From: Arjan van de Ven <arjan@infradead.org>
To: Kyle Moffett <mrmacman_g4@mac.com>
Cc: "Randy.Dunlap" <rdunlap@xenotime.net>,
Pavel Machek <pavel@ucw.cz>,
linux-kernel@vger.kernel.org, nix@esperi.org.uk, rob@landley.net,
mmazur@kernel.pl, llh-discuss@lists.pld-linux.org
Subject: Re: [RFC][PATCH 1/2] Create initial kernel ABI header infrastructure
Date: Sun, 02 Apr 2006 15:09:04 +0200 [thread overview]
Message-ID: <1143983345.2994.11.camel@laptopd505.fenrus.org> (raw)
In-Reply-To: <5AC6F01C-8010-4A28-A303-52F7F11B84BF@mac.com>
On Sun, 2006-04-02 at 00:53 -0500, Kyle Moffett wrote:
> deal with compatibility: the kernel/userspace boundary. I'd like to
> strictly define all of that compatibility within a single set of
> _upstream_ headers in linux/include/linux-kabi or similar.
yes that was what I was hoping you meant
>
> But returning to your statements above, user code in that (C
> standards) context means anything not standard-specified.
I think you misread the C standard then. This only applies when you ONLY
include the headers defined in the standard, and no others.
> Basically,
> from userspace, a program that includes _any_ standard-specified
> header must bring in *only* those symbols specified by the standard
> as being present in that header, as well as any implementation-
> defined double-underscore symbols.
if you follow this argument then ANY library, be it glibc or gtk or
mysql or .. can only add __ things. That is not the case. Think of
"non-indirect" kernel header includes as including a header for a
library.
Maybe an example is needed: include/mtd/mtd-abi.h
Well it's a slightly bad example since it uses stdint.h types not __u64
types. But assume it used __u64 types, then this header would be
perfectly fine, including from a standards point of view.
> So unless I "#include
> <stdint.h>", my user program can typedef uint32_t to anything it
> wants, even a struct, if it likes.
yes this the example where __u64 needs underscores, and headers need to
use __u64 as they have to today.
> I've actually done that before on
> a very strange Cray platform with non-GCC compiler and OS that only
> supported 64-bit integers (a char was a 64-bit integer with ops to
> mask out the upper 56 bits), just to make bzip2 work as expected.
> The same rules apply for _every_ other standard-specified header
> (stdlib.h, stdio.h, sys/types.h, etc).
exactly. Nowhere is linux/* or asm/* in the standards as specified.
> to get access to them. Unfortunately this made it impossible to use
> those headers in userspace.
ok now I'm confused. How is providing FUNCTIONALITY such as setting
bitops the right thing for such headers? I would argue that the clean
headers should absolutely NOT do that but only provide the INTERFACE.
(this is also there for a license angle btw, if you actually start to
provide large bodies of code this way the lawyers will get unhappy wrt
the GPL; it's a LOT better to strictly do interface things only, eg
types, defines and prototypes)
In general:
I think you're focusing WAY too much on the __u64 kind of thing, which
is the vast minority of the ABI work, the ioctl structs are the big one.
For the __u64 kind of thing I would strongly suggest that you stick to
the currently already used symbols; anything existing knows to stay
clear of those already, so things are fine there so lets keep it simple
for those.
next prev parent reply other threads:[~2006-04-02 13:09 UTC|newest]
Thread overview: 102+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-03-14 15:19 [ANNOUNCE] linux-libc-headers dead Mariusz Mazur
2006-03-14 15:28 ` Ismail Donmez
2006-03-16 8:37 ` [llh-announce] " Nigel Kukard
2006-03-16 20:20 ` Jan Engelhardt
2006-03-16 20:42 ` Dan Kegel
2006-03-17 7:56 ` DervishD
2006-03-23 17:11 ` State of userland headers Mariusz Mazur
2006-03-23 23:04 ` Rob Landley
2006-03-26 13:12 ` Mariusz Mazur
