From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Tue, 16 Oct 2001 07:44:05 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Tue, 16 Oct 2001 07:43:55 -0400 Received: from spanner.eng.cam.ac.uk ([129.169.8.9]:41225 "EHLO spanner.eng.cam.ac.uk") by vger.kernel.org with ESMTP id ; Tue, 16 Oct 2001 07:43:42 -0400 Message-ID: X-Mailer: XFMail 1.3.2pre1 [p0] on Linux X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit MIME-Version: 1.0 Date: Tue, 16 Oct 2001 12:45:47 +0100 (BST) From: James Scott To: linux-kernel@vger.kernel.org Subject: Bug warning in skbuff.c Cc: James Scott Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi all. (n.b. linux-2.4.10) I am working on a new type of network at Cambridge University, and am writing a linux kernel driver for the NIC we've built. I get the following in /var/log/messages once per packet: Oct 16 12:27:42 cerealkiller kernel: Warning: kfree_skb on hard IRQ c300c6d0 because I am calling dev_kfree_skb from a hard IRQ for every outgoing packet (the TX-complete interrupt). I know that I can (and perhaps should) do as much as possible in a bottom half tasklet, however right now this is all experimental and "inconsiderate" code. What I want to know is - is the use of kfree_skb on a hard IRQ actually that evil? Is the "bug" just to remind me to be considerate, or is it to point out real problems that might occur (e.g. in OOM cases?) Thanks, James PS Please mail me explicitly (Im not on linux-kernel). Thanks for your help.