From: Russell King <rmk@arm.linux.org.uk>
To: Jeff Garzik <jgarzik@pobox.com>
Cc: Greg KH <greg@kroah.com>, "David S. Miller" <davem@redhat.com>,
rddunlap@osdl.org, davej@redhat.com, willy@debian.org,
linux-kernel@vger.kernel.org,
kernel-janitor-discuss@lists.sourceforge.net
Subject: Re: C99 Initialisers
Date: Wed, 13 Aug 2003 22:06:19 +0100 [thread overview]
Message-ID: <20030813220619.F20676@flint.arm.linux.org.uk> (raw)
In-Reply-To: <3F3A952C.4050708@pobox.com>; from jgarzik@pobox.com on Wed, Aug 13, 2003 at 03:44:44PM -0400
On Wed, Aug 13, 2003 at 03:44:44PM -0400, Jeff Garzik wrote:
> Russell King wrote:
> > But what's the point of the extra complexity? People already put
> > references to other structures in the driver_data element, and
> > enums, so completely splitting the device IDs from the module
> > source is not going to be practical.
>
> enums are easy putting direct references would be annoying, but I also
> argue it's potentially broken and wrong to store and export that
> information publicly anyway.
Until new_id occurred, driver_data wasn't public, so this is a new
problem.
> The use of enums instead of pointers is practically required because
> there is a many-to-one relationship of ids to board information structs.
Hmm. Now that driver_data is public, people will bitch when the enums
change. This is _NOT_ something what I want to deal with - if I want
to add a new TI bridge type, I want to add the new TI enumeration
identifier along side the other TI enumeration identifiers, not at the
end. I don't want to worry about whether the user is using the enum
values or not. (eg, so I can use expressions like:
if (id->driver_data >= first_ti_id && id->driver_data <= last_ti_id))
By separating this, it effectively taking away some of the driver authors
freedoms to make choices like this, and this /will/ make driver code more
ugly.
> > Are you thinking of a parser which outputs C code for the module to
> > include? That might be made to work, but it doesn't sound as elegant
> > as the solution being described previously in this thread.
> >
> > "Make the easy things easy (PCI_DEVICE(vendor,device)) and make the
> > not so easy things possible (the long form)"
>
> That ignores the people who also need to get at the data, which must
> first be compiled from C into object code, then extracted via modutils,
> then turned into a computer readable form again, then used.
Could you describe the "user" side of your idea more? ie, what would
get installed onto the filesystem, and how would the tables be used?
--
Russell King (rmk@arm.linux.org.uk) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
next prev parent reply other threads:[~2003-08-13 21:06 UTC|newest]
Thread overview: 71+ messages / expand[flat|nested] mbox.gz Atom feed top
2003-08-12 2:02 C99 Initialisers CaT
2003-08-12 2:18 ` Robert Love
2003-08-12 2:39 ` Matthew Wilcox
2003-08-12 2:45 ` Robert Love
2003-08-12 2:57 ` Dagfinn Ilmari Mannsåker
2003-08-12 5:38 ` Greg KH
2003-08-12 9:01 ` Maciej Soltysiak
