From: Dave Chinner <dchinner@xxxxxxxxxx>
Create v5 filesystems with rmapbt turned on. Document the rmapbt
options to mkfs, and initialize the extra field we added for reflink
support.
v2: Turn on the rmapbt feature when calculating the minimum log size.
Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
[darrick.wong@xxxxxxxxxx: split patch, add commit message and extra fields]
Signed-off-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx>
---
include/xfs_multidisk.h | 2 -
man/man8/mkfs.xfs.8 | 20 +++++++
mkfs/maxtrres.c | 5 +-
mkfs/xfs_mkfs.c | 138 +++++++++++++++++++++++++++++++++++++++++------
4 files changed, 145 insertions(+), 20 deletions(-)
diff --git a/include/xfs_multidisk.h b/include/xfs_multidisk.h
index 4429dab..8dc3027 100644
--- a/include/xfs_multidisk.h
+++ b/include/xfs_multidisk.h
@@ -68,6 +68,6 @@ extern void res_failed (int err);
/* maxtrres.c */
extern int max_trans_res(unsigned long agsize, int crcs_enabled, int
dirversion,
int sectorlog, int blocklog, int inodelog, int dirblocklog,
- int logversion, int log_sunit, int finobt);
+ int logversion, int log_sunit, int finobt, int rmapbt);
#endif /* __XFS_MULTIDISK_H__ */
diff --git a/man/man8/mkfs.xfs.8 b/man/man8/mkfs.xfs.8
index 980b0e1..d88d314 100644
--- a/man/man8/mkfs.xfs.8
+++ b/man/man8/mkfs.xfs.8
@@ -193,6 +193,26 @@ is used, the free inode btree feature is not supported and
is disabled.
.BI uuid= value
Use the given value as the filesystem UUID for the newly created filesystem.
The default is to generate a random UUID.
+.TP
+.BI rmapbt= value
+This option enables the creation of a reverse-mapping btree index in each
+allocation group. The value is either 0 to disable the feature, or 1 to
+create the btree.
+.IP
+The reverse mapping btree maps filesystem blocks to the owner of the
+filesystem block. Most of the mappings will be to an inode number and an
+offset, though there will also be mappings to filesystem metadata. This
+secondary metadata can be used to validate the primary metadata or to
+pinpoint exactly which data has been lost when a disk error occurs.
+.IP
+By default,
+.B mkfs.xfs
+will not create reverse mapping btrees. This feature is only available
+for filesystems created with the (default)
+.B \-m crc=1
+option set. When the option
+.B \-m crc=0
+is used, the reverse mapping btree feature is not supported and is disabled.
.RE
.TP
.BI \-d " data_section_options"
diff --git a/mkfs/maxtrres.c b/mkfs/maxtrres.c
index c0b1b5d..fc24eac 100644
--- a/mkfs/maxtrres.c
+++ b/mkfs/maxtrres.c
@@ -38,7 +38,8 @@ max_trans_res(
int dirblocklog,
int logversion,
int log_sunit,
- int finobt)
+ int finobt,
+ int rmapbt)
{
xfs_sb_t *sbp;
xfs_mount_t mount;
@@ -72,6 +73,8 @@ max_trans_res(
XFS_DFL_SB_VERSION_BITS;
if (finobt)
sbp->sb_features_ro_compat |= XFS_SB_FEAT_RO_COMPAT_FINOBT;
+ if (rmapbt)
+ sbp->sb_features_ro_compat |= XFS_SB_FEAT_RO_COMPAT_RMAPBT;
libxfs_mount(&mount, sbp, 0,0,0,0);
maxfsb = xfs_log_calc_minimum_size(&mount);
diff --git a/mkfs/xfs_mkfs.c b/mkfs/xfs_mkfs.c
index 8b3cad8..634dcfd 100644
--- a/mkfs/xfs_mkfs.c
+++ b/mkfs/xfs_mkfs.c
@@ -680,6 +680,8 @@ struct opt_params mopts = {
"finobt",
#define M_UUID 2
"uuid",
+#define M_RMAPBT 3
+ "rmapbt",
NULL
},
.subopt_params = {
@@ -699,6 +701,12 @@ struct opt_params mopts = {
.conflicts = { LAST_CONFLICT },
.defaultval = SUBOPT_NEEDS_VAL,
},
+ { .index = M_RMAPBT,
+ .conflicts = { LAST_CONFLICT },
+ .minval = 0,
+ .maxval = 1,
+ .defaultval = 0,
+ },
},
};
@@ -1454,6 +1462,7 @@ struct sb_feat_args {
bool crcs_enabled;
bool dirftype;
bool parent_pointers;
+ bool rmapbt;
};
static void
@@ -1524,6 +1533,8 @@ sb_set_features(
if (fp->finobt)
sbp->sb_features_ro_compat = XFS_SB_FEAT_RO_COMPAT_FINOBT;
+ if (fp->rmapbt)
+ sbp->sb_features_ro_compat |= XFS_SB_FEAT_RO_COMPAT_RMAPBT;
/*
* Sparse inode chunk support has two main inode alignment requirements.
