sm-commit AT lists.ibiblio.org
Subject: Source Mage code commit list
List archive
Re: [SM-Commit] GIT changes to master grimoire by Treeve Jelbert (74b825f76b4857af454b027837754d515705994c)
- From: Eric Sandall <eric AT sandall.us>
- To: sm-commit AT lists.ibiblio.org
- Subject: Re: [SM-Commit] GIT changes to master grimoire by Treeve Jelbert (74b825f76b4857af454b027837754d515705994c)
- Date: Sun, 30 Nov 2008 21:02:46 -0800
On Sun, 30 Nov 2008 18:58:56 +0100
Ladislav Hagara <ladislav.hagara AT unob.cz> wrote:
> > Well it works with the newest kernel headers. I just finished
> > compiling it and told it to use the newest stable headers.
>
> When I updated to default glibc 2.9 I had a problem with mc, the keys
> F3 and F4 didn't work.
> The problem disappeared when I recast glibc 2.9 with "sanitized linux
> kernel headers from the latest stable release", it is default off.
> Imho, we should change default to on, or even automatically use
> headers from latest stable kernel.
> Using headers 2.6.25 with linux 2.6.27 is not good.
We had issues in the past with glibc not using the same headers (even
if outdated) every time. However, if we continue to use old headers we
will not got new features from the new kernels (since the headers won't
support it). What happens is glibc gets updated with new headers, but
other apps are not updated to use these new headers. Or worse, some
apps are built with headers X, others with headers Y, and yet more with
the latest and greatest headers Z. Does this even matter, or does it
only matter that glibc is built against the headers and everyone else
is happy to link against glibc?
How much testing has been done with using 'dynamic' sanitized kernel
headers, which glibc uses our linux spell to determine which headers to
download?
-sandalle
Attachment:
signature.asc
Description: PGP signature
- Re: [SM-Commit] GIT changes to master grimoire by Treeve Jelbert (74b825f76b4857af454b027837754d515705994c), Eric Sandall, 12/01/2008
Archive powered by MHonArc 2.6.24.