2001-02-15 17:38:47

by Matt Liotta

[permalink] [raw]
Subject: RE: aic7xxx (and sym53c8xx) plans

All of my boxes with that card are on 2.2.16. The rest are on 2.4.1, so I
don't really have a need to test 2.2.18 as I would rather be on 2.4.x for
all of my boxes.

-Matt

> -----Original Message-----
> From: Justin T. Gibbs [mailto:[email protected]]
> Sent: Thursday, February 15, 2001 9:36 AM
> To: Matt Liotta
> Cc: [email protected]
> Subject: Re: aic7xxx (and sym53c8xx) plans
>
>
> >I am still stuck on 2.2 because of this issue. I would
> really like to see
> >this driver in 2.4.2.
>
> Have you tested the 2.2.18 version of the new driver? The patches
> should work on most 2.2.X kernels, I just haven't gotten around to
> verifying that. The more testers, the merrier! :-)
>
> --
> Justin
>


2001-02-15 17:47:37

by Justin T. Gibbs

[permalink] [raw]
Subject: Re: aic7xxx (and sym53c8xx) plans

>All of my boxes with that card are on 2.2.16. The rest are on 2.4.1, so I
>don't really have a need to test 2.2.18 as I would rather be on 2.4.x for
>all of my boxes.

Well, I'll try and generate patches against 2.2.16 soon. I probably
need to support 2.2.14 too. There are already so many versions to
keep track of, the sooner the driver becomes embedded, the better.

--
Justin

2001-02-15 20:20:49

by J.A. Magallon

[permalink] [raw]
Subject: Re: aic7xxx (and sym53c8xx) plans


On 02.15 Justin T. Gibbs wrote:
> >All of my boxes with that card are on 2.2.16. The rest are on 2.4.1, so I
> >don't really have a need to test 2.2.18 as I would rather be on 2.4.x for
> >all of my boxes.
>
> Well, I'll try and generate patches against 2.2.16 soon. I probably
> need to support 2.2.14 too. There are already so many versions to
> keep track of, the sooner the driver becomes embedded, the better.
>

Please, I think it would be much more useful a patch against the latest
2.2.19-pre (if that one for 2.2.18 does not work, I have not tried)
and the latest 2.4.1-ac14, that is what people experiments with.

People who has still a 2.2.14 or 16 looks like does not worry too much
about updating and building kernels, and it they go into the work, they
can just go to 2.2.18, compatible with 16 and 14 and much more stable.

I think it is better to work for preX kernels than for eldely ones.

--
J.A. Magallon $> cd pub
mailto:[email protected] $> more beer

Linux werewolf 2.4.1-ac14 #1 SMP Thu Feb 15 16:05:52 CET 2001 i686

2001-02-15 20:30:50

by Chip Salzenberg

[permalink] [raw]
Subject: Re: aic7xxx (and sym53c8xx) plans

According to J . A . Magallon:
> Please, I think it would be much more useful a patch against the latest
> 2.2.19-pre (if that one for 2.2.18 does not work, I have not tried)
> and the latest 2.4.1-ac14, that is what people experiments with.

There's no end of versions that people use.

Might I suggest that Justin imitate the maintainers of lm_sensors, and
create a program (shell script, Perl program, whatever) that *creates*
a patch against any given Linux source tree? Obviously it could break
in the face of weird trees, but even minimal flexibility would save him
a lot of work ...
--
Chip Salzenberg - a.k.a. - <[email protected]>
"We have no fuel on board, plus or minus 8 kilograms." -- NEAR tech

2001-02-15 23:42:50

by J.A. Magallon

[permalink] [raw]
Subject: Re: aic7xxx (and sym53c8xx) plans


On 02.15 Chip Salzenberg wrote:
> According to J . A . Magallon:
>
> Might I suggest that Justin imitate the maintainers of lm_sensors, and
> create a program (shell script, Perl program, whatever) that *creates*
> a patch against any given Linux source tree? Obviously it could break
> in the face of weird trees, but even minimal flexibility would save him
> a lot of work ...

So you can end with 1Mb of patch doing

-#endif /* Hello */
+#endif Hello

like happens in i2c-lm

Better a real patch...

--
J.A. Magallon $> cd pub
mailto:[email protected] $> more beer

Linux werewolf 2.4.1-ac14 #1 SMP Thu Feb 15 16:05:52 CET 2001 i686