From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752391AbZHaRT7 (ORCPT ); Mon, 31 Aug 2009 13:19:59 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751934AbZHaRT7 (ORCPT ); Mon, 31 Aug 2009 13:19:59 -0400 Received: from mail-ew0-f206.google.com ([209.85.219.206]:45483 "EHLO mail-ew0-f206.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751046AbZHaRT6 (ORCPT ); Mon, 31 Aug 2009 13:19:58 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=date:from:to:cc:subject:message-id:references:mime-version :content-type:content-disposition:in-reply-to:user-agent; b=p4K604JRWEsqnBExFLHxt0oBztqqqerpXgg3AWNDzA249Kh3lUx7tMx5EN9YKxl+uS zetSS30iwgvuihshdyzNAV/Hj10bgBHbvv3YxYoB6fFCUAVeJ9aJOGUzd2NiquIRmL2r 9BbfpuoC20BVJCXXPShFRyRskUSloqMQUpOFM= Date: Mon, 31 Aug 2009 21:19:56 +0400 From: Alexey Dobriyan To: Arnd Bergmann Cc: Stefani Seibold , Christoph Hellwig , linux-kernel , Andrew Morton Subject: Re: [PATCH] Fix proc_file_write missing ppos update Message-ID: <20090831171956.GA3036@x200.localdomain> References: <1251563892.13287.11.camel@wall-e> <1251659144.20987.18.camel@wall-e> <20090831063334.GB7821@x200.localdomain> <200908311744.52972.arnd@arndb.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200908311744.52972.arnd@arndb.de> User-Agent: Mutt/1.5.19 (2009-01-05) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, Aug 31, 2009 at 05:44:52PM +0200, Arnd Bergmann wrote: > On Monday 31 August 2009, Alexey Dobriyan wrote: > > Out-of-tree argument means almost nothing. > > > > ... > > > > You'd better start converting to struct file_operations::write now. > > Maybe a purely mechanical conversion to file_operations would be a nice first > step, just so we can remove read_proc and write_proc? Taking scsi_proc.c > as an example, this should be really straightforward and lets us get > rid of the write_proc and read_proc callbacks from proc_dir_entry without > having to rewrite all the remaining drivers that use it. > > Obviously, someone who understands the specific driver code better should > then clean up the code by converting to seq_file operations or something > else that is appropriate there. I don't like any churn in that area. Everything is ready for ->proc_fops usage. So we should simply convert users away. Actually most of SCSI and IDE is already done, just wasn't sent yet. After conversion, ->read_proc, ->write_proc, create_proc_read_entry(), create_proc_entry() will become deprecated for a while, then gone.