>> You are aware that RedHat issued a 2.2.5-22 kernel over a month
>> ago? According to RedHat's errata page, this was to fix as
>> vulnerability in the kernel, but I can state that the -22 kernel
>> is far more stable on my systems than the -15 kernel was...
> Grr, when did this turn into a RedHat support list?
Who said it had?
> If they're going to stuff all sorts of patches into their
> kernels then I don't see what the point of posting bug reports
> about them to this list is.
To be honest, my understanding is that ALL of the distributors who use
packaging systems provide packages with patched versions of the
sources, and the fact that they are patched doesn't stop them being
Linux. Normally, the patches in the kernel packages either work their
way back into the kernel source tree soon after (which accounts for
well over 90% of such patches), or are required due to some aspect of
the setup of that distribution that differs from the default.
One example of the latter that I can think of relates to the use and
configuration of modprobe in the 2.0 kernel series under RedHat 5.x
where it would prefer to use modules in /lib/modules/preferred/* to
/lib/modules/`uname -r`/* if the former existed. The patch for that
was the equivalent or replacing the following...
Q> cd /lib/modules/`uname -r`
...with the following...
Q> cd /lib/modules
Q> if [ -d preferred ]; then
Q> cd preferred
Q> else
Q> cd `uname -r`
Q> fi
...and the resultant effect on bug detection doesn't even need a
single finger to count it on...
> I don't use RedHat and likely never will, and I expect that goes
> for many people here (and yes, I agree, I don't know enough to
> do much in the way of tracking down problems - yet - but I can't
> imagine other people don't feel this way too).
I have used both RedHat and Debian regularly, and have also used
Slackware briefly, and in my experience, the distribution one installs
from makes NO difference whatsoever once it has been installed: The
resulting system is LINUX.
> Now, I know Alan works for RedHat, so if his -acXX patches are
> the only things that get added to RedHat kernels then it's a bit
> less of an issue.
The bulk of the patches used by RedHat in the kernels are ones that
have been developed to fix some bug in the raw kernels, although not
necessarily by Alan. Those patches are almost inevitably included in
the following kernel release.
The patches that don't fall into that category are inevitably minor
configuration changes that are specific to the RedHat distribution,
being similar to that quoted above. These are few in number, and are
inevitably of the variety that are incapable of causing other bugs.
> But it would help if the bug reports talked about "2.2.5-ac2"
> instead of "2.2.5-22".
My inderstanding of the differences between 2.2.5-15 and 2.2.5-22 is
that the latter consists of the former patched with SOME of the
patches in the 2.2.5-ac series, plus SOME of the patches generated by
Andrea Archangel, some generated by Peter Anvin, etc., and that ALL of
the differences made their way into the kernel tree by 2.2.7, although
not all were in there for 2.2.6.
It would therefore be WRONG to refer to it as being 2.2.5-ac2 (it
isn't), or in fact to refer to it as anything other than 2.2.5-22.
> AIUI Alan is dealing with the stable series now anyway...
Is he?
Best wishes from Riley.
+----------------------------------------------------------------------+
| There is something frustrating about the quality and speed of Linux |
| development, ie., the quality is too high and the speed is too high, |
| in other words, I can implement this XXXX feature, but I bet someone |
| else has already done so and is just about to release their patch. |
+----------------------------------------------------------------------+
* ftp://ftp.MemAlpha.cx/pub/rhw/Linux
* http://www.MemAlpha.cx/kernel.versions.html
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/