Date: Sun, 29 Jun 2025 20:34:33 -0700
From: Stephen Gildea <[email protected]>
Message-ID: <[email protected]>
| All the leaks I have investigated have been benign leaks at
| the end of main(). But there are so many false positives that
| I cannot be sure there aren't some real leaks hiding in all
| the noise.
Why would anyone ever care? At least for the vast majority of
nmh binaries. MH (nmh) applications tend to run for just a few
seconds and exit, anything accidentally "forgotten" along the way
will be cleaned up then. It is quite likely more effecient (though
testing of that would be needed to know) to avoid releasing almost
anything, than go through the bookkeeping (including internally in
the malloc implementation) to free everything. The apps might use
a bit more RAM, for a few seconds ...
I have certainly never encountered anything I'd ever consider a
problem in this area, and I have some BIG folders that have been
collecting messages for decades that I sort, pack, pick, mark, scan, ...
Seems to me that this would be just a useless academic exercise,
just "because we always should" for no practical benefit, and
I wouldn't like to see the code being mangled to make it simpler
to achieve.
kre
