From: Paul Gortmaker <p_gortmaker@yahoo.com>
To: Rik van Riel <riel@conectiva.com.br>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2.5.5] do export vmalloc_to_page to modules...
Date: Fri, 05 Apr 2002 04:25:08 -0500 [thread overview]
Message-ID: <3CAD6D74.6E834BC5@yahoo.com> (raw)
In-Reply-To: <Pine.LNX.4.44L.0204041217290.18660-100000@imladris.surriel.com>
> As it stands right now it is IMPOSSIBLE to support binary only
> drivers and I can only see two ways out of this situation:
Two different things "supporting" and "allowing". Currently it is that BOM
are "allowed". (I am not entering the discussion on whether this is good
or bad.) Note that you will not find too many volunteer hackers jumping
in to help "support" users with BOMs - the lk archives will reflect this.
If a vendor decides to release a BOM, then it is implicit in their decision
that by doing so, they have voluntarily denied to accept the support that
comes from having many different persons auditing their code, if it was
available. Hence the issue of "support" lies on their shoulders entirely.
> (1) don't allow binary only modules at all
>
> (2) have a stable ABI for binary only modules and don't allow
> these binary only modules to use other symbols, so people
> in need of binary only modules won't be locked to one
> particular version of the kernel (or have two binary only
> modules locked to _different_ versions of the kernel)
One of the great things about linux is that code quality and performance
have always taken precedence over keeping an interface carved in stone.
We've tried to not make radical inteface changes in stable relases, but
where necessary it has been done, and we were free to do so. If a vendor
with a BOM can't take the time to appropriately update their code and
release a new BOM, then it is clear that they don't really support linux.
If someone has chosen hardware for which a BOM is the only option, then
they need to investigate in detail that vendor's support policy, and get
things in writing if need be (e.g. large installation base, etc) detailing
just how current bugs and new drivers for future kernels will be handled.
If a vendor wants $$$ for such a written support policy, then you have to
factor that into the overall cost of what you are setting up.
If someone is stuck with _two_ unsupported BOMs that are for two different
kernels, then I'd have to say they didn't do their homework when choosing
the hardware and/or determining if the level of support the vendor was
offering was adequate for their particular application. Caveat emptor.
Paul.
next prev parent reply other threads:[~2002-04-05 9:53 UTC|newest]
Thread overview: 94+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-02-20 13:13 [PATCH 2.5.5] do export vmalloc_to_page to modules Stelian Pop
2002-02-20 16:00 ` Hugh Dickins
2002-02-20 16:01 ` Arjan van de Ven
2002-04-03 16:21 ` Andrea Arcangeli
2002-04-03 17:43 ` Alan Cox
2002-04-03 18:13 ` Andrea Arcangeli
2002-04-03 19:11 ` Alan Cox
2002-04-03 19:23 ` Andrea Arcangeli
2002-04-03 20:05 ` Tigran Aivazian
2002-04-03 20:27 ` Tigran Aivazian
2002-04-03 21:22 ` Alan Cox
2002-04-03 21:26 ` Tigran Aivazian
2002-04-03 21:48 ` Alan Cox
2002-04-03 19:03 ` Tigran Aivazian
