From: Matthew Wilcox <willy@infradead.org>
To: linux-mm@kvack.org, linux-fsdevel@vger.kernel.org,
linux-kernel@vger.kernel.org
Cc: Matthew Wilcox <mawilcox@microsoft.com>, Jan Kara <jack@suse.cz>,
Jeff Layton <jlayton@redhat.com>,
Lukas Czerner <lczerner@redhat.com>,
Ross Zwisler <ross.zwisler@linux.intel.com>,
Christoph Hellwig <hch@lst.de>,
Goldwyn Rodrigues <rgoldwyn@suse.com>,
Nicholas Piggin <npiggin@gmail.com>,
Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>,
linux-nilfs@vger.kernel.org, Jaegeuk Kim <jaegeuk@kernel.org>,
Chao Yu <yuchao0@huawei.com>,
linux-f2fs-devel@lists.sourceforge.net
Subject: [PATCH v13 29/72] page cache: Convert filemap_map_pages to XArray
Date: Mon, 11 Jun 2018 07:05:56 -0700 [thread overview]
Message-ID: <20180611140639.17215-30-willy@infradead.org> (raw)
In-Reply-To: <20180611140639.17215-1-willy@infradead.org>
From: Matthew Wilcox <mawilcox@microsoft.com>
Slight change of strategy here; if we have trouble getting hold of a
page for whatever reason (eg a compound page is split underneath us),
don't spin to stabilise the page, just continue the iteration, like we
would if we failed to trylock the page. Since this is a speculative
optimisation, it feels like we should allow the process to take an extra
fault if it turns out to need this page instead of spending time to pin
down a page it may not need.
Signed-off-by: Matthew Wilcox <mawilcox@microsoft.com>
---
mm/filemap.c | 42 +++++++++++++-----------------------------
1 file changed, 13 insertions(+), 29 deletions(-)
diff --git a/mm/filemap.c b/mm/filemap.c
index 67f04bcdf9ef..4204d9df003b 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2516,45 +2516,31 @@ EXPORT_SYMBOL(filemap_fault);
void filemap_map_pages(struct vm_fault *vmf,
pgoff_t start_pgoff, pgoff_t end_pgoff)
{
- struct radix_tree_iter iter;
- void **slot;
struct file *file = vmf->vma->vm_file;
struct address_space *mapping = file->f_mapping;
pgoff_t last_pgoff = start_pgoff;
unsigned long max_idx;
+ XA_STATE(xas, &mapping->i_pages, start_pgoff);
struct page *head, *page;
rcu_read_lock();
- radix_tree_for_each_slot(slot, &mapping->i_pages, &iter, start_pgoff) {
- if (iter.index > end_pgoff)
- break;
-repeat:
- page = radix_tree_deref_slot(slot);
- if (unlikely(!page))
- goto next;
- if (radix_tree_exception(page)) {
- if (radix_tree_deref_retry(page)) {
- slot = radix_tree_iter_retry(&iter);
- continue;
- }
+ xas_for_each(&xas, page, end_pgoff) {
+ if (xas_retry(&xas, page))
+ continue;
+ if (xa_is_value(page))
goto next;
- }
head = compound_head(page);
if (!page_cache_get_speculative(head))
- goto repeat;
+ goto next;
/* The page was split under us? */
- if (compound_head(page) != head) {
- put_page(head);
- goto repeat;
- }
+ if (compound_head(page) != head)
+ goto skip;
/* Has the page moved? */
- if (unlikely(page != *slot)) {
- put_page(head);
- goto repeat;
- }
+ if (unlikely(page != xas_reload(&xas)))
+ goto skip;
if (!PageUptodate(page) ||
PageReadahead(page) ||
@@ -2573,10 +2559,10 @@ void filemap_map_pages(struct vm_fault *vmf,
if (file->f_ra.mmap_miss > 0)
file->f_ra.mmap_miss--;
- vmf->address += (iter.index - last_pgoff) << PAGE_SHIFT;
+ vmf->address += (xas.xa_index - last_pgoff) << PAGE_SHIFT;
if (vmf->pte)
- vmf->pte += iter.index - last_pgoff;
- last_pgoff = iter.index;
+ vmf->pte += xas.xa_index - last_pgoff;
+ last_pgoff = xas.xa_index;
if (alloc_set_pte(vmf, NULL, page))
goto unlock;
unlock_page(page);
@@ -2589,8 +2575,6 @@ void filemap_map_pages(struct vm_fault *vmf,
/* Huge page is mapped? No need to proceed. */
if (pmd_trans_huge(*vmf->pmd))
break;
- if (iter.index == end_pgoff)
- break;
}
rcu_read_unlock();
}
--
2.17.1
next prev parent reply other threads:[~2018-06-11 14:05 UTC|newest]
Thread overview: 79+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-06-11 14:05 [PATCH v13 00/72] Convert page cache to XArray Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 01/72] radix tree test suite: Enable ubsan Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 02/72] dax: Fix use of zero page Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 03/72] xarray: Replace exceptional entries Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 04/72] xarray: Change definition of sibling entries Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 05/72] xarray: Add definition of struct xarray Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 06/72] xarray: Define struct xa_node Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 07/72] xarray: Add documentation Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 08/72] xarray: Add xa_load Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 09/72] xarray: Add XArray tags Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 10/72] xarray: Add xa_store Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 11/72] xarray: Add xa_cmpxchg and xa_insert Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 12/72] xarray: Add xa_for_each Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 13/72] xarray: Add xa_extract Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 14/72] xarray: Add xa_destroy Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 15/72] xarray: Add xas_next and xas_prev Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 16/72] xarray: Add xas_for_each_conflict Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 17/72] xarray: Add xas_create_range Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 18/72] xarray: Add MAINTAINERS entry Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 19/72] page cache: Rearrange address_space Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 20/72] page cache: Convert hole search to XArray Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 21/72] page cache: Add and replace pages using the XArray Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 22/72] page cache: Convert page deletion to XArray Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 23/72] page cache: Convert find_get_entry " Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 24/72] page cache: Convert find_get_entries " Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 25/72] page cache: Convert find_get_pages_range " Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 26/72] page cache: Convert find_get_pages_contig " Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 27/72] page cache; Convert find_get_pages_range_tag " Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 28/72] page cache: Convert find_get_entries_tag " Matthew Wilcox
