To quote Frank on 2002/04/09 :

The memory leak is not in the unit, but in GPC itself. It's a known problem with some string operations (especially concatenation) in a loop. (If you can read /proc/n/maps, you'll see that it's the stack, not the heap that grows.) Moving the thing out of the loop (syntactically, e.g. in a subroutine) is a work-around.

Joe.

-----Original Message----- From: CBFalconer [SMTP:cbfalconer@yahoo.com] Sent: Monday, June 24, 2002 12:45 PM To: gpc@gnu.de Subject: Re: list trouble solved

"da Silva, Joe" wrote:

...

I don't want to enter a debate over this, since clearly not everyone has the same concerns/priorities. However, I regard string handling memory leakage to be more serious than lack of range checking - but that's just my opinion, FWIW. (Which is not to say that range checking isn't also important, of course.)

Are there such leaks, and how do they get triggered (if you know).

-- Chuck F (cbfalconer@yahoo.com) (cbfalconer@worldnet.att.net) Available for consulting/temporary embedded and systems. http://cbfalconer.home.att.net USE worldnet address!