io_uring/rw: fix short rw error handling

[ upstream commit 89473c1a9205760c4fa6d158058da7b594a815f0 ]

We have a couple of problems, first reports of unexpected link breakage
for reads when cqe->res indicates that the IO was done in full. The
reason here is partial IO with retries.

TL;DR; we compare the result in __io_complete_rw_common() against
req->cqe.res, but req->cqe.res doesn't store the full length but rather
the length left to be done. So, when we pass the full corrected result
via kiocb_done() -> __io_complete_rw_common(), it fails.

The second problem is that we don't try to correct res in
io_complete_rw(), which, for instance, might be a problem for O_DIRECT
but when a prefix of data was cached in the page cache. We also
definitely don't want to pass a corrected result into io_rw_done().

The fix here is to leave __io_complete_rw_common() alone, always pass
not corrected result into it and fix it up as the last step just before
actually finishing the I/O.

Cc: stable@vger.kernel.org
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 4d32483..49c0e4e 100644 (file)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -2701,6 +2701,20 @@ static bool __io_complete_rw_common(struct io_kiocb *req, long res)
        return false;
 }
 
+static inline unsigned io_fixup_rw_res(struct io_kiocb *req, unsigned res)
+{
+       struct io_async_rw *io = req->async_data;
+
+       /* add previously done IO, if any */
+       if (io && io->bytes_done > 0) {
+               if (res < 0)
+                       res = io->bytes_done;
+               else
+                       res += io->bytes_done;
+       }
+       return res;
+}
+
 static void io_req_task_complete(struct io_kiocb *req, bool *locked)
 {
        unsigned int cflags = io_put_rw_kbuf(req);
@@ -2724,7 +2738,7 @@ static void __io_complete_rw(struct io_kiocb *req, long res, long res2,
 {
        if (__io_complete_rw_common(req, res))
                return;
-       __io_req_complete(req, issue_flags, req->result, io_put_rw_kbuf(req));
+       __io_req_complete(req, issue_flags, io_fixup_rw_res(req, res), io_put_rw_kbuf(req));
 }
 
 static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
@@ -2733,7 +2747,7 @@ static void io_complete_rw(struct kiocb *kiocb, long res, long res2)
 
        if (__io_complete_rw_common(req, res))
                return;
-       req->result = res;
+       req->result = io_fixup_rw_res(req, res);
        req->io_task_work.func = io_req_task_complete;
        io_req_task_work_add(req);
 }
@@ -2979,15 +2993,6 @@ static void kiocb_done(struct kiocb *kiocb, ssize_t ret,
                       unsigned int issue_flags)
 {
        struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb);
-       struct io_async_rw *io = req->async_data;
-
-       /* add previously done IO, if any */
-       if (io && io->bytes_done > 0) {
-               if (ret < 0)
-                       ret = io->bytes_done;
-               else
-                       ret += io->bytes_done;
-       }
 
        if (req->flags & REQ_F_CUR_POS)
                req->file->f_pos = kiocb->ki_pos;
@@ -3004,6 +3009,7 @@ static void kiocb_done(struct kiocb *kiocb, ssize_t ret,
                        unsigned int cflags = io_put_rw_kbuf(req);
                        struct io_ring_ctx *ctx = req->ctx;
 
+                       ret = io_fixup_rw_res(req, ret);
                        req_set_fail(req);
                        if (!(issue_flags & IO_URING_F_NONBLOCK)) {
                                mutex_lock(&ctx->uring_lock);