Ok, so 2.4.18 was released without a fix that onyl matters to a very
limited number of Linux users, while that's not perfect it's not the end
of the world either.
A few different solutions have been suggested :
1) People have suggested releasing a new 2.4.18 emidiately which
includes the missing fix.
2) Someone proposed to release 2.4.19 imidiately which would be 2.4.18 +
the missing fix.
3) Some people just want to forget about it ;)
All 3 are IMVHO bad ideas for these reasons; nr. 1 would create
confusion by having two different versions of 2.4.18 floating around.
nr. 2 would probably also cause some minor degree of confusion amongst
users and would also give the (wrong) impression that Linux kernel
releases are not thoroughly tested. I'll not comment on 3 ;)
I suggest that instead of the previously proposed solutions an effort
should instead be made to release 2.4.19 relatively fast. Instead of
including a lot of fixes, only include the most critical known fixes in
2.4.19-pre1 (and maybe -pre2), then go to -rc1 as soon as possible and
get 2.4.19 out the door pretty fast with important fixes and leave all
other stuff for 2.4.20-pre1
Just my 0.02 euro-cent
- Jesper Juhl - [email protected] -
On Mon, Feb 25, 2002 at 11:25:25PM +0100, Jesper Juhl wrote:
> I suggest that instead of the previously proposed solutions an effort
> should instead be made to release 2.4.19 relatively fast. Instead of
> including a lot of fixes, only include the most critical known fixes in
> 2.4.19-pre1 (and maybe -pre2), then go to -rc1 as soon as possible and
> get 2.4.19 out the door pretty fast with important fixes and leave all
> other stuff for 2.4.20-pre1
>
Does anyone know how long this bug has been in the kernel?
If it's an old bug, 2.4.19-pre1 already has the fix, just like rc4 did...
This *one* bug isn't big enough to hurry, IMO.
Mike
On Mon, 2002-02-25 at 17:34, Mike Fedyk wrote:
> Does anyone know how long this bug has been in the kernel?
For a very short time -- it was introduced around pre6.
> If it's an old bug, 2.4.19-pre1 already has the fix, just like rc4 did...
>
> This *one* bug isn't big enough to hurry, IMO.
Agreed.
Robert Love
> > I suggest that instead of the previously proposed solutions an effort
> > should instead be made to release 2.4.19 relatively fast. Instead of
> > including a lot of fixes, only include the most critical known fixes
> in
> > 2.4.19-pre1 (and maybe -pre2), then go to -rc1 as soon as possible and
>
> > get 2.4.19 out the door pretty fast with important fixes and leave all
>
> > other stuff for 2.4.20-pre1
> >
>
> Does anyone know how long this bug has been in the kernel?
>
> If it's an old bug, 2.4.19-pre1 already has the fix, just like rc4
> did...
>
> This *one* bug isn't big enough to hurry, IMO.
You are probably right, a resonably harmless bug should not be enough to
justify a rushed 2.4.19 release. But the confusion and "bad publicity"
/might/ justify it.
I personally don't think that the publicity issue is something that the
kernel people should be concerned about, but user confusion probably is
- confused users submitting bug reports and/or patches against two
different 2.4.18 versions sounds to me like something that you want to
keep to a minimum, and the only way I see of removing that confusion is
to get 2.4.18 out of the way and replaced with 2.4.19 as soon as possible.
- Jesper Juhl - [email protected] -