@@ -211,6 +211,7 @@
#define trace_xfs_trans_cancel(a,b) ((void) 0)
#define trace_xfs_trans_brelse(a) ((void) 0)
#define trace_xfs_trans_binval(a) ((void) 0)
+#define trace_xfs_trans_bdetach(a) ((void) 0)
#define trace_xfs_trans_bjoin(a) ((void) 0)
#define trace_xfs_trans_bhold(a) ((void) 0)
#define trace_xfs_trans_bhold_release(a) ((void) 0)
@@ -114,6 +114,7 @@ int libxfs_trans_roll_inode (struct xfs_trans **, struct xfs_inode *);
void libxfs_trans_brelse(struct xfs_trans *, struct xfs_buf *);
void libxfs_trans_binval(struct xfs_trans *, struct xfs_buf *);
void libxfs_trans_bjoin(struct xfs_trans *, struct xfs_buf *);
+void libxfs_trans_bdetach(struct xfs_trans *tp, struct xfs_buf *bp);
void libxfs_trans_bhold(struct xfs_trans *, struct xfs_buf *);
void libxfs_trans_bhold_release(struct xfs_trans *, struct xfs_buf *);
void libxfs_trans_dirty_buf(struct xfs_trans *, struct xfs_buf *);
@@ -246,3 +246,68 @@ xmbuf_verify_daddr(
return daddr < (xf->maxbytes >> BBSHIFT);
}
+
+/* Discard the page backing this buffer. */
+static void
+xmbuf_stale(
+ struct xfs_buf *bp)
+{
+ struct xfile *xf = bp->b_target->bt_xfile;
+ loff_t pos;
+
+ ASSERT(xfs_buftarg_is_mem(bp->b_target));
+
+ pos = BBTOB(xfs_buf_daddr(bp)) + xf->partition_pos;
+ fallocate(xf->fcb->fd, FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE, pos,
+ BBTOB(bp->b_length));
+}
+
+/*
+ * Finalize a buffer -- discard the backing page if it's stale, or run the
+ * write verifier to detect problems.
+ */
+int
+xmbuf_finalize(
+ struct xfs_buf *bp)
+{
+ xfs_failaddr_t fa;
+ int error = 0;
+
+ if (bp->b_flags & LIBXFS_B_STALE) {
+ xmbuf_stale(bp);
+ return 0;
+ }
+
+ /*
+ * Although this btree is ephemeral, validate the buffer structure so
+ * that we can detect memory corruption errors and software bugs.
+ */
+ fa = bp->b_ops->verify_struct(bp);
+ if (fa) {
+ error = -EFSCORRUPTED;
+ xfs_verifier_error(bp, error, fa);
+ }
+
+ return error;
+}
+
+/*
+ * Detach this xmbuf buffer from the transaction by any means necessary.
+ * All buffers are direct-mapped, so they do not need bwrite.
+ */
+void
+xmbuf_trans_bdetach(
+ struct xfs_trans *tp,
+ struct xfs_buf *bp)
+{
+ struct xfs_buf_log_item *bli = bp->b_log_item;
+
+ ASSERT(bli != NULL);
+
+ bli->bli_flags &= ~(XFS_BLI_DIRTY | XFS_BLI_ORDERED |
+ XFS_BLI_STALE);
+ clear_bit(XFS_LI_DIRTY, &bli->bli_item.li_flags);
+
+ while (bp->b_log_item != NULL)
+ xfs_trans_bdetach(tp, bp);
+}
@@ -24,5 +24,7 @@ int xmbuf_map_page(struct xfs_buf *bp);
void xmbuf_unmap_page(struct xfs_buf *bp);
bool xmbuf_verify_daddr(struct xfs_buftarg *btp, xfs_daddr_t daddr);
+void xmbuf_trans_bdetach(struct xfs_trans *tp, struct xfs_buf *bp);
+int xmbuf_finalize(struct xfs_buf *bp);
#endif /* __XFS_BUF_MEM_H__ */
@@ -224,6 +224,7 @@
#define xfs_trans_alloc_empty libxfs_trans_alloc_empty
#define xfs_trans_alloc libxfs_trans_alloc
#define xfs_trans_alloc_inode libxfs_trans_alloc_inode
+#define xfs_trans_bdetach libxfs_trans_bdetach
#define xfs_trans_bhold libxfs_trans_bhold
#define xfs_trans_bhold_release libxfs_trans_bhold_release
#define xfs_trans_binval libxfs_trans_binval
@@ -614,6 +614,46 @@ libxfs_trans_brelse(
libxfs_buf_relse(bp);
}
+/*
+ * Forcibly detach a buffer previously joined to the transaction. The caller
+ * will retain its locked reference to the buffer after this function returns.
+ * The buffer must be completely clean and must not be held to the transaction.
+ */
+void
+libxfs_trans_bdetach(
+ struct xfs_trans *tp,
+ struct xfs_buf *bp)
+{
+ struct xfs_buf_log_item *bip = bp->b_log_item;
+
+ ASSERT(tp != NULL);
+ ASSERT(bp->b_transp == tp);
+ ASSERT(bip->bli_item.li_type == XFS_LI_BUF);
+
+ trace_xfs_trans_bdetach(bip);
+
+ /*
+ * Erase all recursion count, since we're removing this buffer from the
+ * transaction.
+ */
+ bip->bli_recur = 0;
+
+ /*
+ * The buffer must be completely clean. Specifically, it had better
+ * not be dirty, stale, logged, ordered, or held to the transaction.
