-04 17:22:56 EST
From Bugzilla Helper:
User-Agent: Opera/8.5 (X11; Linux i686; U; en)
Description of problem:
After installing the kernel-smp-2.6.14-1.1637_FC4 kernel, irqbalance no longer
balances interrupts between the 4 (dual-Xeon /w hyper-threading) CPUs. The
balancing worked correctly in prior kernel versions.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1.Install package kernel-smp-2.6.14-1.1637_FC4.i686.rpm
2.Reboot to use new kernel
3.Wait a while, then check interrupt counts via /proc/interrupt
Actual Results: With 35 minutes of uptime:
$ cat /proc/interrupts
CPU0 CPU1 CPU2 CPU3
0: 487603 0 0 0 IO-APIC-edge timer
1: 2457 0 0 0 IO-APIC-edge i8042
7: 2 0 0 0 IO-APIC-edge parport0
8: 1 0 0 0 IO-APIC-edge rtc
9: 0 0 0 0 IO-APIC-level acpi
14: 64735 0 0 0 IO-APIC-edge ide0
169: 56014 0 0 0 IO-APIC-level aic7xxx, aic7xxx, eth0
177: 46919 0 0 0 IO-APIC-level uhci_hcd:usb1
185: 0 0 0 0 IO-APIC-level uhci_hcd:usb2
201: 5153 0 0 0 IO-APIC-level CS46XX
209: 104812 0 0 0 IO-APIC-level nvidia
NMI: 0 0 0 0
LOC: 487487 487486 487485 487484
Expected Results: The interrupts should be balanced between the 4 CPUs (except for the eth0
No errors or warnings are seen in system log.
# ps auxS | grep irqbalance
root 1797 0.0 0.0 1556 288 ? Ss 22:26 0:00 irqbalance
root 3790 0.0 0.0 3768 700 pts/1 S+ 23:09 0:00 grep irqbalance
FYI, still broken in 2.6.14-1.1644_FC4.
FYI, still broken in 2.6.14-1.1653_FC4.
Try with this patch:
or disable MSI in kernel.
This patch seems to have fixed it. I left MSI enabled (the default) and
rebuilt. The rebuilt 2.6.14-1.1656_FC4 kernel and the irqbalance service seem
to be balancing the interrupts correctly.
that patch should be in the 2.6.15 kernel in updates-testing. Can you test that
too please ?
This is a mass-update to all currently open kernel bugs.
A new kernel update has been released (Version: 2.6.15-1.1830_FC4)
based upon a new upstream kernel release.
Please retest against this new kernel, as a large number of patches
go into each upstream release, possibly including changes that
may address this problem.
This bug has been placed in NEEDINFO_REPORTER state.
Due to the large volume of inactive bugs in bugzilla, if this bug is
still in this state in two weeks time, it will be closed.
Should this bug still be relevant after this period, the reporter
can reopen the bug at any time. Any other users on the Cc: list
of this bug can request that the bug be reopened by adding a
comment to the bug.
If this bug is a problem preventing you from installing the
release this version is filed against, please see bug 169613.
Kernel 2.6.15-1.1830_FC4smp is balancing the interrupts correctly. In my view,
this bug can be closed.