[Disksim-users] Problems Using DIXTrac

[abpost at mpi-sws.org]

Hello everyone,

I have been using DiskSim for a while and have found it to be a very
useful tool.  I am currently trying to extract and additional model, and I
am running it to some problems.  I'm not sure if anyone can help me with
this, but if you have any ideas or advice, please let me know.  My
detailed experience is below.  My high level question for the list, is if
you
have done a successful extraction, what setup (Disk, OS version, Python,
gcc versions, etc) did you have? As I would be interested in replicating
the setup as best I can.  It is important to me that I have both an
extracted model, and the physical hardware so that
I can do experiments on both.

I would appreciate any help that someone might be able to provide,

Thanks,

Ansley

--- Details of problem below ------

To start I run the initial scripts to set up the directory, and
then run 'make extract', it takes a while to build the layout.mappings
file and then begins to make the passes over it.  The first problem
I have is the following, while running the python scripts I got the
following exception:

~/Projects/disksim-4.0/diskmodel/layout_g4_tools/g4_analysis.py:38:
DeprecationWarning: the sets module is deprecated
  from sets import Set
Traceback (most recent call last):
  File "~/Projects/disksim-4.0/diskmodel/layout_g4_tools/g4_analysis.py",
line 491, in <module>
    idx = outer(lin)
  File
"/home/abpost/Projects/disksim-4.0/diskmodel/layout_g4_tools/g4_analysis.py",
line 445, in outer
    if i > lastgood+1:
UnboundLocalError: local variable 'lastgood' referenced before assignment

Looking at the code, there does exist a code path for which this variable
is undeclared.  So to try fixing it, I've put a declaration in.  This
might be the wrong thing to do, but it allows me to go a bit further. 
After doing this the process manages to build a few more files, and then
fails once it tries to run dx_seek with the following error:

~/Projects/disksim-4.0/dixtrac/dx_seeks --mode=mtbrc --cyl=1 --cylstep=1
--cylmax=100 -d /dev/sg2 -l base-ibm-2.model -s seeks.1
Disk (/dev/sg2): IBM IC35L036UWD210-0, rev. S5CQ, SCSI-3, SN:VMZD4790
 ------- Internal error -------
diskinfo_get_rand_cyl: tried 50 times, giving up.
dx_seeks: dxtools.c:127: internal_error: Assertion `internal_error == 0'
failed.
Aborted

I've stepped through this with gdb, and don't see anything obvious, but
I'm not an expert.  This violation could be the result of adding the
variable in the previous step.