public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
From: cem@kernel.org
To: linux-xfs@vger.kernel.org
Cc: djwong@kernel.org, hch@lst.de
Subject: [PATCH 33/67] xfs: constrain dirty buffers while formatting a staged btree
Date: Mon, 22 Apr 2024 18:25:55 +0200	[thread overview]
Message-ID: <20240422163832.858420-35-cem@kernel.org> (raw)
In-Reply-To: <20240422163832.858420-2-cem@kernel.org>

From: "Darrick J. Wong" <djwong@kernel.org>

Source kernel commit: e069d549705e49841247acf9b3176744e27d5425

Constrain the number of dirty buffers that are locked by the btree
staging code at any given time by establishing a threshold at which we
put them all on the delwri queue and push them to disk.  This limits
memory consumption while writing out new btrees.

Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Carlos Maiolino <cem@kernel.org>
---
 libxfs/xfs_btree_staging.c | 50 +++++++++++++++++++++++++++++---------
 libxfs/xfs_btree_staging.h | 10 ++++++++
 repair/agbtree.c           |  1 +
 3 files changed, 50 insertions(+), 11 deletions(-)

diff --git a/libxfs/xfs_btree_staging.c b/libxfs/xfs_btree_staging.c
index a6f0d7d3b..d4164e37b 100644
--- a/libxfs/xfs_btree_staging.c
+++ b/libxfs/xfs_btree_staging.c
@@ -333,24 +333,41 @@ xfs_btree_commit_ifakeroot(
 /*
  * Put a btree block that we're loading onto the ordered list and release it.
  * The btree blocks will be written to disk when bulk loading is finished.
+ * If we reach the dirty buffer threshold, flush them to disk before
+ * continuing.
  */
-static void
+static int
 xfs_btree_bload_drop_buf(
-	struct list_head	*buffers_list,
-	struct xfs_buf		**bpp)
+	struct xfs_btree_bload		*bbl,
+	struct list_head		*buffers_list,
+	struct xfs_buf			**bpp)
 {
-	if (*bpp == NULL)
-		return;
+	struct xfs_buf			*bp = *bpp;
+	int				error;
+
+	if (!bp)
+		return 0;
 
 	/*
 	 * Mark this buffer XBF_DONE (i.e. uptodate) so that a subsequent
 	 * xfs_buf_read will not pointlessly reread the contents from the disk.
 	 */
-	(*bpp)->b_flags |= XBF_DONE;
+	bp->b_flags |= XBF_DONE;
 
-	xfs_buf_delwri_queue_here(*bpp, buffers_list);
-	xfs_buf_relse(*bpp);
+	xfs_buf_delwri_queue_here(bp, buffers_list);
+	xfs_buf_relse(bp);
 	*bpp = NULL;
+	bbl->nr_dirty++;
+
+	if (!bbl->max_dirty || bbl->nr_dirty < bbl->max_dirty)
+		return 0;
+
+	error = xfs_buf_delwri_submit(buffers_list);
+	if (error)
+		return error;
+
+	bbl->nr_dirty = 0;
+	return 0;
 }
 
 /*
@@ -422,7 +439,10 @@ xfs_btree_bload_prep_block(
 	 */
 	if (*blockp)
 		xfs_btree_set_sibling(cur, *blockp, &new_ptr, XFS_BB_RIGHTSIB);
-	xfs_btree_bload_drop_buf(buffers_list, bpp);
+
+	ret = xfs_btree_bload_drop_buf(bbl, buffers_list, bpp);
+	if (ret)
+		return ret;
 
