* [PATCH] iomap: get/put the page in iomap_page_create/release()
@ 2019-01-21 15:17 Christoph Hellwig
2019-01-31 23:28 ` Darrick J. Wong
0 siblings, 1 reply; 5+ messages in thread
From: Christoph Hellwig @ 2019-01-21 15:17 UTC (permalink / raw)
To: linux-xfs; +Cc: linux-fsdevel, Piotr Jaroszynski
From: Piotr Jaroszynski <pjaroszynski@nvidia.com>
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.
Fixes: 82cb14175e7d ("xfs: add support for sub-pagesize writeback without buffer_heads")
Signed-off-by: Piotr Jaroszynski <pjaroszynski@nvidia.com>
[hch: actually get/put the page iomap_migrate_page() to make it work
properly]
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
fs/iomap.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/fs/iomap.c b/fs/iomap.c
index 987fefc054b4..47362397cb82 100644
--- a/fs/iomap.c
+++ b/fs/iomap.c
@@ -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);
}
--
2.20.1
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] iomap: get/put the page in iomap_page_create/release()
2019-01-21 15:17 [PATCH] iomap: get/put the page in iomap_page_create/release() Christoph Hellwig
@ 2019-01-31 23:28 ` Darrick J. Wong
0 siblings, 0 replies; 5+ messages in thread
From: Darrick J. Wong @ 2019-01-31 23:28 UTC (permalink / raw)
To: Christoph Hellwig; +Cc: linux-xfs, linux-fsdevel, Piotr Jaroszynski
On Mon, Jan 21, 2019 at 04:17:02PM +0100, Christoph Hellwig wrote:
> From: Piotr Jaroszynski <pjaroszynski@nvidia.com>
>
> 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.
>
> Fixes: 82cb14175e7d ("xfs: add support for sub-pagesize writeback without buffer_heads")
> Signed-off-by: Piotr Jaroszynski <pjaroszynski@nvidia.com>
> [hch: actually get/put the page iomap_migrate_page() to make it work
> properly]
> Signed-off-by: Christoph Hellwig <hch@lst.de>
Looks ok (apparently I forgot to ack this explicitly...)
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
--D
> ---
> fs/iomap.c | 9 +++++++++
> 1 file changed, 9 insertions(+)
>
> diff --git a/fs/iomap.c b/fs/iomap.c
> index 987fefc054b4..47362397cb82 100644
> --- a/fs/iomap.c
> +++ b/fs/iomap.c
> @@ -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);
> }
>
> --
> 2.20.1
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH] iomap: get/put the page in iomap_page_create/release()
@ 2018-11-15 0:30 p.jaroszynski
2018-11-15 9:30 ` Christoph Hellwig
0 siblings, 1 reply; 5+ messages in thread
From: p.jaroszynski @ 2018-11-15 0:30 UTC (permalink / raw)
To: Christoph Hellwig, Michal Hocko
Cc: linux-fsdevel, linux-mm, Piotr Jaroszynski
From: Piotr Jaroszynski <pjaroszynski@nvidia.com>
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 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().
Fixes: 82cb14175e7d ("xfs: add support for sub-pagesize writeback
without buffer_heads")
Signed-off-by: Piotr Jaroszynski <pjaroszynski@nvidia.com>
---
fs/iomap.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/fs/iomap.c b/fs/iomap.c
index 90c2febc93ac..23977f9f23a2 100644
--- a/fs/iomap.c
+++ b/fs/iomap.c
@@ -117,6 +117,12 @@ 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);
+
+ /*
+ * At least 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;
@@ -133,6 +139,7 @@ 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);
}
--
2.11.0.262.g4b0a5b2.dirty
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] iomap: get/put the page in iomap_page_create/release()
2018-11-15 0:30 p.jaroszynski
@ 2018-11-15 9:30 ` Christoph Hellwig
2018-11-15 18:44 ` Piotr Jaroszynski
0 siblings, 1 reply; 5+ messages in thread
From: Christoph Hellwig @ 2018-11-15 9:30 UTC (permalink / raw)
To: p.jaroszynski
Cc: Christoph Hellwig, Michal Hocko, linux-fsdevel, linux-mm,
Piotr Jaroszynski
> Fixes: 82cb14175e7d ("xfs: add support for sub-pagesize writeback
> without buffer_heads")
I've never seen line breaks in Fixes tags, is this really a valid format?
> + /*
> + * At least migrate_page_move_mapping() assumes that pages with private
> + * data have their count elevated by 1.
> + */
I'd drop the "At least".
Otherwise this looks fine to me:
Reviewed-by: Christoph Hellwig <hch@lst.de>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] iomap: get/put the page in iomap_page_create/release()
2018-11-15 9:30 ` Christoph Hellwig
@ 2018-11-15 18:44 ` Piotr Jaroszynski
0 siblings, 0 replies; 5+ messages in thread
From: Piotr Jaroszynski @ 2018-11-15 18:44 UTC (permalink / raw)
To: Christoph Hellwig, p.jaroszynski; +Cc: Michal Hocko, linux-fsdevel, linux-mm
On 11/15/18 1:30 AM, Christoph Hellwig wrote:
>> Fixes: 82cb14175e7d ("xfs: add support for sub-pagesize writeback
>> without buffer_heads")
>
> I've never seen line breaks in Fixes tags, is this really a valid format?
Probably not, fixed in v2.
>
>> + /*
>> + * At least migrate_page_move_mapping() assumes that pages with private
>> + * data have their count elevated by 1.
>> + */
>
> I'd drop the "At least".
Fixed in v2.
>
> Otherwise this looks fine to me:
>
> Reviewed-by: Christoph Hellwig <hch@lst.de>
>
Thanks!
Thanks,
Piotr
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2019-01-31 23:28 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-01-21 15:17 [PATCH] iomap: get/put the page in iomap_page_create/release() Christoph Hellwig
2019-01-31 23:28 ` Darrick J. Wong
-- strict thread matches above, loose matches on Subject: below --
2018-11-15 0:30 p.jaroszynski
2018-11-15 9:30 ` Christoph Hellwig
2018-11-15 18:44 ` Piotr Jaroszynski
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).