Messages in this thread | | | Date | Wed, 20 Feb 2013 20:16:46 +0800 | From | Ric Mason <> | Subject | Re: [patch] mm: shmem: use new radix tree iterator |
| |
Hi Hugh, On 02/04/2013 10:01 AM, Hugh Dickins wrote: > On Fri, 1 Feb 2013, Johannes Weiner wrote: > >> In shmem_find_get_pages_and_swap, use the faster radix tree iterator >> construct from 78c1d78 "radix-tree: introduce bit-optimized iterator". >> >> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> > Yes, that looks fine, and is testing out fine, thanks. > Acked-by: Hugh Dickins <hughd@google.com>
Could you share your testcase with me? It seems that you always can test shmem patches.
> >> --- >> mm/shmem.c | 25 ++++++++++++------------- >> 1 file changed, 12 insertions(+), 13 deletions(-) >> >> diff --git a/mm/shmem.c b/mm/shmem.c >> index a368a1c..c5dc8ae 100644 >> --- a/mm/shmem.c >> +++ b/mm/shmem.c >> @@ -336,19 +336,19 @@ static unsigned shmem_find_get_pages_and_swap(struct address_space *mapping, >> pgoff_t start, unsigned int nr_pages, >> struct page **pages, pgoff_t *indices) >> { >> - unsigned int i; >> - unsigned int ret; >> - unsigned int nr_found; >> + void **slot; >> + unsigned int ret = 0; >> + struct radix_tree_iter iter; >> + >> + if (!nr_pages) >> + return 0; >> >> rcu_read_lock(); >> restart: >> - nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree, >> - (void ***)pages, indices, start, nr_pages); >> - ret = 0; >> - for (i = 0; i < nr_found; i++) { >> + radix_tree_for_each_slot(slot, &mapping->page_tree, &iter, start) { >> struct page *page; >> repeat: >> - page = radix_tree_deref_slot((void **)pages[i]); >> + page = radix_tree_deref_slot(slot); >> if (unlikely(!page)) >> continue; >> if (radix_tree_exception(page)) { >> @@ -365,17 +365,16 @@ static unsigned shmem_find_get_pages_and_swap(struct address_space *mapping, >> goto repeat; >> >> /* Has the page moved? */ >> - if (unlikely(page != *((void **)pages[i]))) { >> + if (unlikely(page != *slot)) { >> page_cache_release(page); >> goto repeat; >> } >> export: >> - indices[ret] = indices[i]; >> + indices[ret] = iter.index; >> pages[ret] = page; >> - ret++; >> + if (++ret == nr_pages) >> + break; >> } >> - if (unlikely(!ret && nr_found)) >> - goto restart; >> rcu_read_unlock(); >> return ret; >> } >> -- >> 1.7.11.7 > -- > To unsubscribe, send a message with 'unsubscribe linux-mm' in > the body to majordomo@kvack.org. For more info on Linux MM, > see: http://www.linux-mm.org/ . > Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
| |