netdev
[Top] [All Lists]

Re: [PATCH] Break 'budget' dependency on netdev_max_backlog.

To: Ben Greear <greearb@xxxxxxxxxxxxxxx>
Subject: Re: [PATCH] Break 'budget' dependency on netdev_max_backlog.
From: jamal <hadi@xxxxxxxxxx>
Date: Tue, 22 Oct 2002 21:18:40 -0400 (EDT)
Cc: "'netdev@xxxxxxxxxxx'" <netdev@xxxxxxxxxxx>
In-reply-to: <3DB5F605.1010507@xxxxxxxxxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx

On Tue, 22 Oct 2002, Ben Greear wrote:

> Actually, with the changes the way they are, if you have a backlog value
> of 300, and the first NAPI driver reads 300 packets, then all the rest of the
> NAPI drivers will just throw away packets because the max-backlog has
> already been hit (if I understand things correctly).
>

The 300 is shared amongst all the drivers which have something on their
rx DMA. This is done on a roundrobin fashion.
Also all the drivers which are non-napi are given the same quota.

> If we make the backlog big enough to handle a large burst of incomming
> traffic w/out dropping them, then the first NAPI driver can hog a very
> large amount of CPU because its budget is now huge.  The rest of
> the devices in the poll loop will not be serviced in time, and will
> drop packets (I saw this when testing out the 4-port tulip driver with
> the NAPI patch).
>

I think you misunderstood. Look at the dev->weight.

cheers,
jamal


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