Subject: Re: port-hp700/35239 (audiorecord(1) with harmony(4) on hp700 doesn't seem to wo
To: None <,,>
From: David H. Gutteridge <>
List: netbsd-bugs
Date: 08/21/2007 03:40:02
The following reply was made to PR port-hp700/35239; it has been noted by GNATS.

From: "David H. Gutteridge" <>
Subject: Re: port-hp700/35239 (audiorecord(1) with harmony(4) on hp700 doesn't seem to wo
Date: Mon, 20 Aug 2007 23:36:51 -0400

 Testing the change (pulled into the nick-csl-alignment branch)
 triggered a kernel panic:
 [disciple@arcusiv:disciple]$ audiorecord foo.wav
 panic: trap: uvm_fault(0xc47ab0, 0, 1): 14
 Stopped in pid 0.2 (system) at  netbsd:cpu_Debugger+0x10:       ldo     
 db> bt
 cpu_Debugger(a4a5ac,98000000,0,0) at netbsd:cpu_Debugger+0x10
 panic(a5c9d4,c47ab0,0,1) at netbsd:panic+0x1bc
 panic(f,5b95440,c70078,0) at netbsd:panic+0x14c
 trap(1,0,c70090,0) at netbsd:trap+0xb6c
 -- trap #15
 harmony_intr(31b9400,17,933ad7e0,1718400) at netbsd:harmony_intr+0x1b0
 harmony_intr(0,98000000,0,0) at netbsd:harmony_intr+0x158
 hp700_intr_dispatch(0,98000000,c03bd220,653ed60) at 
 spllower(c81c80,0,0,0) at netbsd:spllower+0x58
 mutex_vector_exit(653ed60,c87ed0,0,c87ec0) at netbsd:mutex_vector_exit+0x8c
 lwp_unlock(5c21d20,0,0,0) at 0x70a180
 mi_switch(5c21d20,5c21d20,6e0f2c,5c21d20) at netbsd:mi_switch+0x37c
 -- trap #0
 This is with source from August 19th.  I'm no kernel hacker, but I
 assume this has exposed an MD bug.  I tried testing with a DEBUG kernel
 instead of GENERIC, but that triggered another bug that caused a panic
 during the initial boot, so no luck getting more detailed output, at
 least not with a serial console...
 (I tested with nick-csl-alignment because I had read this:
 but if HEAD will work I could try that.)