2006-03-26 20:59 ` Rob Landley
2006-03-24 18:51 ` Kyle Moffett
2006-03-24 21:23 ` Rob Landley
[not found] ` <878xqzpl8g.fsf@hades.wkstn.nix>
2006-03-24 22:46 ` Kyle Moffett
2006-03-24 23:01 ` Randy.Dunlap
2006-03-25 6:48 ` Kyle Moffett
2006-03-28 20:17 ` Jim Gifford
2006-03-25 1:36 ` Jeff Dike
2006-03-25 6:33 ` Kyle Moffett
2006-03-25 16:03 ` Jeff Dike
2006-03-25 3:19 ` Rob Landley
2006-03-25 6:27 ` Kyle Moffett
2006-03-26 11:52 ` [RFC][PATCH 0/2] KABI example conversion and cleanup Kyle Moffett
2006-03-26 11:54 ` [RFC][PATCH 1/2] Create initial kernel ABI header infrastructure Kyle Moffett
2006-03-26 12:32 ` Arjan van de Ven
2006-03-26 12:50 ` Kyle Moffett
2006-03-26 12:59 ` Martin Mares
2006-03-26 13:14 ` Kyle Moffett
2006-03-26 15:38 ` Martin Mares
2006-03-26 16:16 ` Kyle Moffett
2006-03-26 14:39 ` Arjan van de Ven
2006-03-26 15:23 ` Kyle Moffett
2006-03-29 22:26 ` Pavel Machek
2006-04-02 0:22 ` Randy.Dunlap
2006-04-02 2:42 ` Kyle Moffett
2006-04-02 3:01 ` Arjan van de Ven
2006-04-02 5:53 ` Kyle Moffett
2006-04-02 13:09 ` Arjan van de Ven [this message]
2006-04-02 10:32 ` Pavel Machek
2006-04-02 11:16 ` Kyle Moffett
2006-03-26 20:05 ` Sam Ravnborg
2006-03-26 20:39 ` Kyle Moffett
2006-03-26 21:26 ` Sam Ravnborg
2006-03-27 0:27 ` Kyle Moffett
2006-03-26 11:55 ` [RFC][PATCH 2/2] Generalize fd_set handling across architectures Kyle Moffett
2006-03-26 12:06 ` [RFC][PATCH 0/2] KABI example conversion and cleanup Kyle Moffett
2006-03-26 13:43 ` Nix
2006-03-26 12:26 ` Arjan van de Ven
2006-03-26 12:30 ` Arjan van de Ven
2006-03-26 12:34 ` Kyle Moffett
2006-03-26 13:22 ` Giuseppe Bilotta
2006-03-26 13:29 ` Avi Kivity
2006-03-26 13:47 ` Kyle Moffett
2006-03-26 13:53 ` Giuseppe Bilotta
2006-03-26 14:30 ` Kyle Moffett
2006-03-26 14:45 ` Giuseppe Bilotta
2006-03-26 17:24 ` Avi Kivity
2006-03-26 17:29 ` Arjan van de Ven
2006-03-26 17:57 ` Avi Kivity
2006-03-26 18:32 ` Arjan van de Ven
2006-03-26 21:18 ` Rob Landley
2006-03-27 0:18 ` Kyle Moffett
2006-03-27 6:19 ` Avi Kivity
2006-03-27 19:48 ` Rob Landley
2006-03-28 20:04 ` Mariusz Mazur
2006-03-28 20:13 ` Kyle Moffett
2006-03-28 22:57 ` Rob Landley
2006-03-26 20:55 ` Rob Landley
2006-03-27 0:12 ` Kyle Moffett
2006-03-26 14:31 ` Eric Piel
2006-03-26 21:09 ` Rob Landley
2006-03-26 23:06 ` Eric Piel
2006-03-27 0:40 ` Kyle Moffett
2006-03-27 3:12 ` Jeff Dike
2006-03-28 14:20 ` Jan Engelhardt
2006-03-28 15:57 ` [OT] Non-GCC compilers used for linux userspace Kyle Moffett
2006-03-28 16:13 ` Eric Piel
2006-03-28 16:20 ` Kyle Moffett
2006-03-28 16:59 ` Jason L Tibbitts III
2006-03-28 17:13 ` Kyle Moffett
2006-03-28 17:28 ` Daniel Jacobowitz
2006-03-28 17:41 ` Kyle Moffett
2006-04-05 17:01 ` Bryan O'Sullivan
2006-03-28 17:08 ` Jan-Benedict Glaw
2006-03-28 17:56 ` Jesper Juhl
2006-03-28 21:47 ` Rob Landley
2006-03-29 21:23 ` Nix
2006-03-30 1:36 ` Rob Landley
2006-03-30 7:24 ` Nix
2006-03-30 20:26 ` Rob Landley
2006-03-30 22:02 ` Nix
2006-03-30 23:00 ` Harald Arnesen
2006-03-30 23:16 ` Rob Landley
2006-03-29 13:25 ` Mathieu Chouquet-Stringer
2006-03-28 18:44 ` Eric W. Biederman
2006-03-29 4:26 ` Peter Chubb
2006-03-30 15:15 ` Roger Heflin
2006-03-28 17:16 ` [RFC][PATCH 0/2] KABI example conversion and cleanup Ben Pfaff
2006-03-28 17:08 ` Catalin Marinas
2006-03-31 0:20 ` Arch-specific header inconsistency (asm-*/termios.h) Kyle Moffett
2006-04-02 17:58 ` [RFC][PATCH 0/2] KABI example conversion and cleanup Sam Ravnborg
2006-04-02 19:30 ` Kyle Moffett
2006-04-02 20:47 ` Arnd Bergmann
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1143983345.2994.11.camel@laptopd505.fenrus.org \
--to=arjan@infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=llh-discuss@lists.pld-linux.org \
--cc=mmazur@kernel.pl \
--cc=mrmacman_g4@mac.com \
--cc=nix@esperi.org.uk \
--cc=pavel@ucw.cz \
--cc=rdunlap@xenotime.net \
--cc=rob@landley.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®