All of lore.kernel.org
 help / color / mirror / Atom feed
From: Adrian Bunk <bunk@fs.tum.de>
To: Dennis Grant <trog@wincom.net>
Cc: linux-kernel@vger.kernel.org
Subject: Re: Identifying/activating faster ATAxx modes (WAS kernel config tale of woe)
Date: Tue, 26 Nov 2002 13:44:25 +0100	[thread overview]
Message-ID: <20021126124424.GJ24796@fs.tum.de> (raw)
In-Reply-To: <3de2eee3.16fd.0@wincom.net>

On Mon, Nov 25, 2002 at 10:31:12PM -0500, Dennis Grant wrote:

>...
> VP_IDE: chipset revision 6
> VP_IDE: not 100% native mode: will probe irqs later
> VP_IDE: Unknown VIA SouthBridge, contact Vojtech Pavlik <vojtech@ucw.cz>
>...      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

This is the information that explains it. As Christian already mentioned
your motherboard has a VIA KT8235 Southbridge which wasn't supported
until 2.4.20-rc2 (second Release Candidate for 2.4.20).

>...
> Quick philosphical point - it was expressed to me that "users shouldn't be compiling
> kernels, that's the distro manager's job"
> 
> I could not POSSIBLY disagree more. What is Linux if not putting more power
> in the hands of the people who actually use the system? Users should ABSOLUTELY
> be compiling kernels, if they want to or need to. The trick is to pass on enough
> information so as to give the guy on the pointy end of the compiler as best
> a chance of success as can be done.

IMHO we should distinguish two different kinds of users (simplified,
but I hope you get the difference) that often get mixed:

1. The novice user who wants to use applications
This user wants to install Linux with as few clicks as possible. He
doesn't know what a kernel is and he doesn't want to know. All he wants
is a stable OS that runs OpenOffice and Netscape.

2. The more experienced user
There are good reasons (e.g. kernel patches) why a system administrator
might want to compile his own kernel. Or a computer science student 
wants to learn a bit more about the kernel he's using.


For people like you in the second category it should always be possible
to compile a custom kernel. A user in the first category should whenever
possible never need to compile his own kernel.


> Is asking for a dmesg that states "this is an ATAxxx drive, and your interfacer
> is using ATAyyy" really such a horrible thing to ask?

No it isn't horrible. The problem I can imagine is that it's possible
that a startup script on a user's computer (e.g. provided by the
distribution) uses hdparm or a similar program to manipulate the IDE
devices. The information provided by "dmesg" would then not reflect the
state of the system after bootup is completed.

> In any case, I will provide any other info anybody might wish on my system,
> and thanks to those people who are offering help. It is appreciated.

The information whether DMA is used is available via "hdparm -v", I'm
sure it will show that "using_dma" is off on your computer.

There are two solutions:
- safe solution:
  Apply the 2.4.20-rc3 patch [1] on top of 2.4.19 and do a
    cp .config /tmp  # because "make mrproper" will delete the .config
    make mrproper
    cp /tmp/.config .
    make oldconfig dep bzImage modules
- more risky solution:
  enable DMA with "hdparm -d1"

> DG

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed


  parent reply	other threads:[~2002-11-26 12:37 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-11-26  3:31 Identifying/activating faster ATAxx modes (WAS kernel config tale of woe) Dennis Grant
2002-11-26  3:49 ` Tomas Szepe
2002-11-26 12:13   ` Adrian Bunk
2002-11-26 12:15     ` Tomas Szepe
2002-11-26 12:48       ` Adrian Bunk
2002-11-26 13:05         ` Tomas Szepe
2002-11-26 12:44 ` Adrian Bunk [this message]
2002-11-26 12:53   ` Adrian Bunk
  -- strict thread matches above, loose matches on Subject: below --
2002-11-26 10:53 Christian Guggenberger
2002-11-26 13:36 Dennis Grant
2002-11-26 13:42 Dennis Grant

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=20021126124424.GJ24796@fs.tum.de \
    --to=bunk@fs.tum.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=trog@wincom.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.