]> pilppa.com Git - linux-2.6-omap-h63xx.git/commit
[PATCH] mm: follow_page with inner ptlock
authorHugh Dickins <hugh@veritas.com>
Sun, 30 Oct 2005 01:16:33 +0000 (18:16 -0700)
committerLinus Torvalds <torvalds@g5.osdl.org>
Sun, 30 Oct 2005 04:40:41 +0000 (21:40 -0700)
commitdeceb6cd17e6dfafe4c4f81b1b4153bc41b2cb70
tree2a722f50e8edef8609a49f65bfcb222e499c44cc
parentc34d1b4d165c67b966bca4aba026443d7ff161eb
[PATCH] mm: follow_page with inner ptlock

Final step in pushing down common core's page_table_lock.  follow_page no
longer wants caller to hold page_table_lock, uses pte_offset_map_lock itself;
and so no page_table_lock is taken in get_user_pages itself.

But get_user_pages (and get_futex_key) do then need follow_page to pin the
page for them: take Daniel's suggestion of bitflags to follow_page.

Need one for WRITE, another for TOUCH (it was the accessed flag before:
vanished along with check_user_page_readable, but surely get_numa_maps is
wrong to mark every page it finds as accessed), another for GET.

And another, ANON to dispose of untouched_anonymous_page: it seems silly for
that to descend a second time, let follow_page observe if there was no page
table and return ZERO_PAGE if so.  Fix minor bug in that: check VM_LOCKED -
make_pages_present ought to make readonly anonymous present.

Give get_numa_maps a cond_resched while we're there.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
fs/proc/task_mmu.c
include/linux/mm.h
kernel/futex.c
mm/memory.c
mm/nommu.c