Hi all,
I was recently writing a log parser and discoverd a large memory leak in my code. When I found the problem I was rather surprised. Hopefully someone can explain the details to me.

Here's where the leak was:

while (my $line = <FILE>) { ... }
[download]

while (<FILE>) { my $line = $_; ... }
[download]

Apparently in the first example a new block of memory was being allocated to each $line, but was never reclaimed. I was unable to determine whether perl ever reclaimed the memory, reclaimed after the end of the loop, or at the end ot the process. According to top (on FreeBSD) the memory was never reclaimed, but that could just be a peculiarity of how top reports memory usage.

I always thought the above two lines were functionally and lexically identical. Can anyone explain why they behave differently?

Thanks,
-M