2018-06-11 14:05 ` Matthew Wilcox [this message]
2018-06-11 14:05 ` [PATCH v13 30/72] radix tree test suite: Convert regression1 " Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 31/72] page cache: Convert delete_batch " Matthew Wilcox
2018-06-11 14:05 ` [PATCH v13 32/72] page cache: Remove stray radix comment Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 33/72] page cache: Convert filemap_range_has_page to XArray Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 34/72] mm: Convert page-writeback " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 35/72] mm: Convert workingset " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 36/72] mm: Convert truncate " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 37/72] mm: Convert add_to_swap_cache " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 38/72] mm: Convert delete_from_swap_cache " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 39/72] mm: Convert __do_page_cache_readahead " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 40/72] mm: Convert page migration " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 41/72] mm: Convert huge_memory " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 42/72] mm: Convert collapse_shmem " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 43/72] mm: Convert khugepaged_scan_shmem " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 44/72] mm: Convert is_page_cache_freeable " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 45/72] pagevec: Use xa_tag_t Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 46/72] shmem: Convert shmem_radix_tree_replace to XArray Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 47/72] shmem: Convert shmem_confirm_swap " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 48/72] shmem: Convert find_swap_entry " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 49/72] shmem: Convert shmem_add_to_page_cache " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 50/72] shmem: Convert shmem_alloc_hugepage " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 51/72] shmem: Convert shmem_free_swap " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 52/72] shmem: Convert shmem_partial_swap_usage " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 53/72] memfd: Convert memfd_wait_for_pins " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 54/72] memfd: Convert memfd_tag_pins " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 55/72] shmem: Comment fixups Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 56/72] btrfs: Convert page cache to XArray Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 57/72] fs: Convert buffer " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 58/72] fs: Convert writeback " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 59/72] nilfs2: Convert " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 60/72] f2fs: " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 61/72] dax: Rename some functions Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 62/72] dax: Hash on XArray instead of mapping Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 63/72] dax: Convert dax_insert_pfn_mkwrite to XArray Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 64/72] dax: Convert dax_layout_busy_page " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 65/72] dax: Convert __dax_invalidate_entry " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 66/72] dax: Convert dax writeback " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 67/72] dax: Convert page fault handlers " Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 68/72] page cache: Finish XArray conversion Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 69/72] radix tree: Remove radix_tree_update_node_t Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 70/72] radix tree: Remove split/join code Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 71/72] radix tree: Remove radix_tree_maybe_preload_order Matthew Wilcox
2018-06-11 14:06 ` [PATCH v13 72/72] radix tree: Remove radix_tree_clear_tags Matthew Wilcox
2018-06-12 10:40 ` [PATCH v13 00/72] Convert page cache to XArray David Sterba
2018-06-12 11:31 ` Matthew Wilcox
2018-06-12 19:37 ` Ross Zwisler
2018-06-12 19:46 ` Matthew Wilcox
2018-06-13 20:10 ` Ross Zwisler
2018-06-16 12:07 ` Matthew Wilcox
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180611140639.17215-30-willy@infradead.org \
--to=willy@infradead.org \
--cc=hch@lst.de \
--cc=jack@suse.cz \
--cc=jaegeuk@kernel.org \
--cc=jlayton@redhat.com \
--cc=konishi.ryusuke@lab.ntt.co.jp \
--cc=lczerner@redhat.com \
--cc=linux-f2fs-devel@lists.sourceforge.net \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nilfs@vger.kernel.org \
--cc=mawilcox@microsoft.com \
--cc=npiggin@gmail.com \
--cc=rgoldwyn@suse.com \
--cc=ross.zwisler@linux.intel.com \
--cc=yuchao0@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).