From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 20 Dec 2001 17:08:57 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 20 Dec 2001 17:08:47 -0500 Received: from lacrosse.corp.redhat.com ([12.107.208.154]:40987 "EHLO lacrosse.corp.redhat.com") by vger.kernel.org with ESMTP id ; Thu, 20 Dec 2001 17:08:41 -0500 Date: Thu, 20 Dec 2001 17:08:36 -0500 From: Benjamin LaHaise To: torvalds@transmeta.com, Linux Kernel Subject: RFC: bio_vec bits, kmap_atomic Message-ID: <20011220170836.C6276@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hello, It looks like 2.5.2-pre1 is working towards recreating all the kvec helpers that I already wrote and offered in the kvec series of patches. Do you want to continue recreating code, or would you like to merge? Another issue: would you accept a patch to convert kmap_atomic into a more kmap-like function that only allows 2 kmaps per "context" for bounce copies? As it stands, the fact that many kmap_atomic users have to disable interrupts is a gross wart that makes things stupidly slow. Oh, and by context I'm thinking of kernel mode, bottom half, softirq and interrupt. This probably means hard limiting the maximum irq depth to 8 or so, but I suspect we need to do that -ben -- Fish.