From: Christoph Hellwig Subject: Re: [PATCH 18/23] init: open code setting up stdin/stdout/stderr Date: Mon, 27 Jul 2020 08:49:57 +0200 Message-ID: <20200727064957.GB2317@lst.de> References: <20200714190427.4332-1-hch@lst.de> <20200714190427.4332-19-hch@lst.de> <20200727030534.GD795125@ZenIV.linux.org.uk> <20200727062425.GA2005@lst.de> <366377E2-6F19-45E1-9285-CFA5E660C6B5@zytor.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <366377E2-6F19-45E1-9285-CFA5E660C6B5@zytor.com> Sender: linux-kernel-owner@vger.kernel.org To: hpa@zytor.com Cc: Christoph Hellwig , Al Viro , linux-kernel@vger.kernel.org, Song Liu , Linus Torvalds , linux-raid@vger.kernel.org, linux-fsdevel@vger.kernel.org List-Id: linux-raid.ids On Sun, Jul 26, 2020 at 11:36:15PM -0700, hpa@zytor.com wrote: > >Err, why? The changes have been pretty simple, and I'd rather not come > >up with new crazy ways just to make things complicated. > > Why? To avoid this neverending avalanche of special interfaces and layering violations. Neatly deals with non-contiguous contents and initramfs in device memory, etc. etc. etc. I don't think it will be all that simple. But given that linux-next is just missing one series Al was already ok with to kill off set_fs entirely for about half of our architectures I'd rather go ahead with this series. If you can send a series mapping user memory that actually cleans things up on top of it I'm not going to complain, but I'm not sure it really is going to be all that much cleaner.