1 files changed, 115 insertions, 68 deletions

diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c
index 4f33c32affe3..393055fe3aef 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/xfs_alloc.c
@@ -430,6 +430,60 @@ xfs_alloc_fixup_trees(
         return 0;
 }
 
+static void
+xfs_agfl_verify(
+        struct xfs_buf  *bp)
+{
+#ifdef WHEN_CRCS_COME_ALONG
+        /*
+         * we cannot actually do any verification of the AGFL because mkfs does
+         * not initialise the AGFL to zero or NULL. Hence the only valid part of
+         * the AGFL is what the AGF says is active. We can't get to the AGF, so
+         * we can't verify just those entries are valid.
+         *
+         * This problem goes away when the CRC format change comes along as that
+         * requires the AGFL to be initialised by mkfs. At that point, we can
+         * verify the blocks in the agfl -active or not- lie within the bounds
+         * of the AG. Until then, just leave this check ifdef'd out.
+         */
+        struct xfs_mount *mp = bp->b_target->bt_mount;
+        struct xfs_agfl *agfl = XFS_BUF_TO_AGFL(bp);
+        int             agfl_ok = 1;
+
+        int             i;
+
+        for (i = 0; i < XFS_AGFL_SIZE(mp); i++) {
+                if (be32_to_cpu(agfl->agfl_bno[i]) == NULLAGBLOCK ||
+                    be32_to_cpu(agfl->agfl_bno[i]) >= mp->m_sb.sb_agblocks)
+                        agfl_ok = 0;
+        }
+
+        if (!agfl_ok) {
+                XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, agfl);
+                xfs_buf_ioerror(bp, EFSCORRUPTED);
+        }
+#endif
+}
+
+static void
+xfs_agfl_write_verify(
+        struct xfs_buf  *bp)
+{
+        xfs_agfl_verify(bp);
+}
+
+static void
+xfs_agfl_read_verify(
+        struct xfs_buf  *bp)
+{
+        xfs_agfl_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_agfl_buf_ops = {
+        .verify_read = xfs_agfl_read_verify,
+        .verify_write = xfs_agfl_write_verify,
+};
+
 /*
  * Read in the allocation group free block array.
  */
@@ -447,7 +501,7 @@ xfs_alloc_read_agfl(
         error = xfs_trans_read_buf(
                         mp, tp, mp->m_ddev_targp,
                         XFS_AG_DADDR(mp, agno, XFS_AGFL_DADDR(mp)),
-                        XFS_FSS_TO_BB(mp, 1), 0, &bp);
+                        XFS_FSS_TO_BB(mp, 1), 0, &bp, &xfs_agfl_buf_ops);
         if (error)
                 return error;
         ASSERT(!xfs_buf_geterror(bp));
@@ -1866,6 +1920,7 @@ xfs_alloc_fix_freelist(
         /*
          * Initialize the args structure.
          */
+        memset(&targs, 0, sizeof(targs));
         targs.tp = tp;
         targs.mp = mp;
         targs.agbp = agbp;
@@ -2090,6 +2145,63 @@ xfs_alloc_put_freelist(
         return 0;
 }
 
+static void
+xfs_agf_verify(
+        struct xfs_buf  *bp)
+ {
+        struct xfs_mount *mp = bp->b_target->bt_mount;
+        struct xfs_agf  *agf;
+        int             agf_ok;
+
+        agf = XFS_BUF_TO_AGF(bp);
+
+        agf_ok = agf->agf_magicnum == cpu_to_be32(XFS_AGF_MAGIC) &&
+                XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) &&
+                be32_to_cpu(agf->agf_freeblks) <= be32_to_cpu(agf->agf_length) &&
+                be32_to_cpu(agf->agf_flfirst) < XFS_AGFL_SIZE(mp) &&
+                be32_to_cpu(agf->agf_fllast) < XFS_AGFL_SIZE(mp) &&
+                be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp);
+
+        /*
+         * during growfs operations, the perag is not fully initialised,
+         * so we can't use it for any useful checking. growfs ensures we can't
+         * use it by using uncached buffers that don't have the perag attached
+         * so we can detect and avoid this problem.
+         */
+        if (bp->b_pag)
+                agf_ok = agf_ok && be32_to_cpu(agf->agf_seqno) ==
+                                                bp->b_pag->pag_agno;
+
+        if (xfs_sb_version_haslazysbcount(&mp->m_sb))
+                agf_ok = agf_ok && be32_to_cpu(agf->agf_btreeblks) <=
+                                                be32_to_cpu(agf->agf_length);
+
+        if (unlikely(XFS_TEST_ERROR(!agf_ok, mp, XFS_ERRTAG_ALLOC_READ_AGF,
+                        XFS_RANDOM_ALLOC_READ_AGF))) {
+                XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp, agf);
+                xfs_buf_ioerror(bp, EFSCORRUPTED);
+        }
+}
+
+static void
+xfs_agf_read_verify(
+        struct xfs_buf  *bp)
+{
+        xfs_agf_verify(bp);
+}
+
+static void
+xfs_agf_write_verify(
+        struct xfs_buf  *bp)
+{
+        xfs_agf_verify(bp);
+}
+
+const struct xfs_buf_ops xfs_agf_buf_ops = {
+        .verify_read = xfs_agf_read_verify,
+        .verify_write = xfs_agf_write_verify,
+};
+
 /*
  * Read in the allocation group header (free/alloc section).
  */
@@ -2101,44 +2213,19 @@ xfs_read_agf(
         int                     flags,  /* XFS_BUF_ */
         struct xfs_buf          **bpp)  /* buffer for the ag freelist header */
 {
-        struct xfs_agf  *agf;           /* ag freelist header */
-        int             agf_ok;         /* set if agf is consistent */
         int             error;
 
