netdev
[Top] [All Lists]

Re: Route cache performance under stress

To: Robert.Olsson@xxxxxxxxxxx
Subject: Re: Route cache performance under stress
From: "David S. Miller" <davem@xxxxxxxxxx>
Date: Mon, 09 Jun 2003 10:19:59 -0700 (PDT)
Cc: sim@xxxxxxxxxxxxx, netdev@xxxxxxxxxxx, linux-net@xxxxxxxxxxxxxxx, kuznet@xxxxxxxxxxxxx
In-reply-to: <16091.11735.721251.925522@robur.slu.se>
References: <20030522.153330.74735095.davem@redhat.com> <20030529205125.GA30058@netnation.com> <16091.11735.721251.925522@robur.slu.se>
Sender: netdev-bounce@xxxxxxxxxxx
   From: Robert Olsson <Robert.Olsson@xxxxxxxxxxx>
   Date: Mon, 2 Jun 2003 12:58:31 +0200
   
    And later GC have to remove all enties with spin_lock_bh hold (no
    packet processing runs). I see packet drops exactly when GC
    runs. Tuning GC might help but it's something to observe.

Please note, in 2.5.x, holding of this lock on one cpu does
not prevent packet processing (even for routes on same hash
chain) on another cpu because we use RCU there.

<Prev in Thread] Current Thread [Next in Thread>