block: add a bi_error field to struct bio

Currently we have two different ways to signal an I/O error on a BIO:

 (1) by clearing the BIO_UPTODATE flag
 (2) by returning a Linux errno value to the bi_end_io callback

The first one has the drawback of only communicating a single possible
error (-EIO), and the second one has the drawback of not beeing persistent
when bios are queued up, and are not passed along from child to parent
bio in the ever more popular chaining scenario.  Having both mechanisms
available has the additional drawback of utterly confusing driver authors
and introducing bugs where various I/O submitters only deal with one of
them, and the others have to add boilerplate code to deal with both kinds
of error returns.

So add a new bi_error field to store an errno value directly in struct
bio and remove the existing mechanisms to clean all this up.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: NeilBrown <neilb@suse.com>
Signed-off-by: Jens Axboe <axboe@fb.com>

1 files changed, 26 insertions, 24 deletions

diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index b33c0cf02668..6b8becfe2057 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1845,8 +1845,10 @@ static int __btrfs_submit_bio_done(struct inode *inode, int rw, struct bio *bio,
         int ret;
 
         ret = btrfs_map_bio(root, rw, bio, mirror_num, 1);
-        if (ret)
-                bio_endio(bio, ret);
+        if (ret) {
+                bio->bi_error = ret;
+                bio_endio(bio);
+        }
         return ret;
 }
 
@@ -1906,8 +1908,10 @@ mapit:
         ret = btrfs_map_bio(root, rw, bio, mirror_num, 0);
 
 out:
-        if (ret < 0)
-                bio_endio(bio, ret);
+        if (ret < 0) {
+                bio->bi_error = ret;
+                bio_endio(bio);
+        }
         return ret;
 }
 
@@ -7689,13 +7693,13 @@ struct btrfs_retry_complete {
         int uptodate;
 };
 
-static void btrfs_retry_endio_nocsum(struct bio *bio, int err)
+static void btrfs_retry_endio_nocsum(struct bio *bio)
 {
         struct btrfs_retry_complete *done = bio->bi_private;
         struct bio_vec *bvec;
         int i;
 
-        if (err)
+        if (bio->bi_error)
                 goto end;
 
         done->uptodate = 1;
@@ -7744,7 +7748,7 @@ try_again:
         return 0;
 }
 
-static void btrfs_retry_endio(struct bio *bio, int err)
+static void btrfs_retry_endio(struct bio *bio)
 {
         struct btrfs_retry_complete *done = bio->bi_private;
         struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
@@ -7753,7 +7757,7 @@ static void btrfs_retry_endio(struct bio *bio, int err)
         int ret;
         int i;
 
-        if (err)
+        if (bio->bi_error)
                 goto end;
 
         uptodate = 1;
@@ -7836,12 +7840,13 @@ static int btrfs_subio_endio_read(struct inode *inode,
         }
 }
 
-static void btrfs_endio_direct_read(struct bio *bio, int err)
+static void btrfs_endio_direct_read(struct bio *bio)
 {
         struct btrfs_dio_private *dip = bio->bi_private;
         struct inode *inode = dip->inode;
         struct bio *dio_bio;
         struct btrfs_io_bio *io_bio = btrfs_io_bio(bio);
+        int err = bio->bi_error;
 
         if (dip->flags & BTRFS_DIO_ORIG_BIO_SUBMITTED)
                 err = btrfs_subio_endio_read(inode, io_bio, err);
@@ -7852,17 +7857,14 @@ static void btrfs_endio_direct_read(struct bio *bio, int err)
 
         kfree(dip);
 
-        /* If we had a csum failure make sure to clear the uptodate flag */
-        if (err)
-                clear_bit(BIO_UPTODATE, &dio_bio->bi_flags);
-        dio_end_io(dio_bio, err);
+        dio_end_io(dio_bio, bio->bi_error);
 
         if (io_bio->end_io)
                 io_bio->end_io(io_bio, err);
         bio_put(bio);
 }
 
-static void btrfs_endio_direct_write(struct bio *bio, int err)
+static void btrfs_endio_direct_write(struct bio *bio)
 {
         struct btrfs_dio_private *dip = bio->bi_private;
         struct inode *inode = dip->inode;
@@ -7876,7 +7878,8 @@ static void btrfs_endio_direct_write(struct bio *bio, int err)
 again:
         ret = btrfs_dec_test_first_ordered_pending(inode, &ordered,
                                                    &ordered_offset,
-                                                   ordered_bytes, !err);
+                                                   ordered_bytes,
+                                                   !bio->bi_error);
         if (!ret)
                 goto out_test;
 
@@ -7899,10 +7902,7 @@ out_test:
 
         kfree(dip);
 
-        /* If we had an error make sure to clear the uptodate flag */
-        if (err)
-                clear_bit(BIO_UPTODATE, &dio_bio->bi_flags);
-        dio_end_io(dio_bio, err);
+        dio_end_io(dio_bio, bio->bi_error);
         bio_put(bio);
 }
 
@@ -7917,9 +7917,10 @@ static int __btrfs_submit_bio_start_direct_io(struct inode *inode, int rw,
         return 0;
 }
 
-static void btrfs_end_dio_bio(struct bio *bio, int err)
+static void btrfs_end_dio_bio(struct bio *bio)
 {
         struct btrfs_dio_private *dip = bio->bi_private;
+        int err = bio->bi_error;
 
         if (err)
                 btrfs_warn(BTRFS_I(dip->inode)->root->fs_info,
@@ -7948,8 +7949,8 @@ static void btrfs_end_dio_bio(struct bio *bio, int err)
         if (dip->errors) {
                 bio_io_error(dip->orig_bio);
         } else {
-                set_bit(BIO_UPTODATE, &dip->dio_bio->bi_flags);
-                bio_endio(dip->orig_bio, 0);
+                dip->dio_bio->bi_error = 0;
+                bio_endio(dip->orig_bio);
         }
 out:
         bio_put(bio);
@@ -8220,7 +8221,8 @@ free_ordered:
          * callbacks - they require an allocated dip and a clone of dio_bio.
          */
         if (io_bio && dip) {
-                bio_endio(io_bio, ret);
+                io_bio->bi_error = -EIO;
+                bio_endio(io_bio);
                 /*
                  * The end io callbacks free our dip, do the final put on io_bio
                  * and all the cleanup and final put for dio_bio (through
@@ -8247,7 +8249,7 @@ free_ordered:
                         unlock_extent(&BTRFS_I(inode)->io_tree, file_offset,
                               file_offset + dio_bio->bi_iter.bi_size - 1);
                 }
-                clear_bit(BIO_UPTODATE, &dio_bio->bi_flags);
+                dio_bio->bi_error = -EIO;
                 /*
                  * Releases and cleans up our dio_bio, no need to bio_put()
                  * nor bio_endio()/bio_io_error() against dio_bio.