[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Deadlock free routing
- To: occam-com@xxxxxxxxx (Receipt notification requested)
- Subject: Deadlock free routing
- From: Øyvind Teig <Oyvind.Teig@xxxxxxxxxxxx>
- Date: 26 Jun 1997 12:39:18 Z
- Alternate-recipient: Allowed
- Conversion: Allowed
- Disclose-recipients: Prohibited
- Original-encoded-information-types: IA5-Text
- Original-x400-recipients: non-disclosure;
- Pp-warning: Parse error in original version of preceding X400-Recipients line
- Priority: normal
- Return-receipt-to: "Øyvind Teig" <Oyvind.Teig@xxxxxxxxxxxx>
- X400-content-type: P2-1984 ( 2 )
- X400-mts-identifier: [/c=NO/admd=TELEMAX/prmd=autronica/; 3087 97/06/26 14:39]
- X400-originator: Oyvind.Teig@xxxxxxxxxxxx
- X400-received: by /c=no/admd=telemax/prmd=internet/; converted ( IA5-Text); Relayed; 26 Jun 1997 14:41:11 +0200
- X400-received: by /c=NO/admd=TELEMAX/; Relayed; 26 Jun 1997 12:40:05 Z
- X400-received: by /c=NO/admd=TELEMAX/prmd=autronica/; Relayed; 26 Jun 1997 12:39:18 Z
- X400-received: by mta MHNORWAY in /c=no/admd=telemax/prmd=internet/; converted ( IA5-Text); Relayed; 26 Jun 1997 14:41:11 +0200
Deadlock free routing
We have a discussion about deadlock-free routing.
Sirs 26 June 1997
In order to guarantee deadlock-free routing:
Q1. What is the relationship between how many buffers we need
and the blocking characteristics of application-code?
Specifically (same questions elaborated):
Q2. Can we have deadlock-free routing and blocking communication
at application level? (Assuming that application level
doesn't deadlock).
Q3. VCR under occam is deadlock free. Occam is blocking on communication.
Does VCR depend on a certain amount of buffering to achieve
the deadlock free properties?
Q3. We have a system here that is written in C altogether.
Packets are routed, but it is the assumption that sufficient
buffers exist that guarantees deadlock free routing.
Could this unbounded buffer solution have been swapped
with a bounded buffer (or zero buffer) - and blocking communication
at application level, and still guaranteed deadlock freedom?
Q4. Are there any more questions I should have asked?
Cheers, > 0yvind Teig, Autronica, Trondheim, Norway <
Oyvind > Oyvind.Teig@xxxxxxxxxxxx <
> Tel.: +47 73 58 12 68 <
> Fax.: +47 73 91 93 20 <