2012-03-02 16:20:39

by Curt Wohlgemuth

Subject: [PATCH] ext4: Don't release page refs in ext4_end_bio()

We can clear PageWriteback on each page when the IO
completes, but we can't release the references on the page
until we convert any uninitialized extents.

Without this patch, the use of the dioread_nolock mount
option can break buffered writes, because extents may
not be converted by the time a subsequent buffered read
comes in; if the page is not in the page cache, a read
will return zeros if the extent is still uninitialized.

I tested this with a (temporary) patch that adds a call
to msleep(1000) at the start of ext4_end_io_work(), to delay
processing of each DIO-unwritten work queue item. With this
msleep(), a simple workload of

fallocate
write
fadvise
read

will fail without this patch, succeeds with it.

Signed-off-by: Curt Wohlgemuth <[email protected]>
---
fs/ext4/page-io.c | 7 ++++---
1 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c
index e6271fe..0991c22 100644
--- a/fs/ext4/page-io.c
+++ b/fs/ext4/page-io.c
@@ -63,7 +63,6 @@ void ext4_ioend_wait(struct inode *inode)
static void put_io_page(struct ext4_io_page *io_page)
{
if (atomic_dec_and_test(&io_page->p_count)) {
- end_page_writeback(io_page->p_page);
put_page(io_page->p_page);
kmem_cache_free(io_page_cachep, io_page);
}
@@ -235,9 +234,9 @@ static void ext4_end_bio(struct bio *bio, int error)
} while (bh != head);
}

- put_io_page(io_end->pages[i]);
+ if (atomic_read(&io_end->pages[i]->p_count) == 1)
+ end_page_writeback(io_end->pages[i]->p_page);
}
- io_end->num_io_pages = 0;
inode = io_end->inode;

if (error) {
@@ -429,6 +428,8 @@ int ext4_bio_write_page(struct ext4_io_submit *io,
* PageWriteback bit from the page to prevent the system from
* wedging later on.
*/
+ if (atomic_read(&io_page->p_count) == 1)
+ end_page_writeback(page);
put_io_page(io_page);
return ret;
}
--
1.7.7.3

2012-03-05 15:42:17

by Theodore Ts'o

Subject: Re: [PATCH] ext4: Don't release page refs in ext4_end_bio()

On Fri, Mar 02, 2012 at 08:19:59AM -0800, Curt Wohlgemuth wrote:
> We can clear PageWriteback on each page when the IO
> completes, but we can't release the references on the page
> until we convert any uninitialized extents.
>
> Without this patch, the use of the dioread_nolock mount
> option can break buffered writes, because extents may
> not be converted by the time a subsequent buffered read
> comes in; if the page is not in the page cache, a read
> will return zeros if the extent is still uninitialized.
>
> I tested this with a (temporary) patch that adds a call
> to msleep(1000) at the start of ext4_end_io_work(), to delay
> processing of each DIO-unwritten work queue item. With this
> msleep(), a simple workload of
>
> fallocate
> write
> fadvise
> read
>
> will fail without this patch, succeeds with it.
>
> Signed-off-by: Curt Wohlgemuth <[email protected]>

Thanks, applied.

- Ted