+ */
+ ASSERT(!test_bit(XFS_LI_DIRTY, &bip->bli_item.li_flags));
+ ASSERT(!(bip->bli_flags & XFS_BLI_DIRTY));
+ ASSERT(!(bip->bli_flags & XFS_BLI_HOLD));
+ ASSERT(!(bip->bli_flags & XFS_BLI_ORDERED));
+ ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
+
+ /* Unlink the log item from the transaction and drop the log item. */
+ xfs_trans_del_item(&bip->bli_item);
+ xfs_buf_item_put(bip);
+ bp->b_transp = NULL;
+}
+
/*
* Mark the buffer as not needing to be unlocked when the buf item's
* iop_unlock() routine is called. The buffer must already be locked
@@ -225,3 +225,122 @@ xfbtree_get_maxrecs(
return xfbt->maxrecs[level != 0];
}
+
+/* If this log item is a buffer item that came from the xfbtree, return it. */
+static inline struct xfs_buf *
+xfbtree_buf_match(
+ struct xfbtree *xfbt,
+ const struct xfs_log_item *lip)
+{
+ const struct xfs_buf_log_item *bli;
+ struct xfs_buf *bp;
+
+ if (lip->li_type != XFS_LI_BUF)
+ return NULL;
+
+ bli = container_of(lip, struct xfs_buf_log_item, bli_item);
+ bp = bli->bli_buf;
+ if (bp->b_target != xfbt->target)
+ return NULL;
+
+ return bp;
+}
+
+/*
+ * Commit changes to the incore btree immediately by writing all dirty xfbtree
+ * buffers to the backing xfile. This detaches all xfbtree buffers from the
+ * transaction, even on failure. The buffer locks are dropped between the
+ * delwri queue and submit, so the caller must synchronize btree access.
+ *
+ * Normally we'd let the buffers commit with the transaction and get written to
+ * the xfile via the log, but online repair stages ephemeral btrees in memory
+ * and uses the btree_staging functions to write new btrees to disk atomically.
+ * The in-memory btree (and its backing store) are discarded at the end of the
+ * repair phase, which means that xfbtree buffers cannot commit with the rest
+ * of a transaction.
+ *
+ * In other words, online repair only needs the transaction to collect buffer
+ * pointers and to avoid buffer deadlocks, not to guarantee consistency of
+ * updates.
+ */
+int
+xfbtree_trans_commit(
+ struct xfbtree *xfbt,
+ struct xfs_trans *tp)
+{
+ struct xfs_log_item *lip, *n;
+ bool tp_dirty = false;
+ int error = 0;
+
+ /*
+ * For each xfbtree buffer attached to the transaction, write the dirty
+ * buffers to the xfile and release them.
+ */
+ list_for_each_entry_safe(lip, n, &tp->t_items, li_trans) {
+ struct xfs_buf *bp = xfbtree_buf_match(xfbt, lip);
+
+ if (!bp) {
+ if (test_bit(XFS_LI_DIRTY, &lip->li_flags))
+ tp_dirty |= true;
+ continue;
+ }
+
+ trace_xfbtree_trans_commit_buf(xfbt, bp);
+
+ xmbuf_trans_bdetach(tp, bp);
+
+ /*
+ * If the buffer fails verification, note the failure but
+ * continue walking the transaction items so that we remove all
+ * ephemeral btree buffers.
+ */
+ if (!error)
+ error = xmbuf_finalize(bp);
+
+ xfs_buf_relse(bp);
+ }
+
+ /*
+ * Reset the transaction's dirty flag to reflect the dirty state of the
+ * log items that are still attached.
+ */
+ tp->t_flags = (tp->t_flags & ~XFS_TRANS_DIRTY) |
+ (tp_dirty ? XFS_TRANS_DIRTY : 0);
+
+ return error;
+}
+
+/*
+ * Cancel changes to the incore btree by detaching all the xfbtree buffers.
+ * Changes are not undone, so callers must not access the btree ever again.
+ */
+void
+xfbtree_trans_cancel(
+ struct xfbtree *xfbt,
+ struct xfs_trans *tp)
+{
+ struct xfs_log_item *lip, *n;
+ bool tp_dirty = false;
+
+ list_for_each_entry_safe(lip, n, &tp->t_items, li_trans) {
+ struct xfs_buf *bp = xfbtree_buf_match(xfbt, lip);
+
+ if (!bp) {
+ if (test_bit(XFS_LI_DIRTY, &lip->li_flags))
+ tp_dirty |= true;
+ continue;
+ }
+
+ trace_xfbtree_trans_cancel_buf(xfbt, bp);
+
+ xmbuf_trans_bdetach(tp, bp);
+ xfs_buf_relse(bp);
+ }
+
+ /*
+ * Reset the transaction's dirty flag to reflect the dirty state of the
+ * log items that are still attached.
+ */
+ tp->t_flags = (tp->t_flags & ~XFS_TRANS_DIRTY) |
+ (tp_dirty ? XFS_TRANS_DIRTY : 0);
+}
@@ -65,6 +65,9 @@ int xfbtree_free_block(struct xfs_btree_cur *cur, struct xfs_buf *bp);
int xfbtree_init(struct xfs_mount *mp, struct xfbtree *xfbt,
struct xfs_buftarg *btp, const struct xfs_btree_ops *ops);
void xfbtree_destroy(struct xfbtree *xfbt);
+
+int xfbtree_trans_commit(struct xfbtree *xfbt, struct xfs_trans *tp);
+void xfbtree_trans_cancel(struct xfbtree *xfbt, struct xfs_trans *tp);
#else
# define xfbtree_verify_bno(...) (false)
#endif /* CONFIG_XFS_BTREE_IN_MEM */