All of lore.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: David Sterba <dsterba@suse.com>,
	Josef Bacik <josef@toxicpanda.com>, Qu Wenruo <wqu@suse.com>
Cc: linux-btrfs@vger.kernel.org
Subject: [PATCH 05/10] btrfs: defer I/O completion based on the btrfs_raid_bio
Date: Wed,  4 May 2022 05:25:19 -0700	[thread overview]
Message-ID: <20220504122524.558088-6-hch@lst.de> (raw)
In-Reply-To: <20220504122524.558088-1-hch@lst.de>

Instead of attaching a an extra allocation an indirect call to each
low-level bio issued by the RAID code, add a work_struct to struct
btrfs_raid_bio and only defer the per-rbio completion action.  The
per-bio action for all the I/Os are trivial and can be safely done
from interrupt context.

As a nice side effect this also allows sharing the boilerplate code
for the per-bio completions

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 fs/btrfs/ctree.h   |   2 +-
 fs/btrfs/disk-io.c |  12 ++---
 fs/btrfs/disk-io.h |   1 -
 fs/btrfs/raid56.c  | 109 ++++++++++++++++++---------------------------
 4 files changed, 48 insertions(+), 76 deletions(-)

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 6f539ddf7467a..a3739143bf44c 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -853,7 +853,7 @@ struct btrfs_fs_info {
 	struct btrfs_workqueue *flush_workers;
 	struct btrfs_workqueue *endio_workers;
 	struct btrfs_workqueue *endio_meta_workers;
-	struct btrfs_workqueue *endio_raid56_workers;
+	struct workqueue_struct *endio_raid56_workers;
 	struct workqueue_struct *rmw_workers;
 	struct btrfs_workqueue *endio_meta_write_workers;
 	struct btrfs_workqueue *endio_write_workers;
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 56d4d4db976bd..ef117eaab468c 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -753,14 +753,10 @@ static void end_workqueue_bio(struct bio *bio)
 			wq = fs_info->endio_meta_write_workers;
 		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE)
 			wq = fs_info->endio_freespace_worker;
-		else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
-			wq = fs_info->endio_raid56_workers;
 		else
 			wq = fs_info->endio_write_workers;
 	} else {
-		if (end_io_wq->metadata == BTRFS_WQ_ENDIO_RAID56)
-			wq = fs_info->endio_raid56_workers;
-		else if (end_io_wq->metadata)
+		if (end_io_wq->metadata)
 			wq = fs_info->endio_meta_workers;
 		else
 			wq = fs_info->endio_workers;
@@ -2273,7 +2269,8 @@ static void btrfs_stop_all_workers(struct btrfs_fs_info *fs_info)
 	btrfs_destroy_workqueue(fs_info->hipri_workers);
 	btrfs_destroy_workqueue(fs_info->workers);
 	btrfs_destroy_workqueue(fs_info->endio_workers);
-	btrfs_destroy_workqueue(fs_info->endio_raid56_workers);
+	if (fs_info->endio_raid56_workers)
+		destroy_workqueue(fs_info->endio_raid56_workers);
 	if (fs_info->rmw_workers)
 		destroy_workqueue(fs_info->rmw_workers);
 	btrfs_destroy_workqueue(fs_info->endio_write_workers);
@@ -2476,8 +2473,7 @@ static int btrfs_init_workqueues(struct btrfs_fs_info *fs_info)
 		btrfs_alloc_workqueue(fs_info, "endio-meta-write", flags,
 				      max_active, 2);
 	fs_info->endio_raid56_workers =
-		btrfs_alloc_workqueue(fs_info, "endio-raid56", flags,
-				      max_active, 4);
+		alloc_workqueue("btrfs-endio-raid56", flags, max_active);
 	fs_info->rmw_workers = alloc_workqueue("btrfs-rmw", flags, max_active);
 	fs_info->endio_write_workers =
 		btrfs_alloc_workqueue(fs_info, "endio-write", flags,
diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h
index 4ee8c42c9f783..809ef065f1666 100644
--- a/fs/btrfs/disk-io.h
+++ b/fs/btrfs/disk-io.h
@@ -21,7 +21,6 @@ enum btrfs_wq_endio_type {
 	BTRFS_WQ_ENDIO_DATA,
 	BTRFS_WQ_ENDIO_METADATA,
 	BTRFS_WQ_ENDIO_FREE_SPACE,
-	BTRFS_WQ_ENDIO_RAID56,
 };
 
