On Mon, Mar 5, 2012 at 9:42 AM, sparsh mittal <[email protected]>wrote:
> Hello
>
> I wanted to ask if it is possible in marss that kernel and user access at
> least some common address-space?
>
> Yes. That's how 'read' and 'write' and other system calls work. One of
them setup the cache and other read/write to it.
> If yes, then kernel may bring a block in cache, which may be used by
> user-space or vice-versa. Since the matching logic in cache (copied below)
> does not check whether it by kernel or user, it will be considered hit. Is
> it correct?
>
>
>
> int match(T target) {
> int way = 0;
> foreach (i, ways) {
> way += (tags[i] == target) ? (i + 1) : 0;
> }
>
> return way - 1;
> }
>
> Yes it looks good to me. Caches store their tag based on physical address
so it doesn't have to check for user/kernel level.
- Avadh
>
> Thanks and Regards
> Sparsh Mittal
>
>
>
> _______________________________________________
> http://www.marss86.org
> Marss86-Devel mailing list
> [email protected]
> https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
>
>
_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
