Red Hat Bugzilla – Bug 217373
irqbalance segmentation fault
Last modified: 2008-05-06 12:57:55 EDT
I recently stuck FC6 on my x86_64 (actually a Pentium D 950 which supports 64 bit) I cannot get irqbalance to run under a variety of kernels including the stock FC6. (atm I'm using 2.6.19-rc6). I als tried recompiling it from the src.rpm [root@emerald-x64 irqbalance]# ./irqbalance Segmentation fault [root@emerald-x64 irqbalance]# gdb ./irqbalance (gdb) run Starting program: /usr/src/redhat/BUILD/irqbalance-1.13/irqbalance/irqbalance Program received signal SIGSEGV, Segmentation fault. 0x0000555555555ac9 in parse_proc_interrupts (incremental=0) at procinterrupts.c:108 108 interrupts[irqnumber].count += count; (gdb) info locals word = <value optimized out> count = 37386 cursor = 0x7fff3aef4031 " 0 PCI-MSI-edge eth0\n" column = 1 irqnumber = 8411 file = (FILE *) 0x55555575b010 linebuffer = "8411:\000\000\000\000\000\00037386\000 0 PCI-MSI-edge eth0\n\000\000\n\000sb1,\000ehci_hcd:usb5\n\000e36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm constant_tsc pni monitor ds_cpl vmx cid cx16 xtpr lahf_lm\n", '\0' <repeats 59 times>... __PRETTY_FUNCTION__ = "parse_proc_interrupts" (gdb) list 103 if (!ret) /* non numeric end stuff */ 104 irqnumber = MAX_INTERRUPTS-1; 105 /* then N columns of counts, where N is the number of cpu's */ 106 } else if (column <= cpucount) { 107 sscanf(word,"%lli",&count); 108 interrupts[irqnumber].count += count; 109 /* and lastly the names of the drivers */ 110 } else if ( ( (incremental==0) || (interrupts[irqnumber].type==IRQ_INACTIVE) ) 111 && column>cpucount+1) 112 classify_type(irqnumber, word); (gdb) quit The program is running. Exit anyway? (y or n) y [root@emerald-x64 irqbalance]# cat /proc/interrupts CPU0 CPU1 0: 6685870 0 IO-APIC-edge timer 1: 12375 0 IO-APIC-edge i8042 6: 5 0 IO-APIC-edge floppy 7: 0 0 IO-APIC-edge parport0 8: 0 0 IO-APIC-edge rtc 9: 0 0 IO-APIC-fasteoi acpi 12: 133628 0 IO-APIC-edge i8042 14: 19918 0 IO-APIC-edge ide0 17: 791 0 IO-APIC-fasteoi uhci_hcd:usb2 18: 7 0 IO-APIC-fasteoi uhci_hcd:usb3 19: 45666 0 IO-APIC-fasteoi uhci_hcd:usb4, HDA Intel 20: 222106 0 IO-APIC-fasteoi uhci_hcd:usb1, ehci_hcd:usb5 21: 3 0 IO-APIC-fasteoi ohci1394 23: 36848 0 IO-APIC-fasteoi libata 24: 21900226 0 IO-APIC-fasteoi libata 8411: 38270 0 PCI-MSI-edge eth0 NMI: 3491 2771 LOC: 13483510 13483783 ERR: 0 [root@emerald-x64 irqbalance]# Ideas? I'm kinda stumped Phil =--=
I think this is a xen enabled kernel right? this is probably a segfault that I recently fixed under RHEL5. Must have forgotten to do this in FC6. Please try the attached patch in irqbalance and confirm that it fixes the problem. Thanks!
Created attachment 142197 [details] patch to increase max interrupts on irqbalance for xen kernels
Sooo close but no cigar 8) Actually I talked to ARjan about it (since he originally wrote the code) the issu is the interrupts number In this case the kernel has assigned the utterly WILD number of 8411 as an interrupt number to the ethernet controller. The way irqbalance works, it won't find a slot for that as even with your patch it only has slots for IRQ's from 0 up to 1023. Now On the good side, Arjan has actually been busy rewriting this code though he's awaiting Intel's lawyers to give him signoff to release it. So in conclusion,.. IRQ number 8411 is WAY too big for irqbalance least in the manner it handles it currently (even with your patch) Replacement irqbalance code is due out in a day or so's time Phil =--=
Well, he's right, 8411 is a wild number. Unfortunately, I'm waiting for arjan to send me his new code too, and I'm not sure that it will solve this problem. To be honest, Barring arjan doing a linked list in the new irqbalance that isn't directly indexed by irq number, this is likely going to be a WONTFIX. Is there anyway you can get that ethernet controller assigned a lower irq value?
We just installed FC6 x86_64 on a whopping huge server at work (4 dual core opterons), and I see what is probably the same irqbalance segfault when the system boots (though it seems to work without it - don't know what the long term implications of no irqbalance might be).
Having antagonized Arjan for ages thers a newer update that I've tried and works http://www.irqbalance.org/download.php Phil =--=
yeah, I pushed -0.55-2 for fc6 last night. This bug will close once the release team gets it into the fc6 updates repository
pushed
Fedora apologizes that these issues have not been resolved yet. We're sorry it's taken so long for your bug to be properly triaged and acted on. We appreciate the time you took to report this issue and want to make sure no important bugs slip through the cracks. If you're currently running a version of Fedora Core between 1 and 6, please note that Fedora no longer maintains these releases. We strongly encourage you to upgrade to a current Fedora release. In order to refocus our efforts as a project we are flagging all of the open bugs for releases which are no longer maintained and closing them. http://fedoraproject.org/wiki/LifeCycle/EOL If this bug is still open against Fedora Core 1 through 6, thirty days from now, it will be closed 'WONTFIX'. If you can reporduce this bug in the latest Fedora version, please change to the respective version. If you are unable to do this, please add a comment to this bug requesting the change. Thanks for your help, and we apologize again that we haven't handled these issues to this point. The process we are following is outlined here: http://fedoraproject.org/wiki/BugZappers/F9CleanUp We will be following the process here: http://fedoraproject.org/wiki/BugZappers/HouseKeeping to ensure this doesn't happen again. And if you'd like to join the bug triage team to help make things better, check out http://fedoraproject.org/wiki/BugZappers
This bug is open for a Fedora version that is no longer maintained and will not be fixed by Fedora. Therefore we are closing this bug. If you can reproduce this bug against a currently maintained version of Fedora please feel free to reopen thus bug against that version. Thank you for reporting this bug and we are sorry it could not be fixed.