On Tue, Jun 20, 2017 at 09:30, David Howells wrote:
> David Rientjes <rientjes@xxxxxxxxxx> wrote:
>
> > I would have expected to see down_read_killable() actually used somewhere
> > after its implementation as part of this patchset.
>
> There are some places we should be using down_{read|write}_interruptible(), if
> it existed, dressed as inode_lock{,_shared}_interruptible().
Then let's use it in iterate_dir():
[PATCH]fs: Use killable down_read() in iterate_dir()
There was mutex_lock_interruptible() initially, and it was changed
to rwsem, but there were not killable rwsem primitives that time.
>From commit 9902af79c01a:
"The main issue is the lack of down_write_killable(), so the places
like readdir.c switched to plain inode_lock(); once killable
variants of rwsem primitives appear, that'll be dealt with"
Use down_read_killable() same as down_write_killable() in !shared
case, as concurrent inode_lock() may take much time, that may be
wanted to be interrupted by user.
Signed-off-by: Kirill Tkhai <ktkhai@xxxxxxxxxxxxx>
---
fs/readdir.c | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/fs/readdir.c b/fs/readdir.c
index 89659549c09d..7c584bbb4ce3 100644
--- a/fs/readdir.c
+++ b/fs/readdir.c
@@ -36,13 +36,12 @@ int iterate_dir(struct file *file, struct dir_context *ctx)
if (res)
goto out;
- if (shared) {
- inode_lock_shared(inode);
- } else {
+ if (shared)
+ res = down_read_killable(&inode->i_rwsem);
+ else
res = down_write_killable(&inode->i_rwsem);
- if (res)
- goto out;
- }
+ if (res)
+ goto out;
res = -ENOENT;
if (!IS_DEADDIR(inode)) {
