All of lore.kernel.org
 help / color / mirror / Atom feed
From: Qu Wenruo <wqu@suse.com>
To: linux-btrfs@vger.kernel.org
Subject: [PATCH v3 10/22] btrfs: extent_io: support subpage for extent buffer page release
Date: Wed,  6 Jan 2021 09:01:49 +0800	[thread overview]
Message-ID: <20210106010201.37864-11-wqu@suse.com> (raw)
In-Reply-To: <20210106010201.37864-1-wqu@suse.com>

In btrfs_release_extent_buffer_pages(), we need to add extra handling
for subpage.

To do so, introduce a new helper, detach_extent_buffer_page(), to do
different handling for regular and subpage cases.

For subpage case, the new trick is to clear the range of current extent
buffer, and detach page private if and only if we're the last tree block
of the page.
This part is handled by the subpage helper,
btrfs_subpage_clear_and_test_tree_block().

Signed-off-by: Qu Wenruo <wqu@suse.com>
---
 fs/btrfs/extent_io.c | 57 +++++++++++++++++++++++++++++++-------------
 fs/btrfs/subpage.h   | 24 +++++++++++++++++++
 2 files changed, 64 insertions(+), 17 deletions(-)

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 9b6b8f6b6da3..3158461ebf4c 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -4991,25 +4991,13 @@ int extent_buffer_under_io(const struct extent_buffer *eb)
 		test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags));
 }
 
-/*
- * Release all pages attached to the extent buffer.
- */
-static void btrfs_release_extent_buffer_pages(struct extent_buffer *eb)
+static void detach_extent_buffer_page(struct extent_buffer *eb,
+				      struct page *page)
 {
-	int i;
-	int num_pages;
-	int mapped = !test_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags);
-
-	BUG_ON(extent_buffer_under_io(eb));
-
-	num_pages = num_extent_pages(eb);
-	for (i = 0; i < num_pages; i++) {
-		struct page *page = eb->pages[i];
+	struct btrfs_fs_info *fs_info = eb->fs_info;
+	bool last = false;
 
-		if (!page)
-			continue;
-		if (mapped)
-			spin_lock(&page->mapping->private_lock);
+	if (fs_info->sectorsize == PAGE_SIZE) {
 		/*
 		 * We do this since we'll remove the pages after we've
 		 * removed the eb from the radix tree, so we could race
@@ -5028,6 +5016,41 @@ static void btrfs_release_extent_buffer_pages(struct extent_buffer *eb)
 			 */
 			detach_page_private(page);
 		}
+		return;
+	}
+
+	ASSERT(PagePrivate(page));
+	/*
+	 * For subpage case, clear the range in tree_block_bitmap,
+	 * and if we're the last one, detach private completely.
+	 */
+	last = btrfs_subpage_clear_and_test_tree_block(fs_info, page,
+					eb->start, eb->len);
+	if (last)
+		btrfs_detach_subpage(fs_info, page);
+}
+
+/*
+ * Release all pages attached to the extent buffer.
+ */
+static void btrfs_release_extent_buffer_pages(struct extent_buffer *eb)
+{
+	int i;
+	int num_pages;
+	int mapped = !test_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags);
+
+	ASSERT(!extent_buffer_under_io(eb));
+
+	num_pages = num_extent_pages(eb);
+	for (i = 0; i < num_pages; i++) {
+		struct page *page = eb->pages[i];
+
+		if (!page)
+			continue;
+		if (mapped)
+			spin_lock(&page->mapping->private_lock);
+
+		detach_extent_buffer_page(eb, page);
 
 		if (mapped)
 			spin_unlock(&page->mapping->private_lock);
diff --git a/fs/btrfs/subpage.h b/fs/btrfs/subpage.h
index e49d4a7329e1..c1556b3b68f3 100644
--- a/fs/btrfs/subpage.h
+++ b/fs/btrfs/subpage.h
@@ -78,4 +78,28 @@ static inline void btrfs_subpage_set_tree_block(struct btrfs_fs_info *fs_info,
 	spin_unlock_irqrestore(&subpage->lock, flags);
 }
 