@@ -1784,6 +1795,7 @@ main(
.crcs_enabled = true,
.dirftype = true,
.parent_pointers = false,
+ .rmapbt = false,
};
platform_uuid_generate(&uuid);
@@ -2073,6 +2085,10 @@ main(
if (platform_uuid_parse(value, &uuid))
illegal(optarg, "m uuid");
break;
+ case M_RMAPBT:
+ sb_feat.rmapbt = getnum(
+ value, &mopts, M_RMAPBT);
+ break;
default:
unknown('m', value);
}
@@ -2409,6 +2425,20 @@ _("sparse inodes not supported without CRC support\n"));
}
sb_feat.spinodes = 0;
+ if (sb_feat.rmapbt) {
+ fprintf(stderr,
+_("rmapbt not supported without CRC support\n"));
+ usage();
+ }
+ sb_feat.rmapbt = false;
+ }
+
+
+ if (sb_feat.rmapbt && xi.rtname) {
+ fprintf(stderr,
+_("rmapbt not supported with realtime devices\n"));
+ usage();
+ sb_feat.rmapbt = false;
}
if (nsflag || nlflag) {
@@ -2890,7 +2920,8 @@ an AG size that is one stripe unit smaller, for example
%llu.\n"),
min_logblocks = max_trans_res(agsize,
sb_feat.crcs_enabled, sb_feat.dir_version,
sectorlog, blocklog, inodelog, dirblocklog,
- sb_feat.log_version, lsunit, sb_feat.finobt);
+ sb_feat.log_version, lsunit, sb_feat.finobt,
+ sb_feat.rmapbt);
ASSERT(min_logblocks);
min_logblocks = MAX(XFS_MIN_LOG_BLOCKS, min_logblocks);
if (!logsize && dblocks >= (1024*1024*1024) >> blocklog)
@@ -2965,7 +2996,7 @@ _("size %s specified for log subvolume is too large,
maximum is %lld blocks\n"),
mp->m_sectbb_log = sbp->sb_sectlog - BBSHIFT;
/*
- * sb_versionnum and finobt flags must be set before we use
+ * sb_versionnum, finobt and rmapbt flags must be set before we use
* xfs_prealloc_blocks().
*/
sb_set_features(&mp->m_sb, &sb_feat, sectorsize, lsectorsize, dsunit);
@@ -3025,7 +3056,7 @@ _("size %s specified for log subvolume is too large,
maximum is %lld blocks\n"),
printf(_(
"meta-data=%-22s isize=%-6d agcount=%lld, agsize=%lld blks\n"
" =%-22s sectsz=%-5u attr=%u, projid32bit=%u\n"
- " =%-22s crc=%-8u finobt=%u, sparse=%u\n"
+ " =%-22s crc=%-8u finobt=%u, sparse=%u, rmapbt=%u\n"
"data =%-22s bsize=%-6u blocks=%llu, imaxpct=%u\n"
" =%-22s sunit=%-6u swidth=%u blks\n"
"naming =version %-14u bsize=%-6u ascii-ci=%d ftype=%d\n"
@@ -3036,6 +3067,7 @@ _("size %s specified for log subvolume is too large,
maximum is %lld blocks\n"),
"", sectorsize, sb_feat.attr_version,
!sb_feat.projid16bit,
"", sb_feat.crcs_enabled, sb_feat.finobt,
sb_feat.spinodes,
+ sb_feat.rmapbt,
"", blocksize, (long long)dblocks, imaxpct,
"", dsunit, dswidth,
sb_feat.dir_version, dirblocksize, sb_feat.nci,
@@ -3217,6 +3249,12 @@ _("size %s specified for log subvolume is too large,
maximum is %lld blocks\n"),
agf->agf_levels[XFS_BTNUM_CNTi] = cpu_to_be32(1);
pag->pagf_levels[XFS_BTNUM_BNOi] = 1;
pag->pagf_levels[XFS_BTNUM_CNTi] = 1;
+ if (xfs_sb_version_hasrmapbt(&mp->m_sb)) {
+ agf->agf_roots[XFS_BTNUM_RMAPi] =
+ cpu_to_be32(XFS_RMAP_BLOCK(mp));
+ agf->agf_levels[XFS_BTNUM_RMAPi] = cpu_to_be32(1);
+ }
+
agf->agf_flfirst = 0;
agf->agf_fllast = cpu_to_be32(XFS_AGFL_SIZE(mp) - 1);
agf->agf_flcount = 0;
@@ -3404,24 +3442,88 @@ _("size %s specified for log subvolume is too large,
maximum is %lld blocks\n"),
/*
* Free INO btree root block
*/
- if (!sb_feat.finobt) {
- xfs_perag_put(pag);
- continue;
+ if (sb_feat.finobt) {
+ buf = libxfs_getbuf(mp->m_ddev_targp,
+ XFS_AGB_TO_DADDR(mp, agno,
XFS_FIBT_BLOCK(mp)),
+ bsize);
+ buf->b_ops = &xfs_inobt_buf_ops;