 static inline u64 btrfs_sb_offset(int mirror)
diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
index a5b623ee6facd..8e089fb4212ff 100644
--- a/fs/btrfs/raid56.c
+++ b/fs/btrfs/raid56.c
@@ -164,6 +164,9 @@ struct btrfs_raid_bio {
 	atomic_t stripes_pending;
 
 	atomic_t error;
+
+	struct work_struct end_io_work;
+
 	/*
 	 * these are two arrays of pointers.  We allocate the
 	 * rbio big enough to hold them both and setup their
@@ -1552,15 +1555,7 @@ static void set_bio_pages_uptodate(struct btrfs_raid_bio *rbio, struct bio *bio)
 	}
 }
 
-/*
- * end io for the read phase of the rmw cycle.  All the bios here are physical
- * stripe bios we've read from the disk so we can recalculate the parity of the
- * stripe.
- *
- * This will usually kick off finish_rmw once all the bios are read in, but it
- * may trigger parity reconstruction if we had any errors along the way
- */
-static void raid_rmw_end_io(struct bio *bio)
+static void raid56_bio_end_io(struct bio *bio)
 {
 	struct btrfs_raid_bio *rbio = bio->bi_private;
 
@@ -1571,23 +1566,34 @@ static void raid_rmw_end_io(struct bio *bio)
 
 	bio_put(bio);
 
-	if (!atomic_dec_and_test(&rbio->stripes_pending))
-		return;
+	if (atomic_dec_and_test(&rbio->stripes_pending))
+		queue_work(rbio->bioc->fs_info->endio_raid56_workers,
+			   &rbio->end_io_work);
+}
 
-	if (atomic_read(&rbio->error) > rbio->bioc->max_errors)
-		goto cleanup;
+/*
+ * End io handler for the read phase of the rmw cycle.  All the bios here are
+ * physical stripe bios we've read from the disk so we can recalculate the
+ * parity of the stripe.
+ *
+ * This will usually kick off finish_rmw once all the bios are read in, but it
+ * may trigger parity reconstruction if we had any errors along the way
+ */
+static void raid56_rmw_end_io_work(struct work_struct *work)
+{
+	struct btrfs_raid_bio *rbio =
+		container_of(work, struct btrfs_raid_bio, end_io_work);
+
+	if (atomic_read(&rbio->error) > rbio->bioc->max_errors) {
+		rbio_orig_end_io(rbio, BLK_STS_IOERR);
+		return;
+	}
 
 	/*
-	 * this will normally call finish_rmw to start our write
-	 * but if there are any failed stripes we'll reconstruct
-	 * from parity first
+	 * This will normally call finish_rmw to start our write but if there
+	 * are any failed stripes we'll reconstruct from parity first.
 	 */
 	validate_rbio_for_rmw(rbio);
-	return;
-
-cleanup:
-
-	rbio_orig_end_io(rbio, BLK_STS_IOERR);
 }
 
 /*
@@ -1662,11 +1668,9 @@ static int raid56_rmw_stripe(struct btrfs_raid_bio *rbio)
 	 * touch it after that.
 	 */
 	atomic_set(&rbio->stripes_pending, bios_to_read);
+	INIT_WORK(&rbio->end_io_work, raid56_rmw_end_io_work);
 	while ((bio = bio_list_pop(&bio_list))) {
-		bio->bi_end_io = raid_rmw_end_io;
-
-		btrfs_bio_wq_end_io(rbio->bioc->fs_info, bio, BTRFS_WQ_ENDIO_RAID56);
-
+		bio->bi_end_io = raid56_bio_end_io;
 		submit_bio(bio);
 	}
 	/* the actual write will happen once the reads are done */
@@ -2108,25 +2112,13 @@ static void __raid_recover_end_io(struct btrfs_raid_bio *rbio)
 }
 
