Re: why no gcc in 9.1 personal?
From: Moritz Franosch (mail_at_Franosch.org)
Date: 05/25/04
- Next message: Fred Emmott: "Re: Laptop for SuSE"
- Previous message: Malke: "Re: Laptop for SuSE"
- In reply to: houghi: "Re: why no gcc in 9.1 personal?"
- Next in thread: houghi: "Re: why no gcc in 9.1 personal?"
- Reply: houghi: "Re: why no gcc in 9.1 personal?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: 25 May 2004 19:34:24 +0200
houghi <houghi@houghi.org.invalid> writes:
> Moritz Franosch wrote:
> > First, a lot of vendors do not open up their hardware.
>
> Why do they not do that?
- They licensed the driver or part of it from someone else.
- They fear that broken drivers could damage the hardware, which may
increase the cost for support.
- They want to sell the same hardware with different capabilities,
controlled by the driver.
- They have certified hardware/driver combinations, e.g. passive ISDN
cards may have to be certified by the telephone company and wireless
equipment may have to be certified to fulfill the law. Hardware
vendors don't want to be liable when a broken driver damages third
party equipment or is used to break the law.
- They fear that competitors use the information they get from an open
source driver for competitive products.
- They save the paperwork that would be necesary to explain the
interface of the hardware to driver developers.
- They don't want to be liable for wrong descriptions of the hardware.
- When the user installs a new OS, they want to sell new hardware by
not porting the driver to the new OS.
> > Second, Suse does not offer new drivers for download even if they
> > are available.
>
> SUSE does not have to do that. The hardware vendor can do that.
The point is: A driver or at least the kernel interface of the driver
has to be compiled on the target machine with the very same compiler
the kernel was compiled with. Without gcc, a hardware vendor can not
install his driver.
> >> Why is it that you do not need to compile if you have Windows and need
> >> specific drivers? Because the manufactoror already compiled them for
> >> you.
> >
> > No. It's because Windows has a stable binary driver interface and/or
> > there are fewer versions of Windows for which the driver has to be
> > compiled.
>
> Why does Windows has a binary driver? Because the hardware manufactoror
> wrote one.
>
> Linux has effectively 3 different versions, if you look at kernels. 2.2,
> 2.4 or 2.6. A driverdeveloper will know how to only make one verion for
> all of those.
No. There is no stable binary driver interface in Linux even during a
stable kernel series. This is a design decision.
For example the 4k kernel stack patch may go into 2.6 which would
break binary driver compatibility.
http://www.kerneltraffic.org/kernel-traffic/kt20040522_259.html#6
http://lwn.net/Articles/80290/
Binary modules and derived works
http://lwn.net/Articles/62446/
http://www.ibrado.com/lkml/#s2-11
Do I need to test my driver against all distributions?
(REG, MEA) There are minor detail changes in between each kernel
version (even in stable series), and depending on what configuration
options are used (basically SMP or not), certain things like
spinlocks may or may not reserve space in structures, and may or may
not need to be called (are even optimized away in non-SMP systems),
meaning that a binary driver compiled for SMP might not work with a
non-SMP kernel. And vice versa.
Also different vendors tend to inject different things into their
kernel patch-sets, which again may subtly change data layouts,
etc. In stable kernel series great pains are suffered at maintenance
so that data layouts of in-kernel APIs (and API calls themselves)
are not changed. Nevertheless something may change making binary
drivers to fail in mysterious ways.
Subtle memory changes may appear with i386-PAE mode (large memory
machines which can't map all of RAM into the kernel at the same
time).
Because of these differences, a driver compiled for one version of
the kernel, or one vendor's kernel, is not likely to work with
another kernel. Thus, if you are distributing a binary-only driver,
you will have a significant support load compiling drivers for
different kernels. If you are distributing a driver in source form,
then, provided the driver is well-written (i.e. does not make
assumptions about byte ordering or word sizes and uses standard
kernel interfaces), the driver should be portable across kernel
versions and architecture types. It will of course have to be
compiled by end-users for their particular kernel. Distribution
maintainers are likely to provide pre-compiled drivers, thus most
end-users won't need to compile the driver themselves.
Moritz
-- Dipl.-Phys. Moritz Franosch http://www.askos.de
- Next message: Fred Emmott: "Re: Laptop for SuSE"
- Previous message: Malke: "Re: Laptop for SuSE"
- In reply to: houghi: "Re: why no gcc in 9.1 personal?"
- Next in thread: houghi: "Re: why no gcc in 9.1 personal?"
- Reply: houghi: "Re: why no gcc in 9.1 personal?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
