How are interrupts handled on SMP (Symmeteric multiprocessor/multicore) machines? Is there only one memory management unit or more?
Say two threads, A and B running on d
I didn't get the "multiple instances of the kernel", usually the kernel rules em' all. meaning that it has no instance, instead one should think about the kernel as a global reactive system, which provides services to applications.
As far as I know, the memory handing is one unit (though each core has its own interrupt vector), the pages are locked using page_table_lock
, so the page retrieving is executed only once, in the order of locking.
** EDIT: After seeing the other comments, my answer might be out-dated: anyway you should check: http://www.xml.com/ldd/chapter/book/ch13.html