From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from MTA-12-4.privateemail.com ([198.54.127.107]) by ewsd; Sat Aug 1 01:35:31 EDT 2020 Received: from mta-12.privateemail.com (localhost [127.0.0.1]) by mta-12.privateemail.com (Postfix) with ESMTP id A12048005A for <9front@9front.org>; Sat, 1 Aug 2020 01:35:20 -0400 (EDT) Received: from localhost (unknown [10.20.151.241]) by mta-12.privateemail.com (Postfix) with ESMTPA id B9DA380058 for <9front@9front.org>; Sat, 1 Aug 2020 05:35:19 +0000 (UTC) Date: Fri, 31 Jul 2020 22:35:12 -0700 From: Anthony Martin To: 9front@9front.org Subject: Re: [9front] CDC ACM nusb/serial driver Message-ID: <20200801053512.GA11004@alice> References: <5a20aa0c-7882-404c-ae40-6542542cea6f@www.fastmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <5a20aa0c-7882-404c-ae40-6542542cea6f@www.fastmail.com> X-Virus-Scanned: ClamAV using ClamSMTP List-ID: <9front.9front.org> List-Help: X-Glyph: ➈ X-Bullshit: descriptor grid storage high-performance-based layer Ethan Gardener once said: > i remember being told you can't use 9p over serial directly, you need > another layer. i can't clearly remember what that layer should do, > but i think it was preserve message boundaries. this was all long ago, > before 9front. Before the 9P2000 rewrite, the kernel required the underlying transport protocol to preserve message boundaries. You needed a little something extra between 9P and TCP, for example, but not for IL. To quote Charles, "all you need is a transport protocol that reliably preserves content and order". Cheers, Anthony