Re: Thou shalt have no other gods before the ANSI C standard
From: Brian Inglis (Brian.Inglis_at_SystematicSW.Invalid)
Date: 02/12/05
- Next message: Douglas A. Gwyn: "Re: Thou shalt have no other gods before the ANSI C standard"
- Previous message: Brian Inglis: "Re: Thou shalt have no other gods before the ANSI C standard"
- In reply to: David Wagner: "Re: Thou shalt have no other gods before the ANSI C standard"
- Next in thread: David Wagner: "Re: Thou shalt have no other gods before the ANSI C standard"
- Reply: David Wagner: "Re: Thou shalt have no other gods before the ANSI C standard"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: Sat, 12 Feb 2005 00:58:44 GMT
On Fri, 11 Feb 2005 20:22:20 +0000 (UTC) in alt.folklore.computers,
daw@taverner.cs.berkeley.edu (David Wagner) wrote:
>Patrick Scheible wrote:
>>There's smaller hardware in embedded systems and
>>DSPs, 64-bit hardware in many new servers and desktops now.
>
>Let's take those one by one.
>
>Embedded systems: Ok, we're having a discussion on that elsewhere.
>Cellphones are a good example. But I want to find out which of DJB's
>assumptions people are worried might lead to security holes. The only
>one I immediately spotted was 32-bit vs 16-bit int. Any others?
The assumption of "obsolete trivia" in the C standard, despite decades
of preaching that "not all the world's a VAX", basically DJB's
assumption.
>DSPs: How many programers ever write a spot of DSP code in their life?
>And are those DSPs connected to the Internet?
The internet is interconnected by DSPs: what do you think is on the
network chips/cards, routers, cable modems, dialup modems, etc.
>And are those DSPs connected to the Internet? It would surprise me
>if this is a top risk factor for security. For the tiny minority of
>programmers who write code for non-32-bit DSPs, sure, they shouldn't
>use DJB's assumptions -- but most programmers will never write a line
>of security-critical code for a non-32-bit DSP.
What do Cisco et al use for their high speed gear?
>64-bit servers: But DJB already talked about this. For instance, he
>mentions that those 64-bit servers use a 32-bit int.
But not 32 bit addresses any longer; 32 bit fast FP is becoming more
widespread with SIMD operations, so the easy to use 64 bit double is
going to lose favour with the performance crowd including all those
doing graphics; 128 bit CPUs are coming out so the 32 bit int is going
to be a short soon.
--
Thanks. Take care, Brian Inglis Calgary, Alberta, Canada
Brian.Inglis@CSi.com (Brian[dot]Inglis{at}SystematicSW[dot]ab[dot]ca)
fake address use address above to reply
- Next message: Douglas A. Gwyn: "Re: Thou shalt have no other gods before the ANSI C standard"
- Previous message: Brian Inglis: "Re: Thou shalt have no other gods before the ANSI C standard"
- In reply to: David Wagner: "Re: Thou shalt have no other gods before the ANSI C standard"
- Next in thread: David Wagner: "Re: Thou shalt have no other gods before the ANSI C standard"
- Reply: David Wagner: "Re: Thou shalt have no other gods before the ANSI C standard"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|