public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Jeff Hartmann <jhartmann@valinux.com>
To: unlisted-recipients:; (no To-header on input)@localhost.localdomain
Cc: John Cavan <johnc@damncats.org>, linux-kernel@vger.kernel.org
Subject: Re: 4.1.0 DRM (was Re: Linux 2.4.6-ac3)
Date: Mon, 16 Jul 2001 12:32:41 -0600	[thread overview]
Message-ID: <3B533349.2070201@valinux.com> (raw)
In-Reply-To: <E15M6jC-0005PK-00@the-village.bc.nu> 	<3B532BB7.1050300@valinux.com> <995307155.32445.36.camel@nomade>

Xavier Bestel wrote:

> On 16 Jul 2001 12:00:23 -0600, Jeff Hartmann wrote:
> [...]
> 
>> will do this.  This will make the 4.0 -> 4.1 have to be a compile time 
>> decision, but 4.1 -> 4.1.1 and higher will just coexist with each 
>> other.  I'm currently working out integrating this into the kernel 
>> build, and I should hopefully have a patch for Linus and Alan soon.
> 
> 
> I would have preferred if you were just versionning vs the API, not the
> X version (and keep the API rather stable).

The problem is that the API can't be completely stable because of the 3D 
client-side drivers, 2D ddx drivers, and the drm kernel modules have to 
be tightly coupled.  The drm kernel module has to provide a secure 
method of sending data to the card, but we also want it to be efficent.  
For example, the mesa-3.5 versions of the drm kernel modules can send 
multiple states inside one dma buffer, this greatly increases the 
efficency of the driver, but it costs API compatibility.  We do version 
the API, and will not attempt to use the drivers if their API versions 
are wrong.  However we want multiple versions of the API to coexist.  
Adding the version number to the module name will allow for all the 
versions to coexist.

>  I wouldn't like to update my
> kernel just to go from X 4.1.1 to X 4.1.1-pl1.

Well if there are changes to the drm kernel drivers, you have to update 
your kernel or compile the kernel modules by hand.  We could easily 
create a stable API if cards were secure, or we did not care about 
security.  Unfortunately neither one is the case, so the API is not 
completely static.

-Jeff


  reply	other threads:[~2001-07-16 18:34 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-07-14 17:36 Linux 2.4.6-ac3 Alan Cox
2001-07-14 20:01 ` Zilvinas Valinskas
2001-07-14 20:05   ` Alan Cox
2001-07-15  1:45     ` Gareth Hughes
2001-07-15 13:12       ` Alan Cox
2001-07-15 14:01         ` Gareth Hughes
2001-07-15 15:31           ` Alan Cox
2001-07-16  1:29             ` 4.1.0 DRM (was Re: Linux 2.4.6-ac3) Gareth Hughes
2001-07-16  1:51               ` Keith Owens
2001-07-16  2:07                 ` Gareth Hughes
2001-07-16 11:23               ` John Cavan
2001-07-16 11:39                 ` Alan Cox
2001-07-16 18:00                   ` Jeff Hartmann
2001-07-16 18:12                     ` Xavier Bestel
2001-07-16 18:32                       ` Jeff Hartmann [this message]
2001-07-16 18:42                     ` John Cavan
2001-07-16 19:32                       ` Jeff Hartmann
2001-07-16 19:34                         ` Xavier Bestel
2001-07-16 20:18                           ` Jeff Hartmann
2001-07-17  2:37                             ` Gareth Hughes
2001-07-17  8:31                           ` Mike A. Harris
2001-07-16 19:49                         ` John Cavan
2001-07-17  7:19                     ` 4.1.0 DRM Mike A. Harris
2001-07-17  5:28               ` 4.1.0 DRM (was Re: Linux 2.4.6-ac3) Juan Quintela
2001-07-18  9:06                 ` Gareth Hughes
2001-07-18 16:21                   ` Juan Quintela
2001-07-18 13:30                 ` Mike A. Harris
2001-07-17 13:19             ` Linux 2.4.6-ac3 Zdenek Kabelac
2001-07-15  1:31 ` Linux 2.4.6-ac3 - some unresolved Eyal Lebedinsky
2001-07-15 13:09   ` Alan Cox

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=3B533349.2070201@valinux.com \
    --to=jhartmann@valinux.com \
    --cc=johnc@damncats.org \
    --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