Re: dmesg queries
> > > > kernel: Interrupt storm detected on "irq19: uhci0 uhci2"; > > throttling interrupt source > > kernel: Interrupt storm detected on "irq18: bge0 uhci1+"; > > throttling interrupt source > > I get this as well. I don't use USB so I turned that off at the BIOS > and removed it from my kernel, but I still get "interrupt storm" on > my bge0 device. No idea why. I had a similar problem on an itx board, enabling DEVICE_POLLING fixed it. I found on this system around 8-10% of CPU time - when idle was used for interrupts alone without this change. That was achieved without disabling USB, which I also need. > Performance is not all that great on this box disk-wise, but CPU wise > it is acceptably fast. I run FreeBSD/amd64 on it rather than > FreeBSD/ i386. > > > Vivek Khera, Ph.D. > +1-301-869-4449 x806 -- Dominic GoodforBusiness.co.uk I.T. Services for SMEs in the UK. ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: dmesg queries
On Sat, 25 Jun 2005 01:39 am, Chris Phillips wrote: > Hi All, > > ... > Finally, is it considered "bad form", to ask multiple > questions like I have, or should I have separated them & sent > them in multiple emails? I don't know about "bad form"; but human nature being what it is many potential respondents won't reply unless they can answer all. So you reduce your chances of obtaining what you want! Malcolm ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: dmesg queries
On Jun 24, 2005, at 12:09 PM, Chris Phillips wrote: Here's what has me writing... looking in /var/run/dmesg I see: - kernel: ioapic0: Changing APIC ID to 2 kernel: ioapic1: Changing APIC ID to 3 kernel: ioapic1: WARNING: intbase 32 != expected base 24 kernel: ioapic0 irqs 0-23 on motherboard kernel: ioapic1 irqs 32-55 on motherboard Should I be worried by that WARNING? I see this as well on my PE800. Also in /var/run/dmesg I see: - kernel: Interrupt storm detected on "irq19: uhci0 uhci2"; throttling interrupt source kernel: Interrupt storm detected on "irq18: bge0 uhci1+"; throttling interrupt source I get this as well. I don't use USB so I turned that off at the BIOS and removed it from my kernel, but I still get "interrupt storm" on my bge0 device. No idea why. Performance is not all that great on this box disk-wise, but CPU wise it is acceptably fast. I run FreeBSD/amd64 on it rather than FreeBSD/ i386. Vivek Khera, Ph.D. +1-301-869-4449 x806
dmesg queries
Hi All, I'm trying to figure out a couple of things & would like some advice please. My server was on FreeBSD 5.3 STABLE #1 this morning (I only took it to STABLE, because at the time, my GigNIC was not supported fully @RELEASE). I upgraded today & it's now looking like this: - % uname -a FreeBSD venus.rainbow-it.net 5.4-RELEASE-p2 FreeBSD 5.4-RELEASE-p2 #2: Fri Jun 24 13:43:08 BST 2005 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/VENUS i386 It's a lowly Dell PowerEdge 800 Here's what has me writing... looking in /var/run/dmesg I see: - kernel: ioapic0: Changing APIC ID to 2 kernel: ioapic1: Changing APIC ID to 3 kernel: ioapic1: WARNING: intbase 32 != expected base 24 kernel: ioapic0 irqs 0-23 on motherboard kernel: ioapic1 irqs 32-55 on motherboard Should I be worried by that WARNING? Also in /var/run/dmesg I see: - kernel: Interrupt storm detected on "irq19: uhci0 uhci2"; throttling interrupt source kernel: Interrupt storm detected on "irq18: bge0 uhci1+"; throttling interrupt source I think the top device is a Logitech QuickCam Express & the bottom one (using irq18), is my onboard Gigabit NIC. Would these lines in dmesg suggest that there is a problem and if so, is there anything that I can do to combat it? Is it likely that this 'throttling', is slowing my NIC at all? I have seen this kind of notification (throttling), when printing. This is the dmesg output for that device: - kernel: ppc0: port 0x778-0x77f,0x378-0x37f irq 7 drq 1 on acpi0 kernel: ppc0: Generic chipset (ECP/PS2/NIBBLE) in COMPATIBLE mode kernel: ppc0: FIFO with 16/16/8 bytes threshold kernel: ppbus0: on ppc0 kernel: ppbus0: IEEE1284 device found /NIBBLE/ECP kernel: Probing for PnP devices on ppbus0: kernel: ppbus0: PRINTER MLC,PCL,PML kernel: lpt0: on ppbus0 kernel: lpt0: Interrupt-driven port kernel: ppi0: on ppbus0 Now, the upgrade may well have stopped the behavior mentioned below, but if not, does anyone know what I might have done wrong, to be getting (a lot of) messages like: - sio0: 1848 more interrupt-level buffer overflows (total 11269) ? This is the device: - kernel: sio0: <16550A-compatible COM port> port 0x3f8-0x3ff irq 4 flags 0x10 on acpi0 kernel: sio0: type 16550A Finally, is it considered "bad form", to ask multiple questions like I have, or should I have separated them & sent them in multiple emails? Kind Regards, Chris Phillips ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"