+/*
+ * Clear the bits in tree_block_bitmap and return if we're the last bit set
+ * int tree_block_bitmap.
+ *
+ * Return true if we're the last bits in the tree_block_bitmap.
+ * Return false otherwise.
+ */
+static inline bool btrfs_subpage_clear_and_test_tree_block(
+			struct btrfs_fs_info *fs_info, struct page *page,
+			u64 start, u32 len)
+{
+	struct btrfs_subpage *subpage = (struct btrfs_subpage *)page->private;
+	u16 tmp = btrfs_subpage_calc_bitmap(fs_info, page, start, len);
+	unsigned long flags;
+	bool last = false;
+
+	spin_lock_irqsave(&subpage->lock, flags);
+	subpage->tree_block_bitmap &= ~tmp;
+	if (subpage->tree_block_bitmap == 0)
+		last = true;
+	spin_unlock_irqrestore(&subpage->lock, flags);
+	return last;
+}
+
 #endif /* BTRFS_SUBPAGE_H */
-- 
2.29.2


  parent reply	other threads:[~2021-01-06  1:03 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-06  1:01 [PATCH v3 00/22] btrfs: add read-only support for subpage sector size Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 01/22] btrfs: extent_io: rename @offset parameter to @disk_bytenr for submit_extent_page() Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 02/22] btrfs: extent_io: refactor __extent_writepage_io() to improve readability Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 03/22] btrfs: file: update comment for btrfs_dirty_pages() Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 04/22] btrfs: extent_io: update locked page dirty/writeback/error bits in __process_pages_contig() Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 05/22] btrfs: extent_io: merge PAGE_CLEAR_DIRTY and PAGE_SET_WRITEBACK into PAGE_START_WRITEBACK Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 06/22] btrfs: extent_io: introduce a helper to grab an existing extent buffer from a page Qu Wenruo
2021-01-12 15:08   ` David Sterba
2021-01-06  1:01 ` [PATCH v3 07/22] btrfs: extent_io: introduce the skeleton of btrfs_subpage structure Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 08/22] btrfs: extent_io: make attach_extent_buffer_page() to handle subpage case Qu Wenruo
2021-01-06  6:54   ` Qu Wenruo
2021-01-07  1:35   ` Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 09/22] btrfs: extent_io: make grab_extent_buffer_from_page() " Qu Wenruo
2021-01-06  1:01 ` Qu Wenruo [this message]
2021-01-06  1:01 ` [PATCH v3 11/22] btrfs: extent_io: attach private to dummy extent buffer pages Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 12/22] btrfs: subpage: introduce helper for subpage uptodate status Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 13/22] btrfs: subpage: introduce helper for subpage error status Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 14/22] btrfs: extent_io: make set/clear_extent_buffer_uptodate() to support subpage size Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 15/22] btrfs: extent_io: make btrfs_clone_extent_buffer() to be subpage compatible Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 16/22] btrfs: extent_io: implement try_release_extent_buffer() for subpage metadata support Qu Wenruo
2021-01-06  8:24   ` Qu Wenruo
2021-01-06  8:43     ` Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 17/22] btrfs: extent_io: introduce read_extent_buffer_subpage() Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 18/22] btrfs: extent_io: make endio_readpage_update_page_status() to handle subpage case Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 19/22] btrfs: disk-io: introduce subpage metadata validation check Qu Wenruo
2021-01-06  1:01 ` [PATCH v3 20/22] btrfs: introduce btrfs_subpage for data inodes Qu Wenruo
2021-01-06  5:04   ` kernel test robot
2021-01-06  5:04     ` kernel test robot
2021-01-06  5:32     ` Qu Wenruo
2021-01-06  6:48       ` Rong Chen
2021-01-06  6:48         ` Rong Chen
2021-01-09  9:53   ` kernel test robot
2021-01-09  9:53     ` kernel test robot
2021-01-06  1:02 ` [PATCH v3 21/22] btrfs: integrate page status update for read path into begin/end_page_read() Qu Wenruo
2021-01-06  1:02 ` [PATCH v3 22/22] btrfs: allow RO mount of 4K sector size fs on 64K page system Qu Wenruo
2021-01-12 15:14 ` [PATCH v3 00/22] btrfs: add read-only support for subpage sector size David Sterba
2021-01-13  5:06   ` Qu Wenruo

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=20210106010201.37864-11-wqu@suse.com \
    --to=wqu@suse.com \
    --cc=linux-btrfs@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.