And allocating the memory outside GFP we could avoid taking so many
memory order-list uptodate all the time, no?
On some of the systems I've tested the 2.3.9 code on, the page cache
hash table is 1MB or 2MB in size. How do you suggest that I allocate
this memory whose size is dynamic based upon available memory and
needs to be sized after the free page pools are setup?
Also once you get into the "huge" range, like 4MB, GFP is the only way
you can guarentee that the table will use the minimal number of TLB
entries.
I'm more than open to suggestions which retain the attributes I've
just described. But I have yet to hear one.
Later,
David S. Miller
davem@redhat.com
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/