Re: [vm] writing to UDF DVD+RW (/dev/sr0) while under memorypressure: box ==> doorstop

[Mike Galbraith]

On Thu, 2008-01-10 at 15:41 +0100, Jan Kara wrote:
> > On Wed, 09 Jan 2008 12:11:20 +0100
> > 
> > 
> > I wonder why UDF was doing a synchronous write in there.  In fact I wonder
> > why it's writing the inode at all?  extN doesn't do that.  If for some
> > reason it really does want to make the inode immediately reclaimable then
> > simply shoving it down into the /dev/hda1 pagecache should be sufficient
> > (ie: what you did)..
>   Looking at the code, I think UDF change is correct. UDF has to call
> write_inode_now() because by the time clear_inode() is called, inode is
> already written by VFS and prepared to be freed. But then UDF modifies
> it in udf_clear_inode() (removes preallocation) and for these changes to
> get to disk you have to write the inode explicitely. 
>   But there's really no need to wait on IO. We only have to copy all
> data from inode structure into buffers and that happens even if we don't
> wait on sync.

Perhaps I should go ahead and submit it then.  There are 5 other async
callers as well, so VM/UDF reclaim buglet can die, and those others can
get what they asked for with net diffstat of 0.

Fix udf_clear_inode() to request asynchronous writeout in icache reclaim
path, and ensure that write_inore_now() honors that request, lest
allocators needlessly block on iprune_mutex.

Signed-off-by: Mike Galbraith <efault@xxxxxx>

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 0fca820..f1cce24 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -657,7 +657,7 @@ int write_inode_now(struct inode *inode, int sync)
 	int ret;
 	struct writeback_control wbc = {
 		.nr_to_write = LONG_MAX,
-		.sync_mode = WB_SYNC_ALL,
+		.sync_mode = sync ? WB_SYNC_ALL : WB_SYNC_NONE,
 		.range_start = 0,
 		.range_end = LLONG_MAX,
 	};
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index 6ff8151..d1fc116 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -117,7 +117,7 @@ void udf_clear_inode(struct inode *inode)
 		udf_discard_prealloc(inode);
 		udf_truncate_tail_extent(inode);
 		unlock_kernel();
-		write_inode_now(inode, 1);
+		write_inode_now(inode, 0);
 	}
 	kfree(UDF_I_DATA(inode));
 	UDF_I_DATA(inode) = NULL;


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/