Re: syscall messaging interface API

[Adrian Chadd <adrian@xxxxxxxxxxx>]

In article <20030722220937.GA40696@xxxxxxxxxxxxx>, Diane Bruce wrote:
> In article <200307231926.h6NJQppi038541@xxxxxxxxxxxxxxxxxxxx>,
>   Matthew Dillon wrote:
>> >     Here is my idea for the system messaging interface.  I will use a
>> >     new trap gate (0x81) to implement it, because it occurs to me that
> ...>     and the contents of the message).  And a new trap gate isolates us from
>> >     the old syscall mechanism.
>> >
>> >     int 0x81 to dispatch, arguments in eax, ecx, edx, return value in eax.
> 
>> How will that affect optimisation w/ the limited register set on ia32?
> 
>   You can pay me now, or you can pay me later.
>   Doing a context (a full stateful context) is isomorphic to a message
> passing scheme. A message is going to cost you less in needed context saving
> than a full context.
>   Anyway, I explained the other nasty consequences to you elsewhere already...

Yeah, now that the fever has worn off I understand the situation better.
Sorry.





Adrian

-- 
Adrian Chadd			<angryskul> learning is bad
<adrian@xxxxxxxxxxx>		  <angryskul> it just makes the people around you dumber
(angryskul == alfred@irc)	    <angryskul> :(