Re: Memory leak?

[Jason Rumney]

Richard Stallman wrote:
> It is not surprising that that change would have this effect
> if a program uses the function this way.
>   
I'm not sure what you mean by "this way". I wonder if you are meaning 
the explanation of a different bug I gave last week where I discovered 
that the semantic-idle timer seemed to be continously running. But my 
explanation then was of the buggy behaviour, I had attributed the 
problem to a different change.

From my understanding of the workaround that has been posted, it works 
by changing a call to cancel-timer to instead increase the timeout 
period to 1 hour. I think that means that cancel-timer is not actually 
cancelling the timer in the case where the timer is repeating due to 
Emacs still being idle. Or maybe cancel-timer was never supposed to deal 
with this case, and the author of semantic only thought it was working 
because idle-timer functions were only running once each time Emacs 
became idle anyway.