iov_iter: add ITER_BVEC_FLAG_NO_REF flag

For ITER_BVEC, if we're holding on to kernel pages, the caller
doesn't need to grab a reference to the bvec pages, and drop that
same reference on IO completion. This is essentially safe for any
ITER_BVEC, but some use cases end up reusing pages and uncondtionally
dropping a page reference on completion. And example of that is
sendfile(2), that ends up being a splice_in + splice_out on the
pipe pages.

Add a flag that tells us it's fine to not grab a page reference
to the bvec pages, since that caller knows not to drop a reference
when it's done with the pages.

Signed-off-by: Jens Axboe <axboe@kernel.dk>

2 files changed, 22 insertions, 5 deletions

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 4c6a5e60ddbe..c592a0933b0d 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -855,6 +855,9 @@ static int io_import_fixed(struct io_ring_ctx *ctx, int rw,
         iov_iter_bvec(iter, rw, imu->bvec, imu->nr_bvecs, offset + len);
         if (offset)
                 iov_iter_advance(iter, offset);
+
+        /* don't drop a reference to these pages */
+        iter->type |= ITER_BVEC_FLAG_NO_REF;
         return 0;
 }
 
diff --git a/include/linux/uio.h b/include/linux/uio.h
index ecf584f6b82d..4e926641fa80 100644
--- a/include/linux/uio.h
+++ b/include/linux/uio.h
@@ -23,14 +23,23 @@ struct kvec {
 };
 
 enum iter_type {
-        ITER_IOVEC = 0,
-        ITER_KVEC = 2,
-        ITER_BVEC = 4,
-        ITER_PIPE = 8,
-        ITER_DISCARD = 16,
+        /* set if ITER_BVEC doesn't hold a bv_page ref */
+        ITER_BVEC_FLAG_NO_REF = 2,
+
+        /* iter types */
+        ITER_IOVEC = 4,
+        ITER_KVEC = 8,
+        ITER_BVEC = 16,
+        ITER_PIPE = 32,
+        ITER_DISCARD = 64,
 };
 
 struct iov_iter {
+        /*
+         * Bit 0 is the read/write bit, set if we're writing.
+         * Bit 1 is the BVEC_FLAG_NO_REF bit, set if type is a bvec and
+         * the caller isn't expecting to drop a page reference when done.
+         */
         unsigned int type;
         size_t iov_offset;
         size_t count;
@@ -84,6 +93,11 @@ static inline unsigned char iov_iter_rw(const struct iov_iter *i)
         return i->type & (READ | WRITE);
 }
 
+static inline bool iov_iter_bvec_no_ref(const struct iov_iter *i)
+{
+        return (i->type & ITER_BVEC_FLAG_NO_REF) != 0;
+}
+
 /*
  * Total number of bytes covered by an iovec.
  *