Re: enbd_ioctl 2.6.9 kernel fails to modprobe enbd - seg fault

Dear Peter,

On Fri, 2006-03-03 at 16:06 +0100, Peter T. Breuer wrote:
> "Also sprach scunacc:"
> > enbd-server  4086: #1765 do_srv_ioctl: (ERROR) server returns error
> > reply (-22) for ioctl type 0x4004ab01
> > 
> > enbd-server  4086: #1765 do_srv_ioctl: (ERROR) server returns error
> > reply (-22) for ioctl type 0x4004ab01
> > 
> > in the make test o/p
> 
> Don't worry about it. It's supposed to error.

Oh. OK.

> > writing....5%....10%....15%....20%....25%....30%....35%....40%....45%....50%....55%....60%....65%....70%....75%....80%....85%....90%....95%....done
> > test 1 success:  0 incorrect blocks
> > flushing buffers..
> > done
> > enbd-client  4079: client (-1) opened device /dev/nda
> > flushing buffers..done
> > reading....5%....10%....15%....20%....25%....30%....35%....40%....45%....50%....55%....60%....65%....70%....75%....80%....85%....90%....95%....done
> > test 2 success:  0 incorrect blocks
> > flushing buffers..done
> > flushing buffers..done
> > writing....5%....10%....15%....20%....25%....30%....35%....40%....45%....50%....55%....60%....65%....70%....75%....80%....85%....90%....95%....done
> > test 3 success:  0 incorrect blocks
> > flushing buffers..done
> > flushing buffers..done
> > reading....5%....10%....15%....20%....25%....30%....35%....40%....45%....50%....55%....60%....65%....70%....75%....80%....85%....90%....95%....done
> > test 4 success:  0 incorrect blocks
> > flushing buffers..done
> > flushing buffers..done
> > seeking and
> > writing....5%....10%....15%....20%....25%....30%....35%....40%....45%....50%....55%....60%....65%....70%....75%....80%....85%....90%....95%....done
> > flushing buffers..nbd/hash  4084: hash saturation reached at 2337
> > entries
> > nbd/hash  4084: hash hiwater set at 2337/2337
> > nbd/hash  4084: hash lowater set at 2104/2337
> > done
> > reading and
> > checking....5%....10%....15%....20%....25%....30%....35%....40%....45%....50%....55%....60%....65%....70%....75%....80%....85%....90%....95%....
> > (1000KB written in 0.048705s, 20.050559 MB/s, 41063.545837 seeks/s)
> > (1000KB read in 0.011381s, 85.806388 MB/s)done
> > test 5 success:  0 incorrect blocks
> > flushing buffers..done
> 
> All seems fine.

Good to know.

> > enbd-server  4086: #1765 do_srv_ioctl: (ERROR) server returns error
> > reply (-22) for ioctl type 0x4004ab01
> > ioctl 1 (write direct) testing write illegal ..OK (fail Invalid argument
> > with result -1)
> > ioctl 1 (write direct) testing write 0x7ff192bc ..OK (success with
> > result 0)
> > ioctl 2 (read indirect) testing read 0x7ff192bc ..OK (success with
> > result 0)
> > enbd-server  4087: #1765 do_srv_ioctl: (ERROR) server returns error
> > reply (-22) for ioctl type 0xc004ab12
> > ioctl 3 (write + read indirect) testing illegal ..OK (fail Invalid
> > argument with result -1)
> > ioctl 3 (write + read indirect) testing w/r 0x62174006 ..OK (wrote
> > 0x62174006 read 0x9de8bff9 expected 0x9de8bff9)
> > ioctl 4 (large read indirect) testing ..OK (success with result 0)
> > ioctl 5 (large write + read indirect) testing ..OK (success with result
> > 0)
> > done
> 
> All OK.

That's great!

> What did you do?

As per my previous email, - I made the change to the gcc version from 4
to 0400 so it did that comparison "properly" (!) and it worked.

Yup.

I'm now setting it up to run across a Beowulf cluster...

I'll let you know how it goes.

I appreciate you sticking with this and prodding in the right direction.

Kind regards

Derek.