 /*
- * This is called only for stripes we've read from disk to
- * reconstruct the parity.
+ * This is called only for stripes we've read from disk to reconstruct the
+ * parity.
  */
-static void raid_recover_end_io(struct bio *bio)
+static void raid_recover_end_io_work(struct work_struct *work)
 {
-	struct btrfs_raid_bio *rbio = bio->bi_private;
-
-	/*
-	 * we only read stripe pages off the disk, set them
-	 * up to date if there were no errors
-	 */
-	if (bio->bi_status)
-		fail_bio_stripe(rbio, bio);
-	else
-		set_bio_pages_uptodate(rbio, bio);
-	bio_put(bio);
-
-	if (!atomic_dec_and_test(&rbio->stripes_pending))
-		return;
+	struct btrfs_raid_bio *rbio =
+		container_of(work, struct btrfs_raid_bio, end_io_work);
 
 	if (atomic_read(&rbio->error) > rbio->bioc->max_errors)
 		rbio_orig_end_io(rbio, BLK_STS_IOERR);
@@ -2209,11 +2201,9 @@ static int __raid56_parity_recover(struct btrfs_raid_bio *rbio)
 	 * touch it after that.
 	 */
 	atomic_set(&rbio->stripes_pending, bios_to_read);
+	INIT_WORK(&rbio->end_io_work, raid_recover_end_io_work);
 	while ((bio = bio_list_pop(&bio_list))) {
-		bio->bi_end_io = raid_recover_end_io;
-
-		btrfs_bio_wq_end_io(rbio->bioc->fs_info, bio, BTRFS_WQ_ENDIO_RAID56);
-
+		bio->bi_end_io = raid56_bio_end_io;
 		submit_bio(bio);
 	}
 
@@ -2583,7 +2573,6 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio,
 
 	while ((bio = bio_list_pop(&bio_list))) {
 		bio->bi_end_io = raid_write_end_io;
-
 		submit_bio(bio);
 	}
 	return;
@@ -2671,24 +2660,14 @@ static void validate_rbio_for_parity_scrub(struct btrfs_raid_bio *rbio)
  * This will usually kick off finish_rmw once all the bios are read in, but it
  * may trigger parity reconstruction if we had any errors along the way
  */
-static void raid56_parity_scrub_end_io(struct bio *bio)
+static void raid56_parity_scrub_end_io_work(struct work_struct *work)
 {
-	struct btrfs_raid_bio *rbio = bio->bi_private;
-
-	if (bio->bi_status)
-		fail_bio_stripe(rbio, bio);
-	else
-		set_bio_pages_uptodate(rbio, bio);
-
-	bio_put(bio);
-
-	if (!atomic_dec_and_test(&rbio->stripes_pending))
-		return;
+	struct btrfs_raid_bio *rbio =
+		container_of(work, struct btrfs_raid_bio, end_io_work);
 
 	/*
-	 * this will normally call finish_rmw to start our write
-	 * but if there are any failed stripes we'll reconstruct
-	 * from parity first
+	 * This will normally call finish_rmw to start our write, but if there
+	 * are any failed stripes we'll reconstruct from parity first
 	 */
 	validate_rbio_for_parity_scrub(rbio);
 }
@@ -2758,11 +2737,9 @@ static void raid56_parity_scrub_stripe(struct btrfs_raid_bio *rbio)
 	 * touch it after that.
 	 */
 	atomic_set(&rbio->stripes_pending, bios_to_read);
+	INIT_WORK(&rbio->end_io_work, raid56_parity_scrub_end_io_work);
 	while ((bio = bio_list_pop(&bio_list))) {
-		bio->bi_end_io = raid56_parity_scrub_end_io;
-
-		btrfs_bio_wq_end_io(rbio->bioc->fs_info, bio, BTRFS_WQ_ENDIO_RAID56);
-
+		bio->bi_end_io = raid56_bio_end_io;
 		submit_bio(bio);
 	}
 	/* the actual write will happen once the reads are done */
-- 
2.30.2


