diff options
| author | Hugh Dickins <hugh@veritas.com> | 2005-01-07 21:58:02 -0800 |
|---|---|---|
| committer | Linus Torvalds <torvalds@evo.osdl.org> | 2005-01-07 21:58:02 -0800 |
| commit | b37e39b03bcd49397cac7b937d9b180157705e08 (patch) | |
| tree | bff70e85ba5286bb0c150b260eb1ace00242a6da /mm | |
| parent | de146a08f93f5c4ab71503a79a7c5e300065b39a (diff) | |
| download | history-b37e39b03bcd49397cac7b937d9b180157705e08.tar.gz | |
[PATCH] vmtrunc: truncate_count not atomic
Why is mapping->truncate_count atomic? It's incremented inside i_mmap_lock
(and i_sem), and the reads don't need it to be atomic.
And why smp_rmb() before call to ->nopage? The compiler cannot reorder the
initial assignment of sequence after the call to ->nopage, and no cpu (yet!)
can read from the future, which is all that matters there.
And delete totally bogus reset of truncate_count from blkmtd add_device.
truncate_count is all about detecting i_size changes: i_size does not change
there; and if it did, the count should be incremented not reset.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diffstat (limited to 'mm')
| -rw-r--r-- | mm/memory.c | 12 |
1 files changed, 5 insertions, 7 deletions
diff --git a/mm/memory.c b/mm/memory.c index 6224cb1933c145..5c6a4db990acfa 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1424,7 +1424,7 @@ void unmap_mapping_range(struct address_space *mapping, spin_lock(&mapping->i_mmap_lock); /* Protect against page fault */ - atomic_inc(&mapping->truncate_count); + mapping->truncate_count++; if (unlikely(!prio_tree_empty(&mapping->i_mmap))) unmap_mapping_range_list(&mapping->i_mmap, &details); @@ -1726,7 +1726,7 @@ do_no_page(struct mm_struct *mm, struct vm_area_struct *vma, struct page * new_page; struct address_space *mapping = NULL; pte_t entry; - int sequence = 0; + unsigned int sequence = 0; int ret = VM_FAULT_MINOR; int anon = 0; @@ -1738,9 +1738,8 @@ do_no_page(struct mm_struct *mm, struct vm_area_struct *vma, if (vma->vm_file) { mapping = vma->vm_file->f_mapping; - sequence = atomic_read(&mapping->truncate_count); + sequence = mapping->truncate_count; } - smp_rmb(); /* Prevent CPU from reordering lock-free ->nopage() */ retry: cond_resched(); new_page = vma->vm_ops->nopage(vma, address & PAGE_MASK, &ret); @@ -1774,9 +1773,8 @@ retry: * invalidated this page. If unmap_mapping_range got called, * retry getting the page. */ - if (mapping && - (unlikely(sequence != atomic_read(&mapping->truncate_count)))) { - sequence = atomic_read(&mapping->truncate_count); + if (mapping && unlikely(sequence != mapping->truncate_count)) { + sequence = mapping->truncate_count; spin_unlock(&mm->page_table_lock); page_cache_release(new_page); goto retry; |