SIO Interrupt storms and unhanded interrupts
jhb at FreeBSD.org
Thu Sep 9 12:10:24 PDT 2004
On Wednesday 08 September 2004 07:25 pm, Mike Tancsa wrote:
> I think we have been bouncing around this issue for the past few months
> both on RELENG_4 and now RELENG_5. In the past it has been somewhat
> difficult to reproduce, but now we can do it reliably. I dont think its
> a hardware issue as I can take the exact same 2 boxes with the exact same
> IRQ assignments and boot with OpenBSD and not run into an interrupt storm
> or freeze up the box. Swap back the RELENG_4 or RELENG_5 HD and again, I
> can produce an interrupt storm at will.
> I can also reproduce it on 2 different chipsets as well (VIA and
> Intel). The problem seems to be around how a PUC device (either a PCI
> modem or a PCI serial card) and the sharing of an interrupt (usually an USB
> controller, but not always).
> On RELENG_4, the box just locks up in a race trying to service an interrupt
> on IRQ 12 but remains unhandled.
> On RELENG_5, I actually catch an interrupt storm. e.g. I attach to sio4
> (PUC modem) and
> Interrupt storm detected on "irq12: uhci1"; throttling interrupt source
> Looking at vmstat -i does indeed show a the rate getting throttled
> releng-5-pioneer# vmstat -i
> interrupt total rate
> irq0: clk 596719 99
> irq1: atkbd0 2 0
> irq4: sio0 1079 0
> irq6: fdc0 1 0
> irq8: rtc 763812 127
> irq12: uhci1 5825 0
> irq13: npx0 1 0
> irq14: ata0 38727 6
> irq15: vr0 ata1 1984 0
> Total 1408150 235
> where irq12 is the IRQ shared by the modem and the USB port. However,
> because all IRQ 12s get throttled, the modem is unusable. e.g. trying to cu
> -l /dev/cuaa4 and typing atz takes about 5 seconds.
> Is there some way to safely tell the kernel that the PUC device that its
> shareable ? We did this perhaps very ugly hack on RELENG_4
> @@ -1431,15 +1431,19 @@
> rid = 0;
> com->irqres = bus_alloc_resource(dev, SYS_RES_IRQ, &rid, 0ul,
> ~0ul, 1, - RF_ACTIVE);
> +/* RF_ACTIVE); */
> + RF_SHAREABLE);
> to /usr/src/sys/isa/sio.c
> and at least we can talk to the sio device. However, on RELENG_5 there
> does not seem to be the same fix.
> My question is this-- Is the root cause the same issue on RELENG_4 and
> RELENG_5 ? Are we going about it the best way to fix the problem ? Or is
> the underlying problem something else ?
> Attached is a dmesg and acpidump
You can do a similar hack to puc/sio in 5.x, but you also need to remove the
'INTR_FAST' flag to bus_setup_intr() for it to work.
John Baldwin <jhb at FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve" = http://www.FreeBSD.org
More information about the freebsd-current