Thu Jun 3 02:13:38 EDT 2004
mlists at bizintegrators.com wrote:
>On Wed, Jun 02, 2004 at 08:48:26PM -0500, Pete Wright wrote:
>>boy i wish i was at the bug and not doing a night shift :( here is an
>>interesting link regarding monolithic kernels relating to OpenBSD:
>>i pretty much agree with alot of the issues raised, but in my industry i
>>need fully funtional 3D drivers provided by nvidia for linux. yea it
>>sucks that they are not GPL'd, but hey it's better than using NT for
>They seem to have missed one important post on the list.
>Here it is, from Kamo Hiroyasu <wd at ics.nara-wu.ac.jp>
>Chuck, you are misusing the term `monolithic'. A monolithic kernel
>with loadable modules is still a monolithic kernel because the loaded
>codes will run on the kernel space. It is far from a micro kernel.
>That implies: you cannot gain any advantage of mocro kernels from
>Please consult any textbook on operating systems for detail.
> Kamo Hiroyasu
> [Kamo is the family name and Hiroyasu the given name.]
>Therefore, Linux kernel is also monolithic.
good point, hence the whole tanenbaum/torvalds split....
>I'm very gratefull for OpenBSD's integrity, meaning things like binary
>only drivers will never be accepted.
>talk mailing list
>talk at lists.nycbug.org
email: pete at nomadlogic.org
More information about the talk