Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 26 Nov 2024 12:40:29 +0000
From:      bugzilla-noreply@freebsd.org
To:        bugs@FreeBSD.org
Subject:   [Bug 282984] [PATCH] pfctl: add -T `makezero` to touch pfras_tzero _only_ for non-zero entries
Message-ID:  <bug-282984-227-rLSnIFaOLV@https.bugs.freebsd.org/bugzilla/>
In-Reply-To: <bug-282984-227@https.bugs.freebsd.org/bugzilla/>
References:  <bug-282984-227@https.bugs.freebsd.org/bugzilla/>

next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D282984

--- Comment #2 from Leonid Evdokimov <leon+freebsd@darkk.net.ru> ---
(In reply to Kristof Provost from comment #1)

I also dislike the name, but I failed to come up with a better one. I'd
appreciate help here.

"touch" is a bad one as it actually changes counters.

"clear" sounds like an option, but might be confused with "flush", so IMO i=
t's
even worse than "makezero".

"mark" might come from mark-n-sweep gc, but it's confusing in this context.

"rearm" comes with watchdog/timer semantics that is kinda close, but still =
not
100% applicable.

"reset" is almost the winner, but TCP has already taken the word for RST. I
would say "reset" is my 2nd preferred option after "makezero".

So I'm kinda out of reasonable options.


> come with some performance penalty

It's probably my mistake made under assumption that counter-aware tables ha=
ve
different memory layout and handling.

I was unaware of pfr_get_astats() saying that

> It was possible to have a table without per-entry counters. Now they are=
=20
> always allocated, we just discard data when reading it if table is not=20
> configured to have counters.

ACK for tests & commit message. Will do.

--=20
You are receiving this mail because:
You are the assignee for the bug.=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-282984-227-rLSnIFaOLV>