  parent reply	other threads:[~2022-05-04 12:25 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-04 12:25 cleanup btrfs bio handling, part 2 v3 Christoph Hellwig
2022-05-04 12:25 ` [PATCH 01/10] btrfs: don't double-defer bio completions for compressed reads Christoph Hellwig
2022-05-04 12:25 ` [PATCH 02/10] btrfs: move more work into btrfs_end_bioc Christoph Hellwig
2022-05-04 12:25 ` [PATCH 03/10] btrfs: cleanup btrfs_submit_dio_bio Christoph Hellwig
2022-05-04 12:25 ` [PATCH 04/10] btrfs: split btrfs_submit_data_bio Christoph Hellwig
2022-05-04 12:38   ` Qu Wenruo
2022-05-04 14:08     ` Christoph Hellwig
2022-05-04 22:41       ` Qu Wenruo
2022-05-04 22:44         ` Qu Wenruo
2022-05-05 15:08         ` Christoph Hellwig
2022-05-04 12:25 ` Christoph Hellwig [this message]
2022-05-04 12:25 ` [PATCH 06/10] btrfs: don't use btrfs_bio_wq_end_io for compressed writes Christoph Hellwig
2022-05-11 19:20   ` Nikolay Borisov
2022-05-11 19:28     ` Nikolay Borisov
2022-05-12  5:56       ` Christoph Hellwig
2022-05-04 12:25 ` [PATCH 07/10] btrfs: centralize setting REQ_META Christoph Hellwig
2022-05-04 12:25 ` [PATCH 08/10] btrfs: remove btrfs_end_io_wq Christoph Hellwig
2022-05-05  2:34   ` Qu Wenruo
2022-05-05 15:09     ` Christoph Hellwig
2022-05-04 12:25 ` [PATCH 09/10] btrfs: refactor btrfs_map_bio Christoph Hellwig
2022-05-04 12:46   ` Qu Wenruo
2022-05-04 12:25 ` [PATCH 10/10] btrfs: do not allocate a btrfs_bio for low-level bios Christoph Hellwig
2022-05-05 11:23   ` Qu Wenruo
2022-05-05  6:56 ` cleanup btrfs bio handling, part 2 v3 Qu Wenruo
2022-05-05 15:11   ` Christoph Hellwig
2022-05-12  6:22     ` Anand Jain
2022-05-12  6:30       ` Anand Jain
2022-05-05 15:34   ` David Sterba
  -- strict thread matches above, loose matches on Subject: below --
2022-04-29 14:30 cleanup btrfs bio handling, part 2 v2 Christoph Hellwig
2022-04-29 14:30 ` [PATCH 05/10] btrfs: defer I/O completion based on the btrfs_raid_bio Christoph Hellwig
2022-05-01  4:40   ` Qu Wenruo
2022-05-01  4:53     ` Qu Wenruo
2022-05-02 16:44       ` Christoph Hellwig
2022-06-03 16:44       ` David Sterba
2022-06-03 16:45         ` David Sterba
2022-04-25  7:54 cleanup btrfs bio handling, part 2 Christoph Hellwig
2022-04-25  7:54 ` [PATCH 05/10] btrfs: defer I/O completion based on the btrfs_raid_bio Christoph Hellwig

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220504122524.558088-6-hch@lst.de \
    --to=hch@lst.de \
    --cc=dsterba@suse.com \
    --cc=josef@toxicpanda.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=wqu@suse.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.