bug-cvs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: SSL?


From: Larry Jones
Subject: Re: SSL?
Date: Mon, 19 Mar 2001 15:46:22 -0500 (EST)

Derek R. Price writes:
> 
> The advice in the HACKING file is still to aim in that direction:
> 
>      Although this is a long-term goal, it also would be nice to move CVS
>      in the direction of reentrancy.  This reduces the size of the data
>      segment and will allow a multi-threaded server if that is desirable.
>      It is also useful to write the code so that it can be easily be made
>      reentrant later.  For example, if you need to pass data from a
>      Parse_Info caller to its callproc, you need a static variable.  But
>      use a single pointer so that when Parse_Info is fixed to pass along a
>      void * argument, then the code can easily use that argument.
> 
> Also, there do some to be some advantages to eventual reentrancy, mostly speed
> and memory profile advantages available from using threads instead of separate
> processes, I believe.

Correct.  When you mentioned a "reentrant server", I assumed you meant a
multi-threaded server that uses one thread per client connection as
opposed to the current architecture where there's an entire process per
client connection.  Note that the above is talking about CVS code in
general, not just the server code; and I generally agree with what it
says.

(Actually, the current architecture uses multiple processes per client
connection because it forks to execute individual CVS commands.  There
are many reasons for that, mostly having to do with the fact that
client/server was an afterthought rather than being designed in from the
start.  I think it makes a lot more sense to concentrate on fixing that
rather than worrying about a multi-threaded server.)

> Are you suggesting that reentrancy no longer be a long-term goal?  Simply that
> pratical considerations make a security design like Alexey's more important in
> the short term?

No, all I'm suggesting is that a multi-threaded server is not desirable
and thus should not be used as an argument against designs like
Alexey's.  (Although to be honest, it's because I agree with designs
like Alexey's that I don't think a multi-threaded server is desirable.)

-Larry Jones

He doesn't complain, but his self-righteousness sure gets on my nerves.
-- Calvin



reply via email to

[Prev in Thread] Current Thread [Next in Thread]