[PATCH 48/51] xfs_repair: check the refcount btree against our observed reference counts when -n
Darrick J. Wong
darrick.wong at oracle.com
Wed Oct 7 00:10:26 CDT 2015
Check the observed reference counts against whatever's in the refcount
btree for discrepancies.
Signed-off-by: Darrick J. Wong <darrick.wong at oracle.com>
---
repair/phase4.c | 20 ++++++++
repair/rmap.c | 135 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
repair/rmap.h | 4 ++
repair/scan.c | 2 +
4 files changed, 161 insertions(+)
diff --git a/repair/phase4.c b/repair/phase4.c
index caa4221..dd03ca4 100644
--- a/repair/phase4.c
+++ b/repair/phase4.c
@@ -213,6 +213,21 @@ _("%s while fixing inode reflink flags.\n"),
}
static void
+check_refcount_btrees(
+ work_queue_t *wq,
+ xfs_agnumber_t agno,
+ void *arg)
+{
+ int error;
+
+ error = check_refcounts(wq->mp, agno);
+ if (error)
+ do_error(
+_("%s while checking reference counts"),
+ strerror(-error));
+}
+
+static void
process_rmap_data(
struct xfs_mount *mp)
{
@@ -239,6 +254,11 @@ process_rmap_data(
for (i = 0; i < mp->m_sb.sb_agcount; i++)
queue_work(&wq, process_inode_reflink_flags, i, NULL);
destroy_work_queue(&wq);
+
+ create_work_queue(&wq, mp, libxfs_nproc());
+ for (i = 0; i < mp->m_sb.sb_agcount; i++)
+ queue_work(&wq, check_refcount_btrees, i, NULL);
+ destroy_work_queue(&wq);
}
void
diff --git a/repair/rmap.c b/repair/rmap.c
index b582b51..9103a10 100644
--- a/repair/rmap.c
+++ b/repair/rmap.c
@@ -45,6 +45,7 @@ struct xfs_ag_rmap {
static struct xfs_ag_rmap *ag_rmaps;
static bool rmapbt_suspect;
+static bool refcbt_suspect;
/*
* Compare rmap observations for array sorting.
@@ -1132,6 +1133,140 @@ _("Unable to fix reflink flag on inode %"PRIu64".\n"),
}
/**
+ * refcount_record_count() -- Return the number of refcount objects for an AG.
+ *
+ * @mp: XFS mount object
+ * @agno: AG number
+ */
+size_t
+refcount_record_count(
+ struct xfs_mount *mp,
+ xfs_agnumber_t agno)
+{
+ return slab_count(ag_rmaps[agno].ar_refcount_items);
+}
+
+/**
+ * init_refcount_cursor() -- Return a slab cursor that will return refcount
+ * objects in order.
+ * @agno: AG number.
+ * @cur: The new cursor.
+ */
+int
+init_refcount_cursor(
+ xfs_agnumber_t agno,
+ struct xfs_slab_cursor **cur)
+{
+ return init_slab_cursor(ag_rmaps[agno].ar_refcount_items, NULL, cur);
+}
+
+/**
+ * refcount_avoid_check() -- Disable the refcount btree check.
+ */
+void
+refcount_avoid_check(void)
+{
+ refcbt_suspect = true;
+}
+
+/**
+ * check_refcounts() -- Compare the observed reference counts against
+ * what's in the ag btree.
