Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933163AbcKVNHo (ORCPT ); Tue, 22 Nov 2016 08:07:44 -0500 Received: from port70.net ([81.7.13.123]:60746 "EHLO port70.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932480AbcKVNHm (ORCPT ); Tue, 22 Nov 2016 08:07:42 -0500 Date: Tue, 22 Nov 2016 14:07:38 +0100 From: Szabolcs Nagy To: Dmitry Vyukov Cc: syzkaller , linux-api@vger.kernel.org, LKML , Michael Kerrisk-manpages , Thomas Gleixner , Sasha Levin , Mathieu Desnoyers , scientist@fb.com, Steven Rostedt , Arnd Bergmann , carlos@redhat.com, Kostya Serebryany , Mike Frysinger , Dave Jones , Tavis Ormandy Subject: Re: Formal description of system call interface Message-ID: <20161122130738.GT5749@port70.net> References: <20161107002839.GN5749@port70.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.6.0 (2016-04-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1306 Lines: 32 * Dmitry Vyukov [2016-11-21 16:03:04 +0100]: > On Mon, Nov 7, 2016 at 1:28 AM, Szabolcs Nagy wrote: > > * Dmitry Vyukov [2016-11-06 14:39:28 -0800]: > >> For the reference, current syzkaller descriptions are in txt files here: > >> https://github.com/google/syzkaller/tree/master/sys > > ... > >> Taking the opportunity, if you see that something is missing/wrong > >> in the descriptions of the subsystem you care about, or if it is not > >> described at all, fixes are welcome. > > > > abi variants are missing (abi variation makes a lot of > > syscall interface related work painful). > > > What exactly do you mean by "abi variants"? Is it architecture? > What exactly needs to be added to the descriptions support "abi variants? > abi variant is a supported syscall abi (linux supports several abis on the same arch: e.g. x86 has i386,x86_64,x32 one might also call different endianness a different abi, but i'd expect le/be to be handled together.) available syscalls, argument types and ordering can be different across abis i may be wrong, but i did not see those handled in sys.txt, looking at it now i see missing paddings in ipc structs (*_ds) so it may not work the way i assumed (these are types which have some abi variation).