2002-04-03 19:10 ` Linus Torvalds
2002-04-03 19:19 ` Tigran Aivazian
2002-04-03 19:24 ` Linus Torvalds
2002-04-03 21:05 ` Tigran Aivazian
2002-04-03 21:25 ` Alan Cox
2002-04-04 6:43 ` Keith Owens
2002-04-04 10:22 ` Tigran Aivazian
2002-04-04 10:35 ` Arjan van de Ven
2002-04-04 11:54 ` Alan Cox
2002-04-04 12:01 ` Tigran Aivazian
2002-04-04 12:31 ` Adrian Bunk
2002-04-04 12:48 ` Russell King
2002-04-04 12:40 ` Russell King
2002-04-04 12:46 ` Tigran Aivazian
2002-04-05 7:29 ` David Schwartz
2002-04-05 8:24 ` Adrian Bunk
2002-04-05 8:28 ` David Schwartz
2002-04-04 13:26 ` Ingo Molnar
2002-04-04 15:21 ` Rik van Riel
2002-04-05 9:25 ` Paul Gortmaker [this message]
2002-04-04 15:35 ` Tigran Aivazian
2002-04-04 16:55 ` Andrea Arcangeli
2002-04-04 17:16 ` Christoph Hellwig
2002-04-04 17:46 ` Ingo Molnar
2002-04-04 17:59 ` Arjan van de Ven
2002-04-04 18:15 ` Rik van Riel
2002-04-05 0:47 ` Linux kernel and binary drivers (was: [PATCH 2.5.5] do export vmalloc_to_page to modules...) Steffen Persvold
2002-04-04 15:55 ` [PATCH 2.5.5] do export vmalloc_to_page to modules Richard B. Johnson
2002-04-04 16:14 ` Alan Cox
2002-04-04 16:15 ` Peter Horton
2002-04-04 16:23 ` Ingo Molnar
2002-04-04 16:38 ` Michael Clark
2002-04-04 20:57 ` Adrian Bunk
2002-04-04 16:44 ` Andrea Arcangeli
2002-04-04 17:16 ` Ingo Molnar
2002-04-04 18:00 ` Alan Cox
2002-04-04 16:42 ` Alexander Viro
2002-04-03 19:29 ` Andrea Arcangeli
2002-04-03 19:38 ` Alan Cox
2002-04-03 19:25 ` Linus Torvalds
2002-04-03 19:44 ` Alan Cox
2002-04-03 19:39 ` Linus Torvalds
2002-04-03 20:35 ` Gerd Knorr
2002-04-03 22:17 ` Richard B. Johnson
2002-04-03 22:24 ` Rik van Riel
2002-04-03 22:33 ` Tigran Aivazian
2002-04-03 22:35 ` David S. Miller
2002-04-06 9:08 ` David Woodhouse
2002-04-03 22:39 ` Rik van Riel
2002-04-04 5:59 ` Chris Wedgwood
2002-04-04 12:06 ` Alan Cox
2002-04-04 9:33 ` Ingo Molnar
2002-04-03 19:26 ` Andrea Arcangeli
2002-04-03 19:35 ` Alan Cox
2002-04-05 6:06 ` David Schwartz
2002-04-05 12:48 ` Alan Cox
2002-04-05 18:46 ` David Schwartz
2002-04-05 19:27 ` Alan Cox
2002-04-05 20:05 ` David Schwartz
2002-04-06 17:55 ` Alan Cox
2002-04-03 19:19 ` Alexander Viro
2002-04-03 21:07 ` David Schwartz
2002-04-03 21:33 ` Alan Cox
2002-04-03 21:28 ` Daniel Jacobowitz
2002-04-03 22:09 ` David Schwartz
2002-04-04 6:26 ` Kai Henningsen
2002-04-04 8:44 ` Adrian Bunk
2002-04-04 12:01 ` Alan Cox
2002-02-20 16:27 ` Ingo Molnar
-- strict thread matches above, loose matches on Subject: below --
2002-04-03 23:47 Petr Vandrovec
2002-04-04 0:19 ` Rik van Riel
2002-04-04 11:54 Petr Vandrovec
[not found] <Pine.LNX.4.44L.0204041217290.18660-100000@imladris.surriel .com>
2002-04-04 16:11 ` Anton Altaparmakov
2002-04-04 16:29 ` Ingo Molnar
[not found] <Pine.LNX.4.44.0204041123410.6422-100000@devserv.devel.redh at.com>
2002-04-04 17:06 ` Anton Altaparmakov
2002-04-04 17:55 ` Alan Cox
2002-04-04 17:52 ` Anton Altaparmakov
2002-04-04 17:27 Nicholas Berry
2002-04-05 13:22 Gareth Hughes
2002-04-09 6:55 Rick A. Hohensee
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=3CAD6D74.6E834BC5@yahoo.com \
--to=p_gortmaker@yahoo.com \
--cc=linux-kernel@vger.kernel.org \
--cc=riel@conectiva.com.br \
/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