Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262275AbTHaJOJ (ORCPT ); Sun, 31 Aug 2003 05:14:09 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262309AbTHaJOJ (ORCPT ); Sun, 31 Aug 2003 05:14:09 -0400 Received: from AMontsouris-108-1-16-60.w80-15.abo.wanadoo.fr ([80.15.145.60]:22400 "EHLO paldrick.research.newtrade.nl") by vger.kernel.org with ESMTP id S262275AbTHaJOI (ORCPT ); Sun, 31 Aug 2003 05:14:08 -0400 From: Duncan Sands To: Alan Stern Subject: Re: 2.6.0-test4: uhci-hcd.c: "host controller process error", slab call trace Date: Sun, 31 Aug 2003 11:15:24 +0200 User-Agent: KMail/1.5.1 Cc: Fredrik Noring , Linux Kernel Development , Johannes Erdfelt , References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200308311115.24188.baldrick@wanadoo.fr> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 734 Lines: 18 > My feeling is that a reasonably large change may end up being the best > thing to do. In particular, we probably only need to have one QH per > queue, instead of one for each URB. But it'll be a while before that > stuff gets done. But won't that result in starvation of some endpoints in favour of those with vast numbers of urbs queued on them? At the moment the per-urb QHs mean that the hc works on only one urb per endpoint before moving on to the next endpoint. Ciao, Duncan. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/