>> We always, always use page-aligned lookups for the page cache.
>> (Actually there is one exception: certain obsolete a.out binaries, which
>> are demand paged with the pages beginning at offset 1K into the binary.
>> We don't support cache coherency for those and we don't support them at
>> all on filesystems with a >1k block size. It doesn't impact on the hash
>> issue.)
>
>i guess i'm confused then. what good does this change do:
Hmm I think I misunderstood you point, Chuck. I thought you was
complaining about the fact that some hash entry could be unused and other
overloaded but I was just assuming that the offset is always page-aligned.
I could write a simulation to check the hash function...
Andrea Arcangeli
-
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/