 	/* Initialize the new btree block. */
 	xfs_btree_init_block_cur(cur, new_bp, level, nr_this_block);
@@ -770,6 +790,7 @@ xfs_btree_bload(
 	cur->bc_nlevels = bbl->btree_height;
 	xfs_btree_set_ptr_null(cur, &child_ptr);
 	xfs_btree_set_ptr_null(cur, &ptr);
+	bbl->nr_dirty = 0;
 
 	xfs_btree_bload_level_geometry(cur, bbl, level, nr_this_level,
 			&avg_per_block, &blocks, &blocks_with_extra);
@@ -808,7 +829,10 @@ xfs_btree_bload(
 			xfs_btree_copy_ptrs(cur, &child_ptr, &ptr, 1);
 	}
 	total_blocks += blocks;
-	xfs_btree_bload_drop_buf(&buffers_list, &bp);
+
+	ret = xfs_btree_bload_drop_buf(bbl, &buffers_list, &bp);
+	if (ret)
+		goto out;
 
 	/* Populate the internal btree nodes. */
 	for (level = 1; level < cur->bc_nlevels; level++) {
@@ -850,7 +874,11 @@ xfs_btree_bload(
 				xfs_btree_copy_ptrs(cur, &first_ptr, &ptr, 1);
 		}
 		total_blocks += blocks;
-		xfs_btree_bload_drop_buf(&buffers_list, &bp);
+
+		ret = xfs_btree_bload_drop_buf(bbl, &buffers_list, &bp);
+		if (ret)
+			goto out;
+
 		xfs_btree_copy_ptrs(cur, &child_ptr, &first_ptr, 1);
 	}
 
diff --git a/libxfs/xfs_btree_staging.h b/libxfs/xfs_btree_staging.h
index bd5b3f004..f0a500728 100644
--- a/libxfs/xfs_btree_staging.h
+++ b/libxfs/xfs_btree_staging.h
@@ -112,6 +112,16 @@ struct xfs_btree_bload {
 	 * height of the new btree.
 	 */
 	unsigned int			btree_height;
+
+	/*
+	 * Flush the new btree block buffer list to disk after this many blocks
+	 * have been formatted.  Zero prohibits writing any buffers until all
+	 * blocks have been formatted.
+	 */
+	uint16_t			max_dirty;
+
+	/* Number of dirty buffers. */
+	uint16_t			nr_dirty;
 };
 
 int xfs_btree_bload_compute_geometry(struct xfs_btree_cur *cur,
diff --git a/repair/agbtree.c b/repair/agbtree.c
index 10a0c7e48..981d8e340 100644
--- a/repair/agbtree.c
+++ b/repair/agbtree.c
@@ -23,6 +23,7 @@ init_rebuild(
 	memset(btr, 0, sizeof(struct bt_rebuild));
 
 	bulkload_init_ag(&btr->newbt, sc, oinfo);
+	btr->bload.max_dirty = XFS_B_TO_FSBT(sc->mp, 256U << 10); /* 256K */
 	bulkload_estimate_ag_slack(sc, &btr->bload, est_agfreeblocks);
 }
 