+ * @mp: XFS mount object
+ * @agno: AG number
+ */
+int
+check_refcounts(
+ struct xfs_mount *mp,
+ xfs_agnumber_t agno)
+{
+ struct xfs_slab_cursor *rl_cur;
+ struct xfs_btree_cur *bt_cur = NULL;
+ int error;
+ int have;
+ int i;
+ struct xfs_buf *agbp = NULL;
+ struct xfs_refcount_irec *rl_rec;
+ struct xfs_refcount_irec tmp;
+ struct xfs_perag *pag; /* per allocation group data */
+
+ if (!xfs_sb_version_hasreflink(&mp->m_sb))
+ return 0;
+ if (refcbt_suspect) {
+ if (no_modify && agno == 0)
+ do_warn(_("would rebuild corrupt refcount btrees.\n"));
+ return 0;
+ }
+
+ /* Create cursors to refcount structures */
+ error = init_refcount_cursor(agno, &rl_cur);
+ if (error)
+ return error;
+
+ error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp);
+ if (error)
+ goto err;
+
+ /* Leave the per-ag data "uninitialized" since we rewrite it later */
+ pag = xfs_perag_get(mp, agno);
+ pag->pagf_init = 0;
+ xfs_perag_put(pag);
+
+ bt_cur = xfs_refcountbt_init_cursor(mp, NULL, agbp, agno, NULL);
+ if (!bt_cur) {
+ error = -ENOMEM;
+ goto err;
+ }
+
+ rl_rec = pop_slab_cursor(rl_cur);
+ while (rl_rec) {
+ /* Look for a refcount record in the btree */
+ error = xfs_refcountbt_lookup_le(bt_cur,
+ rl_rec->rc_startblock, &have);
+ if (error)
+ goto err;
+ if (!have) {
+ do_warn(
+_("Missing reference count record for (%u/%u) len %u count %u\n"),
+ agno, rl_rec->rc_startblock,
+ rl_rec->rc_blockcount, rl_rec->rc_refcount);
+ goto next_loop;
+ }
+
+ error = xfs_refcountbt_get_rec(bt_cur, &tmp, &i);
+ if (error)
+ goto err;
+ if (!i) {
+ do_warn(
+_("Missing reference count record for (%u/%u) len %u count %u\n"),
+ agno, rl_rec->rc_startblock,
+ rl_rec->rc_blockcount, rl_rec->rc_refcount);
+ goto next_loop;
+ }
+
+ /* Compare each refcount observation against the btree's */
+ if (tmp.rc_startblock != rl_rec->rc_startblock ||
+ tmp.rc_blockcount < rl_rec->rc_blockcount ||
+ tmp.rc_refcount < rl_rec->rc_refcount)
+ do_warn(
+_("Incorrect reference count: saw (%u/%u) len %u nlinks %u; should be (%u/%u) len %u nlinks %u\n"),
+ agno, tmp.rc_startblock, tmp.rc_blockcount,
+ tmp.rc_refcount, agno, rl_rec->rc_startblock,
+ rl_rec->rc_blockcount, rl_rec->rc_refcount);
+next_loop:
+ rl_rec = pop_slab_cursor(rl_cur);
+ }
+
+err:
+ if (bt_cur)
+ xfs_btree_del_cursor(bt_cur, XFS_BTREE_NOERROR);
+ if (agbp)
+ libxfs_putbuf(agbp);
+ free_slab_cursor(&rl_cur);
+ return 0;
+}
+
+/**
* fix_freelist() - Regenerate the AGFL, so that we don't run out of it while
* rebuilding the rmapbt.
* @mp: XFS mount object
diff --git a/repair/rmap.h b/repair/rmap.h
index d0bcde1..df7d489 100644
--- a/repair/rmap.h
+++ b/repair/rmap.h
@@ -40,6 +40,10 @@ extern void rmap_avoid_check(void);
extern int check_rmaps(struct xfs_mount *, xfs_agnumber_t);
extern int compute_refcounts(struct xfs_mount *, xfs_agnumber_t);
+extern size_t refcount_record_count(struct xfs_mount *, xfs_agnumber_t);
+extern int init_refcount_cursor(xfs_agnumber_t, struct xfs_slab_cursor **);
+extern void refcount_avoid_check(void);
+extern int check_refcounts(struct xfs_mount *, xfs_agnumber_t);
extern void record_inode_reflink_flag(struct xfs_mount *, struct xfs_dinode *,
xfs_agnumber_t, xfs_agino_t, xfs_ino_t);
diff --git a/repair/scan.c b/repair/scan.c
index 54b9b68..3e8633c 100644
--- a/repair/scan.c
+++ b/repair/scan.c
@@ -1257,6 +1257,8 @@ _("%s btree block claimed (state %d), agno %d, bno %d, suspect %d\n"),
}
}
out:
+ if (suspect)
+ refcount_avoid_check();
return;
}
More information about the xfs
mailing list