On 4/6/2018 1:32 AM, Barbora Murinov√° wrote:
Hi all,

I've been running se.py with the arguments --caches --cacheline_size=8 on a 
small program:

int a = 42;
int *p = &a;
*a = 10;

And trying to understand the data cache output for this piece of code. For every access, the data cache claims to have a hit, although the first two are supposed to be cold misses. No prefetching is enabled as far as I'm aware. Do you know what could be causing this?

There is a lot of library setup and so on going on before, but that shouldn't 
affect this?

Of course it can!  It gets the stack region into the cache.

Regards - Eliot Moss
_______________________________________________
gem5-users mailing list
gem5-users@gem5.org
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to