iomap: get/put the page in iomap_page_create/release()
PATCH 1/1 iomap: get/put the page in iomap_page_create/release()
From: Piotr Jaroszynski <[email protected]>
migrate_page_move_mapping() expects pages with private data set to have a page_count elevated by 1. This is what used to happen for xfs through the buffer_heads code before the switch to iomap in commit 82cb14175e7d ("xfs: add support for sub-pagesize writeback without buffer_heads"). Not having the count elevated causes move_pages() to fail on memory mapped files coming from xfs.
Make iomap compatible with the migrate_page_move_mapping() assumption by elevating the page count as part of iomap_page_create() and lowering it in iomap_page_release().
It causes the move_pages() syscall to misbehave on memory mapped files from xfs. It does not not move any pages, which I suppose is "just" a perf issue, but it also ends up returning a positive number which is out of spec for the syscall. Talking to Michal Hocko, it sounds like returning positive numbers might be a necessary update to move_pages() anyway though.
[hch: actually get/put the page iomap_migrate_page() to make it work
properly]
--- fs/iomap.c | 9 +++++++++ 1 file changed, 9 insertions(+)
fs/iomap.c +9 -0
@@ -116,6 +116,12 @@ iomap_page_create(struct inode *inode, struct page *page)atomic_set (&iop->read_count,0 );atomic_set (&iop->write_count,0 );bitmap_zero (iop->uptodate,PAGE_SIZE /SECTOR_SIZE ); + +/* +* migrate_page_move_mapping() assumes that pages with private data have +* their count elevated by 1. +*/ +get_page (page);set_page_private (page, (unsigned long )iop);SetPagePrivate (page);return iop;
@@ -132,6 +138,7 @@ iomap_page_release(struct page *page)WARN_ON_ONCE (atomic_read (&iop->write_count));ClearPagePrivate (page);set_page_private (page,0 ); +put_page (page);kfree (iop); }
@@ -569,8 +576,10 @@ iomap_migrate_page(struct address_space *mapping, struct page *newpage,if (page_has_private (page)) {ClearPagePrivate (page); +get_page (newpage);set_page_private (newpage,page_private (page));set_page_private (page,0 ); +put_page (page);SetPagePrivate (newpage); }
Signature · 1 line
2.20.1
Quoted 20 lines · Christoph Hellwig
From: Piotr Jaroszynski <[email protected]>
migrate_page_move_mapping()expects pages with private data set to have apage_countelevated by 1. This is what used to happen for xfs through thebuffer_headscode before the switch to iomap in commit82cb14175e7d("xfs: add support for sub-pagesize writeback withoutbuffer_heads"). Not having the count elevated causesmove_pages()to fail on memory mapped files coming from xfs.Make iomap compatible with the
migrate_page_move_mapping()assumption by elevating the page count as part ofiomap_page_create()and lowering it iniomap_page_release().It causes the
move_pages()syscall to misbehave on memory mapped files from xfs. It does not not move any pages, which I suppose is "just" a perf issue, but it also ends up returning a positive number which is out of spec for the syscall. Talking to Michal Hocko, it sounds like returning positive numbers might be a necessary update tomove_pages()anyway though.
Quoted 5 lines
Fixes 82cb14175e7d ("xfs: add support for sub-pagesize writeback without buffer_heads")Signed-off-by Piotr Jaroszynski <[email protected]>[hch: actually get/put the page
iomap_migrate_page()to make it work
properly]Signed-off-by Christoph Hellwig <[email protected]>
Looks ok (apparently I forgot to ack this explicitly...)
--D
Quoted 43 lines
--- fs/iomap.c | 9 +++++++++ 1 file changed, 9 insertions(+)fs/iomap.c +9
-0@@ -116,6 +116,12 @@ iomap_page_create(struct inode *inode, struct page *page)atomic_set (&iop->read_count,0 );atomic_set (&iop->write_count,0 );bitmap_zero (iop->uptodate,PAGE_SIZE /SECTOR_SIZE ); + +/* +* migrate_page_move_mapping() assumes that pages with private data have +* their count elevated by 1. +*/ +get_page (page);set_page_private (page, (unsigned long )iop);SetPagePrivate (page);return iop;@@ -132,6 +138,7 @@ iomap_page_release(struct page *page)WARN_ON_ONCE (atomic_read (&iop->write_count));ClearPagePrivate (page);set_page_private (page,0 ); +put_page (page);kfree (iop); }@@ -569,8 +576,10 @@ iomap_migrate_page(struct address_space *mapping, struct page *newpage,if (page_has_private (page)) {ClearPagePrivate (page); +get_page (newpage);set_page_private (newpage,page_private (page));set_page_private (page,0 ); +put_page (page);SetPagePrivate (newpage); }Signature · 1 line
2.20.1