Even more hdparm weirdness !
Well, the hdparm issue turns out to be real fun ;)
stracing the hdparm yields an illegal instruction, causing the program to stop
( as seen in the previous post ).
Now, I donloaded to correct source to hdparm 3.6 and applied the 3.6-1 diff to
it, in order to recreate the failure from source ( as opposed to binary-deb ).
Now for the _real strange_ stuff. I unpack the source, apply the patch and run the
resulting executable, right in the directory of the source.
Of course, the hdparm fails with the illegal instruction, and leaves the binary
corrupted. But, in addition to that, it corrupts *every other file* in that directory as
well. It leaves the filesizes intact, but zaps everything to NULL.
Now, I'm confused ;)
Thomas Weyergraf email@example.com
My Favorite IA64 Opcode-guess ( see arch/ia64/lib/memset.S )
"br.ret.spnt.few" - got back from getting beer, did not spend a lot.