reading IRIS tapes on a Sun

[Ruth Milner]

In v7n1, Liz Allen-Mitchell (elroy!grian!liz at ames.arc.nasa.gov) writes:
> We have some software on a cartridge tape that was written on a Silicon
> Graphics Workstation running UN*X.  We would like to read it on a Sun
> 3/60.

For at least the older Iris workstations, if the tape was written using
tar, you may be out of luck. The Iris tar swaps bytes as it writes them
onto the tape. This is not just a question of data format; plain ASCII
text is actually written out with the bytes swapped. We demonstrated this
by tarring a single simple ASCII file onto tape on the Iris, and then
dumping it on the Sun. You can quite clearly see the swapped bytes. You
may even be able to see it just by doing a tar tvf on it. I doubt you can
hurt anything by trying that. I believe /dev/rst0 was what we used
(QIC-11).

Possibly this has been fixed in more recent workstations or software
releases from Silicon Graphics. We did not have one handy to test with.
Silicon Graphics was apparently not aware of this idiosyncrasy until Mike
Peterson in our Chemistry Dept. reported it to them. He discovered that
his Apollo couldn't read his Iris' tapes. So he brought a tape from each
over to me to read on a Sun, with the result that the Iris tape was
garbage, but the Apollo tape was perfectly readable.

You might be able to get around it using dd with conv=swab and piping that
into tar.

Ruth Milner
Systems Manager
University of Toronto Physics