Boot hangs on ips0: resetting adapter, this may take up to 5 minutes

[Oleg Sharoiko]

On Thu, 6 Apr 2006, Oleg Sharoiko wrote:

OS>It looks like the mis-routed interrupt is the one from ips. In my kernel 
OS>ips is on vector 49 and bge is on vector 60. I've added 
OS>	if (vector == 60)
OS>		vector = 49;
OS>to sys/amd64/amd64/local_apic.c and I have no more interrupt storm until 

I forgot to mention that I inserted this 'if' into lapic_handle_intr().

OS>bge really generates interrupt. Am I right with my conclusion about ips 
OS>interrupt being mis-directed to bge?
OS>
OS>There's also another interesting point: it looks like ips triggers 
OS>interrupt on both vectors (49 and 60 - irq 28 and irq 16). Why do I think 
OS>so?
OS>
OS>1. ips works fine even when there's no bge in kernel (I suppose irq 16 is 
OS>not activated in this case). I suppose this should mean that interrupts 
OS>are properly delivered to ips driver.
OS>
OS>2. I've added debug printf to bge_intr and in single mode when preemption 
OS>is disabled I see exact the same number of interrupts delivered to ips 
OS>(checked counters with showintrcnt) and to bge (incorrectly delivered - 
OS>bge is not in UP state and bge registers say "no interrupt"). 
OS>
OS>This seems really strange to me, how can this be possible? Is there any 
OS>way to fix this?
OS>
OS>On Tue, 28 Mar 2006, Oleg Sharoiko wrote:
OS>
OS>OS>
OS>OS>On Tue, 28 Mar 2006, John Baldwin wrote:
OS>OS>
OS>OS>JB>Is this an Intel box?
OS>OS>
OS>OS>This is IBM xSeries 226. Motherboard is labeled 'MS 9151 Ver.1'. It 
OS>OS>appears to be MSI E7525 Master-S2 or a modification of it. I couldn't find 
OS>OS>information about it in the net. Just few short notes and nothing else. I 
OS>OS>suppose it's based on Intel's E7525 chipset.
OS>OS>
OS>OS>JB>It only detects it if the interrupt line is constantly on.  If the interrupt
OS>OS>JB>line goes off for a little bit and then back on often enough the istorm logic
OS>OS>JB>won't kick in.  You can try lowering the threshold via the tunable to see if
OS>OS>JB>you can get it to kick in perhaps.
OS>OS>
OS>OS>No luck here.
OS>OS>
OS>OS>
OS>
OS>

-- 
Oleg Sharoiko.
Software and Network Engineer
Computer Center of Rostov State University.