Re: port-amd64/53155: OS wedges after <12h uptime when >2 bnx network interfaces in use

[Havard Eidnes <he%NetBSD.org@localhost>]

The following reply was made to PR port-amd64/53155; it has been noted by GNATS.

From: Havard Eidnes <he%NetBSD.org@localhost>
To: gnats-bugs%NetBSD.org@localhost, hannken%eis.cs.tu-bs.de@localhost
Cc: port-amd64-maintainer%netbsd.org@localhost
Subject: Re: port-amd64/53155: OS wedges after <12h uptime when >2 bnx
 network interfaces in use
Date: Thu, 17 May 2018 15:54:45 +0200 (CEST)

 >  >> Could you show the stack trace of each softser/N? In particular softser/0
 >  >> looks running and is a suspect.
 >  >
 >  > Getting the tracebacks of these LWPs appears to be somewhat
 >  > problematic; I don't get much useful info from crash:
 >  <snip>
 >  >
 >  > 0    >   6 7   0       200   fffffe822f6eb440          softser/0
 >  > 0    >   5 7   0       200   fffffe822f6eb860          softclk/0
 >
 >  Crash reports two threads on cpu 0.
 >
 >  If you get this type of hang again please try to get the corresponding
 >  stack traces from DDB: if softXXX/N is on cpu (>) use "mach cpu N; bt"
 
 Hmm, looking back at earlier submitted information where I dupmd
 various information from DDB, I find this identical situation:
 
 0    >   6 7   0       200   fffffe822f6eb440          softser/0
 0    >   5 7   0       200   fffffe822f6eb860          softclk/0
 
 and the backtraces for these were:
 
 db{0}> trace/a fffffe822f6eb440
 trace: pid 0 lid 6 at 0xffff80008f69bc30
 breakpoint() at netbsd:breakpoint+0x5
 comintr() at netbsd:comintr+0x562
 Xhandle_ioapic_edge13() at netbsd:Xhandle_ioapic_edge13+0x66
 --- interrupt ---
 Xspllower() at netbsd:Xspllower+0xe
 uhci_intr() at netbsd:uhci_intr+0x19
 intr_biglock_wrapper() at netbsd:intr_biglock_wrapper+0x1d
 Xhandle_ioapic_level5() at netbsd:Xhandle_ioapic_level5+0x6a
 --- interrupt ---
 Xspllower() at netbsd:Xspllower+0xe
 DDB lost frame for netbsd:Xsoftintr+0x4f, trying 0xffff80008f69bff0
 Xsoftintr() at netbsd:Xsoftintr+0x4f
 --- interrupt ---
 0:
 db{0}> 
 db{0}> trace/a fffffe822f6eb860
 trace: pid 0 lid 5 at 0xffff80009edadd60
 Xsoftintr() at netbsd:Xsoftintr+0x4f
 --- interrupt ---
 ffff80009edadf00:
 db{0}> 
 
 Not sure this provides any more useful information, though...
 
 Regards,
 
 - Havard