Re: MVME2100 locks up during interrupt setup

[Eric Norum <norume at aps dot anl dot gov>]

On Jul 28, 2005, at 2:34 PM, Till Straumann wrote:

> openpic_init() does poll indefinitely
>
> while ( openpic_read() & OPENPIC_ACTIVITY )
>    /* wait until openpic idle */ ;
>
> Either you experience a hardware problem or a
> caching/volatility bug. Disassembly helps finding
> the latter...


Yep -- that's where it's stuck:

WARNING: unable to allocate page table, keeping DBAT0
WARNING: unable to setup page tables VME bridge must share PCI space
openpic_read(0x00040FFF) = 0x01218196
openpic_read(0x0004107F) = 0x0181A878
OpenPIC Version ? (2 CPUs and 290 IRQ sources) at 0x0003FFFF
OpenPIC Vendor 120 (Unknown), Device 168 (Unknown), Stepping 129
openpic_read(0x000410EF) = 0x0181A878
OpenPIC timer frequency is 25274488 Hz
openpic_read(0x0004111F) = 0x00C03878
openpic_write(0x0004111F, 0x80C03878)
openpic_read(0x0004111F) = 0x80C03878
openpic_read(0x0004111F) = 0x80C03878
openpic_write(0x0004111F, 0x80C03840)
openpic_write(0x0004112F, 0x00000000)
openpic_read(0x0004115F) = 0x0181A878
openpic_write(0x0004115F, 0x8181A878)
openpic_read(0x0004115F) = 0x8181A878
openpic_read(0x0004115F) = 0x8181A878
openpic_write(0x0004115F, 0x8180A841)
openpic_write(0x0004116F, 0x00000000)
openpic_read(0x0004119F) = 0xF0FF4B0A
openpic_write(0x0004119F, 0xF0FF4B0A)
openpic_read(0x0004119F) = 0xF0FF4B0A
openpic_read(0x0004119F) = 0xF0FF4B0A
openpic_read(0x0004119F) = 0xF0FF4B0A
openpic_read(0x0004119F) = 0xF0FF4B0A
openpic_read(0x0004119F) = 0xF0FF4B0A
openpic_read(0x0004119F) = 0xF0FF4B0A
openpic_read(0x0004119F) = 0xF0FF4B0A
openpic_read(0x0004119F) = 0xF0FF4B0A
.
.
.



I doubt that this is a caching/volatility problem since the first two  
passes through the loop seemed to work fine (0x0004111F, 0x0004115F)  
but the third (0x0004119F) did not.
So, is there something peculiar about the MVME2100 that's causing  
this problem?

--
Eric Norum <norume@aps.anl.gov>
Advanced Photon Source
Argonne National Laboratory
(630) 252-4793