TCB size in ix86 L4

[Michael Hohmuth]

Adam, 

thanks for your replies!

"Adam 'WeirdArms' Wiggins" <awiggins at cse.unsw.edu.au> writes:

> 	Question here. Is Jochen's L4 got higher interupt latencies then
> fiasco due to masking out interupts or is time in the kernel small enough
> for real time applications. 

Our measurements with L4/x86 version 2 indicate that this kernel is
not suitable for real-time work.  The ``Taming Linux'' pager details
some of these concerns
<URL:http://os.inf.tu-dresden.de/papers_ps/part98.ps>.  Recently,
Torsten Paul did a series of measurements which show that L4/x86 often 
can't meet real-time requirements; the task_delete operation seems to
be especially bad, sometimes disabling interrupts for more than one
second.

(Torsten, if you're reading this: Is this data correct?)

Real-time requirements in our DROPS system are one reason why Fiasco
tries to be interruptible at almost any time.

Michael
-- 
hohmuth at innocent.com, hohmuth at inf.tu-dresden.de
http://home.pages.de/~hohmuth/