14 messages in org.freebsd.freebsd-archRemoving kernel thread stack swapping| From | Sent On | Attachments |
|---|---|---|
| David Schultz | Mar 3, 2005 7:42 am | |
| Ryan Sommers | Mar 3, 2005 2:12 pm | |
| David Xu | Mar 3, 2005 2:37 pm | |
| Scott Long | Mar 3, 2005 2:51 pm | |
| John Baldwin | Mar 3, 2005 2:52 pm | |
| David Schultz | Mar 3, 2005 3:35 pm | |
| M. Warner Losh | Mar 3, 2005 4:47 pm | |
| Stephan Uphoff | Mar 3, 2005 4:51 pm | |
| Brooks Davis | Mar 3, 2005 4:58 pm | |
| David Xu | Mar 3, 2005 11:29 pm | |
| Scott Long | Mar 4, 2005 12:34 am | |
| Christoph Hellwig | Mar 5, 2005 9:17 am | |
| Peter Wemm | Mar 15, 2005 5:43 pm | |
| Stephan Uphoff | Mar 15, 2005 7:19 pm |
| Subject: | Removing kernel thread stack swapping | |
|---|---|---|
| From: | Scott Long (sco...@samsco.org) | |
| Date: | Mar 4, 2005 12:34:07 am | |
| List: | org.freebsd.freebsd-arch | |
David Xu wrote:
Brooks Davis wrote:
On Thu, Mar 03, 2005 at 09:54:07AM -0500, John Baldwin wrote:
Hence, don't kill this whole feature just because someone is too lazy to fix a bug.
It would be very useful and informative if someone were to write a high level description of the ways in which the kernel is not a POSIX C programming environment.
when you are importing a POSIX C programmed piece of code into kernel, this will be a problem, I always think problem from another side, not just stand at my side. :)
The kernel is not a POSIX C environment. It's not just a program that happens to do special things. I know that Linux goes to great lengths to support this concept, but it's simply not true. Programming in the kernel ideally is not much different from other forms of embedded system programming where resources are finite and expensive and it's the responibility of the programmer to develop efficient and safe code.
Scott