>> If I fill a file with much a characters and then I see something
>> others at the end of teh file:

>> aaaaaaaaaaaaaaaaaaaaaaa
>> @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
>> @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
>> @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
>> @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
> These are NULL characters: all-zero bytes.

Depends.  @ is not all-zero; most things that display anything for NUL
display ^@, not just @.  (I have, though, occasionally seen peculiar
failure modes that end up displaying a bare @.)

Not that NUL versus @ really makes much difference here.

> So this is a bug, and what the bug is is that on filesystem full, the
> very last block or fragment allocated is not filled with the data
> that the caller of write() wrote, but it is still left allocated to
> the file.

Probably. :-)  It could also be whatever happened to be hanging around
in some kernel memory or some such. :-(

Once I get a machine pushed the rest of the way to -current I may be
able to have a bash at this - but I don't know when that will be....

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B