On 16 May 2010 10:39, Karanbir Singh mail-lists@karan.org wrote:
On 05/15/2010 11:48 PM, Ron Loftin wrote:
What would be the proper way to request such a thing?
s/request/offer to do this/ and its game on.
Open an issue at bugs.centos.org, with the details, and we can help from there on.
- KB
Whilst it may be that CentOS as a community can pull together to offer a non-PAE kernel (presumably as a CentOS-Plus) it seems odd that RH would have insist on something that has been present since the Pentium Pro (ffs) but which isn't part of the first 2 iterations of the pentium M (i believe the post dothan CPUs that went with the sonoma chipset was the first to have pae "enabled"). The idea that they are doing this deliberatly to ensure newish hardware is poppycock (sorry about the swearing). What happens if intel decide to ship new supa-budget chips which also have no pae or emt-64? Why make it a i386 kernel (with pae) rather than a i686 without or are there some 80(2/3/4)86 revisions with pae that i don't know of? It just seems to be a bad engineering decision.
Anyways wrong place for a rant never liked Mondays
On a more useful note: I would be more than happy to help out in any effort to provide a non-pae kernel for CentOS 6 :)
mike