[rescue] Happy New Year! RIP, Sun/Solaris...

[Joshua Boyd]

On Wed, Jan 05, 2011 at 04:06:11PM +0000, Peter Corlett wrote:

> > What I see in USB and cifs is that some unexpected event happens, or an
> > expected event doesn't happen, or there is garbage in a response, and the
> > whole thing throughs up its hands and quits with no way to restart except
> > a reboot.
> 
> CIFS? That's nothing to do with NFS. CIFS is handled by the samba userspace
> daemon. Samba has its own share of disgusting hacks, although most of those
> are thanks to Microsoft.

I didn't say it has anything to do with CIFS.  It is just an example of
a kernel module that faults without halting the whole machine and can't
be restarted without a reboot.

With USB, I suppose that hypothetically the hardware itself could be
wedged and require the power cycle to make it work, but CIFS has no such
issue.

If I've ever seen the NFS client show signs of having crashed, I don't
think it was anytime in the last 5-10 years.
 
> [...]
> > OTOH, maybe I don't need to care about context switches on a quad core
> > machine.
> 
> Context switches are extremely expensive operations and best avoided. Your
> shiny quad core box takes about the same time as a 6502 to do a context
> switch.

I meant more that it might not need to context switch as much with
multiple cores.