path: root/fs/xfs/scrub/repair.c

// SPDX-License-Identifier: GPL-2.0+
/*
 * Copyright (C) 2018 Oracle.  All Rights Reserved.
 * Author: Darrick J. Wong <darrick.wong@oracle.com>
 */
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_shared.h"
#include "xfs_format.h"
#include "xfs_trans_resv.h"
#include "xfs_mount.h"
#include "xfs_btree.h"
#include "xfs_log_format.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
#include "xfs_inode.h"
#include "xfs_alloc.h"
#include "xfs_alloc_btree.h"
#include "xfs_ialloc.h"
#include "xfs_ialloc_btree.h"
#include "xfs_rmap.h"
#include "xfs_rmap_btree.h"
#include "xfs_refcount_btree.h"
#include "xfs_extent_busy.h"
#include "xfs_ag_resv.h"
#include "xfs_quota.h"
#include "scrub/scrub.h"
#include "scrub/common.h"
#include "scrub/trace.h"
#include "scrub/repair.h"
#include "scrub/bitmap.h"

/*
 * Attempt to repair some metadata, if the metadata is corrupt and userspace
 * told us to fix it.  This function returns -EAGAIN to mean "re-run scrub",
 * and will set *fixed to true if it thinks it repaired anything.
 */
int
xrep_attempt(
	struct xfs_inode	*ip,
	struct xfs_scrub	*sc)
{
	int			error = 0;

	trace_xrep_attempt(ip, sc->sm, error);

	xchk_ag_btcur_free(&sc->sa);

	/* Repair whatever's broken. */
	ASSERT(sc->ops->repair);
	error = sc->ops->repair(sc);
	trace_xrep_done(ip, sc->sm, error);
	switch (error) {
	case 0:
		/*
		 * Repair succeeded.  Commit the fixes and perform a second
		 * scrub so that we can tell userspace if we fixed the problem.
		 */
		sc->sm->sm_flags &= ~XFS_SCRUB_FLAGS_OUT;
		sc->flags |= XREP_ALREADY_FIXED;
		return -EAGAIN;
	case -EDEADLOCK:
	case -EAGAIN:
		/* Tell the caller to try again having grabbed all the locks. */
		if (!(sc->flags & XCHK_TRY_HARDER)) {
			sc->flags |= XCHK_TRY_HARDER;
			return -EAGAIN;
		}
		/*
		 * We tried harder but still couldn't grab all the resources
		 * we needed to fix it.  The corruption has not been fixed,
		 * so report back to userspace.
		 */
		return -EFSCORRUPTED;
	default:
		return error;
	}
}

/*
 * Complain about unfixable problems in the filesystem.  We don't log
 * corruptions when IFLAG_REPAIR wasn't set on the assumption that the driver
 * program is xfs_scrub, which will call back with IFLAG_REPAIR set if the
 * administrator isn't running xfs_scrub in no-repairs mode.
 *
 * Use this helper function because _ratelimited silently declares a static
 * structure to track rate limiting information.
 */
void
xrep_failure(
	struct xfs_mount	*mp)
{
	xfs_alert_ratelimited(mp,
"Corruption not fixed during online repair.  Unmount and run xfs_repair.");
}

/*
 * Repair probe -- userspace uses this to probe if we're willing to repair a
 * given mountpoint.
 */
int
xrep_probe(
	struct xfs_scrub	*sc)
{
	int			error = 0;

	if (xchk_should_terminate(sc, &error))
		return error;

	return 0;
}

/*
 * Roll a transaction, keeping the AG headers locked and reinitializing
 * the btree cursors.
 */
int
xrep_roll_ag_trans(
	struct xfs_scrub	*sc)
{
	int			error;

	/* Keep the AG header buffers locked so we can keep going. */
	if (sc->sa.agi_bp)
		xfs_trans_bhold(sc->tp, sc->sa.agi_bp);
	if (sc->sa.agf_bp)
		xfs_trans_bhold(sc->tp, sc->sa.agf_bp);
	if (sc->sa.agfl_bp)
		xfs_trans_bhold(sc->tp, sc->sa.agfl_bp);

	/*
	 * Roll the transaction.  We still own the buffer and the buffer lock
	 * regardless of whether or not the roll succeeds.  If the roll fails,
	 * the buffers will be released during teardown on our way out of the
	 * kernel.  If it succeeds, we join them to the new transaction and
	 * move on.
	 */
	error = xfs_trans_roll(&sc->tp);
	if (error)
		return error;

	/* Join AG headers to the new transaction. */
	if (sc->sa.agi_bp)
		xfs_trans_bjoin(sc->tp, sc->sa.agi_bp);
	if (sc->sa.agf_bp)
		xfs_trans_bjoin(sc->tp, sc->sa.agf_bp);
	if (sc->sa.agfl_bp)
		xfs_trans_bjoin(sc->tp, sc->sa.agfl_bp);

	return 0;
}

/*
 * Does the given AG have enough space to rebuild a btree?  Neither AG
 * reservation can be critical, and we must have enough space (factoring
 * in AG reservations) to construct a whole btree.
 */
bool
xrep_ag_has_space(
	struct xfs_perag	*pag,
	xfs_extlen_t		nr_blocks,
	enum xfs_ag_resv_type	type)
{
	return  !xfs_ag_resv_critical(pag, XFS_AG_RESV_RMAPBT) &&
		!xfs_ag_resv_critical(pag, XFS_AG_RESV_METADATA) &&
		pag->pagf_freeblks > xfs_ag_resv_needed(pag, type) + nr_blocks;
}

/*
 * Figure out how many blocks to reserve for an AG repair.  We calculate the
 * worst case estimate for the number of blocks we'd need to rebuild one of
 * any type of per-AG btree.
 */
xfs_extlen_t
xrep_calc_ag_resblks(
	struct xfs_scrub		*sc)
{
	struct xfs_mount		*mp = sc->mp;
	struct xfs_scrub_metadata	*sm = sc->sm;
	struct xfs_perag		*pag;
	struct xfs_buf			*bp;
	xfs_agino_t			icount = NULLAGINO;
	xfs_extlen_t			aglen = NULLAGBLOCK;
	xfs_extlen_t			usedlen;
	xfs_extlen_t			freelen;
	xfs_extlen_t			bnobt_sz;
	xfs_extlen_t			inobt_sz;
	xfs_extlen_t			rmapbt_sz;
	xfs_extlen_t			refcbt_sz;
	int				error;

	if (!(sm->sm_flags & XFS_SCRUB_IFLAG_REPAIR))
		return 0;

	pag = xfs_perag_get(mp, sm->sm_agno);
	if (pag->pagi_init) {
		/* Use in-core icount if possible. */
		icount = pag->pagi_count;