-- 
2.44.0


  parent reply	other threads:[~2024-04-22 16:39 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-22 16:25 [PATCH 00/67] libxfs: Sync to Linux 6.8 cem
2024-04-22 16:25 ` [PATCH 01/67] xfs: use xfs_defer_pending objects to recover intent items cem
2024-04-22 16:25 ` [PATCH 02/67] xfs: recreate work items when recovering " cem
2024-04-22 16:25 ` [PATCH 03/67] xfs: use xfs_defer_finish_one to finish recovered work items cem
2024-04-22 16:25 ` [PATCH 04/67] xfs: move ->iop_recover to xfs_defer_op_type cem
2024-04-22 16:25 ` [PATCH 05/67] xfs: hoist intent done flag setting to ->finish_item callsite cem
2024-04-22 16:25 ` [PATCH 06/67] xfs: hoist ->create_intent boilerplate to its callsite cem
2024-04-22 16:25 ` [PATCH 07/67] xfs: use xfs_defer_create_done for the relogging operation cem
2024-04-22 16:25 ` [PATCH 08/67] xfs: clean out XFS_LI_DIRTY setting boilerplate from ->iop_relog cem
2024-04-22 16:25 ` [PATCH 09/67] xfs: hoist xfs_trans_add_item calls to defer ops functions cem
2024-04-22 16:25 ` [PATCH 10/67] xfs: move ->iop_relog to struct xfs_defer_op_type cem
2024-04-22 16:25 ` [PATCH 11/67] xfs: make rextslog computation consistent with mkfs cem
2024-04-22 16:25 ` [PATCH 12/67] xfs: fix 32-bit truncation in xfs_compute_rextslog cem
2024-04-22 16:25 ` [PATCH 13/67] xfs: don't allow overly small or large realtime volumes cem
2024-04-22 16:25 ` [PATCH 14/67] xfs: elide ->create_done calls for unlogged deferred work cem
2024-04-22 16:25 ` [PATCH 15/67] xfs: don't append work items to logged xfs_defer_pending objects cem
2024-04-22 16:25 ` [PATCH 16/67] xfs: allow pausing of pending deferred work items cem
2024-04-22 16:25 ` [PATCH 17/67] xfs: remove __xfs_free_extent_later cem
2024-04-22 16:25 ` [PATCH 18/67] xfs: automatic freeing of freshly allocated unwritten space cem
2024-04-22 16:25 ` [PATCH 19/67] xfs: remove unused fields from struct xbtree_ifakeroot cem
2024-04-22 16:25 ` [PATCH 20/67] xfs: force small EFIs for reaping btree extents cem
2024-04-22 16:25 ` [PATCH 21/67] xfs: ensure logflagsp is initialized in xfs_bmap_del_extent_real cem
2024-04-22 16:25 ` [PATCH 22/67] xfs: update dir3 leaf block metadata after swap cem
2024-04-22 16:25 ` [PATCH 23/67] xfs: extract xfs_da_buf_copy() helper function cem
2024-04-22 16:25 ` [PATCH 24/67] xfs: move xfs_ondisk.h to libxfs/ cem
2024-04-22 16:25 ` [PATCH 25/67] xfs: consolidate the xfs_attr_defer_* helpers cem
2024-04-22 16:25 ` [PATCH 26/67] xfs: store an ops pointer in struct xfs_defer_pending cem
2024-04-22 16:25 ` [PATCH 27/67] xfs: pass the defer ops instead of type to xfs_defer_start_recovery cem
2024-04-22 16:25 ` [PATCH 28/67] xfs: pass the defer ops directly to xfs_defer_add cem
2024-04-22 16:25 ` [PATCH 29/67] xfs: force all buffers to be written during btree bulk load cem
2024-04-22 16:25 ` [PATCH 30/67] xfs: set XBF_DONE on newly formatted btree block that are ready for writing cem
2024-04-22 16:25 ` [PATCH 31/67] xfs: read leaf blocks when computing keys for bulkloading into node blocks cem
2024-04-22 16:25 ` [PATCH 32/67] xfs: move btree bulkload record initialization to ->get_record implementations cem
2024-04-22 16:25 ` cem [this message]
2024-04-22 16:25 ` [PATCH 34/67] xfs: repair free space btrees cem
2024-04-22 16:25 ` [PATCH 35/67] xfs: repair inode btrees cem
2024-04-22 16:25 ` [PATCH 36/67] xfs: repair refcount btrees cem
2024-04-22 16:25 ` [PATCH 37/67] xfs: dont cast to char * for XFS_DFORK_*PTR macros cem
