Date: Sun, 09 Apr 2000 11:10:13 +0200
From: Manfred Spraul <manfreds@colorfullife.com>
I don't understand the purpose of flush_page_to_ram():
It's a (bad) attempt to deal with virtually indexed caches which
are larger than the page size of the machine. Generally, when
the kernel writes to a page which potentially can subsequently
accessed from user mappings, it must call flush_page_to_ram.
It flushes the kernel-view page out of the caches and thus
makes main memory up to date, this way when the user accesses
the page from his mapping he won't see stale data if he happens
to have the page mapped at a bad "virtual alias" of what the
kernel maps it at.
It sucks, I'd like to kill it along with flush_icache_page.
I have been working a lot recently on something which is clean and
hopefully can allow these things to die. But I don't want to talk
more about it until I am able to come up with an implementation which
I am happy with, because until that time it may as well not exist.
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/
This archive was generated by hypermail 2b29 : Sat Apr 15 2000 - 21:00:12 EST