+ block = XFS_BUF_TO_BLOCK(buf);
+ memset(block, 0, blocksize);
+ if (xfs_sb_version_hascrc(&mp->m_sb))
+ xfs_btree_init_block(mp, buf,
XFS_FIBT_CRC_MAGIC, 0, 0,
+ agno,
XFS_BTREE_CRC_BLOCKS);
+ else
+ xfs_btree_init_block(mp, buf, XFS_FIBT_MAGIC,
0, 0,
+ agno, 0);
+ libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
}
- buf = libxfs_getbuf(mp->m_ddev_targp,
- XFS_AGB_TO_DADDR(mp, agno, XFS_FIBT_BLOCK(mp)),
+ /* RMAP btree root block */
+ if (xfs_sb_version_hasrmapbt(&mp->m_sb)) {
+ struct xfs_rmap_rec *rrec;
+
+ buf = libxfs_getbuf(mp->m_ddev_targp,
+ XFS_AGB_TO_DADDR(mp, agno, XFS_RMAP_BLOCK(mp)),
bsize);
- buf->b_ops = &xfs_inobt_buf_ops;
- block = XFS_BUF_TO_BLOCK(buf);
- memset(block, 0, blocksize);
- if (xfs_sb_version_hascrc(&mp->m_sb))
- xfs_btree_init_block(mp, buf, XFS_FIBT_CRC_MAGIC, 0, 0,
+ buf->b_ops = &xfs_rmapbt_buf_ops;
+ block = XFS_BUF_TO_BLOCK(buf);
+ memset(block, 0, blocksize);
+
+ xfs_btree_init_block(mp, buf, XFS_RMAP_CRC_MAGIC, 0, 0,
agno, XFS_BTREE_CRC_BLOCKS);
- else
- xfs_btree_init_block(mp, buf, XFS_FIBT_MAGIC, 0, 0,
- agno, 0);
- libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
+
+ /*
+ * mark the AG header regions as static metadata
+ * The BNO btree block is the first block after the
+ * headers, so it's location defines the size of region
+ * the static metadata consumes.
+ */
+ rrec = XFS_RMAP_REC_ADDR(block, 1);
+ rrec->rm_startblock = 0;
+ rrec->rm_blockcount = cpu_to_be32(XFS_BNO_BLOCK(mp));
+ rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_FS);
+ rrec->rm_offset = 0;
+ be16_add_cpu(&block->bb_numrecs, 1);
+
+ /* account freespace btree root blocks */
+ rrec = XFS_RMAP_REC_ADDR(block, 2);
+ rrec->rm_startblock = cpu_to_be32(XFS_BNO_BLOCK(mp));
+ rrec->rm_blockcount = cpu_to_be32(2);
+ rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_AG);
+ rrec->rm_offset = 0;
+ be16_add_cpu(&block->bb_numrecs, 1);
+
+ /* account inode btree root blocks */
+ rrec = XFS_RMAP_REC_ADDR(block, 3);
+ rrec->rm_startblock = cpu_to_be32(XFS_IBT_BLOCK(mp));
+ rrec->rm_blockcount = cpu_to_be32(XFS_RMAP_BLOCK(mp) -
+ XFS_IBT_BLOCK(mp));
+ rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_INOBT);
+ rrec->rm_offset = 0;
+ be16_add_cpu(&block->bb_numrecs, 1);
+
+ /* account for rmap btree root */
+ rrec = XFS_RMAP_REC_ADDR(block, 4);
+ rrec->rm_startblock = cpu_to_be32(XFS_RMAP_BLOCK(mp));
+ rrec->rm_blockcount = cpu_to_be32(1);
+ rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_AG);
+ rrec->rm_offset = 0;
+ be16_add_cpu(&block->bb_numrecs, 1);
+
+ /* account for the log space */
+ if (loginternal && agno == logagno) {
+ rrec = XFS_RMAP_REC_ADDR(block, 5);
+ rrec->rm_startblock = cpu_to_be32(
+ XFS_FSB_TO_AGBNO(mp, logstart));
+ rrec->rm_blockcount = cpu_to_be32(logblocks);
+ rrec->rm_owner = cpu_to_be64(XFS_RMAP_OWN_LOG);
+ rrec->rm_offset = 0;
+ be16_add_cpu(&block->bb_numrecs, 1);
+ }
+
+ libxfs_writebuf(buf, LIBXFS_EXIT_ON_FAILURE);
+ }
+
xfs_perag_put(pag);
}
@@ -3646,7 +3748,7 @@ usage( void )
{
fprintf(stderr, _("Usage: %s\n\
/* blocksize */ [-b log=n|size=num]\n\
-/* metadata */ [-m crc=0|1,finobt=0|1,uuid=xxx]\n\
+/* metadata */ [-m crc=0|1,finobt=0|1,uuid=xxx,rmapbt=0|1]\n\
/* data subvol */ [-d agcount=n,agsize=n,file,name=xxx,size=num,\n\
(sunit=value,swidth=value|su=num,sw=num|noalign),\n\
sectlog=n|sectsize=num\n\
|