2024-04-22 16:26 ` [PATCH 38/67] xfs: set inode sick state flags when we zap either ondisk fork cem
2024-04-22 16:26 ` [PATCH 39/67] xfs: zap broken inode forks cem
2024-04-22 16:26 ` [PATCH 40/67] xfs: repair inode fork block mapping data structures cem
2024-04-22 16:26 ` [PATCH 41/67] xfs: create a ranged query function for refcount btrees cem
2024-04-22 16:26 ` [PATCH 42/67] xfs: create a new inode fork block unmap helper cem
2024-04-22 16:26 ` [PATCH 43/67] xfs: improve dquot iteration for scrub cem
2024-04-22 16:26 ` [PATCH 44/67] xfs: add lock protection when remove perag from radix tree cem
2024-04-22 16:26 ` [PATCH 45/67] xfs: fix perag leak when growfs fails cem
2024-04-22 16:26 ` [PATCH 46/67] xfs: remove the xfs_alloc_arg argument to xfs_bmap_btalloc_accounting cem
2024-04-22 16:26 ` [PATCH 47/67] xfs: also use xfs_bmap_btalloc_accounting for RT allocations cem
2024-04-22 16:26 ` [PATCH 48/67] xfs: return -ENOSPC from xfs_rtallocate_* cem
2024-04-22 16:26 ` [PATCH 49/67] xfs: indicate if xfs_bmap_adjacent changed ap->blkno cem
2024-04-22 16:26 ` [PATCH 50/67] xfs: move xfs_rtget_summary to xfs_rtbitmap.c cem
2024-04-22 16:26 ` [PATCH 51/67] xfs: split xfs_rtmodify_summary_int cem
2024-04-22 16:26 ` [PATCH 52/67] xfs: remove rt-wrappers from xfs_format.h cem
2024-04-22 16:26 ` [PATCH 53/67] xfs: remove XFS_RTMIN/XFS_RTMAX cem
2024-04-22 16:26 ` [PATCH 54/67] xfs: make if_data a void pointer cem
2024-04-22 16:26 ` [PATCH 55/67] xfs: return if_data from xfs_idata_realloc cem
2024-04-22 16:26 ` [PATCH 56/67] xfs: move the xfs_attr_sf_lookup tracepoint cem
2024-04-22 16:26 ` [PATCH 57/67] xfs: simplify xfs_attr_sf_findname cem
2024-04-22 16:26 ` [PATCH 58/67] xfs: remove xfs_attr_shortform_lookup cem
2024-04-22 16:26 ` [PATCH 59/67] xfs: use xfs_attr_sf_findname in xfs_attr_shortform_getvalue cem
2024-04-22 16:26 ` [PATCH 60/67] xfs: remove struct xfs_attr_shortform cem
2024-04-22 16:26 ` [PATCH 61/67] xfs: remove xfs_attr_sf_hdr_t cem
2024-04-22 16:26 ` [PATCH 62/67] xfs: turn the XFS_DA_OP_REPLACE checks in xfs_attr_shortform_addname into asserts cem
2024-04-22 16:26 ` [PATCH 63/67] xfs: fix a use after free in xfs_defer_finish_recovery cem
2024-04-22 16:26 ` [PATCH 64/67] xfs: use the op name in trace_xlog_intent_recovery_failed cem
2024-04-22 16:26 ` [PATCH 65/67] xfs: fix backwards logic in xfs_bmap_alloc_account cem
2024-04-22 16:26 ` [PATCH 66/67] xfs: reset XFS_ATTR_INCOMPLETE filter on node removal cem
2024-04-22 16:26 ` [PATCH 67/67] xfs: remove conditional building of rt geometry validator functions cem
  -- strict thread matches above, loose matches on Subject: below --
2024-04-17 21:16 [PATCHSET 04/11] libxfs: sync with 6.8 Darrick J. Wong
2024-04-17 21:30 ` [PATCH 33/67] xfs: constrain dirty buffers while formatting a staged btree Darrick J. Wong
2024-03-26  2:55 [PATCHSET 02/18] libxfs: sync with 6.8 Darrick J. Wong
2024-03-26  3:11 ` [PATCH 33/67] xfs: constrain dirty buffers while formatting a staged btree Darrick J. Wong
2024-03-13  1:47 [PATCHSET 02/10] libxfs: sync with 6.8 Darrick J. Wong
2024-03-13  2:01 ` [PATCH 33/67] xfs: constrain dirty buffers while formatting a staged btree Darrick J. Wong

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=20240422163832.858420-35-cem@kernel.org \
    --to=cem@kernel.org \
    --cc=djwong@kernel.org \
    --cc=hch@lst.de \
    --cc=linux-xfs@vger.kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox