From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <9803121733.AA16133@clio.rice.edu>
Subject: Re: Temporary files considered unsafe
To: Vik DOT Heyndrickx AT rug DOT ac DOT be (Vik Heyndrickx)
Date: Thu, 12 Mar 1998 11:33:44 -0600 (CST)
Cc: eliz AT is DOT elta DOT co DOT il, bill AT taniwha DOT tssc DOT co DOT nz, djgpp-workers AT delorie DOT com
In-Reply-To: <3507CAC3.CF9@rug.ac.be> from "Vik Heyndrickx" at Mar 12, 98 12:45:07 pm
Content-Type: text
Precedence: bulk

> First an update on the STR instruction:
> - it is valid regardless of CPL, according to official documentation and
> according to tests I performed.

True, it is an unprotected instruction.

> The second: all DPMI programs share the same processor task under W95,
> meaning that a prg's TSS is definitely not a good PID. Nor is the LDT
> selector.

True.  Things which are most likely to be unique:

1) The cs selector value 
2) The base virtual address of the cs selector
3) PSP
4) clock at startup

A combination (sum? xor?) of these might be best for the PID.  There may
be a pathaloigcal 1 in a million odds with something like this, but
get a grip.

> Packing different values into an 8+3 filename is not really a problem
> (assuming that the information content of these values together is not
> larger than 39^8).

The file name should be a unique PID plus some counter or something, 
which should work fine with 8.3 names