> I though I should send a send-pr on this problem and fix.

You probably should.

> It seems that it is necessary to get a ID to give to the pr.

It isn't; that is a somewhat confusing aspect of NetBSD having lifted
the gnats mechanism from elsewhere.  Let the ID default; it should
default to "net".

> I know from the list that sometimes there can be significant delay
> between recording a PR and having the solution migrate into the
> sources but that recently there has been a renewed effort to improve
> the situation.  What is the time frame from submission to inclusion
> in source code which can be expected if the fix is something simple
> like adding "* 2" to a line?

As far as I can tell the complexity of the fix has very little to do
with the promptness with which it gets handled.  For example, my PR
1356, which stops join from coredumping in some circumstances, is still
open, but 1366, which made fsck use fs_csaddr correctly, got applied.
They're both approximately the same level of complexity.

I conjecture that it really depends on when someone is working in that
area of the tree anyway and notices the PR sitting around.

					der Mouse

			    mouse@collatz.mcrcim.mcgill.edu