From: Constantine Gavrilov <const-g@optibase.com>
To: Jeff Garzik <jgarzik@mandrakesoft.com>
Cc: Anton.Safonov@bestlinux.net, linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: PCMCIA problems on IBM ThinkPad 600X
Date: Sun, 01 Apr 2001 20:45:19 +0200 [thread overview]
Message-ID: <3AC7773F.9090401@optibase.com> (raw)
In-Reply-To: <Pine.LNX.3.96.1010401123249.28121B-100000@mandrakesoft.mandrakesoft.com>
Jeff Garzik wrote:
> On Sun, 1 Apr 2001, Constantine Gavrilov wrote:
>
>> There are problems with some PCMCIA drivers included in the kernel. For
>> example, support for cardbus 3com cards was moved to 3c59x.o driver. It
>> works (on 600X at least) only of you compile it in. It will not work as
>> a module.
>
>
> It works just fine as a module. What problems are you seeing?
>
Exactly as reported by Anton. "cs: socket XXXXX timed out during reset"
messages on the console when loading the module. This is at least on IBM
Thinkpad 600X. 16-bit cards work fine.
>
>> I think a much better solution right now is to use drivers from
>> pcmcia-cs package. It always works. If you do not configure any support
>> for pcmcia in your kernel, when you build pcmcia-cs it will build kernel
>> drivers from its own source tree. Just make sure you use the latest
>> version. This also allows configuration files interoperbility with 2.2.x
>> kernel, if you wish to use that as well.
>
>
> pcmcia-cs does not always work, and it puts your nice 32-bit hardware
> into 16-bit compatibility mode AFAIK.
>
> If you have 2.4 bugs, please report them instead of spewing B.S.
>
> Jeff
>
>
Several points:
* this bug and the workaround have been reported several times on
several mailing lists, probably on linux-kernel as well. Explanations
also stated that it has been broken and reported since 2.4.0-preX (I do
not remember which pre-release). So it is not a hidden knowledge and I
do not have to report a known bug.
* I do not think pcmcia-cs puts cardbus cards into 16-bit compatibility
mode. According to David Hinds, pcmcia code has been integrated into
2.4, so 2.4 uses a similar code base. My tests of bonding code showed 2
Mbit/sec with PCMCIA (100% CPU utilization) and 12 Mbit/sec with CardBus
(<5% CPU utilization).
* The letter has not been addressed to you, but to the list. Why are you
taking this personal? What I said is no BS. The bug has been known and
reported. I personally use multiple versions of 2.2.x and 2.4.x kernels
installed on my machine for research and development. These include
various experimental patches and pre-releases. For people in my
situation, it is more convenient to use drivers from pcmcia-cs mainly
for two reasons: 1) I can use the same PCMCIA configuration for all
kernels; 2) I do not have to recompile kernel to upgrade PCMCIA drivers.
Why should it bother you?
David's stuff happens to work better right now. So what? There are
several quite logical reasons for it:
* PCMCIA code has been integrated relatively recently and not all
integration problems have been solved yet. "Official" and "unofficial"
Linux documentation state this and recommend pcmcia-cs in the case of
problems.
* Since 2.4 has come out, a lot of efforts are made to fix bugs. Some
changes in the code incidentally break "other" stuff. Since David has to
concentrate on PCMCIA only, he can respond quickly to fix integration
problems. He is not bound to kernel release schedules and can release
more frequently. In the current situation, it helps.
* When you (or somebody else) update network drivers, you cannot
possibly make sure that changes work across all card models. For David,
on the other hand, it is by far much easier to insure compatibility,
since he has to deal with CardBus and PCMCIA only. He also has had a lot
of experince doing this and his releases have always being of high quality.
So, you do not have to get angry. I did not reflect on the quality of
your code and the thought has not even occurred to me. After all, if
you update epro100 code, for instance, these changes appear in pcmcia-cs
package rather quickly. Part of David's job has been to make sure that
network drivers written in whole or in large part by other people work
WELL with PCMCIA and CardBus cards. He has been doing an excellent job
-- why should it bother you?
--
----------------------------------------
Constantine Gavrilov
Linux Leader
Optibase Ltd
7 Shenkar St, Herzliya 46120, Israel
Phone: (972-9)-970-9140
Fax: (972-9)-958-6099
----------------------------------------
next prev parent reply other threads:[~2001-04-01 18:52 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-04-01 14:27 PCMCIA problems on IBM ThinkPad 600X Constantine Gavrilov
2001-04-01 17:34 ` Jeff Garzik
2001-04-01 18:45 ` Constantine Gavrilov [this message]
-- strict thread matches above, loose matches on Subject: below --
2001-03-30 12:36 Anton Safonov
2001-03-30 17:42 ` Jeff Garzik
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=3AC7773F.9090401@optibase.com \
--to=const-g@optibase.com \
--cc=Anton.Safonov@bestlinux.net \
--cc=jgarzik@mandrakesoft.com \
--cc=linux-kernel@vger.kernel.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