         ASSERT(agno != NULLAGNUMBER);
         error = xfs_trans_read_buf(
                         mp, tp, mp->m_ddev_targp,
                         XFS_AG_DADDR(mp, agno, XFS_AGF_DADDR(mp)),
-                        XFS_FSS_TO_BB(mp, 1), flags, bpp);
+                        XFS_FSS_TO_BB(mp, 1), flags, bpp, &xfs_agf_buf_ops);
         if (error)
                 return error;
         if (!*bpp)
                 return 0;
 
         ASSERT(!(*bpp)->b_error);
-        agf = XFS_BUF_TO_AGF(*bpp);
-
-        /*
-         * Validate the magic number of the agf block.
-         */
-        agf_ok =
-                agf->agf_magicnum == cpu_to_be32(XFS_AGF_MAGIC) &&
-                XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) &&
-                be32_to_cpu(agf->agf_freeblks) <= be32_to_cpu(agf->agf_length) &&
-                be32_to_cpu(agf->agf_flfirst) < XFS_AGFL_SIZE(mp) &&
-                be32_to_cpu(agf->agf_fllast) < XFS_AGFL_SIZE(mp) &&
-                be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp) &&
-                be32_to_cpu(agf->agf_seqno) == agno;
-        if (xfs_sb_version_haslazysbcount(&mp->m_sb))
-                agf_ok = agf_ok && be32_to_cpu(agf->agf_btreeblks) <=
-                                                be32_to_cpu(agf->agf_length);
-        if (unlikely(XFS_TEST_ERROR(!agf_ok, mp, XFS_ERRTAG_ALLOC_READ_AGF,
-                        XFS_RANDOM_ALLOC_READ_AGF))) {
-                XFS_CORRUPTION_ERROR("xfs_alloc_read_agf",
-                                     XFS_ERRLEVEL_LOW, mp, agf);
-                xfs_trans_brelse(tp, *bpp);
-                return XFS_ERROR(EFSCORRUPTED);
-        }
         xfs_buf_set_ref(*bpp, XFS_AGF_REF);
         return 0;
 }
@@ -2207,7 +2294,7 @@ xfs_alloc_read_agf(
  * group or loop over the allocation groups to find the result.
  */
 int                             /* error */
-__xfs_alloc_vextent(
+xfs_alloc_vextent(
         xfs_alloc_arg_t *args)  /* allocation argument structure */
 {
         xfs_agblock_t   agsize; /* allocation group size */
@@ -2417,46 +2504,6 @@ error0:
         return error;
 }
 
-static void
-xfs_alloc_vextent_worker(
-        struct work_struct      *work)
-{
-        struct xfs_alloc_arg    *args = container_of(work,
-                                                struct xfs_alloc_arg, work);
-        unsigned long           pflags;
-
-        /* we are in a transaction context here */
-        current_set_flags_nested(&pflags, PF_FSTRANS);
-
-        args->result = __xfs_alloc_vextent(args);
-        complete(args->done);
-
-        current_restore_flags_nested(&pflags, PF_FSTRANS);
-}
-
-/*
- * Data allocation requests often come in with little stack to work on. Push
- * them off to a worker thread so there is lots of stack to use. Metadata
- * requests, OTOH, are generally from low stack usage paths, so avoid the
- * context switch overhead here.
- */
-int
-xfs_alloc_vextent(
-        struct xfs_alloc_arg    *args)
-{
-        DECLARE_COMPLETION_ONSTACK(done);
-
-        if (!args->userdata)
-                return __xfs_alloc_vextent(args);
-
-
-        args->done = &done;
-        INIT_WORK_ONSTACK(&args->work, xfs_alloc_vextent_worker);
-        queue_work(xfs_alloc_wq, &args->work);
-        wait_for_completion(&done);
-        return args->result;
-}
-
 /*
  * Free an extent.
  * Just break up the extent address and hand off to xfs_free_ag_extent