2003-08-12 10:03 ` Geert Uytterhoeven
2003-08-12 10:19 ` Jakub Jelinek
2003-08-12 11:27 ` Matthew Wilcox
2003-08-12 16:54 ` Ian Hastie
2003-08-12 18:01 ` Greg KH
2003-08-12 23:53 ` Dave Jones
2003-08-13 0:08 ` Matthew Wilcox
2003-08-13 0:23 ` Greg KH
2003-08-13 0:31 ` Matthew Wilcox
2003-08-14 5:45 ` H. Peter Anvin
2003-08-13 15:52 ` Timothy Miller
2003-08-13 17:50 ` Jeff Garzik
2003-08-13 0:02 ` Jeff Garzik
2003-08-13 0:14 ` Randy.Dunlap
2003-08-13 0:31 ` Jeff Garzik
2003-08-13 0:37 ` Randy.Dunlap
2003-08-13 0:49 ` Dave Jones
2003-08-13 1:25 ` Jeff Garzik
2003-08-13 3:02 ` Randy.Dunlap
2003-08-13 3:26 ` Jeff Garzik
2003-08-13 10:14 ` David S. Miller
2003-08-13 17:31 ` Greg KH
2003-08-13 17:36 ` David S. Miller
2003-08-13 17:47 ` Jeff Garzik
2003-08-13 18:02 ` Greg KH
2003-08-13 18:26 ` Jeff Garzik
2003-08-13 18:38 ` Russell King
2003-08-13 19:44 ` Jeff Garzik
2003-08-13 19:54 ` Matthew Wilcox
2003-08-13 20:15 ` Greg KH
2003-08-13 20:16 ` Dave Jones
2003-08-13 20:30 ` Matt Domsch
2003-08-13 20:29 ` Jeff Garzik
2003-08-13 21:05 ` Sam Ravnborg
2003-08-13 22:24 ` Roman Zippel
2003-08-14 20:31 ` Sam Ravnborg
2003-08-14 10:05 ` Geert Uytterhoeven
2003-08-14 10:25 ` Gene Heskett
2003-08-14 10:52 ` jw schultz
2003-08-14 12:34 ` Geert Uytterhoeven
2003-08-14 12:57 ` Andrey Panin
2003-08-14 18:45 ` H. Peter Anvin
2003-08-13 21:06 ` Russell King [this message]
2003-08-13 21:17 ` Eduardo Pereira Habkost
2003-08-13 17:50 ` Sam Ravnborg
2003-08-13 17:54 ` Jeff Garzik
2003-08-13 17:54 ` Matthew Wilcox
2003-08-13 17:58 ` Jeff Garzik
2003-08-13 18:03 ` Greg KH
2003-08-13 17:58 ` Greg KH
2003-08-13 18:21 ` Sam Ravnborg
2003-08-13 18:09 ` Russell King
2003-08-13 20:21 ` Krzysztof Halasa
2003-08-13 21:17 ` David S. Miller
2003-08-13 21:26 ` Greg KH
2003-08-14 22:46 ` Krzysztof Halasa
2003-08-12 17:37 ` Dave Jones
2003-08-12 17:48 ` Matthew Wilcox
2003-08-12 22:06 ` Ian Hastie
2003-08-13 15:54 ` CaT
2003-08-14 6:57 ` Maciej Soltysiak
-- strict thread matches above, loose matches on Subject: below --
2003-08-12 16:52 Shureih, Tariq
[not found] <jFFu.7t8.15@gated-at.bofh.it>
[not found] ` <jLKX.4KI.13@gated-at.bofh.it>
[not found] ` <jRnj.2dx.11@gated-at.bofh.it>
[not found] ` <jRwZ.2kJ.15@gated-at.bofh.it>
[not found] ` <jRQi.2zQ.5@gated-at.bofh.it>
[not found] ` <jRZY.2Hw.5@gated-at.bofh.it>
[not found] ` <jS9J.2Np.5@gated-at.bofh.it>
[not found] ` <jUbt.57S.7@gated-at.bofh.it>
[not found] ` <jUuT.5kZ.13@gated-at.bofh.it>
[not found] ` <k13k.22O.3@gated-at.bofh.it>
[not found] ` <k7Lq.7Gr.7@gated-at.bofh.it>
2003-08-13 21:19 ` junkio
2003-08-13 22:18 ` Greg KH
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=20030813220619.F20676@flint.arm.linux.org.uk \
--to=rmk@arm.linux.org.uk \
--cc=davej@redhat.com \
--cc=davem@redhat.com \
--cc=greg@kroah.com \
--cc=jgarzik@pobox.com \
--cc=kernel-janitor-discuss@lists.sourceforge.net \
--cc=linux-kernel@vger.kernel.org \
--cc=rddunlap@osdl.org \
--cc=willy@debian.org \
/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